PDF_Issue_14_3_Visconti_Pellegata Italian Political Science, VOLUME 14 ISSUE 3, FEBRUARY 2020 * This essay is part of the research project titled ‘Reconciling Economic and Social Europe: The Role of Val- ues, Ideas and Politics’, funded by an Advanced Grant of the European Research Council (Grant No. 340534, P.I. Maurizio Ferrera). © 2020 Italian Political Science. ISSN 2420-8434. Volume 14, Issue 3, 188-205. Contact Author: Francesco Visconti, University of Milan. E-mail address: francesco.visconti@unimi.it Representation in hard times: party-voter distance on support for Social Europe in Italy* Francesco Visconti UNIVERSITY OF MILAN Alessandro Pellegata UNIVERSITY OF MILAN Abstract Political representation is a fundamental virtue of contemporary democracies. The policy preferences of politi- cians should converge to some extent with those of voters. In the last twenty years, and in particular in the aftermath of the multiple crises the European Union (EU) has recently experienced, the integration process has become an increasingly polarizing issue for both voters and political representatives. While the existing literature has investigated party-voter distance on diffuse support for the EU, this article focuses on preferences for EU- level policies aiming to strengthen European solidarity. We argue that although Italian voters and their political representatives tend to diverge on their general support for the EU, they are closer over their willingness to share the burden across EU member states of the multiple crises Italy has recently experienced. Employing original mass and elite survey data collected between the end of 2016 and the beginning of 2018 in the framework of the REScEU project, our empirical findings show that both MPs and their voters strongly support European solidarity, though they also detect differences across parties. The most important implication of this study is that the mounting Euroscepticism of Italian voters is not an outright rejection of the EU but a call for the EU’s proac- tive role in protecting weaker countries and peoples. 1. Introduction n representative democracies the decisions made by policy-makers should, to some degree, mirror the preferences of their voters (Downs 1957). The present article looks at collective representation (Wlezien and Soroka 2016) by comparing the (general and specific) attitudes towards the European Union (EU) of Italian citizens with those of their parliamentary representatives. While most of the extant literature analyses the gap between masses and elites in their diffuse support for the EU (Dolny and Babos 2015; Mat- tila and Raunio 2006; 2012; McEvoy 2012; Thomassen and Schmitt 1999), our study takes a less travelled road by focusing also on the differences in how the demand side and the supply side of the chain of representation evaluate the policy-making role played by EU institutions (see Sanders and Toka 2013 or Muller et al. 2012). More precisely, we investi- gate party-voter issue proximity on preferences for policies aiming to foster European I VISCONTI and PELLEGATA 189 solidarity, an issue that in the aftermath of the Euro crisis has become highly salient and contentious in several EU member states. In the last two decades, a broad literature has discussed how EU integration and other macro-level phenomena such as globalisation and massive migration have contrib- uted to reshaping party competition in several European countries by generating a new political divide that deals with the opening of national borders and cross cuts the tradi- tional left-right dimension (Kriesi 2008; 2012; Hooghe and Marks 2018; Otjes and Katsanidou 2017). Considering its increasing policy-making role during the Euro and the refugee crises, the EU has further polarised the attitudes of both voters and political elites. What is increasingly debated are the policy initiatives and measures that the EU has adopted (or should adopt) to face the negative externalities of the multiple crises it has experienced and to avoid future predicaments: in other words, policies fostering solidar- ity across EU member states as well as European citizens. Financial help to states facing severe economic difficulties, the harmonisation of welfare policies to guarantee a high standard of social security to all Europeans in need, and the redistribution of extra-EU migrants and asylum seekers among EU member states are all items on the agenda of po- litical parties that campaign on these issues to gain electoral consensus. However, while several recent studies investigate the determinants of public support for European soli- darity (Baute et al. 2019; Ciornei and Recchi 2017; Ferrera and Pellegata 2018; Gerhards et al. 2019), we have significantly less knowledge about the attitudes expressed by political elites (see Conti 2018 for an exception). To the best of our knowledge, moreover, no study expressly compares voter and party preferences on this highly contentious topic. Italy represents an interesting case to study party-voter distance on the topic of Eu- ropean solidarity. It was badly hit by the economic recession, it implemented a series of fiscal austerity measures and structural reforms to keep its public debt under control, and it is one of the first landing places for migrants coming from the African continent. On the political side, Italy has shown a dramatic decrease in popular support for the EU and trust in its institutions and, at the same time, a significant increase in electoral consent for Eurosceptic parties, though with different nuances, such as the Five Star Movement (M5S), the League (L) and Brothers of Italy (FdI). Thus, do voter and party preferences tend to converge or diverge on the issues of Eu- ropean integration and solidarity? Are there differences across parties in how close political representatives are to their voters? We expect to find a significant divergence be- tween Italian voters and elites on a general evaluation of the benefits of the EU project with the former being more sceptical than the latter. However, we expect to find more convergence on support for policies strengthening solidarity in the EU. Given Italy’s ex- posure to the crises, both voters and their representatives demand EU-policy intervention to correct economic and social imbalances among EU member states that the recent challenges experienced by the EU have exacerbated. We provide answers to these research questions by employing original public opin- ion and political elite survey data collected between the end of 2016 and the beginning of 2018 within the framework of the REScEU (Reconciling Economic and Social Europe: Values, Ideas and Politics) research project conducted at the University of Milan in the 2014-2019 period. While exploratory in nature due to the limitations of the data, we be- lieve that the present study provides useful contributions to the literature on both EU Representation in hard times 190 support and representation in Italy. Regarding the former, by comparing party-voter dis- tance not only on the general view of the EU but also on its policy role, we shed light on the nature of Euroscepticism in Italy. Concerning the latter, we provide important insights on how the integration-demarcation dimension is contributing to restructure the prefer- ences of political actors in the Italian party system. 2. Background and expectations 2.1. Italy and the crises The multiple crises that the EU has experienced in the last decade have contributed to exacerbate the line of tension over the integration process and to generate a new conflict constellation within the EU (Armingeon and Ceka 2014; Ferrera 2017). The bailout of the Greek economy in exchange for severe austerity measures, Merkel’s decision to allow more than one million asylum seekers into Germany and the outcome of the Brexit refer- endum as well as other critical junctures became salient political issues for national governments and affected the everyday life of European citizens. However, the consequences of the multifaceted European crisis were unevenly dis- tributed across EU member states. Italy, as well as other weak economies of the Eurozone periphery, was strongly affected by the detrimental consequences of the global economic downturn and the sovereign debt crisis. Since 2009, the main macro-economic indicators have abruptly deteriorated. The sharp decline in GDP caused the Italian economy to enter recession and in 2012 – the peak of the sovereign debt crisis – Italian public debt rose to 126.16 per cent of GDP – a figure considerably higher not only than the average value of the Eurozone, but also than the average value of southern European member states (Pedrazzani et al. 2018). Even though Italy, differently from Greece, Ireland, Portugal and Spain, did not formally sign a bailout agreement with the ‘Troika’ (EU Commission, Eu- ropean Central Bank and International Monetary Fund), the country committed itself to an ‘implicit conditionality’ programme that implied fiscal austerity measures and struc- tural (labour market and pensions) reforms (Moschella 2017; Sacchi 2015). The economic downturn and the policies implemented by the technocratic Monti government (2011-2013) had relevant negative social consequences (Natili 2019). Unem- ployment rose from 6.1 per cent of the labour force in 2007 to 13.1 per cent in 2014 and the scenario was even more alarming concerning youth unemployment, which rose from 20.4 per cent in 2007 to a remarkable 42.7 per cent in 2014. Poverty rose as well: the share of households living in absolute poverty increased from 4.7 per cent in 2009 to 7.9 per cent in 2013 (ISTAT 2019a). The deterioration of the objective indicators was paralleled by mounting public concern about the situation of the national economy, with a peak of neg- ative evaluations – over 92 per cent of Italian citizens – in 20121. Besides the economic crisis, Italy is facing other important challenges deriving from increasing migration flows into the country. Given its geographic position, in the after- math of the Arab Spring, Italy became one of the countries of first landing for migrants and asylum seekers coming from the African continent. Immigration contributed to 1 See Eurobarometer data on Italian citizens’ evaluation of their national economy: http://ec.eu- ropa.eu/commfrontoffice/publicopinion/index.cfm/Chart/index . VISCONTI and PELLEGATA 191 change the composition of Italian society. While in the early 2000s foreign-born residents were around 1.5 million, at the beginning of 2018 they were around 5 million, about 8.5% of the total population (ISTAT 2019b). This mutated social landscape contributed to affect the Italians’ attitude towards immigration. A series of IPSOS surveys highlight that Ital- ian citizens tend to overestimate the real number of immigrants living in their country.2 In the weeks before the 2018 national elections, 62 per cent of respondents who partici- pated in the PASTEL survey believed that Italy had already hosted too many immigrants and 43 per cent of them believed that foreigners represent a threat to Italian culture.3 The consequences of these multiple challenges on party system configuration and electoral competition dynamics are manifest for the observers of Italian politics. Indeed, the 2013 and 2018 general elections caused a political earthquake (Chiaramonte et al. 2018; Emanuele 2018; Schadee et al.2019). In line with a pattern emerging also in other EU member states badly hit by the economic crisis (Hutter et al. 2018), the last two elec- tion rounds marked a significant increase in votes for Eurosceptic parties at the expense of the mainstream parties – the Democratic Party (PD) and Go Italy (FI) – that had rotated in government since the mid-1990s. In 2013 M5S – a populist and anti-elite party – ob- tained 25.6 per cent of the vote, which is the best result ever obtained by a newcomer party in national elections in Europe. After five years in opposition, in 2018 M5S managed to again increase their vote share (32.7 per cent) becoming the pivot in the parliament and the major party in the two different government coalitions it has experienced so far. The 2018 elections also marked a victory for the League – a right-wing Eurosceptic party ren- ovated under Matteo Salvini’s leadership – which quadrupled its vote share (17.4%) compared with the 2013 elections (4.1%). 2.2. Diffuse and Policy support for the EU during the crisis Against this background, a number of recent studies depict how the average level of public support for the EU among Italian citizens – traditionally among the highest across EU member states from the 70s to the 90s – sharply decreased after the crises that invested Eu- rope in the last decade (Olmastroni and Pellegata 2018; Conti et al. 2020; Serricchio 2018). The deterioration in the level of support for the EU is striking whichever dimension we con- sider: polity identification, trust in EU institutions, benefit from EU membership or the deepening of the integration process (Olmastroni and Pellegata 2018; Conti et al. 2020). A similar trend is detected in all southern EU member states, while in northern and continen- tal countries the average support for the EU remained high also during the crisis (2011- 2016). The literature argues that economic concerns and perceived threats to national iden- tity are the main drivers of Euroscepticism (Bellucci 2014; Conti and Memoli 2015). In hard times, individuals blame the EU because they consider it responsible for the worsening of both their personal and their country’s economic conditions. While a broad literature has investigated trends of public support for the EU, we know incredibly less about political elites’ preferences on this increasingly contentious issue. Figure 1 compares the attitudes of Italian voters and their representatives in the Chamber of Deputies towards the EU. Data on voters are taken from the Eurobarometer 2 See the IPSOS ‘perils of perceptions’ project (https://perils.ipsos.com/). 3 Participated platform for the study of Italian Elections 2018, a project of the Department of Social and Political Sciences of the University of Milan and Ipsos Italia. REScEU project sample. Representation in hard times 192 88.1 (October 2017) while data on national representatives come from the REScEU elite survey conducted between 2017 and 20184. The column chart displays the share of voters and political representatives who believe that Italy has benefited or has not benefited from EU membership. We chose this question to compare mass and elite support for the EU for two main reasons. First, this is one of the most commonly used survey items in the literature on EU support. By tapping the utilitarian calculation with which individuals evaluate their country’s membership in the EU, this is particularly suitable for measuring how Italians’ judgement of the EU has changed during the crisis. Second, this is one of the few survey items for which data are available at both mass and political elite level. Figure 1. Diffuse support for the EU among Italian masses and elites. Note: ‘Taking everything into consideration, would you say that [COUNTRY] has on balance benefited or not from being a member of the European Union? 1) It has benefited; 2) It has not benefited.’ Not surprisingly, in late 2017, 46.2 per cent of Italian respondents believed that their country had not benefited from being a member of the EU. This percentage is one of the highest among the 28 EU member states and it reflects a pattern common to other south- ern EU member states, such as Cyprus (49.7 per cent) and Greece (45.9 per cent). However, citizens of other countries that have been badly hit by the Euro crisis such as Portugal (14.2 per cent) and Spain (20.4 per cent) tend to evaluate much more positively their country’s membership in the EU. On average, instead, Italian political representa- tives are much more satisfied than their electorate with Italy’s membership in the EU. 73.7 per cent of the MPs interviewed believe that Italy has benefited from being a mem- ber of the EU, while only 26.3 per cent express a negative view about the EU. Although the support for the EU expressed by Italian political elites has also wors- ened in the last decade, even after the crisis an overwhelming majority of MPs tend to express a positive evaluation of Italy’s membership in the EU.5 More than 30 points sep- arate Italian MPs from their voters, with the former being more supportive of the EU 4 Detailed information about the survey is reported in the next section. 5 Data taken from the IntUne elite survey show that the share of MPs who believe that Italy has not bene- fited from being a member of the EU was 4.1 per cent in 2007 and 2.6 per cent in 2009 (Best et al. 2012). 40.2 73.7 46.2 26.3 13.6 0 10 20 30 40 50 60 70 80 90 100 Voters MPs Sh ar e of r es po nd en ts Benefited Not benefited Don't know VISCONTI and PELLEGATA 193 than the latter. This result confirms that the EU can still be considered an elite-driven project (Haller 2008). The masses’ permissive consensus that allowed a deepening and a widening of the integration process for more than three decades after the Treaty of Maastricht (1992) turned into an explicit dissensus that started to constrain the work of the EU institutions (Hooghe and Marks 2009; Müller et al. 2012; Sanders et al. 2012; Conti 2014). Opposition to the integration process became explicit especially in those countries badly hit by the multiple crises experienced by the EU. In these countries citi- zens voiced criticism of the role played by the EU institutions during the crisis and Eurosceptic parties entered the government. However, if we turn to analyse support for specific policies that can be implemented by the EU institutions to challenge the most contentious and detrimental aspects of the different European crises, we expect to find closer views between Italian citizens and their representatives in parliament. We start from the assumption that, since Italy found itself in the eye of the storm of simultaneous crises, the relevance of the utilitarian cal- culation over the role of the EU increases both for public opinion and among political elites. Given that the EU is a supranational project built with the aim of providing solu- tions to large-scale problems, such as the economic recession and the humanitarian crisis related to massive migration flows, in a socio-tropic perspective Italian masses and elites should be more supportive of EU-level policies aiming to strengthen solidarity across EU member states by sharing among them the burden of the crises. In an appar- ently paradoxical scenario, while Italian citizens are more negative than their official representatives when expressing a general judgment of the EU as a polity project, they share preferences closer to those of the elite for policies that can help Italy, as well as the other countries in difficulties, to face the detrimental consequences of the crises. Recent studies have investigated which factors contribute to explain mass and elite support for burden sharing and solidarity measures across EU member states and for delegating policy initiative to the EU (Basile and Olmastroni 2019; Conti et al. 2020). The degree of exposure to the crisis at both the country and individual levels correlate signif- icantly with support for European solidarity. The present study, instead of assessing mass and elite correlates of support for a more proactive role of the EU, remains inter- ested in the existence of representation gaps between voters and parties over EU-level policies strengthening European solidarity. In doing so we mainly focus on differences across parties. We are firstly interested in comparing the level of party-voter distance between mainstream and challenger parties. Following the definition provided by Hobolt and Tilley (2016), challenger parties are not necessarily new but are those parties that have ‘sought to reshape the political landscape by putting new issues on the agenda’ (Hobolt and Tilley 2016, 974). In the following anal- yses we consider the M5S and the L as two challenger parties given the issue- entrepreneurship role they played in the recent developments in Italian politics. They have been able to mobilise the electorate on new issues like European integration and im- migration, by breaking the previous consensus.6 Challenger parties should also not have 6 The 2017 Chapel Hill Expert Survey (Polk et al. 2017) considers the L and the M5S as the two parties among those investigated with the lowest values in overall orientation towards European integration (variable goes from 1 ‘Strongly opposes’ to 7 ‘Strongly in favour’). Party scores are: L 1.5; FI 4.714; M5S 2.643; PD 6.5; SI 4.5; MDP 5.363. Representation in hard times 194 had previous government experience. While this was true for the M5S at the time the two surveys were conducted, this is not the case for the L who held important government po- sitions in all governments led by Silvio Berlusconi. Still, the new leadership of Matteo Salvini has given a new and distinct characterisation to the party, and thus we consider it a challenger.7 Due to their role as issue entrepreneurs, we expect challenger parties to re- duce the gap with their voters, as well as with the median Italian voter, compared to mainstream parties (PD and FI) that have registered major electoral setbacks. 3. Defining and measuring European solidarity The concept of solidarity refers to individuals’ agreement to share resources with others by personal contribution through taxation and redistribution organised by the state (Ge- lissen, 2000; Stjernø, 2009). It is a definition grounded in the very notion of nation state, and historically solidarity has been institutionalised through the development of national welfare states. Since the 1980s, the successive waves of European integration have put under stress nation-based welfare systems. This was primarily due to the tension between the ‘logic of closure’ upon which nation-based welfare states developed and the ‘logic of openness’ of EU integration, represented by the completion of the internal market and then by the in- stitutionalisation of the monetary union. This tension threatening the social sovereignty of EU member states has forced institutions and citizens to reframe the concept of soli- darity in European terms (Ferrera 2017). In the EU wide context, solidarity has been (re-)defined as the individual willingness to share risks across the EU (Ciornei and Recchi 2017:470). Sangiovanni (2013) differen- tiates between two dimensions of European solidarity: member state and transnational solidarity. The former refers to risk sharing across EU member states. The latter entails the sharing of obligations among EU citizens (see also Ciornei and Recchi 2017; Baute et al. 2019). Here we consider both these dimensions of European solidarity. The following analyses draw on two surveys conducted within the REScEU project. The first is a public opinion survey carried out in the autumn of 2016 that evaluates citi- zens’ preferences on EU solidarity. It includes 1,320 Italian respondents interviewed through the CAWI method and sampled around age, education, and area of residence. The second is an elite survey conducted on national members of parliament (MPs) be- tween 2017 and 2018. It includes 87 Italian MPs of the lower chamber (Camera dei Deputati) elected in 2013, of whom 76 answered all relevant EU solidarity items. In both surveys we identified five items designed to measure Italian citizens’ and MPs’ support for policy programmes introducing cross-national and trans-national forms of European solidarity. Table A1 in the Appendix lists these five items along with their response categories. The first question asks respondents whether they are in favour of or against the introduction of common European bonds. While undoubtedly a technical issue, so-called Eurobonds became a salient topic after the Euro crisis in the Italian de- bate. Think for instance about the M5S manifesto prepared for the 2014 European Parliament elections that explicitly mentioned the introduction of Eurobonds as one of 7 Brothers of Italy can also be considered a challenger party, but due to the low number of MPs of this party that took part in the elite survey we could not consider them in our analyses. VISCONTI and PELLEGATA 195 their seven key proposals (Della Porta et al.2017). The second question asks respondents to indicate whether they are in favour of or against the introduction of a common EU fund compensating national governments and local communities for the costs related to im- migration from other EU member states. In the last five years in Italy as well as in other countries both intra and extra EU immigration have become a more salient topic of polit- ical debate on which right-wing populist parties have capitalised. The third EU solidarity item measures respondents’ preferences on the institution of common EU social insur- ance schemes (such as healthcare, unemployment or pensions) covering intra-EU migrant workers. This question captures preferences over forms of trans-national soli- darity, and preferences over one of the cardinal achievements of the EU: free movement. The fourth question asks citizens and MPs whether or not they agree with the introduc- tion of an EU budget large enough to provide substantial financial help to member states facing a sudden rise in unemployment rates, which remains among the top concerns in the minds of Italian citizens. The fifth and final question asks citizens and MPs whether they agree that in the case of a severe financial crisis in a given member state, the EU should make sure that no citizen remains without means of subsistence (food, shelter, es- sential medicines etc.). This item depicts a social crisis triggered by severe economic turmoil, a scenario similar to what happened in Greece during the Euro crisis. Responses were given using 4-point Likert scales (1 - Strongly disagree/against, 2 - Somewhat disagree/against, 3 - Somewhat agree/in favour, 4 - Strongly agree/in fa- vour), where lower values mean anti-solidarity preferences, and higher values pro- solidarity views.8 In the next section we gauge the congruence between Italian voters and party repre- sentatives both on the single dimensions, as well as on a comprehensive measure of EU solidarity. This consists of an additive index based on the summation of respondents’ an- swers on the five solidarity items presented above. The index has then been rescaled to vary between ‘0 - Anti-EU solidarity’ and ’10 - Pro-EU solidarity’. The internal con- sistency of this index is ensured by a factor analysis run on polychoric correlations separately for the elites and the mass datasets, confirming that these items are captured by only one underlying dimension.9 To investigate the congruence between Italian MPs and their electorate, we have matched them based on the party group of the MP at the moment of the interview, and on the future vote intention of citizens. To ensure consistency we included in our anal- yses only parties for whom more than three MPs and at least ten voters have been interviewed. Table A2 in the Appendix lists the political parties with the number of vot- ers and MPs considered in the analysis: Sinistra Italiana + Movimento Democratico e Progressista (SI+MDP), Partito Democratico (PD), Movimento 5 Stelle (M5S), Forza Ita- lia (FI), and Lega (L). 8 Employing short answers scales to survey questions protects against the risk of excessive influence of outliers. This is a way to make findings relatively more robust given the difficulties of collecting elite data and the relatively small number of observations that does not allow us to conduct more sophisticate tests. 9 Table A3 in the Appendix displays results obtained from factor analysis. The correlation between the additive index of EU solidarity and the factor extracted from the factor analysis is of 0.995 for MPs and 0.993 for citizens. Representation in hard times 196 4. Assessing party-voter differences Here we are interested in comparing Italian public opinion and political elites on the Eu- ropean solidarity issue. To this end we look at two dyads: issue congruence between citizens and MPs, and between voters and parties. Firstly, we compare the average levels of support for European solidarity expressed by our samples of Italian citizens and politi- cal representatives, irrespective of their different political orientations and partisan affiliations. With this analysis we aim to detect whether political elites are also more sup- portive of policies fostering European solidarity than citizens, as they are for the EU in general. Then we turn to compare the distance in support for European solidarity between major Italian parties and their voters to assess whether there are differences between mainstream and challenger parties. Rather than focusing on the individual representa- tive, we scrutinise party-level representation because the electoral system in place at the time of the elite survey emphasised the central role of parties in the chain of representa- tion. The ‘Calderoli electoral law’ was indeed a proportional representation system with a majority prize for the plurality party/coalition, large districts and blocked lists in which the voter could not express any preference vote. While directly comparing distributions within two groups is a viable approach (Golder and Stramski 2010; Lupu et al. 2017) we follow the classic representation literature that resorts to measures of central tendency like the mean and the median (Huber and Powell 1994; Müller et al. 2012) to evaluate ab- solute congruence. Table 1 reports summary statistics for the additive EU solidarity index and each of its five components for the Italian masses and elites in the aggregate. They show that on aver- age, both the Italian public and elites favour a very high level of EU solidarity. On average, Italian citizens share an EU solidarity score of 7.35 on a scale from 0 to 10. At the same time, MPs are even closer to the pro-EU solidarity pole, with an average value of 8.33. Looking at each solidarity item separately, we note that the interpolated median10 is always closer to the pro-solidarity pole of the measurement scale for both groups with only slight mass-elite differences. Italian political elites are a bit more pro-EU solidarity than the average Italian citizen on all policies but one – preferences for an EU budget large enough to provide financial help to MSs in case of a sudden rise in unemployment. In this case, no relevant difference between the two groups emerges. The introduction of Eurobonds is the solidarity policy towards which both the public and elites are rela- tively more sceptical. Instead, the one receiving the broadest support is the policy programme ensuring the EU will guarantee to all its citizens that no one is left behind in the aftermath of an economic crisis. The two European solidarity policies relating to the free movement of European citizens, the fund compensating for the costs of man- aging intra-EU migration and the institutionalisation of a supra-national social insurance, are instead the two proposals with the relatively most significant discrep- ancy between masses and elites. At the aggregate level, we find support for our expectation of observing closer views between Italian citizens and their representatives when it comes to specific EU level pol- icy programmes fostering solidarity, rather than when looking at diffuse support. 10 When dealing with ordinal variables with a limited number of responses, comparing simple medians may not be informative: two distributions with equal median may be heavily weighted above or below the median. The interpolated median provides an alternative measure of centre which takes into account the percentage of the data strictly below or strictly above the median. It gives a measure within the upper and lower bound of the median, in the direction that the data is more heavily weighted. VISCONTI and PELLEGATA 197 Table 1. Summary statistics for EU solidarity index and its five components. EU solidarity index Mean Standard Error Median St. Dev. N Elite 8.33 0.154 8.67 1.34 76 Mass 7.35 0.043 7.33 1.56 1295 Introduction of Eurobonds Mean Standard Error Interpolated median St. Dev. N Elite 3.37 0.086 3.36 0.75 76 Mass 2.86 0.02 2.94 0.85 1295 EU fund to compensate for immigration costs Mean Standard Error Interpolated median St. Dev. N Elite 3.55 0.076 3.70 0.66 76 Mass 3.19 0.022 3.25 0.80 1295 EU social insurance covering intra-EU migrants Mean Standard Error Interpolated median St. Dev. N Elite 3.47 0.081 3.58 0.70 76 Mass 3.01 0.022 3.07 0.80 1295 EU should provide financial help to MSs in case of sud- den rise in unemployment Mean Standard Error Interpolated median St. Dev. N Elite 3.34 0.078 3.44 0.69 76 Mass 3.39 0.018 3.46 0.66 1295 EU should ensure subsist- ence in case of financial crisis Mean Standard Error Interpolated median St. Dev. N Elite 3.76 0.049 3.85 0.42 76 Mass 3.57 0.017 3.70 0.63 1295 Of course, aggregate level differences can hinder variation at a lower level of analysis. The next figures present data disaggregated by party, portraying the median position of parties (blue dots), of their voters (red triangles), and the median position of all Italian citizens interviewed in the mass survey (dotted line) on all measures of EU solidarity. Fig- ure 2 portrays the median on the EU solidarity index for the five parties meeting the conditions described in the previous section and listed from left to right on the vertical axis: SI+MDP, PD, M5S, FI, L.11 The horizontal axis records instead the median on the EU solidarity additive index for both party members and for citizens who would vote for these parties. Four insights emerge from the figure. First, there is more between-party variation in support of European solidarity among political elites than among voters. While median supporters of parties tend to be closer to the median citizen, the median representatives of the same parties, though they tend to be more pro-solidarity than the median citizen, present more differences. Second, while in the aggregate the elites were more pro- 11 The left-right positioning of Italian parties is based on the 2017 Chapel Hill Expert FLASH Survey (Polk et al. 2017) available at https://www.chesdata.eu/1999-2014-chapel-hill-expert-survey-ches-trend-file- 1.The position of the parties on the LRGEN variable measuring parties’ ideology – where 0 means ‘Ex- treme left’ and 10 ‘Extreme right’ – is the following: SI 1.36; MDP 2.23; PD 3.8; M5S 5.2; FI 6.533; L 8.26. Representation in hard times 198 European solidarity than the masses, this is not always true when we look at each party separately. The aggregate level result holds for SI+MDP, PD, and L. The two parties lo- cated closer to the left pole of the ideological axis are those with the largest distance between elites and masses (for both parties, voters and MPs shared the same median po- sition). Solidarity policies are the preferred terrain of battle for left-wing parties, issues that therefore they tend to emphasise more in their political activities. The same also holds for the League of Matteo Salvini. Interestingly, L voters are the only group whose median lies below the overall median of Italian citizens. Third, MPs of the L, together with the M5S representatives, are the least supportive of European solidarity. Nevertheless, compared to the left-wing MPs, they are closer both to the L median voter and to the me- dian citizen. Finally, FI and M5S show perfect congruence between national MPs and their voters. Voters of Berlusconi’s party present the same median score on the solidarity index as the left-wing voters (PD and SI+MDP). However, FI MPs are less prone to sup- porting pro-solidarity policies than their left-wing counterparts. The perfect congruence of the position of M5S representatives not only with their electoral base but also with the median Italian citizen on the issue of EU solidarity is probably related to their electoral success in the general elections of 8 March. It falls in line with the pivotal role played by the M5S in the Italian political system since the 2013 elections Figure 2. Medians for voters and elites on the EU solidarity index. Looking at the additive index may hinder differences on the specific policy pro- posals considered. While the underlying structure of preferences on the five issues is similar, it is not exactly the same and there is some independent variation across the five EU solidarity policies. Therefore, Figures 3 and 4 show the interpolated medians on each of the five components of European solidarity. The first figure depicts the three items measuring preferences over the introduction of forms of cross-national ● ● ● ● ●SI+MDP PD M5S FI LEGA 5 6 7 8 9 10 Median Pa rt y ● Elite Voters EU solidarity · · · Country median VISCONTI and PELLEGATA 199 solidarity, while the second looks at the transnational forms of EU solidarity. First, we find that there is greater congruence on issues related to poverty and unemployment, two problems that the economic crisis has exacerbated since 2009. Both parties’ and voters’ preferences on these issues are very close to the pro-EU-solidarity extreme. Therefore, regardless of the polity-level Euroscepticism characterising Italian citizens, it emerges that a solution to these problems lies at the supra-national level through fur- ther integration in the minds not only of MPs but also of voters. Second, we find lower congruence when we look at support for the introduction of Eurobonds to pool risks re- lated to public debt. This is a complex issue to be understood and processed by citizens, and across parties they tend to align close to the overall country median. Third, the two policies related to freedom of movement are the ones with more variation across parties in terms of congruence. Figure 3. Interpolated medians for voters and elites on cross-national components of EU solidarity. ● ● ● ● ●SI+MDP PD M5S FI LEGA Strongly disagree 2 3 Strongly agree Interpolated median Pa rt y ● Elite Voters Eurobonds · · · Country interpolated median ● ● ● ● ●SI+MDP PD M5S FI LEGA Strongly disagree 2 3 Strongly agree Interpolated median Pa rt y ● Elite Voters EU fund compensating for immigration's costs · · · Country interpolated median ● ● ● ● ●SI+MDP PD M5S FI LEGA Strongly disagree 2 3 Strongly agree Interpolated median Pa rt y ● Elite Voters EU financial help to face unemployment · · · Country interpolated median Representation in hard times 200 Concerning the first policy, the introduction of an EU fund compensating member states for immigration costs, we find a low level of congruence for all parties but SI+MDP, which perfectly match their voters, and the M5S. As regards this policy, of particular in- terest is the distance between the interpolated median of MPs of the L and their voters, which is the most extreme distance between elites and voters found in our data. L’s voters are only moderately in favour of more integration policies in the field of immigration, and less so than the median citizen. Instead, their representatives hold a very favourable opin- ion about the introduction of this programme. This result could hint at the outright rejection of immigration on behalf of voters, notwithstanding refunds for its manage- ment. On the second policy item related to free movement, the introduction of a social insurance scheme covering workers moving from one EU member state to another, our data suggest an association between the interpolated median position of MPs and the left- right positioning of their party: the more a party locates itself on the left, the more it sup- ports such a form of transnational solidarity. Still, parties located on the left of the ideological spectrum are those less congruent with their voters. Even if M5S representa- tives are less in favour of EU solidarity policies than left-wing parties (apart from the item on subsistence in case of a severe crisis), they tend to be very much attuned to their voters and the general citizenry. Figure 4. Interpolated medians for voters and elites on transnational components of EU solidarity These results suggest the M5S has been able to catalyse the priorities and preferences of Italian citizens on the fight against poverty and unemployment, and the need for new welfare instruments which are much better than its competitors. These issues have been central in the M5S policy agenda and key to its electoral success. Instead, these results somewhat contradict the official M5S rhetoric of a party that wants a return to the princi- ples of solidarity and community in the EU (Della Porta et al.2017). When compared to the other parties included in our analysis, their MPs, as well as their voters, present posi- tions that are either less pro-EU solidarity or not significantly different from other parties. ● ● ● ● ●SI+MDP PD M5S FI LEGA Strongly disagree 2 3 Strongly agree Interpolated median Pa rt y ● Elite Voters EU should provide means of subsistence · · · Country interpolated median ● ● ● ● ●SI+MDP PD M5S FI LEGA Strongly disagree 2 3 Strongly agree Interpolated median Pa rt y ● Elite Voters EU social insurance · · · Country interpolated median VISCONTI and PELLEGATA 201 5. Concluding remarks This article investigated party-voter distance on support for policies strengthening the so- cial dimension of Europe in Italy. In the aftermath of the multiple crises that the EU has experienced in the last decade, solidarity and redistribution of resources across EU member states and citizens have become salient and highly contentious issues among both political elites and citizens. While Italian voters are much more sceptical than their representatives about the benefit brought by their country’s membership in the EU, they are much closer to them when it comes to supporting policies strengthening European solidarity. In a country such as Italy, badly hit by the Euro crisis and by the negative externalities of fiscal austerity, both public opinion and political elites blame EU institutions for being unable to face the unevenly distributed challenges coming from the crisis and ask for policies aiming to redis- tribute resources and correct imbalances across EU member states. By means of a public opinion and elite survey under the framework of the REScEU project, we analysed the Ital- ian mass-elite gap on support for policies fostering European solidarity, focusing on differences across parties. We found a high level of congruence between mass and elite pref- erences towards pro-solidarity policies. Left-wing parties (PD and SI+MDP) are the most supportive of European solidarity. For these parties as well as for the League we detected a certain amount of divergence between MPs and their voters, with the former tending to be more pro-EU solidarity than the latter. Our analysis revealed instead a perfect match-up of the M5S MPs’ preferences and those of both their voters and the Italian median citizen. Vot- ers of the League show they are the most Eurosceptic, while M5S party representatives show on overage a lower level of support for pro-EU solidarity policies, even when they were part of their official electoral manifesto as in the case of Eurobonds. These findings seem to cor- roborate works on the issue-entrepreneurial role of challenger parties and their ability to better represent the median citizen (Hobolt and Tilley 2016). While exploratory in nature, our study suggests two relevant and interrelated implica- tions. First, our findings show that, though salient, EU-related issues do not seem to contribute to reshaping political competition in Italy. Although we detected few cross-par- tisan differences, voters and MPs have preferences in favour of policies that can contribute to sharing the burden of the multiple crises Italy is experiencing with other EU member states by redistributing risks, resources and responsibilities. Second, this study sheds light on the nature of the Euroscepticism that is increasingly characterizing large sectors of Ital- ian public opinion as well as important parties. It is plausible that Italians are sceptical about the EU institutions for being unable, or unwilling, to implement policies that can help those member states that, more than others, are facing the challenges provided by the multiple crises that recently invested Europe. However, both Italian voters and their political repre- sentatives do not consider the EU a project deemed to fail but, on the contrary, they ask for a proactive role of the EU in protecting weaker countries and peoples. References Armingeon, K., and B. Ceka (2014) ‘The loss of trust in the European Union during the great recession since 2007: The role of heuristics from the national political system’, European Union Politics 15(1) 82–107. Representation in hard times 202 Basile, L. and F. Olmastroni (2019) ‘‘Sharing the burden in a free riders’ land: The EU migra- tion and asylum policy in the views of public opinion and politicians’, early published online ahead of print in European Journal of Political Science. Baute, S., K. Abst and B. Meleuman (2019) ‘Public Support for European Solidarity: Between Euroscepticism and EU Agenda Preferences?’, JCMS: Journal of Common Market Stud- ies, 57(3), 533-550. Bellucci, P. (2014). ‘The political consequences of blame attribution for the economic crisis in the 2013 Italian national election.’ Journal of Elections, Public Opinion and Parties 24(2), 243-263. Chiaramonte, A., Emanuele, V., Paparo, A. and Maggini, N. (2018) ‘Populist Success in a Hung Parliament: The 2018 General Election in Italy’, South European Society and Politics, 23(4), 479-501. Ciornei, I., and Recchi, E. (2017). ‘At the source of European solidarity: Assessing the effects of cross-border practices and political attitudes.’ JCMS: Journal of Common Market Studies, 55(3), 468-485. Conti, N. (ed.) (2014), Party Attitudes Towards the EU in the Member States. Parties for Eu- rope, arties against Europe, London, Routledge. Conti, N., and V. Memoli (2015). ‘Show the money first! Recent public attitudes towards the EU in Italy.’ ItalianPolitical Science Review/Rivista Italiana di Scienza Politica 45(2), 203-222. Conti, N. (2018). ‘National political elites, the EU, and the populist challenge.’ Politics, 38(3), 361-377. Conti, N., D. Di Mauro and V. Memoli (2020) ‘Immigration, security and the economy: who should bear the burden of global crises? Burden sharing and citizens’ support for EU in- tegration in Italy’. Contemporary Italian Politics, 12(1), 77-97. Della Porta, D., Fernández, J., Kouki, H., and Mosca, L. (2017) Movement parties against aus- terity, John Wiley and Sons, Cambridge. Dolný, B., and Baboš, P. (2015). ‘Voter–Representative Congruence in Europe: A Loss of Insti- tutional Influence?’, West European Politics, 38(6), 1274-1304. Downs, A. (1957) An Economic Theory of Democracy, New York, NY: Harper and Row. Emanuele, V. (2018) ‘Who’s the winner? An analysis of the 2018 Italian general election’, Ital- ian Political Science 13(1), 1-7. Ferrera, M. (2017). ‘The Stein Rokkan Lecture 2016. Mission impossible? Reconciling eco- nomic and social Europe after the euro crisis and Brexit.’ European Journal of Political Research, 56(1), 3-22. Ferrera, M. and Pellegata, A. (2018) “Worker mobility under attack? Explaining labour mar- ket chauvinism in the EU”, Journal of European Public Policy 25(10), 1461-1480. Gelissen, J. (2000). Popular support for institutionalised solidarity: a comparison between Eu- ropean welfare states. International Journal of Social Welfare, 9(4), 285-300. Gerhards, J., H. Lengfeld, Z.S. Ignacz, F. Kley, and M. Priem (2019) European Solidarity in Times of Crisis: Insights from a thirteen-country survey. Abington: Routledge. Golder, M. and Stramski, J. (2010) ‘Ideological Congruence and Electoral Institutions’. Amer- ican Journal of Political Science, 54(1), 90–106. Haller, M. (2008) European Integration as an Elite Process. The Failure of a Dream? New York: Routledge. VISCONTI and PELLEGATA 203 Hobolt, S. B., and Tilley, J. (2016). Fleeing the centre: the rise of challenger parties in the af- termath of the euro crisis. West European Politics, 39(5), 971-991. Hooghe, L., and G. Marks (2009). ‘A Postfunctionalist Theory of European Integration: From Permissive Consensus to Constraining Dissensus.’ British Journal of Political Science 39(1), 1-23. Hooghe, L., and G. Marks (2018). ‘Cleavage theory meets Europe’s crises: Lipset, Rokkan, and the transnational cleavage.’ Journal of European Public Policy, 25(1), 109-135. Huber, J. D. and Powell, G. B. (1994) ‘Congruence Between Citizens and Policymakers in Two Visions of Liberal Democracy’. World Politics 46(3), 291–326. Hutter, S., H. Kriesi and G. Vidal (2018) ‘Old versus New Politics: The Political Spaces in Southern Europe in Times of Crises.’, Party Politics, 24(1), 10-22. ISTAT (2019a) La povertà in Italia. Istituto Nazionale di Statistica. Available at: https://www.istat.it/it/files//2019/06/La-povert%C3%A0-in-Italia-2018.pdf . ISTAT (2019b) Bilancio Demografico Nazionale Anno 2018. Istituto Nazionale di Statistica. Available at: https://www.istat.it/it/files//2019/07/Statistica-report-Bilancio-demogra- fico-2018.pdf . Kriesi, H., Grande, E., Lachat, R., Dolezal, M., Bornschier, S., and Frey, T. (2008). West Euro- pean politics in the age of globalization. Cambridge: Cambridge University Press. Kriesi, H., Grande, E., Dolezal, M., Helbling, M., Höglinger, D., Hutter, S., and Wüest, B. (2012). Political conflict in western Europe. Cambridge: Cambridge University Press. Lupu, N., Selios, L., and Warner, Z. (2017). A New Measure of Congruence: The Earth Mover’s Distance. Political Analysis, 25(1), 95-113. Mattila, M., and Raunio, T. (2006). ‘Cautious voters-supportive parties: Opinion congruence between voters and parties on the EU dimension.’, European Union Politics, 7(4), 427- 449. Mattila, M., and Raunio, T. (2012). ‘Drifting further apart: National parties and their elec- torates on the EU dimension.’ West European Politics, 35(3), 589-606. McEvoy, C. (2012). ‘Unequal Representation in the EU: A Multi-Level Analysis of Voter–Party Congruence in EP Elections.’ Representation, 48(1), 83-99. Natili, M. (2019) The Politics of Minimum Income. Explaining path departure and policy rever- sal in the age of austerity. London: Palgrave Macmillan. Moschella, M. (2017). ‘Italy and the fiscal compact: Why does a country commit to permanent austerity?’ Italian Political Science Review, 47(2), 205–225. Müller, W. C., Jenny, M., and Ecker, A. (2012). The elites–masses gap in European integration, in Best, H., Lengyel, G., and Verzichelli, L. (Eds.). (2012). The Europe of elites: A study into the Europeanness of Europe's political and economic elites. Oxford University Press.167-191. Olmastroni, F., and A. Pellegata (2018) ‘Members apart: a mass-elite comparison of mutual perceptions and support for the European Union in Germany and Italy’, Contemporary Italian Politics 10(1), 56-75. Otjes, S. and Katsadinou, A. (2017) ‘Beyond Kriesiland: EU integration as a super issue after the Eurocrisis’, European Journal of Political Research, 56(2), 301-319. Pedrazzani, A., A. Pellegata and L. Pinto (2018) ‘Economic crisis and lawmaking. The impact of crisis on legislative agenda in Italy’, The Journal of Legislative Studies 24(3), 315-337. Polk, J., Rovny, J., Bakker, R., Edwards, E., Hooghe, L., Jolly, S., and Steenbergen, M. (2017). Explaining the salience of anti-elitism and reducing political corruption for political Representation in hard times 204 parties in Europe with the 2014 Chapel Hill Expert Survey data. Research and Politics, 4(1), 2053168016686915. Sacchi, S. (2015). ‘Conditionality by other means: EU involvement in Italy’s structural re- forms in the sovereign debt crisis.’ Comparative European Politics, 13(1), 77–92. Sanders, D., Bellucci, P., Tóka, G. and Torcal, M. (eds.) 2012, The Europeanization of National Polities? Citizenship and Support in a Post-Enlargement Union, Oxford, Oxford Univer- sity Press. Sanders, D., and Toka, G. (2013). Is anyone listening? Mass and elite opinion cueing in the EU. Electoral Studies, 32(1), 13-25. Sangiovanni, A. (2013). Solidarity in the European Union. Oxford Journal of Legal Studies, 33(2), 213-241. Schadee, H., Segatti, P. and Vezzoni, C. (2019) L’Apocalisse della Democrazia Italiana. Alle Origini di Due Terremoti Elettorali. Bologna: Il Mulino. Serricchio, F. (2018) Gli italiani, l’Europa e la crisi. Torino: Giappichelli. Stjernø, S. (2009). Solidarity in Europe: The history of an idea. Cambridge: Cambridge Uni- versity Press. Thomassen, J. and Schmitt, H. (1999) ‘Issue congruence’, in H. Schmitt and J. Thomassen (Eds.), Political Representation and Legitimacy in the European Union, Oxford: Oxford University Press, 186–208. Wlezien, C., and Soroka, S. N. (2016). Public opinion and public policy. In Oxford research en- cyclopedia of politics. VISCONTI and PELLEGATA 205 Appendix Table A1. Descriptive statistics of the dependent variable's components. Question wording Response categories Thinking about the European Union over the next 10 years, can you indicate whether you are in favour or against the following: Item 1 The introduction of common European bonds (aka Eu- robonds). 1- Strongly against; 2- Somewhat against; 3- Somewhat in favour; 4- Strongly in favour Item 2 The introduction of a common EU fund compensating national governments and local communities for the costs related to immigration from other EU member states. Item 3 The introduction of common EU social insurance schemes (such as healthcare, unemployment or pen- sions) that cover intra-EU migrant workers. EU member states have decided that their economic and social poli- cies should be brought closer together. Please indicate whether you agree or disagree with the following statements: Item 4 The EU should equip itself with a budget large enough to provide substantial financial help to member states facing a sudden rise in unemployment rates. 1- Strongly disagree; 2- Somewhat disagree; 3- Somewhat agree; 4- Strongly agree Item 5 In the case of a very severe financial crisis in a given member state, the EU should make sure that no citizen of that state remains without means of subsistence (food, shelter, essential medicines etc...). Table A2. Number of MPs and voters for each party considered. Party Number of MPs Number of voters SI+MDP 9 21 PD 39 235 M5S 13 305 FI 7 51 LEGA 4 99 Total 72 711 Table A3. Results from factor analyses on elites and masses items. Elites Masses Variables Factor loadings Uniqueness Factor loadings Uniqueness Eurobonds 0.580 0.586 0.371 0.810 Financial help 0.334 0.612 0.575 0.504 Subsistence 0.411 0.597 0.583 0.489 Immigration 0.738 0.442 0.679 0.473 Insurance 0.756 0.400 0.605 0.506 Eigenvalues 1.733 1.636 Note: factor analyses based on polychoric correlations.