Keywords

This book started from a puzzle that has recently emerged in Europe. Since 2009, the EU has undergone a series of crises, which posed challenges in different policy areas, from monetary and fiscal, to foreign, migration and health (Genschel & Jachtenfuchs, 2016; Schmidt, 2020). The response to these challenges has generally been to deepen EU integration, with a greater supranationalisation of policies (Laffan & Schlosser, 2016). In doing so, the EU has become more political, acting in a purposeful way, rather than simply following the Treaties, as its institutions have often intervened in a discretionary way to respond to the Eurozone crisis, COVID-19, or even Russia’s invasion of Ukraine (Middelaar, 2019).

Still, there has been no institutional change since the Treaty of Lisbon to match these political and policy trends. In addition, there is evidence to show that the European Parliament elections remain second-order (Nielsen & Franklin, 2017). How, then, is the EU being held accountable for these increased competencies, given the lack of institutional change, and the unchanged “second order” character of EP elections? In this book, we argue that it is fundamental to shift our academic gaze to the domestic level of politics in order to understand how Europe is being held accountable by national institutions. There has been research on the way in which European legitimacy is achieved through national institutions (De Vries, 2007; De Vries & Hobolt, 2016), yet this book contributes to that discussion by offering novel/unique perspectives and evidence on how national institutions are holding the EU accountable.

Following Powell’s classification (Powell, 2004), our objective was to examine the chain of EU responsiveness which can be established through national institutions, namely media, parliamentary debates and legislative elections, in the following countries: Belgium, Germany, Greece, Ireland, Portugal and Spain. We show that the EU politicisation has an impact at the individual voting level, with citizens using their EU preferences to express their party choices at the national level. This means that national politicians, MPs and governments are being (s)elected, and held accountable, for their stance on European integration. What this volume concludes is that the European Union is being held accountable nationally.

We now discuss the conclusions of our research to engage and assess the empirical results from a comparative perspective. We will consider first our findings about the context within which EU politicisation occurs across Europe in the media and parliamentary arenas, and second our findings on the consequences of EU politicisation for the vote at the national level.

Chapters 2, 3 and 4 allowed us to understand the differentiated EU information context within which electors have been placed over time. Research on EU politicisation has evolved, with current studies focusing on the “differentiated” forms it may assume (Braun et al., 2019; Hutter & Kriesi, 2019). Among theorists of EU integration, there was a sense that Euroscepticism fed on “constitutive” issues of membership and institutional design (Mair, 2000), whereas, “politicizing European policies” was likely to lead to a European public sphere (Risse, 2015). We took these perspectives on board to evaluate salience, tone, as well as the types of issues which are being politicised over time, by mainstream media and by parliamentary parties in plenary debates in each country in the twenty-first century.

Our analysis was driven by a media-centric approach which considered all articles (news and op-eds) that focused on the EU. Indeed, other studies of EU politicisation (Hutter & Grande, 2014) have measured the concept in terms of “how parties discuss the EU in the media”, which we have shown conflates measures of parties and media which are not shaped similarly. In each country’s mainstream media, salience and contestation are substantially correlated in each newspaper analysed, independent of their ideological leaning. Distinguishing between news articles and op-eds, it becomes clear that the EU has been, in general, more salient, as well as more contested, in opinion articles, and that the differences between left leaning and right leaning newspapers increase when the focus is exclusively on op-eds. These results suggest that, while the journalistic coverage of the EU may be similar during legislative campaigns, op-eds may be a better object of enquiry when researching the way media shape EU politicisation. Further, focusing on salience and contestation provides a relatively incomplete picture of how the traditional media politicised the EU before and after the eurozone crisis, since it ignores the topics being covered. Our analysis shows that all countries’ media are focusing on policies rather than membership, signalling an increasing tendency for the EU to be discussed in terms of the consequences of its policies, rather than questioned on membership per se. Yet, a more fine-grained analysis shows that while Germany and Ireland discuss EU policies per se to a larger extent, the Southern European countries discuss the EU from the perspective of domesticated policies. Indeed, even within the domain of “policies”, it is possible to discuss their design from a European perspective, and only a few countries are doing that. In the Southern European countries, debates on Europe tend to take the EU policy for granted, and the news focuses on the domestic consequences of EU policies.

Turning to the longitudinal analysis of EU politicisation in parliamentary plenary debates, Chapter 3 by Kartalis and Silva shows that Parliamentary parties behave rather strategically in that arena. Namely, salience tends to be higher when EU contestation is low. Ie the larger mainstream parties will discuss the EU less when they share the plenary with an Eurosceptic party which devotes part of their speeches to contesting the EU. Thus, in Parliamentary debates, salience tends to be higher at lower levels of contestation. In the countries which we analyse, and with the partial exception of Greece, national parliaments have not made a very noticeable contribution to the politicisation of the EU, nor did this change with the Eurozone crisis. Concerning the determinants of the politicisation of the EU in parliament, a Eurosceptic position is the strongest determinant of both salience and, especially, contestation.

Assessing the salience of the EU’s multiple dimensions, results suggest policy-related issues are the most salient topics in both the media and parliaments. Within the realm of policies, Chapter 4’s findings by Santos and Rogeiro Nina confirm that these mostly concern economic and financial matters. Importantly, the greatest differences between parliaments and the media are found in levels of tone. With the exception of Greece, the share of texts (articles or speeches) about the EU that have a negative tone is much greater in the media than it is in parliaments, confirming the negative bias in political news coverage, and the strategic behaviour of parties in Parliament. The picture which emerges then from this in-depth longitudinal analysis of mainstream media and parliamentary debates EU politicisation in Europe is one of differentiation between media and parliaments, rather than parallelism between the two arenas in each country.

Overall, the chapters suggest that EU politicisation is established broadly in a similar fashion across Europe. Yet, it does not function similarly in the media and parliamentary arenas. While salience and contestation are substantially correlated in the mainstream media, this does not tend to be the case in Parliaments. This suggests that parties in Parliament act strategically to reduce EU contestation, whereas this does not occur to the same extent in the media since it operates under a different logic. Thus, the media are contributing to a greater extent to EU politicisation than Parliamentary debates, while the latter are the ideal arena to observe parties’ preferences unfiltered, but in dialogue with other parliamentary parties. This matters for the overall nature of politicisation, as citizens may be receiving mixed messages from different sources, which becomes obvious when both arenas are considered simultaneously.

In the second part of our book, we established the relative importance of EU issue voting using a variety of methods and focuses on the consequences which EU politicisation may have for that factor of voting behaviour.

Chapter 5’s experimental results by Pannico and Lobo showed unequivocally that there is EU issue voting, with parties being punished on average if their EU position is different from the voter’s. Analysing the data per country produces similar findings, with the exception of Spain where no EU issue voting is detected. Overall, this constitutes a major finding that confirms the indirect path of EU representation and accountability via the national elections. We have therefore unequivocally demonstrated, in the experimental analysis, that EU attitudes are a cause rather than a consequence of voting behaviour. Chapter 5’s analysis was then confirmed in Chapter 6, by Heyne, Lobo and Pannico. In it, observational data from post-election surveys fielded in the four debtor countries is employed to understand the relative importance of the EU issue. Results show that respondents’ EU position proximity vis-à-vis the party they voted for matters significantly in determining vote choice, but to a lesser degree than left–right proximity. Thus, while the exogenous impact of the EU issue is clear, it is not a predominant factor of voting behaviour. Moreover, there are no clear differences between voters of challenger and mainstream parties when it comes to the relevance of the EU position proximity in the vote calculus. Indeed, in the bailout countries not only is left–right proximity always a more important explanatory factor than EU issue proximity, generally for voting behaviour this hierarchy is maintained both for mainstream and challenger parties.

When the importance of the EU has been considered for national politics, the perspective has been mostly on parties and party systems (Hooghe and Marks, Hutter and Kriesi). Further, these analyses consider the importance of the EU not on its own, but as part of a larger cleavage which has been variously called the “globalisation”, “integration-demarcation” or “GAL-TAN” cleavage, their goal is to understand the degree to which this new cleavage is supplanting the left–right cleavages in terms of party competition. There is a view that this cleavage is replacing the left–right cleavage as a determinant of voting behaviour. While we are not evaluating the importance of this (variously denominated) cleavage per se, vis-à-vis the left–right dimension, we do evaluate, in the most thorough way possible the importance of European preferences for voting. As we explained in the previous paragraphs, we have established EU issue voting in such a robust way as to be sure that it matters, but we can also qualify its importance vis-à-vis other factors which are determinant of voting behaviour, in particular in relation to the left–right dimension. We establish that the EU issue cross-cuts the left–right dimension, as has been determined for the party systems. Yet, we do not find that the EU issue is replacing the left–right dimension in terms of its importance for voters, whether they vote for mainstream or more extreme parties. This simply is not happening at the voter level.

Whereas EU issue voting has been established comparatively in Chapters 5 and 6, some differences between countries emerged. Therefore, the rest of the book was dedicated to country case chapters. The country case chapters are very rich in detail on the intricacies of EU issue voting across Europe. Taken together, the message is similar to the one emerging from the comparative chapters. Using full multivariate models that include other relevant political variables, we show that EU issue voting is occurring not only for voters of Eurosceptic parties, that politicise the issue in Parliament, but also for mainstream parties, in every single country considered. Namely, Vlams Beland (Flanders) and the Christian Democrats (Wallonia) in Belgium, as shown in Chapter 7 by Stiers; the AfD, CDU and FDP in Germany as shown in Chapter 8 by Navarrete and Debus; Syriza, Kinal and ND in Greece, as shown by Nezi in Chapter 9; Sinn Fein and Fine Gael in Ireland as shown by Heyne in Chapter 10; PCP and PSD in Portugal as shown by Lobo in Chapter 11; and finally Vox, PP and Cs in Spain as shown by Marne in Chapter 12. Thus, we have managed to congregate very strong multimethod evidence in favour of EU issue voting in each of these countries, in the post-crisis period.

While Chapter 3 showed that the parties’ Eurosceptic position explains both salience and tone, leading to strategic behaviour, what is interesting is that, as both Chapter 5 and the country chapters show, EU issue voting is not circumscribed to voters of Eurosceptic parties. Instead, citizens voting for different types of parties, namely mainstream and challenger parties, use the vote to express EU preferences. This was shown experimentally in Chapter 5, with the exception of Spain. It was also shown observationally, in Chapter 6, and in the country case chapters.

Both the comparative Chapters, as well as the Country case studies also analyse in different ways how the EU politicisation context matters for voting. In Chapter 5, the association between EU media and salience and EU issue voting is tested. Results suggest that there is an association between the availability of EU issues and the strength of EU issue voting, while the tone appears less relevant. Chapter 6 interacts the parties’ tone in plenary debates and the strength of EU issue voting, showing that for parties that have a more negative tone towards the EU (hence, have a more Eurosceptic discourse), the EU proximity matters more in determining the voters’ choice than for parties with a positive or neutral EU tone in parliament.

Overall, the book’s findings concerning the consequences of politicisation for voting behaviour matter for different reasons. First and foremost, they have consequences for the debates about EU legitimacy which are ongoing today. Most of the efforts for democratising Europe involve supranational reforms, at the level of the European Parliament, or citizens’ initiatives at the EU level (Hennette et al., 2019). Yet, what our research suggests is that there is vertical accountability in national legislative elections regarding the EU. Namely, voters of both mainstream and challenger parties are using their positions on the EU to determine the vote. Therefore, the national channel of EU accountability, and namely the domestic institutions which contribute to it, need to become a central part of the debate on EU legitimacy.

Secondly, our findings have consequences for the study of EU politicisation. We have demonstrated that EU politicisation media and parliamentary debates suggest that they do not work in tandem, they follow different trends and that citizens are taking cues from both arenas. Importantly, the media has a negative bias as issues become more salient. Yet, even when parties actively tried to depoliticize the issue, such as in the cases of Ireland, Germany or Portugal, the EU issue still appears as relevant for vote choices, not only for challenger but also for mainstream parties. This suggests that politicisation works through multiple channels, depending on the arena being considered, and it impacts voters differently, depending on whether we consider the media or parliaments.

Finally, our findings suggest that while the chains of accountability are similar from Greece to Germany, in many respects, there are important differences too. Namely, whereas all countries mostly discuss “policies” rather than “membership” in the media and parliamentary debates, suggesting the EU’s legitimacy, they do so differently. Whereas Spain and Portugal mostly discuss “domesticated EU policies”, they hardly discuss “EU policies”, while the latter tend to dominate debates in Germany. These findings suggest again that not all politicisation is alike, and it is necessary to go beyond salience and tone, to see the dimensions which are being debated as we do in this volume. Also, it is the case that in the experimental study, EU issue voting is not significant in Spain. Further, while EU issue voting is significant using observational data in all countries, both for mainstream parties and extreme ones, whether these belong to the left or right side of the ideological spectrum varies between countries.

Further Research

Our research findings notwithstanding, they also raise issues that call for more investigation. Namely, we should focus on deepening the patterns of differentiation in the different arenas and between countries. Thus, even when salience and tone are converging across Europe, there may be fundamentally different perspectives on Europe originating from different countries that can only be understood from a closer, qualitative analysis of discourse. These deeper insights into the topics and frames of EU politicisation may give important insights on the nature of politicisation and whether it helps or hinders EU legitimacy. Further, the way we analyse the relationship between the EU media salience and tone and the vote is at the aggregate level, and should be explored further at the individual level. In addition, studies should take into account other arenas of politicisation such as social networks, given their enormous and ever-growing importance. While the comparison of the media and parliamentary debates’ patterns of politicisation opened up new perspectives on how voters are receiving multiple and different cues, we still did not explore these alongside social networks patterns of politicisation. We expect that the latter may follow a similar pattern to the mainstream media, but this needs to be confirmed. Finally, other countries should be included in the analysis in order to consolidate the overall finding concerning the importance of EU issues for voting in national elections.