Keywords

Introduction

Previous chapters in this book have considered from a comparative perspective the extent to which EU politicization is taking place in media and parliaments and, adopting an experimental standpoint, whether EU issue voting exists. Evidence contained in these chapters indicates that Spain has witnessed an increase in the salience of EU issues over the past years that is not necessarily accompanied by more contestation in parliaments and media. It also suggests that we must be careful when considering the existence of EU issue voting, as heuristics may be operating. This chapter delves into the political dynamics of Spain to shed light on the effects of EU issues for voting in the November 2019 elections. For that, it presents an overview of the perceptions and positions towards the EU among Spanish citizens and parties, disaggregates indicators of politicization in both media and parliamentary debates, and conducts an innovative electoral analysis using a complete set of variables that includes key issues for voting in 2019, such as general preferences towards the territorial organization of the State, perceptions of the degree of autonomy and the state of civil and political rights in Catalonia, immigration, and feminism (the last one with a focus on the perception of violence against women). The rest of this section revolves around the evolution of EU perceptions among parties and citizens in Spain.

Spain was for a long time an exceptional case due to the positive consensus existing around the European Union (EU), at least among mainstream state-wide political parties (Benedetto & Quaglia, 2007; Elias, 2008; Gomez-Reino et al., 2008; Jerez Mir et al., 2008). EU membership appeared linked to ideas of democratization and modernization (Farrell, 2004; Powell, 1986; Royo & Manuel, 2007) and putting an end to the international isolation experienced under the dictator Franco (Powell, 2003), which made contestation around the issue almost nonexistent in the years immediately before and after the accession (VĆ”zquez GarcĆ­a et al., 2010; Verney, 2011). Although Spain joined the EU amid negotiations conducive to the adoption of the Single European Act (SEA), concerns about its economic impact were balanced out by the existence of European funds that would be particularly beneficial for less developed countries to adapt to the single market. Therefore, under the presidencies of Felipe GonzĆ”lez (1982ā€“1996) and JosĆ© MarĆ­a Aznar (1996ā€“2004) that happened before and after the effective accession, both public opinion and a large majority of political actors in Spain maintained a markedly Euro-enthusiast position (Powell, 2003). In fact, less than 10% of the Spanish population declared to oppose EU membership until 2008 (Real-Dato & Sojka, 2020). However, the economic and political earthquakes that started with the economic and financial crisis of 2008 caused some tremor in the Euro-enthusiasm of Spain.

Spain had the dubious honor of being among the European countries most affected by the crisis, which hit hard the labor market. Unemployment rose by 37% in just one year, and the situation only got worse afterwards (Royo, 2009). By the beginning of 2013, the unemployment rate had more than tripled 2007 figures. Besides rising poverty and economic inequality (Zamora-Kapoor & Coller, 2014), the negative effects of the crisis were (and still are) visible in fundamental aspects such as public health, or gender inequality (Gili et al., 2013; Karamessini & Rubery, 2013). It thus stands to reason that the political sphere would not be immune to the situation.

While mainstream political parties of the left (Partido Socialista Obrero EspaƱol, PSOE) and the right (Partido Popular, PP) changed little their discourses towards the EU (Bakker et al., 2012; Volkens et al., 2020), public opinion became much more critical of the EU after 2008. As a textbook example of this trend, the percentage of people who declared to trust the European Parliament in Spain fell from 69 to 44% between 2007 and 2011, the second biggest decline of the EU after Greece (Armingeon & Ceka, 2013), and trust towards the EU shrunk from 66 to 16% between 2008 and 2014 (Cordero & Montero, 2015). In fact, the 15Ā M-indignados mobilization that spread after 2011 was a paradigmatic example of a social movement challenging EU austerity while asking for pro-democratic reforms (Flesher Fominaya, 2017).

Connected with this public mobilization, the ongoing reconfiguration of the Spanish party system that started with the 2014 European elections affected the general consensus around the EU. Podemos, a new party characterized by its radical left-wing ideology and populist discourse was born with a soft Eurosceptic discourse directed towards neoliberal and adjustment policies (Plaza-Colodro et al., 2018). The radical-right VOX also displayed a soft Eurosceptic discourse that, unlike that of Podemos, focused much more on sovereignty and national identity issues (Marcos-Marne et al., 2021; Real-Dato & Sojka, 2020). On the contrary, the third new party that gained importance in this period, the center-right Ciudadanos (Cs), never questioned the mainstream consensus around the EU (Polk et al., 2017). Overall, by the turn of the decade a new scenario emerged in Spain that, far from representing a Eurosceptic turn, looked more prone to EU politicization along the lines of increased issue salience, actor expansion, and actor polarization (Hutter & Kriesi, 2019). On the one hand, public opinion towards the EU slightly changed from its monolithic positive view. On the other hand, new political actors emerged that were willing to compete on EU issues, a strategy situated in sharp contrast with the no-contestation approach largely followed util then (VƔzquez Garcƭa et al., 2010).

Politicization of the EU in Media and Parliamentary Debates

To further delve into the indicators of EU politicization in media and parliamentary debates, this chapter presents two types of evidence. First, a graphical representation of the salience and tone of the coverage of the EU in the two most important newspapers in the country, El Mundo, and El PaĆ­s, in the period between 2004 and 2016 (Fig.Ā 12.1). Second, a party-disaggregated representation of EU salience and tone for the five most important state-wide parties (PSOE, PP, Cs, Podemos, VOX), in the period between 2002 and 2019 (Figs.Ā 12.2 and 12.3). The salience variable represents the percentage of news/interventions that refer to the EU. Meanwhile, the tone variable represents the overall perception of the EU in each unit of analysis (be it parliamentary intervention or newspaper news). Values over zero represent positive perceptions, and values below zero are negative ones. More detailed information about the coding procedure can be found in Chapter 3.

Fig. 12.1
Two bar graphs reflect the increasing coverage of EU issues in media between 2004 and 2019. It plots the percentage of EU salience with EU tone for both E I mundo and E I Pais.

EU salience and tone in SpanishĀ media (Note self-elaboration with data from MAPLE)

Full size image
Fig. 12.2
A line graph displays the different political parties like Cs, I U, Pedemos, P P, and P S O E speaking about E U salience in parliamentary debates from the year 2000 to 2019. P P plots the highest values in 2012.

EU salience in parliamentary debates by partyĀ in Spain (Note self-elaboration with data from MAPLE)

Full size image
Fig. 12.3
A line graph compares the E U tone in debates by different political parties like Cs, I U, Pedemos, P P, and P S O E from 2000 to 2019. P P exhibits the highest values from 2000 to 2003 and in 2014.

EU tone in parliamentary debates by partyĀ in Spain (Note self-elaboration with data from MAPLE)

Full size image

Regarding salience, data from Fig.Ā 12.1 reflects the increasing coverage of EU issues in media between 2004 and 2016, with a moderation in 2015, and an ending point in 2019 somewhat higher than 2004. This trend can be seen in both El Mundo and El PaĆ­s. Data from parliamentary debates gathered in Spain evidence that political parties speak about the EU, although the percentage of EU interventions was rarely above 10% (Fig.Ā 12.2). For the two most important political parties, PSOE and PP, mentions about the EU increased after 2008 (and especially after 2011, coinciding with the most evident effects of the Eurocrisis) (Braun & Tausendpfund, 2014). Mentions of the EU reached its maximum in 2012 both for the PP (11%), and the PSOE (9.3%), coinciding with the official request of help from the Spanish government for the financial system.

As for the tone variable, the indicator for media reached the lowest point in 2011 both in El Mundo and El PaĆ­s, thus coinciding again with the Eurocrisis. EU tone in political parties is positive for both the PP and the PSOE except in 2014 (PSOE), with no evident evolution trend (Fig.Ā 12.3). Cs and Podemosā€™ interventions, conversely, show a negative trend (but there are only four points in time available for these parties). Data on VOX is lacking since they only accessed the national Parliament after the April 2019 elections (Turnbull-Dugarte et al., 2020).

Overall, data from both newspapers and parliamentary discourses evidence an increase in the salience of EU issues after 2008 that declines again after 2012, and a not-so-clear pattern regarding negative mentions of the EU. However,Ā it is evident that Euroscepticism never dominated public debates in Spain. Considering this analysis of media and parliamentary data, this chapter now asks, is there evidence of EU issue voting in the general elections that took place in November 2019?

Did EU Preferences Matter for Voting Decisions?

An increasing bulk of comparative empirical research suggests that EU issues matter for voting decisions, both at the European and national level (Beach et al., 2018; De Vries & Tillman, 2011; Hobolt & Rodon, 2020). In a nutshell, these studies find that whenever information about the EU is available, individuals may ponder EU issues when deciding whom to vote. Overall, the framing and intensity of the information about the EU are expected to influence both voting decisions and the comparative strength of EU issues vis-Ć -vis other variables (van Elsas et al., 2019). Basically, theory predicts that EU politicization will correlate with an increased importance of EU issues in voting, which explains for example the relatively low salience of EU issues in times of the ā€œgeneral consensusā€ before the 90s (Hooghe & Marks, 2009). However, it is also suggested that not all parties will benefit the same from this situation. The issue salience and polarization trends that define politicization (Hutter & Grande, 2014) imply that parties adopting extreme positions and emphasizing EU issues can benefit the most from it. In this sense, the general picture of partiesā€™ position towards the EU in Spain in 2019 largely corresponded to the inverted U-shaped proposed by Hooghe et al. (2002), with Izquierda Unida (IU), Podemos, and VOX displaying comparatively more critical ideas about the EU (Bakker et al., 2020). However, Euroscepticism was not particularly strong in either of the extremes (Carrasco et al., 2021; Fitzgibbon, 2013). In fact, even in 2018 an overwhelming majority of Members of Parliament in Spain (91% of the interviewed) expressed a positive view of the EU (Carrasco et al., 2021).

To unravel if EU issues influenced voting decisions in the November 2019 national elections, I used data from an online survey conducted in Spain in May 2019. Data was gathered by the company Netquest using a quota sampling strategy that took into account gender, age, education, and NUTS1 regions to replicate as much as possible the composition of the census of the country (Nā€‰=ā€‰3,006). Respondents were selected from a large sample of individuals available to the panel provider (opt-in panel). While opt-in panel data has proven to be potentially problematic to calculate population estimates, due to sampling issues, these are far less troublesome when studies focus on relationships between variables, as it is the case in this chapter (Baker et al., 2010).

Multinomial logistic regression models were conducted using voteĀ recall in the past elections as a dependent variable (Which party or coalition did you vote for in the last legislative elections?), and binomial regression ones were also run to see the relevance of EU issues in voting for new parties within ideological blocks (Cs vs PP; VOX vs PP; Unidas Podemos vs PSOE). The key independent variable in all models is attitudes towards the EU, for which I used a question that asks respondents about the future of the European integration process (Some people believe that the process of European integration should move forward to the creation of the United States of Europe. Others believe that the European Union should be dissolved in order to return to a situation in which states are fully sovereign. In which point of the following scale would you place yourself?). This question taps into the strengthening dimension identified by Lubbers and Scheepers (2010).

The models include a rigorous battery of controls that allow considering socio-demographic features and important predictors of voting in Spain. I controlled for age (continuous), gender (female as reference), education (up to secondary, secondary, more than secondary), religiosity (from 1, I have no religious beliefs to 4, I am very religious), trade union membership (dichotomous), self-location in the leftā€“right ideological scale (from 0, extreme left, to 10, extreme right), evaluation of the leader of the party voted (from 0, I strongly dislike, to 10, I strongly like), assessment of the economic situation (when compared to the last 12Ā months, the economy is nowā€¦ from 1, much better, to 5, much worse), preferences about the territorial organization of the state (from 1, a state without autonomies, to 5, a state in which autonomies can become independent), evaluations of the political situation in Catalonia (as far as relations between Catalonia and Spain are concerned, do you think Catalonia has reachedā€¦ from 1, too much autonomy, to 3, an insufficient level of autonomy), assessment of rights in Catalonia using a dichotomous question (do you think that during the last year the citizens of Catalonia have seen a reduction of their individual and collective freedoms as well as their fundamental rights?), and perceptions of violence against women (many women exaggerate the problem of male violence, responses ranging from 1, strongly disagree, to 5, strongly agree). Listwise deletion was applied to deal with missing cases, which results in a final sample of 1512 cases. The reduction of the cases from the general sample is mainly caused by the number of people who said they did not vote or did not declare the party they voted for.

Descriptive data is shown in Table 12.1, and an analysis of preferences on EU strengthening considering vote recall is further included in Table 12.2. Data in Table 12.2 indicate that voters of VOX rank the lowest in the scale of EU strengthening, even if they also show the highest Standard Deviation (SD). Voters of Unidas Podemos (UP, a coalition lead by Podemos and IU) come in second place, followed by voters of PSOE. Voters of Cs and PP rank the highest in the EU strengthening scale. This lends additional support for the expectation that EU issues may influence voting for new radical parties that question, at least to a certain extent, the general positive consensus around the EU.

Table 12.1 Descriptive statisticsĀ of the Spanish sample
Full size table
Table 12.2 EU strengthening by partyĀ in Spain (mean of votersā€™ responses)
Full size table

To see if EU issue voting was operating in the 2019 elections, I first conducted two multinomial regression models using vote choice as dependent variable (using voters of the incumbent PSOE as reference). The first model is the most stringent one, as it included all control variables referred above (Appendix: Table 12.3). The second model maintains the same structure, but two powerful predictors of vote (evaluation of the leader of the party voted in the past elections, and assessment of the economic situation) were removed to see if the electoral effect of EU issues increased (only minor differences were found) (Appendix: Table 12.4).Footnote 1 A graphical representation of the Average Marginal Effects (AMEs)Footnote 2 of main covariates were included in Fig.Ā 12.4. Results show that preferring a more integrated EU is negatively related with having voted for one of the two new radical parties, VOX, and positively related with having voted for PP and Cs. This is partially connected to the U-shape relationship existing between EU positions and ideological radicalism among voters (Hooghe et al., 2002; Rooduijn et al., 2017), clearer in Spain for voters of the radical right (the AME of EU integration on having voted for UP is negative but it does not reach statistical significance at 0.05). It also indicates that left-wing voters, even if they are not radical, may have become more critical of the EU in Southern Europe because of the orthodox economic policies adopted after 2008 (Hutter & Kriesi, 2019).

Fig. 12.4
Five forest plot graphs display the average marginal effects with a confidence interval of 95 percent for different political parties like P P, P O S E, U P, ciudadanos, and V O X.

AMEs visual representation. Vote choice in the past national electionsĀ in Spain (Note Average marginal effects with confidence intervals (95%). The visual representation is based on results included in Appendix: Table 12.3)

Full size image

To further delve into these results, three binomial logistic models were run that separated voters of new parties from voters of mainstream ones within the same ideological space. Accordingly, voters for VOX and Cs were compared to voters of PP, and voters of UP with voters of PSOE (see Fig.Ā 12.5). While choosing between UP/PSOE and Cs/PP seems not influenced by preferences on European integration, results confirm that voting for VOX, and not for PP, is more likely the less individuals favor European federalism (Bā€‰=ā€‰āˆ’0.14, pā€‰<ā€‰0.01). Furthermore, the within-blocks comparison is helpful to shed additional light on the reasons that trigger voting for new parties in Spain. Using PSOE voters as reference, individuals are more likely to vote for UP if they are younger, less religious, more critical of the economic situation, located more to the left, more in favor of more competences for Catalonia, more likely to understand there was a reduction of rights in Catalonia, and more favorable towards immigration (see Appendix: Table 12.5, model 1). Using voters of PP as reference, individuals are more likely to vote for VOX if they are younger, less educated, value more their leader, perceive that violence against women is an overstated issue, favor a more centralized state and crucial for this chapter, oppose more the EU integration process (see Appendix: Table 12.5, model 2). Meanwhile, individuals are more likely to choose Cs over PP if they are younger, less religious, assess the economic situation more negatively, value more their leader, are located more to the left, and understand that violence against women is not an overstated issue (see Appendix: Table 12.5, model 3).

Fig. 12.5
Three forest plot graphs display the average marginal effects with a confidence interval of 95 percent for new mainstream political parties like C s, U P, and V O X.

AMEs visual representation. Voting for new/mainstream parties within ideological blocksĀ in Spain (Note Average marginal effects with confidence intervals (95%). The visual representation is based on results included in Table A.3)

Full size image

Conclusion: European Integration Discourses and EU Issue Voting in Spain

The results shown in the previous section support that EU issue voting was not absent from the November 2019 general elections in Spain. While not the most powerful explanatory variable, which makes total sense considering the stringency of the models proposed, preferences about the supranational character of the EU influenced voting for parties in a way that is coherent with their electoral platforms. Preferring less integration increases the likelihood of voting for the radical-right VOX. On the contrary, favoring a federal EU is positively related with having voted for PP and Cs, and has no effect on having voted for PSOE and UP. Although more studies are needed to confirm it, I suggest these results can be linked to left-wing voters being comparatively more critical of the EU after the Eurocrisis and the measures implemented to tackle it, particularly visible in Southern Europe.

Considering electoral competition within ideological blocks, preferences about the EU integration process play a role only when choosing between right-wing parties (VOX over PP), which may have important consequences for electoral competition. Crucially, all these electoral effects can be observed in 2019, years after debates around the EU peaked in both newspapers and parliamentary debates. This seems to suggest that preferences about the EU can be a (modest) contributor to voting decisions in Spain, especially among voters of right-wing parties, even when politicization remains at low levels.

A fundamental implication of these results is the apparent persistence of a nuanced general consensus around the EU in Spain. The two most voted parties in the past national elections, likely to repeat according to the polls, still hold positive views of the EU, and nothing in current trends make us expect a radical change in that regard (even less with Covid funds being mobilized under the program NextGeneration EU and defense challenges evidenced after the invasion of Ukraine). Combined with the no-contestation strategy followed by mainstream parties, it is likely that EU issue voting maintains a modest role vis-Ć -vis other variables such as self-positioning in the leftā€“right scale (Lancaster & Lewis-Beck, 1986). This is even more the case in light of recent studies that emphasize the importance of heuristics for the formation of attitudes towards the EU (Armingeon & Ceka, 2013; Pannico, 2017; Torcal & Christmann, 2018). Evidence presented in Chapter 5 points in this direction. Once party cues are removed from the calculus preferences about the EU did not explain voting. Accordingly, it cannot be ruled out that part of the strength of the EU coefficients found in voting models is a consequence of which is the preferred party, rather than a prior attitude that explains voting for it.

Importantly, and in line with the overall general consensus mentioned above (in fact, it should be explicitly mentioned that no strong Eurosceptic party emerged in Spain even amid the worst of the crisis), discourses of UP and VOX do not question the EU in its entirety. However, they are clearly more critical towards the EU, and such discourses are likely to maintain at least some relevance in the future. For example, UP may still criticize neoliberal policies of the EU in the implementation of neoliberal policies to tackle the effects of the COVID-19 crisis (e.g., benefiting large companies over affected sectors of the population). From a different perspective, VOX could easily combine anti-immigration and anti-EU rhetoric to reinforce its positioning in the cosmopolitan-parochial divide that is increasingly salient in West European Politics (Ford & Jennings, 2020). These strategies could in turn influence the effect of EU preferences on voting, provided that party-cueing is operating (Pannico, 2017; Hobolt, 2007; Steenbergen et al., 2007). In that sense, a combination of salient and critical discourses towards the EU among radical parties of the left and the right could translate into a more critical public opinion, especially among their voters, which could be reflected again in voting patterns. At the moment of writing these lines the anti-European strategy seems much more likely in the case of VOX, which has recently started to campaign along the lines of the debate between globalism or motherland (globalismo o patria). If successful, this strategy could incorporate relevant changes to the electoral scenario in Spain, a country where was historically feasible to combine a strong national identity with a positive perception of the EU (Carey, 2002; McLaren, 2004).