In the name of the people: left populists versus right populists

European Politics and Society ISSN: 2374-5118 (Print) 2374-5126 (Online) Journal homepage: https://www.tandfonline.com/loi/rpep21 In the name of the people: left populists versus right populists José Rama & Andrés Santana To cite this article: José Rama & Andrés Santana (2020) In the name of the people: left populists versus right populists, European Politics and Society, 21:1, 17-35, DOI: 10.1080/23745118.2019.1596583 To link to this article: https://doi.org/10.1080/23745118.2019.1596583 Published online: 22 Mar 2019. Submit your article to this journal Article views: 428 View related articles View Crossmark data Citing articles: 1 View citing articles Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=rpep21 EUROPEAN POLITICS AND SOCIETY 2020, VOL. 21, NO. 1, 17–35 https://doi.org/10.1080/23745118.2019.1596583 In the name of the people: left populists versus right populists José Rama and Andrés Santana Edificio de Ciencias Jurídicas, Autonomous University of Madrid, Madrid, Spain ABSTRACT KEYWORDS Using the 8th Round of the European Social Survey, we analyse the electoral competition between left-wing (LWPPs) and rightwing populist parties (RWPPs) in the six European countries where this type of competition has occurred (France, Germany, Iceland, Italy, Lithuania and The Netherlands). By focusing on the competition between populist parties in the same sample of countries we can resolve what tells apart a LWPP voter from a RWPP voter. Thus, we study the factors that explain their different voting behaviour and confine our attention to the four main factors that affect populist competition: anti-elite attitudes, material deprivation, euro-scepticism and anti-immigration attitudes. Our results show that LWPPs and RWPPs voters are similar in their attitudes towards immigration and the European Union (EU) but differ in their attitudes towards elites and material deprivation. Populism; left-wing populist parties; right-wing populist parties; vote; electoral behaviour; globalisation; immigration; material deprivation; European Union; political elites Populism is on the rise. From Central and Eastern European countries (e.g. Jobbik in Hungary, and Law and Justice in Poland) to Western and Southern ones (e.g. Party for Freedom in The Netherlands, and Podemos in Spain), populist parties have attracted a notable electoral support in recent elections. Accordingly, a growing corpus of academic work has examined the determinants of their electoral success.1 Given that the populist upsurge has involved parties with different ideologies, some studies have analysed the cultural and economic determinants of the support for populist parties at large (Inglehart & Norris, 2016; Rovny & Rovny, 2017; Springford & Tilford, 2017; van der Waal & Koster, 2017). Indeed, there are scholars who defend that it is not useful to distinguish between populism on the left and on the right (Clark, Bottom, & Copus, 2008; Judis, 2016; Rivero, Zarzalejos, & del Palacio, 2017) and populist parties’ leaders often argue that the traditional left-right divide is no longer valid. Nevertheless, most works have pointed out that populist parties are not only characterised by their populist element but also by their thick ideology (Mudde, 2004). There is also increasing agreement that belonging to one or the other breed of populist parties has important implications, which even affect how the pure people are portrayed (Otjes & Louwerse, 2015, p. 61). These differences should not obscure the fact that populist parties from different ideological blocks share important characteristics. Rooduijn and Akkerman (2015) argue that CONTACT José Rama jose.rama@uam.es Edificio de Ciencias Jurídicas, Autonomous University of Madrid, 1ª Planta, Office D01, C/Marie Curie, 1, Ciudad Universitaria de Cantoblanco, Madrid 28049, Spain © 2019 Informa UK Limited, trading as Taylor & Francis Group 18 J. RAMA AND A. SANTANA radical left and radical right parties ‘do not differ significantly from each other when it comes to their populism’ (p. 8) and have the same message: ‘corrupt elites neglect the interests of ordinary people’ (p. 9). Let us illustrate this with a few examples: in Great Britain, Clark et al. (2008) compare the British Nationalist Party (BNP) and the Respect Party (the first one a RWPP and the second, a LWPP), and find that they are remarkably similar in terms of authoritarianism, anti-elitism and community-focused politics (p. 527). Likewise, the coalition built by Syriza and ANEL after the 2015 Greek general elections (respectively, a LWPP and a RWPP) has reinforced the so-called theory of the two extremes whereby left and right populists have essential (proto-extremist) commonalities (Anastasakis, 2013). The coalition between Di Maio’s Five Star Movement (M5S) and Salvini’s Lega is another recent case of a government conformed by populist parties which differ in their ideological dimension. In fact, these examples are consistent with the socalled horseshoe hypothesis (Faye, 2002), which states that supporters of LWPPs and RWPPs are quite similar and share individual characteristics (hence, the predictors of voting for a LWPP or RWPP would be largely the same). The question is: to what extent do these common characteristics imply that LWPPs and RWPPs compete for a similar electoral base? And, when they do, how does their ideological leaning shape this competition? The bulk of our knowledge on the electoral bases of populist parties stems from case studies and comparative studies that are circumscribed to either LWPPs (March, 2011) or RWPPs (Mudde, 2007). Comparative studies highlight that young, highly educated urban dwellers are more prone to support LWPPs (Segatti & Capuzzi, 2016). Those with lower class positions have also been shown to be more likely to vote for populist radical left parties (Lubbers & Scheepers, 2007), and those who identify with the working class are more likely to vote for the radical left (Ramiro, 2016). Several case studies point in the same direction: for instance, the contributions of Fernández-Albertos (2015) and Orriols and Cordero (2016) for Podemos in Spain, or Teperoglou, Tsatsanis, and Nicolacopoulos (2015) for Syriza in Greece. Comparative investigations focused on RWPPs point out that their support grows among males who belong to the youngest and oldest age cohorts (Arzheimer & Carter, 2006), the low educated (Arzheimer, 2009), the unemployed (Lubbers, Gijsberts, & Scheepers, 2002) and the lower social classes (Van der Brug et al., 2000).2 By and large, the common understanding is that LWPPs attract people with high levels of education, middle class and young, whereas RWPPs gain their strongest support from the less educated, old and working class (Hernández & Kriesi, 2016). Both types of voters do, however, come from similar social strata, in that both are more likely to be manual workers, unemployed and with low incomes (Visser, Lubbers, Kraaykamp, & Jaspers, 2014). Although these are certainly relevant insights, it is essential not to overlook that they rest ultimately on the comparison of different sets of countries. In the case of RWPP, the sample of countries is likely to include Austria, Belgium, Finland, Great Britain, Hungary, Slovenia, Sweden and Switzerland, as well as some that are also considered in studies on LWPPs. Studies on LWPPs, in turn, are bound to include countries not present in studies on RWPPs, such as Spain and Estonia. Visser et al. (2014) as well as Rooduijn (2017) include both LWPPs and RWPPs in their studies with the explicit goal of comparing voters from both, but most of the countries analysed have only one type of populist party. EUROPEAN POLITICS AND SOCIETY 19 Hence, despite the important progress in the state of the art, comparisons of the supporters’ characteristics of the two populist families have usually been indirect, and we still lack a systematic comparison of LWPPs and RWPPs with the same sample of countries for both. Filling this gap is important because the conclusions regarding the effects of any set of independent variables upon a variable of interest (in this case, support for LWPPs or RWPPs) may be an artefact of the characteristics of electoral competition within countries. For instance, lower social strata do not have the option to vote for a LWPP in countries where there only is a RWPP. Perhaps they would have voted for a LWPP if there had been one in their country. This problem is solved in the current study: we offer more robust conclusions because we focus on those countries where both kinds of populist parties exist, i.e. our inferences are derived from the same sample (in this case, countries), not from indirect comparisons of different cases. Our work concentrates on electoral populist competition. The main question we address is: what tells apart a LWPP voter from a RWPP voter?3 We confine our attention to four factors that are especially related to populist competition: anti-elite and anti-immigration attitudes, material deprivation, and euro-scepticism. We draw on the 8th round of the ESS (ESS Round 8: European Social Survey Round 8 Data, 2016), from now on, ESS (2016). 1. Theory and hypotheses 1.1. Material deprivation Economic factors have been recurrently singled out to understand the renewed success of populist parties. Some investigations have underlined the importance of economic downturns (Kriesi & Pappas, 2015), the worsening of personal economic conditions (Eatwell, 2003, p. 53), or the perception that the country’s economy is performing poorly (Mols & Jetten, 2017). At the aggregate level, numerous scholars have held the Great Recession (the global financial crisis that set out in 2008) responsible for the rise of populist parties (Funke, Schularick, & Trebesch, 2015). Country case-studies show that the regions most severely exposed to the crisis have witnessed more significant increases in the support for populist parties – see Kestilä and Söderlund (2007) and Ivalid, Lanzone, and Dutozia (2016) for the 2004 and 2015 French regional elections, respectively. For instance, in the 2016 American Presidential elections, Trump obtained the largest increase in support (as compared to the Republican record four years before) in the poorest and most hardly hit regions, i.e. the most depressed regions and those offering fewer economic opportunities (Fernández-Albertos, 2018, pp. 37–38). Moreover, Trump was most effective at capturing voters among the most economically vulnerable individuals, which indicates that the strong negative correlation between support for him and good economic performance extends to the individual level. More generally, changes in the occupational structure in the West that derive from the technological economy have generated rising economic insecurity and social deprivation among the most vulnerable citizens (Oesch & Rennwald, 2018). This certainly includes the unemployed, but also unskilled workers, whose wages and social prestige have been falling steadily under Republicans and Democrats, conservatives and social democrats, mainstream right-wing and left-wing parties (Kitschelt & Rhem, 2015). Feeling that 20 J. RAMA AND A. SANTANA social democrats and socialists are no longer able or willing to improve their situation (McGann & Kitschelt, 2005), unskilled workers are increasingly ready to listen to whomever promises to address their concerns. RWPPs leaders have identified the culprits of unskilled workers’ worsening situation and sketched out a plan to revert it. Professional politicians of traditional parties, international elites, and immigrants are among the usual culprits. Solutions emphasise restrictions to the mobility of goods (more protectionism) and persons (barriers to immigrants, who steal ‘our’ work). As it happened with the sirens, whose song lured Ulysses’ sailors to shipwreck, the promises of easy solutions have been for many voters too attractive to ignore. Millions have abandoned their social democrat and socialist parties and embraced RWPPs in the developed world (Ivarsflaten, 2005). That vulnerable individuals may also be lured by the populist promises of LWPPs should come as no surprise. Hobolt and Tilley (2016) argue that, after the Great Recession, the voters most affected by the economic crisis were more prone to support challenger parties of the left than cast a vote for mainstream parties. Consistently, Bosch and Durán (2017) show that economic variables were the key explanatory factors to account for the success of Podemos in the 2015 Spanish general elections, and Teperoglou et al. (2015) report that unemployed people were more likely to support Syriza in the 2014 European elections in Greece. Santana and Rama (2018) extend this line of thought and argue that the vulnerable electors attracted by LWPPs may have a different social profile. Especially in Southern Europe, were youth unemployment is remarkably high4, a significant contingent of youngsters with university education feels attracted to LWPPs because their discourse focuses on the economy (Hobolt & Tilley, 2016) while mainstream parties are unable to solve their economic and social problems. Considering that vulnerable electors are mobilised by RWPP and LWPP, we expect the following: H1: Those who anticipate material deprivation will tend to support LWPPs and RWPPs in the same way. In operational terms, we expect that being unemployed or lacking the money to make ends meet will not have statistically significant effects on the decision to vote for LWPPs or RWPPs. 1.2. Protest voting and anti-elite attitudes The most characteristic defining trait of populism is the distinction between a pure, virtuous people and a corrupt, dishonest elite (Mudde, 2004, p. 543). Moved perhaps by the desire to show their discontent with the elites, people who distrust the political elite are especially prone to vote for populist parties (Bergh, 2004; Bowler, Denemark, Donovan, & McDonnell, 2016; Oesch, 2008). Given that the distinction between the two antagonistic groups, the people and the elite, should apply to both LWPPs and RWPPs, anti-elite attitudes and beliefs should favour both. Yet, little is known as to which of the populist families is more successful at capitalising the potential of protest voting. The depiction of elites as a corrupt cartel acting only on behalf of their own interests and against the bulk of the population is perhaps the most characteristic trait of LWPPs. Indeed, leaders such as Pablo Iglesias (Podemos) in Spain, Beppe Grillo (M5S) in Italy, and Alexis Tsipras (Syriza) in Greece have coined the term casta to refer to the political establishment as a whole. Likewise, the diatribes against the traditional politicians are part of the daily menu of the speeches of Geert Wilders (PVV) in the Netherlands, Marine Le Pen (FN) in France, or Nigel Farage (UKIP) in the UK. EUROPEAN POLITICS AND SOCIETY 21 Thus, many studies have shown that the discourse of RWPPs recurrently employs a Manichean distinction between the pure people and the corrupt elite (Mudde & Rovira Kaltwasser, 2018). Taking all this into account, we advance our first hypothesis: H2: Those who mistrust in political parties will tend to support both LWPPs and RWPPs, but this will not provide any comparative advantage to either of them. In operational terms, we do not expect distrust to have statistically significant effects on the decision to vote for LWPPs or RWPPs. 1.3. Parochialism, globalisation, and attitudes towards the EU The role of cultural factors has also been emphasised by many scholars. According to the globalization hypothesis, the processes of denationalisation and political and economic integration have produced a contrast between the winners and the losers of globalisation. The losers (low-skilled) are supposed to be mobilised by populist right parties, whereas the winners (high-skilled) by left (populist or not) parties (Hernández & Kriesi, 2016). One of the key explanatory factors in this hypothesis, in its Western European version, is the European Union process of integration (the other is immigration and will be discussed later). Many scholars have found that negative attitudes towards EU integration explain the support for RWPP and that the link between RWPPs and their voters tends to be stronger when the former adopt extremely negative positions on this issue (Gómez-Reino & Llamazares, 2015). Yet, there are reasons to expect that LWPPs will also mobilise those who are critical of European integration. According to Hooghe and Marks (2017), ‘radical left parties reject European integration on the grounds that it hurts those who cannot take advantage of transnational mobility’ (p. 17). Moreover, protectionism has historically been an important issue in LWPPs’ rhetoric (March, 2008, p. 128) and seems to be a relevant dimension on radical left parties’ competition (Burgoon, 2013). Empirical evidence also suggests that those who are critical towards EU integration are more likely to vote for LWPP (March & Rommerskirchen, 2015). For Halikiopoulou, Nanou, and Vasilopoulou (2012, p. 524, 531), ‘issues of cultural identity and European integration’, including the ‘national-populism of the left’ mean that ‘Radical right- and left-wing parties side together’ (see in March, 2017). Taking all this into consideration, it seems reasonable to expect that critical attitudes towards globalisation and EU integration will be important for both LWPPs and RWPPs. Hence, we formulate our third hypothesis: H3: Those with overall negative attitudes towards the EU will be equally likely to vote for LWPPs and RWPPs. In operational terms, we do not expect attitudes towards the EU to have statistically significant effects on the decision to vote for LWPPs or RWPPs. 1.4. Nativism and attitudes towards migration One of the leitmotivs of the research on populism is that those who perceive immigration as a threat to their way of life tend to support challenger parties of the right (Hobolt & Tilley, 2016). Indeed, as Paul Taggart has pointed out in his contribution to the Oxford Handbook of Populism, ‘there has been an almost universal tendency to treat the parties that focus on this issue as being synonymous with populist parties in Western Europe’ (2017, p. 251). Not in vain, ‘the rise of populist radical right parties is linked to 22 J. RAMA AND A. SANTANA mass immigration and multiculturalism, and support for these parties is mostly an expression of nativism’ (Mudde & Rovira Kaltwasser, 2018, p. 13). The issue is less clear for LWPPs. On the one hand, Hooghe and Marks (2017) defend that ‘radical left parties (…) retain a commitment to working-class internationalism and do not take a strong position against immigration’ (p.17). However, Halikiopoulou et al. (2012) argue that ‘nationalism is the underlying feature that unites the radical right and the radical left’ (p. 505), and some case studies suggest that, even if the leaders of LWPPs do not emphasise the issue of migration in their programmes and speeches, their voters may feel more strongly about it: ‘Sinn Féin supporters are more likely to feel that there are already too many immigrants in Ireland’ (O’Malley, 2008, p. 971). In fact, in a comparative study of the 2014 European Parliament elections, Santana and Rama (2018) show that voters with anti-immigration attitudes are more prone to support LWPPs than other left parties. Given that the arguments are consistent for RWPPs but highly controversial for LWPPs, we expect the following: H4: Negative attitudes towards migration and immigrants will work in favour of RWPPs. In operational terms, we expect that having negative attitudes towards immigrants will have statistically significant positive effects on the decision to vote for a RWPP instead of a LWPP. 2. Data We use Round 8th of the ESS, 2016, which contains data on 44,387 individuals from 23 countries, including six European countries where RWPPs and LWPPs coexist (Germany, France, Italy, Iceland, Lithuania and The Netherlands). Data were collected during 2016 and 2017,5 using random probability sampling and face to face interviews. Our dependent variable distinguishes between respondents who declare having voted for a LWPP (Y = 1) and those who have opted for a RWPP (Y = 0). Abstainers and voters of non-populist parties are recoded as missing. We have drawn on other scholars’ work for the classification of parties as LWPP or RWPP. Classifications are liable to potential criticism and different authors have put forward different classifications of populist parties (Inglehart & Norris, 2016; Mair, 2013; van Kessel, 2015). Our approach has been rather conservative because we have followed the most widely accepted criteria in this field. Table 1 exhibits the parties analysed in this inquiry. The operationalisation of our independent variables is as follows (see Table A1 in the Appendix with the number of observations, mean, standard deviation, minimum and maximum for each variable). In model 1, we consider two indicators of expected material deprivation: the perception that one is likely to be unemployed in the next 12 months (a four-item ordinal variable, recoded as a dummy, 1 = likely or very likely, 0 = not at all or not very likely) and the belief that one’s household necessities will be uncovered due to money shortages in the next 12 months (again, a four-item ordinal variable recoded as a dummy, 1 = likely or very likely, 0 = not at all or not very likely). Model 2 focuses on the anti-elite dimension which, to a varying extent, characterises populist parties. The best measure available in the ESS, 2016 questionnaire is mistrust in political parties (an eleven-point scale variable, rescaled to grow with mistrust, 0 = complete trust, 10 = no trust at all). Although it would have been interesting to include mistrust in the national parliament as well, its correlation with the former was above 0.7. Table 1. Classification of parties into LWPPs and RWPPs. Family France (FR) LWPPb RWPPc Iceland (IS) LWPP RWPP Italy (IT) LWPPd RWPPe,f Party name GEa EP LR LRE AE CS EUP EUB RI 1.7 9.6 1.1 5.7 9.0 9.5 7.2 7.9 3.3 2.2 1.3 1.1 2.4 9.8 4.7 8.9 6.7 7.9 3.4 7.3 7.0 5.6 10.0 8.8 4.0 6.3 9.8 6.4 3.2 5.8 3.3 2.8 n.a. 3.8 1.3 1.1 2.1 1.4 4.3 9.5 7.8 8.8 Front de Gauche (FG); then La France insoumise (FI) Front National (FN) Left Front; then Unbowed France National Front 19.6 (#4) 21.3 (#2) 6.6 (#6) 24.8 (#1) Píratar (P) Framsóknarflokkurinn Flokkur fólksins (F) The Pirate Party Progressive Party People’s Party 9.2 (#6) 10.7 (#5) 6.9 (#7) n.a. (not applicable) Five Star Movement Northern League; then League Forward Italy / Let’s Go Italy Brothers of Italy 32.7 17.4 14.0 4.4 21.2 6.2 16.8 3.7 Socialist Party 50PLUS Party for Liberty 9.1 (#6) 3.1 (#10) 13.1 (#2) 9.6 (#5) 3.7 (#10) 13.3 (#3) 1.0 5.3 9.3 1.0 3.7 4.6 6.6 5.8 9.4 3.3 2.5 5.0 3.3 n.a. 1.3 1.5 5.5 1.0 4.4 5.0 9.9 The Left National Democratic Party of Germany Alternative for Germany 7.4 (#5) 0.1 (#11) 12.6 (#3) 9.7 (#4) 1.0 (#11) 7.1 (#5) 1.2 n.a. 8.9 1.3 5.3 8.3 5.4 9.1 7.8 3.1 3.0 3.8 5.3 1.9 2.4 2.7 1.2 2.5 4.0 9.0 9.3 4.4 6.6 4.7 4.4 4.7 7.5 3.6 6.3 4.7 n.a. 4.3 2.9 4.4 6.5 Movimento 5 Stelle (M5S) Lega Nord (LN); then La Lega (L) Forza Italia (FI) Fratelli d’Italia (FdI) The Netherlands (NL) LWPP Socialistische Partij (SP) RWPP 50PLUS Partij voor de Vrijheid (PVV) Germany (DE) LWPP Die Linke RWPP Nationaldemokratische Partei Deutschlands (NPD) Alternative für Deutschland (AfD) Lithuania (LT) g LWPP Darbo Partija (DP) RWPP Tvarka ir Teisingumas (TT) Labour Party Order and Justice (#1) (#3) (#4) (#5) 4.9 (#8) 5.6 (#7) (#2) (#4) (#3) (#7) 12.4 (#5) 14.3 (#4) 23 GE: percentage of votes and position in the ranking of votes obtained in the country’s most recent general elections, which took place in April 2017 (France, presidential, 1st round), March 2018 (Italy), March 2017 (The Netherlands), September 2017 (Germany) and October 2016 (Lithuania). EP: As in GE, but for the 2014 European Parliament elections. LR: Overall LR placement. LRE: Economic LR placement. AE: Anti-elite salience (salience of anti-establishment and anti-elite rhetoric: 0 = not salient at all, 10 = extremely important). CS: Salience of reducing political corruption (0 = not salient at all, 10 = extremely important). EUP: position of the party leadership in 2014 on the powers of the European Parliament (1 = strongly opposes, 7 = strongly in favour). EUB: position of the party leadership on EU authority over member states’ economic and budgetary policies (1 = strongly opposes, 7 = strongly in favour). RI: Restrictive immigration policy (0 = fully opposed to a restrictive policy on immigration, 10 = fully in favour). b FG was a coalition of the Parti Communiste Français (PCF), Parti de Gauche (PG) and Gauche Unitaire (GU), among other far, radical or illiberal left parties. Jean-Luc Mélenchon was its leading candidate in the 2014 EP elections. In 2016, he founded FI (known in English as Unbowed France, Unsubmissive France, or Untamed France), being its leading candidate in the 2017 Presidential elections. c On 1 June 2018, the FN was renamed as Rassemblement National (RN), i.e. National Rally. d Although we justify why we consider the M5S as a LWPP (see the source of this table) we want to reinforce our decision by the results exhibited on the expert survey conducted by Di Virgilio, Giannetti, and Pedrazzani (2015) during the 2013 national elections, which shows that the mainstream centre-left Democratic Party (PD) and the M5S are very close to each other on the economic left-right dimension. In this vein, the Comparative Manifestos Project (CMP) locates the M5S on the left of both the PD and Left Ecology Freedom (SEL). For more information, see (Franchino & Negri, 2018). e On 28 January 2018, LN was rebranded as Lega, although the official name was not formally changed. f On 29 March 2009, Il Popolo della Libertà was created to assemble Berlusconi’s Forza Italia (FI) and Fini’s Alleanza Nazionale. On 16 November 2013, it was dissolved and FI was relaunched. g Following van Kessel (2015, p. 58) we do not include The Way of Courage as an RWPP due to its idiosyncratic character. Source: own elaboration based on Segatti and Capuzzi (2016, p. 54) and Fella and Ruzza (2013, p. 49) for the case of the M5S, Bobba and Roncarlo (2018) for the PdL (also Pappas, 2016, p. 34) and the FdI, and Rensmann, de Lange, and Couperus (2017, p. 155) for the League in Italy; Otjes and Louwerse (2015) for the SP, Hameleers, Bos, and de Vreese (2017, p. 138) for 50PLUS and Rensmann et al. (2017, p. 155) for the PVV in the Netherlands; Alekonis and Matkeviciené (2016) for the DP and the TT in Lithuania; Rensmann et al. (2017, p. 155) for the FN and Stockemer (2018) for the FDG in France; March (2007) for Die Linke, Rensmann et al. (2017, p. 155) for AfD and Nociar and Thomeczek (2018) for the NPD in Germany; finally, Eiermann, Mounk, and Gultchin (2017) for P, F and the Progressive Party in Iceland. Data on the last columns have been obtained from the 2014 Chapell-Hill Expert Survey, or CHES 2014 (Bakker et al., 2015). EUROPEAN POLITICS AND SOCIETY a Party name (English) 24 J. RAMA AND A. SANTANA In model 3, we contemplate two eleven-point scale variables capturing attitudes towards the EU, which we inverted so that smaller values correspond to more cosmopolitan and bigger values to more parochial orientations: how unfavourable to EU unification are the interviewees (10 = EU unification has gone too far, 0 = unification should be pushed further) and emotional detachment from Europe (10 = not at all emotionally attached, 0 = very emotionally attached). In model 4, we include two variables related to attitudes towards migrants, which we rescaled so that smaller values correspond to more positive attitudes towards them and bigger ones to more negative ones. The first one is an eleven-point scale on the effect of immigrants on the country’s economy (rescaled as 10 = bad, 0 = good), which captures the economic dimension of citizens’ attitudes towards migration. The second is a question on how inclined interviewees are to allow immigrants of different races to live in the country (a four-item ordinal variable, recoded as a dummy, 0 = allow some or many, 1 = allow a few or none), and captures the cultural dimension of attitudes towards migration. Model 5 is the full model, in which all the variables of models 1–4 are simultaneously considered; this is possible, as the VIFs (variance inflation factors) of all independent variables are well below the levels that would rise concerns of collinearity problems (Table A1). In all models, we control for ideology (0 = left, 10 = right) and for a series of sociodemographic variables, namely gender (1 = woman, 0 = man), age (in years), and education (university level = 1). All the models include survey weights (post-stratification weights, which subsume design weights and therefore redress the different probabilities of being surveyed as well as the differences in the response rates), country dummies and country clusters (to address the hierarchical structure of the data).6 3. Results Figure 1 displays the results of the logistic regressions of the main models7 (see also Table A2 in the Appendix). The graph shows the independent variables standardised (standard deviation = 1). As we have mentioned, we run the logistic regression with country clusters and employing the survey weights. Additionally, we run all the models with the same number of observations to prevent results from changing because of the number of cases. Model 1, focused on the economic deprivation thesis, evinces that neither the probability of being unemployed in the next 12 months nor the probability that household necessities will be uncovered in the next 12 months have a statistically significant effect on the probability of voting for a LWPP instead of a RWPP. Model 2 reveals that the mistrust in political parties does not condition the casting of a vote in favour of a LWPP or a RWPP. Model 3, related to the EU dimensions, exhibits a negative relationship of parochial attitudes to EU unification with LWPP support: the belief that EU unification has gone too far reduces the propensity to vote for LWPPs as compared to RWPPs. Model 4 shows that attitudes towards immigrants (including perceptions about the impact of immigrants on the country’s economy and attitudes towards allowing immigrants of different races to live in their country) do not condition the support for a LWPP as against a RWPP. However, the full model (Model 5) amends the results of Model 2: those who mistrust political parties are more prone to support a LWPP than a RWPP. The elite dimension fits better with LWPPs than RWPPs, at least when both types of parties compete in the same country. Thus, although RWPPs also employ a discourse that criticises the corrupt elite and EUROPEAN POLITICS AND SOCIETY 25 Figure 1. Coefficient plots for LWPP support vs RWPP. extols the good people, LWPPs enjoy an advantage on this dimension. Additionally, Model 5 reveals that those with difficulties to pay for their household necessities are more likely to vote for a LWPP, whereas the probability of being unemployed is not statistically significant. On top of the country dummies, all models include controls for gender, age, education and left-right ideology (see Table A2 for the detailed results). Age displays consistently negative effects, indicating that older individuals are less prone to vote for LWPPs than for RWPPs. Unsurprisingly, left-right ideological self-placement also displays a significantly negative effect. Although the effects of being female and having high education are negative, they fail to attain statistical significance in any of the models (a more nuanced measure of education, distinguishing among four levels of education, also fails to attain statistical significance). Given that Lithuania is the only Eastern European country and that its parties appear relatively centrist from the position data (Table 1), we have rerun Model 5 excluding Lithuania: as Figure A1 in the Appendix show, results remain almost unaltered. Figures 2–5 display the effects of our four key groups of independent variables upon the probability of voting for LWPPs vis à vis RWPPs (all other variables are held at their means). Figure 2 confirms that in those countries in which LWPPs and RWPPs coexist, voters feel that both can solve their unemployment problems equally well (or bad). The 95 confidence intervals of the predicted probability of voting of those who feel likely or very likely to become unemployed clearly overlap with the ones for those who feel unlikely or very unlikely to lose their jobs and, as the footnote clarifies, a formal contrast of the difference between both clearly fails to attain statistical significance. Yet, as the right panel shows, those who fear to experience difficulties to pay for their household necessities are more prone to vote for LWPP. Hence, our first hypothesis is partially rejected. The significant effect of the second measure of material deprivation may owe to the left-wing parties’ traditional emphasis on economic redistribution (Lipset, Lazarsfeld, Barton, & Linz, 1954) and the consequent alignment between voters and parties along this dimension. Figure 3 underlines that the propensity to cast a vote for a LWPP increases as the mistrust in political parties grows. This effect is not only statistically but also substantively significant: whereas for those who have complete trust in political parties the predicted 26 J. RAMA AND A. SANTANA Figure 2. Predictive margins for the effects of key independent variables upon the probability of voting LWPPs vs RWPPs, material deprivation. Figure 3. Predictive margins for the effects of key independent variables upon the probability of voting LWPPs vs RWPPs, elite dimension. probability of voting for a LWPP is 0.37, this figure rises to 0.54 for those who do not trust in them at all. Thus, in those countries in which LWPPs and RWPPs coexist, LWPPs take advantage of their competitors of the right on the populist anti-elite dimension. This runs EUROPEAN POLITICS AND SOCIETY 27 Figure 4. Predictive margins for the effects of key independent variables upon the probability of voting LWPPs vs RWPPs, EU dimension. Figure 5. Predictive margins for the effects of key independent variables upon the probability of voting LWPPs vs RWPPs, immigration dimension. 28 J. RAMA AND A. SANTANA counter to our second hypothesis, which must therefore be rejected. What could explain this unexpected result? Our take is that the anti-elite discourses of RWPP leaders are often intertwined with Eurosceptic and anti-immigration arguments; for instance, they justify their criticism towards elites because they claim that the latter facilitate the uncontrolled entrance of migrants. Using Mudde and Rovira Kaltwasser’s words: ‘these parties use populist rhetoric to blame “the elite” for help bringing the “aliens” into the country and for ignoring the problems that “natives” are suffering’ (2018, p. 12). As compared to this somewhat diffuse reproach, the criticism of LWPPs tends to be more focused and recurrent as they portray the casta as enemy number one. However, Figure 4 underlines that none of the measures of parochialism affect the competition among populist parties, given the very large confidence intervals associated to the predictions related to both variables. At most, it can be noticed that those who believe that EU integration should be pushed further or has not gone too far (values 0–5 of the ‘EU integration has gone too far’ variable) are more likely to vote for LWPPs than those who think that it has gone too far (values 6–10). These minor differences may owe to the fact that LWPPs voters do not consider belonging to the EU as a bad thing per se; however, the overlapping of the confidence intervals and the lack of significance of the variable in Figure 1 and Table A2 suggest that their acceptance of the EU integration process is, at most, conditional. That is, they believe in a different type of EU which is not an elitist capitalist project that ignores the interests of ‘the common working man’ (Kriesi & Pappas, 2015). As to the other measure of parochialism (right panel of Figure 4), even if the slope of the effect of emotional detachment to Europe is also negative, the slope is rather flat, and the 95 per cent confidence intervals are very large, precluding any possible effects of this variable on the propensity to vote for a LWPP or a RWPP. Hence, our third hypothesis is confirmed, and it is so for the two measures of parochialism. Finally, Figure 5 displays the propensity to cast a vote for a LWPP instead of a RWPP upon the immigration dimensions. Contrary to our fourth hypothesis, the perception that immigrants are bad for the county’s economy and the opposition to the prospect that immigrants of different race come and live in the country, do not condition the support for a LWPP or a RWPP. Although the left panel illustrates that on a 0–10 scale, the propensity to vote for a LWPP is higher among those who consider immigrants good for the country’s economy, the 95 per cent confidence intervals overlap for all values of this first measure of nativism. The overlapping among the two categories of the binary measure of the opposition to the entrance of immigrants of other races is even more evident (see the right panel). Again, the lack of significance is confirmed by a formal contrast of the predictive margins of this variable (the p-value of the Wald test being 0.120). Thus, nativist attitudes do not affect the left versus right populist competition, and our fourth hypothesis must be rejected. This result fits well the thesis of Dancygier and Margalit (2018), who argue that mainstream parties from the left and right converge on immigration issues since the good days of anti-immigration parties. It is also consistent with the situation in Italy, Germany or the Netherlands, where LWPPs have adopted the common RWPPs’ discourses against immigration. This is the recent case of the German party Die Linke: notwithstanding the fact that almost 400,000 electors switched from Die Linke to AfD, the co-founder of the radical left party, Sahra Wagenknecht, created a new movement inside the party, called Aufstehen (Get Up) that favours tighter controls for migrants trying to enter the labour market.8 EUROPEAN POLITICS AND SOCIETY 29 4. Conclusions In this paper, we focus on the factors favouring the vote for left-wing populist parties (LWPPs) versus right-wing populist parties (RWPPs). Instead of drawing conclusions about the profiles of RWPP and LWPP voters from the comparison of results obtained for different sets of countries and, therefore, samples, we study only those countries where both LWPPs and RWPPs compete. Although there are other interesting questions in the comparative electoral sociology of LWPPs and RWPPs voters, such as the extent to which different sociodemographic characteristics shape the competition between the two populist families, our endeavour focuses on the factors related to the support for populist parties, namely material deprivation, attitudes towards political elites, parochial and anti-globalisation motivations, and nativist attitudes. Figure 6 will help to summarise our main results and enrich the scope of our conclusions beyond the direct competition of LWPPs and RWPPs by using a nominal specification of the dependent variable (LWPPs = 2, RWPPs = 1, non-populist parties = NPP = 0): Our results have important theoretical implications. First, material deprivation theories state that the so-called losers of globalisation tend to support RWPPs; our results show that, when they compete with LWPPs, this effect is fully absorbed by LWPPs (material deprivation increases the likelihood of voting for LWPPs instead of either RWPPs or NPPs, without significant differences among the latter two). Second, the core distinction between populist and non-populist parties should relate to attitudes towards elites; our results show that when both breeds of populist parties Figure 6. Coefficient plots for LWPP vs RWPP vs NPP support (baseline, NPP). 30 J. RAMA AND A. SANTANA compete, such attitudes only increase the likelihood of voting LWPPs (again, RWPPs and NPPs are indistinguishable). Third, the globalisation hypothesis has been used to justify support for both types of populist parties. Our results confirm that this is true for the detachment from Europe, which increases the likelihood of voting a populist party (RWPPs and LWPPs being indistinguishable) versus NPPs. Nevertheless, attitudes against EU unification only favour RWPPs. Finally, nativist arguments have recurrently proven to be useful to understand support for RWPPs. Interestingly, negative attitudes towards immigration, whether cultural or economic oriented, do not have a statistically significant effect when we focus the attention on the populist ground, but the second measure of nativism (restricting immigrants of other races) increases the propensity of voting for RWPPs instead of NPPs. All in all, when the two breeds of populist parties are on the same playing field, they appear to be more similar in the eyes of their voters than what it is usually assumed. Notes 1. See Rydgren (2007) for a review of demand- and supply-side explanations of the success of radical right-wing parties. 2. Findings are neither monolithic for LWPPs nor for RWPPs. For instance, Cordero, Orriols, and Teperoglou (2018) fail to find statistically significant effects of age for SYRIZA or education for Podemos. And while the main electoral support for the United Kingdom Independence Party (UKIP) is among the oldest and without university education (Ford & Goodwin, 2014), the supporters of Golden Down in Greece and the National Front in France are young and highly educated (Stockemer, 2017; Teperoglou et al., 2015, pp. 348–350). 3. In empirical terms it means that we select those voters of a LWPP and those who vote for a RWPP in those countries where both parties coexist. Thus, our dependent variable values 1 when a voter supports a LWPP and 0 when (s)he supports a RWPP. In other words, our aim is not to find differences among the electoral bases of right-wing parties, socialdemocratic parties, radical left parties, and populist from the left and the right, but to determine the effect of a set of variables on the competition among populist parties. 4. See the ECB report ‘The impact of the Economic Crisis in the Euro Area Labour Market’. Online access: https://goo.gl/7n3A3X. 5. The specific months for each country can be consulted in the ESS’ Data Documentation Report: https://www.europeansocialsurvey.org/docs/round8/survey/ESS8_data_documentat ion_report_e02_1.pdf. 6. Given the low number of level two observations, random effects multilevel models are strongly discouraged. We also estimated a multilevel fixed effects regression which yielded qualitatively the same results. However, as the statistical program (Stata) does not allow us to include weights at the individual level using this kind of technique, and given the importance of weights in the ESS, we opt to report the results for the most similar approximation: logistic regression with country dummies. 7. Each horizontal line in Figure 1 represents an independent variable of the model, the point standing for the best estimation of its effect upon the dependent variable, and the line, for its 95 per cent confidence interval. If a confidence interval crosses the vertical line drawn at the origin (zero) of the horizontal axis, the effect of the variable is not statistically significant. If it does not cross it and is located at its right, the effect is positive, whereas if it is located at its left, the effect is negative. 8. See the articles published in The Guardian: https://www.theguardian.com/world/2018/sep/04/ german-politicians-launch-leftwing-stand-up-movement and CNN: https://edition.cnn.com/ 2018/09/07/opinions/sahra-wagenknecht-opinion-intl/index.html. EUROPEAN POLITICS AND SOCIETY 31 Acknowledgement We wish to express our gratitude to Susana Aguilar (UCM) for her assistance on the revision of this manuscript, to the anonymous reviewers for their insightful comments on the first version of this manuscript, and to the participants in the First Spanish Congress of the European Social Survey held in September 2018 in Madrid. ORCID José Rama http://orcid.org/0000-0002-3990-1696 Andrés Santana http://orcid.org/0000-0003-2594-1360 References Alekonis, G., & Matkeviciené, R. (2016). Populism in Lithuania: Defending the research tradition. Baltic Journal of Law and Politics, 9(1), 26–48. Anastasakis, O. (2013). The far right in Greece and the theory of the two extremes. OpenDemocracy. Retrieved from http://www.opendemocracy.net/othon-anastasakis/far-right-in-greece-andtheory-of-twoextremes-0 Arzheimer, K. (2009). Contextual factors and the extreme right vote in Western Europe, 1980–2002. American Journal of Political Science, 53(2), 259–275. Arzheimer, K., & Carter, E. (2006). Political opportunity structures and right-wing extremist party success. European Journal of Political Research, 45, 419–443. Bakker, R., Edwards, E., Hooghe, L., Jolly, S., Marks, G., Polk, J., … Vachudova, M. (2015). 2014 Chapel Hill expert survey. Version 2015.1. Chapel Hill, NC: University of North Carolina, Chapel Hill. Retrieved from chesdata.eu. Bergh, A. (2004). The universal welfare state: Theory and the case of Sweden. Political Studies, 52(4), 745–766. Bobba, G., & Roncarlo, F. (2018). The likeability of populism on social medial in the 2018 Italian General Election. Italian Political Science, 13(1), 51–62. Bosch, A., & Durán, I. M. (2017). How does economic crisis impel emerging parties on the road to elections? The case of the Spanish Podemos and Ciudadanos. Party Politics. doi:10.1177/ 1354068817710223 Bowler, S., Denemark, D., Donovan, T., & McDonnell, D. (2016). Right-wing populist party supporters: Dissatisfied but not direct democrats. European Journal of Political Research, 56, 70–91. Burgoon, B. (2013). Inequality and anti-globalization backlash by political parties. European Union Politics, 14(3), 408–435. Clark, A., Bottom, K., & Copus, C. (2008). More similar than they’d like to admit? Ideology, policy and populism in the trajectories of the British National Party and respect. British Politics, 3, 511–534. Cordero, G., Orriols, L. L., & Teperoglou, E. (2018). Syriza, M5S y Podemos: Los tres campeones del votante enfadado en el Sur de Europa. Revista Jurídica de la Universidad Autónoma de Madrid, 37, 105–131. Dancygier, R. M., & Margalit, Y. (2018). The evolution of the immigration debate: A study of party positions over the last half-century. doi:10.2139/ssrn.3192913 Di Virgilio, A., Giannetti, D., & Pedrazzani, A. (2015). Party competition in the 2013 Italian elections: Evidence from an expert survey. Government and Opposition, 50(1), 65–89. Eatwell, R. (2003). Ten theories of the extreme right. In P. Merkl & L. Weinberg (Eds.), Right-wing extremism in the twenty-first century. London: Frank Cass. Retrieved from http://opus.bath.ac.uk/11126/ Eiermann, M., Mounk, Y., & Gultchin, L. (2017). European populism: Trends, threats and future prospects. Tony Blair Institute for Global Change. Retrieved from https://institute.global/insight/renewingcentre/european-populism-trends-threats-and-future-prospects ESS Round 8: European Social Survey Round 8 Data. (2016). Data file edition 2.0. NSD – Norwegian Centre for Research Data, Norway – Data Archive and distributor of ESS data for ESS ERIC. 32 J. RAMA AND A. SANTANA Faye, J.-P. (2002). Le Siècle des idéologies. Paris: Pocket Agora. Fella, S., & Ruzza, C. (2013). Populism and the fall of the centre-right in Italy: The end of the Berlusconi model or a new beginning? Journal of Contemporary European Studies, 21(1), 38–52. Fernández-Albertos, J. (2015). Los Votantes De Podemos: Del Partido De Los Indignados Al Partido De Los Excluidos (Colección Alternativas, 12). Madrid: Los Libros de la Catarata y Fundación Alternativas. Fernández-Albertos, J. (2018). Antisistema. Madrid: Libros de la Catarata. Ford, R., & Goodwin, M. (2014). Understanding UKIP: Identity, social change and the left behind. The Political Quarterly, 85(3), 277–284. Franchino, F., & Negri, F. (2018). The fiscally moderate Italian populist voter: Evidence from a survey experiment. Party Politics. doi:10.1177/1354068818761180 Funke, M., Schularick, M., & Trebesch, C. (2015). Going to extremes: Politics after financial crisis, 1870– 2014 (CESIFO Working Papers n° 5553 Category 7). Monetary Policy and International Finance. Gómez-Reino, M., & Llamazares, I. (2015). New left populism in contemporary Spain? The upsurge of Podemos. Paper presented at the 22nd international conference of Europeanists, Paris. Halikiopoulou, D., Nanou, K., & Vasilopoulou, S. (2012). The paradox of nationalism: The common denominator of radical right and radical left euroscepticism. European Journal of Political Research, 51(4), 504–539. Hameleers, M., Bos, L., & de Vreese, C. (2017). The Netherlands: A heartland full of insights into populist communication. In T. Aalberg, F. Esser, C. Reinemann, J. Stromback, & C. de Vreese (Eds.), Populist political communication in Europe (pp. 138–150). New York, NY: Routledge. Hernández, E., & Kriesi, H. (2016). The electoral consequences of the financial and economic crisis in Europe. European Journal of Political Research, 55(2), 203–224. doi:10.1111/1475-6765.12122 Hobolt, S. B., & Tilley, J. (2016). Fleeing the centre: The rise of challenger parties in the aftermath of the euro crisis. West European Politics, 39(5), 971–991. Hooghe, L., & Marks, G. (2017). Cleavage theory meets Europe’s crises: Lipset, Rokkan, and the transnational cleavage. Journal of European Public Policy, 25(1), 109–135. Inglehart, R., & Norris, P. (2016). Trump, Brexit and the rise of populism: Economic have-nots and cultural backlash (Faculty Research Working Paper Series). Ivalid, G., Lanzone, M. E., & Dutozia, J. (2016). Explaining regional support for national-populist parties: The case of FN in France. Populism, Regionalism and Nationalism in Western European Party Mobilisation. The Territorial Dimension. Lausanne. Ivarsflaten, E. (2005). The vulnerable populist right parties: No economic realignment fuelling their electoral success. European Journal of Political Research, 44(3), 465–492. Judis, J. B. (2016). The populist explosion: How the great recession transformed American and European politics. New York, NY: Columbia Global Reports. Kestilä, E., & Söderlund, P. (2007). Subnational political opportunity structures and the success of the radical right: Evidence from the March 2004 regional elections in France. European Journal of Political Research, 46(6), 773–796. Kitschelt, H., & Rhem, P. (2015). Party alignments: Change and continuity. In P. Beramendi, S. Häusermann, H. Kitschelt, & H. Kriesi (Eds.), Politics of advanced capitalism (pp. 179–201). Cambridge: Cambridge University Press. Kriesi, H., & Pappas, T. (Eds.). (2015). European populism in the shadow of the great recession. Colchester: ECPR Press. Lipset, S. M., Lazarsfeld, P. F., Barton, A. H., & Linz, J. J. (1954). The psychology of voting: An analysis of political behavior. In G. Lindzey (Ed.), Handbook of social psychology (pp. 1124–1175). Cambridge, MA: Addison-Wesley. Lubbers, M., Gijsberts, M., & Scheepers, P. (2002). Extreme right-wing voting in Western Europe. European Journal of Political Research, 41(3), 345–378. Lubbers, M, & Scheepers, P. (2007). Explanations of Political Euroescepticism at the Individual Regional and National Levels. European Societies, 9(4), 643–689. Mair, P. (2013). Ruling the void. The hollowing of western democracy. London: Verso Books. March, L. (2007). From Vanguard of the proletariat to Vox Populi: Left-populism as a ‘shadow’ of contemporary socialism. SAIS Review of International Affairs, 27(1), 63–77. EUROPEAN POLITICS AND SOCIETY 33 March, L. (2011). Radical left parties in contemporary Europe. Abingdon: Routledge. March, L. (2017). Left and right populism compared: The British case. The British Journal of Politics and International Relations, 19(2), 282–203. March, L. (2008). Contemporary far left parties in Europe: From marxism to the mainstream. Bonn: Friedrich Ebert Stiftung. March, L., & Rommerskirchen, C. (2015). Out of left field? Explaining the variable electoral success of European radical left parties. Party Politics, 21(1), 40–53. McGann, A. J., & Kitschelt, H. (2005). The radical right in the Alps. Party Politics, 11(2), 147–171. Mols, F., & Jetten, J. (2017). The wealth paradox. Economic prosperity and the hardening of attitudes. Cambridge: Cambridge University Press. Mudde, C. (2004). The populist zeitgeist. Government and Opposition, 39(4), 541–563. doi:10.1111/j. 1477-7053.2004.00135.x Mudde, C. (2007). Populist radical right parties in Europe. Cambridge: Cambridge University Press. Mudde, C., & Rovira Kaltwasser, C. (2018). Studying populism in comparative perspective: Reflections on the contemporary and future research agenda. Comparative Political Studies. doi:10.1177/ 0010414018789490 Nociar, T., & Thomeczek, P. (2018). Far right politics in Germany: From fascism to populism? The London School of Economics and Political Science. Retrieved from http://blogs.lse.ac.uk/europpblog/2018/ 01/24/far-right-politics-in-germany-from-fascism-to-populism/ Oesch, D. (2008). Explaining workers’ support for right-wing populist parties in Western Europe: Evidence from Austria, Belgium, France, Norway, and Switzerland. International Political Science Review, 29(3), 349–373. Oesch, D., & Rennwald, L. (2018). Electoral competition in Europe’s new tripolar political space: Class voting for the left, centre-right and radical right. European Journal of Political Research, 57(4), 783–807. O’Malley, E. (2008). Why is there no radical right party in Ireland? West European Politics, 31(5), 960–977. Orriols, L., & Cordero, G. (2016). The breakdown of the Spanish two-party system: The upsurge of Podemos and Ciudadanos in the 2015 general election. South European Society and Politics, 21 (4), 469–492. Otjes, S., & Louwerse, T. (2015). Populists in parliament: Comparing left-wing and right-wing populism in the Netherlands. Political Studies, 63(1), 60–79. Pappas, T. (2016). Distinguishing liberal democracy’s challengers. Journal of Democracy, 27(4), 22–36. Ramiro, L. (2016). Support for radical left parties in Western Europe: Social Background, ideology and political orientations. European Political Science Review, 8(1), 1–23. Rensmann, L., de Lange, S., & Couperus, S. (2017). Populism and the remaking of (il)liberal democracy in Europe. Politics and Governance, 5(4). doi:10.17645/pag.v5i4.1328 Rivero, A., Zarzalejos, J., & del Palacio, J. (2017). Geografía del Populismo: Un viaje por el universo del populismo desde sus orígenes hasta Trump. Madrid: Tecnos. Rooduijn, M. (2017). What unites the voter bases of populist parties? Comparing the electorates of 15 populist parties. European Political Science Review. doi:10.1017/S1755773917000145 Rooduijn, M., & Akkerman, T. (2015). Flank attacks: Populism and left-right radicalism in Western Europe. Party Politics. doi:10.1177/1354068815596514 Rovny, A., & Rovny, J. (2017). Outsiders at the ballot box: Operationalizations and political consequences of the insider–outsider dualism. Socio-Economic Review, 15(1), 161–185. https://doi.org/ 10.1093/ser/mww039 Rydgren, J. (2007). The sociology of the radical right. Annual Review of Sociology, 33(1), 241–262. Santana, A., & Rama, J. (2018). Electoral support for left wing populist parties in Europe: Addressing the globalization cleavage. European Politics and Society, 19(5), 558–576. Segatti, P., & Capuzzi, F. (2016). Five star movement, Syriza and Podemos: A mediterranean model? In A. Martinelli (Ed.), Populism on the rise. Democracies under challenge? (pp. 47–73). Milan: Instituto per gli Studi di Politica Internazionale. Springford, J., & Tilford, S. (2017). Populism – culture or economics? Insight, Centre for European Reform. 34 J. RAMA AND A. SANTANA Stockemer, D. (2017). The economic crisis (2009–2013) and electoral support for the radical right in Western Europe – some new and unexpected findings. Social Science Quarterly, 98(5), 1536– 1553. Stockemer, D. (2018). Populism around the world. A comparative perspective (pp. 27–48). Cham: Springer. Taggart, P. (2017). Populism in Western Europe. In C. Rovira Kaltwasser, P. A. Taggart, P. Ochoa Espejo, & P. Ostiguy (Eds.), The Oxford handbook of populism (pp. 248–264). Oxford: Oxford University Press. Teperoglou, E., Tsatsanis, E., & Nicolacopoulos, E. (2015). Habituating to the new normal in a postearthquake party system: The 2014 European election in Greece. South European Society and Politics, 20(3), 333–355. Van der Brug, W, Fennema, M, & Tillie, J. (2000). Anti-immigrant parties in Europe: Ideological or protest vote? European Journal of Political Research, 37, 77–102. van der Waal, J., & Koster, W. (2017). Populism and support for protectionism: The relevance of opposition to trade openness for leftist and rightist populist voting in the Netherlands. Political Studies. doi:10.1177/0032321717723505 van Kessel, S. (2015). Dutch populism during the crisis. In H. Kriesi & T. S. Pappas (Eds.), Populism in the shadow of the great recession (pp. 109–124). Colchester: ECPR Press. Visser, M., Lubbers, M., Kraaykamp, G., & Jaspers, E. (2014). Support for radical left ideologies in Europe. European Journal of Political Research, 53, 541–558. Appendix Figure A1. Coefficient plots for LWPP support vs RWPP with and without Lithuania. 35 EUROPEAN POLITICS AND SOCIETY Table A1. Summary statistics of the dependent, independent and control variables Count Weight Dependent variable - LWPP instead of RWPP 1372 1287 Main independent variables - Probability of being unemployed 34,100 34,347 - Probability of necessities uncovered 40,612 40,546 - Mistrust in parties 43,532 43,465 - EU unification gone too far 40,776 40,648 - Detachment from Europe 43,699 43,668 - Immigrants bad for economy 42,825 42,791 - Restrict immigrants of other races 43,136 43,131 Control variables: - Sex (female) 44,378 44,379 - Age (years) 42,993 42,510 - Education (university) 44,170 44,157 - LR Ideology 38,583 38,415 Notes: statistics computed using post-stratification weights (including Source: Own elaboration (ESS, 2016). Mean SD Min Max vif 0.48 0.50 0 1· 0.22 0.25 6.35 5.08 4.53 4.98 0.46 0.42 0.43 2.38 2.66 2.72 2.52 0.50 0 0 0 0 0 0 0 1 1 10 10 10 10 1 1.37 1.40 1.16 1.39 1.31 1.75 1.51· 0.52 0.50 48.30 18.02 0.22 0.41 5.17 2.23 design weights). 0 18 0 0 1 100 1 10 1.02 1.06 1.08 1.19 Table A2. Logistic regressions for LWPP support vs RWPP M1 Material deprivation Likely to be unemployed Likely to have financial needs Anti-elitism Mistrust in parties Parochialism EU unification gone too far Detachment from Europe Nativism Immigrants bad for economy Restrict immigrants of other races Control variables Female Age (years) University LR Ideology Country fixed effects Germany M2 M3 M4 0.136 (0.078) 0.238 (0.152) 0.141 (0.088) 0.391* (0.155)· 0.058 (0.044) 0.106* (0.052)· −0.075 (0.039) −0.090* (0.039) −0.010 (0.039) −0.213 (0.174) −0.027* (0.013) 0.395 (0.345) −0.617*** (0.087) M5 −0.200 (0.168) −0.028* (0.013) 0.379 (0.359) −0.620*** (0.088) −0.201 (0.171) −0.028* (0.013) 0.259 (0.385) −0.595*** (0.082) −0.012 (0.054)· −0.105 (0.069) −0.100 (0.073) −0.397 (0.279) −0.363 (0.249)· −0.175 (0.145) −0.028* (0.011) 0.172 (0.286) −0.581*** (0.069) −0.130 (0.137) −0.026* (0.011) 0.244 (0.281) −0.576*** (0.068)· 1.743*** 1.622*** 1.499*** 1.595*** (0.074) (0.084) (0.069) (0.073) Iceland 1.283*** 1.100*** 0.719*** 0.963*** (0.213) (0.130) (0.129) (0.215) Italy 2.009*** 1.960*** 1.971*** 1.975*** (0.212) (0.236) (0.221) (0.212) Lithuania 0.492*** 0.371*** 0.337*** 0.277*** (0.082) (0.109) (0.051) (0.079) 1.181*** 1.293*** 1.154*** 0.989*** 1.228*** (0.146) The Netherlands (0.107) (0.119) (0.104) (0.086) N 972 972 972 972 972 aic 953 953 945 936 921 Notes: Robust models clustered by country, using post-stratification weights (including design weights). An additional model with an ordinal specification of education (less than lower secondary, lower secondary, upper secondary and university) has also been tested and yields the same results. This is also true for a model including only the control variables. Missing values are delated, i.e. not considered in the analyses. Standard errors in parentheses. Source: Own elaboration (ESS, 2016). *p < .05, **p < .01, ***p < .001. 1.691*** (0.081) 1.223*** (0.123) 1.974*** (0.241) 0.249** (0.082)