East Versus West on the European Populism Scale

Abstract

We study a sample of individuals in 24 European countries that includes nine Eastern European countries in order to identify whether these countries differ from their Western counterparts as regards popularity of populist right-wing parties once we have controlled for personal attributes. The results show that there is variation among the Eastern European countries but that, as a whole, they are not distinct from Western Europe. However, there is greater support of populist right-wing parties in Hungary and Poland once account is taken of personal attributes and we discuss some possible reasons for this observation. When it comes to personal identities, we find that a right-wing identity, a negative view of immigrants, dissatisfaction with democracy, a negative view of homosexuality, and mistrust in both national and European parliaments seem to correlate heavily with voting for a populist right-wing party in Europe. Moreover, men are more likely to vote for a populist right-wing party, as are the old and the less educated.

1 Introduction

The aim of this paper is to use individual-level survey data to describe broad patterns and regularities in political attitudes towards right-wing populism, defined as the electoral success of populist right-wing (PRW) parties in the European Economic Area (EEA) between 2002 and 2016.¹ We focus on personal values, economic factors, and country of residence. We include nine Eastern European countries and explore whether they differ fundamentally from the Western European ones. Our main empirical question is whether, because of their communist heritage, the Eastern European nations are distinct as regards voting for populist right-wing parties.

We control for several individual characteristics. We focus on trust in domestic and EU institutions because lack of trust has been found to be correlated with voting for populist parties (see Dustmann et al. 2016), placement on the left/right scale, and satisfaction with democracy as representing confidence in the political establishment. Attitudes towards homosexuals and immigrants and religiosity measure traditional values. We include two economic factors: placement in the income distribution and whether the individual has ever been unemployed for 3 months or more. Finally, we include belonging to a minority group, level of education, gender, and age. Country dummy variables capture factors that remain to be explained. The main objective of the paper is to compare the country dummy variables between individual Eastern European nations, on the one hand, and between Eastern European nations and Western European ones, on the other hand, in order to determine whether there are significant differences between Eastern and Western Europe.

The main innovation of the paper over those surveyed in the following section is the inclusion of Eastern European nations—nations that turn out to be quite diverse in their propensity to vote for populist parties. The attitudes and voting patterns of these nations are important for decision-making within the European Union (EU), and it is of some interest to see whether they share a populist sentiment that could disrupt the operations of the EU. These nations share the experience of having had communist societies involving central planning, absence of democracy, and limited human rights in the form of freedom of expression and freedom of movement. They may also have enjoyed more economic security because unemployment did not exist and education and health care were free of charge in the communist states. This shared history may make these nations more or less prone to vote for populist right-wing parties, which then affects collective decision making at the EU level.

2 Literature

The Brexit referendum in the UK in June 2016 and the election of Donald Trump as president of the US have generated intense interest in the reasons for the success of populist politicians and parties. Below we give a brief overview of some recent contributions.

2.1 Populism and Its Causes

According to the political scientist Cas Mudde (see Mudde 2016), populist parties tend to challenge prevailing elites and institutions such as the media, universities, mainstream political parties, and international organisations.² Populists also tend to share a tendency to claim to represent the “people” against the prevailing authorities and institutions and “outsiders”—such as foreign countries, immigrants, or minority groups—and to be led by charismatic leaders. They share nativist policies directed at different outside groups, such as illegal Mexican immigrants in the US, Americans in Venezuela, or immigrants from Europe in the UK.

We are interested in exploring to what extent economic and cultural factors may fuel the emergence of populist parties, particularly to include the lingering effects of a communist past. Inglehart and Norris (2016) propose two explanations for the rise of populism. The first is based on economic factors that create insecurity, such as international trade. The other is based on opposition to progressive or socio-liberal values, such as feminism and environmentalism. They use the 2014 Chapel Hill Expert Survey to identify the ideological location of 268 political parties in 31 countries—the EU member states and Norway, Switzerland, and Turkey—and use the European Social Survey (ESS) from 2002 to 2014 to test whether it is economic insecurity or cultural factors that predict voting for populist parties. They control for gender, age and education, experience of unemployment, measures of feeling of income security, and values that were meant to separate populist and liberal values. Their regression model pools responses to the ESS conducted between 2002 and 2014 in order to test the two hypotheses, and it finds more evidence for the cultural hypothesis. This leads us to believe that past economic systems may have a long-lasting effect on values and attitudes in a country.³

2.2 Values and Trust as a State Variable

Inglehart and Baker (2000) use data from three waves of the World Values Survey, including 65 countries and 75% of the world’s population. They find evidence for economic development as a factor affecting cultural values, as well as some persistence of distinctive cultural traditions. Economic development is found to be associated with shifts from absolute norms and values to values that are more rational, tolerant, trusting, and participatory. This supports what in sociology is called the modernization theory.⁴ Not only do cultural values respond to economic development, however; they are persistent, so that the cultural heritage of a society—be it in the form of a religion or an economic system—leaves an imprint on values, which endure in spite of increased economic development. These cross-country differences—that is, cross-cultural differences—are transmitted from one generation to the next through schools and the media. Voigtländer and Voth (2015) demonstrated that stronger anti-Semitic beliefs among Germans who grew up with racial propaganda under the Nazi regime were still prevalent more than 50 years after the end of the Second World War. Inglehart and Baker mention the emergence of fundamentalist Islam as an example of the persistence of cultural heritage in spite of economic development. Another example is given by Fukuyama (1995), who argued that societies suffering from low levels of trust are at a competitive disadvantage in global markets because of difficulties in developing large, complex institutions such as corporations.

2.3 The Lingering Effects of Communism

Communism may have a lingering effect on values and attitudes. According to Inglehart and Baker (2000), the former communist societies have more traditional values than protestant European Union nations, with the latter leaning away from traditional values and towards self-expression values. They also find that the Catholic societies of Eastern Europe form a sub-cluster of the Catholic world between Western European Catholic societies and Orthodox societies. The collapse of communism in the early 1990s brought about changes. Following German unification and the fall of the Soviet Union, both the former West Germany and the former East Germany experienced a shift towards rational values and an emphasis on self-expression and away from traditional values. Another example mentioned by Inglehart and Baker is that East Germany is much closer to the ex-communist countries of the Czech Republic and the Baltic States than West Germany in terms of “traditional/secular” versus “self-expression” values. Thus the cultural heritage of a country appears to matter—in this case, their communist past.

2.4 Populism and Trade

Economic shocks, trade, and crises have an effect on political developments. There is a rapidly growing literature on the effect of trade on values, particularly to include the vote for populist parties. Clearly, a populist party that is nativist and anti-establishment may oppose free trade, as recent examples show. The negative income and employment effects of trade may affect subgroups of the labour force, as is demonstrated in a rapidly growing literature that shows how international trade is having a negative effect on local economies. Pessoa (2014) finds that UK workers in industries that became exposed to Chinese import competition earned significantly less over the period 2000–2007 because of fewer years of employment and lower hourly earnings while employed. The economic effects of import competition can also have political effects by creating protectionist sentiments and can increase the share of voters who choose populist parties. Dippel et al. (2015) find that trade integration with China and Eastern Europe affected voting in Germany from 1987 to 2009. The share of voters who favoured extreme-right parties responds significantly to trade integration as measured by changes in manufacturing employment. Curtice (2016) studies public attitudes to the European Union in Britain and finds that voters are concerned about the cultural consequences of EU membership but are inclined to consider membership economically beneficial. In a study of voting patterns in Western Europe, Colantone and Stanig (2016) find that voters in areas more exposed to competition from Chinese imports tend to vote in a more protectionist and nationalist direction.

2.5 Populism and Economic Cycles

Yann et al. (2017) find a relationship between increases in unemployment and voting for populist parties. Moreover, they find a correlation between the increase in unemployment and a decline in trust in national and European political institutions. Overall, these authors find that crisis-driven economic insecurity is a driver of populism and political distrust. Frieden (2016) uses data from Eurobarometer surveys since 2004 to explore changes in attitudes before and after the recent crisis. He found that the crisis reduced trust in both national governments and the EU. He also found that less educated and less skilled citizens, along with the unemployed, are particularly lacking in trust and that those in the southern periphery—the debtor nations—are uniformly disappointed with their national political institutions. The UK is again an outlier in terms of lack of trust towards the EU. In another recent paper, Foster and Frieden (2017) analyse individuals’ responses in Eurobarometer surveys conducted from 2004 to 2015 in order to study the reasons for changes in trust during the recent financial crisis. The authors confirm the results of previous studies: that the better educated have the highest levels of trust in both their national governments and the EU, while those with lower levels of skills and education have less trust. Economic variables such as unemployment help explain the variation in trust among Europeans over time and across countries.

In a recent paper, Dustmann et al. (2017) find that growth in GDP per capita increases support for European integration and enhances trust in both European and national parliaments, while an increase in the unemployment rate has a negative effect on these same variables. The economic situation matters more in regions where people have traditional and autocratic values. If political populism is associated with less trust in parliamentary institutions and more Euroscepticism, then adverse macroeconomic shocks tend to increase the demand for populist political parties. The authors find that the effect of macroeconomic shocks is almost twice as large on trust towards national parliaments as it is on trust towards the European parliament. In essence, then, citizens mainly blame national politicians for adverse economic conditions. These authors conclude that anti-EU sentiment is more sensitive to national identity and personal attributes than to economic factors, so that economic growth will not fully restore support for the European Union. Once again, in this study the UK is clearly an outlier in terms of lack of trust towards the EU and diminishing trust in recent years.

2.6 Populism and Financial Crises

Financial crises tend to reduce trust in societies and have a stronger impact on voters than ordinary recessions. Hence it is possible that financial crises also erode trust in domestic institutions, political parties, and international institutions. Funke et al. (2016) study election data for 20 developed economies going back to 1870. They find that polarization rises following financial crises and that voters seem to move towards populist right-wing parties. Hernandez and Kriesi (2016) reach similar conclusions in their study of election outcomes in 30 European countries in three elections: two immediately preceding the latest crisis and one immediately following it. They find that falling output, increased unemployment, and increased debt resulted in losses for incumbent parties in Western Europe, but less so in Central and Eastern Europe. Moreover, Funke and Trebesch (2017) compared the 2008 financial crisis to other European financial crises in the late 1980s and early 1990s and found a similar pattern of a surge in populist right-wing party vote shares in national elections. Another study was done by Bartels (2014), who found, in a sample of 42 elections in 28 OECD countries before and after the Great Recession, that 1% growth of GDP increased the voting share of the incumbent party by 1.2%.

2.7 Populism and the Welfare State

The emergence of populism in the wake of economic recessions and financial crises may be prevented by the creation of a welfare state. Swank and Betz (2003) analysed national elections in 16 European countries from 1981 to 1988 and found that a welfare state weakens the link between international trade and immigration, on the one hand, and support for the populist right, on the other hand. Mayda et al. (2007) found that in small countries with higher levels of government expenditures, the population tends to be less risk-averse with respect to international trade. Finally, Rodrik (1998) argued that, because governments can reduce aggregate risk through redistribution and stable provision of publicly provided goods and services, more open economies tended to have larger governments.

2.8 Populism and Immigration

Immigration seems to have a mixed effect on populist voting. According to Dustmann et al. (2016), most Danish municipalities in the period 1986–1998 experienced an increase in the vote share for parties with an anti-immigration agenda when their share of refugee immigrants increased. In urban municipalities, however, an increase in refugees had the opposite effect on nativist voting. Davis and Deole (2017) also found a mixed relationship between a country’s immigrant share and the vote for far-right parties in 14 European countries in 2002–2014, when controlled for individual characteristics. In societies that the author classified as collectivist, a larger share of immigrants led to an increase in the vote share for far-right parties. An increased presence of immigrants had the opposite effect in societies that were classified as individualistic.

3 Populist Parties

We are interested in the propensity of individuals and nations to vote populist right-wing parties (PRW) into power. Table 1 lists the vote share of PRW parties found in 24 countries contained in the dataset. ⁵ The list of parties is given in Appendix 1. Their election results in the most recent parliamentary elections in 2002 and 2016 are also shown, showing an increase in support in 14 out of 20 countries.⁶ Hungary tops the list in terms of both the share of votes in 2014 and the increase from 2002. There is also a PRW party in Greece that did not exist in 2002 but had a vote share of 20.5% in 2014. In third place is Finland, where the “True Finns” have around a fifth of the voting share. Perhaps surprisingly, Sweden comes next, with the Swedish Democrats carrying 12.9% of votes. After Poland are Bulgaria, Austria, and Lithuania. At the bottom of the list is Italy, where the Lega Nord lost many votes during this period.⁷ Just above Italy are Belgium, the Netherlands, Slovenia, France, Estonia, and Denmark. Germany and the UK are close to the centre of the list.

Table 1 Populist parties in different countries

4 Explanatory Variables

Our data come from the European Social Survey (ESS), which contains answers from individuals in 24 EEA member countries between 2002 and 2016, 15 Western European countries and nine Eastern European countries. The ESS, carried out every 2 years, measures the attitudes and behavioral patterns of more than 300,000 persons in various European countries. We use 268,513 observations from the survey.

Table 2 lists the names and definition of selected variables. The dependent variable, pop, takes the value 1 if an individual voted for a PRW party in the last election, but otherwise it is 0. Variables meant to capture cultural traits and trust in institutions are continuous variables taking a value between 0 and 1, except for those measuring religion. These are trust in the national parliament, trust in the EU Parliament, placement on the left/right scale of the political spectrum, satisfaction with democracy, attitude towards homosexuals, attitudes towards immigrants, and place in the income distribution. In addition, we have three dummy variables for each of the four following features: the respondents’ placement on the left/right scale, their education level, their placement in the country’s income distribution, and their religiosity. Age is measured in the number of years at the time of the election and normalised to take values between 0 and 1, and a squared term for age is added to capture the possible non-linear effects of age on PRW voting.

Table 2 Definition of variables

There are several other dummy variables. They include belonging to a minority group, gender (1 denoting females), and a history of unemployment for 3 months or more.⁸ Finally, there is a dummy variable for each country and each wave of the ESS, starting in 2002.

5 Empirical Analysis

Pop is a dependent variable that only takes the values 0 and 1, so it does not follow a normal distribution. Therefore, a regression by least squares would produce the wrong standard errors. Running a logistic regression would counter this problem, but the interpretation of the coefficients would be more complicated. Therefore, we choose to use a least-squares regression in order to simplify the interpretation, even though the standard errors for the estimated coefficients might be wrong and the regression might produce probabilities outside the interval between 0 and 1. We try to account for this problem by running a generalised least-squares regression and using heteroscedasticity-consistent standard errors in the tables below.

As a robustness check, we computed the marginal effects obtained in a logit regression for the same statistical model and compared them to the coefficients of our generalised least-squares regression (see Appendix 2). Some discrepancies in marginal effects were found between the models in three variables. These are listed in Table 5. In order to account for unequal inclusion probabilities in the survey and differences in the countries’ population size, post-stratification and population weights provided by the ESS are used. The weighted dataset is considered to be a random sample of the European population. The variance inflation factors (VIF) for each variable revealed that the model contained little multicollinearity.

The generic estimation equation is given by:

$$ {pop}_{it}={\beta}_0+{X}_{it}^{\prime }{\beta}_1+{Z}_{it}^{\prime }{\beta}_2+{T}_t+{C}_i+{u}_{it}, $$

where pop _it takes the value 1 if the individual voted for a PRW party, X is a matrix with the values and attitudes variables and religion listed in Table 2, Z has the demographic and economic variables (age, income distribution, education, gender, unemployment, minority group), T _t has the years of interview dummies, and C _i are country dummies.

Table 3 shows the regression results for the cultural and demographic variables along with their significance and heteroscedasticity-consistent standard errors. The coefficients of the country and time dummy variables are shown in Table 4. The analysis of the full model contains 223,794 observations, as 44,719 observations were dropped due to one or more missing variables.

Table 3 OLS Regression with sample weights. Dependent variable: Pop

Table 4 OLS Regression with sample weights. Dependent variable: Pop

The coefficients of the independent variables are mostly in line with previous results in the literature. A right-wing identity, a negative view of immigrants, dissatisfaction with democracy, a negative attitude towards homosexuality, and mistrust in the European parliament seem to be the factors heavily correlated with voting for a PRW party. Surprisingly, though, trust in a country’s national parliament does not seem to have a significant effect on voting for a PRW party. In addition, women are less likely to vote for these parties, as are the young, while people with low- and mid-level education are more likely to vote for them than those with higher education. Having been unemployed for at least 3 months in the past does not seem to have a significant effect on voting for a PRW party. Moreover, age seems to have a positive but decreasing effect on the probability of voting for a PRW party. Quantitatively, being right-wing raises the probability of voting for a PRW party by 8%, being in favour of immigration reduces the probability by 6%, and being satisfied with democracy reduces the probability by more than 2%. Trusting the European Parliament fully lowers the probability of voting for a PRW party by 4%. Having a negative attitude towards homosexuals increases the probability by 2%, having a university education reduces it by 3%, and being female lowers it by 1%. Finally, being halfway through life (age = 0.5) increases the probability by 2%. Other variables, although statistically significant, are numerically less important. Thus, a typical PRW party voter will be someone who is male, against immigration, dissatisfied with democracy, right-wing, negative towards homosexuals, without a university education, and distrustful of the European Parliament.

The only result that is perhaps puzzling is that individuals are more likely to vote for these parties if they place themselves higher in the income distribution. There is also the question why people with “medium religiosity” are less likely to vote for a PRW party than the group of respondents with low and high values in this variable.

Table 4 shows the coefficients of the country and ESS-round dummy variables. The time dummies show that support of PRW parties increased between 2008 and 2010 following the global financial crisis, and also between 2012 and 2016, which can possibly be attributed to the euro crisis as well as the increasing inflow of refugees.

Comparing the Eastern and Western European nations, the average value of the dummy variable for the 11 Western European nations is −0.017, while the average for the nine Eastern European nations is 0.01. Therefore, on average, the Eastern European nations have a slightly lower country effect in absolute terms. However, there is variation within the group. Hungary has the largest country dummy coefficient, followed by Norway and Poland, while the Czech Republic, Estonia, the UK, Greece, Slovakia, and Portugal have the smallest dummies. Both Hungary and Poland were in the top half of Table 1, but Finland, Greece, and Sweden, also at the top of that table, have negative coefficients in Table 4, which suggests that the explanatory variables account for the populist sentiments in these countries. In contrast, unexplained factors appear to make Hungary and Poland vote for PRW parties.

Note that two Eastern European countries—Hungary and Poland—rank high in Table 1 and also have large positive coefficients of the country dummies in Table 4, as is shown in Fig. 1 below. The other Eastern European countries—Bulgaria, Lithuania, the Czech Republic, Estonia, Croatia, Slovenia, and Slovakia—have negative country dummies, which indicates that time-constant country-specific factors are not pulling them in that direction. The average value for Eastern Europe excluding Hungary and Poland is −0.039; that is to say, more negative than the average for the Western European countries. The countries that are furthest from the fitted line are Norway, Denmark, and the Netherlands, which tend to lean more strongly towards PRW parties, once the personal attributes have been taken into account, than the electoral success of these parties in 2016 would lead us to expect.

Fig. 1

Country fixed effects and voting for PRW parties. The sample includes observations for Croatia (HRV), Portugal (PRT), Estonia (EST), Germany (DEU), The Czech Republic (CZE), Greece (GRE), Lithuania (LTU), Italy (ITA), Belgium (BEL), France (FRA), Sweden (SVE), Great Britain (GBR), The Netherlands (NLD), Denmark (DNK), Norway (NOR), Slovakia (SVK), Bulgaria (BGR), Finland (FIN), Slovenia (SVN), Austria (AUT), Switzerland (CHE), Poland (POL) and Hungary (HUN). Sources for the shares of dominant ethnic groups: National statistical bureaus. We exclude Bulgaria from the figure since its values are very similar to those of Lithuania

Table 5 shows the largest differences in estimated marginal effects between a weighted logit regression—see Appendix 2—and our generalised least-squares regression. According to the logit regression, trust in the national parliament is a significant and negative variable, whereas the unemployment dummy is significant and positive. The medium education dummy is also significant in the logit regression, but its effect on voting for a PRW party are marginally positive, in contrast to the marginally negative effects in the GLS regression.

Table 5 Discrepancies between ordinary least-squares estimates and estimated marginal effects in logit

6 Specificities of Eastern Europe

Results in the existing literature would suggest that the socio-economic environment in Eastern European countries is conducive to the emergence and electoral success of populist political movements, right-wing parties in particular. For example, Inglehart and Norris (2016) suggest that countries that have been exposed to major economic displacement and change, along with countries whose culture is traditional and conservative, tend to have a larger electoral base for populist right-wing parties. Eastern Europe seems to qualify on both dimensions. Yet the reality happens to be very different.

Eastern European countries are small or medium-sized open economies that depend largely on international markets—European markets in particular—for both essential inputs and the sale of their final or intermediate products. Since the fall of the Soviet bloc, they have been the recipients of major foreign direct investment and in turn have experienced at least one major episode of capital flight in the wake of the financial crisis of 2008–2009. Moreover, the structure of their economies, dependent on traditional heavy and light industry, has exposed Eastern European countries to strong competitive pressures as a result of China’s entry into the World Trade Organization. In combination, these factors have contributed to a rise in inequality on a number of dimensions: rural vs. urban areas, declining vs. emerging or growing industries, young vs. old generations and so forth.

In addition, Eastern Europe also underwent a painful transition from central planning to market- and price-based economic coordination. Eastern European countries vary considerably in the speed, dynamics, and success of their transition, but there are several common features across the region. The economic changes have led to substantial growth in inequality and economic displacement. Another source of social tension and bitterness is that in all countries, abuse of political power and influence has often given rise to abuse of market power. For these reasons, it is still rare that incumbents are re-elected. Thus Eastern Europe has been exposed to economic pressures that are at least as severe as those facing old EU member states.

The conservative culture and prevailing social norms in Eastern European societies also suggest that the advance and popularisation of social-liberal ideas and policies would provoke a political backlash. Dustmann et al. (2017) show how traditional values magnify the effect of economic downturns on voters, making them distrust the European Union and national parliaments more and vote for populist parties. Due to the relatively late transition to modernity and the influence of the Soviet bloc, Eastern Europeans were subject to more traditional and conservative standards of behaviour in society and in the family. Furthermore, for somewhat complex reasons, education and the mass media before 1989 emphasized patriotism and even nationalism as opposed to internationalism. In addition, as is pointed out by Inglehart and Baker (2000), Eastern European cultures and social norms up to the 2000s are strikingly oriented towards social survival and cohesion rather than self-expression.

After the fall of the Berlin Wall, Eastern European countries have experienced major social and cultural changes: the elimination of national borders, a reduction in national sovereignty, the emigration and immigration of large numbers of people, an increase in the social acceptability of cohabitation without marriage, abortion, same-sex relations, and even same-sex marriage. In other words, these societies have experienced a convergence with beliefs and norms in Western Europe, with emphasis on the individual and self-expression rather than on some sort of collective identity. In such circumstances of major and rapid change, one would expect cultural friction and opposition that could foster the growth of right-wing parties.

Yet our results do not lend support to those hypotheses: They show that Eastern European countries are not more susceptible to right-wing populism than Western European countries. We believe that several factors may account for this. One of these factors is the turbulent history of the region and, in particular, its turbulent relationship with nationalism. On the one hand, the spread of nationalism from Western Europe to Eastern Europe is largely responsible for the creation of national identities in the region, which eventually led to the demise of the four great empires of the East: Ottoman, Russian, Habsburg, and German. The final result was the establishment of the modern nation states of the region by the end of WWI. The entire process proved particularly violent and destructive, however: Both WWI and WWII were much bloodier and more socially disruptive in the East than in the West. Moreover, the events of the twentieth century emphasize a key political feature of the region: Namely, economic, political, and cultural life in the region is caught up in the interplay of Great Powers, specifically Germany and Russia, which cannot be opposed by any single regional nation state. For example, Germany became a major export market and creditor for most Eastern European countries in the 1930s. As a result, long before WWII, the Nazis used the resulting economic influence to promote their political agenda. Naturally, since their inception, all nation states in the region have been actively looking for allies and forms of international cooperation, with various degrees of success, in order to offset and limit the impact of foreign interference in their own affairs.

In this context, both the Cold War and WWII remain constant reminders of the consequences when chauvinism runs rampant and Eastern European countries fail to form strong international alliances. After the fall of the Berlin Wall, membership of NATO and the European Union have been the two pillars of the foreign policy of all Eastern European countries, and they have enjoyed wide public support precisely because of the traumatic memories of the past. It is important to point out that even the most extreme Eurosceptic in Hungary and Poland, for example, do not question the value of NATO or EU membership. Instead, they defend traditional social values, the preservation of the idea of a Europe of nations rather than a federal Europe.

Other factors that limit the susceptibility of Eastern Europeans to right-wing populism are socio-economic in origin. Many Eastern Europeans are aware that they have directly or indirectly benefited from the EU common market, mainly through inward FDI and the opening of Western European markets to Eastern European labour. In addition, opening the borders has meant an increase in travel and interaction with other Europeans, intermarriages, and much a greater awareness of ‘the other.’ In this context, the electoral map of Poland is revealing. Eastern Poland has been and remains the stronghold of right-wing populism, while Western Poland has been much less susceptible to right-wing ideas and much more prone to vote based on economic issues. This does not seem to be a coincidence, given the increasing cross-border integration of Western Poland with Germany, the large-scale reallocation of people, mainly Poles across the border, and the influx of FDI to the region.

Finally, yet importantly, while Eastern Europeans may exhibit fear and unwillingness to accept immigrants from non-European or non-Christian countries, they seem to take no issue with the immigration of Western Europeans to their countries for economic and family reasons. In addition, this acceptance of greater European social and economic integration can be traced back to a residual belief that some Western European countries—Germany and the Scandinavian countries in particular—remain role models that should be emulated. Finally, the Eastern European countries have a chequered state and institutional tradition. Thus the alternative to European integration appears much less attractive to the general public than it does in some of the old European great powers, such as the UK, France, or Germany. On a related note, political interests in Eastern Europe are not institutionalised through political parties, and political parties represent clusters around certain influential political leaders. Consequently, while authoritarian tendencies may very well be present and even accepted by much of the electorate, right-wing populists in Eastern Europe simply do not have the institutional capacity to impose total control on political, economic, and cultural life.

7 What Makes Hungary and Poland Different?

On most dimensions, Hungary and Poland appear similar to most other Eastern European countries. In what follows, we explore possible explanations for the high susceptibility of Hungarians and Poles to right-wing populism. We believe that the observed patterns can be accounted for by a combination of traditional culture, strong nationalist tradition, and an extremely ethnically homogeneous population. We suggest that the Europeans in less ethnically homogeneous societies are less prone to support right-wing populism not only because exposure to immigrants and minorities somehow makes them more enlightened. Rather, we believe that ethnic diversity and migration makes everyone aware of economic, political, and social difficulties prior to the implementation of a right-wing nationalist agenda. In this sense, ethnic diversity undermines the credibility of right-wing policies.

Poland and Hungary experienced high rates of economic growth in the 2000s and the 2010s. Moreover, they did not perform worse than the rest of the Eastern or Western European countries during and after the Great Recession. In fact, Poland is the only European country that did not experience even a technical recession following the financial crisis of 2008–2009. Although the public finances of Hungary are still a cause for concern, the Hungarian economy has also performed reasonably well during the same period. In this context, it is all the more surprising that these two countries have turned strongly to the extreme right in the past decade or so.

Another puzzling factor is that the migrant crisis after 2013 did not affect either country. Poland has been completely isolated from the main channels of immigration from the Middle East and Africa to the EU, and Hungary was only briefly a transit destination in 2015–2016. Even during this period, the influx of temporary immigrants to Hungary pales in comparison with that in Greece and Italy. Consequently, unlike in the case of Germany, it does not appear likely that the rise of right-wing populism can be traced back to the increase in immigration to the EU after the Arab Spring.

In terms of standard explanations, Poland and Hungary stand out for their traditional culture and social norms. Inglehart and Baker (2000) show that Poland and Hungary, along with Romania, are the Eastern European countries with the most traditional and conservative social norms and attitudes. It comes as no surprise, then, that large sections of these societies have felt deeply uncomfortable with the rise of social liberalism. These developments have certainly contributed to the popularity of political parties that appear to defend the certainty of established institutions, such as the church and the state, or established social norms based on (patriarchal) hierarchy, order, and the value of the community as opposed to individual expression. Nevertheless, one is bound to ask why other societies that score high on Inglehart and Baker (2000) index of traditionalism, such as Romania, have not experienced a similar rise in right-wing populism.

In this context, it is interesting to note that both Poland and Hungary have had a long historical tradition as regional powers. While it is true that by 1815 both countries had become parts of the multinational Russian and Habsburg empires, their intellectual and political elites largely survived intact and, in turn, they were the first Eastern European countries to develop a very strong sense of national identity. The strength and popular nature of Polish and Hungarian nationalism is evident, for example, from the Polish rebellions of 1830–1831, 1846, 1848–1849, and 1863–1864 and the Hungarian revolution of 1848–1849. The resistance to Soviet rule and interference after 1945 also bears a strong flavour of national resistance against the successors of the traditional imperial enemy.

There are also other ‘old’ Eastern European countries, but their historical tradition has either been interrupted by long foreign rule and socio-economic dominance, as in the case of Bulgaria, Croatia, Czechia, Lithuania, and Serbia, or current states that are the union of older states with a distinct historical tradition, as in the case of Romania. For example, among the other Eastern European countries, Czechia and Romania also have histories that can be traced back at least to the Middle Ages through the kingdom of Bohemia, in the former instance, and the principalities of Walachia and Moldova, in the latter. Still, the Czech lands experienced sustained Germanization after 1620, characterized by strong influence from the central government of the Habsburgs and from the local German aristocracy. Thus the Czech national revival only gathered pace in the nineteenth century. Similarly, historical differences between Walachia, Moldova, and Transylvania slowed down the formation of a Romanian identity during the nineteenth century.

The only dimension on which Hungary and Poland are completely different from the rest of the Eastern European countries, and in fact from most Western European countries, is their ethnic homogeneity. As regional powers, both Hungary and Poland ruled over other nations until the eighteenth and the nineteenth century, respectively. After WWI, however, Hungary lost all regions with mixed populations under the formal jurisdiction of Austria-Hungary. Although Poland after 1920 included sizeable Ukrainian, Lithuanian, and German minorities, the end of WWII saw the creation of a largely homogeneous nation state, albeit displaced to the West. The result is that by the early 1990s, ethnic Hungarians accounted for 98 percent of the population of Hungary and ethnic Poles for 97 percent of the population of Poland. In this respect, the only Western European countries with similar dominance by major ethnic groups are Norway and Finland, both of which also turn out in our statistical tests to be susceptible to right-wing nationalism.

We investigate further this issue in Fig. 2 by plotting the relationship between the estimated country-specific susceptibility to right-wing populism and the share of dominant ethnic groups in the population. A visual inspection of Fig. 1 confirms that the tendency to support right-wing populism does not increase with the ethnic diversity of a country. This finding contradicts the notion that ethnic differences on their own increase social tension and the probability of civil conflict. On the contrary: We find that the propensity to support right-wing populist parties increases with the ethnic homogeneity of the country. This applies to both Eastern and Western Europe. A linear regression of the susceptibility to right-wing populism on the share of the dominant ethnic groups shows a positive and statistically significant correlation between the two variables. Moreover, Fig. 1 suggests that support for right-wing populism may grow exponentially with the share of the dominant ethnic groups.

Fig. 2

Share of dominant group and the country-specific effect. Relationship between country-specific susceptibility to right-wing populism in Table 4 and the share of dominant ethnic groups in the population. Country-specific susceptibility is measured by the country effects reported in Table 4. The sample includes observations for Austria (AUT), Belgium (BEL), Croatia (HRV), Czech Republic (CZE), Denmark (DEN), Estonia (EST), Finland (FIN), France (FRA), Germany (DEU), Greece (GRE), Hungary (HUN), Italy (ITA), Lithuania (LTU), Netherlands (NLD), Norway (NOR), Poland (POL), Slovenia (SVN), Slovakia (SVK), Sweden (SWE), United Kingdom (UK), Portugal (PRT). Sources for the shares of dominant ethnic groups: national statistical bureaus. We exclude Bulgaria from the figure since its values are very similar to those of Lithuania

We suggest two hypotheses for the observed pattern. First, it may be that interacting with numerous people of diverse background in daily life makes it harder to demonize ‘the other’ as the source of all social evils. Thus ethnic diversity may actually promote the (liberal) notion of common human nature and, in turn, universal human rights. Along with this optimistic hypothesis, we also consider a second hypothesis: that ethnic diversity does not promote mutual understanding, but its existence increases the costs and dangers associated with promoting right-wing policies. In ethnically diverse societies, potential sympathizers are aware from a practical standpoint that the implementation of right-wing nationalistic and populist policies is likely to increase social tension, cause social disruption, and lead to international isolation, if not intervention by more powerful neighbours. The practical difficulties in putting such policies in place in diverse societies may actually undermine the electoral credibility of right-wing populism. We leave it to future research to test which of these hypotheses holds, if either one does.

8 Conclusions

In this paper, we have focused on support for PRW parties and ignored the populist parties on the left of the political spectrum because the accepted wisdom has been that Eastern Europe is particularly fond of PRW parties and that these parties pose an existential threat to the European Union. Our results show that the Eastern European nations differ internally in their propensity to vote for a PRW party. They have a slightly higher average country effect but vary greatly internally. Thus Hungary and Poland have a greater affinity with such parties, while the Baltics, Bulgaria, Slovenia, Croatia, and Slovakia have much less. However, we can also find comparable countries in Western Europe, such as Norway and Denmark, which also are inclined to vote for a PRW party. Poland and Hungary are both former regional powers with homogeneous populations where power tends to lie in individual politicians rather than organised parties, as in the West. We argue that the diversity of the population of other Eastern European countries makes their culture more liberal; that is, more accepting of other ethnic group and less prone to vote for right-wing populists.

The coefficients of the personal attributes have a familiar pattern. A right-wing identity, a negative view of immigrants, dissatisfaction with democracy, a negative view of homosexuality, and mistrust in institutions seem to be the factors heavily correlated with voting for a PRW party. In addition, women are less likely to vote for these parties, as are the young, while people with low- and mid-level education are more likely to vote for them than the highly educated are. The only surprising result is that individuals are more likely to vote for these parties if they place themselves higher in the income distribution. There is also the question why people with “medium religiosity” are less likely to vote for a PRW party than the group of respondents with low and high religiosity.

One limitation of the study is that some political parties that are not considered PRW have adopted more radical policies in order to win votes from PRW parties. Therefore, overall support for populism could be underestimated. A good example is the UK, where the Conservative Party became more populist in response to the challenge presented by the UK Independence Party. In fact, in the recent study by Dustmann et al. (2017), the Conservative Party is counted among populist parties based on its manifesto.

In future work, we plan to study the determinants of support for left-wing populist parties in Europe and test whether they differ from the determinants of PRW parties. Left-wing populist parties have been more popular in Southern Europe and less so in Eastern Europe; therefore, the focus of our research would shift from the latter to the former.

Appendices

Appendix 1 List of Populist Parties

Country	Party
Hungary	Fidesz (F) after 2006, Jobbik (J), Magyar Igazság és Élet Pártja (MIÉP)
Poland	Prawo i Sprawiedliwość (PiS) after 2007 Samoobrona Rzeczpospolitej Polskiej (SRP) Polska Wspolnota Narodowa (PWN) Ruch Patriozinsky (RP), Narodowe Odrodzenie Polski (NOP), Polska Partia Narodowa (PPN), Kongres Nowej Prawicy (KNP) Kukiz ‘15
Slovenia	Slovenska Nacionalna Stranka (SNSi), Slovenska demokratska stranka (SDS) Lipa (L)
Austria	Freiheitliche Partei Österreichs (FPÖ), Bündnis Zukunft Österreich (BZÖ), Team Stronach für Österreich (TS)
Finland	Perussuomalaiset (PS), Suomen Kansan Sinivalkoiset (SKS), Sinivalkoinen Rintama (SR), Muutos 2011 (M)
Bulgaria	Ataka, Balgarski Natsionalen Saiuz, Natzionalen Front za Spasenie na Bulgaria (NFSB), Natzionalen Ideal za Edinstvo (NIU), Edinna narodna partia (ENP), Partia za horata ot naroda (PHN), Prezaredi Balgariya (PB)
Slovakia	Slovenská národná strana (SNSk), Ľudová strana—Naše Slovensko (L’SNS)
Greece	Laikós Orthódoxos Synagermós (LAOS), Anexartitoi Ellines (ANEL), Chrysí Avgí (C)
Sweden	Sverigedemokraterna
UK	The Democratic Unionist Party (DUP), United Kingdom Independence Party (UKIP)
Germany	Nationaldemokratische Partei Deutschlands (NPD), Alternative für Deutschland (AfD)
Czech Rep.	Sdružení pro republiku (SPR), Pravý blok (PB), Úsvit (U)
Lithuania	Partija tvarka ir teisingumas (PTT), Jaunoji Lietuva (JL)
Switzerland	Schweizerische Volkspartei (SVP), Schweizer Demokraten (SD), Eidgenössisch-Demokratische Union (EDU), Freiheits-Partei der Schweiz (FPS), Lega Dei Ticinesi (LT), Partei National Orientierter Schweizer (PNOS)
Estonia	Eesti Iseseisvuspartei (EIP), Eesti Konservatiivne Rahvaerakond (EKRE)
Norway	Fremskrittspartiet (FRPn)
Portugal	Partido Nacional Renovador (PNR), Partido da Nova Democracia (PND), Partido Democrático do Atlântico (PDA)
Denmark	Dansk Folkeparti (DF), Fremskridtspartiet (FRPd)
France	Mouvement National Républicain (MNR), Front National (FNf)
Croatia	Hrvatski demokratski savez Slavonije i Baranje (HDSSB), Hrvatska stranka prava (HSP1)
Luxembourg	Alternative Demokratesch Reformpartei (ADR)
Belgium	Vlaams Blok (VB1), Vlaams Belang (VB2), Parti Populaire (PP), Front National (FNb)
Netherlands	Partij voor de Vrijheid (PVV), List Pim Fortuyn (PF), Staatkundig Gereformeerde Partij (SGP)
Italy	Allianza Nazionale (AN), Fratelli D’Italia (FI), Lega Nord (LN), Fiamma Tricolore (FT)

Appendix 2 GLS and a Marginal-Effects Logit Regression

Variable name	GLS	Logit	GLS	Logit
	Estimated values		Significance
Intercept	0.061		∗∗∗
Trust in national parliament	0.005	−0.007		∗∗∗
Trust in EU parliament	−0.040	−0.019	∗∗∗	∗∗∗
Left-wing views on the left/right scale	−0.013	−0.017	∗∗∗	∗∗∗
Right-wing views on the left/right scale	0.077	0.037	∗∗∗	∗∗∗
Satisfaction with democracy	−0.025	−0.013	∗∗∗	∗∗∗
Attitude towards homosexuals	0.023	0.006	∗∗∗	∗∗∗
Attitude towards immigrants	−0.061	−0.040	∗∗∗	∗∗∗
In middle of income distribution	0.011	0.006	∗∗∗	∗∗∗
In top 30% of income distribution	0.006	0.003	∗∗	∗∗∗
Medium religiosity	−0.003	0.000	.
High religiosity	0.005	0.002	∗∗	∗∗
Belongs to a minority group	−0.009	−0.011	∗∗∗	∗∗∗
Female	−0.008	−0.007	∗∗∗	∗∗∗
Age	0.256	0.117	∗∗∗	∗∗∗
Age, squared	−0.384	−0.155	∗∗∗	∗∗∗
Medium education	−0.012	0.006	∗∗∗	∗∗∗
High education	−0.026	−0.007	∗∗∗	∗∗∗
Has ever been unemployed for 3 months	−0.001	0.003		∗∗∗
Belgium	−0.012	−0.009	∗∗∗	∗∗∗
Bulgaria	−0.026	−0.015	∗∗∗	∗∗∗
Switzerland	0.057	0.026	∗∗∗	∗∗∗
Czech Rep.	−0.066	−0.024	∗∗∗	∗∗∗
Germany	−0.047	−0.027	∗∗∗	∗∗∗
Denmark	0.029	0.006	∗∗∗	∗∗
Estonia	−0.060	−0.024	∗∗∗	∗∗∗
Finland	−0.024	−0.009	∗∗∗	∗∗∗
France	−0.022	−0.013	∗∗∗	∗∗∗
UK	−0.052	−0.023	∗∗∗	∗∗∗
Greece	−0.068	−0.023	∗∗∗	∗∗∗
Croatia	−0.047	−0.020	∗∗∗	∗∗∗
Hungary	0.155	0.047	∗∗∗	∗∗∗
Italy	−0.017	−0.009	∗∗∗	∗∗∗
Lithuania	−0.026	−0.012	∗∗∗	∗∗∗
Luxembourg	−0.017	−0.011	∗∗∗	∗∗∗
Netherlands	0.014	0.001	∗∗∗
Norway	0.049	0.018	∗∗∗	∗∗∗
Poland	0.103	0.034	∗∗∗	∗∗∗
Portugal	−0.071	−0.031	∗∗∗	∗∗∗
Sweden	−0.024	−0.011	∗∗∗	∗∗∗
Slovenia	−0.022	−0.014	∗∗∗	∗∗∗
Slovakia	0.000	−0.007		∗∗∗
Round 2: 2004	0.007	0.000	∗
Round 3: 2006	0.004	0.006		∗∗∗
Round 4: 2008	0.017	0.013	∗∗∗	∗∗∗
Round 5: 2010	0.025	0.015	∗∗∗	∗∗∗
Round 6: 2012	0.023	0.015	∗∗∗	∗∗∗
Round 7: 2014	0.040	0.021	∗∗∗	∗∗∗
Round 8: 2016	0.043	0.016	∗∗∗	∗∗∗