Keywords

1 Introduction

Until around the 1992 Treaty of Maastricht, public opinion on European integration was described as “permissive consensus” (Lindberg and Scheingold 1970), implying that citizens in Western Europe tacitly supported their elite’s efforts to push integration further. In the countries of East Central Europe (ECE), citizens had strongly supported EU membership as a “return to Europe” since the 1990s at least until the accession in 2004. Nevertheless, the support was not consensual (Guerra 2013). Nowadays, public opinion on European integration in both ECE and the EU in general is characterised by a “constraining dissensus” (Hooghe and Marks 2008); that is, significant parts of the population evaluate European integration negatively or even oppose it. This public contestation constrains pro-European elite’s leeway in pursuing further steps of integration and encourages parties—both at the fringes and from the mainstream—to politicise European integration and its liberal constitutional foundations.

Anti-EU and illiberal policies of governments may thus be backed by negative public opinion on European integration. With institutional reforms that contest the rule of law or by open non-compliance with EU immigration policy, national governments in ECE—encompassing Czechia, Hungary, Poland and Slovakia—may thus be responsive to domestic Eurosceptic public opinion or actively mould the latter to get domestic support for such policies. The present chapter, accordingly, describes patterns, trends and determinants of public Euroscepticism in ECE to analyse whether public opinion on European integration in these countries is related to the contestation of both the immigration policies and the constitutional principles of the EU by the respective governments. By using longitudinal and cross-sectional, cross-country and country-by-country analyses, the chapter makes two contributions to the research on public opinion on European integration and to regional studies on ECE. Firstly, three dimensions of attitudes towards European integration are delineated and empirically validated: diffuse regime, input- and policy-specific support. Secondly, the chapter analyses the results of ECE in a comparative perspective to investigate whether Euroscepticism in ECE is characterised by regional or by European-wide patterns. It shows only weak indications for a ECE-specific Euroscepticism, i.e. similar patterns among all ECE countries that are distinct from the EU average. In contrast, ECE is less characterised by regional similarities but rather by country differences with regard to both the degree of Euroscepticism and its linkage to the issue of immigration and conceptions of democracy. Additionally, the standard explanatory factors for Euroscepticism need to be amended for and are mediated by country-specific factors. For instance, negative assessments of domestic democracy are positively related to Euroscepticism in Czechia and Slovakia while negatively related to in Hungary and Poland.

In order to develop these findings in more detail, the remainder is structured as follows: the following part introduces the multidimensional concept of Euroscepticism. The third paragraph delineates the hypotheses to explain Euroscepticism and introduces preferences towards immigration and conceptions of democracy as explanatory factors that are directly linked to the ECE governments’ contestation. The final three parts present the data, the results of the descriptive and multivariate analysis and provide some conclusions.

2 Euroscepticism: Concept

Since Euroscepticism implies different meanings, it requires conceptual clarification before being applied in empirical research. It has to be delineated whether Euroscepticism includes fundamental opposition to European integration as well as a critique of its current praxis and performance, or just reluctance or doubts. Furthermore, the term Euroscepticism suggests unidimensionality, i.e. people either support or oppose European integration generally. However, research on public opinion has demonstrated that attitudes towards European integration are multidimensional. They can display ambivalence and ambiguity: instead of consistent pro- or anti-European sentiments, citizens may reject certain facets or objects of European integration but favour others (De Vries 2018, pp. 40–41; Boomgaarden et al. 2011; Stoeckel 2013).

An important starting point for conceptual clarification is provided by Easton’s (1975) concept of political support that distinguishes between authorities, regime, and community as three different objects of evaluation, as well as between specific and diffuse as two different modes of support. According to this perspective, scepticism is the lack or withdrawal of support. While specific support is based on the positive evaluation of the performance of the authorities, institutions or the entire regime, diffuse support is not referring to the actual performance of an object but is derived from the support of the values and norms it represents. This conceptual distinction was refined when applied to attitudes on European integration, but the difference between “utilitarian/specific/output-oriented support [and] affective/diffuse/input-oriented support” (Boomgaarden et al. 2011) still appears in most of these approaches. For instance, Kopecky and Mudde (2002), in their work regarding party stances on European integration, differentiate between support (or lack of) that is affective/value based (Europhiles and Europhobes) or performance based (Optimists and Pessimists). The combination of both dimensions results in a typology in which Euroscepticism encompasses fundamental opposition towards European integration (Europhobe pessimists) and rejection of the actual practice of integration (Europhile pessimists). This dichotomy is resembled in the distinction between “Hard” and “Soft” Euroscepticism (Taggart and Szczerbiak 2004) or in the category of “critical citizens” (Norris 1999), who support the idea of democracy but evaluate the practice of democracy in their own country negatively. Weßels (2007) underlines that each mode of support—value and performance based—can be applied to each object except the community, i.e. for authorities and institutions as well as for the entire process of European integration.

An additional dimension of specific support is policy-related support, which is derived from the dichotomy of input- vs output-legitimacy (Scharpf 1999) or regime- vs policy-evaluations (Dahl 1998). On the one hand, specific support is input-based on the assessment of the responsiveness and inclusiveness of a regime. For democratic systems such as the EU, input-based support is particularly important, since it is normatively based on democratic equality in terms of participation and representation. On the other hand, specific support is based on the output a regime delivers, for instance, the provision of economic goods or social security. Recent studies underline that the demand for a shift of competencies to the EU level is influenced by the comparison between the performance of the domestic government in a particular policy field and the (presumed) performance of the EU institutions in that policy field. The demand for further integration is, accordingly, a specific form of policy support based on the presumed superiority of policy solutions at EU level compared to national solutions—averaged for multiple policies (De Vries 2018, pp. 45ff).

In order to capture the described multidimensionality of attitudes towards European integration, we distinguish (I) diffuse regime support, (II) input-specific regime support and (III) policy-specific regime support, which are further combined in a typology of Euroscepticism (s. below).

3 Explaining Euroscepticism: Hypotheses

Theoretical approaches to explain public support of European integration either focus on economic considerations, identity issues, such as anti-Muslim attitudes (see Chapter 3), benchmarking, cue-taking or, rather seldom, democratic conceptions as explanatory factors (Ejrnæs and Jensen 2019; Hobolt and Vries 2016).

Economic or utilitarian approaches assume that citizens support European integration, should the perceived benefits of membership of a given country in the EU outweigh the costs (Gabel 1998). These approaches distinguish between individuals and collective benefits. Individual benefits of European integration are unequally distributed among the population and conditional on socioeconomic status traits such as, for instance, education, occupation and income. The higher the education and income and the more skilled the occupation, the more advantages individuals can retrieve from trade liberalisation in the course of European integration and the more they support it. In contrast, the less skilled and educated people are, the more their job security is endangered by liberalisation, which makes them less supportive. Sociotropic assessments of the benefits of EU membership for ones’ own country can be based on the attributed responsibility of the EU for the macroeconomic performance of this country or on its status as net-recipient or net-contributor in the EU system of fiscal transfers. Other approaches have focused on the welfare system and assume that European integration is less supported in countries with a developed welfare state, which is perceived as threatened by further liberalisation. In countries with high levels of inequality, European integration is perceived as stimulating further redistribution. The underlying assumption is that individuals’ support is linked to their attitudes to the welfare state (Garry and Tilley 2015).

In the ECE countries with comparatively weak welfare states and high levels of inequality, support for the EU is, accordingly, supposed to be stronger among citizens who support further redistribution. Accordingly, we assume that policy-specific support for European integration decreases, the lower the individual socioeconomic status becomes (H1a) and the more one supports welfare-state redistribution (H1b). Diffuse regime support and input-specific regime support are presumed as less influenced by these factors.

Identity approaches (Hobolt and Vries 2016) emphasise cultural issues such as identity, immigration and religion (Carey 2002). In this view, citizens perceive European integration not primarily in terms of economic liberalisation but as a more general process to overcome national and cultural boundaries. Citizens who share conceptions of distinct and culturally homogeneous nations are, accordingly, more likely to feel threatened by European integration than those who favour cosmopolite conceptions of culturally heterogeneous societies (s. also Chapter 3 in this volume). Accordingly, Euroscepticism is expected to be higher among citizens with nationalist world views compared to those with cosmopolite orientations (H2). The former share homogeneous and exclusive conceptions of national identity that are expressed, inter alia, by sceptic to hostile attitudes towards immigration.

Scholars of public opinion have emphasised that citizens lack knowledge, time, interest and commitment to gather sufficient information on the basis of which they are able to evaluate distant objects such as the EU and its institutions. They rather rely on informational shortcuts or cues that are more easily accessible in their daily experience (Zaller 1992; Anderson 1998). The impact of several cues has been investigated, including the media coverage of the EU, attitudes of domestic elites, stances of established and challenger parties, and the performance of national governments (Hobolt and Vries 2016). These cues are primarily derived from objects in the domestic context and presumed to enfold two opposite influences. The cueing argument was developed in research on voting behaviour in European elections: if people are satisfied with their domestic government, they vote for the parties that form this government (Franklin et al. 1995). Generalising this argument assumes that the less one is satisfied with the national government, the less one supports European integration. In contrast, benchmarking approaches (De Vries 2018; Sánchez-Cuenca 2000) stress that the national contexts provide a benchmark for the evaluation of the EU: if citizens are not satisfied with their national government or the democratic regime in their country, they may perceive a shift of competencies to the supranational level of the EU as an appealing alternative. Vice versa, citizens who are satisfied with their national government are likely to oppose shifts of competencies from the national to the supranational level; that is, they are Eurosceptic concerning the policy-specific dimension of political support. Whether cueing or benchmarking dominates is conditional on the national context (Ejrnæs and Jensen 2019). We therefore assume that satisfaction with the government is linked to both input- and policy-specific support for the EU (H3a), satisfaction with the economic situation is linked to policy-specific support (H3b), while satisfaction with national democracy is linked to input-specific support (H3c). However, whether the domestic evaluation is a cue that is directly transferred to the EU level or serves as an indirect benchmark is conditional upon the national context.

Parties and governments also actively mould public opinion towards European integration (Gabel and Scheve 2007; Steenbergen et al. 2007). Since the consensus among established political elites to support European integration has remained longer than the “permissive consensus” among the population (Vogel and Rodríguez-Teruel 2016), attempts to politicise European integration were pursued primarily at the fringes of the national party systems. Parties and elites on the left mobilise voters against the neoliberal and, since the financial crisis, the austerity policy of the EU, whereas on the right, they mobilise by accusing the EU of undermining the national identity and sovereignty (Hobolt and Vries 2016). If citizens take cues from parties they support, the degree of Euroscepticism is supposed to be higher for citizens who support a Eurosceptic party (H4). The causality may, however, also be reversed: Eurosceptic attitudes may originate endogenously and subsequently determine the party choice. Although we cannot completely rule out this alternative interpretation, two caveats are made. Firstly, if citizens’ Eurosceptic attitudes were endogenous to party stances, the choice for a Eurosceptic party would be significantly influenced by the other explanatory factors discussed in this section. Thus, the absence of multicollinearity would support the cueing interpretation. Secondly, voters’ choice for Eurosceptic parties is in any case not completely influenced by the Eurosceptic stances of these parties. This is more likely for Eurosceptic parties with a broad programmatic package compared to niche parties and in electoral competitions, which are not centred on issues of European integration, i.e. rather in national than European elections.

Input-specific support is linked to the perception of responsiveness in the policy-making process; that is, both private interests and the interests of the own country are regarded in the policy-making process (see Chapter 7). Empirical evidence shows that perceived external efficacy of the EU indeed decreases Euroscepticism (Rohrschneider 2002; McEvoy 2016). We assume that input-specific support is further influenced by the conceptions of democracy. This assumption links the debate on Euroscepticism with the one on democratic backsliding and contestation of EU law by the ECE governments. We distinguish analytically between an illiberal conception of democracy that stresses the primacy of national popular sovereignty and majoritarian rule, and a liberal conception which puts universal civil and minority rights as well as institutional control and division of power in the foreground (Mény and Surel 2002). Citizens may perceive the EU as a threat to illiberal conceptions, since qualified majority voting can result in EU decisions that are contrary to national majorities. Additionally, economic liberalisation is inextricably linked to the protection of individual civil rights such as, for instance, free movement, which constrains national sovereignty further. We therefore assume that citizens with an illiberal conception of democracy are less supportive of European integration on the input-specific dimension (H5).

Combining these assumptions in a longitudinal perspective, diffuse regime support for the EU is presumed to have remained stable in the ECE countries in the period under investigation, but specific support, both input- and policy-related, is volatile and dependent on the domestic context. Since diffuse regime support is driven by value-based evaluations, it should be more stable than specific support. Accordingly, the economic developments of the ECE countries are presumed to contribute in particular to policy-specific support. In particular, the economic difficulties since the financial crisis of 2007/2008 may have changed the perception of economic benefits resulting from EU membership. Less benefits are perceived, if the responsibility for economic decline is attributed to the EU, but also, if the responsibility for successfully coping with economic hardships is attributed to the national governments instead of to the EU (Vogel and Göncz 2018). Additionally, cultural issues may have influenced policy-specific support in the course of the so-called refugee crisis of 2015/2016 and the efforts at EU-level to reform the relocation scheme for immigrants and refugees. In the ECE countries, in which some national governments portrayed the EU as undermining national ethnic homogeneity and thus the foundation of cultural identity, policy-specific support is considered to decrease in the aftermath of 2015. Input-specific support might have decreased too, since governments in ECE countries fuse their accusation of the EU undermining national identity and sovereignty with their conception of an alternative, illiberal view of democracy that challenges the liberal foundations of the EU enshrined in Article 2 TEU.

4 Data Basis: The European Election Voter Studies 2004–2019

The data for this investigation is derived from the four Voter Studies of the European Election Studies (2004, 2009, 2014 and 2019).Footnote 1 Our sample includes all countries in which the survey was conducted at each election to the European Parliament since 2004, thereby including 24 states that were EU members in 2019 (except for Bulgaria, Croatia, Malta and Romania). Each Voter Study was conducted after the respective European election and contains around 1000 randomly selected respondents in each wave in each country. The recommended weighting procedures were applied for post-stratification and to take into account the different number of inhabitants of the countries in the case of reporting EU averages.

4.1 Dependent Variables

Diffuse regime support is measured by the answer to the question “Generally speaking, do you think that [country]’s membership of the European Union is a good thing, a bad thing, or neither?” and input-specific support by the question “All in all again, are you very satisfied, fairly satisfied, not very satisfied or not at all satisfied with the way democracy works in the European Union?” To capture policy-specific support, the following standard item was selected: “Some say European unification should be pushed further. Others say it already has gone too far. What is your opinion?” Since policy evaluations in the multi-level context of the EU are linked to policy allocation, respondents who evaluate the problem-solving capacities of the EU better than that of their national political system are supposed to favour further integration. The membership question was asked in all four waves; satisfaction with democracy was not asked in 2014 and neither was the question on unification in 2004.

4.2 Independent Variables

Individual utilitarian judgments are not directly assessed but measured by the length of formal education and the subjective social class (Table 3 in the Appendix). Self-assessment of class belonging serves as a valid indicator for the access to resources that is otherwise measured by occupation and income (Wright 2000). Sociotropic utilitarian considerations are measured by the preference for welfare-state redistribution. Nationalist and cosmopolite orientations are distinguished by measuring respondents’ attitudes towards immigration. Cueing and benchmarking are assessed by asking for the satisfaction with democracy within a given country, the disapproval of the national government and an assessment of the national economic situation. Party cues are assessed by asking for which party respondents have voted in the most recent national election. By this focus on the national election instead of on the European elections, it is less likely that the electoral choice was influenced by parties’ positions on European integration (see above). The classification of parties as Eurosceptic is based on their belongingness to one of the following party groups in the European Parliament: GUE/NGL, ID and ECR (Table 4 in the appendix). The exception being Fidesz, which was categorised as Eurosceptic despite their affiliation to the EPP. If parties do not belong to any parliamentary group, the classification follows the overview provided by Barbieri (2015). Finally, the measurement of democratic conceptions is constrained by the indicators that are available in the EES. We use the support for the restriction of civil rights to combat crime as an indication of illiberal conceptions. Citizens who approve restrictions are more likely to share illiberal conceptions of democracy, since they provide individual civil rights not universally, but only if this is not seen as an obstacle to the realisation of certain policy goals.

5 Euroscepticism in ECE 2004–2019: Description and Typology

Regime support increased in all ECE countries since their accession in 2004 even though in Hungary it is only marginally higher in 2019 compared to 2004 (Fig. 1). In Poland, Hungary and Slovakia, diffuse support in 2019 is even above the average of the remaining EU member states, while support in Czechia remains remarkably lower. With the exception of Hungary, neither in ECE nor in the remaining EU, diffuse regime support dropped at the peak of the economic crisis in 2009. Instead, it decreased slightly afterwards (except in Hungary and Poland), since the economic crisis has developed into a financial and state budget crisis. Furthermore, neither the conflicts during or in the aftermath of 2015’s massive influx of asylum seekers nor the contestation of the rule of law by constitutional reforms in Hungary and Poland nor the non-compliance of the Visegrád countries with the relocation scheme have been accompanied by a decrease in public diffuse regime support for the EU since 2015. Moreover, diffuse support has even increased, while it decreased in the average of the remaining EU countries.

Fig. 1
figure 1

Diffuse regime support: EU membership is a good thing (Means 2004–2019, EES) Note Generally speaking, do you think that [country]’s membership of the European Union is a good thing (3), a bad thing (1) or neither (2)? (answers are recorded). EU without ECE countries, Bulgaria, Croatia, Malta and Romania

Full size image

In contrast, satisfaction with EU democracy has decreased in almost all observed ECE countries since 2004, except in Poland, which parallels the development of the average in the other member states (Fig. 2). Similar to the level of diffuse support, the input-specific support in Hungary and, especially, in Poland was always above the average of the remaining EU countries including Slovakia and Czechia. Finally, policy-specific support is the most volatile dimension: in all ECE countries and in the average of the other EU countries, it decreased significantly in 2014 with a re-rise in 2019 (Fig. 3). Only in Czechia did policy-specific support not re-rise after the drop from 2009 to 2014.

Fig. 2
figure 2

Input-specific support: Satisfaction with EU democracy (Means 2004/9/19, EES) Note All in all again, are you very satisfied (4), fairly satisfied (3), not very satisfied (2) or not at all satisfied (1) with the way democracy works in the European Union? (answers recoded). EU without ECE countries, Bulgaria, Croatia, Malta and Romania

Full size image
Fig. 3
figure 3

Policy-specific support: Evaluation of European integration (Means 2009–2019, EES) Note Some say European unification should be pushed further. Others say it already has gone too far. What is your opinion? Please indicate your views using a scale from 0 “has already gone too far” to 10 “should be pushed further”. EU without ECE countries, Bulgaria, Croatia, Malta and Romania

Full size image

These different and, accordingly, loosely coupled developments in ECE—increasing diffuse support, decreasing input-specific support and volatile output-specific support—corroborate the multidimensional structure of attitudes towards European integration at the macro-level. To investigate this multidimensionality at the individual level, input-specific support and policy-specific support are combined into three categories of specific support: (1) citizens who are not satisfied with EU democracy and who judge European integration as too far-reaching,Footnote 2 (2) citizens who are both satisfied with EU democracy and with the current status of European integration (or even want it to be pushed further) and (3) those with mixed evaluations of input and policy. In the observed ECE countries in 2019, the relation between those three categories of specific support and diffuse regime support is rather close (Table 1): 69.3% of those who are sceptical towards the regime display no specific support, while 70.7% of the diffuse supporters are simultaneously satisfied with both EU democracy and policy performance. Those who are undecided about the benefits of EU membership judge input and policy overwhelmingly mixed or negative. In the average of the other EU member states (no figure), 65.1% of those without diffuse support show no specific support, which is close to the ECE average of 69.3%. But only 47.6% of those with diffuse regime support display outright specific support. This suggests that in the ECE countries, regime support is more closely coupled with performance assessment than in the remaining member states, which is presumably due to the shorter experience with European integration in ECE.

Based on these observations, a typology of Euroscepticism, i.e. of those who explicitly judge at least one dimension of European integration negatively, is applied to the EES of 2019 (see Table 1). Relying on the categories proposed by Taggart and Szczerbiak (2004) and Weßels (2007), “Hard Eurosceptics”, who neither display diffuse nor specific support, constitute a minority of 6.2% of all ECE citizens. “Soft Eurosceptics”, who lack specific support but do not judge the membership of their country in the EU negatively (but also not positively), are the majority of Eurosceptics constituting a slightly greater share of the ECE population (8.4%). “Critical EU supporters” are those who evaluate EU membership positively but who are critical both about policy and democratic performance of the EU. The few citizens who perceive no benefits from EU membership, but are either satisfied with EU democracy or policy output or both are “Sceptical Instrumentalists”.

Table 1 Types of Euroscepticism in ECE countries (Column- and (Total-) Percentages, EES 2019)
Full size table
Fig. 4
figure 4

Types of Euroscepticism in ECE and EU-20 (Percentages, EES 2009/19) Note EU without ECE countries, Bulgaria, Croatia, Malta and Romania

Full size image

Taking these four types together, Eurosceptics appear as a minority in the ECE countries and in the average of the remaining EU countries (Fig. 4). Nevertheless, in Czechia and Slovakia, as in the EU average, its amount increased in the past ten years. In 2019, it even encompasses a huge minority of 43.5% in Czechia. In contrast, Euroscepticism has decreased in Hungary and Poland in the same period, which rules out that rising Eurosceptic public opinion has fuelled the non-compliance in immigration policy of the national governments in these two countries. Despite these differences, the composition of Eurosceptics differs only marginally: in most of the countries, Euroscepticism is primarily based on a lack of diffuse regime support (“Hard” and “Soft” Eurosceptics) and its increase is primarily due to a growing amount of “Hard” Eurosceptics. The “Critical EU supporters”, who support membership, but who are critical about both the EU’s input and policy performance, constitute the minority of Eurosceptic citizens in most countries.

6 Findings: Multivariate Analysis for 2019

To analyse the determinants of these attitudes to European integration, we use single-level linear regression models for the country-by-country analyses and two-level random intercept models for the cross-country analyses (Table 2). We first investigate the impact of individual utilitarian considerations based on the own social status. The higher the self-estimated social class-affiliation, the higher the diffuse regime and policy-specific support in the average of the EU. Formal education additionally increases input-specific support. In contrast, in the ECE countries, both factors enfold an impact only in Czechia. Here, education increases diffuse regime support and higher social class increases policy-specific support. Thus, in ECE, with the exception of Czechia, all types of EU support are independent from individual utilitarian judgements, which is a remarkable difference to the EU average for which H1a is confirmed.

Table 2 Single- and Multi-level linear regression models explaining support for European Integration (EES 2019, beta values)a
Full size table

Proceeding to sociotropic judgements, H1b is partially confirmed; that is, welfare-state preferences are linked to diffuse regime support in Slovakia but not in the EU average, while they are linked to specific support in the EU and in Czechia, Slovakia and Hungary. There is no link between preferences for redistribution and EU support in Poland. Furthermore, the domestic context mediates not only the degree but also the direction of the impact: the more citizens reject redistribution, the more they support the EU in Czechia and Slovakia, while the support is lowered among Hungarian opponents of redistribution. Following the explanations suggested above, the EU is perceived by Czech and Slovakian citizens—and in the EU average—as an agent of economic liberalisation and is thus valued by those who want to constrain redistribution. In contrast, the Hungarians perceive the EU rather as means for fostering redistribution that makes the EU more valued among those who support redistribution. This mediating effect is presumably due to the differing degree of inequality and welfare-state institutions in the respective countries.

In the EU average, citizens who approve restriction of immigration—used as proxy for nationalistic orientations—are less likely to display both diffuse support and specific support compared to citizens with cosmopolite orientations. In ECE, the impact is limited to specific support and, furthermore, to Hungary and Poland. The more Hungarian and Polish citizens favour restrictions of immigration, the more they are dissatisfied with EU democracy and the more they perceive integration as too far-reaching. The results corroborate H2 in that citizens who share conceptions of culturally homogeneous national identity are more reluctant to support European integration—presumably due to its border-transcending impact. The lack of such a link in Czechia and Slovakia is an exception from the general picture in the EU.

Satisfaction with national democracy, the government and the economy are presumed to influence the attitudes towards the EU either via the cueing or the benchmarking mechanism. Dissatisfaction with democracy decreases diffuse support and specific support in the EU average as well as in Czechia and Slovakia. This underlines the cueing effect of the domestic context. As presumed in H3c, the evaluation of domestic democracy has the strongest impact on satisfaction with EU democracy. It is, however, also relevant for diffuse support for which it is even one of the most influential predictors. The Polish and Hungarian cases, in contrast, support the benchmarking interpretation: dissatisfaction with democracy increases policy-specific support of the EU; that is, the more one is dissatisfied with Hungarian or Polish democracy, the more EU integration is requested to be pushed further. In Hungary, however, dissatisfaction with domestic democracy is also connected positively to dissatisfaction with EU democracy. While EU democracy is judged by the Hungarians by the same measure as domestic democracy, there is no evidence for such a link in Poland. In both countries, further EU integration seems to be perceived to solve democratic deficits at national level and in Hungary at both levels.

Disapproval of the national government fuels diffuse support in Czechia only, accordingly, Czech citizens’ satisfaction with their government goes along with increased Euroscepticism. This negative effect corroborates the benchmarking approach for Czechia, but the effect prevails in no other ECE country. Rather, the opposite cueing effect dominates in the EU average: dissatisfaction with the own national government decreases enthusiasm for further integration.

Slovakia is the only ECE country in which the perception of economic decline goes along with lower EU support in all three dimensions. In Poland, a negative sociotropic assessment is connected to decreased diffuse support, while in the EU average it goes along with diffuse and policy-specific EU support. Sociotropic assessments are not linked to the EU evaluation in Czechia and Hungary.

Voting for a Eurosceptic party in the most recent national election is strongly and consistently linked to lower EU support in all dimensions, which holds true for the EU average and for almost all ECE countries—with Slovakia as a remarkable exception. It should be underlined that the voting decision is controlled for other explanatory factors and shows only low multicollinearity. Thus, the decision to vote for Eurosceptic parties in the last national election is just partially motivated by the utilitarian and benchmarking considerations or by the policy- and politics-related attitudes, which are analysed here to explain Euroscepticism.

Finally, illiberal conceptions of democracy, measured by the approval of civil rights restrictions in favour of reaching a certain policy goal, go along with decreased diffuse regime and input-specific support in the EU average. In contrast, Hungary is the only country in ECE in which illiberal conceptions of democracy are linked to attitudes on European integration: Hungarians who accept restrictions of civil rights are more likely to display lower input-specific support.

The multi-level cross-country analysis shows a significant and positive impact of the domestic context in Hungary and Poland for input-specific support and, for Hungary, for diffuse support. The attitudes towards European integration in Czechia and Slovakia are, in contrast, explained thoroughly by the explanatory factors valid in the EU average. This result corroborates observations of a “Europeanisation of Euroscepticism” in Czechia; that is, Eurosceptic attitudes are decreasingly linked to the domestic historic, social and political characteristics of a country but are increasingly connected to a Eurosceptic political discourse with similar patterns and narratives in all EU member states (Havlík et al. 2017, pp. 163–164). In contrast, the higher levels of support in Hungary and in Poland indicate additional domestic factors that decrease Euroscepticism in these countries vis-á-vis their explicitly Eurosceptic governments.

Furthermore, the country-by-country analysis of the ECE countries has demonstrated that the impact of many general explanatory factors is conditional upon the national context—with the exception of the vote for a Eurosceptic party. In Czechia and Slovakia, utilitarian considerations and cueing effects of domestic democracy are most important. However, in Slovakia, the sociotropic benefits are important, including the preferences for redistribution, while in Czechia only individual utility is important. The two countries differ further as Slovakia is the only ECE country in which the stances of the (two) Eurosceptic parties do not decrease EU support among their voters, and Czechia is the only country in which the government serves as a benchmark instead of as a cue for EU evaluation.

By contrast, in Poland and Hungary, utilitarian considerations are not decisive for attitudes towards European integration—except the preferences for redistribution in Hungary—but preferences for immigration enfold significant impact. Furthermore, in both countries, dissatisfaction with democracy serves as a benchmark instead of as a cue for policy-specific support implying that dissatisfaction with domestic democracy goes along with the demand for further integration, which is in strong contrast to Czechia and Slovakia and to the EU average—even though the picture in Hungary is more complex due to the cueing effect on input-specific support. The peculiarity of Hungary is further underlined by the negative connection between input-specific support and illiberal conceptions of democracy, which is unique in ECE.

7 Conclusion

The aim of this chapter was to describe the structure, development and determinants of attitudes towards European integration in ECE in order to analyse whether and how the contentious politics of the governments in this region vis-à-vis the EU are related to public opinion on European integration. The chapter focused especially on the connection between Euroscepticism and illiberal conceptions of national identity and democracy in the context of standard explanatory factors. Using data from a series of European Election Studies, the chapter distinguished Euroscepticism analytically in diffuse regime, input- and policy-specific support. It demonstrated that public support for European integration in ECE is more closely linked to instrumental performance assessments than in the EU average. This result may be explained by the shorter EU experience of citizens in the ECE countries, which provided less time to decouple diffuse regime support from performance evaluations.

Given this coupling, governments in ECE may back their contentious politics by drawing on negative assessments of the EU in immigration policy and its constitutional regulations. The results indicate, however, that the relation between public opinion and government politics is less straightforward. Firstly, Euroscepticism is not widespread and has even decreased in Hungary and Poland, i.e. in those countries that pursue the fiercest politics of EU policy contestation. It is, however, remarkably high—and has even increased—in Czechia and Slovakia. Secondly, although contestation began to escalate in the course and aftermath of the so-called refugee crisis of 2015, Euroscepticism in ECE shows only a moderate increase in its input-specific dimension since then. Thirdly, the determinants for Euroscepticism reveal rather country-specific than region-specific patterns. Especially, the cultural issues that are most salient in ECE governmental politics—immigration and conceptions of democracy—are only relevant for public Euroscepticism in Hungary and Poland, but not in Czechia and Slovakia, where socioeconomic and utilitarian considerations are more important.

The country-specific domestic context is also relevant for the outcome of comparisons between the performance at the national and the EU level. While dissatisfaction with democracy is generalised to the EU level in Czechia and Slovakia, dissatisfied citizens in Poland and in Hungary are more likely to support further integration. Obviously, the domestic public discourse and its narratives, including the politics of the national governments, are built upon and revolve around country-specific criteria for citizens’ evaluation of European integration. National grievances that serve as cues to increase Euroscepticism in some countries increase support of European integration in others, presumably as a means to overcome them by further integration. Given this intra-regional heterogeneity, the level and the determinants of public Euroscepticism do not provide a thorough explanation for the similarity of the Visegrád governments in their illiberal and anti-EU politics.