ARTÍCULO / ARTICLE

POPULIST ATTITUDES, POLICY PREFERENCES, AND PARTY SYSTEMS IN SPAIN, FRANCE, AND ITALY

ACTITUDES POPULISTAS, PREFERENCIAS POLÍTICAS Y SISTEMAS DE PARTIDOS EN ESPAÑA, FRANCIA E ITALIA

Guillermo Boscán

Universidad de Salamanca

guillermoboscan@gmail.com

ORCID iD: http://orcid.org/0000-0002-4199-2160

Iván Llamazares

Universidad de Salamanca

illamaz@usal.es

ORCID iD: http://orcid.org/0000-0003-2947-6830

Nina Wiesehomeier

IE School of International Relations (Madrid)

ninawiesehomeier@faculty.ie.edu

ORCID iD: http://orcid.org/0000-0002-0134-5647

CONTENTS

INTRODUCTION Top

European party systems have witnessed the increasing prevalence of populist discursive elements among political parties with quite diverse ideological platforms (Mudde 2004Mudde, C. 2004. "The Populist Zeitgeist". Government and Opposition 39(4):541-563. https://doi.org/10.1111/j.1477-7053.2004.00135.x.; Mudde and Rovira Kaltwasser 2013Mudde, C. and C. Rovira Kaltwasser. 2013. "Populism". Pp. 493-512 in Oxford Handbook of Political Ideologies, edited by M. Freeden, L. Tower Sargent and M. Stears. Oxford: Oxford University Press.; Andreadis and Stavrakakis 2017Andreadis, I. and Y. Stavrakakis. 2017. "European Populist Parties in Government: How Well are Voters Represented? Evidence from Greece". Swiss Political Science Review 23(4):485-508. https://doi.org/10.1111/spsr.12255.; Anselmi 2017Anselmi, M. 2017. Populism: An Introduction. London: Routledge. https://doi.org/10.1007/978-981-287-588-4_209.). Scholars discuss a myriad of factors that have favored this political transformation, pinpointing to structural processes connected to globalization (Kriesi et al. 2008Kriesi, H., E. Grande, R. Lachat, M. Dolezal, S. Bornschier and T. Frey. 2008. West European Politics in the Age of Globalization. Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9780511790720.; Inglehart and Norris 2016Inglehart, R. and P. Norris. 2016. Trump, Brexit, and the Rise of Populism, Faculty Research Paper:RWP16-026.), the crisis of European party systems (Mudde and Rovira Kaltwasser 2013Mudde, C. and C. Rovira Kaltwasser. 2013. "Populism". Pp. 493-512 in Oxford Handbook of Political Ideologies, edited by M. Freeden, L. Tower Sargent and M. Stears. Oxford: Oxford University Press.; Roberts 2015Roberts, K.M. 2015. Changing Course in Latin America: Party Systems in the Neoliberal Era. Cambridge: Cambridge University Press. https://doi.org/10.1017/gov.2012.11.; Anselmi 2017Anselmi, M. 2017. Populism: An Introduction. London: Routledge. https://doi.org/10.1007/978-981-287-588-4_209.; Bornschier 2018Bornschier, S. 2018. "Ideational and Party-System-Centered Explanations of Populist Success: Latin America and Western Europe Compared". In The Ideational Approach to Populism: Concept, Theory, and Method, edited by K.A. Hawkins, R. Carlin, L. Littvay and C. Rovira Kaltwasser. London: Routledge.) and conjunctural events, such as the Great Recession (Hobolt and Tilley 2016Hobolt, S. B. and J. Tilley. 2016. "Fleeing the Centre: the Rise of Challenger Parties in the Aftermath of the Euro Crisis". West European Politics 39(5):971-991. https://doi.org/10.1080/01402382.2016.1181871.; Andreadis and Stavrakakis 2017Andreadis, I. and Y. Stavrakakis. 2017. "European Populist Parties in Government: How Well are Voters Represented? Evidence from Greece". Swiss Political Science Review 23(4):485-508. https://doi.org/10.1111/spsr.12255.) and the migration waves arriving from Africa and the Middle East triggering heightened perceptions of threat (Brubaker 2017Brubaker, R. 2017. "Between nationalism and civilizationism: the European Populist moment in comparative perspective". Ethnic and Racial Studies 40(8):1191-1226. https://doi.org/10.1080/01419870.2017.1294700.; Ivaldi et al. 2017Ivaldi, G., M.E. Lanzone and D. Woods. 2017. "Varieties of Populism across a Left-Right Spectrum: The Case of the Front National, the Northern League, Podemos and Five Star Movement". Swiss Political Science Review 23(4):354-376. https://doi.org/10.1111/spsr.12278.).

These studies echo previous comparative analyses on West European party systems that revealed a cleavage structure around new globalization-demarcation divides in two main dimensions, one of economic and another of political-cultural content (Kriesi et al. 2008Kriesi, H., E. Grande, R. Lachat, M. Dolezal, S. Bornschier and T. Frey. 2008. West European Politics in the Age of Globalization. Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9780511790720.). More recently, Häusermann and Kriesi (2015Häusermann, S. and H. Kriesi. 2015. "What do voters want? Dimensions and configurations in individual-level preferences and party choice". In The Politics of Advanced Capitalism, edited by P. Beramendi, S. Häusermann, H. Kitschelt and H. Kriesi. Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9781316163245.009.) have also revealed the crucial role that divisions between voters with particularistic and universalist orientations play in West European party systems.

The ideational theory of populism (Hawkins and Rovira Kaltwasser 2018Hawkins, K.A. and C. Rovira Kaltwasser. 2018. "Concept, Theory, and Method". In The Ideational Approach to Populism: Concept, Theory, and Method, edited by K. A. Hawkins, R. E. Carlin, L. Littvay and C. Rovira Kaltwasser. London: Routledge. https://doi.org/10.4324/9781315196923.; Mudde and Rovira Kaltwasser 2018Mudde, C. and C. Rovira Kaltwasser. 2018. "Studying Populism in Comparative Perspective: Reflections on the Contemporary and Future Research Agenda". Comparative Political Studies 51(13):1667-1693. https://doi.org/10.1177/0010414018789490.) indeed agrees that social contexts are necessary to explain support for populist actors. Yet, to fully understand the interplay between populist messages that draw on such contexts and the motivation for individuals to support populist parties, the ideational theory also distinguishes between populist positions on the supply-side (political parties) and on the demand-side (individual level). Once activated by the appropriate framing, individual-level latent populist attitudes can play an important role for vote choice. Recent empirical studies have explored the explanatory value of such latent dispositions and found evidence for the direct and indirect impact of populist attitudes on party support (Akkerman et al. 2014Akkerman, A., C. Mudde and A. Zaslove. 2014. "How Populist are the People? Measuring Populist Attitudes in Voters". Comparative Political Studies 47:1324-1353. https://doi.org/10.1177/0010414013512600.; Van Hauwaert and Van Kessel 2017Hauwaert, S. and S. Van Kessel. 2017. "Beyond Protest and Discontent: A Cross-National Analysis of the Effect of Populist Attitudes and Issue Positions on Populist Party Support". European Journal of Political Research 57(1):68-92. https://doi.org/10.1111/1475-6765.12216.; Lavezzolo and Ramiro 2017Lavezzolo, S. and L. Ramiro. 2017. "Stealth Democracy and the Support for New and Challenger Parties". European Political Science 10(2):267-289. https://doi.org/10.1017/S1755773917000108.; Andreadis et al. 2018Andreadis, I., K. A. Hawkins, I. Llamazares and M. Singer. 2018. "Conditional Populist voting in Chile, Greece, Spain, and Bolivia". In The Ideational Approach to Populism: Concept, Theory, and Method, edited by K. A. Hawkins, R. E. Carlin, L. Littvay and C. Rovira Kaltwasser. London: Routledge.).

What is less known so far, however, is how these populist attitudes match up with both the social-structural characteristics of voters and their policy stances. Kriesi et al. (2008Kriesi, H., E. Grande, R. Lachat, M. Dolezal, S. Bornschier and T. Frey. 2008. West European Politics in the Age of Globalization. Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9780511790720.), for instance, have identified the location of radical right voters and parties in spatial competition, yet they have not specifically addressed the populist content of the attitudes of such voters. In addition, these authors focused only on party systems in which populist forces were located on the right side of the political spectrum, a bias that has led researchers to confuse the explanation of voting for radical right populist parties with the explanation of populist attitudes (Mudde and Rovira Kaltwasser 2018Mudde, C. and C. Rovira Kaltwasser. 2018. "Studying Populism in Comparative Perspective: Reflections on the Contemporary and Future Research Agenda". Comparative Political Studies 51(13):1667-1693. https://doi.org/10.1177/0010414018789490.). In this article, by contrast, we analyze populist attitudes in party systems in which the parties embracing populist discourses (as identified by Hawkins and Castanho Silva 2018Hawkins, K.A. and B. Castanho Silva. 2018. "Textual Analysis: Big Data Approaches". In The Ideational Approach to Populism: Concept, Theory, and Method, edited by K. A. Hawkins, R. E. Carlin, L. Littvay and C. Rovira Kaltwasser. London: Routledge. https://doi.org/10.4324/9781315196923.) adopt very different ideological platforms. Other scholars (Rico and Anduiza 2017Rico, G. and E. Anduiza. 2017. "Economic Correlates of Populist Attitudes: an Analysis of Nine European Countries in the Aftermath of the Great Recession". Acta Politica:1-27. https://doi.org/10.1057/s41269-017-0068-7. and Rico et al 2017Rico, G., M. Guinjoan and E. Anduiza. 2017. "The Emotional Underpinnings of Populism: How Anger and Fear Affect Populist Attitudes". Swiss Political Science Review 23(4):444-461. https://doi.org/10.1111/spsr.12261.) have thoroughly explored the role that different attitudinal variables play in the explanation of populism. Their work has shown the importance of economic perceptions and feelings of anger in the formation of populist attitudes in Europe, and it has also led to emphasize the importance of subjective perceptions of economic hardship and relative deprivation over objective characteristics distinguishing between “losers” and “winners” of the globalization processes (Mudde and Rovira Kaltwasser 2018Mudde, C. and C. Rovira Kaltwasser. 2018. "Studying Populism in Comparative Perspective: Reflections on the Contemporary and Future Research Agenda". Comparative Political Studies 51(13):1667-1693. https://doi.org/10.1177/0010414018789490.: 6 and 8).

Our analyses continue and complement these different studies by examining the role of policy preferences and populist attitudes in the structuring of public attitudes and in the articulation of party system maps in these three countries. While these three countries have experienced important party system transformations in the face of economic crises and in the context of encompassing globalization processes, populist discursive elements (Hawkins and Castanho Silva 2018Hawkins, K.A. and B. Castanho Silva. 2018. "Textual Analysis: Big Data Approaches". In The Ideational Approach to Populism: Concept, Theory, and Method, edited by K. A. Hawkins, R. E. Carlin, L. Littvay and C. Rovira Kaltwasser. London: Routledge. https://doi.org/10.4324/9781315196923.; Ivaldi et al. 2017Ivaldi, G., M.E. Lanzone and D. Woods. 2017. "Varieties of Populism across a Left-Right Spectrum: The Case of the Front National, the Northern League, Podemos and Five Star Movement". Swiss Political Science Review 23(4):354-376. https://doi.org/10.1111/spsr.12278.) have been articulated by parties adopting very different ideological and programmatic platforms (leftist in the case of the France Insoumise, Podemos, and Unidad Popular-Izquierda Unida, rightist in the case of the Front National and the Lega Nord, and centrist and/or ambiguous in the case of the Movimento 5 Stelle) and appealing to voters with different types of ideological, attitudinal, and socioeconomic characteristics. Furthermore, scholars have identified these countries as belonging to two different political-economic constellations connected to the economic role of the state (“capture” in the case of Italy and Spain, and “status” in the case of France; Beramendi et al. 2015Beramendi, P., S. Häusermann, H. Kitschelt and H. Kriesi. 2015. "Introduction. The Politics of Advanced Capitalism". In The Politics of Advanced Capitalism, edited by P. Beramendi, S. Häusermann, H. Kitschelt and H. Kriesi. Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9781316163245.002.). These political-economic characteristics could also have effects on individual policy preferences and the socio-economic coalitions articulated by political parties (Beramendi et al. 2015Beramendi, P., S. Häusermann, H. Kitschelt and H. Kriesi. 2015. "Introduction. The Politics of Advanced Capitalism". In The Politics of Advanced Capitalism, edited by P. Beramendi, S. Häusermann, H. Kitschelt and H. Kriesi. Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9781316163245.002.). Given the important differences that exist between these West European societies and party systems, it becomes a central task for comparative analyses to determine whether populist attitudes are a crucial element of the political-ideological maps of these societies and whether the party system articulation of populist attitudes and policy preferences is entirely contingent upon the idiosyncratic characteristics of each party and party system or whether, on the contrary, it obeys some consistent cross-national pattern.

Furthermore, if populist attitudes are a key dimension of European public opinions in general and of the attitudinal maps of these societies in particular, then it becomes necessary to identify their determinants by extending the analytical frameworks that have been classically used in the analysis of preference formation in European politics. Scholars have rightly emphasized the thin-ideological character of populist ideas (Hawkins and Rovira Kaltwasser 2018Hawkins, K.A. and C. Rovira Kaltwasser. 2018. "Concept, Theory, and Method". In The Ideational Approach to Populism: Concept, Theory, and Method, edited by K. A. Hawkins, R. E. Carlin, L. Littvay and C. Rovira Kaltwasser. London: Routledge. https://doi.org/10.4324/9781315196923.; Mudde and Rovira Kaltwasser 2018Mudde, C. and C. Rovira Kaltwasser. 2018. "Studying Populism in Comparative Perspective: Reflections on the Contemporary and Future Research Agenda". Comparative Political Studies 51(13):1667-1693. https://doi.org/10.1177/0010414018789490.) and the presence of important qualitative differences in their articulation (Ivaldi et al. 2017Ivaldi, G., M.E. Lanzone and D. Woods. 2017. "Varieties of Populism across a Left-Right Spectrum: The Case of the Front National, the Northern League, Podemos and Five Star Movement". Swiss Political Science Review 23(4):354-376. https://doi.org/10.1111/spsr.12278.). But this does not exclude the possibility that socio-structural characteristics do affect public populist attitudes. This also means that, even if it is the case that subjective perceptions and feelings provide us with the best empirical predictors of populist attitudes, we are not exempt from the task of analyzing and underscoring, if it is empirically justified to do so, the role played by different types of socio-structural determinants (such as education, income, and occupation), in ways that are similar to those used by Kitschelt and Rehm (2014Kitschelt, H. and P. Rehm. 2014. "Occupations as a Site of Political Preference Formation". Comparative Political Studies 47(2):1670-1706. https://doi.org/10.1177/0010414013516066.) or Häusermann and Kriesi (2015Häusermann, S. and H. Kriesi. 2015. "What do voters want? Dimensions and configurations in individual-level preferences and party choice". In The Politics of Advanced Capitalism, edited by P. Beramendi, S. Häusermann, H. Kitschelt and H. Kriesi. Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9781316163245.009.) regarding the formation of public policy preferences in Europe. Certainly, very different kinds of substantive grievances could be connected to populist attitudes in contemporary democracies. However, given the fact that current economic transformations (including globalization and economic shocks) have hit much more severely individuals with less economic resources, lower levels of education, typically employed in less qualified occupations, we are led to hypothesize that it is the less educated, worse qualified and poorer individuals who will be more prone to holding populist attitudes. Of course, this does not exclude the possibility that more complex, interactive effects may be at work in the formation of populist attitudes. Furthermore, this expectation is also perfectly compatible with the fact that subjective perceptions will be, in the end, much more directly related to populist attitudes than socio-structural predictors. Finally, the effects of socio-structural characteristics do not preclude the possibility that such attitudes can be channeled by parties that articulate very different ideological platforms, in particular, as far as issues related to immigration and national sovereignty are concerned.

By exploring the role of populist attitudes in these countries’ public opinions, their party system articulation, and their association with different types of socio-structural, non-attitudinal variables, our work aims at filling a gap between two different and parallel types of analysis of European party systems; one focuses on the structuring of policy preferences and the other on the attitudinal determinants of populist attitudes.

Our work is organized as follows. In the first section we conduct factor analyses in order to identify the public opinion maps of policy preferences and political attitudes (including populism) in Spain, France, and Italy. In the second section we examine the positions of different partisan groups in the main attitudinal dimensions identified by our analyses. This results in a comparative map of populist attitudes and policy preferences in public opinions, as well as in empirical evidence as to the political articulation of these different types of attitudes in these three countries. In the final section we analyze and compare the strictly socio-structural determinants of populist attitudes in Spain, France, and Italy.

1. POPULISM AND THE ATTITUDINAL MAP OF PUBLIC OPINIONS Top

We understand populism in line with the so-called ideational theory (Hawkins and Rovira Kaltwasser 2018Hawkins, K.A. and C. Rovira Kaltwasser. 2018. "Concept, Theory, and Method". In The Ideational Approach to Populism: Concept, Theory, and Method, edited by K. A. Hawkins, R. E. Carlin, L. Littvay and C. Rovira Kaltwasser. London: Routledge. https://doi.org/10.4324/9781315196923.; Mudde and Rovira Kaltwasser 2018Mudde, C. and C. Rovira Kaltwasser. 2018. "Studying Populism in Comparative Perspective: Reflections on the Contemporary and Future Research Agenda". Comparative Political Studies 51(13):1667-1693. https://doi.org/10.1177/0010414018789490.). Accordingly, populism is a particular understanding of politics as a moral struggle based on a Manichean outlook that conceives of ‘the people’ as a virtuous homogenous group which is confronting an evil, corrupt and self-serving elite. Thus, full manifestations of populist ideas and narratives necessarily combine these three essential elements of people-centered, anti-elitist, and Manichean discourses that will attach themselves to more complex ideologies (Hawkins and Rovira Kaltwasser 2018Hawkins, K.A. and C. Rovira Kaltwasser. 2018. "Concept, Theory, and Method". In The Ideational Approach to Populism: Concept, Theory, and Method, edited by K. A. Hawkins, R. E. Carlin, L. Littvay and C. Rovira Kaltwasser. London: Routledge. https://doi.org/10.4324/9781315196923.). Most importantly, the ideational theory puts at the forefront the individual level and thus allows not only to explore the interplay between the supply and demand sides, but also to focus in particular on voters’ characteristics in conjunction with populist attitudes.

Based on previous theoretical and comparative contributions, and in light of the recent trajectories of electoral behavior and party system change in the three countries covered in this study, we can lay out the main expectations that will guide and structure our exploration. In the first place, given the recent political trajectories in Spain, France, and Italy (public mistrust precipitated by economic and responsiveness crises, erosion of traditional parties and emergence of new political actors channeling discontent with established parties) we assume that a populist dimension that is clearly different from the policy preferences divisions, will be present in the public opinions of these three countries. The analysis of this dimension is thus necessary for the understanding of the attitudinal maps of these countries. Following previous contributions on the socio-demographic correlates of social discontent (Kriesi et al. 2008Kriesi, H., E. Grande, R. Lachat, M. Dolezal, S. Bornschier and T. Frey. 2008. West European Politics in the Age of Globalization. Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9780511790720.) and populist attitudes (Spruyt et al. 2016Spruyt, B., G. Keppens and F. Van Droogenbroeck. 2016. "Who Supports Populism and What Attracts People to It". Political Research Quarterly 69(2):335-34. https://doi.org/10.1177/1065912916639138.; Rico and Anduiza 2017Rico, G. and E. Anduiza. 2017. "Economic Correlates of Populist Attitudes: an Analysis of Nine European Countries in the Aftermath of the Great Recession". Acta Politica:1-27. https://doi.org/10.1057/s41269-017-0068-7.), we expect negative associations between populist attitudes and educational levels and between populist attitudes and occupational and professional skills.

To measure populist attitudes at the individual level in the case of Spain and France we rely on an inventory by Hawkins, Riding, and Mudde (2012Hawkins, K.A., S. Riding and C. Mudde. 2012. "Measuring populist attitudes". Political Concepts Committee on Concepts and Methods Working Paper Series 55:1-35.), subsequently developed into a six-item version (Akkerman, Mudde and Zaslove 2014Akkerman, A., C. Mudde and A. Zaslove. 2014. "How Populist are the People? Measuring Populist Attitudes in Voters". Comparative Political Studies 47:1324-1353. https://doi.org/10.1177/0010414013512600.). These studies show that populist attitudes are widespread and indeed pre-exist short-term changes in social context (Akkerman et al. 2014Akkerman, A., C. Mudde and A. Zaslove. 2014. "How Populist are the People? Measuring Populist Attitudes in Voters". Comparative Political Studies 47:1324-1353. https://doi.org/10.1177/0010414013512600., Hawkins et al. 2012Hawkins, K.A., S. Riding and C. Mudde. 2012. "Measuring populist attitudes". Political Concepts Committee on Concepts and Methods Working Paper Series 55:1-35.). Furthermore, these items have been used successfully to measure populist attitudes and to explain vote choice (e.g. Andreadis and Stavrakakis 2017Andreadis, I. and Y. Stavrakakis. 2017. "European Populist Parties in Government: How Well are Voters Represented? Evidence from Greece". Swiss Political Science Review 23(4):485-508. https://doi.org/10.1111/spsr.12255.; Spruyt et al. 2016Spruyt, B., G. Keppens and F. Van Droogenbroeck. 2016. "Who Supports Populism and What Attracts People to It". Political Research Quarterly 69(2):335-34. https://doi.org/10.1177/1065912916639138.; Andreadis et al. 2018Andreadis, I., K. A. Hawkins, I. Llamazares and M. Singer. 2018. "Conditional Populist voting in Chile, Greece, Spain, and Bolivia". In The Ideational Approach to Populism: Concept, Theory, and Method, edited by K. A. Hawkins, R. E. Carlin, L. Littvay and C. Rovira Kaltwasser. London: Routledge.), they have been adapted to measure populist attitudes at the elite level (Andreadis and Ruth-Lovell 2018Andreadis, I. and S.P. Ruth-Lovell. 2018. "Elite Surveys". In The Ideational Approach to Populism: Concept, Theory, and Method, edited by K. A. Hawkins, R. E. Carlin, L. Littvay and C. Rovira Kaltwasser. London: Routledge.) and also used in experimental research that relates populist attitudes to wider behavioural outcomes (Busby et al. 2018Busby, E., D. Doyle, K. Hawkins and N. Wiesehomeier. 2018. "Activating Populist Attitudes: The Role of Corruption". In The Ideational Approach to Populism: Concept, Theory, and Method, edited by K. A. Hawkins, R. E. Carlin, L. Littvay and C. Rovira Kaltwasser. London: Routledge.).

Table 1. Items measuring populist attitudes (scales are reported in parentheses) [Descargar tamaño completo]

Table 1 shows the wording of the survey questions for each of these concepts. We implemented these questions in online surveys of about 1200 respondents in Spain and about 800 respondents in France. The samples were selected from a commercial, by invitation only, on-line pool of respondents (ISO 26362 certified) aged 18-65 and included sex, age and regional quotas. In Spain the survey was fielded in the week before the general election in December 2015 and in France in the week before the first round of the presidential election in April 2017. For Italy unfortunately we cannot rely on the same battery. We therefore matched questions as close as possible using the 2013 Italian National Election Survey (ITANES), which provides us with just two items for populist attitudes. For policy preferences we rely on a broad list of issues related to the economic left-right, cultural attitudes, and attitudes towards the European Union. Again, as Table 2 indicates, the questions for Italy differ slightly due to availability.

Table 2. Policy issues (scales are reported in parentheses) [Descargar tamaño completo]

In this first section, we present the results of the factor analysis conducted for each one of these three countries. Table 3, Table 4 and Table 5 show the structure of three principal components for each country that have been arrived at by implementing a varimax rotation as a method of representation. We discuss first the Spanish and French cases, since both analyses are based on exactly the same batteries of variables, and address the Italian case afterwards. Both in France and Spain the battery of populist indicators loads onto a single factor, always above the 0.50 threshold. This factor provides us with a strong uniform scale (eigenvalues of 3.16 in Spain and 2.86 in France), and it accounts for slightly more than the 0.20% of the variance of these variables. A second factor measures classical left-right issues of predominantly, but not exclusively, economic content. Attitudes towards redistribution and the tradeoff between economic growth and environmental protection play decisive roles in this factor in both countries, in combination with attitudes towards gay marriage (particularly in Spain) and state economic intervention (particularly in France), and to a lesser degree, but also above the 0.4 level, privacy protection. Differences mainly concern the role of attitudes towards taxes (more important in Spain and absent in this factor in France) and privacy concerns (more important in France). The third and least important factor for this battery of questions has immigration attitudes at its center in both cases. However, there are important disparities in the other variables loading onto this factor. In both cases attitudes towards EU control of public budgets load onto this factor, but they do so in varying ways for each country (with the same sign as immigration attitudes in Spain and with an opposing one in France). In other words, individuals favorable to immigration are also favorable to having full national control of budgets (positive coefficient in both variables), whereas in France the opposite association holds. This is most likely the result of the different impact of EU imposed austerity policies in Spain and France, which may have led individuals with multi-cultural and universalist values to reject EU economic policies in the former. Finally, and contrary to Spain, attitudes towards taxes also load on this factor in France. In the latter case, opposition to immigration (negative coefficient in this variable), and assertion of national budget controls and tax cuts (positive coefficients for the last two variables) go hand in hand. Overall, these results show the presence of an attitudinal dimension revealing populist attitudes that is clearly distinguishable from policy preferences in both countries. In both cases a classical left-right dimension that integrates economic redistribution and traditional versus libertarian topics is also present. Finally, a third dimension relates to immigration and, to a lesser degree, European integration issues, but the association between these two variables is disparate in these two countries, most likely as a result of the different role of EU institutions during the Great Recession in each one.

In the Italian case we rely on a battery of questions that most closely resembles the issues covered in the French and Spanish surveys. For populist attitudes we use one item that measures orientations towards direct democracy and anti-elite sentiment (If Italians could decide directly about important political matters, instead of relying on politicians, the country would be much better off) and another one that measures the rejection of political negotiations and compromises (In politics settling for a compromise actually means selling out one’s principles). For policy issues we use relatively similar indicators for attitudes towards redistribution, state economic intervention, taxes, immigration, European unification (this time in a broad way), and the environment.

As in Spain and France, the factor analysis reveals for Italy a single populist attitudes factor in which our two populism indicators load above 0.6. However, in the Italian case attitudes towards state intervention and economic redistribution do also load on this factor, albeit to a lesser degree. The eigenvalue for this factor is lower than in France and Spain (1.4), but strong enough to speak of a uniform scale. As in the French case, immigration, attitudes towards the EU, and towards taxes load on a single factor, anti-immigration, anti-EU, and anti-taxes attitudes being positively associated with each other. Finally, as in Spain and France, a separate dimension maps classical economic left-right attitudes (regarding redistribution and state economic intervention) and attitudes towards environmental protection.

Table 3. Factor analysis Populism and Policy issues, Spain. Varimax rotation with Kaiser normalization. Factor loadings greater than 0.5 are marked in bold [Descargar tamaño completo]

Table 4. Factor analysis Populism and Policy issues, France. Varimax rotation with Kaiser normalization. Factor loadings greater than 0.5 are marked in bold [Descargar tamaño completo]

Table 5. Factor analysis Populism and Policy issues, Italy. Varimax rotation with Kaiser normalization. Factor loadings greater than 0.5 are marked in bold [Descargar tamaño completo]

2. ISSUE DIMENSIONS AND PARTY VOTERS Top

In the previous section we established the presence of a populist dimension in the attitudinal maps of these three countries’ public opinions. Previous analyses have also shown that populist attitudes have direct effects on vote-choices in many European countries, including, at the least, Spain and France among these cases (Andreadis et al. 2018Andreadis, I., K. A. Hawkins, I. Llamazares and M. Singer. 2018. "Conditional Populist voting in Chile, Greece, Spain, and Bolivia". In The Ideational Approach to Populism: Concept, Theory, and Method, edited by K. A. Hawkins, R. E. Carlin, L. Littvay and C. Rovira Kaltwasser. London: Routledge.). Some of these analyses have already shown that the voting effects of populist attitudes are often conditional on other ideological orientations or policy preferences of voters. In this section we examine the characteristics of these three party systems at the demand-side level by presenting the distribution of party voters in the three dimensions we identified above. Our purpose here is not to assess the causal direct or indirect impact of populist attitudes on party choices, but to identify, from a comparative perspective, the different party system constellations of populist attitudes and policy preferences in these three cases. This analysis will allow us to locate the positions of party voters in the three dimensions identified by our analysis, irrespective of the causal effects that populist attitudes may have on voting once all controls and explanatory factors are taken into consideration.

In the following graphs we present the average positions of party voters in the three attitudinal dimensions identified by our factor analysis. We have labeled as left-right the dimension that maps preferences for redistribution and state economic intervention and also, to a lesser degree, environmental protection and civic rights. The other dimensions map populist attitudes and particularism versus universalism preferences. In order to combine all countries in one single graph we have standardized (at the national level) average values in these dimensions.

Figure 1 maps the positions of party voters in the left-right and populist attitudes dimensions and highlights a strong negative association between the positions of party voters in both dimensions. The diagonal pits populist leftists and conservative anti-populists against each other and results show that voters of economically leftist parties tend to hold more populist orientations in all three countries (Pearson correlation of -0.76, significant at the .001 level). At one extreme we find the left populist voters of Podemos (Pds), Unidad Popular-Izquierda Unida (UP-IU), and France Insoumise (FI) and at the opposite extreme the right conservative voters of Les Républicains (LR) and the Partido Popular (PP). The main cases deviating from the line that sets populist leftists and conservative anti-populists against each other are the voters of the Front National (FN), who hold populist attitudes and center-right preferences, the Italian Partito Democratico (PD) and En Marche!, comprised of voters who hold anti-populist attitudes and center-left economic preferences, and, to a lesser degree, the Movimento 5 Stelle (M5S), whose voters are center-left and highly populist.

The voters of the Italian coalition Popolo della Libertà (PDL) occupy a more moderate position than those of French and Spanish conservative parties. In this case, it must be taken into consideration that this coalition, led by Silvio Berlusconi, also integrated the voters of the Lega Nord (LN).[1] The off-diagonal quadrants of rightist populism and leftist anti-populism are largely empty. Put differently, one of the most remarkable aspects of the location of party voters is that there is not a curvilinear pattern of locations, and that it is not the case that populist attitudes and generally extreme ideological locations go hand in hand in these party systems. Overall, these analyses reveal that for these three cases there is a clear association, at the partisan level, between populist attitudes and economically leftist, pro-redistributive attitudes.

Figure 1. Average positions of party voters in the general left-right and populist attitudes dimensions [Descargar tamaño completo]

Figure 2, in turn, shows the positions of party voters in the populist attitudes and particularism versus multiculturalism/universalism dimensions. It shows that their locations are characterized by their orthogonal character, i.e. all quadrants are populated by party voters. We find populist particularist party voters, such as those of FN and M5S, populist multicultural/universalist voters, such as those of UP-IU, Podemos or FI, anti-populist universalists (EM, SC, PD), and anti-populist particularists (LR, PP). What this disposition suggests is that contrary to positioning on the left-right dimension, populist attitudes can be attached to totally opposite preferences regarding cultural inclusiveness and universalism of national institutions and policies. We cannot address here whether the causes of this open pattern of articulation between populist demands and particularist orientations lie in structural socio-economic factors, individual features, or the characteristic party discourses and strategies. Whatever the causes, what we found is a remarkable contrast between a strong association of left-right economic preferences and populist attitudes and a lack of connection between populist attitudes and particularist-multiculturalist/universalist orientations.

Figure 2. Average positions of party voters in the particularism versus multiculturalism and populist attitudes dimensions [Descargar tamaño completo]

3. THE SOCIAL DETERMINANTS OF POPULIST ATTITUDES Top

The previous section has revealed that populist attitudes are one of the dimensions structuring public opinion in Spain, France, and Italy. It has also shown that the positions of party voters in the populist dimension and the mostly economic left-right dimension are strongly connected despite the fact that these countries present in theory very different kinds of partisan ideological configurations. Analyses of populism (Mudde and Rovira Kaltwaser 2018Mudde, C. and C. Rovira Kaltwasser. 2018. "Studying Populism in Comparative Perspective: Reflections on the Contemporary and Future Research Agenda". Comparative Political Studies 51(13):1667-1693. https://doi.org/10.1177/0010414018789490; Hawkins and Rovira Kaltwasser 2018Hawkins, K.A. and C. Rovira Kaltwasser. 2018. "Concept, Theory, and Method". In The Ideational Approach to Populism: Concept, Theory, and Method, edited by K. A. Hawkins, R. E. Carlin, L. Littvay and C. Rovira Kaltwasser. London: Routledge. https://doi.org/10.4324/9781315196923.) have stressed the thin-ideological character of populism and its contingent articulation to different kinds of political and ideological platforms. However, our previous empirical results underscore the need to assess the degree to which populist attitudes may be linked to a common constellation of underlying socio-structural features affecting both economic preferences and populist attitudes. By conducting an analysis of this type our study complements works that have focused strictly on the determinants of economic and cultural preferences -without examining populist attitudes- (Häusermann and Kriesi 2015Häusermann, S. and H. Kriesi. 2015. "What do voters want? Dimensions and configurations in individual-level preferences and party choice". In The Politics of Advanced Capitalism, edited by P. Beramendi, S. Häusermann, H. Kitschelt and H. Kriesi. Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9781316163245.009.) and works that have focused on the determinants of populist attitudes, which have particularly addressed the role of different kinds of economic judgments and perceptions (Rico and Anduiza 2017Rico, G. and E. Anduiza. 2017. "Economic Correlates of Populist Attitudes: an Analysis of Nine European Countries in the Aftermath of the Great Recession". Acta Politica:1-27. https://doi.org/10.1057/s41269-017-0068-7.; Rico et al 2017Rico, G., M. Guinjoan and E. Anduiza. 2017. "The Emotional Underpinnings of Populism: How Anger and Fear Affect Populist Attitudes". Swiss Political Science Review 23(4):444-461. https://doi.org/10.1111/spsr.12261.).

Therefore, in this section we turn our analysis to the study of the factors conditioning populist attitudes in these cases. With that purpose in mind we have conducted a general linear model analysis in order to ascertain the effects of different socio-demographic variables on populist attitudes. Thus, we are not interested in developing a full-fledged explanation of populist attitudes on the bases of other attitudes and perceptions, such as economic evaluations, perceptions of corruption, or political mistrust, a topic on which several insightful contributions already exist (Spruyt et al 2016Spruyt, B., G. Keppens and F. Van Droogenbroeck. 2016. "Who Supports Populism and What Attracts People to It". Political Research Quarterly 69(2):335-34. https://doi.org/10.1177/1065912916639138.; Rico et al 2017Rico, G., M. Guinjoan and E. Anduiza. 2017. "The Emotional Underpinnings of Populism: How Anger and Fear Affect Populist Attitudes". Swiss Political Science Review 23(4):444-461. https://doi.org/10.1111/spsr.12261.). Rather, we are interested in examining whether the locations of individuals in social structures predispose them to embrace populist attitudes in these countries.

In order to assess the influence of social and economic characteristics on populist attitudes we run an analysis of variance focused on the qualitative, nominal variables that could influence populist attitudes (a continuous variable in this case). These factors are, for all countries, education, occupation, gender, and age. In the case of Spain and France we also control for income, although our indicator maps household income in the first case and individual income in the second. Unfortunately, we lack any indicator for income in the Italian survey. We use a main effects generalized linear model ANOVA analysis aimed at ascertaining the effects of each of these factors on our dependent variable. This type of technique allows us to conduct further analyses that consider all possible combinations of levels and not just contrasts with one reference category. Compared to a simple t test, the generalized ANOVA analysis allows us to also reduce the risk of type 1 error, that is, of wrongly rejecting the null hypothesis when it is true (Lindsey 1997Lindsey, J.K. 1997. Applying Generalized Linear Models. New York: Springer.).

Table 6. Generalized Linear Model (ANOVA Main Effects). Dependent variable: Populist attitudes (higher values indicate more populist attitudes) [Descargar tamaño completo]

Our results show that income levels help to account for differences in populist attitudes scores in France and Spain. In the Spanish case, occupation does also affect populist attitudes. In the case of Italy we lack an indicator for personal or family income, which may be the reason why in addition to occupation also age and education are significantly linked to populist attitudes. Post hoc tests focusing on income, education, and occupation allow us to interpret better the specific effects that social characteristics have on populist attitudes.[2] We start first with the effects of income differences reported in Table 7, the variable that shows statistically significant effects in both France and Spain. The fact that our indicators for France and Spain are different (personal income in Spain and household income in France) does not constrain our ability to make some comparative inferences on the effects of income on populist attitudes. In the case of Spain, the analysis shows that individuals earning less than 1200 euros and therefore at the bottom level of income are more populist than individuals earning more than1800 euros (significant at the .01 level). Analyses also show that individuals earning from 1200 to 1800 euros are more populist than individuals earning more than1800 euros (significant at the .01 level). By contrast, there are no significant differences between the groups with lower and middle income levels in Spain. In the case of France, the difference between individuals from households earning less than 2000 euros and between 2000 and 5000 euros are significant at just the 0.109 level. Differences are highly significant, however, between individuals from these two groups and individuals from households earning more than 5000 euros (at the .01 level).

Table 7. Multiple comparisons. Post hoc analysis: income and populist attitudes [Descargar tamaño completo]

Table 8 shows group differences for education levels. The only common feature for these three cases is that college graduates tend to be less populist than individuals with upper secondary education (differences significant at the .01 levels in Spain and Italy and at the .10 level France). In addition, in Italy individuals with basic education also tend to be more populist than individuals with college education. Overall, the data suggests that individuals with the highest education credentials are less populist than individuals with secondary, intermediate studies, whereas the contrasts with very low education groups are strong only in the Italian case.

Table 8. Multiple comparisons. Post hoc analysis: education and populist attitudes [Descargar tamaño completo]

Finally, our analysis turns to the effects of occupational differences on populist attitudes. We do not report all coefficients in Table 9, and leave out of our report and analysis groups, such as farmers and the self-employed, for which we do not have information on all countries. If we focus first on the differences between higher occupations (professionals and top managers) and the rest, we observe that professionals tend to be less populist than routine non-manual workers (in France and Spain), skilled workers (in all cases), and unskilled workers (in France and Italy). As for top managers and bureaucrats, they also tend to be less populist than skilled workers (in Spain and Italy), and routine non-manual and unskilled workers in Italy. Across all three countries, these two groups are the least populist. By contrast, skilled workers show the most consistent pattern of positive differences with professionals (statistically significant in the three cases) and managers and top bureaucrats (significant in Spain and Italy).

Contrasts among different groups of workers (routine non-manual, unskilled, and skilled) are less consistent across these cases. Skilled workers tend to be more populist than routine non-manual and unskilled workers in Spain, but less so than unskilled workers in Italy. And in the case of Italy, routine non-manual workers also tend to be less populist than unskilled workers. Contrasts among these groups are therefore less marked. Perhaps the most interesting characteristic for these groups lies in the presence of cross-national differences. In the case of Spain populist attitudes reach their highest levels among skilled workers, whereas in Italy it is among the unskilled that populist attitudes reach their highest scores. In France, by contrast, no differences between these groups of workers show up in this analysis.

Table 9. Multiple comparisons. Post hoc analysis: occupation and populist attitudes [Descargar tamaño completo]

CONCLUSION Top

Despite the presence of some broad political and ideological commonalities, the party systems of Spain, France, and Italy have been characterized by important differences in the partisan and programmatic articulation of populist discourses. The quite different orientation of parties adopting populist discourses (radical right and left in France, radical left in Spain, and centrist and radical right in Italy) reveals the presence of wide variations in the political articulation of populist ideas in this region. We started our analysis with the intention of assessing whether, despite the presence of these important differences, populist attitudes played a role in the structuring of the ideological attitudinal maps in these countries, in combination with the policy preferences dimensions that have been usually established by comparative analyses on European party systems. Our empirical results are consistent with this expectation, even if the measurement tools we had at our disposal, particularly in the case of Italy, presented some limitations. In all these countries populist attitudes proved to be a clearly discernible attitudinal dimension, along with the most classical left-right dimension, strongly connected to redistribution and state intervention preferences, and with another dimension mostly mapping attitudes towards immigration and globalization. That is, despite the presence of some distinctive national characteristics, the attitudinal maps of these three countries presented strong similarities in terms of the substantive contents of their main attitudinal dimensions and of the role of populist orientations in them.

Our second, exploratory goal was to go beyond classical vote-choice models on the effects of populist attitudes and map the positions of party voters in the populist and policy preferences dimensions. Our exploration revealed a clear association, at the party level, between classical left-right, mostly economic preferences, and populist attitudes. In this case, party voters tended to populate a diagonal leading from redistributive populism to pro-market non-populism. Symmetrically, no party voters adopted populist and anti-redistributive positions or non-populist and pro-redistributive positions. By contrast, our data showed the absence of associations between populist attitudes and preferences regarding immigration and globalization. Overall, these results endorse the interpretation that, at least for this group of cases, populist orientations and pro-redistributive attitudes tend to go hand in hand. Important disparities among the parties representing populist orientations are observed in cultural demarcation attitudes. In other words, it is in this political-cultural dimension that the thin-ideological character of populism is more apparent and that the discursive content of parties representing populist orientations becomes more diverse (Ivaldi et al. 2017Ivaldi, G., M.E. Lanzone and D. Woods. 2017. "Varieties of Populism across a Left-Right Spectrum: The Case of the Front National, the Northern League, Podemos and Five Star Movement". Swiss Political Science Review 23(4):354-376. https://doi.org/10.1111/spsr.12278.). Whether these different patterns of discursive and programmatic articulation result from socio-structural factors or from partisan agency is a question that we cannot address here.

Finally, our analysis of the structural factors affecting populist attitudes showed that it is among groups with fewer occupational skills, lower income levels and lower education levels that populist attitudes tend to be stronger. These facts endorse the importance of structural factors in the formation of populist attitudes, notwithstanding the role that other subjective experiences and feelings can play in the emergence of such orientations. The effects of education, occupation, and income on populist attitudes help to understand why there are significant associations, at the party-level, between the populist attitudes of party voters and their preferences for redistribution. By contrast, the lack of associations between populist attitudes and immigration preferences is consistent with the fact that, in the cultural domain, structurally based economic grievances can be channeled by both universalist and particularist political parties.

This comparative and empirical study has aimed at analyzing the presence, political articulation, and socio-economic determinants of populist attitudes in Spain, France, and Italy. Future developments of this work should seek to include other types of societies and party systems in the analysis and to develop more complex and interactive multivariate models focusing on the socio-economic determinants of populist attitudes. These two goals remain the objective of posterior developments of this work.

ACKNOWLEDGMENTSTop

The authors acknowledge financial support by the Spanish Ministry for the Economy and Competitiveness (Research Grant CSO2013-47667-P).

NOTES Top

REFERENCESTop

ABOUT THE AUTHORSTop