Public Support for European Integration in Greece and Italy Between 2015 and 2020

The consecutive crises of the last decade have eroded public support for the EU, especially so among Southern European member states. A long‐standing scholarly debate centers around whether it is economic or cultural considerations that drive public support for the EU. However, it is still unclear whether public attitudes toward European integration are driven primarily by economic evaluations or concerns associated with growing immigration flows. To explore this question, we draw on data from the Eurobarometer in Greece and Italy between 2015 and 2020. We find consistent evidence that diffuses public support for the EU and specific support for EU institutions are associated positively with economic evalua‐ tions of the European economy and household finances and negatively with opposition to immigration. Our study provides further insights into the dynamics of public support for the EU in the European periphery during critical times


Introduction
Since the 2009 eurozone crisis and the 2015 immigration crisis, public support for the EU has been questioned across member states, especially in Southern Europe. The increasing public opposition to the EU has manifested itself in the rise of nationalist-Eurosceptic parties across the continent (Schmitter & Lefkofridi, 2016). Euroscepticism further proliferated during the Covid-19 pandemic crisis, even though the EU's response to the pandemic has been evaluated relatively positively (Jones et al., 2021;Rhodes, 2021). This dynamic was particularly salient in Greece and Italy, two European countries that were hit hard by all three crises. In Greece, the eruption of the financial crisis had an impact on the rise of anti-EU sentiments (Georgiadou, 2019;Halikiopoulou, 2020). Similarly, public opinion in Italy has been increasingly Eurosceptic since the eurozone crisis, and in some cases, it has even been starkly negative against deeper European integration (Conti, Di Mauro, & Memoli, 2020;Conti, Marangoni, & Verzichelli, 2020;Giannetti et al., 2017).
However, it is still unclear in the existing literature whether economic evaluations or opposition to immigration is a better predictor of public attitudes toward the EU. This article attempts to fill the gap in the literature by exploring the economic and immigration-related cultural antecedents of public support for the EU in Greece and Italy between 2015 and 2020. This period was marked by considerable suffering caused by the financial, immigration, and Covid-19 crises in both member states. More specifically, the article aims to answer the question: Does the impact of economic evaluations or opposition to immigration better explain public support for the EU in Greece and Italy? To investigate the correlates of public support for European integration, we draw on cross-sectional and longitudinal data from the Eurobarometer. We provide consistent evidence that diffuses public support for the EU and specific support for EU institutions are positively associated with favorable evaluations of the European economy and household finances and negatively associated with opposition to immigration. We argue that anti-immigrant sentiments are the primary suppressor of public support for the EU in both countries. As these sentiments increase, diffuse support for the EU and specific support for EU institutions diminish, even for institutions that are not directly responsible for the management of the immigration crisis, such as the European Central Bank. Furthermore, positive economic assessments of the EU economy and household finances are positively correlated with trust in the EU.
We specifically focus on Italy and Greece from 2015 to 2020 for three reasons. Firstly, we consider this period critical due to the outbreak of the immigration crisis in 2015 which mostly affected Greece and Italy. In fact, both countries received the largest numbers of migrants and asylum seekers and reported the highest percentages of such populations among all first-arrival European countries (International Organization for Migration, 2016, 2021. During this period, the effects of the 2009 eurozone crisis were still being felt in both countries (Bull, 2018;Maris et al., 2022;Serapioni & Hespanha, 2019). In terms of gross domestic product, the fiscal years of 2013-2016 were the worst for both Greece and Italy in the pre-Covid-19 era (The World Bank, 2023).
Finally, the Covid-19 pandemic, an unprecedented crisis for the entire world, hit Greece and Italy hard, resulting in high fatalities and an economic downturn. At the beginning of the pandemic, fatalities were higher in Italy than in Greece, but, as the crisis progressed, Greece performed significantly worse in terms of both health and economic outcomes according to Chantzaras and Yfantopoulos (2022). Overall, these consecutive crises have harmed public support for deeper EU integration (Dimitrakopoulos & Lalis, 2022;Lefkofridi & Schmitter, 2015;Schimmelfennig, 2018).

Public Support for European (Dis)Integration
In the conceptual framework initially introduced by Easton (1975), diffuse and specific support represent a stable political self-orientation toward political objects in a system of governance. Norris (1999) expanded this classification into a five-fold framework, introducing the notions of political support for the community, principles, performance, institutions, and actors. As "critical citizens" express their support for distinct levels of governance, this conceptual framework is useful for measuring public support for the EU for two reasons.
First, during times of economic stress, "without a reservoir of goodwill towards democratic institutions" (Norris, 1999, p. 203), the buffer zone between people's demands and the performance of the political sys-tem weakens. Second, regime (diffuse) support is taken for granted in established democracies. However, in the case of a hybrid multilevel political system such as the EU, this type of support is highly uncertain (Hobolt & de Vries, 2016). Further, specific support for EU institutions is also affected during times of crisis as EU policies affect national policies and practices (Hooghe & Marks, 2001) and strengthen the ambiguity around the EU's polity (de Wilde & Trenz, 2012).
Therefore, the two-dimensional approach of public support for the EU project is relevant because it allows for a nuanced understanding of public attitudes toward the EU. Regime support refers to overall support for the establishment of the EU, while institutional support measures public trust in the institutional framework of the EU (Hobolt & de Vries, 2016).
Public opposition to the European project has been steadily increasing since the 2000s (Hobolt & de Vries, 2016). According to Webber (2018), EU disintegration occurs when one or more EU institutions lose power or authority, when a member state withdraws from the union (e.g., Brexit), or when a process for renationalization of EU common policy starts. Scholarship also argues that European disintegration is a multifaceted process that involves economic, socio-cultural, territorial, political, and legal dimensions, rather than a unidimensional political outcome (Vollaard, 2018).
The ambiguity surrounding policymaking and decision-making processes in the EU has been increasingly politicized and mediatized, even before the eurozone crisis in 2009(de Vries, 2007Hobolt, 2009;Hooghe & Marks, 2009). The lack of consensus between elites and the public about the nature of the EU and its institutions has spurred an almost existential crisis within the EU (de Wilde & Trenz, 2012;Hobolt & de Vries, 2016). Although EU integration has deepened and the public interest in EU affairs has increased over the years, public opinion has remained highly reluctant toward the EU project (de Vries & Steenbergen, 2013).
Since 2009, the future of the EU has been put into question due to the rise of Eurosceptic parties (Lefkofridi & Schmitter, 2015;Schmitter, 2012), the politicization of EU issues at the national level, and the decrease in trust in national and EU institutions (Cramme & Hobolt, 2015). During the eurozone crisis, the issue of integration became increasingly politicized by parties challenging the EU project (Hobolt & de Vries, 2015). Indeed, the impact of the financial crisis demonstrated how public opinion can be a "leverage" for national governments to negotiate with EU institutions (Halikiopoulou, 2020;Hobolt & de Vries, 2016;Vollaard, 2014;Webber, 2018).

Economic Hardship and European (Dis)Integration
Public support for European integration is often thought to have a utilitarian explanation. The main argument of this approach is based on the benefits of the liberalization of EU trade and how it rewards citizens with higher levels of income and human capital, particularly in terms of education and professional skills Tucker et al., 2002). Indeed, high-skilled workers and entrepreneurs with capital are more supportive of European integration than blue-collar workers (Hobolt & de Vries, 2016;Hooghe & Marks, 2005).
Economic assessments of support or opposition to the EU project can be divided into two main categories: egocentric and socio-tropic (Hooghe & Marks, 2005). According to the former, an individual's perception of having benefited or not from the EU project can affect their trust in EU integration (Aiello et al., 2019). The latter refers to the fact that individuals' support depends on the evaluation of national economies and whether or not their countries received economic aid from the EU (Aiello et al., 2019). Economic hardship at the national level predicts opposition toward the EU project (Hobolt & de Vries, 2016). Public support for the EU is higher in countries where economic performance is positive, due to the benefits of EU integration (Anderson & Kaltenthaler, 1996;Eichenberg & Dalton, 1993;Hooghe & Marks, 2004). Short-term economic performance is an important determinant of EU support, according to this approach , as well as a strong predictor for the future of national and EU economies (Aiello et al., 2019;de Vreese et al., 2008;Loveless, 2010). The political context can also influence public support or opposition and interacts with economic evaluations. Specifically, discontent is more pronounced when accountability mechanisms are clear (Anderson, 2000), i.e., when public opinion can evaluate the performance of an institution in a straightforward manner.
Since the eurozone crisis, economic insecurity has clouded many of the benefits of EU integration. In Greece, the economic crisis has undermined trust in the traditional political system (Kriesi, 2012) and served as a springboard for far-right ideologies advocating for further disintegration (Ellinas, 2013). This impact increased the electoral base of anti-austerity parties, such as Syriza on the far left and Golden Dawn on the far right of the political spectrum (Bedock & Vasilopoulos, 2015). The percentage of Greek citizens with a negative opinion about the EU increased from 2009 onwards, and this trend continued with the discussion of a potential Grexit after the referendum of 2015 (Katsanidou & Lefkofridi, 2020;Vasilopoulou, 2018). Making austerity measures a prerequisite for EU membership hardened Greek Euroscepticism (Lefkofridi & Nezi, 2020), and the EU became a target of blame (Verney, 2015) as Greek public opinion rejected EU economic policies and interference in national politics (Vasilopoulou, 2018).
In Italy, economic performance was already a strong predictor of EU evaluations even before the crisis. However, the financial crisis and the subsequent austerity measures imposed by the EU have strengthened Italians' negative attitudes toward the EU (Bull, 2018) and increased Eurosceptic voting (Bellucci, 2014). Similar to the Greeks, Italians blamed the EU for being an entity that imposes austerity measures (Conti, Di Mauro, & Memoli, 2020). As a result, EU-led austerity policies increased Eurosceptic attitudes in Italian public opinion (Conti, Di Mauro, & Memoli, 2020). In the past, public opinion often associated economic motivations and material benefits with European integration (Bellucci et al., 2012), but this link was disrupted after the eurozone crisis, enhancing Eurosceptic sentiments in Italy (Giannetti et al., 2017).
In line with these findings, we expect that public support for the EU and EU institutions will be positively (negatively) associated with (un)favorable assessments of the European (H1a) and national (H1b) economy as well as household finances (H1c).

Immigration and European (Dis)Integration
The immigration crisis of 2015 has increased the importance of the "cultural threat" and aggravated anti-EU sentiments (Brack, 2020). This has given rise to a new cultural cleavage that opposes "open" positions of integration to "closed" positions of demarcation, which has had a significant impact on identity politics in European countries (Kriesi et al., 2008;Loch & Norocel, 2015). As argued by Loch and Norocel (2015), this cleavage involves an internal dimension (immigration) and an external dimension (European integration). Perceived cultural threats play a major role in the articulation of national identities, as they trigger the rejection of different cultures and the protection of national ideals (Loch & Norocel, 2015).
Furthermore, immigration is often associated with Eurosceptic attitudes characterized by hostility toward immigrants and minority groups (de Vreese & Boomgaarden, 2005;Hobolt et al., 2011), as well as toward different cultures in general (McLaren, 2002(McLaren, , 2006. Carey (2002) also demonstrated that people who express strong national attachment and pride are less supportive of EU integration. Moreover, left-wing and right-wing Eurosceptics perceive immigration differently. On the one hand, left-wing Eurosceptics argue that the EU should create a "safe path" for people to immigrate (Brack, 2020). On the other hand, right-wing supporters claim that immigration poses a threat to national identity and security and publicly oppose the free movement of foreign populations in the EU (Brack, 2020).
In Greece, immigration became an issue of significant concern for public opinion mostly after 2015 and the outbreak of the migration and refugee crisis on the Greek shores and islands (Dimitriadi & Sarantaki, 2019). This crisis, similar to the debt crisis, incited the discussion around Grexit and turned the EU into the scapegoat considered responsible for the crisis and for failing to provide adequate response and assistance to Greece (Dimitriadi & Sarantaki, 2019). This disaffection was manifested by an increase in public support for anti-immigrant and neo-fascist political parties (Dinas et al., 2019). The Popular Orthodox Rally (LAOS) and Golden Dawn based their political programs and rallies on an antiimmigration and anti-EU agenda (Dennison & Geddes, 2019;Ellinas, 2013;Vasilopoulou & Halikiopoulou, 2015), as did the new far-right party Greek Solution more recently (Georgiadou, 2019).
Similarly, in Italy, the public perceived the management of the refugee crisis by the EU and the rules established by the Dublin regulation as the cause of disproportionate costs for the country (Dixon et al., 2018). The crisis of 2015 increased perceptions of the cultural threat of immigration and added to the fear of its economic impact on the country, in a context already burdened by socio-economic concerns, particularly by the conservative and nationalist segments of Italian society (Conti, Marangoni, & Verzichelli, 2020). Italians also consider immigration to be one of the two most important issues affecting their country and the EU, and they have a less positive image of the EU than other Europeans (Geddes & Pettrachin, 2020). The EU was also held accountable for leaving Italy alone to handle the migration and refugee crisis, and far-right supporters openly challenged EU migration policies and approaches (Brunazzo & Mascitelli, 2020).
Following these findings, we expect that public support for the EU and EU institutions will be negatively associated with anti-immigrant sentiments (H2).

The Interaction Between Economic Evaluations and Anti-Immigrant Sentiment
Opposition to immigration also has an economic dimension, often referring to it as a threat to "economic competition" (Dennison & Geddes, 2021;Pardos-Prado & Xena, 2019;Ruist, 2016). Immigrants are considered to have negative effects on individual and societal material well-being. Far-right Euroscepticism is mostly driven by the fear of cultural and economic losses (Bremer & Schulte-Cloos, 2019;Lefkofridi & Michel, 2017), as it perceives national sovereignty not only as cultural but also as an economic response against external threats (Mazzoleni & Ivaldi, 2020). Far-right supporters see immigrants, particularly low-educated immigrants from non-Western economies (Edo et al., 2019), as potential "threats" to the national economy and labor market (Mazzoleni & Ivaldi, 2020).
In contrast, far-left Euroscepticism focuses solely on the negative economic impacts of European integration (Brack, 2020). For the far-left opposition to the EU, immigration is not a real concern and has a negligible economic dimension. Far-left supporters believe that immigration is not a "crisis" and advocate for creating a safe passage to Europe for immigrants while addressing the underlying causes of immigration (Brack, 2020).
In Greece, the far-right contestation responds to attitudes against immigration in the EU. Although the cultural threat is more prevalent, Greek far-right supporters believe that immigration can have negative effects on the national economy because immigrants could affect the redistribution of the welfare system to natives (Sekeris & Vasilakis, 2016;Vasilopoulou & Halikiopoulou, 2015). Such discourses were evident in the political rallies of various far-right parties, including LAOS, Golden Dawn, and, most recently, Greek Solution (Georgiadou, 2019;Roumanias et al., 2018).
Similarly, in Italy, opposition to immigration due to economic grievances comes from far-right Eurosceptic parties, as the left and far left see positive effects of immigration on the domestic economy (Barone et al., 2016). Immigrants are often accused of "stealing" jobs and housing from Italian citizens in various political discourses of far-right parties (Caiani & Kröll, 2017;Castelli Gattinara & Froio, 2016). These fears increase during times of crisis as immigrants are perceived by far-right supporters as competitors against Italian citizens for the same limited resources (Conti, Marangoni, & Verzichelli, 2020).

Data and Methods
To explore the correlates of public support for the EU and EU institutions, we turn to 10 representative surveys of the Greek and Italian population that cover the period between 2015 and 2020 (N = 20,337 This six-year period was critical for public support for the EU. On the one hand, the effects of the economic crisis of 2008 were still strong on the population of the European periphery. On the other hand, this period captures the European migrant crisis and the years that immediately precede it. In the EU, the migrant crisis peaked in 2015 but the pressure from massive migration flows is still felt in both countries. Finally, the 2020 survey wave studies attitudes during the first phase of the Covid-19 pandemic, which represents a critical juncture for public trust in the EU.

Measuring Anti-Immigrant Sentiment and Economic Evaluations
Previous literature suggests that public support for the EU hinges on identity-related factors as well as on utilitarian evaluations of the economy (for an overview, see Hobolt & de Vries, 2016). To gauge anti-immigrant sentiments, we combine two items in an additive scale (Cronbach's alpha = 0.674; for exploratory factor analysis, see Table B1 in the Supplementary File). The first question asks respondents whether they agree that immigrants contribute a lot to their country on a four-point scale, ranging from totally agree to totally disagree. The second item asks how respondents feel about the immigration of people from outside the EU on a five-point scale ranging from very positive to very negative.
Further, we measure socio-tropic and egocentric evaluations of the economy by constructing three additive scales. The first two scales tap into retrospective and prospective evaluations of the national and European economy by adding assessments of the current situation in the national/European economy (measured on a four-point scale) and relevant expectations for the next 12 months (measured on a three-point scale). Accordingly, we gauge egocentric economic evaluations by adding two similar items that ask respondents about their assessment and expectations regarding their household finances (for exploratory factor analyses and the respective Cronbach's alpha reliability coefficients, see Tables B2-B4 in the Supplementary File).
Finally, we measure respondents' ideology (measured on a 10-point scale), sex, age, education, occupation, and self-reported belonging to the working class, the middle class, or the upper class of society (for the correlation matrices, see Figures B1-B3 in the Supplementary File).

Measuring Diffuse and Specific Public Support for the European Union
Our main dependent variables of interest measure diffuse and specific public support for the EU. In the Eurobarometer, the classic measure of diffuse public support for European integration is an item that asks respondents if their country's membership in the EU is a good thing, a bad thing, or neither good nor bad. However, this item is not available in any of the Eurobarometer waves we draw on. Moreover, the fact that this question is measured with a three-point scale would make the regression analysis more complicated as the dependent variable would be ordinal. To overcome these shortcomings, we create a nine-point scale that combines two items that arguably tap into a closely related construct.
Indeed, the first item asks respondents to rate the image of the EU on a five-point scale ranging from very good to very bad while the second question is a dummy that asks if respondents tend to trust or not to trust the EU (Cronbach's alpha = 0.699; for exploratory factor analysis, see Table B5 in the Supplementary File).
Accordingly, we use three measures of specific public support that ask whether participants tend to trust or not to trust the European Commission, the European Parliament, and the European Central Bank. In addition, we construct a scale that gauges public support for European institutions in general (Cronbach's alpha = 0.881; for exploratory factor analysis, see Table  B6 in the Supplementary File). Figures 1 and 2 show how average public support for the EU and EU institutions fluctuates between 2015 and 2020. Across all four targets, public support steadily decreases to reach an average level of almost 40%. This trend resonates with previous trends reported by Hobolt and de Vries (2016) who found that public support for EU membership plummeted among Southern member states after the 2010 financial crisis. Indeed, Greeks and Italians seem to be reluctant to trust the EU even after 2015 as the migration crisis unfolds.

Analytic Strategy
To study public support for the EU in Greece and Italy, we run a series of linear regressions. We pool observations from 2015 to 2020 and fit a series of linear models that include anti-immigrant sentiment, evaluations of the national and European economy, and evaluations of the household finances as predictors while adjusting for wave, year, and country fixed effects as well as for ideology, sex, age, education, occupation, and self-reported class identity. We choose to estimate fixed-effects models as we employ a similar systems design, which allows us to reduce the heterogeneity between the cases we analyze and to study the associations of main interest in depth (Sartori, 1991). Moreover, fixed-effects models are more appropriate than mixed-effects models for our analysis as they are less vulnerable to omitted variable bias, depend on fewer assumptions, and the sample size of the superordinate variables (countries, years, waves) is relatively small (McNeish & Kelley, 2019). The dependent variables of the models are diffuse public support for the EU and specific support for the European Commission, the European Parliament, the European Central Bank, and European institutions in general. The formula of the multivariate fixed-effects OLS equations is structured as follows: where t, w, and c index years, waves, and countries (with Greece assigned in the reference category), respectively; 1 -4 report the associations of anti-immigrant sentiments and evaluations of national and European economy and household finances; X i,t,w,c is a set of controls outlined above; t , w , and c are the year, wave, and country fixed effects, respectively; and i,t,w,c is the error term. In addition, to study how the association between the dependent variables and the main regressors of interest vary across years, we estimate a series of conditional fixed-effects OLS models, that is models that include interactive terms. The structure of the conditional equa-tions is as follows: t,w,c Prior to estimating the models, all variables were normalized to range from 0 to 1, and therefore coefficients should be interpreted as percentage points. Although our analysis has a descriptive value, an important caveat is that in the absence of random assignment, we cannot support any causal claims.

Empirical Results
We begin the analysis by showing the results of diffuse public support for the EU. We then turn our attention to specific public support for the European Commission, the European Parliament, the European Central Bank, and European institutions in general. This differentiation between levels of trust provides the opportunity to provide a more comprehensive analysis of the dynamics of public support for the EU and EU institutions. Due to the large size of the sample, we focus on the substantive rather than the statistical significance of associations and report related 95% confidence intervals. The large size of our sample increases the precision of reported associations and thus many of them achieve statistical significance. However, in the subsequent analysis, we show that anti-immigrant sentiments and economic evaluations are substantively stronger predictors of diffuse and specific public support for the EU than other variables included in the models, even ideology and social class. In the following tables, we present results from multivariate and conditional OLS fixed-effects models. Table 1 presents the correlates of diffuse public support for the EU in Greece and Italy from 2015 to 2020. Anti-immigrant sentiments and economic evaluations are the primary drivers of public support for the EU. However, the strength of these associations varies considerably in predicting public support. Results from the conditional fixed-effects OLS model resemble closely those obtained from the multivariate linear model. Two points stand out. As the European economy recovers from the decade-long financial crisis, Table 1. Anti-immigrant sentiment, economic evaluations, and public support for the EU in Greece andItaly (2015-2020 Greece andItaly (2015-2020  In Tables C1 and C2 in the Supplementary File, we also investigate whether economic assessments interact with opposition to immigration or country to predict public support for the EU. Although we find little evidence in favor of interaction between anti-immigrant sentiments and economic assessments, the results suggest that country variation is substantive but relatively small. More specifically, anti-immigrant sentiments and positive assessments of the national economy are more negatively associated with support for the EU in Italy than in Greece while the inverse association is found regarding assessments of the EU economy. Greece andItaly, 2015-2020 We now turn our focus to specific public support for EU institutions. Table 2 presents the results from the first set of tests. In line with previous findings, opposition to immigration is negatively associated with support for the European Commission (b = −0.284, 95% CI Table 2. Fixed-effects OLS models: Anti-immigrant sentiment, economic evaluations, and public support for EU institutions in Greece andItaly (2015-2020  These results support H2 as they show that opposition to immigration is the strongest and most persistent sup-pressor of public support for the EU in Italy and Greece during the six-year period. The results concerning economic evaluations follow a similar pattern as in previous tests. Positive assessments of the European economy are the strongest predictor of public support for the European Commission Table 3. Fixed-effects conditional OLS models: Anti-immigrant sentiment, economic evaluations, and public support for EU institutions in Greece andItaly (2015-2020  Further, egocentric assessments of household finances, as well as socio-tropic assessments of the national economy, are correlated positively with public trust in EU institutions but to a lesser degree (b ranging between 0.072 and 0.197). Taken together, this evidence is in favor of H1a, H1b, and H1c. Accordingly, self-identification with the upper social class is the only demographic attribute that predicts moderate but consistent public support for EU institutions (b ranging from 0.104 to 0.137).

Specific Public Support for European Union Institutions in
The results from our final tests are presented in Table 3 and essentially replicate the analysis above. One exception is worth highlighting: The association between assessments of the national economy and public support for EU institutions is found to be substantively negligible. Similarly, to diffuse public support, the association between positive evaluations of the EU economy and public support for EU institutions grows stronger as the European economy recovers from the financial crisis starting in 2017.
In Tables C1 and C2 in the Supplementary File, we extend our analysis to study whether anti-immigrant sentiment or country moderate the association of economic assessments with public support for EU institutions. We do not find evidence that the conditional effect between anti-immigrant sentiments and economic assessments is substantive. In contrast, evidence supports that there is variation across Greece and Italy, albeit relatively small. Similar to previous results, antiimmigrant sentiments and assessments of the national economy predict support for the EU more negatively in Italy than Greece while the opposite is true regarding assessments of the EU economy.

Discussion
The economic depression, immigration crisis, and Covid-19 pandemic have all contributed to a decline in public support for deeper EU integration, particularly in the periphery. These challenges have diminished the reservoir of goodwill that once legitimized European institutions and initiatives for greater integration. To alleviate the socio-economic repercussions of the Covid-19 pandemic in EU countries, the NextGenerationEU recovery program was introduced as a mitigation policy. However, as this program was adopted in December 2020, during a period that is outside the scope of this study, its impact on public opinion in Greece and Italy should be explored in future research.
To understand whether public support for the EU is based on economic evaluations or anti-immigrant sentiments, we examined data from the Eurobarometer survey conducted between 2015 and 2020 in Greece and Italy. Indeed, our study focuses on the association between subjective economic evaluations and opposition to immigration with public support for the EU at the individual level. Future studies should investigate whether objective measures of migration flows and changes in GDP also predict public trust in the EU at the aggregate level as well as whether subjective perceptions and objective measures are sufficiently correlated with one another, or whether public opinion suffers from widespread misperceptions.
Opposition to immigration and economic evaluations are the primary correlates of public support for the EU and its institutions in Greece and Italy. More specifically, favorable economic evaluations of the European economy and household finances are positively associated with the trust of Greeks and Italians in the EU, the European Commission, the European Parliament, and the European Central Bank. In contrast, economic evaluations of the national economy play only a negligible role.
Further, anti-immigrant sentiments are negatively correlated with public support for the EU. As antiimmigrant sentiments increase, diffuse public support for the EU, as well as specific support for EU institutions, decreases, even in the case of institutions such as the European Central Bank that are not responsible for addressing immigration issues. Indeed, the case of the European Central Bank highlights the pervasive symbolic consequences of anti-immigrant sentiments. These considerations have become so central in how the Greek and Italian publics perceive and evaluate the EU that citizens do not seem to differentiate between institutions based on their policy area of responsibility. Future research should delve deeper into the mechanisms that underlie responsibility attribution across different policy areas and institutions in the EU.
Finally, ideology and social class are the only remaining factors that correlate with public support for the EU. Individuals who identify as right-wing and those from the upper classes are more likely to express pro-European opinions, but the relationship is not linear. Instead, people who adhere to extreme ideologies, either right or left, are more critical of the EU.
In this article, we find that Italian and Greek citizens are capable of distinguishing between the performance of the EU economy and their national economies. Specifically, the Italian and Greek publics assess the economic performance of both national and European economies but only the latter assessments predict support for EU institutions. In addition to socio-tropic evaluations, both publics take into account egocentric assessments of their household finances. This finding is significant because it demonstrates that Italians and Greeks are able to use information and experiences from their daily lives to evaluate European institutions, which are often believed to have complex and overlapping responsibilities that hinder the effective attribution of accountability. Future research should examine how multilevel governance in the EU affects public support and accountability attribution for EU institutions. Moreover, future investigations should study whether the associations reported here generalize beyond the cases we analyze, and more precisely to countries of the European north such as Germany and the Netherlands, which were differentially affected by the consecutive crises of economy, migration, and the pandemic.
Overall, the increasing immigration flows driven by the war in Ukraine and the ongoing climate crisis are likely to exacerbate public opposition to immigration and shape European politics in the immediate future. As international, ethnic, and social tensions become more acute, security threats will increase, and cultural divisions will deepen within European societies. Furthermore, with the advent of automation and artificial intelligence, immigration will put additional pressure on the European labor force, particularly on less-skilled workers who have precarious jobs in sectors disrupted by the digital revolution. It is, therefore, important for future research to investigate the consequences of immigration, international instability, automation, and climate change in the European job market, and how the structure and characteristics of the European economy may condition these consequences.