EU Funding and Euroskeptic Vote Choice

Abstract

Why do voters in Central and Eastern Europe (CEE) vote for Euroskeptic political parties? Existing explanations of Euroskepticism suggest that those benefiting economically due to the European Union (EU) are less likely to be Euroskeptic. These approaches fail to take into account the social purpose of EU economic transfers. I argue that the minority advancement realized through EU funding drives voters toward Euroskeptic electoral options. I provide evidence of this relationship through two methods: a large-N statistical analysis and a survey experiment. The large-N analysis employs time-series, regional data from ten CEE member states. The survey experiment tests the hypothesis with a nationally representative sample of the Slovak population. Results from both methods corroborate the hypothesized relationship. Importantly, results suggest that reactionary voters may undermine the long-term institutional goals of the EU due to the short-term consequences of EU policies.

Keywords

euroskepticism political parties voting European Union nationalism survey experiment

After the end of state socialism, the countries of Central and Eastern Europe (CEE) marched—albeit at different paces—toward the same goals: first, the North Atlantic Treaty Organization and, second, the European Union (EU).¹ At the time, both CEE political elites and publics considered EU accession a heralded return to Europe. Thus, elites undertook the necessary reforms; obstructionist governments (like that of Vladimír Mečiar in Slovakia) were (eventually) voted out of office; and “permissive consensus” reigned supreme (Hooghe and Marks 2008). However, consensus did not last. Euroskepticism in CEE is on the rise. In recent elections, new Euroskeptic parties, such as SPD in the Czech Republic or EKRE in Estonia, have achieved electoral representation for the first time. Mainstream parties, like Fidesz in Hungary and PiS in Poland, have likewise tempered their support for European integration. The challenge these parties pose to the European project is significant in its own right; moreover, their electoral success threatens minority rights and portends possible democratic backsliding (Börzel and Schimmelfennig 2017). At the same time, the EU has transferred billions of euros to CEE through EU Regional Policy. In this study, I focus on the relationship between EU funding and Euroskeptic voting—what is the effect of EU funding on Euroskeptic vote choice?

Scholars employ three approaches to explain Euro-skepticism: economic cost–benefit, national identity, and cueing (see Hobolt and De Vries 2016). Likewise, a new literature on public support for Eurozone bailouts has emerged (e.g., Bechtel, Hainmueller, and Margalit 2014; Kleider and Stoeckel 2019; Kuhn, Solaz, and van Elsas 2018). I draw on these literatures to suggest that a politicized EU policy contributes to the rise of Euroskepticism in CEE. Specifically, I contend that EU funding for the advancement of minority groups—allocated through EU Regional Policy—drives CEE electorates to vote for Euroskeptic electoral options. I argue that ethnonationalist political parties politicize these funds, thus creating Euroskeptic grievances among the electorate.

While EU funding serves an important economic purpose—supporting agriculture and infrastructure—it also promotes minority integration and multicultural programming. Recent work suggests that non-cosmopolitan individuals in economically disadvantaged or corrupt EU member states opposed the Eurozone bailouts—an economic transfer between EU members (e.g., Bauhr and Charron 2018; Bechtel, Hainmueller, and Margalit 2014; Kleider and Stoeckel 2019). However, little work has addressed how EU Regional Policy affects behavior within EU members. Bustikova (2014) finds that minority advancement mobilizes far-right voters. Similarly, I suggest voters choose Euroskeptic electoral options when they possess sufficient grievances with minority groups. These grievances are more likely to manifest in regions where minorities are politically or economically accommodated. EU Regional Policy provides this accommodation. Domestic actors play an important role—citizens must be “cued” to consider minority advancement. Specifically, I investigate the cues of parties that privilege the rights of titular majorities over the rights of ethnic minorities and support exclusive conceptions of the “nation.” I term these parties ethnonationalist. To bolster their own electoral fortunes, ethnonationalist parties politicize minority transfers. By doing so, ethnonationalists create grievances that drive Euroskeptic vote choice. Thus, regions characterized by high amounts of EU funding and strong ethnonationalist cues are more likely to vote Euroskeptic. In addition to testing the impact of EU transfers on the general Euroskeptic vote, I also test for differential effects across Euroskeptic party type. Results suggest that EU funding drives Euroskeptic vote choice generally, and that ethnonationalist Euroskeptics benefit most. I also show that the effect of EU transfers is moderated by the salience of ethnonationalist cues in a given country. EU funding has a larger impact when the ethnic cue in a given region is stronger.

I make three contributions to the Euroskepticism literature. First, I modify utilitarian explanations of Euro-skepticism. If individuals voted solely according to the economic benefits conferred by the EU, Euroskepticism could not be marketable in CEE. Instead, I combine economic, identity, and cueing explanations to show that the social purpose of EU transfers creates reactionary Euroskepticism. Second, I support my theory through time-series, cross-national data from ten CEE states and through a survey experiment in Slovakia. Combining large-N and experimental methods allows for generalizability, suggests causal effects, and addresses endogeneity. Finally, I build on the EU economic redistribution literature by addressing the political consequences of a different redistributive transfer, EU Regional Policy.

The article proceeds as follows. First, I discuss extant explanations for Euroskepticism. While this work sheds substantial light on the determinants of Euroskeptic public opinion, cross-national comparative studies of Euroskeptic vote choice are far less prevalent. I suggest that economic explanations treat all economic transfers equally, ignoring the backlash minority funding can create. I also suggest that while cueing is important for explaining Euroskepticism, it works best when parties/elites have tangible “material” with which to prime electorates. Next, I present my theory. I provide a brief background on EU funding prior to discussing how projects realized through Regional Policy accommodate minorities. I then discuss how ethnonationalist parties use this accommodation to cue electorates. The combination of minority transfers and elite cues generates the grievances driving Euroskeptic vote choice. Subsequently, I introduce my research design and enumerate results. I employ two empirical tests: a time-series, cross-sectional model and a survey experiment. The last section concludes.

Existing Explanations of Euroskepticism

Economic cost–benefit explanations suggest that individuals who accrue material benefits from the EU are unlikely to be Euroskeptics. This includes individuals that benefit from increased trade or those with higher levels of income and education (Gabel 1998; Tucker, Pacek, and Berinsky 2002). These explanations suggest that economic considerations, whether sociotropic or egocentric, explain Euroskepticism. This approach is relevant for explaining Euroskepticism during accession. At the time, joining the EU had normative significance, and it also promised economic integration and concomitant growth (Schimmelfennig and Sedelmeier 2004, 665). Due to the normative and economic importance of the EU, few political parties held salient Euroskeptic positions. However, the 2010 Eurozone crisis shifted opinion about the purported economic benefits of EU membership (Usherwood and Startin 2013, 2). These doubts continued post crisis. The durability of Euroskepticism in CEE indicates that economic explanations alone cannot account for Euroskepticism in the region. According to the distributional approach, CEE publics should be pro-EU; they benefit immensely from economic transfers. I argue that not all economic transfers are seen positively by CEE electorates.

Euroskepticism may also be a response to the real (or perceived) EU threat to national identity (Hooghe and Marks 2005; McLaren 2006). National sovereignty is threatened through integration, a process that transfers national competencies to the European level. The reforms introduced after the Eurozone crisis were a turning point for deepening European integration (Buti and Carnot 2012). These reforms introduced EU-level supervision of national financial sectors and strict regulations concerning fiscal policy. Individuals with stronger national identity attachments, or those hostile to other cultures, are more likely to oppose this transfer and thus are more likely to be Euroskeptic (e.g., De Vreese and Boomgaarden 2005; McLaren 2006).

It is important to consider the tangible manifestations of a policy—as well as where it originates—before determining its impact. Ethnonationalists politicize the outcomes of projects realized through EU funding, motivating Euroskeptic vote choice. Euroskeptic vote choice in CEE is not a response to EU funding decisions being made at the EU level. In fact, much of the decision-making concerning disbursement is made at the national (or subnational) level. Rather, it is a response to the minority advancement policies realized through EU funding.

Finally, Euroskepticism may be more likely to manifest when publics are cued by Euroskeptic elites. Citizens rely on political elites to provide cues, or informational shortcuts, about the EU (e.g., Anderson 1998; De Vries and Edwards 2009; Hobolt 2009, 52–56; Stoeckel and Kuhn 2018). European publics cannot possess complete information about the EU. The role of political parties and elites in driving Euroskepticism is critical. I build upon cueing explanations by suggesting that the extent to which EU transfers drive Euroskepticism is contingent on the salience of ethnonationalist cues in a given country. When EU transfers to minority outgroups are politicized, Euroskepticism is more likely to develop.

The Eurozone crisis spawned a body of research about the determinants of public support for economic redistribution across the EU (e.g., Baute, Abts, and Meuleman 2019). Findings conclude that altruistic cosmopolitans in economically well-to-do countries are generally supportive of EU redistribution (Daniele and Geys 2015; Kuhn and Stoeckel 2014). Additionally, perceptions of corruption and party cues play an important role in evaluations (Bauhr and Charron 2018; Stoeckel and Kuhn 2018). Findings concerning income and left-right positioning are more mixed (e.g., Bechtel, Hainmueller, and Margalit 2014 vs. Kleider and Stoeckel 2019).

These findings provide predictions about which individuals are most likely to oppose EU transfers to minority groups. While the recipients of the economic transfer—minorities within a country versus the populace of another member state—differ, grievance formation can function similarly. For example, cosmopolitanism and ideology may moderate individual perceptions of, and subsequent responses to, within-country EU social transfers.

Conceptualizing Euroskeptic Vote Choice

The Euroskepticism literature generally focuses on public opinion rather than vote choice. Studies of Euroskeptic vote choice require the construction of new data sets and require the scholar to code whether parties are Euroskeptic. This is a costly and subjective endeavor.

However, the proliferation of cross-national data sets on party issue positioning can increase confidence in the validity of Euroskeptic party coding. Nonetheless, there are clear trade-offs between using public opinion versus electoral returns. Surveys tell us how publics feel about the EU at one singular point. Unfortunately, they do not tell us whether these sentiments translate into electoral actions. Electoral returns tell us when Euroskeptic parties become stronger, but it is more difficult to trace the reason(s) behind their success—after all, a voter may choose a Euroskeptic party for reasons other than its EU stance. Likewise, individuals can hold Euroskeptic positions and not vote for Euroskeptic parties (De Vries 2018).

I use Euroskeptic vote choice as the dependent variable for two reasons. First, while Euroskeptic public opinion is a precursor to Euroskeptic vote choice, electoral returns are linked to policy. The electoral success of Euroskeptic parties, independent of voter motivations, leads to Euroskeptics in government. Euroskeptic governments may lead to Euroskeptic policy outputs. Second, the ability to disaggregate Euroskeptic vote choice into different categories is advantageous. By dividing Euroskeptic vote choice into different categories, we can identify what factors drive different types of Euroskepticism. This is more difficult with the survey questions generally used in public opinion research, which do not always neatly correspond to the different types of Euroskepticism.

To identify the potentially differential impact of EU transfers on different types of Euroskeptic parties, I rely on Taggart and Szczerbiak’s (2002) division between soft and hard Euroskeptic parties. While scholars have put forth alternatives to this formulation—for example, reject versus skeptic (Kopecky and Mudde 2002) or diffuse versus specific (Gabel 1998)—the division between qualified opposition and steadfast rejection is appropriate for representing the varieties of Euroskepticism in CEE party systems.

While Euroskepticism continues to be prevalent on the extremes of the political spectrum (De Vries and Edwards 2009, 6), mainstream parties have increasingly adopted Euroskeptic issue positions. Euroskeptic extremist parties are frequently far-right ethnonationalist movements, though they also take the form of anti-globalist communist successor parties. Examples of far-right Euroskeptic parties include ATAKA in Bulgaria and Jobbik in Hungary.

Perhaps the most successful Euroskeptic party on the far-left is KSČM, the Czech communist successor party. Mainstream Euroskeptic parties are generally right-leaning “Eurorealists.” Examples include Law and Justice (PiS) in Poland and Freedom and Solidarity (SaS) in Slovakia. Figure 1 shows the average (across ten CEE countries) vote totals of Euroskeptic parties from 2004 to 2017. The vote share of Euroskeptic parties has more than doubled, from under 9 percent to over 20 percent.

Figure 1.

Euroskeptic vote choice in Central and Eastern Europe (2004–2017; 4/5 year averages).

EU Funding and Euroskeptic Vote Choice

I argue that the accommodation of marginalized groups—minorities, migrants, and refugees—realized through EU Regional Policy, drives Euroskeptic vote choice in CEE. The minority advancement aims and outcomes of EU transfers create opportunities for ethnonationalist political parties to politicize these funds. This politicization creates grievances among the majority which translate to Euroskeptic vote choice.

EU Regional Policy in CEE

Funding is the central policy mechanism of EU Regional Policy. Regional Policy’s goal is to reduce economic disparity between EU regions. Through five EU Structural and Investment Funds, the EU transfers over one-third of its annual budget to member-state regions.² In 2017, this was over €55 billion. Funding levels are determined by the “Berlin” formula. The formula allocates funding on the basis of three economic classifications. A region is classified as less developed (LDR), transition, or more developed (MDR) primarily on the basis of regional gross domestic product (GDP) per capita. In the 2014-2020 budgeting period, fifty-two of the fifty-eight regions in CEE are LDRs.³ In comparison, only eighteen (out of 217) West European regions are LDRs.

As fifty-two of the fifty-eight CEE regions are LDRs, the majority of Regional Policy funding is allocated to CEE. Poland is the largest net recipient. Hungary is second. The majority of EU funding goes toward economic purposes—building roads and schools or subsidizing agriculture. However, EU funding has an enumerated social purpose as well. In the 2014–2020 budget, social inclusion exceeded 10 percent of total EU funding (author calculations, European Commission 2018d). Social inclusion funding is disbursed to programs that support five groups: migrants, refugees and asylum seekers, ethnic minorities, religious minorities, and Roma (Glytsos and Viela 2010, 27). Social inclusion projects are funded through the EU Fund for Regional Development, The EU Social Fund, and the EU Cohesion Fund. Funding is allocated on a project basis. Projects are generally proposed by governments (local, regional, and national) in partnership with civil society organizations.

Social inclusion projects fall into three broad categories: culture (e.g., opening a community center); inclusion (e.g., language training for migrants); and infrastructure development (e.g., building water canalization in Roma communities). Specific examples of projects noticeable to voters include the following: the building of the World Roma Cultural Center in the Czech Republic, the funding of The Migration Information Center in Slovakia or the Eurocities Project in Latvia and Poland, and funding for The National Minorities’ Cultural and Integration Centers in Lithuania or the “Education Integration for Ethnic Minorities” project in Bulgaria.

How EU Funding Motivates Euroskeptic Vote Choice

I argue that the social purpose of EU funding motivates Euroskeptic vote choice by promoting the advancement of minorities, migrants, and other marginalized populations. This advancement is politicized by ethnonationalist political parties. Euroskeptic voting is a reactionary decision made by CEE electorates to a certain, tangible subset of EU policies—those that promote the advancement of nonmajority social groups. The advancement of such groups creates two types of grievances among CEE electorates: economic and cultural (Lubbers and Scheepers 2002; Rydgren 2008).

Economic grievance explanations are generally predicated on realist theories of economic competition. When economic conditions are unfavorable, majorities may blame minorities and also believe that they are in direct economic competition with them. Thus, if either sociotropic or egocentric economic conditions are unfavorable, majorities may develop grievances with minorities. Cultural grievances refer to the sentiments that develop when an individual considers their culture or national way of life to be threatened (Ivarsflaten 2008; Lubbers and Scheepers 2002; Rydgren 2008). Grievances do not automatically manifest because minorities exist. Rather, grievances manifest when minorities advance their position relative to the majority (Bustikova 2014). EU funding provides such opportunities to minority groups. This opportunity is then politicized by ethnonationalists. This politicization translates to Euroskeptic grievances, and thus to support for Euroskeptic parties. As a result, I expect regions that receive higher levels of EU social funding to vote more Euroskeptic.

The Public’s Awareness

An underlying assumption in my argument is that individuals are aware of these minority transfers. To ensure the validity of this assumption, I employ a three-prong approach. First, I provide evidence from Eurobarometer surveys. Second, I contend that the public’s awareness is a function of four factors: the amount of funding, the large number of projects, media attention, and the politicized nature of ethnicity in the region. Third, and most importantly, I suggest that ethnonationalist parties politicize these funds.

Eurobarometer surveys conducted in 2017 and 2019 (Flash 452 and Flash 480) asked EU publics about their awareness of EU Regional Policy. In 2017, 61 percent of the CEE public was aware of Regional Policy; in 2019, it was 64 percent. The corresponding numbers in Western Europe were 30 and 32 percent. These differences are statistically significant (p < .001). The surveys also asked whether it is important that these funds go to migrants and refugees. In both 2017 and 2019, 72 percent of CEE respondents said it was not important. In Western Europe, the corresponding numbers were 39 and 38 percent. Again, these differences are statistically significant (p < .001). Admittedly, the survey does not measure negative evaluations and it only asks about one outgroup. Nonetheless, it is clear that Regional Policy is salient and that social transfers are less of a priority in CEE.

CEE publics are also aware of EU-funded minority advancement projects due to the high monetary value of social transfers—about €64 billion between 2014 and 2020 (author calculations, European Commission 2018d). These transfers fund thousands of projects. I draw upon data provided by the Czech Government (DotaceEU.cz) and the Slovak Government (Úrad Vlády Slovenskej Republiky). Both sources provide data on all EU-funded projects realized or ongoing between 2014 and 2024. In the Czech Republic, 10,348 of 32,904 projects are dedicated to social inclusion (31%). In Slovakia, the EU funds 601 social inclusion projects (12% of all EU projects).⁴

Additionally, EU-realized projects are highly visible (accompanied by EU billboards, shown in Figure 2). Per their grant conditions, the EU requires that projects identify their funding source and feature the EU Regional Policy logos. This branding means that the EU can be linked to projects promoting minority advancement.

Figure 2.

European Union branding: Prague, Czech Republic.

Social transfers are covered by the news media. To demonstrate the extent of media attention, I trace the coverage of two different social transfers in the Czech Republic and Slovakia. I provide an overview of this coverage here and list all the news stories and their sources in the Supplemental Information. In the Czech Republic, the creation of a new EU Asylum and Migration Fund was announced in June 2019. At least eight major news sources covered the creation of this fund, including the largest online daily (idnes.cz), the second largest tabloid (Blesk), the Europe-focused online news (euractiv.cz), and two pro-Russian sites (cz.sputniknews.com and parlamentnilisty.cz)

The creation of a new fund by the EU is a major story. Less drastic social transfers also receive news coverage. In April 2019, the Slovak Ministry of Interior announced the financing of Roma community centers through €6 million of EU funds. At least six media sources varying in political ideology and readership covered this story. The largest daily tabloid, Nový Čas covered it, as did the Russian mouthpiece, Hlavné Spravy. The story also received coverage in the left-leaning, third largest newspaper, SME. These examples show that EU transfers receive coverage from a diversity of news sources and thus reach a large audience.

Finally, ethnicity is historically salient in the region (Kymlicka and Opalski 2002). While state socialism muted ethnic differences, democratization reignited historically derived ethnic tensions. Some of these tensions resulted in relatively benign outcomes, like the dissolution of Czechoslovakia. Others, like the Yugoslav Wars, had tragic consequences. This history did not go unnoticed by the EU. Minority incorporation was a central criterion for accession. Specifically, potential member states had to ensure that they afforded national minorities protections, including official recognition (Hughes and Sasse 2003). The importance of ethnicity, combined with the extent and scope of funding, renders minority transfers a salient issue among CEE electorates.

Most importantly, I suggest that ethnonationalist political parties are responsible for politicizing EU funding. Ethnonationalist parties rely on exclusive appeals to the majority in-group to garner electoral support. By defining the titular majority as a “deserving” group, and minorities as otherwise “not deserving,” these parties reinforce rigid group boundaries. EU funding—and its implications for minorities—serves as politicizeable fodder for ethnonationalists. By suggesting that the EU supports “undeserving” minorities, ethnonationalists mobilize the grievances that lead to their electoral success.

Ethnonationalists politicize Regional Policy as an example of EU “control” or liberal agenda-setting. Primary frames include references to loss of sovereignty, misplaced priorities, and challenges to “traditional” culture. These frames are accompanied by language that contrasts “deserving” ingroups with “undeserving” outgroups. Appeals also focus on how this transfer is at odds with the national interest. The politicization of EU transfers makes up a subset of Euroskeptic rhetoric. Other themes include the supremacy of EU law and the loss of national competencies. Party manifestos address the topic in a limited and less-outgroup-oriented manner. First, parties use rhetoric about how EU funds are not essential: “In fact, 80% of EU funds are made by the Slovak Republic’s contributions to the European Union” (L’SNS 2016 manifesto). Second, parties suggest they should have more control over EU finances: “Law and Justice will strive to verify the EU budget policy, taking into account the principle of even distribution of costs and equal development opportunities” (PiS 2011 manifesto).⁵ Examples of politicization span project types and the targeted out-group in multiple countries:

“For example, in Slovakia, there are calls for projects financed by European funds, European Structural Funds, that are designated only for communities which have a certain proportion of Roma, or Gypsy, populations. I do not understand how in the 21^st century, funds are allocated to some communities only on the basis of what skin color the residences of the community have.” (Milan Uhrik, L’SNS, Slovakia)

“This means we have succeeded in advancing the interests of those advocates and grantees who, for 30 years, are only spending money on European programs to integrate the Gypsies.” (VRMO, Bulgaria)

“The Czech NGO named Organization for Helping Refugees (Organizace pro pomoc uprchlikum), which supports the importation of migrants to the Czech Republic is financed by the European Union . . . instead of giving money to Czech citizens, they spend millions on migrants, whom no one invited.” (Tomio Okamura, SPD, Czech Republic)

“[The EU] intends to spend the money saved on integrating migrants, which is unacceptable for Hungary.” (Fidesz, Hungary)

“The European Union is demanding the allocation of EU funds, valued at millions of euros [for African integration and culture] . . . this means that Slovak citizens will have to use their taxes to pay for African inhabitants in Europe and in Slovakia.” (Martin Belusky, L’SNS, Slovakia)⁶

Previous work has shown that party positions and statements are an important moderator in how electorates perceive the EU (De Vries and Edwards 2009). Likewise, I expect the effect of EU funding to be contingent on the strength of ethnonationalist appeals within a given party system. When minority issues are politicized, electorates are more likely to be aware of EU social transfers. The effect of EU funding in countries with strong ethnonationalist parties should be stronger than in countries with relatively weak ethnonationalists.

Euroskeptic Electoral Options

My theory suggests that individuals from the majority react negatively to EU transfers for minority groups. However, CEE electorates have a variety of Euroskeptic parties from which to choose. To test whether EU funding has differential effects across Euroskeptic party type, I operationalize Euroskeptic vote choice in five ways. First, I test whether EU funding affects Euroskeptic vote choice in the aggregate. I expect EU funding to exhibit a positive relationship with aggregate Euroskeptic vote choice due to the grievances created by minority advancement. Next, I define two pairs of Euroskeptic subsets: soft versus hard and ethnonationalist versus non-ethnonationalist (mainstream).

I predict that soft Euroskeptics will benefit relative to hard Euroskeptics. While opposition has strengthened in recent years, the EU remains relatively popular in CEE. In addition, voters may eschew voting for hard Euroskeptics for strategic reasons. A strategic voter may, for example, hesitate to cast a vote for a radical party if that party is unlikely to surpass the threshold of inclusion (Cox 1997). A voter may instead vote for a soft Euroskeptic alternative. Likewise, more established parties may collude to exclude extremist challengers from accessing resources and from participating in government (Katz and Mair 1995, 16–18). I also suggest that parties with anti-minority platforms (ethnonationalists) should benefit more than other Euroskeptic parties (mainstream). If individuals vote for Euroskeptic parties due to grievances created by minority advancement, it is likely that their ballots go to parties seeking to limit the advancement of minorities. For this reason, EU funding may have a stronger effect on the electoral success of ethnonationalist parties. I provide a complete listing of all parties in each category in the Supplemental Information and operationalize the categories in the following section.

From the preceding discussion, I enumerate the following hypotheses:

Hypothesis 1 (H1): EU funding has a positive relationship with aggregate Euroskeptic vote choice

Hypothesis 2 (H2): EU funding has a positive relationship with soft Euroskeptic vote choice

Hypothesis 3 (H3): EU funding has a positive relationship with ethnonationalist Euroskeptic vote choice

Hypothesis 4 (H4): The effect of EU funding on Euroskeptic vote choice is moderated by the strength of ethnonationalist cues

Research Design: Cross-National Data

In order to test the hypotheses derived from my theory, I conduct a time-series, cross-sectional analysis of fifty-three regions, across forty-six elections in ten CEE states between 2001 and 2017. I use a regional unit of analysis. Specifically, I use the EU NUTS (Nomenclature of Territorial Units for Statistics) 2 regional classifications. NUTS regions are classified at three levels in the order of descending size: NUTS 1, 2, and 3. NUTS 2 regions are defined as “macroregions,” ranging from a population of 800,000 to 3 million. In the three Baltic States, the NUTS 2 level corresponds to the entirety of the state. The number of NUTS 2 regions in a given CEE state ranges from one (the Baltics) to sixteen (Poland).⁷ EU funding is disbursed according to indicators measured at the NUTS 2 level.

Operationalizing Euroskeptic Vote Choice

I use the Chapel Hill Expert Survey (CHES) data set to identify Euroskeptic political parties (Polk et al. 2017). The position variable in the CHES asks country experts to identify the “overall orientation of the party leadership towards European integration.” Experts code the parties along a 7-point ordinal scale, ranging from 1 (strongly opposed) to 7 (strongly in favor). A value of 4 denotes a neutral position. I coded a party as Euroskeptic if the average value assigned by all country experts was below 4 (neutral). Only parties that received greater than 1.5 percent of the national vote were included. In addition, parties competing in the 2017 Bulgarian and 2016 Romanian election have not been coded. Parties in these elections were included in previous iterations of the CHES. I rely on these previous codings.

My theory also suggests that the effect of EU funding on Euroskeptic vote choice may be contingent on party subtype. First, I differentiate between hard and soft Euroskeptics. For each election-year I coded Euroskeptic parties as hard if the mean position score assigned by country experts was less than 2.⁸ I consider a Euroskeptic party to be soft if they receive a mean position score greater than or equal to 2 but less than 4.

Second, I identify ethnonationalists through CHES’s ethnic minorities variable. Country experts are asked to code parties along a 0 to 10 ordinal scale, based on the party’s position toward ethnic minorities. A 0 denotes a position of strong support for ethnic minority rights, while a 10 denotes strong opposition. Ethnonationalist parties are coded as such if their mean expert score is greater than 7.⁹ All parties with a mean score of 7 or below are coded as mainstream. Unfortunately, CHES did not code the ethnic minorities variable for the first set of elections in the sample. Thus, these observations are dropped when analyzing the ethnonationalist subsample. Table 1 enumerates the average vote share of all five Euroskeptic categories and their inclusion criteria.

Table 1.

Varieties of Euroskeptic Vote Choice in CEE between 2001 and 2017.

Type	Inclusion criterion	Average vote share in sample (%)
Aggregate	position < 4	23.21
Soft	2 ≥ position < 4	19.93
Hard	position < 2	3.28
Mainstream	position < 4; ethnic minorities ≤ 7	6.08
Ethnonationalist	position < 4; ethnic minorities > 7	18.29

CEE = Central and Eastern Europe.

For each NUTS 2 region and a given election, I sum the percentage of votes received by each Euroskeptic party category. Consider the 2017 Czech election. Three Euroskeptic parties—ODS, KSČM, and SPD—won 32 percent of the vote in the NUTS 2 region of Central Bohemia. This is the aggregate Euroskeptic variable. SPD and KSČM are hard Euroskeptic parties, while ODS is a soft Euroskeptic party. Together, SPD and KSČM won around 16 percent of the vote. This is the hard Euroskeptic dependent variable for Central Bohemia. The remaining 16 percent won by ODS is the soft Euroskeptic dependent variable. In this case, the ethnonationalist versus mainstream variables mirror the hard versus soft variables—SPD and KSČM are ethnonationalist, while ODS is mainstream.

EU Funding, Ethnonationalist Cues, and Relevant Controls

The primary independent variable of interest is EU social inclusion funding. Using data on Regional Policy provided by the European Commission (2018c), I calculated the total allocation of EU funds for each CEE NUTS 2 region from 2002 to 2017. Allocations from the European Social Fund, the Cohesion Fund, and the EU Fund for Regional Development were included. These are the primary sources of social inclusion funding in CEE. Due to their unrelated purposes, the European Agricultural Fund for Rural Development, and the European Maritime and Fisheries Fund were excluded. EU Funding is calculated as the amount of euros (measured in the hundreds of millions) received by a NUTS 2 region in the year prior to an election (t − 1).

CEE regions started receiving EU funding prior to accession. These allocations allow me to include elections prior to EU accession in the statistical models. The allocations prior to accession are minimal, averaging €3.9 million annually per region. Post accession, the average allocation is €320 million. The largest recipient in the post-accession period was Mazowieckie Voivodeship in 2014 (€2.1 billion), and the lowest was Prague in 2005 (€9.7 million).

To operationalize the ethnonationalist cue in a given country, I rely on CHES’s ethnic minorities variable. Following De Vries and Edwards (2009), I operationalize the ethnonationalist cue as the average Euroskeptic party position on ethnic minorities. Unlike De Vries and Edwards, I am not solely interested in the effect of extremist cues, and thus I include all Euroskeptic parties (weighted by vote share) in the measure. The values of the ethnonationalist cue can range from 0 (when no Euroskeptic parties win more than 1.5% of the vote) to 10 (when all Euroskeptic parties have the maximum value).¹⁰

I include GDP per capita, the percent change in GDP per capita, unemployment (%), the percent change in unemployment, the percentage of minorities, the number of years since accession, and a post-2010 dummy (to signify the Eurozone crisis) as control variables. The economic variables directly affect the amount of EU funding a region receives and may also affect Euroskeptic vote choice through sociotropic evaluations of the economy. I include the percentage of minority groups to control for the possible direct effects of minority–majority interaction. Furthermore, if there are more minorities in a given region, said region could receive more funding. I include a measure of how many years a state has been in the EU. Prior to accession, the variable takes on a value of 0, and increases by 1 for each year of EU membership. This variable is meant to capture the effects of EU “fatigue,” a possible determinant of Euroskeptic vote choice. The number of years since accession may also affect a region’s funding levels. Finally, I include a dummy variable measuring the onset of the Eurozone crisis. The crisis had a profound effect on both Euroskeptic public opinion and Euroskeptic electoral success (Usherwood and Startin 2013). Likewise, it altered the extent of EU funding allocated to EU regions.

EU funding, GDP per capita, and unemployment are all measured at t − 1. Change in GDP per capita and change in unemployment measure the extent to which either changed between the year prior to the election and the election year (e.g., t − (t − 1)). Percent minority is measured using census data, and thus the time of measurement varies. The crisis variable takes on a value of 0 prior to 2010 and a value of 1 after 2010. Per the results of a Hausman test, I estimate the models using fixed-effects ordinary least squares (OLS) estimation. It is important to note that the dependent variable is censored below values of 1.5 percent. I report random-effects censored tobit regressions in the Supplemental Information. Results are consistent between the two estimations. All models include a lagged dependent variable to address autocorrelation. Coefficients for the fixed-effects estimations are reported with robust standard errors clustered by region.

The included controls help address a possible alternate explanation: sociotropic economic conditions or large minority populations may create grievances that mobilize Euroskeptic vote choice and also motivate the disbursement of EU funds. In other words, these regions may be disadvantaged more generally, motivating both the independent and dependent variables. Egocentric grievances (cultural or economic) may also affect Euroskepticism and EU transfers. To determine whether egocentric grievances confound the relationship between EU funding and Euroskeptic vote choice, I use the Eurobarometer survey. Specifically, I code individual evaluations of household financial status and whether an individual associates the EU with a loss of cultural identity. These variables control for personal economic and cultural grievances. Unfortunately, Eurobarometer did not systematically collect these indicators in CEE until 2009, shrinking the sample size by 40 percent. Thus, I report these estimations in the Supplemental Information. Results remain robust to the inclusion of these variables.

Results: Large-N Analysis

I present the results of the statistical analysis in Tables 2 to 4. Table 2 presents the fixed-effect estimations for aggregate Euroskeptic vote choice. Table 3 presents soft versus hard Euroskeptic vote choice. Table 4 presents mainstream versus ethnonationalist Euroskeptic vote choice. For aggregate vote choice, I present three-model specifications: a bivariate skeleton (Lenz and Sahn 2018), a model with all controls, and an interaction model. For the remaining models, due to space constraints, I present two specifications: control and interaction.

Table 2.

Aggregate Euroskeptic Vote Choice.

	Bivariate	Controls	Interaction
EU funding	0.013 (0.005)**	0.021 (0.004)***	−0.018 (0.023)
GDP per capita		−0.019(0.016)	−0.018(0.015)
% Change GDP per capita		0.612(0.269)*	0.531(0.264)*
Unemployment		0.812(0.371)*	0.961(0.408)*
% Change Unemployment		−0.923(0.584)	−0.882(0.617)
% Minority		1.934(0.766)*	1.522(0.758)*
Years since accession		0.016(0.007)*	0.016(0.007)*
Post 2010		0.015(0.043)	0.120(0.045)
EN cue		0.011(0.007)	−0.006(0.014)
EN cue × EU funding			0.005(0.003)
Region fixed effects	✓	✓	✓
Lagged dependent variable	✓	✓	✓
Observations	200	192	192
Adjusted R²	.056	.388	.411
F statistic	5.65 (df = 2; 52)	6.89 (df = 10; 52)	15.67 (df = 11; 52)

Fixed-effects regression with lagged dependent variable (not shown); robust standard errors clustered by region in parentheses. EU = European Union; GDP = gross domestic product; EN = ethnonationalist.

p ≤ .05, **p ≤ .01, ***p ≤ .001

Table 3.

Soft versus Hard Euroskeptic Vote Choice.

	Soft-control	Soft-interaction	Hard-control	Hard-interaction
EU funding	0.022(0.004)***	−0.021(0.019)	−0.004(0.002)	0.003(0.006)
GDP per capita	−0.018(0.016)	−0.017(0.014)	−0.002(0.001)	−0.002(0.001)
% Change GDP per capita	0.356(0.210)	0.259(0.213)	0.309(0.099)**	0.326(0.098)**
Unemployment	0.730(0.309)*	0.878(0.383)*	0.241(0.139)	0.222(0.137)
% Change Unemployment	−0.331(0.516)	−0.323(0.561)	−0.165(0.279)	−0.187(0.278)
% Minority	1.180(0.594)	0.704(0.564)	0.828(0.336)*	0.920(0.372)*
Years since accession	0.010(0.007)	0.010(0.006)	0.006(0.003)*	0.006(0.003)*
Post 2010	−0.027(0.033)	−0.034(0.035)	0.058 (0.020)**	0.058 (0.019)**
EN cue	0.003(0.007)	−0.017(0.012)	0.106(0.003)**	0.014(0.005)**
EN cue × EU funding		0.006(0.002)*		−0.001(0.000)
Region fixed effects	✓	✓	✓	✓
Lagged dependent variable	✓	✓	✓	✓
Observations	192	192	192	192
Adjusted R²	.321	.365	.447	.449
F statistic	8.52 (df = 10; 52)	14.99 (df = 11; 52)	9.70 (df = 10; 52)	9.01 (df = 11; 52)

p ≤ .05, **p ≤ .01, ***p ≤ .001

Table 4.

Mainstream versus Ethnonationalist Euroskeptic Vote Choice.

	MS-control	MS-interaction	EN-control	EN-interaction
EU funding	−0.000(0.002)	−0.018(0.015)	0.029(0.006)***	−0.000(0.013)
GDP per capita	0.031(0.022)	0.030(0.021)	−0.042(0.016)*	−0.047(0.016)**
% Change GDP per capita	0.139(0.099)	0.164(0.105)	−0.279(0.220)	−0.262(0.206)
Unemployment	0.762(0.280)**	0.702(0.253)**	−1.606(0.490)**	−1.803(0.467)***
% Change Unemployment	−0.403(0.275)	−0.453(0.240)	−4.350(1.027)***	−4.637(1.005)***
% Minority	0.617(0.329)	0.448(0.245)	1.365(0.746)	0.935(0.756)
Years since accession	−0.013(0.004)*	0.013(0.004)**	0.021(0.007)**	0.019(0.006)**
Post 2010	0.011(0.015)	0.012(0.016)	−0.037(0.037)	−0.022(0.040)
EN cue	−0.001(0.005)	−0.011(0.006)	0.019(0.005) **	−0.000(0.008)
EN cue × EU funding		0.003(0.002)		0.005(0.001)**
Region fixed effects	✓	✓	✓	✓
Lagged dependent variable	✓	✓	✓	✓
Observations	150	150	150	150
Adjusted R²	.403	.432	.511	.532
F statistic	16.64 (df = 10; 52)	17.33 (df = 11; 52)	12.88 (df = 10; 52)	10.66 (df = 11; 52)

Fixed-effects regression with lagged dependent variable (not shown); robust standard errors clustered by region in parentheses. MS = mainstream; EU = European Union; GDP = gross domestic product; EN = ethnonationalist.

p ≤ .05, **p ≤ .01, ***p ≤ .001

Taken together, the models provide strong evidence for a positive relationship between EU funding and Euroskeptic vote choice. The substantive effect of EU funding is largest for ethnonationalist parties. Finally, the relationship between EU funding and Euroskeptic vote choice is conditional on the ethnic cue in a given country. The stronger the ethnic cue, the larger the impact of EU funding on Euroskeptic vote choice.

EU funding has a statistically significant and positive relationship with aggregate Euroskeptic vote choice. Although the coefficient on EU funding is nonsignificant in the interaction model, I follow Brambor et al. (2006, 74), and graph the marginal effect. The first panel in Figure 3 shows that the effect of EU funding becomes statistically significant and positive at an approximate ethnic cue of 6. Substantively, the effect ranges from 1 to 4 percent per €100 million. Since 2010, the average CEE region receives €420 million in funding. This translates to a 4 to 16 percent increase in Euroskeptic vote. These findings support H1. H4 is also supported in the case of aggregate Euroskeptic vote choice. EU funding has a statistically nonsignificant effect at low levels of ethnic cue. The effect becomes statistically significant and larger in magnitude as ethnic cue increases.

Figure 3.

Marginal effect of European Union funding on aggregate, soft, and ethnonationalist Euroskeptic vote choice at different levels of ethnic cue.

The results provide evidence for H2. EU funding has a statistically significant and substantive effect on soft Euroskeptic vote choice. In the model with all controls (Table 3), an additional €100 million in funding translates to a 2.2 percent increase in soft Euroskeptic vote choice. I again graph the marginal effect for the interaction model. Figure 3, panel 2, confirms that the effect of EU funding varies as a function of the ethnic cue in a given country. Results parallel those of aggregate Euroskeptic vote choice and support H4. EU funding does not have a significant effect on hard Euroskeptic vote choice.

The results also provide evidence for H3. EU funding has a statistically significant effect on ethnonationalist Euroskeptic vote choice. In fact, the coefficient value is the largest across the estimated models. Substantively, an additional €100 million in funding translates to nearly a 3 percent increase in ethnonationalist vote choice. Plotting the interaction term confirms H4 (Figure 3, panel 3). At the highest values of ethnic cue, EU funding has a 4 percent effect on ethnonationalist vote choice. EU funding does not have a statistically significant effect on mainstream Euroskeptic vote choice.

Survey Experiment in Slovakia

To test the individual-level causal mechanism, I employ a survey experiment on a nationally representative sample of the Slovak population. Slovakia is one of the least Euroskeptic countries in CEE and thus makes for a stringent test of my hypotheses. I designed the experiment to examine the effect of EU funding on individual evaluations of the EU. The sample for this experiment was collected by Focus Research, a Slovak survey company.

Respondents were randomly assigned to three groups: control, Treatment A, and Treatment B. Standard randomization checks suggest that randomization on observed covariates was successful.¹¹ Respondents in all groups were asked the following question:

“To what extent do you agree or disagree with the following statement: Slovak politicians have a responsibility to maintain a positive relationship between the EU and Slovakia”

The response scale consisted of four items. I assigned each item a numerical value, ranging from 1 to 4. Values of 1 and 2 correspond to strongly agree and agree, denoting pro-EU positions. Values of 3 and 4 correspond to disagree and strongly disagree, denoting Euroskeptic positions. The nonresponse rate was less than 5 percent. After dropping these responses, the control group had 305 respondents, the Treatment A group had 324 respondents, and the Treatment B group had 307 respondents. The treatments consisted of two different vignettes about the role of EU funding in Slovakia. The control group was not given a vignette, while the vignette differed between the Treatment A group and Treatment B group. The vignette is enumerated below, with the additional treatment administered to the B group in bold:

“As you may know, Slovakia receives funding from the European Union. One of these funds is called the Structural Fund Regional Development Programme. This Programme provides funding for various projects including roads, railways, and agricultural subsidies. A high-ranking European politician recently stated that these funds should be tied to the fair treatment of migrants, refugees, and Roma populations, and to how well a region represents “European Values.”

The vignette for the Treatment A group primes respondents to consider the economic purpose of EU funding. The vignette was written using wording taken from the EU Commission website on EU Regional Policy (European Commission 2018a). The treatment for the B group primes respondents to consider the social purpose of EU funding, alongside the economic goals. This treatment was likewise written using material taken from the EU Commission website on the social inclusion aims of EU Regional Policy (European Commission 2018d).

This survey experiment tests the individual-level mechanism behind the EU funding–Euroskeptic vote choice relationship. By priming respondents with different treatments concerning the role of EU funding, the survey experiment should identify whether, and under what conditions, EU funding precipitates a Euroskeptic response. Theoretical expectation suggests that Euroskepticism should manifest when respondents are exposed to Treatment B. Respondents should be less likely to support a positive relationship between Slovakia and the EU when primed to consider how EU funding aids minority groups. When this social inclusion treatment is absent, I expect respondents to react favorably. Focusing solely on the economic purposes of EU funding should render respondents less Euroskeptic. Thus, I expect Treatment A respondents to respond in a more pro-EU manner than respondents in the control and Treatment B groups.

Average Treatment Effects

I analyze the results of the survey experiment using difference-of-means tests.¹² Table 5 compares the mean responses of respondents in each of the three groups. The mean response for the control group is 1.87, while the mean for the Treatment A group is 1.75. This difference is statistically significant at a p value of .047. Respondents become more favorable toward the EU when exposed to the economic benefits of EU funding.

Table 5.

Survey Experiment Results.

Control mean	Treatment A mean	p value	df
1.87	1.75	.047	626.35
Control mean	Treatment B mean	p value	df
1.87	1.84	.637	606.73
Treatment A mean	Treatment B mean	p value	df
1.75	1.84	.150	622.07

There is no statistically significant difference between the control group and the Treatment B group. The pro-EU effect of EU funding disappears when respondents are primed to consider the social inclusion aims of EU funding. This suggests that respondents revert to their baseline attitudes when exposed to the social purposes of EU funding. Likewise, there is no statistically significant difference between the Treatment A and Treatment B groups. However, the mean response for the Treatment A group is less Euroskeptic than the mean response for the Treatment B group (1.75 vs. 1.84).

Importantly, the experiment does not include politicization or ethnonationalist appeals. While a treatment that adds this type of frame would have been a more accurate representation of the theory, it would have made for a less stringent test. Using only neutral information, as opposed to providing a politicized frame, makes it more difficult to identify a significant difference between groups.

Subgroup Analysis

I explore variations in treatment effects across different levels of (1) political ideologies and (2) education. Previous work on financial transfers between EU member states suggests that these factors may influence the relationship between the treatments and Euroskepticism (e.g., Kuhn, Solaz, and van Elsas 2018; Stoeckel and Kuhn 2018).

The effects of the treatments may be contingent on respondents’ political ideology. Leftists may be more supportive of redistributive transfers than right-leaning respondents. Likewise, respondents on the right of the political spectrum may respond more negatively to minority accommodation than left-leaning respondents. To test this prediction, I divide respondents into three groups: left, center, and right. I measure ideology on a 5-point scale. I code 1 and 2 as left, 3 as center, and 4 and 5 as right. The left group had 172 individuals; center, 389; and right, 279.¹³ Results contradict expectation: left-leaning individuals “move” in response to the treatments, while right-leaning individuals do not.

Specifically, comparing the control and Treatment A groups, left-leaning individuals are less Euroskeptic when exposed to the economic purposes of EU funding, indicating support for EU economic transfers. However, and unlike right-leaning respondents, left-leaning respondents exposed to Treatment B are more Euroskeptic than those exposed to Treatment A (2.11 vs. 1.64, p = .007). It seems that left-leaning individuals, generally pro-EU and pro-redistribution, have room to shift their stances in response to information about EU accommodation of minority groups. Rightists, generally more Euroskeptic, may be less likely to respond to the same information.

Previous research has shown that cosmopolitanism plays an important role in support for EU economic redistribution across member states. Bechtel, Hainmueller, and Margalit (2014, 841) define cosmopolitanism as the “extent to which people have an interest in and orientations towards, groups of individuals who are distant from them”. Such individuals may react more favorably to EU minority transfers. Concurrently, cosmopolitans may be less likely to manifest economic or cultural grievances with minorities. I test whether cosmopolitanism (operationalized as university education) moderates the effects of the treatments. I divide respondents into two groups: those with and those without a university degree. Two interesting findings emerge. First, amongst nonuniversity graduates, exposure to the social and economic vignette (Treatment B) results in more Euroskeptic responses than exposure to only the economic vignette (Treatment A). Mentioning minority accommodation renders these individuals more Euroskeptic in spite of economic benefits (p = .028). Second, when comparing the control group with the treatment groups, university graduates are less Euroskeptic when exposed to Treatment B (p = .018), but not when exposed to Treatment A. It seems that university graduates may care more about social goals, as their economic position is relatively secure. Nonuniversity graduates, on the other hand, seem to weigh cultural grievances more than economic benefits. I present these results, along with the political ideology subgroup analysis, in Figure 4.¹⁴

Figure 4.

Subgroup treatment effects: ideology and education.

Conclusion

Economic theories of Euroskepticism argue that those who benefit from EU transfers are unlikely to be Euroskeptics. National identity approaches contend that Euroskepticism results from the transfer of national competencies to the EU. Cueing explanations suggest that ordinary citizens do not, on their own, formulate evaluations of the EU. Thus far, these approaches have failed to consider how EU transfers fit into the equation. In this article, I suggest that Euroskepticism is a result of anti-minority backlash. Specifically, I contend that the minority advancement realized through EU funding, when politicized by ethnonationalists, promotes Euroskeptic vote choice in CEE. Ethnicity is a salient issue in the region and forms one of the primary cleavages of political competition (Rovny 2014). Yet, the presence of minority groups alone is not sufficient for the development of grievances. Grievances manifest when minorities are politically, economically, or socially accommodated (Bustikova 2014). Ethnonationalists use this accommodation to advance their own aims.

Prior to accession, CEE states had to ensure legal protections for minorities. Before the Eurozone crisis, the economic and normative appeal of EU membership tempered backlash against these policies. Contemporarily, accession criteria no longer drive minority advancement. Instead, advancement is realized through EU Regional Policy. This advancement is politicized by ethnonationalists which, in turn, creates grievances amongst the electorate. Finally, grievances drive Euroskeptic vote choice.

To test these predictions, I used two empirical strategies. First, I conducted a large-N analysis of ten CEE states, broken down into NUTS 2 regions. I controlled for a variety of sociotropic factors that may confound the relationship between EU funding and Euroskeptic voting. Egocentric grievances—both economic and cultural—may also be relevant. I briefly explored the role of these grievances with public opinion data. While results remained consistent to their inclusion, the interaction of EU funding and egocentric grievances is a promising avenue for future research. Across all models, I found that EU funding has a robust positive relationship with aggregate, soft, and ethnonationalist Euroskeptic vote choice. I also found that this effect is contingent on the salience of the ethnic cue in a given region—the stronger the ethnic cue, the larger the effect of EU funding.

To test the individual-level relationship, I conducted a survey experiment on a nationally representative sample of the Slovak population. Results suggested that respondents primed to consider the economic purpose of EU funding—in the absence of any mention of social inclusion—become less Euroskeptic. This pro-EU effect, however, disappeared when minority social inclusion was included in the treatment. At the same time, differential effects across ideological and educational subgroups suggest that these factors play important roles in grievance formation.

Importantly, these results suggest that the long-term goals of the EU are undermined by the short-term consequences of EU policy. EU funding serves an important economic purpose. Funds are used to develop infrastructure, subsidize agriculture, and mitigate economic disparities between EU regions. However, the social inclusion aims of these policies result in unintended consequences. The intended consequence of EU funds is minority advancement. However, minority advancement creates a backlash that drives voters toward Euroskeptic electoral options. The success of Euroskeptic parties also has important consequences for a democratic CEE. Although not all Euroskeptic parties eschew democratic standards, some do. The long-term success of these parties potentially portends the erosion of democracy and minority rights.

Supplemental Material

REVISED_SUPPLEMENTAL_MATERIAL – Supplemental material for EU Funding and Euroskeptic Vote Choice

Supplemental material, REVISED_SUPPLEMENTAL_MATERIAL for EU Funding and Euroskeptic Vote Choice by Roman Hlatky in Political Research Quarterly

Footnotes

Acknowledgements

I thank Lenka Buštiková, Terrence Chapman, Kenneth Greene, Nate Jensen, Tse-min Lin, Amy Liu, Tatiana Londoño, Zeynep Somer-Topcu, Chris Wlezien, and the reviewers and editors for their valuable comments. Earlier versions of this article were presented at Comenius University, at the IPSA Colloquium on Diversity and Democratic Governance, and at the 2019 Texas Comparative Circle Conference. I thank all participants for their comments. All remaining errors are my own responsibility.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethical Approval

Institutional review board (IRB) approval was granted at the University of Texas at Austin.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Support for this research was provided by the Texas Politics Project and the Center for Russian, East European, and Eurasian Studies at the University of Texas at Austin.

ORCID iD

Roman Hlatky

Notes

Supplemental Material

Data, replication scripts, and supporting information are available on the author’s website: .

References

Anderson

Christopher

. 1998. “When in Doubt Use Proxies: Attitudes to Domestic Politics and Support for the EU.” Comparative Political Studies 31:569–601.

Bauhr

Monika

Charron

Nicholas

. 2018. “Why Support International Redistribution? Corruption and Public Support for Aid in the Eurozone.” European Union Politics 19 (2): 233–54.

Baute

Sharon

Abts

Koen

Meuleman

Bart

. 2019. “Public Support for European Solidarity: Between Euroscepticism and EU Agenda Preferences?” JCMS: Journal of Common Market Studies 57 (3): 533–50.

Bechtel

Michael M.

Hainmueller

Jens

Margalit

Yotam

. 2014. “Preferences for International Redistribution: The Divide over the Eurozone Bailouts.” American Journal of Political Science 58 (4): 835–56.

Börzel

Tanja A.

Schimmelfennig

Frank

. 2017. “Coming Together or Drifting Apart? The EU’s Political Integration Capacity in Eastern Europe.” Journal of European Public Policy 24 (2): 278–96.

Brambor

Thomas

Clark

William Roberts

Golder

Matt

. 2006. “Understanding Interaction Models: Improving Empirical Analyses.” Political Analysis 14 (1): 63–82.

Bustikova

Lenka

. 2014. “Revenge of the Radical Right.” Comparative Political Studies 47 (12): 1738–65.

Buti

Marco

Carnot

Nicolas

. 2012. “The EMU Debt Crisis: Early Lessons and Reforms.” Journal of Common Market Studies 50 (6): 899–911.

Cox

Gary

. 1997. Making Votes Count: Strategic Coordination in the World’s Electoral Systems. New York: Cambridge University Press.

10.

Daniele

Gianmarco

Geys

Benny

. 2015. “Public Support for European Fiscal Integration in Times of Crisis.” Journal of European Public Policy 22 (5): 650–70.

11.

De Vreese

Claes H.

Boomgaarden

Hajo G.

2005. “Projecting EU Referendums: Fear of Immigration and Support for European Integration.” European Union Politics 6 (1): 59–82.

12.

De Vries

Catherine E.

2018. Euroscepticism and the Future of European Integration. Oxford: Oxford University Press.

13.

De Vries

Catherine E

Edwards

Erica E.

2009. “Taking Europe to Its Extremes: Extremist Parties and Public Euroscepticism.” Party Politics 15 (1): 5–28.

14.

European Commission. 2018a. “EU Regional Policy.” https://ec.europa.eu/regional_policy/en/.

15.

European Commission. 2018b. “European Commission Budget.” https://ec.europa.eu/budget/graphs/annual_life_cycle.html.

16.

European Commission. 2018c. “Historic EU payments—Regionalized and Modeled.” https://cohesiondata.ec.europa.eu/Other/Historic-EU-payments-regionalised-and-modelled/tc55-7ysv.

17.

European Commission. 2018d. “Social Inclusion.” https://ec.europa.eu/regional_policy/en/policy/themes/social-inclusion/.

18.

Fidesz. 2018. “Magyarorszag jol safarkodik az unios pénzekkel, és elutasítja a forrasok tervezett atcsoportosítasat [Hungary is well funded with EU money and rejects the planned reallocation of resources].” https://www.fidesz.hu/hirek/magyarorszag-jol-safarkodik-az-unios-penzekkel-es-elutasitja-a-forrasok-tervezett-atcsoportositasat.

19.

Gabel

Matthew J.

1998. “Public Support for European Integration: An Empirical Test of Five Theories.” Journal of Politics 60:333–54.

20.

Glytsos

Nicholas

Viela

Bruno

. 2010. “The European Social Fund: Migrants and Minorities, a Study Produced by Bernhard Brunhes International (BBI) for the European Commission.” https://ec.europa.eu/employment_social/esf/docs/br_migrants_minorities_en.pdf

21.

Hobolt

Sara

. 2009. Europe in Question: Referendums on European Integration. Oxford: Oxford University Press.

22.

Hobolt

Sara

De Vries

Catherine E.

2016. “Public Support for European Integration.” Annual Review of Political Science 19:413–32.

23.

Hooghe

Liesbet

Marks

Gary

. 2005. “Calculation, Community and Cues: Public Opinion on European Integration.” European Union Politics 6 (4): 419–43.

24.

Hooghe

Liesbet

Marks

Gary

. 2008. “A Postfunctional Theory of European Integration: From Permissive Consensus to Constraining Dissensus.” British Journal of Political Science 39:1–23.

25.

Hughes

James

Sasse

Gwendolyn

. 2003. “Monitoring the Monitors: EU Enlargement Conditionality and Minority Protection in the CEECs.” JEMIE: Journal on Ethnopolitics and Minority Issues in Europe 1:1–35.

26.

Ivarsflaten

Elisabeth

. 2008. “What Unites Right-wing Populists in Western Europe? Re-examining Grievance Mobilization Models in Seven Successful Cases.” Comparative Political Studies 41:3–23.

27.

Katz

Richard S.

Mair

Peter

. 1995. “Changing Models of Party Organization and Party Democracy: The Emergence of the Cartel Party.” Party Politics 1 (1): 5–28.

28.

Kleider

Hanna

Stoeckel

Florian

. 2019. “The Politics of International Redistribution: Explaining Public Support for Fiscal Transfers in the EU.” European Journal of Political Research 58 (1): 4–29.

29.

Kopecky

Petr

Mudde

Cas

. 2002. “The Two Sides of Euroscepticism.” European Union Politics 3 (3): 297–326.

30.

Krause

Werner

Lehmann

Pola

Lewandowski

Jirka

Matthieß

Theres

Merz

Nicolas

Regel

Sven

Werner

Annika

. 2018. “Manifesto Corpus.” Version: 2018-2. Berlin: WZB Berlin Social Science Center.

31.

Kuhn

Theresa

Solaz

Hector

van Elsas

Erika J.

2018. “Practising What You Preach: How Cosmopolitanism Promotes Willingness to Redistribute across the European Union.” Journal of European Public Policy 25 (12): 1759–78.

32.

Kuhn

Theresa

Stoeckel

Florian

. 2014. “When European Integration Becomes Costly: The Euro Crisis and Public Support for European Economic Governance.” Journal of European Public Policy 21 (4): 624–41.

33.

Kymlicka

Will

Opalski

Magda

, eds. 2002. Can Liberal Pluralism Be Exported? Western Political Theory and Ethnic Relations in Eastern Europe. Oxford: Oxford University Press.

34.

Lenz

Gabriel

Sahn

Alexander

. 2018. “Achieving Statistical Significance with Covariates and without Transparency.” Working paper. University of California, Berkeley.

35.

L’SNS. 2019a. “Milan Uhrık vyzval politikov z Bruselu, aby nemudrovali o rasizme a radsej si skusili ist’byvat do osad.” “Milan Uhrik asked politicians from Brussels to stop pondering racism and to try living in the settlements”. http://www.naseslovensko.net/nase-nazory/kriminalita-asocialov/milan-uhrik-vyzval-politikov-z-bruselu-aby-nemudrovali-o-rasizme-a-radsej-si-skusili-ist-byvat-do-osad.

36.

L’SNS. 2019b. “POZOR! Europarlament schvalil nariadenie ocernochoch v Europe! [CAREFUL! European parliament approved regulation about blacks in Europe].” http://www.naseslovensko.net/nase-nazory/europska-unia/pozor-europarlament-schvalil-nariadenie-o-cernochoch-v-europe/.

37.

Lubbers

Marcel

Scheepers

Peer

. 2002. “French Front National Voting: A Micro and Macro Perspective.” Ethnic and Racial Studies 25:120–49.

38.

McLaren

Lauren M.

2006. Identity, Interests and Attitudes to European Integration. London: Palgrave Macmillan.

39.

Polk

Jonathan

Rovny

Jan

Bakker

Ryan

Edwards

Erica

Hooghe

Liesbet

Jolly

Seth

Koedam

Jelle

, et al. 2017. “Explaining the Salience of Anti-elitism and Reducing Political Corruption for Political Parties in Europe with the 2014 Chapel Hill Expert Survey Data.” Research & Politics. Published electronically January 1. doi:10.1177/2053168016686915.

40.

Rovny

Jan

. 2014. “Communism, Federalism, and Ethnic Minorities: Explaining Party Competition Patterns in Eastern Europe.” World Politics 66 (4): 669–708.

41.

Rydgren

Jens

. 2008. “Immigration Sceptics, Xenophobes or Racists? Radical Right-wing Voting in Six West European Countries.” European Journal of Political Research 47:737–65.

42.

Schimmelfennig

Frank

Sedelmeier

Ulrich

. 2004. “Governance by Conditionality: EU Rule Transfer to the Candidate Countries of Central and Eastern Europe.” Journal of European Public Policy 11 (4): 661–79.

43.

Stoeckel

Florian

Kuhn

Theresa

. 2018. “Mobilizing Citizens for Costly Policies: The Conditional Effect of Party Cues on Support for International Bailouts in the European Union.” JCMS: Journal of Common Market Studies 56 (2): 446–61.

44.

Taggart

Paul

Szczerbiak

Aleks

. 2002. “The Party Politics of Euroscepticism in EU Member and Candidate States.” Sussex European Institute, Working Paper No. 51. http://www.sussex.ac.uk/sei/publications/seiworkingpapers

45.

Tomio Okamura—SPD. 2019. “Facebook Post.” https://www.facebook.com/tomio.cz/photos/a.185333081477515/2665515160125949/?type=3&theater.

46.

Tucker

Joshua A.

Pacek

Alexander C.

Berinsky

Adam J.

2002. “Transitional Winners and Losers: Attitudes toward EU Membership in Post-communist Countries.” American Journal of Political Science 46 (3): 557–71.

47.

Usherwood

Simon

Startin

Nick

. 2013. “Euroscepticism as a Persistent Phenomenon.” Journal of Common Market Studies 51 (1): 1–16.

48.

VMRO (Internal Macedonian Revolutionary Organization). 2019. “Alexander Sidi: People in the Ghetto Have Categorically Made Their ‘Sensible Choice’—Not to Socialize.” http://www.vmro.bg/александър-сиди-хората-в-гетата-категорично-са-направили-своя-разумен-избор-да-не-се-социализират/.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

1.60 MB

1.71 MB