Is Euroscepticism Contagious?

• 1 Euroscepticism in the Split Belgian Party System

  In Belgium, pro-EU sentiments have long been dominant across most of the party landscape (Franck, Leclercq & Vandevievere, 2003; Randour & Bursens, 2019). One recent example is the 2020 federal coalition agreement, signed by social democrats, liberals, greens and Flemish Christian democrats, which expresses explicit support for deeper European integration (Bodson, 2020). However, in recent years, Euroscepticism has made an entry into some pockets of the political spectrum, especially on the fringes of the landscape such as the far left and far right (Pittoors, Wolfs, Van Hecke & Bursens, 2016). This raises the question whether this awakening Euroscepticism has triggered a response of the other parties and to what extent. Did mainstream parties start to devote more attention to the EU over the first two decades of the 21st century (from 2000 until 2019), and to what extent have their pro-integration positions endured? This article discusses the effects of Eurosceptic challengers on other parties’ EU salience and EU positions. Given that Belgium’s party landscape is split along linguistic lines, we address this question in a comparative design. On the basis of the assumption that parliaments provide parties with both an arena for party competition and a forum for communication (Auel & Raunio, 2014), we use data on contributions of members of parliament (MPs) in the plenary debates of the Belgian federal and regional parliaments.
  We start with a presentation of the relevant literature on Euroscepticism, followed by a presentation of the Belgian party system. After a methodological section, we present our main findings. We find that when facing Eurosceptics, other parties raise salience of the EU fairly equally, while government parties and parties further from the political centre were found to adopt (soft) Euroscepticism more often than other parties. In the conclusion we revisit the literature in the light of our findings, discuss the article’s limitations and present some suggestions for future research.

• 2 Euroscepticism: Scope, Salience and Positions

  2.1 The Scope of Euroscepticism

  Academic literature is rife with definitions of Euroscepticism. It is most often used as a “generic, catch-all term encapsulating a disparate bundle of attitudes opposed to European integration in general and opposition to the EU in particular” (Szczerbiak & Taggart, 2008: 239). Definitions differ significantly, but usually make a distinction between principled opposition to the idea of European integration, on the one hand, and opposition to the policies and practices of the EU, on the other. Such a framework was first advanced and later refined by Taggart and Szczerbiak (2002; Szczerbiak & Taggart, 2008), who dubbed these positions ‘hard’ and ‘soft’ Euroscepticism respectively. A prominent criticism of this approach is that it ignores the rich variety of opinions (Bijsmans, 2017; Kopecký & Mudde, 2002). Multiple authors have proposed more specific classifications. Kopecký and Mudde (2002) identify four categories based on a distinction between diffuse and specific support or opposition, and De Wilde, Michailidou and Trenz (2014) consider even more positions based on support for or opposition against the principal question of integration, the specific institutional arrangement and the goals of further integration. Indeed, Euroscepticism can be based on varying grounds, such as national sovereignty, communitarianism, identitarianism, populism or opposition to neoliberal policy, leading to either diffuse or specific opposition (Carlotti & Gianfreda, 2020; De Vries & Edwards, 2009; Pirone, 2020; Vezzani, 2020).
  While more detailed classifications are suitable for the analysis of positions based on sufficiently detailed sources, they are not necessarily useful for the classification and analysis of mediated debates or isolated statements in parliaments (Bijsmans, 2017). This challenge is expressed in empirical research on Euroscepticism, which despite the many efforts at theoretical sophistication, often sticks to more basic classifications. For example, although Statham and Trenz (2013) provide clear illustrations of more specific types of Euroscepticism, they must rely on a simpler classification for their main empirical contributions: a distinction between Eurocritical, Eurosupportive and neutral claims.
  Out of many classifications, each study must equip one fit for its purposes. The difficulty associated with more sophisticated classifications is even more pressing for the sometimes rather short contributions made by MPs; these do not necessarily reveal the complete and detailed perspective on European integration, which is necessary for classification in more sophisticated models. For this reason, we do not use more specific categories than hard and soft Euroscepticism, despite the wide variety these terms hold. We concede that the loss in variety is a drawback, but since this article aims to uncover the spread of Euroscepticism and not to analyse its character, this drawback is less problematic than the alternative.
  Although there are exceptions, Euroscepticism has long been seen as a phenomenon limited to peripheral or extreme parties on either side of the political spectrum, which use it as a strategy to differentiate themselves from usually pro-EU mainstream parties (Hooghe, Marks & Wilson, 2002; König, Marbach & Osnabrügge, 2017; Taggart, 1998). In recent years, however, a growing set of parties across Europe have voiced Eurosceptic arguments. Some authors even see a new axis of party competition around the EU issue (Giannetti, Pedrazzani & Pinto, 2017; Karlsson & Persson, 2020), introduced by parties that feel they are losing battles on other fronts (Hobolt & De Vries, 2015) in an environment where issue competition is increasingly important (Green-Pedersen, 2007).
  Party positions related to the EU and the salience of EU positions are influenced by political constellations, in general (Giannetti et al., 2017; Taggart, 1998), and by Eurosceptic challengers, in particular (Meijers, 2017; Meijers & Williams, 2020; Treib, 2020; Van de Wardt, 2015). Such peripheral Eurosceptic challengers can be fairly mainstream in terms of socio-economic cleavages, but by positioning themselves on the extreme end of the EU dimension, they become the main drivers of opposition to the EU (Karlsson & Persson, 2020), triggering response from other parties.
  This article investigates the effect of Eurosceptic challengers on mainstream parties’ positions towards the EU. We look at both hard and soft Euroscepticism in the multilevel Belgian political party system. We first discuss how much attention parties devote to the EU dimension (salience) and then turn to the type of arguments they present (position).

  2.2 Salience of the EU

  In many countries, EU salience is traditionally low, for example because of a high level of consensual support for the EU or because other (often domestic) issues grab attention (Bursens, 2002; Wolfs & Van Hecke, 2020). At the same time, increasing European integration is assumed to lead to a growing Europeanisation of national policy debates as well as of EU-related party competition, especially as a reaction to crises (Hutter & Kriesi, 2019; Serricchio, Tsakatika & Quaglia, 2013). National political parties respond in multiple ways, such as by adapting their electoral platforms and actively engaging in competition with other parties, with respect to both EU policies and the EU itself (Ladrech, 2002). One can thus expect the salience of the EU to grow as European integration deepens.
  The theory of issue ownership argues that mainstream parties would simply choose to avoid addressing the EU altogether, even in a context of growing Euroscepticism, because they feel they would be fighting a losing battle on enemy turf (Petrocik, 1996). However, some studies indicate that mainstream parties do react by devoting more attention to the EU (Giannetti et al., 2017; Meijers, 2017) but that this effect might be contingent on the salience Eurosceptic challengers give to the EU. Hence, our first expectation:

  E1: When competing electorally with surging Eurosceptics, other parties will give increasing salience to the EU dimension, even more so if Eurosceptics give increasing attention to the EU dimension themselves.

  At the same time, mainstream parties may show varying degrees of salience to the EU. For example, Dardanelli (2012) points to an interaction between positions on the EU and on regionalisation: pro-EU regionalist parties may grant salience to the EU dimension as they favour a ‘Europe of the Regions’. Moreover, Meijers (2017) argues that since parties do not always compete for the same voters, EU salience will increase most if parties are in direct and intense electoral competition with Eurosceptics. In addition, parties may refrain from taking explicit positions on the EU if they fear dividing their electoral base on the issue. Furthermore, Meijers argues that centre-left parties are more often triggered to take up the EU issue than centre-right parties as the former compete with both the far right and the far left (on cultural and economic issues, respectively), whereas centre-right parties only compete with the far right. In other words, if Euroscepticism comes from the left, only centre-left parties will show increased EU salience; if Euroscepticism comes from the right, both centre-left and centre-right parties have incentives to increase the salience of the EU dimension. Hence, we expect the following:

  E2: The salience given to the EU dimension will increase most for pro-EU regionalist parties and parties in intense electoral competition with a Eurosceptic party.

  2.3 Positions Toward the EU

  Meijers (2017) finds that Eurosceptic rhetoric increases among parties that must compete with Eurosceptics. However, if support for the EU remains rather high among most parties, they could respond not by voicing hard Eurosceptic arguments, but by increasingly making arguments in favour of a specific kind of European project, including calls for a more democratic/social/efficient EU, tailored to their own party platforms (Franck et al., 2003; Treib, 2020). Importantly, Meijers (2017) argued that the effect of Eurosceptic parties is strongly qualified by the degree of their EU salience. This leads to the third expectation:

  E3: When facing increasing Eurosceptic competition, mainstream parties adopt (soft/hard) Euroscepticism, even more so if Eurosceptics’ EU salience is high.

  In addition, Van de Wardt (2015) found that opposition parties and extreme parties are most likely to co-opt Euroscepticism as they are less risk averse and will thus more easily take the risk of changing position on the EU. On the other hand, Hobolt and De Vries (2015) find that taking up new issues is not primarily a function of opposition, but of being on the losing side of the dominant dimension of contestation. We expect the following:

  E4: Opposition parties are more likely to take Eurosceptic positions than government parties and are even more likely to do so when they are located further from the political centre.

  Finally, Meijers (2017) found that Eurosceptic challengers have, above all, differential effects on mainstream parties. Since centre-left parties are in competition with both the far right and the far left (on the cultural and economic fronts, respectively), they are more likely to adopt Euroscepticism than centre-right mainstream parties, which only compete with the far right. This is in line with recent findings that Euroscepticism across the political spectrum comes in different shapes and sizes, based in communitarianism, identitarianism, populism or opposition to neoliberal policy (Carlotti & Gianfreda, 2020; De Vries & Edwards, 2009; Pirone, 2020; Vezzani, 2020). Hence our final expectation:

  E5: With Eurosceptics in parliament, centre-left parties are more likely to adopt Eurosceptic argumentation in parliament than centre-right parties.

• 3 Methodology

  3.1 Case Selection and Data

  This article makes a comparative evaluation of the effect of Eurosceptic challengers on two dimensions of Euroscepticism (salience and positions) among other parties, using parties’ parliamentary groups in several Belgian parliaments as units of analysis. These groups fulfil a communicative role as they are expected to voice opinions on the EU in the plenary with the goal of reaching the public (Auel & Raunio, 2014). In addition, parliaments are a prime locus of strategic party competition, which is a key mechanism in the proliferation of Eurosceptic initiators’ effects as well as a key locus of opposition to the EU (Karlsson & Persson, 2020). Data was compiled by coding parliamentary statements by MPs during plenary sessions of the various parliaments.
  Belgium is an excellent case to examine the effect of Eurosceptic parties on party competition. First, the long-standing pro-EU consensus has recently been challenged by parties at the far ends of the political spectrum (Wolfs & Van Hecke, 2020). Secondly, the federal structure of Belgium enables a comparative approach discussing several party systems and related parliaments. Reflecting the centrifugal dynamic of Belgian federalism, the Belgian electoral system is split into two separate party systems. Previously unitary parties split along linguistic lines (although some still cooperate intensively with their sister parties across the language border, the Flemish and French-speaking Green parties being the main example here). With the exception of the Brussels Capital Region and the German community, political parties operate mostly within their respective language communities, even when they compete in federal parliament elections (Meier & Bursens, 2020). Flemish parties do not compete for the support of Walloon voters and vice versa. As a result, even sister parties can develop divergent strategies and platforms depending on the dynamics of party competition within their respective constituencies.
  This setting allows us to compare the regional party systems and, because the federal parliament is elected within two separate party systems, we have an additional assembly to assess whether the electoral or the parliamentary context matters most. In short, the Belgian setting enables us to assess whether Eurosceptic parties affect other parties through electoral competition or simply through their mere presence in parliament without competing for the same electorate. Tables 1 and 2 in Appendix A list Francophone- and Dutch-speaking parties, respectively. Only parties that managed to win a seat at some point between 2000 and 2019 are listed and considered in the analysis. Independent MPs were scarce and are not included either. From these parties, only Parti du Travail de Belgique/Partij van de Arbeid van België (PTB/PVDA) maintains a national structure.
  Additionally, most Belgian parties have traditionally been very pro-EU (Dardanelli, 2012; Franck et al., 2003). EU discussions have long been rare: Belgian politics was Europeanised only to a surprisingly low degree (Bursens, 2002). Recently, however, Eurosceptic arguments have gained some momentum, starting with fringe parties (Randour & Bursens, 2019), while all parties have increasingly included EU positions in one way or another on their party platforms (Pittoors et al., 2016; Randour & Bursens, 2019). Vlaams Belang (VB) is the biggest Belgian party to openly call itself Eurosceptic (Vlaams Belang, 2018). However, we are not as much interested in which parties may or may not be fundamentally Eurosceptic as in the introduction of hard Euroscepticism in the political debate, regardless of the parties’ official ideology. Also, while arguments to change the EU (soft Euroscepticism) are more common and can even be a sign of pro-European sentiment, hard Euroscepticism, evoking a principled opposition to European integration, is more unequivocal.
  To investigate how mainstream parties react to such Euroscepticism, we identified the parties that were the first to introduce this in each of the parliaments under scrutiny. Our data, presented graphically below and in tables in Appendix B, show that multiple Belgian parties fit this profile. In the federal parliament, there are two distinct sparks of Eurosceptic positioning and therefore of potential Eurosceptic proliferation, once in 2008 by VB and Lijst Dedecker (LDD) and a second time in 2015 by the Parti Populaire (PP) and the unitary PTB/PVDA. In the Flemish parliament, VB introduced Euroscepticism in 2002, and its position has oscillated since. In the French-speaking and Walloon parliaments, the only introduction of Euroscepticism, by PTB/PVDA in 2015, was very short-lived. The Belgian data confirm König et al’s (2017) finding that Eurosceptic initiators are extreme parties at both ends of the political spectrum and are in line with analyses of parties’ platforms (Wolfs & Van Hecke, 2020).
  Since support for Eurosceptic initiators is higher in Flanders than in Wallonia (Randour & Bursens, 2019), we can assess the effect on other parties by comparing the behaviour in the respective regional parliaments. Flemish mainstream parties are more often confronted with intense electoral competition from Eurosceptic challengers. Walloon parties, however, also face the Flemish Eurosceptic parties in the federal parliament, even though they do not compete with each other for federal seats (except in the Brussels Capital constituency). This institutional setting allows us to assess whether effects of Eurosceptic initiators spread through electoral competition or additionally through parliamentary interaction.
  To ensure comparability with the Flemish parliament, the scores for the parliaments of the French Community and the Walloon Region are combined. Finally, Belgium is an example of centrifugal federalism, in which communitarian tensions occurred simultaneously with the deepening of European integration (Beyers & Bursens, 2013). This allows for an assessment of the competing hypotheses of Europeanisation and federalism. Our design does not control for other variables that may affect whether and in which direction mainstream parties develop positions towards European integration, and, therefore, we only posit plausibility claims. Our empirical focus includes the Belgian federal parliament, the Flemish parliament, the parliament of the French Community, and the Walloon parliament. A selection of plenary meetings from 2000 until 2019 was analysed.
  Plenary documents were collected through each parliament’s websites (De Kamer van volksvertegenwoordigers, 2020; Parlement de la Fédération Wallonie-Bruxelles, 2020; Parlement de Wallonie, 2020; Vlaams Parlement, 2020). We took two months, April and November, in each year between 2000 and 2019, avoiding election campaigns and parliamentary recesses. Selecting the same months each year keeps our data stable over the years. The selection includes variation on the share of Eurosceptic initiators over time and between parliaments. Relevant mediating variables such as the party type (mainstream vs. peripheral, government vs. opposition, left vs. right, degree of regionalism) are automatically included at the party level. Each time MPs took the floor, their intervention was counted as one intervention by their party. Based on the content of the intervention, interventions were coded as arguments ‘pro-EU’, ‘change-EU’, ‘anti-EU’ or ‘no-EU’ (when the EU was not discussed). The result is an estimation of the salience each party gives to the EU and its EU position, for every year in each parliament. To facilitate comparison, we combined scores for the Walloon parliament and the parliament of the French-speaking region, since on the Flemish side community and region are merged in a single assembly.

  3.2 Measuring Euroscepticism

  Party Euroscepticism was operationalised as a continuous variable, by measuring the share of party interventions in parliaments that include Eurosceptic arguments. This allows us to see an increase or decrease in parties’ EU salience and positions. We do not establish whether parties are fundamentally Eurosceptic, or whether parties adopt Eurosceptic rhetoric because of ideological or strategic imperatives. This is not relevant for our purpose as in either case Euroscepticism is communicated in an arena of political competition. We operationalise the distinction between hard and soft Euroscepticism as arguments ‘anti-EU’ and ‘change-EU’, respectively. We also measure arguments made in favour of the EU, categorised as ‘pro-EU’ arguments. For the independent variable, we include two types of behaviour from challenger parties in the variable ‘increase of Euroscepticism’: either a shift towards Euroscepticism in terms of content or an increase of EU salience by challenger parties are expected to trigger a response from mainstream parties.

• 4 Results

  We first discuss the evolution of salience and positions in each of the parliaments based on two types of graphs. The first graph on salience shows parties’ proportion of plenary contributions that include arguments about the EU (rounded to the nearest whole number). The second type of graph shows the positions of each party, outlining the share of each party’s contributions that argued pro-EU, change-EU or anti-EU. Null scores are omitted everywhere, and years when a party had no seats in the given parliament are shaded.
  The vertical lines in the graphs indicate a rise of Euroscepticism, which can mean either a position shift towards more Euroscepticism or an increase in EU salience from Eurosceptic challengers. The lines are dotted if the Eurosceptic challenger in question competes in another regional electoral district. We find that a mere positional shift towards Euroscepticism affects only some other parties, while a subsequent rise in EU salience has a more widespread effect. As vertical axes differ across parliaments, we refer to Appendix B with parties’ scores for comparisons between parliaments.

  4.1 Federal Parliament

  EU salience among Flemish parties in the federal parliament.
  
  EU salience among Walloon parties in the federal parliament. For better legibility, one high score was omitted: Défi’s 2000 pro-EU score of 50.
  

  In the federal parliament, an increase in Eurosceptic positioning occurs in 2008 and 2015-2016. In 2008 both challenging parties were Flemish (VB and LDD). While Groen and the Walloon social democrats were the first to raise attention for the EU after this initial introduction of Euroscepticism, it was only after the initiators also raised EU salience (in 2011, 2014 and 2018) that other parties eventually started devoting more attention to the EU. In 2011, the responding parties were all Flemish traditional parties together with the Walloon liberals, and in 2014, when VB’s EU salience was very high, these were joined by the Flemish regionalists. Similarly, when VB increased EU salience shortly after 2018 again to a high level, we see a rise in EU salience across all mainstream parties except the Flemish Christian democrats, who were already at very high levels of EU salience (Figures 1 and 2).
  In terms of positions, the Flemish greens show substantial changes in their position in 2008 as they start to argue more in favour of changing the EU. Both social democrat parties adopt some change-EU rhetoric but quickly return to their previously established, fairly consistently pro-EU courses. The Walloon liberals briefly introduced a small number of ‘change-EU’ arguments after 2008. After the 2011 increase in Eurosceptic EU salience, argumentation to change the EU increased for the Flemish liberals, both Christian democrat parties, the Walloon greens, and, eventually, also the Flemish regionalists. In 2018, when VB increased EU salience once again, the Flemish regionalists even introduced anti-EU argumentation, while no other parties reacted.
  In 2015-2016, the Eurosceptic initiator was PTB/PVDA, which competes in the entire country but has most success in the Walloon/francophone part, and the one-seat PP. Both parties’ Euroscepticism was paired with high EU salience, but PTB/PVDA soon softened its position, voicing only pro-EU arguments by 2018, and PP only argued about the EU in 2015-2016. Only the greens and Walloon Christian democrats increase EU salience in reaction to the PTB/PVDA and PP challenge. Also in terms of positions, there were few effects. The Flemish greens included ‘change-EU’ rhetoric in 2015 and 2016. After PTB/PVDA softened its position, the greens made no more ‘change-EU’ arguments. Nieuw-Vlaamse Alliantie (N-VA) adopted more change-EU arguments, but they were not electorally threatened by either initiator, and their shift outlasted the Eurosceptic challenge, which had disappeared by 2018. In that year, N-VA even voiced anti-EU arguments for the first time (Figures 3 and 4).

  EU positions among Flemish parties in federal parliament.
  
  EU positions among Walloon parties in federal parliament. For better legibility, one high score was omitted: Défi’s 2000 pro-EU score of 50.
  

  4.2 Flemish Parliament

  In the Flemish parliament, VB initiated Eurosceptic rhetoric in 2002, while between 2006 and 2012 it argued both in favour and against the EU. In 2012, VB voiced more unequivocal and salient Euroscepticism, and since 2016 it has run an exclusively anti-EU course. Ramping up Euroscepticism position-wise mostly correlates with an increase in EU salience (Figure 5).

  Although most of them had seen periods of increased EU salience before, the liberals, greens and Flemish regionalists showed a modest increase of EU salience in or shortly after both 2012 and 2016. The social democrats increased EU salience in 2012, but not in 2016, while the Christian democrats did not in 2012 but did in 2016.
  In and shortly after 2012, the Christian democrats, the Flemish regionalists, the liberals, the social democrats and the greens adapted their rhetoric to include change-EU arguments (although for the greens and social democrats such rhetoric had already been voiced a couple times before). In and shortly after 2016, when Eurosceptics became exclusively anti-EU but with a low salience, the liberals and Christian democrats remained pro-EU. The greens, Flemish regionalists, and social democrats, all shifted towards Eurocriticism, with the latter two even including anti-EU arguments (Figure 6).

  Parties’ EU salience in Flemish parliament.
  
  Parties’ EU positions in Flemish parliament.
  

  4.3 Walloon and Francophone Parliaments

  Parties’ EU salience in the Walloon and Francophone parliaments. One high score is omitted for legibility: PTB had an EU salience score of 25 in 2015.
  

  In the Walloon and francophone parliaments, PTB briefly introduced a Eurosceptic challenge in 2015, combined with a very high EU salience. However, its anti-EU stance disappeared quickly: in the subsequent years PTB made no arguments about the EU, except in 2019, when it argued to merely change the EU. All parties in the Walloon and francophone parliaments increased attention to the EU in 2015 and again in 2016, but for most parties this is simply a return to previous levels after particularly low levels in 2014. In 2016, the Walloon parliament blocked the ratification of CETA (Comprehensive Economic and Trade Agreement), but the crisis peaked in October (a month not selected for analysis) and can therefore not explain the increasing EU salience already in 2015. In terms of positions, parties mostly maintained their pre-established courses; changes were very limited. Parti Socialiste (PS) briefly shifted to more change-EU argumentation, while the other parties, if anything, became more pro-EU (Figures 7 and 8).

  Parties’ EU positions in the Walloon and French-speaking parliaments.
  

• 5 Discussion

  5.1 Salience: The Eurosceptic Effect Across the Board

  In almost all assemblies, parties devoted more attention to the EU after 2009 than they did before 2009. The only exception is the Flemish parliament, where salience remained mostly constant. At the same time, Flemish parties did raise salience in the federal parliament, indicating that the EU is increasingly on the radar across the linguistic border. Despite the ongoing and even increasing communitarian tensions, which were seen as impeding EU salience in the past (Bursens, 2002), more recent years do not show any decrease of EU salience. Moreover, the federal parliament, the prime locus of communitarian disputes, shows a bigger increase in salience than the regional parliaments. Although all Belgian government levels are affected by EU policies, the federal level seems to be the most popular place to discuss EU matters. In sum, as European integration deepened, Belgian domestic politics has become more Europeanised, confirming Ladrech’s expectations (2002) and in line with scholars who point to the subsequent financial and refugee crises as explanation for the increased salience (Hutter & Kriesi, 2019; Serricchio et al., 2013).
  In addition, our data suggests an effect of the more manifest Euroscepticism from peripheral political parties. While Groen and the Walloon social democrats were the first to raise attention for the EU after an initial introduction of Euroscepticism (in 2008), it is only after the Eurosceptics raise EU salience (first in 2011 and with a spike in 2014), that all parties eventually end up devoting more attention to the EU. Similarly, when VB increased Eurosceptic salience in 2018, all mainstream parties increased EU salience, except for Christen-Democratisch and Vlaams (CD&V), which was already at a high level. However, Eurosceptic competition might not always have been the cause, as some had already shown oscillations in EU salience before (cdH and Écolo). The very brief Eurosceptic episode of PTB/PVDA in the federal parliament had only a limited effect, as just cdH and Écolo, which had been showing oscillating levels of EU salience before, increased their EU salience.
  At the Flemish level, the response to Eurosceptic competition is characterised by an increase in EU salience for three out of five parties (N-VA, CD&V, Groen), while Open VLD and sp.a seem unaffected in terms of salience. In the Walloon and French-speaking regional parliaments, all parties increased EU salience in 2015-2016 after PTB/PVDA had a short episode of Eurosceptic competition. This finding seems to support the idea that the limited impact of PTB/PVDA Euroscepticism might be explained by the fact that VB had already made parties adapt to Eurosceptic competition. At the French-speaking/Walloon regional level, where VB had not yet forced parties to react to Eurosceptic competition, PTB/PVDA does affect the other parties.
  In short, our first expectation seems to be supported, in line with Meijers (2017) and Giannetti et al. (2017). Although we must be cautious as our design does not control for alternative explanations, it does provide an impetus to reject the theoretical alternative, namely that mainstream parties do not raise attention for the EU when facing an increase in Eurosceptic competition, either in terms of salience or in terms of positions.
  Our second expectation aimed to examine Dardanelli’s (2012) argument with respect to pro-EU regionalist parties and Meijers’ (2017) argument with respect to the centre left. Neither is confirmed. When VB first introduced Eurosceptic competition at the federal level in 2008, its closest competitors were N-VA (the then new, generally pro-EU regionalist party that reintroduced an alternative for voters focused on Flemish-communitarian issues), and Open VLD (the other main right-wing party that in the previous election of 2004 had lost many voters while VB had gained many). However, N-VA was the very last to increase EU salience. Open VLD only noticeably increased EU salience when Eurosceptics made the EU more salient. However, so did other parties involved in less intense competition with VB, such as Groen and CD&V. Similarly, at the Flemish level, Open VLD and N-VA did not increase their attention for the EU more than other parties when confronted with VB’s Euroscepticism.
  As for the French-speaking assemblies, the parties in most intense electoral competition with PTB/PVDA are PS and Écolo. At the federal level, Écolo did indeed increase the salience it gives to the EU in reaction to PTB/PVDA, but so did cdH and Défi. Moreover, PS even decreases the salience it gives to the EU at the federal level. At the regional level, all parties react similarly to PTB/PVDA Euroscepticism in terms of EU salience.
  In short, our analysis does not confirm the expectations of Dardanelli (2012) and Meijers (2017). At least during periods without campaigns, parties in closer electoral competition with Eurosceptics give no more attention to the EU than other parties. Instead, EU salience increased across the board, albeit in varying degrees, not more so for parties in intense competition with Eurosceptics. In other words, when EU salience is raised by Eurosceptics, it affects all other parties’ salience. The Flemish-regionalist response, contrary to Dardanelli’s hypothesis that they would raise salience especially quickly, could be interpreted in light of the cosmopolitan-communitarian divide, as the Flemish-communitarian regionalists have incentives to oppose the more cosmopolitan EU.

  5.2 Positions: The Condition of Salience, Government and Distance From the Centre

  At the federal level, expectation 3 predicts that parties start criticising the EU more from 2008 onwards, after VB became vocally anti-EU. In the federal parliament, the introduction of Eurosceptic competition introduced a shift in one party as Groen started arguing to change the EU. However, it is only when the Eurosceptic challengers also increase EU salience that other parties also start changing their positions and formulating arguments to change the EU. They adapted even more when VB had become unambiguously Eurosceptic. This condition of salience was theorised by Meijers (2017) for all parties. At times we find social democratic parties showing behaviour that does not follow the trend: while in the Flemish parliament, social democrats did shift their position to even include anti-EU argumentation when VB became unambiguously anti-EU, in all other cases, the position shifts of social democratic parties are extremely limited, short-lived or even non-existent. Our findings for the Flemish parliament are similar: only the liberal party clearly does not react to the VB Euroscepticism (while the social democrats, who showed no reaction at the federal level, do react at the Flemish level). PTB/PVDA’s brief period of Euroscepticism had almost no effect as changes in position by other parties were very limited. Finally, in the Walloon and French-speaking parliament, PS barely and only briefly shifts to change-EU arguments after PTB had briefly introduced Eurosceptic opinions. The other parties even became more vocally pro-EU. In short, our findings are a qualified confirmation of the work of Franck et al. (2003), Meijers (2017) and Treib (2020): some, but not all mainstream parties, follow Eurosceptic challengers in their negative evaluation of the EU. Moreover, at least in Belgium, Euroscepticism might have a limited shelf-life as a political tool of competition: while early on, most parties shifted their positions on the EU, the last time VB increased EU salience, only its closest competitor changed its position.
  The fourth expectation, which focuses on opposition parties and parties furthest from the centre, is only partially confirmed. At the federal level, Groen, an opposition party at the left end of the continuum, was the first to react to Euroscepticism, before other parties, which changed positions only after Eurosceptic salience increased. However, it seems Groen’s position further from the centre, rather than its opposition status, explains this, as the other parties that changed their EU position were most often government parties. Most parties whose EU argumentation was unaffected were opposition parties: the Flemish nationalists and social democrats were in the federal opposition at the time, and neither changed their EU position. N-VA became more Eurosceptic only after spikes of Eurosceptic salience in 2014, and when VB had become unambiguously Eurosceptic in 2017. At this point, N-VA was no longer in the opposition. At the Flemish level, a similar dynamic was found: the liberals, the only party that was unaffected by the introduction of Eurosceptic competition, was in the opposition. The greens were also in the opposition. Although the extent to which Groen’s change in EU argumentation was caused by Eurosceptic competition is unclear (since it already showed oscillations before), it seems to be rather connected to their position further from the centre than to their oppositional status.
  In sum, in line with Van de Wardt (2015), we found that being further away from the centre makes parties more likely to react to Eurosceptic competition by including arguments to change the EU. With respect to opposition versus government status, however, we found an evolution contrary to Van de Wardt (2015) who theorised that opposition status increases the likelihood of adopting Eurosceptic arguments, as well as to Meijers, (2017) who hypothesised that oppositional status is unimportant. Our analysis shows that opposition parties are less likely than government parties to criticise the EU when faced with Eurosceptic competition. Our result may be explained by Hobolt and De Vries’ (2015) finding that it is not oppositional status, but rather being on the losing side of the dominant dimension of contestation that affects Euroscepticism. Alternatively, it may also be an effect of two-level game dynamics (Putnam, 1988), whereby government parties can use the EU as a scapegoat for unpopular policies and portray it as a constraining actor for domestic governance. Both situations would imply a critical position towards the EU. Opposition parties, on the other hand, can use the EU to criticise the governing coalition (e.g. by citing critical EU reports or government shortcomings in complying with European requirements), which requires more leniency in criticising the EU. Statham and Trenz (2013) and Brack and Startin (2015) show how criticism of the EU was also a substantive part of the French socialists during Hollande’s term as president.
  Our final expectation, regarding centre-left parties following Meijers (2017), is not confirmed. At the federal level, only one of the four centre-right parties did not change its EU argumentation, while none of the centre-left parties did so significantly. At the Flemish level, the only centre-left party adapted its argumentation, but so did two out of three centre-right parties. In the French-speaking and Walloon parliaments, PS briefly increased argumentation to change the EU, but this change is limited. The other francophone parties (one left, one centre left and one centre-right) increased their pro-EU argumentation. The limited effect in these parliaments is possibly due to the negligible amount of Eurosceptic competition. In short, two scenarios seem plausible: either centre-right parties are more likely to change their position on the EU or centre-right and centre-left parties are equally likely to do so. Meijers’ (2017) logic underpinning this hypothesis is only partly supported by our findings: parties that are in competition with Eurosceptics might be more likely to shift argumentation, but at least in this case, these are not more often centre-left parties that supposedly compete with both the far right and the far left. When Euroscepticism came from the right, centre-left parties did not react as centre-right parties did. This hints at the possibility that Euroscepticism was not connected to competition between the far right and the centre left. A content analysis and classification of types of Eurosceptic frames might serve to clarify whether this is due to the specific type of Eurosceptic frames of far-right Eurosceptics, since Euroscepticism varies (Carlotti & Gianfreda, 2020; De Vries & Edwards, 2009; Pirone, 2020; Vezzani, 2020).

  5.3 North and South

  While we did not present specific expectations with respect to the comparison between the Flemish parliament and the Walloon/French-speaking parliaments, some takeaways can nevertheless be formulated. As expected, Eurosceptic initiations were more common among the Flemish parties (Randour & Bursens, 2019). This stands out especially when comparing the regional parliaments. Hard Eurosceptic initiations were somewhat common in the Flemish parliament, mostly coming from VB and LDD. In the Walloon and French-speaking parliaments, there was one year in which the small PTB voiced hard Euroscepticism, but the party very quickly switched to a pro-EU stance. This translated into very modest reactions from other parties.
  Remarkably, in the federal parliament, parties show fewer substantial reactions to PTB’s short term of Euroscepticism than was the case for earlier Eurosceptic initiators. This might be explained by PTB’s small size at the time but could also suggest decreasing marginal returns of Euroscepticism. The latter could also explain why PTB/PVDA did not continue to pursue a Eurosceptic strategy: as mainstream parties had already adapted, Euroscepticism was no longer an effective strategy for a fringe party trying to break into the electorate of bigger parties (Hobolt & De Vries, 2015). Euroscepticism was already a part of the field of contention, owing to the Flemish Eurosceptics, and the other parties had already reacted to this. At the regional level, parties increased EU salience, but this was largely in an act of resistance against Euroscepticism as their positions remained mostly as pro-EU as before. In any case, this could not have been enough to incentivise PTB to maintain its Euroscepticism as the only party they affected was not a main competitor. Moreover, it might even have provided an additional incentive to abort Euroscepticism as a strategy: they would not want to combat PS in the same way as a centrist party (which maintained soft Euroscepticism).
  On the other hand, the French-speaking parties did change in the federal parliament when reacting to Flemish Eurosceptics. We see two explanations for this. First, the division of competences has put the federal level in charge of areas that are more prone to Euroscepticism (migration, monetary policy) compared with the competences of the regions and communities. In addition, the mere presence of Flemish Eurosceptics in the federal parliament might be enough to trigger Walloon parties’ response, despite the absence of an electoral threat.

• 6 Conclusion

  Euroscepticism has become an increasingly salient issue in Belgian parliaments, despite a traditionally Europhile political landscape and increasing communitarian tensions. We found that mainstream parties raise EU salience when facing Euroscepticism, especially when considering that Euroscepticism is salient itself. Moreover, the salience effect was not more pronounced for parties in intense competition with Eurosceptics. Neither was it for regionalists. Our findings indicate that hard Euroscepticism from peripheral parties, especially when salient, increased soft Euroscepticism among the other parties. We did not find evidence to support the hypotheses that regionalist parties, centre-left parties or opposition parties would become especially Eurosceptic. In contrast, the increase in soft Euroscepticism was especially pronounced for parties in government and parties on the further ends of the political spectrum. However, we argue that hard Eurosceptic challengers face diminishing returns in their attempt to politicise the EU. While they affect mainstream parties a lot at first, subsequent increases in Euroscepticism might be less effective: once parties have adapted to hard Eurosceptic challengers, they are less affected by later increases in Euroscepticism. Lastly, we note that parties’ EU-related behaviour evolved differently in parliaments at different levels and that at times parties reacted to Eurosceptic initiators from the other language group, i.e. outside their electoral arena. This might be ascribed to the division of competences, to the mere fact that parties interact in the federal parliament or to outside events.
  Our study has many limitations. Firstly, we did not control for variables external to the parliamentary setting that might trigger variation in parties’ EU salience and position. For instance, the signing of the Treaty of Lisbon or the consecutive crises (Hutter & Kriesi, 2019; Serricchio et al., 2013) may help to explain the effects of the 2008 rise in Euroscepticism in the federal arena, despite the absence of direct electoral competition. Secondly, future research may ascertain whether other sources such as party platforms or media statements confirm our findings on the basis of parliamentary documents.
  While we reached clear results for the Belgian case, the findings do not seem to be universally supported. For example, Turnbull-Dugarte (2020) shows that German mainstream parties reacted to a Eurosceptic challenge by becoming more vocally pro-EU, arguing that this was due to Europhilia among the supporters of those mainstream parties. Indeed, Europhilia among Belgian voters seems to be decreasing (Brack & Hoon, 2017) and could thereby explain the difference between the behaviour of Belgian and German parties. The direction of causality between parties is difficult to determine. Rohrschneider and Whitefield (2017) argue that parties might feel too constrained by their previous positions to shift strategies to gain votes. However, this explanation seems insufficient as the traditionally Europhile Belgian parties have shifted their positions, and we see the same in the Netherlands and France (Bijsmans, 2017; Brack & Startin, 2015; Statham & Trenz, 2013). Another possibility is a difference in the depiction of the EU in media. For example, Startin (2015) shows how in the British case lopsided coverage of the EU has led to a mainstreaming of Euroscepticism. This mainstreaming of Euroscepticism in UK media is also noted by Bijsmans (2017), who additionally shows that criticism of the EU is increasingly also a part of debates in Dutch media. Both countries have substantial Eurosceptic parties. In any case, there seems to be a country-level variable that determines what strategy mainstream parties will adopt in the case of Eurosceptic competition.
  Finally, it should be noted that the spread of (soft) Euroscepticism to other parties is not the only type of Eurosceptic influence, and it is not the same as policy influence. For example, Biard (2019) illustrates different stages at which such parties can affect policy. Similarly, Carvalho (2014) shows how especially in terms of the influence of challenging parties, there can be a considerable gap between statements made by politicians and effectively implemented policy.
  Nevertheless, despite these limitations, our contribution is clear: when facing Eurosceptic challengers, mainstream parties in Belgian parliaments raise salience fairly equally, with government and peripheral parties adopting (soft) Euroscepticism more often than other parties.

References

• Auel, K. & Raunio, T. (2014). Debating the state of the union? Comparing parliamentary debates on EU issues in Finland, France, Germany and the United Kingdom. The Journal of Legislative Studies, 20(1), 13-28. https://doi.org/10.1080/13572334.2013.871482

• Beyers, J. & Bursens, P. (2013). How Europe shapes the nature of the Belgian federation: Differentiated EU impact triggers both co-operation and decentralization. Regional and Federal Studies, 23(3), 271-291. https://doi.org/10.1080/13597566.2013.773898

• Biard, B. (2019). The influence of radical right populist parties on law and order policy-making. Policy Studies, 40(1), 40-57. https://doi.org/10.1080/01442872.2018.1533110

• Bijsmans, P. (2017). EU Media Coverage in Times of Crisis: Euroscepticism Becoming Mainstream? In M. Caiani & S. Guerra (Eds.), Euroscepticism, democracy and the media (pp. 73-94). London: Palgrave Macmillan.

• Bodson, B. (2020). What Can the EU Expect from the New Belgian Government? Egmont European Policy Brief, 64. Retrieved from www.egmontinstitute.be/what-can-the-eu-ex­pect-from-the-new-belgian-government/

• Brack, N. & Hoon, L. (2017). Euroscepticism in Belgium: Issue-Voting Against the Odds. In J. Hassing Nielsen & M. Franklin (Eds.), The Eurosceptic 2014 European Parliament Elections (pp. 171-195). London: Palgrave Macmillan.

• Brack, N. & Startin, N. (2015). Introduction: Euroscepticism, from the margins to the mainstream. International Political Science Review, 36(3), 239-249. https://doi.org/10.1177/0192512115577231

• Bursens, P. (2002). Belgium’s adaptation to the EU: Does Federalism Constrain Europeanisation. Res Publica, 54(4), 549-598.

• Carlotti, B. & Gianfreda, S. (2020). Do Populists Talk the Same? A Multilevel Analysis of the Northern League and Five Star Movements’ Positions on Immigration and the European Union. In M. Baldassari, E. Castelli, M. Truffelli & G. Vezzani (Eds.), Anti-Europeanism (pp. 163-183). Cham: Springer.

• Carvalho, J. (2014). British and French policies towards high-skilled immigration during the 2000s: Policy outplays politics or politics trumps policy? Ethnic and Racial Studies, 37(13), 2361-2378. https://doi.org/10.1080/01419870.2013.833645

• Dardanelli, P. (2012). Europeanization and the unravelling of Belgium: A comparative analysis of party strategies. Acta Politica, 47(2), 181-209.

• De Kamer van volksvertegenwoordigers (2020). Plenaire vergaderingen, De Kamer. Retrieved from www.dekamer.be/kvvcr/showpage.cfm?section=/cricra&language=nl&cfm=dcricra.cfm?type=plen&cricra=cri&count=all (accessed 19 March 2020).

• De Vries, C. E. & Edwards, E. E. (2009). Taking Europe to its extremes: Extremist parties and public Euroscepticism. Party Politics, 15(1), 5-28.

• De Wilde, P., Michailidou, A. & Trenz, H.-J. (2014). Converging on Euroscepticism: Online polity contestation during European Parliament elections: Converging on Euroscepticism. European Journal of Political Research, 53(4), 766-783. https://doi.org/10.1111/1475-6765.12050

• Franck, C., Leclercq, H. & Vandevievere, C. (2003). Belgium: Europeanisation and Belgian Federalism. In W. Wesselts, M. Maurer & J. Mittag (Eds.), Fifteen into one? (pp. 69-91). Manchester: Manchester University Press.

• Giannetti, D., Pedrazzani, A. & Pinto, L. (2017). Party system change in Italy: Politicising the EU and the rise of eccentric parties. South European Society and Politics, 22(1), 21-42.

• Green-Pedersen, C. (2007). The growing importance of issue competition: The changing nature of party competition in Western Europe. Political studies, 55(3), 607-628.

• Hobolt, S. B. & De Vries, C. E. (2015). Issue entrepreneurship and multiparty competition. Comparative Political Studies, 48(9), 1159-1185.

• Hooghe, L., Marks, G. & Wilson, C. J. (2002). Does left/right structure party positions on European integration? Comparative Political Studies, 35(8), 965-989.

• Hutter, S. & Kriesi, H. (2019). Politicizing Europe in times of crisis. Journal of European Public Policy, 26(7), 996-1017.

• Karlsson, C. & Persson, T. (2020). Comparing oppositional behaviour in European Union affairs: Empirical evidence from six member states. West European Politics, online first, https://doi.org/10.1080/01402382.2020.1814618

• Kopecký, P. & Mudde, C. (2002). The two sides of Euroscepticism: Party positions on European integration in East Central Europe. European Union Politics, 3(3), 297-326. https://doi.org/10.1177/1465116502003003002

• Ladrech, R. (2002). Europeanization and political parties: Towards a framework for analysis. Party politics, 8(4), 389-403. https://doi.org/10.1177/1354068802008004002

• Meier, P. & Bursens, P. (2021). Belgium: the Democratic State of the Federation. In A. Benz & J. Sonnicksen (Eds.), Federalism and Democracy (pp. 180-196). Toronto: University of Toronto Press.

• Meijers, M. J. (2017). Contagious Euroscepticism: The impact of Eurosceptic support on mainstream party positions on European integration. Party Politics, 23(4), 413-423. https://doi.org/10.1177/1354068815601787

• Meijers, M. J. & Williams, C. J. (2020). When shifting backfires: the electoral consequences of responding to niche party EU positions. Journal of European Public Policy, 27(10), 1506-1525. https://doi.org/10.1080/13501763.2019.1668044

• Parlement de la Fédération Wallonie-Bruxelles (2020). Réunions. Fédération Wallonie-Bruxelles: Le parlement. Retrieved from www.pfwb.be/le-travail-du-parlement/seance-pleniere/reunions (accessed 19 March 2020).

• Parlement de Wallonie (2020). Travail parlementaire – Séances plénières et commissions, Parlement de Wallonie. Retrieved from www.parlement-wallonie.be/pwpages?p=pub-form (accessed 19 March 2020).

• Petrocik, J. R. (1996). Issue ownership in presidential elections, with a 1980 case study. American Journal of Political Science, 825-850. https://doi.org/10.2307/2111797

• Pirone, M. (2020). Paths of critical Europeanism: From Blockupy to neo-municipalism. In M. Baldassari, E. Castelli, M. Truffelli & G. Vezzani (Eds.), Anti-Europeanism (pp. 185-191). Cham: Springer.

• Pittoors, G., Wolfs, W., Van Hecke, S. & Bursens, P. (2016). Europa in Vlaanderen: een verkennende analyse van de impact van Europese integratie op Vlaamse politieke partijen. Res Publica, 58(4), 423-444.

• Putnam, R. D. (1988). Diplomacy and domestic politics: The logic of two-level games. International Organization, 42(3), 427-460.

• Randour, F. & Bursens, P. (2019). The Politicisation of the EU within and across Belgian Parliaments. Paper prepared for the UACES Annual Conference, Lisbon, 1-4 September 2019.

• Rohrschneider, R. & Whitefield, S. (2017). Party Positions about European Integration in Germany: An Electoral Quandary? German Politics, 26(1), 83-103. https://doi.org/10.1080/09644008.2016.1189532

• Serricchio, F., Tsakatika, M. & Quaglia, L. (2013). Euroscepticism and the global financial crisis. JCMS: Journal of Common Market Studies, 51(1), 51-64. https://doi.org/10.1111/j.1468-5965.2012.02299.x

• Startin, N. (2015). Have we reached a tipping point? The mainstreaming of Euroscepticism in the UK. International Political Science Review, 36(3), 311-323. https://doi.org/10.1177/0192512115574126

• Statham, P. & Trenz, H.-J. (2013). How European Union Politicization can Emerge through Contestation: The Constitution Case: How EU politicization can emerge through contestation. JCMS: Journal of Common Market Studies, 51(5), 965-980. https://doi.org/10.1111/jcms.12032

• Szczerbiak, A. & Taggart, P. (2008). Theorizing party-based Euroscepticism: Problems of definition, measurement, and causality. In A. Szczerbiak & P. Taggart (Eds.), Opposing Europe? The Comparative Party Politics of Euroscepticism, Volume II: Comparative and Theoretical Perspectives (pp. 238-262). Oxford: Oxford University Press.

• Taggart, P. (1998). A touchstone of dissent: Euroscepticism in contemporary Western European party systems. European Journal of Political Research, 33(3), 363-388.

• Taggart, P. & Szczerbiak, A. (2002). The party politics of Euroscepticism in EU member and candidate states (pp. 7-8). Brighton: Sussex European Institute.

• Treib, O. (2020). Exploring mainstream Euroscepticism: Similarities and differences between Eurosceptic claims of centre-right and radical right parties. Research and Politics, 7(3), 1-7.

• Turnbull-Dugarte, S. J. (2020). A new hope for europhiles? The 2017 German federal elections and the revenge of the pro-European mainstream. Journal of European Integration, 1-26. https://doi.org/10.1080/07036337.2020.1826943

• Van de Wardt, M. (2015). Desperate needs, desperate deeds: Why mainstream parties respond to the issues of niche parties. West European Politics, 38(1), 93-122. https://doi.org/10.1080/01402382.2014.945247

• Vezzani, G. (2020). From the Clash of Cultures to the Culture of Clash: Normative Political Theory and the Question of Muslims’ Citizenship in Europe and the USA. In M. Baldassari, E. Castelli, M. Truffelli & G. Vezzani (Eds.), Anti-Europeanism (pp. 149-162). Cham: Springer.

• Vlaams Belang (2018). ‘Programma: een vrij Europa’, Vlaams Belang. Retrieved from https://www.vlaamsbelang.org/programma (accessed 27 May 2020).

• Vlaams Parlement (2020). Plenaire vergaderingen, Vlaams Parlement. Retrieved from www.vlaamsparlement.be/plenaire-vergadering#verslagen (accessed 19 March 2020).

• Wolfs, W. & Van Hecke, S. (2020). Belgium: Breaking the Consensus? Eurosceptic Parties. In M. Keading, J. Pollak & P. Schmidt (Eds.), Euroscepticism and the Future of Europe (pp. 9-12). Cham: Palgrave Macmillan.