Public Opinion on the EU in the Aftermath of the Brexit Referendum

(1)

Public Opinion on the EU

in the Aftermath of the Brexit Referendum

Annette Hilzinger s1572016

Master Thesis

Public Administration: International & European Governance Leiden University

Supervisor: Dr. R. de Ruiter Date: 10 January 2020

(2)

Abstract

The European Union derives its legitimacy to a large extent from popular support by its inhabitants. This support has been tested on several occasions within the last decade. Arguably, crisis such as the Eurozone crisis and the migration crisis had significant effects on public per-ceptions of the Union and further European integration. Less research has been done on the impact of the UK EU membership referendum on public views on the EU. This research aims to fill the gap in the literature by examining how the Brexit referendum affected public opinion in the remaining EU member states. Furthermore, this research finds that public opinion on the EU has remained relatively stable at the aggregate level.

(3)

List of tables

Table Page

1 Overview of hypothesis, variables and indicators 38

2 Average views on the EU per member state and in the EU27 40

3 Determinants of public opinion on the EU at the aggregate level 42

(6)

1. Introduction

1.1 Public opinion on the European Union in the member states

The European Union (EU), just like any other democratic institution, relies on the sup-port of the citizens it governs. Public supsup-port is crucially imsup-portant for the democratic legiti-macy as well as the stability of the Union (Severs & Mattelaer, 2014). Findings of public opin-ion surveys are, therefore, relevant indicators of the performance of a political entity. Empiri-cally, public political opinion is said to be stable and resilient (Druckman & Leeper, 2012). However, past research suggests that public opinion can be influenced by major exogenous shocks such as an economic or political crisis (Bârgăoanu, Radu & Negrea-Busuioc, 2016; Debomy, 2016; European Commission, n.d.-a; Hobolt & Wratil, 2015; Taggart & Szczerbiak, 2018). While, within the context of the EU, much research has been done on the effect of the financial crisis that started in 2008 and, to a lesser extent, the migration crisis in 2015 on public support for European integration, hardly any scholar has examined how public opinion on the EU is influenced by the United Kingdom’s EU membership referendum (Brexit referendum) which was held in 2016.

This research investigates how the British decision to leave the European integration project affected public opinion in the remaining 27 member states. More specifically, this re-search will test how the relationships between public perception of the EU and socioeconomic as well as political factors that are known to shape pro-Europeanism have been impacted by the Brexit referendum.

It is expected that, overall, the outcome of the Brexit referendum had a strong positive effect on public opinion on the EU in the remaining member states. In other words, this research anticipates that existing biases towards the EU were reinforced. This means that factors that are already known to influence public views on the EU are expected to have an even bigger impact

(7)

on public opinion on the EU after the Brexit referendum. Furthermore, this research expects an increase in positive views on the EU at the aggregate level due to the insecurity and anticipated negative consequences of the UK’s withdrawal from the EU. This expectation is in line with findings by De Vries (2017) who claimed that support for the EU in the EU27 increased right after the British vote to leave the Union. Moreover, according to Bershidsky (2019) and the European Parliament (2018), recent results of the public opinion survey Eurobarometer confirm the positive trend in public opinion on the EU after the Brexit referendum in the remaining member states.

1.2 Brexit and the UK’s EU membership

On 23 June 2016 the British people voted to leave the European Union in a landmark referendum. One year before the referendum was held, the UK prime minister David Cameron from the Conservative Party called for the referendum due to the growing pressure from Euro-sceptic members of the British parliament and the public. The decision of the United Kingdom to leave the European Union, commonly referred to as ‘Brexit’, was made more than 40 years after the British public voted to affirm the new membership in the European Communities in 1975. The United Kingdom, consisting of England, Scotland, Wales and Northern Ireland, joined the European Economic Community (EEC) in 1973. Hence, the opposing Labour Party, which was against European integration, was defeated when two-thirds of the British public voted in favour of the membership.

However, one could argue that Britain has had a distinct relationship with the European supranational organisation all along. By opting out from the Schengen Area, the Economic and Monetary Union (EMU), the Area of Freedom, Security and Justice (AFSJ) and the European Charter of Fundamental Rights, the UK has held a special position within the Union. Several scholars have investigated this special relationship between the United Kingdom and the

(8)

European Union and referred to the UK as “awkward partner” of the EU (George as cited in Startin, 2015, p. 311). Startin (2015) identified three main drivers of opposition towards Euro-pean integration in the UK; the entering into force of the Maastricht Treaty, the EU Eastern enlargements in the early 2000s and the Eurozone crisis in 2008.

1.3 Euroscepticism

Opposition towards European integration and disapproval of the European Union is gener-ally referred to as Euroscepticism (Taggart, 1998). Taggart (1998) defined Euroscepticism as “the idea of contingent or qualified opposition, as well as incorporating outright and unqualified opposition to the process of European integration” (p. 366). The term is a British invention from the 1980s (Vasilopoulou, 2013).

Taggart and Szczerbiak (2002) distinguished between two kinds of Euroscepticism; hard and soft. Hard Euroscepticism refers to “a principled opposition to the EU and European inte-gration” (p. 4). Therefore, hard Eurosceptics oppose their countries’ membership as well as integration and the institution as a whole (Taggert & Szczerbiak, 2002). Soft Euroscepticism, on the other hand, refers to objections of EU policies and concerns about national sovereignty (Taggert & Szczerbiak, 2002). Other scholars differentiated between Euroenthusiasts, Euro-pragmatists, Eurosceptics and Eurorejects (Kopecky & Mudde, 2002). This research employs Taggart’s (1998) definition of Euroscepticism and uses the term to mean the opposite of support for European integration.

1.4 Euroscepticism in the EU27

While some may argue that opposition towards the European project is a peculiar British phenomenon, research and opinion polls suggest that Euroscepticism has been increasing in

(9)

other member states as well (Hobolt, 2016; Torreblanca & Leonard, 2013; BBC, 2016). Euro-scepticism is commonly found in the main political agenda of populist parties. Recent national parliamentary elections showed an upsurge in the support for populist right-wing parties in dif-ferent EU member states. Examples of this are the federal election in Germany in 2017 which resulted in the nationalist Alternative for Germany (AfD) being the third largest party in the German parliament or the French presidential election in the same year which gave rise to the populist party National Front (FN). Another example is the Austrian Freedom Party (FPÖ) which “became the only far-right party in power in Western Europe when it joined a coalition as junior partner with conservative Chancellor Sebastian Kurz in 2017” (BBC, 2019). Moreover, Euroscepticism is prevalent in Eastern Europe and Nordic countries as well, with anti-immi-gration parties winning significant shares of votes during parliamentary elections in Estonia in 2015, in Hungary and Sweden in 2018, and in Finland and Slovenia in 2019.

Potential drivers of such rising Eurosceptic sentiments have been identified as national-istic or conservative political ideology, low feeling of representation at the EU level, negative perceptions of immigration, negative perceptions of the national economic performance, strong feelings of national identity and low levels of education (Goodwin & Milazzo, 2015). Research on the Brexit referendum in the UK showed that those drivers correspond to the characteristics and concerns of those UK citizens who voted to leave the European Union (see for example Alabrese, Becker, Fetzer, & Novy, 2019). Therefore, some studies argued that Brexit “sets a precedent for leaving” the EU (De Vries, 2017, p. 39) and, thus, the rest of the EU feared that other (Eurosceptic) member states could follow the British example (Hobolt, 2016). However, citizens in the EU27 seem to have been largely unaffected by the Brexit vote (De Vries & Hoffmann, 2019). A study conducted on behalf of the Bertelsmann Stiftung in 2019 suggested that citizens of the remaining EU member states are not concerned about potential negative consequences for the EU after Brexit (De Vries & Hoffmann, 2019).

(10)

1.5 Previous research on public perceptions of the EU in the member states

Public support for European integration has been widely studied by European Union schol-ars. Arguably, the main trigger for research on this topic was the entering into force of the Maastricht Treaty in the early 1990s and the revelation of declining support towards further integration as a result of referenda held in Denmark and France (see for example Eichenberg & Dalton, 1993; Taggart, 1998; Franklin, Van der Eijk & Marsh, 1995). First research measured public support for the European Communities in rational utilitarian terms such as the economic performance of a country and individual benefits derived from the membership (see for example Eichenberg & Dalton, 1993; Anderson & Reichert, 1996; Carrubba, 1997; Gabel, 1998a). Other research claimed that the length of membership, education levels and political values are deter-minants of individual support for the European project (see for example Anderson & Kalten-thaler, 1996; Inglehart, 1970; Inglehart, 1977). In the early 2000s the focus of scholars shifted from explaining support to explaining opposition towards European integration (De Vries, 2018). Therefore, more recent studies have investigated how national attachment, political party affiliation and national political systems affect public opinion regarding further integration at the supranational European level (see for example Franklin, van der Eijk & Marsh, 1995; An-derson, 1998; Sanchez-Cuenca, 2000; Rohrschneider, 2002; De Vries, 2018).

1.6 Societal and academic relevance of this study

Studies of public opinion are important for several reasons. Public opinion research can reveal crucial insights into the legitimacy of a political institution. In general terms, legitimacy refers to the “popular acceptance of a government, political regime, or system of governance” (Blatter, n.d.). In the EU context, one can distinguish between input and output legitimacy which refer to “the EU’s responsiveness to citizen concerns” and effective policies, respectively (Schmidt, 2013, p. 2). A common way to measure the legitimacy and, thus, the performance

(11)

and stability of a political institution are public opinion surveys (Von Haldenwang, 2016). Pub-lic opinion surveys, such as the Eurobarometer, provide indications of popular support for the EU and its performance. Moreover, Williams (2012) suggests that public opinion acts as “a driving force in the process of integration” in the EU (p. 1).

As public opinion research showed, the entering into force of the Maastricht Treaty marked the first major turning point in popular support for the EU. Arguably, the Maastricht Treaty in the early 1990s can be regarded as the beginning of citizens’ concerns towards an ever more integrated Europe (Franklin, Marsh & McLaren, 1994; Franklin, Van der Eijk & Marsh, 1995; Taggart, 1998). According to previous research, public support for the European integration project declined significantly for the first time since its formation when the Maas-tricht Treaty was introduced (Taggart, 1998; Eichenberg & Dalton, 2007). Following this, the resilience of the EU has been tested on several occasions such as the rejection of a European-wide constitution in France and the Netherlands in 2005, the eurozone crisis which started in 2009, the migration crisis of 2015 and, most recently, the Brexit referendum in the UK in 2016.

Much research has been done on how big events or crises affect public opinion (see for example Debomy, 2016). In the European context, such research has mainly focused on changes in public support in the aftermath of the implementation of the Maastricht Treaty and the euro-zone crisis (Franklin, Marsh & McLaren, 1994; Franklin, Van der Eijk & Marsh, 1995; Eich-enberg & Dalton, 2007; Hobolt & Wratil, 2015). However, studies of the effect of the Brexit referendum on public perceptions of the EU are still scarce. While there is research examining public views on Brexit in the remaining member states (De Vries & Hoffmann, 2019), most scholarly research concerning the Brexit referendum focused either on public opinion within the UK (Vasilopoulou & Talving, 2019) or party-based Euroscepticism (Taggart & Szczerbiak, 2018). One exception to this is a study by De Vries (2017) who found that, directly after people in the UK decided to leave the EU, support for European integration increased in the EU27 as

(12)

a result of uncertainty. This study, by contrast, aims to fill this gap in the literature by investi-gating how public views on the EU changed two years after the referendum compared to two years before the referendum. In this way, the time frame will be considerably longer than De Vries’s (2017) study who only examined public opinion in 2016. Furthermore, this study high-lights differences in public views on the EU at the national level.

1.7 Research question and main findings

This study aims to answer the following research question: How did the result of the Brexit referendum affect public opinion on the EU in the remaining 27 EU member states? More precisely, this research investigates how the British decision to leave the European Union influenced the relationships between public perceptions of the EU and socioeconomic and po-litical factors that are already known to shape pro-Europeanism in the EU27. Moreover, this research hypothesises that the remaining member states re-evaluate their own EU membership in light of the Brexit vote and, as a result, previous biases are reinforced. Therefore, it is ex-pected that the relationships between public opinion on the EU and its predictors will be strengthened after the referendum. This research assumes that Brexit, which is often referred to as a “crisis” (see for example Taggart & Szczerbiak, 2018), has the potential to influence es-tablished opinions. It is expected that Brexit highlighted people’s concerns about the EU and made EU citizens, who were previously not aware, more conscious about such concerns.

In order to test the hypothesis outlined above, this research evaluates Eurobarometer data, by means of regression analyses, before and after the Brexit referendum in the EU27. Furthermore, this research investigates whether factors that are known to influence public opin-ion on the EU have become weaker or stronger predictors of public views on the EU in 2018 compared to 2014.

(13)

The results show that attitudes towards foreigners, government support and citizens’ value of national sovereignty are strong predictors of public perceptions of the EU in the re-maining member states. However, no significant changes after the Brexit referendum could be found at the aggregate level after the referendum. At the national level, hostilities towards for-eigners predict public opinion on the EU in 18 out of the 27 remaining member states. This association became significantly stronger in Denmark and Poland. The relationship between government support and public view on the EU became moderately stronger in West Germany and Luxembourg. Similarly, citizens’ satisfaction with the domestic political system became a somewhat stronger predictor of public opinion on the EU in Portugal, Austria, the Czech Re-public and Estonia after the Brexit referendum. By contrast, citizens’ value of sovereignty be-came a weaker predictor of public opinion on the EU in all member states that showed signifi-cant results and, most notably, in Ireland and Austria. Economic factors, on the other hand, do not appear to influence public perceptions of the European integration project substantially in any of the remaining EU member states.

This paper proceeds as follows. First, established theories regarding public support for European integration are discussed in order to outline the theoretical foundation of this research. Subsequently, the methodological approach of this study is introduced. This is followed by the presentation of the results. Subsequently, the results are interpreted, and their implications are discussed. This paper concludes by answering the research question and making suggestions for further research.

(14)

2. Theoretical Framework

The following chapter gives an overview of seminal research on determinants of support for European integration. Subsequently, four different categories of theories that explain public support for European integration are distinguished and explained. Each section concludes by stating hypotheses based on the respective theory cluster. It is important to note here that, due to data availability, this research will analyse general perceptions of the EU rather than public support for European integration. However, as the subsequent methodology chapter shows, both indicators are similar enough to each other as substitutes. Therefore, this research relies on theories about public support for European integration to formulate hypotheses and define relevant predictors.

2.1 Early studies on determinants of public support for European integration

Scholarly interest in the study of public support for European integration is almost as old as the European project itself. While previous research was concerned with the support for European integration, more recent studies seek to explain growing popular opposition, that is, the increase in so-called Euroscepticism that has been observed in current years (De Vries, 2018). The most significant early work on public attitudes towards the European Communities appeared in the early 1970s (see Inglehart, 1970, 1977; Inglehart & Rabier, 1978). The prevail-ing theories back then were Inglehart’s (1970, 1977) ideas of cognitive mobilisation and value priorities. Inglehart (1970) claimed that education is conducive to support of European integra-tion. In other words, the higher an “individual’s capacity to receive and interpret messages re-lating to a remote political community”, the more likely it is that this individual will be in favour of European integration (p. 47). Inglehart (1970, 1977) called this cognitive mobilisation. As a consequence, individuals who are highly educated and “exposed to cosmopolitan

(15)

communications” (p. 70) have better access to information about European integration and are, therefore, more likely to be supportive of it. This is reinforced by a positive image of European integration produced by mass media, politicians and universities in Western Europe (Inglehart, 1970). A second argument by Inglehart (1977; Inglehart & Rabier, 1978) concerns the political value orientation of individuals. In his book ‘The Silent Revolution’, Inglehart (1977) distin-guishes between Materialist and Post-Materialist values and argues that these values are shifting from the former to the latter. Whereas Materialist values include existential needs such as phys-ical well-being and economic stability, Post-Materialists prioritise more abstract values such as freedom of speech and environmental protection (Inglehart, 1977). As a result, “the Post-Ma-terialists are more likely to take an interest in remote and abstract causes – European integration being one of many possibilities – and to place relatively heavy emphasis on a sense of European solidarity (…)” (Inglehart & Rabier, 1978, p. 87). Even though, Inglehart’s (1970, 1977) the-ories have been contested by other later scholars (see for example Janssen, 1991; Anderson & Reichert, 1996; Gabel, 1998b), his work is considered a cornerstone in the study of public opin-ion towards European integratopin-ion.

2.2 Attention to drivers of popular support for the European integration project

The public debate surrounding the signing and entering into force of the Maastricht Treaty in the early 1990s accompanied with referenda in Denmark, France and Ireland triggered a renewed scholarly interest into the drivers of positive and negative attitudes towards European integration (see for example Eichenberg & Dalton, 1993; Taggart, 1998; Franklin, Van der Eijk & Marsh, 1995). The Maastricht Treaty, also referred to as the Treaty on European Union (TEU), was an important milestone in the history of European integration as it fundamentally changed the setup of the supranational institution. Most importantly, the Maastricht Treaty con-ferred new competences in the area of Common Foreign and Security Policy and Justice and

(16)

Home Affairs to the European Communities. According to Franklin, Marsh and McLaren (1994), “one of the main arguments for the TEU was that economic union could not advance any further without some form of political union” (p. 455).

While it was previously believed that there is a ‘permissive consensus’ towards further European integration (Lindberg & Scheingold, 1970), the Maastricht Treaty proved to be the first indicator of Euroscepticism, which refers to the rejection of further integration at the Eu-ropean level (Taggert, 1998). As research has shown, public support for EuEu-ropean integration declined dramatically after the Maastricht Treaty had been signed (Eichenberg & Dalton, 2007; Anderson & Kaltenthaler, 1996). Furthermore, referenda in France and Denmark highlighted the discontent of the public with the reform of the European Communities. While the Danes rejected the treaty in a first referendum in 1992, France approved the TEU only with a very narrow margin. Clear support for the Maastricht Treaty could only be seen in the Irish referen-dum result. Consequently, the post-Maastricht period marked the beginning of a new theoretical debate about the drivers of support of European integration as well as the drivers of Euroscep-ticism.

2.3 Public support for European integration after the Maastricht Treaty

Research in the early 1990s focused mainly on economic factors as well as the length of membership in the European Communities to explain public support for European integra-tion. Eichenberg and Dalton (1993) found that national economic performance plays a crucial role in shaping public views regarding the European Union besides various other factors. More-over, Eichenberg and Dalton (2007) claimed that the main reason for declining support and growing concerns about European integration was the establishment of the Economic and Mon-etary Union (EMU) which was accomplished with the entering into force of the Treaty on Eu-ropean Union. It is further argued that the reason for this are concerns about the integration of

(17)

social security policies which would possibly lead to a decline of the welfare state (Eichenberg & Dalton, 2007).

These findings are in contrast with research conducted by Anderson and Kaltenthaler (1996) who argued that the variance in public support for integration in Europe is mainly due to the length of membership and the time period in which a country joined the European Com-munities rather than economic conditions only. This confirmed earlier research which found that countries which joined the European Communities earlier are generally more supportive of it (Inglehart & Rabier, 1978; Anderson & Kaltenthaler, 1996).

Other research in the 1990s focused specifically on the Maastricht Treaty referenda that were held in France, Denmark and Ireland to explain Euroscepticism (Franklin, Marsh & McLaren, 1994; Franklin, Van der Eijk & Marsh, 1995). Franklin, Marsh and McLaren (1994) studied the different referendum outcomes in Denmark, France and Ireland regarding the rati-fication of the Treaty on European Union. As mentioned above, the three referenda resulted in very different outcomes and signified popular opposition towards European integration in two member states for the first time. The study revealed that the “voters were (…) influenced by national politics” as well as political party positions (Franklin, Marsh & McLaren, 1994, p. 469). Similar research suggested that the outcomes of the Maastricht Treaty referenda mirrored the acceptance of the incumbent governments (Franklin, Van der Eijk & Marsh, 1995). Acknowl-edging the crucial role of party positions, Franklin, Van der Eijk and Marsh (1995) claimed that the position of the parties in power are especially important in shaping public opinion towards European integration. As they argued, the major opposition towards the TEU in Denmark and France was due to the “unpopularity of [the] ruling parties in both countries” while the popular government in Ireland led to a positive vote (Franklin, Van der Eijk & Marsh, 1995, p. 101).

(18)

2.4 Four main drivers of public opinion on the EU

Following these insights concerning the influence of the Maastricht reform that would later lead to an even more integrated European Union, Gabel (1998b) distinguished and tested what he considered the five most important theories on public support for European integration. These theories include Inglehart’s (1970, 1977) theories of cognitive mobilisation and political values, utilitarian explanations as well as partisanship and approval of the national government (Gabel, 1998b). Gabel (1998b) concluded that utilitarian explanations as well as party affilia-tion are the most convincing theories to explain differences in public support while the scholar disputed Inglehart’s (1977) theory of a ‘silent revolution’.

As the scholarly interest in the subject of public support towards European integration and the EU progressed, scholars generally identified three broad categories of theories that ex-plain support for European integration and Euroscepticism, respectively. These theories include economic utilitarianism, national identity and political party cues (see for example Hooghe & Marks, 2005; De Vreese & Boomgaarden, 2005; De Vries & Edwards, 2009; Hobolt & De Vries, 2016; Hobolt, 2016). However, it could be argued that there is a fourth category, namely national political context, which refers to EU citizens’ evaluation of the functioning of the na-tional political system and its autonomy (De Vries, 2017, 2018). In the remainder of this chapter, each of the four approaches will be discussed in more detail.

2.4.1 Economic Utilitarianism

One of the oldest theoretical and still prominent approaches towards public support for European integration concerns utilitarian considerations. Proponents of the utilitarian idea posit that support for European unification can be measured in terms of a rational cost-benefit anal-ysis (Gabel 1998b, De Vries 2018). This explanation seems logical since the European project started as a purely economic union.

(19)

One of the first studies using this approach was carried out by Eichenberg and Dalton in 1993. The scholars’ analysis of public support for European integration over a time period of 20 years revealed that economic conditions play an important role in shaping citizens’ atti-tudes (Eichenberg & Dalton, 1993). The main argument put forward is that if European coun-tries perform well economically, due to the creation of the Single Market and trade liberalisation, then support for European integration will increase in those countries (Eichenberg & Dalton, 1993). This can be explained by the acknowledgement of economic benefits that can be derived from being a member of the European Communities. In line with this argument, Carrubba (1997) suggested that financial transfers function as “side-payments to further the integration process” (p. 469). According to the scholar, financial transfers are used to convince member states that are less enthusiastic about European unification to agree on further integration (Car-rubba, 1997). The improved economic condition of those countries which receive such pay-ments will, in turn, lead to a more positive perception of integration by their citizens (Carrubba 1997).

Another stream of research, which relies on utilitarian reasoning, focused at the individ-ual level and how individindivid-ual benefits influence public opinion on unification (Anderson & Reichert, 1996; Gabel & Palmer, 1995; Gabel, 1998a). For example, Anderson and Reichert (1996) claimed that personal economic benefits foster positive views on European integration. Moreover, the same study found that the trade volume of a country and indirectly received individual benefits are also positively correlated with support for European integration (Ander-son & Reichert, 1996). Furthermore, in their seminal work, Gabel and Palmer (as cited in Gable, 1998b) argued that individual experiences impact perceptions of European integration and lib-eralisation differently. Their main argument was that human capital (education level and pro-fessional skills) is closely related to an individual’s perception of integration (Gabel, 1998b). The liberalisation of the labour market that allows people to move freely across borders to work,

(20)

benefits those with higher human capital due to the increased competition. Consequently, peo-ple with higher education levels and more professional skills would be more supportive of fur-ther integration in Europe (Gabel, 1998b).

In sum, scholars who employ utilitarian arguments assume that positive perceptions about the economy lead to positive views about the EU, and vice versa. Based on the utilitarian theories above, this research expects the following:

Hypothesis 1a: The more positive an inhabitant of an EU member state perceives the national economic performance, the more positively he or she views the EU.

This research suggests that the Brexit referendum had a significant impact on public opinion on the EU in the EU27 for the following two reasons. First, Brexit can be regarded a political crisis for both, the UK and the EU (see for example Taggart & Szczerbiak, 2018; McHarg, 2018; Caporaso, 2018). As such, the outcome of the referendum has the potential to influence public opinion dramatically (Debomy, 2016). Research showed that political and eco-nomic crisis such as the entering into force of the Maastricht Treaty or the Eurozone crisis had a profound impact on popular support for European integration (see for example Franklin, Marsh & McLaren, 1994; Hobolt & Wratil, 2015). This research considers the British decision to withdraw from the European integration project as a comparable exogenous shock and, there-fore, proposes that public views on the EU have been affected as a result of the UK’s referen-dum on its European Union membership.

Secondly, this research argues that the media coverage surrounding the Brexit referen-dum raised awareness of certain concern of the general public related to the European Union. An extensive body of literature has examined how public opinion is influenced by media fram-ings of certain issues (see for example Gavin, 2018). It is widely argued that the Eurosceptic stance of the UK tabloid press has an enormous influence on negative perceptions of the EU

(21)

within the United Kingdom (Daddow, 2012). Furthermore, with regard to the British vote on its EU membership, previous studies suggested that media outlets had a substantial impact in shaping people’s views on the EU within the UK and, therefore, on the outcome of the referen-dum (Berry, 2016). This research applies the same hypothesis to the remaining 27 member states. Media coverage on the Brexit referendum in both, the UK and the EU27, mainly focused on four issues; the economy, political views, immigration and sovereignty (Media coverage of the EU referendum, 2016; Reuters Institute, n.d.). Considering the economic prospect of the UK, media as well as experts drew a grim picture, suggesting that the UK economy will face tremendous negative consequences as a result of Brexit (Begg & Mushövel, n.d.). Due to this projection, this research expects that inhabitants of the remaining EU member states make a stronger positive connection between their economic prosperity and the EU. Consequently, the second expectation with regard to economic utilitarianism is the following:

Hypothesis 1b: The association in hypothesis 1a remained positive and has become stronger after the Brexit referendum.

The above hypotheses are tested by means of Eurobarometer survey data of two waves; one before and one after the Brexit referendum. Furthermore, perceptions about the economic performance are measured with the following indicators: Perceptions of the national economy and personal economic circumstances. Public opinion, on the other hand, is measured with a question which asks whether respondents have a positive or negative image of the EU. The methodological setup of this study is further elaborated in the following chapter.

2.4.2 National Identity

In contrast with the utilitarian theories outlined above, the identity approach emphasises “people’s attachment to their nation and their perception of people from other cultures” as key

(22)

drivers of Euroscepticism (De Vries, 2018, p. 15). For example, Carey (2002) argued that an individual’s sense of belonging to his or her country of origin is positively associated with Euroscepticism. In other words, the more attached citizens feel towards their nation, the more sceptical they are with regard to European unification (Carey, 2002). Similar research by Hooghe and Marks (2005) confirmed this finding. The scholars distinguished between inclusive and exclusive national identities and suggested that people who exclusively identify themselves with their own country are more likely to oppose the idea of an integrated Europe (Hooghe & Marks, 2005). The reason for this is that “European integration makes it more difficult for na-tional governments to pursue distinctly nana-tional preferences, it undermines nana-tional self-deter-mination and blurs boundaries between distinct national communities” (Hooghe & Marks, 2005, p. 423).

In line with social identity theory, McLaren (2002) suggested that the main reason for Euroscepticism is a “perceived cultural threat” (p. 551). According to her research findings, opposition towards European integration is caused by factors related to national identity rather than economic benefits (McLaren, 2002). As McLaren (2002) argued, concerns regarding the preservation of one’s nation state, culture and national resources, in light of increasingly blurred national boundaries, are key drivers of Eurosceptic sentiments. Similarly, a study conducted by De Vreese and Boomgaarden (2005) found that opposition towards European integration stems from negative preconceptions of people that do not share the same national identity. Further-more, this research contended that Euroscepticism is strongly related to the hostility towards immigrants (De Vreese & Boomgaarden, 2005). Hence, the second stream of arguments asso-ciated with the identity approach suggests that the absence of support for European integration has less to do with economic considerations but rather with intrinsic feelings of threat and cul-tural biases. With regard to the findings outlined above, this research expects the following:

(23)

Hypothesis 2a: The less an inhabitant of an EU member state associates him- or herself with his/her own nationhood, the more positively he/she views the EU.

Again, this research puts forward two arguments as to why the Brexit referendum is expected to have changed public views on the EU in the remaining member states. The first argument is the same that has been outlined above. In general terms, this research posits that the outcome of the Brexit referendum had an effect on public opinion on the EU due to its nature of being a political crisis. Secondly, this research expects that EU citizens’ association with their own nationhood has become a stronger predictor of public views on the EU after the Brexit referendum. Immigration was an intensely debated subject ahead of the referendum. A huge influx of immigrants in previous years due to the migration crisis combined with pre-existing Eurosceptic sentiments paved the way for an anti-immigration campaign led by Brex-iters (Garrett, 2019). This research proposes that inhabitants of an EU member state, who are hostile towards foreigners, have been made more aware of the connection between immigration and the EU. Therefore, this research expects the following changes:

Hypothesis 2b: The association in hypothesis 2a remained negative and has become stronger after the Brexit referendum.

In order to test the above hypotheses, data from the Eurobarometer 81.2 and Euroba-rometer 90.3, which were conducted in 2014 and 2018, respectively, have been analysed. More-over, EU inhabitants’ association with their own nationhood has been measured with the fol-lowing two variables: Attachment towards the nation state and hostility towards foreigners. The following chapter outlines the methodological approach of this study in more detail.

(24)

2.4.3 Political views and government support

The third main approach towards support for European integration or Euroscepticism, respectively, is concerned with the role of political parties and how their position on the Euro-pean project affects public opinion. An extensive body of research has examined the influence of partisanship, party affiliation and elite cues on public attitudes towards European unification (De Vries & Edwards, 2009; Ray, 2003; Steenbergen, Edwards & De Vries, 2007 ). First re-search using this line of argument put forward that the European unification process is viewed more sceptically by Communist or Left parties than other parties (Inglehart & Reif, 1991; Gabel, 1998b). De Vries and Edwards (2009), on the other hand, contended that there is no such division. Instead, the scholars found that both, parties on the left and on the right, object further integration in Europe (De Vries & Edwards, 2009). According to the scholars, parties on the left of the political spectrum are motivated by economic concerns and right-leaning parties worry about the loss of national sovereignty and identity (De Vries & Edwards, 2009). By con-trast, Franklin, Marsh and McLaren (1994), who investigated Euroscepticism in light of the Maastricht Treaty referenda in France, Denmark and Ireland, suggested that political parties do not hold uniform positions regarding European integration. Instead, the scholars suggested that there are opposing opinions among supporters of a particular party (Franklin, Marsh & McLaren, 1994). Moreover, the scholars argued that the negative vote in Denmark and the slim majority in France were due to the influence of powerful party elites (Franklin, Marsh & McLaren, 1994). Furthermore, Franklin, Van der Eijk and Marsh (1995) found that the approval of national gov-ernments largely influence public attitudes towards the unification of Europe in their analysis of the Maastricht Treaty referenda outcomes in Ireland, Denmark and France. Adding to these studies, Hooghe and Marks (2005) claimed that “[t]he more national elites are divided, the more citizens are cued to oppose European integration (…)” (p. 419).

(25)

Other research on party cues confirmed that public attitudes are affected by party elites and their view towards European unification (Ray, 2003). However, as Ray (2003) argued, “this effect varies with levels of disagreements among parties, party unity, issue salience, and party attachment” (p. 978). More recent research, on the other hand, suggested that this effect goes both ways (Steenbergen, Edwards & De Vries, 2007). A study on the referenda on a European Constitution in the Netherlands and France in 2005, which both rejected the Constitution, found that voters are not only influenced by party cues, but public opinions also impact party positions (Steenbergen, Edwards & De Vries, 2007).

Based on the above insights regarding political aspects that influence Euroscepticism, this research expects the following:

Hypothesis 3a: The more an inhabitant of an EU member state supports the incumbent national government, the more positively he/she views the EU, regardless of their polit-ical ideology.

As outlined above, this research suggests that the outcome of the Brexit referendum had an impact on public opinion in the EU27 based on two arguments. First, the referendum can be regarded an exogeneous shock which has the potential to influence public opinion (Debomy, 2016). Secondly, the Brexit referendum emphasised the deep divisions in UK politics (Ford & Goodwin, 2017). This supports the argument made by Hooghe and Marks (2005) who claimed that citizens of a member state are more likely to be Eurosceptic when their national leaders are divided on this subject. On the contrary, Taggart (1998) contended that big and well-established parties are generally supportive of European integration, whereas less successful parties express opposition towards it in order to dissociate themselves from mainstream parties. This argument is in line with Hobolt (2016) who stated that the majority of traditional political parties in EU member states are supportive of the European integration project. Based on this notion, this

(26)

research suggests that the link between government support and perceptions of the EU has strengthened after the UK referendum.

In line with other hypotheses, Eurobarometer before and after the Brexit referendum has been analysed to test the above hypotheses. Government support was measured with the fol-lowing indicators: Trust in the national government and political ideology.

2.4.4 National politics as proxy

As mentioned earlier in this chapter, scholars have identified the above approaches, namely economic utilitarianism, national identity and political party cues, as the main theories to explain support for as well as opposition towards European integration (Hooghe & Marks, 2005; De Vreese & Boomgaarden, 2005; De Vries & Edwards, 2009; Hobolt, 2016). However, as hinted at by De Vries (2017, 2018), one could argue that there is a fourth, distinct stream of research that attempts to explain public opinion on European integration. Scholars who are as-sociated with this theory cluster are concerned with the distinct national political context that impacts public opinion in the individual member states. A first stream of research in this group-ing suggested that European citizens evaluate the European integration project based on domes-tic polidomes-tical considerations (Franklin, Van der Eijk & Marsh, 1995; Anderson, 1998; Sánchez-Cuenca, 2000). Anderson (1998) claimed that national politics, and more precisely the func-tioning of democratic institutions, serve as a ‘proxy’ for the assessment of the European project. More specifically, Anderson (1998) argued that high levels of approval of the domestic political system correlate positively with support for European integration. By contrast, Sánchez-Cuenca (2000) contended that the reverse is true. According to the scholar, high levels of approval of

(27)

the domestic political system are associated with lower levels of support for integration, whereas support for the supranational institution increases as popularity on the national level declines (Sánchez-Cuenca, 2000). The reason for this is the low “opportunity cost of transfer-ring sovereignty to Europe” (Sánchez-Cuenca, 2000, p. 147). The underlying assumption of this argument is that poor assessments of the national system are the result of corruption and democratic deficits. Therefore, it is assumed that transferring sovereignty to the European level will benefit the citizens of such member states (Sánchez-Cuenca, 2000). Other research, again, suggested that rather than the national performance, it is the performance at the European level and, more precisely its representational performance, that shapes attitudes towards integration (Rohrschneider, 2002). As Rohrschneider (2002) claimed, approval of the European project decreases when people are concerned about its democratic legitimacy.

The most recent approach within this theory cluster, considers the national context as determinant of public opinion towards European integration (De Vries, 2018). In her bench-mark theory, De Vries (2018) suggested that attitudes towards the European project depend on how people compare their current situation as an EU member against a supposed alternative state of being a non-member. As the scholar argued, people are more optimistic about leaving the European Union when their country is doing well in political and economic terms, and vice versa (De Vries, 2018). The reason for this is that “people attribute responsibility for these conditions, good or bad, primarily to national governments rather than to the EU” (De Vries, 2018, p. 206).

Based on research evaluating support for integration in the EU with regard to the na-tional context, I expect the following:

Hypothesis 4a: The more satisfied an inhabitant of an EU member state is with the domestic political system, the more positively he/she views the EU.

(28)

As outlined for previous hypotheses, this research expects a change in public opinion on the EU in the remaining member states based on two arguments. The first argument concerns the potential of the Brexit vote to influence public views as it is a political crisis and political crises are known to have such effects (Debomy, 2016). The second argument, on the other hand, relates to the specific relationship between a national political system and public perceptions of the Union. Another major argument put forward by Vote Leave campaigners in the UK was the gradual surrender of national sovereignty to Brussels (The Economist, 2016). This research suggests that ordinary citizens within the EU27 have been made aware of this connection through the Brexit debate. In light of previous research which suggested that media coverage reinforces existing attitudes (Gavin, 2018), this research expects the following:

As mentioned above, Eurobarometer before and after the Brexit referendum was ana-lysed to test the hypotheses. Satisfaction with the national political system was measured with the following indicators: Approval of the functioning of the national democratic system and value of national sovereignty.

In sum, this chapter has outlined the theoretical foundation of this study. Furthermore, four main theory clusters that explain popular support for the EU have been identified; eco-nomic utilitarianism, national identity, political views and government support, and approval of the national political system. Based on these theories, this chapter argued that the Brexit referendum had a significant impact on public opinion on the EU in the remaining member states. The following chapter describes the methodological approach of this research. More precisely, it specifies which indicators were used to measure the various variables in the hy-potheses, how the data was collected and how it was analysed.

(29)

3. Methodology

The previous chapter summarised seminal work on public opinion on the EU and de-fined four main drivers of popular support for the European integration project; economic util-itarianism, national identity, approval of the domestic government and endorsement of the way the political system works at the national level. This chapter sets out the methodological ap-proach of this study. First, the research question is explained together with the definition of the research problem. Subsequently, this chapter describes how the data was collected and analysed. This is followed by a detailed explanation of which variables were used to test the hypotheses and how these variables were measured.

3.1 Research question and problem definition

This research investigates the effect of the referendum in the United Kingdom on its European Union membership in 2016 on the relationship between citizen’s views on the EU and factors that influence such views. As outlined in the previous chapter, scholars interested in the study of public opinion on European integration have identified various drivers of public support for the EU. The aim of this study is to examine whether the Brexit referendum had an impact on the relationships between the four main predictors outlined above and public opinion on the EU. In doing so, this study tested existing theories on support for the European Union before and after the Brexit referendum. Based on the previous literature and data availability, the following variables were included in the analysis:

Variable 1: EU inhabitants’ perception of the national economy

Variable 2: EU inhabitants’ personal job situation

(30)

Variable 4: EU inhabitants’ hostility towards foreigners

Variable 5: EU inhabitants’ approval of the national government

Variable 6: EU inhabitants’ political ideology

Variable 7: EU inhabitants’ satisfaction with the national democracy

Variable 8: EU inhabitants’ value of national sovereignty

The variables listed above were assessed independently from each other. Table 1 shows how these variables are measured and how they relate to the hypotheses.

The Brexit referendum paved the way for the first withdrawal of a member state from the European Union – with the exception of Greenland’s departure from the EU in 1985. However, it is expected that the UK’s exit from the Union will have much more far-reaching consequences compared to Greenland’s withdrawal for a number of reasons. First of all, the European Com-munities have developed tremendously over the past 30 years. While the main purpose of the ECC in the 1980s was economic cooperation, the EU has evolved into an established and deeply interconnected political entity. Secondly, Greenland is not a sovereign state but belongs to Den-mark. Thirdly, and perhaps most importantly, the UK is far bigger in terms of its population and its economy. Therefore, the impact of the UK’s withdrawal from the EU is expected to be unprecedented for both, the UK and the EU. The purpose of this research is to shed a light on the impact of the UK’s departure on public opinion on the EU in other EU member states. In contrast to recent research on this topic which has focused on public opinion towards the EU within the UK (Carl, Dennison & Evans, 2018), this research reveals how the British decision to leave the Union affected public opinion in the remaining 27 member states.

(31)

3.2 Data collection and analysis

Data for this study have been taken from the following Eurobarometer surveys: Euro-barometer 81.2 (conducted in March 2014) and EuroEuro-barometer 90.3 (conducted in November 2018). The use of existing European-wide surveys is a common method in this field of research due to the large sample size and the ease of data comparability. As De Vreese and Boomgaarden (2005) noted, “all previous research on support for European integration has relied on Euroba-rometer data” (p. 67).

The Eurobarometer is a European-wide survey which measures public opinion in all EU member states since 1974. The Standard Eurobarometer is carried out twice a year by means of interviews in person (European Commission, n.d.-b). The surveys include questions regarding citizens’ attitudes towards the EU institutions, policies and further integration as well as other topics such as environmental, social, health or technological issues. The sample size for each country is roughly 1,000 with the exceptions of Luxembourg, Cyprus and Malta, which have about 500 respondents each (GESIS Leibniz Institute for the Social Sciences, 2017). Moreover, samples for East and West Germany as well as Northern Ireland and the rest of the UK are taken individually (GESIS Leibniz Institute for the Social Sciences, 2017). In recent years, some Eurobarometer surveys have included non-EU member states and candidate countries such as Turkey, Cyprus TCC, Macedonia, Montenegro, Serbia and Albania. These non-EU countries have been excluded from this research.

Critics of the Eurobarometer point out that the data cannot be used for panel studies since the respondents vary from survey to survey and from year to year (Carrubba as cited in Steenbergen, Edwards & De Vries, 2007, p. 22). Besides this, many questions have been mod-ified or omitted over the years which poses a challenge towards comprehensive long-term stud-ies of public support for European integration (Hobolt & De Vrstud-ies, 2016). Furthermore, schol-ars have criticised that the survey questions are biased towards support for integration (Höpner

(32)

& Jurczyk, 2015). Despite such concerns, most scholars in this research area rely on the Euro-barometer surveys (see for example Eichenberg & Dalton, 1993; Anderson & Reichert, 1996; McLaren, 2002; Carey, 2002; De Vries & Edwards, 2009; Rohrschneider, 2002; Sanchez-Cuenca, 2000).

Much of the seminal work conducted on public attitudes towards the European project is based on Eurobarometer data. The reason for this is that the data facilitates the assessment of differences between countries (Hobolt & De Vries, 2016). In fact, according to Hobolt and De Vries (2016), “Eurobarometer surveys (…) constitute the only data source that allows for cross-national and longitudinal comparisons” (p. 416). For this reason, this research will employ Eu-robarometer data for its analysis. Moreover, the respondent selection and the sample size of the Eurobarometer suggest that the data is representative of the EU population. Respondents are chosen based on “a multi-stage, random (probability) sampling design” (GESIS Leibniz Insti-tute for the Social Sciences, 2017, p. VII) and amount to a total of approximately 28.000 re-spondents per survey (GESIS Leibniz Institute for the Social Sciences, 2017).

This study is a quantitative study which was designed following a deductive research approach. The Eurobarometer data for this research were accessed through the data catalogue of the GESIS Leibniz Institute for the Social Sciences. This research compared public opinion in the remaining 27 EU member states of two points in time – one before and one after the Brexit referendum on 23 June 2016. The time frame is about 4.5 years comparing public views 27 months before and 29 months after the referendum on the EU membership of the UK. The time span of roughly two years before and after the referendum has been chosen on the follow-ing grounds: This time frame allows to identify results as effects may not be visible immediately after an event took place. However, it is not possible to use a bigger time frame than this because the referendum was held too recently. The corresponding surveys are Eurobarometer 81.2 (con-ducted in March 2014) and Eurobarometer 90.3 (con(con-ducted in November 2018). These surveys

(33)

have been chosen based on data availability. As outlined above, some Eurobarometer questions are not asked in every round. The selected surveys were the only ones in the respective time period that include all relevant indicators necessary to test the hypotheses in this study.

The Eurobarometer data was analysed using a multiple linear regression model in the statistical software IBM SPSS Statistics 25. Separate analyses were conducted for each of the two surveys. Non-EU member states, the UK (samples for Great Britain and Northern Ireland) as well as respondents who answered “don’t know” or “refusal” to any of the relevant survey questions were excluded from the analysis. After this, the sample size shrank from original 27,932 to 15,616 respondents in 2014 and from 32,600 to 13,967 respondents for the year 2018. Furthermore, the data was tested for multicollinearity (see appendix A) and binary logistic re-gressions were conducted to test the robustness of the model (see appendix D).1

Prior to the main analyses, mean values of the dependent variable were assessed and compared to illustrate general trends in opinions on the EU in different member states. For this, mean values of the dependent variable of the Eurobarometer 86.2 (conducted in November 2016) was included for a more comprehensive comparison. Subsequently, regression analyses were conducted to examine the relationships between the dependent variable and the predictors by analysing the EU27 as a whole. In a second round of multiple linear regression analyses, each country was assessed individually by using the split file function in SPSS.

3.3 Operationalisation

The dependent variable in this research is ‘public opinion on the EU’. This dependent variable is slightly different from ‘public support for European integration’ commonly used by

1_{The dependent variable was transformed into a dichotomous variable; very positive and positive responses}

(34)

other scholars. Past research commonly measures support for European integration in terms of citizens’ support for their country’s membership of the EU. The corresponding Eurobarometer survey questions is:

‘Generally speaking, do you think that [OUR COUNTRY]’s membership of the EU is a good or bad thing or neither?’ (GESIS Leibniz Institute for the Social Sciences, n.d.). However, this question was not available in the relevant surveys used for this research. There-fore, it was substituted with the following indicator which comes closest to the one above:

‘In general, does the EU conjure up for you a very positive, fairly positive, neutral, fairly negative or very negative image?’ (Eurobarometer 81.2; Eurobarometer 90.3)

The above question is measured on a scale from 1 (very positive) to 5 (very negative). Hence, the smaller the value is, the more positive public opinion on the EU. The underlying assumption is that people who have a positive view of the EU are supportive of it, whereas a negative opinion indicates less support. This assumption is justified by the relatively high pos-itive correlation between the membership and the image question (rs = 0.5). The response cat-egories for the dependent variable are normally distributed (see appendix B).

The indicators to measure the independent variables have been principally based on pre-vious research. Table 1 provides an overview of the variables and the indicators used in this study as well as the corresponding hypotheses. In this research the influence of the national economy on citizens’ views on the EU is measured with the following question:

‘How would you judge the situation in each of the following? – The situation of the [NATIONALITY] economy’ (Eurobarometer 81.2; Eurobarometer 90.3).

(35)

A similar indicator has been used by Anderson (1998) in his research on support for European integration in relation to approval of the national government and democracy. An-derson (1998) used the following indicator to measure the economic evaluation of respondents:

‘Compared to 12 months ago, do you think that the general economic situation in this country is [a lot better – a lot worse]?’

Similarly, a question regarding respondents’ personal job situation was used by De Vries and Edwards (2009) and Rohrschneider (2002) to measure economic anxiety. This study will use the following indicator to measure personal economic circumstances:

‘How would you judge the situation in each of the following? – Your personal job situ-ation’ (Eurobarometer 81.2; Eurobarometer 90.3)

Based on utilitarian theories which suggest that positive evaluations about the national economy are positively associated with views on the EU, this research expects that after the Brexit refer-endum, this positive relationship has become stronger.

Previous research argued that there is a negative correlation between people’s opinion on the EU and their association with their own nationhood (Carey, 2002). This research expects that after the Brexit referendum this link intensified. Citizens’ association with their nationhood is measured with two variables; attachment towards the nation state and hostility towards for-eigners. Attachment towards the nation has been commonly measured with the Eurobarometer question ‘Do you see yourself as…?’ to which respondents can choose from the following op-tions: (1) [NATIONALITY] only, (2) [NATIONALITY] and European, (3) European and [NA-TIONALITY] , (4) European only, (5) None (see for example Carey, 2002; De Vries & Ed-wards, 2009). The exact same question was used for this analysis. Hostility towards foreigners, on the other hand, has been previously measured with a question which asks whether or not respondents are afraid of ‘the loss of our national identity and culture’ (Carey, 2002: McLaren,

(36)

2004). This question has been amended to ‘What does the EU mean to you personally?’ with one of the options to choose from being ‘loss of our cultural identity’. Respondents who chose this option were coded as 1 and respondents who did not choose this answer category were coded as 0.

The third hypothesis concerns people’s support of their national government and their political ideology. It is expected that, after the Brexit referendum, perceptions towards the EU are influenced more strongly by the degree of support for the current government, regardless of whether the ruling party is liberal or conservative. Government support was measured with the following question:

‘For each of the following media and institutions, please tell me if you tend to trust it or tend not to trust it. – The [NATIONALITY] Government’ (Eurobarometer 81.2; Euro-barometer 90.3)

This indicator is closely related to the hypothesis established by Sanchez-Cuenca (2000) who suggested that public support for the EU increases when citizens disapprove their national institutions. Another variable to test this hypothesis is political ideology. In this research, polit-ical ideology is associated with the politpolit-ical left-right analysis. Respondents of the Eurobarom-eter were asked to indicate their political bent on a scale from 1 (left) to 10 (right). Research interested in the impact of party positions and party attachment on public support for European integration have used similar questions in their analyses (see for example Ray, 2003).

The final expectation in this study regards people’s evaluation of their national political system and the degree of their country’s sovereignty in relation to their opinion about the EU. The following indicators have been used to assess this relationship. The first indicator is:

(37)

‘On the whole, are you very satisfied, fairly satisfied, not very satisfied or not at all satisfied with the way democracy works in [OUR COUNTRY]?’ (Eurobarometer 81.2; Eurobarometer 90.3)

The exact same question has been previously used by Anderson (1998) to establish the link between peoples’ satisfaction with the national democratic system and support for Euro-pean integration. Finally, the relationship between citizens’ value of sovereignty and their view on the EU was assessed. Citizens’ value of sovereignty was measured with the following indi-cator:

‘Please tell me to what extent you agree or disagree with each of the following state-ments: ‘[OUR COUNTRY] could better face the future outside the EU’ (Eurobarometer 81.2; Eurobarometer 90.3)

This question corresponds to De Vries’s (2018) benchmark theory outlined in the previous chapter.

More recent surveys such as the Eurobarometer 90.1 (conducted in September 2018) have started to ask questions directly related to the Brexit referendum. The following are ex-amples of these questions.

‘If a referendum was held tomorrow regarding [OUR COUNTRY’S] membership of the EU, how would you vote?’ (Eurobarometer 90.1)

‘Would you say that Brexit is a good or a bad thing for the European Union?’ (Euroba-rometer 90.1)

‘From today’s perspective, would you say the British people made the right choice in voting to leave the EU?’ (Eurobarometer 90.1)

These Brexit-related questions allow for a more precise analysis of the impact of the UK’s EU membership referendum on public opinion in the rest of the EU. However, since these

(38)

questions have only been asked after the referendum took place, they could not be included in the main analysis of this research.

(39)

Overview of hypotheses, variables and indicators

Hypothesis Variable Indicator

Public opinion on the EU*

In general, does the EU conjure up for you a very positive, fairly positive, neutral, fairly negative or very negative image?

Very positive (1) – very negative (5)

H1a: The more positive an inhabitant of an EU

member state perceives the national economic per-formance, the more positively he or she views the EU.

H1b: The association in hypothesis 1a remained

positive and has become stronger after the Brexit referendum.

Perceptions of the na-tional economy

How would you judge the situation in each of the following: The situation of the (NA-TIONALITY) economy?

Very good (1) - very bad (4) Personal economic

cir-cumstances

How would you judge the situation of each of the following: Your personal job situa-tion?

Very good (1) – very bad (4)

H2a: The less an inhabitant of an EU member state

associates him- or herself with his/her own nation-hood, the more positively he/she views the EU.

negative and has become stronger after the Brexit referendum.

Attachment towards the nation

Do you see yourself as …?

Nationality only (1) – Nationality and European (2) – European and Nationality (3) – European only (4) – None (5) Hostility towards

foreign-ers

What does the EU mean to you personally? - Loss of our cultural identity Not mentioned (0) – mentioned (1)

H3a: The more an inhabitant of an EU member state

supports the incumbent national government, the more positively he/she views the EU, regardless of their political ideology.

positive and has become stronger after the Brexit referendum.

Government support For each of the following media and institutions, please tell me if you tend to trust it or

tend not to trust it: The (NATIONAL) Government

Tend to trust (1) – tend not to trust (2)

Political ideology In political matters people talk of “the left” and “the right”. Thinking about your

views, how would you place yourself on this scale?

Left (1) – right (10)

H4a: The more satisfied an inhabitant of an EU

member state is with the domestic political system, the more positively he/she views the EU.

positive and has become stronger after the Brexit referendum.

Approval of the national political system

On the whole, are you very satisfied, fairly satisfied, not very satisfied or not at all sat-isfied with the way democracy works in (OUR COUNTRY)?

Very satisfied (1) – not at all satisfied (4) Value of national

sover-eignty

Please tell me to what extent you agree or disagree with each of the following state-ments: (YOUR_COUNTRY) could better face the future outside the EU.

Totally agree (1) – totally disagree (4)

Public Opinion on the EU in the Aftermath of the Brexit Referendum