According to welfare chauvinism, access to the welfare state should be reserved for the native population, whereas immigrants are seen as a drain on resources. The curious aspect of welfare chauvinism in Europe is that it is more prevalent in the East. Why is this the case? This article uses the European Social Survey (ESS) and the Life in Transition Survey (LITS) in order to locate the most robust individual-level determinants of welfare chauvinism for countries of both Eastern and Western Europe. The results suggest that there is no support for the socioeconomic explanation of welfare chauvinism. There is support for the cultural capital explanation of welfare chauvinism, but only for Western Europe. Finally, there is support for the theory that higher levels of trust lessen the likelihood that a person adopts welfare chauvinism. This finding holds for both Eastern and Western Europe.

INTRODUCTION

Welfare chauvinism has become a common political attitude as well as a rather successful political platform for right-wing populist parties in Europe. Those who adopt welfare chauvinism suggest that the services of the welfare state should be provided to “our own.” In other words, they are to be provided to natives and denied to immigrants (Goul Andersen & Bjorklund, 1990, p. 212; Kitschelt, 1997, p. 22). It is a particular form of anti-immigrant sentiment placed in the context of social policy. The curious thing about welfare chauvinism is that it is not restricted to countries where the inflow of immigrants is substantial but can also be found where there is very little immigration. Eastern Europe is one such region. Indeed, surveys suggest that more people in Eastern Europe adopt these attitudes than in Western Europe, despite the fact that Eastern Europe is of little interest to immigrants. Why is this the case?

This article investigates welfare chauvinism in several Eastern European countries and compares them to several Western European countries. The research strategy is based on a comparison of two large-N data sets, the European Social Survey and the Life in Transition Survey (see ESS, 2020; EBRD, 2020). The goal is to replicate, as closely as possible, the same multivariate analysis on both data sets in order to locate the most robust determinants of welfare chauvinism. The focus, therefore, is on the micro-level predictors of welfare chauvinism, that is, the characteristics and attitudes of individuals that make them more or less prone to welfare chauvinism. The analysis focuses on four Western European countries (Germany, France, UK, Sweden) and four Eastern European countries (Czech Republic, Poland, Hungary, Slovenia). These four Eastern countries are chosen because they are the richest countries in the region (as measured by GDP per capita) and thus may become more attractive to immigrants in the mid- to long-term future.1

The analysis proceeds from three possible explanations of welfare chauvinism that can be investigated through an analysis of survey data. First, welfare chauvinism may be a rational response to an economically precarious or disadvantaged position. People who depend on the services of the welfare state and/or are of lower socioeconomic position may see immigrants as a source of competition. The working class, in particular, may be seen as potentially more prone to welfare chauvinism (Olzak, 1992; Kitschelt, 1997; Scheve & Slaughter, 2001; Malchow-Moller et al., 2008; Guiso et al., 2017). Second, the reasons for welfare chauvinism may not be economic but cultural. People who have higher levels of cultural capital, gained primarily through higher levels of education, may be more receptive to diversity. Those without education and the cultural sophistication that it provides may see immigrants in a more negative light (Lamont, 1987; Houtman, 2001; van Oorshot, 2006; Stubager, 2008; van der Waal et al., 2010; Mewes & Mau, 2013; Inglehart & Norris, 2016). Education may broaden the horizons of people, making them more welcoming and tolerant.

And third, welfare chauvinism may not come from the structural position one occupies in the economy nor from the cultural and symbolic resources that derive from education, but rather from a generalized attitude of trust. People who are more trusting of others may be able to resist welfare chauvinism (Uslaner, 2003; Crepaz & Damron, 2009; Crepaz, 2010; Ervasti & Hjerm, 2012). Trust makes it possible to consider problems in more general terms, instead of seeing politics as something that we can do to them (Rothstein & Uslaner, 2005). Trust leads to more cohesive, cooperative communities (Putnam, 2006; Wilkinson & Pickett, 2010). Generalized trust means trust in an unknown citizen, thus potentially encompassing immigrants as well (Kuovo, Kankainen & Niemela, 2012). Therefore, people who report higher levels of generalized trust may be more likely to keep the doors of the welfare state open to immigrants.

Which of these explanations is the most relevant? With regard to Western Europe, an increasing number of studies have pointed to the relevance of the cultural explanation. Which of the three explanations holds in Eastern Europe? Are there differences between East and West in the level of support that each of these explanations receives? To briefly summarize the analysis, the results for Western European countries give support to the second (education) and third explanation (trust), but not the first (socioeconomic class). Those respondents who had a university education and who reported higher levels of trust were less likely to adopt welfare chauvinism.

With regard to Eastern Europe, there is support only for the third explanation (trust). Therefore, in Eastern Europe, the only thing that can be said with a fair degree of certainty is that people who are more trusting are less prone to welfare chauvinism. The weaker impact of education on welfare chauvinism in Eastern Europe points to an important difference between the East and the West: in the East, a key conduit that may be able to weaken welfare chauvinism does not work as it does in the West. Since levels of trust also tend to be lower in Eastern Europe than in Western Europe, the efficacy of this conduit is also more limited. As a note of caution, it should be added that the data are more volatile for Eastern Europe, which suggests that attitudes are not fixed and may undergo further change.

WELFARE CHAUVINISM, EAST AND WEST

In Eastern Europe, anti-immigrant sentiment seems to be on the rise. In comparison with the 2000s, when EU membership suggested convergence with Western ideals of tolerance and multiculturalism (e.g., Mudde, 2005; Ceobanu & Escandell, 2008), in recent years there has been a conservative turn. Since the refugee crisis of 2015, various right-wing politicians and parties have won new momentum with an exclusionary agenda. Major examples include Czech president Miloš Zeman, Hungary's prime minister Viktor Orbán, and the ruling Polish party Law and Justice. Their anti-immigration platform encompasses several elements. They see immigrants, especially Muslim immigrants, as culturally incompatible with the European way of life. They also see immigrants as a security threat, alleging connections with terrorist networks. And finally, they see immigrants as harmful economically, since they take jobs away from natives and/or abuse welfare benefits.2 Such positions are quite popular as well. Indeed, all of the politicians mentioned above have won repeated electoral victories.

The welfare chauvinism evident in Eastern Europe is all the more unusual given that there is so little immigration into the region. Eastern European countries are the most homogeneous countries in the EU with about 3 to 7 immigrants per 1,000 inhabitants.3 Even where large minorities do exist, they tend to be culturally and racially similar groups, such as Moravians in the Czech Republic and Bosnians in Slovenia. Yet, the threat of immigration is presented in terms of the black and brown populations of the Middle East and Northern Africa, despite the fact that such groups have, by and large, preferred to seek a new home farther west. Seen from the viewpoint of basic economics (Bansak et al., 2015; Collier, 2015), there is very little momentum for immigration because (1) wage differentials are not high enough to attract immigrants and (2) migrant diasporas, which could ease the way for new arrivals, are small to nonexistent.

Not only is there very little immigration, but the welfare state is also not very generous in Eastern Europe. For example, with regard to unemployment benefits, Poland, Hungary, and the Czech Republic institute a qualifying period of one year, while in Slovenia it is nine months. By contrast, in Sweden, Luxembourg, and the Netherlands it is six months, in France and Finland two months, and in Germany, UK, and Ireland access is immediate, pending certain conditions. Child benefits, on the other hand, can be claimed immediately in most countries, but are much lower in Eastern Europe: around 20 euros in the Czech Republic, 40 in Hungary, 55 in Poland. In Western Europe they are higher: 90 euros in Austria, 115 in Belgium, 161 in Denmark, 90 in Finland, 155 in Germany, 100 in Sweden.4 Even adjusting for cross-country differences in price levels, it would be quite a stretch to say that people have an incentive to move to Eastern Europe in order to draw benefits.

Despite the lack of substantial immigration and the lack of welfare state generosity, aggregate levels of welfare chauvinism are higher in Eastern Europe than in the West. Table 1 provides an overview of the percentages. It tells us how widespread welfare chauvinism is. The survey questions are explained in more detail later, but it is evident that the percentages are higher in Eastern Europe. In Western Europe, the share of welfare chauvinists is around 30–40%, with the exception of Sweden, where it is lower. In Eastern Europe, the share increases to more than 50%, even above 70%. The percentages in Eastern Europe also seem to be more volatile, which suggests caution with regard to any kind of long-term prediction. At the moment, it is not possible to say if the increase over time observed in Table 1 is a permanent one. It may be a temporary shock following on the heels of the refugee crisis, or it may indeed be a more stable equilibrium.

TABLE 1.

Percent of respondents who are welfare chauvinists

Life in Transition Survey: Percent of people who agree with statement: “Immigrants are a burden for the national protection system”European Social Survey: Percent of people who agree with statement: “Taxes and services: Immigrants take out more than they put in or less” (0–10; 0 = “Generally take out more,” 10 = “Generally put in more”) Recoded: 1 if 0 to 4
Western Europe   
Germany (2016) 36.7 (2016) 32.9 (2014) 
France (2010) 44.2 (2010) 41.9 (2014) 
UK (2010) 44.8 (2010) 41.1 (2014) 
Sweden (2010) 8.3 (2010) 25.4 (2014) 
Eastern Europe   
Poland (2016) 69.5 (2016) 38.6 (2014) 
Czech Republic (2016) 76.8 (2016) 57.4 (2014) 
Hungary (2016) 76.7 (2016) 52.3 (2014) 
Slovenia (2016) 57.2 (2016) 24.4 (2014) 
Life in Transition Survey: Percent of people who agree with statement: “Immigrants are a burden for the national protection system”European Social Survey: Percent of people who agree with statement: “Taxes and services: Immigrants take out more than they put in or less” (0–10; 0 = “Generally take out more,” 10 = “Generally put in more”) Recoded: 1 if 0 to 4
Western Europe   
Germany (2016) 36.7 (2016) 32.9 (2014) 
France (2010) 44.2 (2010) 41.9 (2014) 
UK (2010) 44.8 (2010) 41.1 (2014) 
Sweden (2010) 8.3 (2010) 25.4 (2014) 
Eastern Europe   
Poland (2016) 69.5 (2016) 38.6 (2014) 
Czech Republic (2016) 76.8 (2016) 57.4 (2014) 
Hungary (2016) 76.7 (2016) 52.3 (2014) 
Slovenia (2016) 57.2 (2016) 24.4 (2014) 
Source: Author's calculations based on LITS and ESS (see EBRD, 2020; ESS, 2020).

Note: Sample weights applied.

What can explain the welfare chauvinism of Eastern Europeans? Why does the difference between East and West exist? Some answers, particularly those that have to do with the micro-level determinants of welfare chauvinism, can be gleaned from large-N survey data. In particular, such data can help to adjudicate between several possible explanations. As mentioned, these revolve around (1) socioeconomic position, (2) cultural capital gained through education, and (3) the general tendency to trust other people. The first explanation suggests that those in disadvantaged economic positions turn to welfare chauvinism as a way to protect their livelihoods. The second explanation argues that a certain cultural sophistication gained through education may lessen welfare chauvinism. And the third argues that a sentiment of generalized trust may be the way to lessen welfare chauvinism. These three theories can be transformed into testable hypotheses:

Hypothesis 1. Persons who occupy disadvantaged economic positions will be more likely to adopt welfare chauvinism.

Hypothesis 2. Persons who have a lesser degree of education will be more likely to adopt welfare chauvinism.

Hypothesis 3. Persons who report lesser levels of interpersonal generalized trust will be more likely to adopt welfare chauvinism.

Each of these hypotheses will be tested in both Western and Eastern Europe, on both the LITS and the ESS data. Findings that emerge for all countries and both data sets can be treated as robust.

DATA

The data for this article come from two large survey projects. As mentioned, the goal is to replicate as close as possible the same analysis on both data sets in order to locate the most robust results. The first data set is the Life in Transition Survey. This large comparative data set is headed by the European Bank for Reconstruction and Development with the goal of understanding the transition to liberal democracy in the post-communist region. The data are publicly available.5 LITS encompasses all of Eastern Europe and the former Soviet Union but includes an occasional Western country as well. In 2016 Germany was included, and in 2010 France, UK, and Sweden were included.6 The second source of data is the European Social Survey.7 This long-standing comparative project encompasses most European countries. Data from the seventh round (2014) were used.8

This article focuses on four West European countries and four East European countries. The practical limitation of the LITS data means that only Germany, France, UK, and Sweden can be included, though the ESS, of course, covers many other Western countries. On the other hand, there was more of a choice regarding Eastern Europe. However, it may make the most sense to focus on those countries that are the richest in the post-socialist region and so may, at least in the mid-term future, become recipients of somewhat larger immigrant inflows. This includes Poland, Czech Republic, Hungary, and Slovenia.9

The main dependent variable is welfare chauvinism. LITS includes a question that asks respondents to choose between the following claims: “Immigrants are a burden for the national social protection system” and “Immigrants make a valuable contribution to the national economy of our country.” The first claim captures welfare chauvinist views. In the ESS, there are several questions on immigration. However, the one that most resembles the question posed in LITS is the one in which respondents are asked to consider if immigrants put more into the national social system via taxes or if they take out more via services. Respondents are asked to place themselves on an ordinal scale of 0 to 10, with 0 meaning that “immigrants generally take out more than they put in” and 10 meaning the opposite. Since the question from LITS is not an ordinal scale but a categorical variable, the question from ESS was recoded such that those who placed themselves from 0 to 4 are given a 1 (i.e., they are coded as welfare chauvinists).

The first set of independent variables are basic demographic variables, which function mostly as control variables: gender, age, and status of ethnic minority. The last question tracks all respondents who do not belong to the titular ethnicity of a given country. For instance, in the German sample, this includes everybody who said their ethnicity is something other than German. Analogously, in the Polish sample, for example, this includes everybody who said their ethnicity is something other than Polish. With LITS data from 2010, a work-around solution had to be used since a question on ethnicity was not included.10

The next set of variables operationalizes the socioeconomic explanation of welfare chauvinism. This includes variables for income, blue-collar workers, business owners, the unemployed, and retired people. Regarding income, a categorical variable was constructed for all respondents who belong in the top 10% of the income distribution. This way, it is possible to track the wealthy. Yet, it should be mentioned that this variable is not as informative as would ideally be hoped, since refusal to answer is a problem.11 Therefore, this variable tracks the wealthiest out of those who provided an answer. A categorical variable was constructed to track blue-collar workers. Each survey has its own categorization of occupations, but efforts were undertaken to match them as much as possible. The goal was to include workers who occupy classic industrial positions. The remaining variables are also simple categorical variables: business owners, the unemployed, retired persons.

With regard to education, cross-country variations pose challenges for comparison. Each country has its own specific educational system, and it is not always possible to locate commensurate ranks within the educational hierarchy. This especially concerns the gray area between secondary education (high school) and tertiary education (university). However, a university diploma functions as a similar threshold in most countries. Therefore, in order to ease comparison across countries, the educational variable is operationalized in terms of a university education (bachelor's degree or above).

The analysis also includes several organizational and attitudinal variables. First, a variable for trust is included. LITS uses a five-point ordinal scale (from complete distrust to complete trust), whereas ESS uses a ten-point scale. The answers in ESS were recoded to a five-point scale in order to correspond with LITS. Next, a variable for religiosity is included. LITS tracks active churchgoers, and ESS tracks people who said they were very religious (recoded as 1 for respondents who answered between 6 and 10 on a scale from 0 to 10). Also included are variables for political party membership and labor union membership. And finally, a variable was added that tracks attitudes toward income inequality. Here, the data from LITS had to be transformed from a ten-point ordinal scale to a five-point scale in order to correspond with ESS. This variable makes it possible to trace the impact of preferences for more redistribution (smaller values) to less redistribution (larger values). These final attitudinal and organizational variables are treated primarily as controls, as are the demographic variables. The list of variables was not expanded beyond this since there was a limit to the number of theoretically relevant variables that could be matched in both data sets.

ANALYSIS

The results of the analysis are presented in Tables 2 to 5. Tables 2 and 3 present the results of logistic regression models predicting welfare chauvinism in Western Europe, first using LITS and then using ESS. Tables 4 and 5 present the results of the same models for Eastern Europe, once again using LITS first and ESS second. On a technical note, all standard errors are robust or Huber-White standard errors, in order to maintain conservative inference thresholds (Angrist & Pischke, 2009). Sample weights were applied for the entire analysis. In addition, sample sizes were equalized to about 1,000 randomly selected respondents so that differing sample sizes would not lead to differing levels of statistical significance. No observations had missing data within this subset.

TABLE 2.

Logistic regression models for welfare chauvinism in Western Europe (Life in Transition Survey)

(1)(2)(3)(4)
Germany
(2016)
France
(2010)
UK
(2010)
Sweden
(2010)
Demographic variables     
Gender
(1 = male) 
1.178
(0.170) 
0.857
(0.128) 
0.491***
(0.072) 
1.190
(0.309) 
Age 1.004
(0.006) 
1.009
(0.007) 
1.001
(0.005) 
1.015
(0.012) 
Socioeconomic variables     
Blue-collar worker
(1 = yes) 
1.156
(0.257) 
1.696
(0.582) 
0.539
(0.658) 
9.780**
(6.909) 
Business owner
(1 = yes) 
0.851
(0.266) 
1.108
(0.202) 
1.175
(0.220) 
0.361**
(0.134) 
Unemployed
(1 = yes) 
1.087
(0.224) 
1.034
(0.243) 
0.590**
(0.118) 
0.801
(0.314) 
Retired
(1 = yes) 
1.470
(0.433) 
0.911
(0.266) 
1.176
(0.293) 
1.270
(0.760) 
Education     
University education
(1 = yes) 
0.287***
(0.071) 
0.381***
(0.066) 
0.325***
(0.074) 
0.224***
(0.065) 
Trust     
Trust
(1 to 5; no trust to complete trust) 
0.463***
(0.044) 
0.774***
(0.054) 
0.667***
(0.045) 
0.605**
(0.092) 
Organizational and attitudinal variables     
Active churchgoer
(1 = yes) 
1.087
(0.188) 
0.598
(0.214) 
0.820
(0.148) 
0.627
(0.345) 
Political party member
(1 = yes) 
1.129
(0.235) 
0.830
(0.377) 
1.139
(0.430) 
0.279*
(0.174) 
Labor union member
(1 = yes) 
0.807
(0.279) 
0.690
(0.177) 
0.561
(0.183) 
1.370
(0.439) 
Tolerance for inequality
(1 to 5; prefer equality to prefer inequality) 
1.078
(0.064) 
1.155**
(0.071) 
0.814**
(0.052) 
1.126
(0.124) 
Constant 4.158**
(1.796) 
1.155
(0.452) 
7.840***
(3.121) 
0.361
(0.303) 
Observations 1047 965 1081 946 
Pseudo R-squared 0.104 0.096 0.101 0.159 
(1)(2)(3)(4)
Germany
(2016)
France
(2010)
UK
(2010)
Sweden
(2010)
Demographic variables     
Gender
(1 = male) 
1.178
(0.170) 
0.857
(0.128) 
0.491***
(0.072) 
1.190
(0.309) 
Age 1.004
(0.006) 
1.009
(0.007) 
1.001
(0.005) 
1.015
(0.012) 
Socioeconomic variables     
Blue-collar worker
(1 = yes) 
1.156
(0.257) 
1.696
(0.582) 
0.539
(0.658) 
9.780**
(6.909) 
Business owner
(1 = yes) 
0.851
(0.266) 
1.108
(0.202) 
1.175
(0.220) 
0.361**
(0.134) 
Unemployed
(1 = yes) 
1.087
(0.224) 
1.034
(0.243) 
0.590**
(0.118) 
0.801
(0.314) 
Retired
(1 = yes) 
1.470
(0.433) 
0.911
(0.266) 
1.176
(0.293) 
1.270
(0.760) 
Education     
University education
(1 = yes) 
0.287***
(0.071) 
0.381***
(0.066) 
0.325***
(0.074) 
0.224***
(0.065) 
Trust     
Trust
(1 to 5; no trust to complete trust) 
0.463***
(0.044) 
0.774***
(0.054) 
0.667***
(0.045) 
0.605**
(0.092) 
Organizational and attitudinal variables     
Active churchgoer
(1 = yes) 
1.087
(0.188) 
0.598
(0.214) 
0.820
(0.148) 
0.627
(0.345) 
Political party member
(1 = yes) 
1.129
(0.235) 
0.830
(0.377) 
1.139
(0.430) 
0.279*
(0.174) 
Labor union member
(1 = yes) 
0.807
(0.279) 
0.690
(0.177) 
0.561
(0.183) 
1.370
(0.439) 
Tolerance for inequality
(1 to 5; prefer equality to prefer inequality) 
1.078
(0.064) 
1.155**
(0.071) 
0.814**
(0.052) 
1.126
(0.124) 
Constant 4.158**
(1.796) 
1.155
(0.452) 
7.840***
(3.121) 
0.361
(0.303) 
Observations 1047 965 1081 946 
Pseudo R-squared 0.104 0.096 0.101 0.159 
Source: Author's calculations based on LITS and ESS (see EBRD, 2020; ESS, 2020).

Note: Odds ratios and robust standard errors in parentheses. Statistical significance: * p < .05, ** p < .01, *** p < .001

TABLE 3.

Logistic regression models for welfare chauvinism in Western Europe (European Social Survey)

(1)(2)(3)(4)
Germany
(2014)
France
(2014)
UK
(2014)
Sweden
(2014)
Demographic variables     
Gender
(1 = male) 
0.839
(0.129) 
1.042
(0.995) 
0.985
(0.146) 
1.109
(0.186) 
Age 0.992
(0.005) 
0.995
(0.006) 
1.000
(0.006) 
1.011
(0.006) 
Socioeconomic variables     
Blue-collar worker
(1 = yes) 
1.044
(0.180) 
1.307
(0.219) 
1.170
(0.201) 
1.347
(0.253) 
Business owner
(1 = yes) 
1.055
(0.256) 
0.967
(0.248) 
1.006
(0.205) 
0.823
(0.223) 
Unemployed
(1 = yes) 
1.427
(0.465) 
1.298
(0.359) 
1.234
(0.438) 
1.163
(0.477) 
Retired
(1 = yes) 
1.219
(0.290) 
1.341
(0.359) 
0.832
(0.204) 
1.091
(0.329) 
Education     
University education
(1 = yes) 
0.372***
(0.092) 
0.629**
(0.107) 
0.369***
(0.072) 
0.577**
(0.101) 
Trust     
Trust
(1 to 5; no trust to complete trust) 
0.717***
(0.042) 
0.705***
(0.044) 
0.825**
(0.045) 
0.686***
(0.045) 
Organizational and attitudinal variables     
Religious person
(1 = yes) 
1.011
(0.157) 
0.987
(0.151) 
0.703*
(0.112) 
0.905
(0.179) 
Political party member
(1 = yes) 
0.922
(0.348) 
0.398
(0.173) 
0.308*
(0.176) 
1.237
(0.365) 
Labor union member
(1 = yes) 
0.756
(0.170) 
0.888
(0.275) 
1.447
(0.322) 
1.126
(0.199) 
Tolerance for inequality
(1 to 5; prefer equality to prefer inequality) 
1.034
(0.077) 
0.985
(0.061) 
1.079
(0.073) 
1.368
(0.110) 
Constant 1.963
(0.719) 
2.314*
(0.899) 
1.079
(0.073) 
0.307*
(0.146) 
Observations 980 850 914 1023 
Pseudo R-squared 0.067 0.065 0.071  
(1)(2)(3)(4)
Germany
(2014)
France
(2014)
UK
(2014)
Sweden
(2014)
Demographic variables     
Gender
(1 = male) 
0.839
(0.129) 
1.042
(0.995) 
0.985
(0.146) 
1.109
(0.186) 
Age 0.992
(0.005) 
0.995
(0.006) 
1.000
(0.006) 
1.011
(0.006) 
Socioeconomic variables     
Blue-collar worker
(1 = yes) 
1.044
(0.180) 
1.307
(0.219) 
1.170
(0.201) 
1.347
(0.253) 
Business owner
(1 = yes) 
1.055
(0.256) 
0.967
(0.248) 
1.006
(0.205) 
0.823
(0.223) 
Unemployed
(1 = yes) 
1.427
(0.465) 
1.298
(0.359) 
1.234
(0.438) 
1.163
(0.477) 
Retired
(1 = yes) 
1.219
(0.290) 
1.341
(0.359) 
0.832
(0.204) 
1.091
(0.329) 
Education     
University education
(1 = yes) 
0.372***
(0.092) 
0.629**
(0.107) 
0.369***
(0.072) 
0.577**
(0.101) 
Trust     
Trust
(1 to 5; no trust to complete trust) 
0.717***
(0.042) 
0.705***
(0.044) 
0.825**
(0.045) 
0.686***
(0.045) 
Organizational and attitudinal variables     
Religious person
(1 = yes) 
1.011
(0.157) 
0.987
(0.151) 
0.703*
(0.112) 
0.905
(0.179) 
Political party member
(1 = yes) 
0.922
(0.348) 
0.398
(0.173) 
0.308*
(0.176) 
1.237
(0.365) 
Labor union member
(1 = yes) 
0.756
(0.170) 
0.888
(0.275) 
1.447
(0.322) 
1.126
(0.199) 
Tolerance for inequality
(1 to 5; prefer equality to prefer inequality) 
1.034
(0.077) 
0.985
(0.061) 
1.079
(0.073) 
1.368
(0.110) 
Constant 1.963
(0.719) 
2.314*
(0.899) 
1.079
(0.073) 
0.307*
(0.146) 
Observations 980 850 914 1023 
Pseudo R-squared 0.067 0.065 0.071  
Source: Author's calculations based on LITS and ESS (see EBRD, 2020; ESS, 2020).

Note: Odds ratios and robust standard errors in parentheses. Statistical significance: * p < .05, ** p < .01, *** p < .001

TABLE 4.

Logistic regression models for welfare chauvinism in Eastern Europe (Life in Transition Survey)

(1)(2)(3)(4)
Poland
(2016)
Czech Rep.
(2016)
Hungary
(2016)
Slovenia
(2016)
Demographic variables     
Gender
(1 = male) 
1.291
(0.191) 
1.038
(0.163) 
1.149
(0.185) 
0.785
(0.103) 
Age 1.008
(0.006) 
0.998
(0.006) 
1.004
(0.007) 
0.999
(0.005) 
Socioeconomic variables     
Blue-collar worker
(1 = yes) 
1.070
(0.249) 
1.106
(0.244) 
0.981
(0.247) 
0.678
(0.165) 
Business owner
(1 = yes) 
0.462*
(0.137) 
0.981
(0.293) 
1.112
(0.491) 
0.521*
(0.148) 
Unemployed
(1 = yes) 
2.172**
(0.495) 
1.451
(0.392) 
1.208
(0.306) 
1.144
(0.195) 
Retired
(1 = yes) 
0.408**
(0.109) 
0.801
(0.249) 
0.612
(0.196) 
0.737
(0.165) 
Education     
University education
(1 = yes) 
1.018
(0.180) 
0.503**
(0.107) 
0.585
(0.183) 
0.367***
(0.077) 
Trust     
Trust
(1 to 5; no trust to complete trust) 
0.814**
(0.062) 
0.736***
(0.060) 
0.775**
(0.061) 
0.845*
(0.058) 
Organizational and attitudinal variables     
Active churchgoer
(1 = yes) 
1.383
(0.236) 
0.740
(0.325) 
2.139
(1.271) 
0.786
(0.220) 
Political party member
(1 = yes) 
2.322*
(0.939) 
0.950
(0.458) 
1.333
(1.396) 
2.596
(1.323) 
Labor union member
(1 = yes) 
3.645**
(1.808) 
2.380*
(0.850) 
0.706
(0.304) 
1.114
(0.270) 
Tolerance for inequality
(1 to 5; prefer equality to prefer inequality) 
0.814***
(0.043) 
0.866**
(0.043) 
0.955
(0.049) 
0.963
(0.054) 
Constant 3.838**
(1.527) 
2.277***
(0.412) 
7.471***
(3.088) 
3.963
(1.376) 
Observations 1058 1049 1032 1074 
Pseudo R-squared 0.055 0.041 0.019 0.060 
(1)(2)(3)(4)
Poland
(2016)
Czech Rep.
(2016)
Hungary
(2016)
Slovenia
(2016)
Demographic variables     
Gender
(1 = male) 
1.291
(0.191) 
1.038
(0.163) 
1.149
(0.185) 
0.785
(0.103) 
Age 1.008
(0.006) 
0.998
(0.006) 
1.004
(0.007) 
0.999
(0.005) 
Socioeconomic variables     
Blue-collar worker
(1 = yes) 
1.070
(0.249) 
1.106
(0.244) 
0.981
(0.247) 
0.678
(0.165) 
Business owner
(1 = yes) 
0.462*
(0.137) 
0.981
(0.293) 
1.112
(0.491) 
0.521*
(0.148) 
Unemployed
(1 = yes) 
2.172**
(0.495) 
1.451
(0.392) 
1.208
(0.306) 
1.144
(0.195) 
Retired
(1 = yes) 
0.408**
(0.109) 
0.801
(0.249) 
0.612
(0.196) 
0.737
(0.165) 
Education     
University education
(1 = yes) 
1.018
(0.180) 
0.503**
(0.107) 
0.585
(0.183) 
0.367***
(0.077) 
Trust     
Trust
(1 to 5; no trust to complete trust) 
0.814**
(0.062) 
0.736***
(0.060) 
0.775**
(0.061) 
0.845*
(0.058) 
Organizational and attitudinal variables     
Active churchgoer
(1 = yes) 
1.383
(0.236) 
0.740
(0.325) 
2.139
(1.271) 
0.786
(0.220) 
Political party member
(1 = yes) 
2.322*
(0.939) 
0.950
(0.458) 
1.333
(1.396) 
2.596
(1.323) 
Labor union member
(1 = yes) 
3.645**
(1.808) 
2.380*
(0.850) 
0.706
(0.304) 
1.114
(0.270) 
Tolerance for inequality
(1 to 5; prefer equality to prefer inequality) 
0.814***
(0.043) 
0.866**
(0.043) 
0.955
(0.049) 
0.963
(0.054) 
Constant 3.838**
(1.527) 
2.277***
(0.412) 
7.471***
(3.088) 
3.963
(1.376) 
Observations 1058 1049 1032 1074 
Pseudo R-squared 0.055 0.041 0.019 0.060 
Source: Author's calculations based on LITS and ESS (see EBRD, 2020; ESS, 2020).

Note: Odds ratios and robust standard errors in parentheses. Statistical significance: * p < .05, ** p < .01, *** p < .001

TABLE 5.

Logistic regression models for welfare chauvinism in Eastern Europe (European Social Survey)

(1)(2)(3)(4)
Poland
(2014)
Czech Rep.
(2014)
Hungary
(2014)
Slovenia
(2014)
Demographic variables     
Gender
(1 = male) 
1.046
(0.161) 
0.130
(0.159) 
0.966
(0.135) 
1.072
(0.177) 
Age 0.988
(0.006) 
1.006
(0.006) 
0.999
(0.006) 
1.002
(0.007) 
Socioeconomic variables     
Blue-collar worker
(1 = yes) 
1.198
(0.211) 
1.244
(0.186) 
0.912
(0.128) 
1.392
(0.234) 
Business owner
(1 = yes) 
1.648*
(0.324) 
1.381
(0.339) 
0.761
(0.211) 
0.633
(0.188) 
Unemployed
(1 = yes) 
1.287
(0.414) 
0.966
(0.294) 
0.840
(0.287) 
0.712
(0.206) 
Retired
(1 = yes) 
1.717*
(0.414) 
0.842
(0.190) 
0.878
(0.188) 
0.652
(0.170) 
Education     
University education
(1 = yes) 
1.178
(0.412) 
0.710
(0.142) 
0.780
(0.208) 
0.863
(0.243) 
Trust     
Trust
(1 to 5; no trust to complete trust) 
0.820**
(0.047) 
0.901*
(0.046) 
0.819***
(0.041) 
0.836**
(0.049) 
Organizational and attitudinal variables     
Religious person
(1 = yes) 
1.007
(0.156) 
0.611**
(0.112) 
0.969
(0.131) 
1.176
(0.194) 
Political party member
(1 = yes) 
0.688
(0.360) 
0.411
(0.205) 
2.609
(1.945) 
1.669
(0.641) 
Labor union member
(1 = yes) 
1.130
(0.347) 
0.703
(0.221) 
0.648
(0.207) 
1.420
(0.319) 
Tolerance for inequality
(1 to 5; prefer equality to prefer inequality) 
0.868
(0.068) 
0.909
(0.053) 
0.993
(0.079) 
1.129
(0.093) 
Constant 1.114
(0.409) 
1.869
(0.620) 
2.035*
(0.709) 
0.376*
(0.147) 
Observations 943 1004 996 971 
Pseudo R-squared 0.034 0.032 0.019 0.039 
(1)(2)(3)(4)
Poland
(2014)
Czech Rep.
(2014)
Hungary
(2014)
Slovenia
(2014)
Demographic variables     
Gender
(1 = male) 
1.046
(0.161) 
0.130
(0.159) 
0.966
(0.135) 
1.072
(0.177) 
Age 0.988
(0.006) 
1.006
(0.006) 
0.999
(0.006) 
1.002
(0.007) 
Socioeconomic variables     
Blue-collar worker
(1 = yes) 
1.198
(0.211) 
1.244
(0.186) 
0.912
(0.128) 
1.392
(0.234) 
Business owner
(1 = yes) 
1.648*
(0.324) 
1.381
(0.339) 
0.761
(0.211) 
0.633
(0.188) 
Unemployed
(1 = yes) 
1.287
(0.414) 
0.966
(0.294) 
0.840
(0.287) 
0.712
(0.206) 
Retired
(1 = yes) 
1.717*
(0.414) 
0.842
(0.190) 
0.878
(0.188) 
0.652
(0.170) 
Education     
University education
(1 = yes) 
1.178
(0.412) 
0.710
(0.142) 
0.780
(0.208) 
0.863
(0.243) 
Trust     
Trust
(1 to 5; no trust to complete trust) 
0.820**
(0.047) 
0.901*
(0.046) 
0.819***
(0.041) 
0.836**
(0.049) 
Organizational and attitudinal variables     
Religious person
(1 = yes) 
1.007
(0.156) 
0.611**
(0.112) 
0.969
(0.131) 
1.176
(0.194) 
Political party member
(1 = yes) 
0.688
(0.360) 
0.411
(0.205) 
2.609
(1.945) 
1.669
(0.641) 
Labor union member
(1 = yes) 
1.130
(0.347) 
0.703
(0.221) 
0.648
(0.207) 
1.420
(0.319) 
Tolerance for inequality
(1 to 5; prefer equality to prefer inequality) 
0.868
(0.068) 
0.909
(0.053) 
0.993
(0.079) 
1.129
(0.093) 
Constant 1.114
(0.409) 
1.869
(0.620) 
2.035*
(0.709) 
0.376*
(0.147) 
Observations 943 1004 996 971 
Pseudo R-squared 0.034 0.032 0.019 0.039 
Source: Author's calculations based on LITS and ESS (see EBRD, 2020; ESS, 2020).

Note: Odds ratios and robust standard errors in parentheses. Statistical significance: * p < .05, ** p < .01, *** p < .001

What are the results for Western Europe? As can be seen, there is little evidence to support Hypothesis 1 (the socioeconomic hypothesis). There are several variables that yield statistically significant results in Table 2, but these are not consistent throughout. Certainly, it does not seem that the working class is especially prone to welfare chauvinism. The variable that tracks blue-collar workers was statistically significant for Sweden in Table 2, but this result was not replicated in Table 3.

However, the situation is very different with regard to Hypothesis 2 (the education hypothesis). In all four models in Tables 2 and 3, the variable that tracks respondents with a university education leads to the same results: odds ratios were statistically significant and below 1 (i.e., the coefficient is negative). Therefore, it seems that higher education is a consistent negative determinant of welfare chauvinism: those respondents who had a higher education were less likely to adopt welfare chauvinism. Odds ratios presented in Tables 2 and 3 range from 0.63 to 0.29. Thus, respondents with a university education were anywhere from 37% to 71% less likely to adopt welfare chauvinism, relative to respondents who did not have a university education.

With regard to Hypothesis 3 (the trust hypothesis), there is also a lot of support. In all four models in Tables 2 and 3, the variable that tracks trust is statistically significant and negative (odds ratios below 1). Therefore, more-trusting people are less likely to adopt welfare chauvinism. Odds ratios range from 0.83 to 0.46. Given that the variable for trust is an ordinal scale (1 to 5), these odds ratios mean that for an increase of 1 on the independent variable (e.g., from 1 to 2, or from 2 to 3), the likelihood that a respondent adopts welfare chauvinism decreases anywhere from 17% to 54%.

What are the results for Eastern Europe? As can be seen, once again, very little support can be found for Hypothesis 1 (socioeconomic hypothesis). In some models in Tables 4 and 5, certain socioeconomic variables lead to statistically significant results, but there is nothing consistent across countries and data sets. Additionally, the variable that tracks blue-collar workers was not statistically significant in any of the models.

The results for Hypothesis 2 (the education hypothesis) are more surprising: university education was not a robust determinant of welfare chauvinism as it was in Western Europe. As can be seen, odds ratios for this variable were below 1 (i.e., coefficient was negative) and statistically significant in Models 2 and 4 (Czech Republic and Slovenia) in Table 4 (LITS), but these results were not replicated in Table 5 (ESS). Therefore, it cannot be concluded that education makes respondents less prone to welfare chauvinism in Eastern Europe.

The single robust finding for Eastern Europe concerns trust. The variable for trust produced statistically significant results with odds ratios less than 1 (i.e., negative coefficients) in all four models in Table 4 as well as in Table 5. Odds ratios range from 0.90 to 0.74. This means that an increase of 1 on the ordinal scale for trust (e.g., from 1 to 2, or from 2 to 3) may lead to a decrease in the likelihood that a respondent adopts welfare chauvinism of anywhere between 10% and 26%. Therefore, Hypothesis 3 (the trust hypothesis) is the only one that received support in Eastern Europe. People who have higher levels of trust are less likely to adopt welfare chauvinism. However, it should also be noted that the size of the effect is smaller in Eastern Europe than in Western Europe, which means that even the one conduit of chauvinism reduction that works in Eastern Europe, “does so” less effectively. The only other variable that produced robust results was the variable that tracks ethnic minorities in Slovenia. As would be expected, they were less prone to adopt welfare chauvinism.

Table 6 presents an overview and summary of the results, by combining those variables that were statistically significant determinants (with the same sign) for both data sets. As can be seen, in Western Europe, university education and trust were consistent negative determinants of welfare chauvinism. This result is valid for all four countries. With regard to Eastern Europe, the only consistent determinant in all four countries, also negative, was trust. In Slovenia, the status of ethnic minority status was an additional (negative) determinant of welfare chauvinism.

TABLE 6.

Overview of results

Statistically significant determinants of welfare chauvinism in both LITS and ESS
Western Europe  
Germany University education (negative), trust (negative) 
France University education (negative), trust (negative) 
UK University education (negative), trust (negative) 
Sweden University education (negative), trust (negative) 
Eastern Europe  
Poland Trust (negative) 
Czech Republic Trust (negative) 
Hungary Trust (negative) 
Slovenia Ethnic minority (negative), trust (negative) 
Statistically significant determinants of welfare chauvinism in both LITS and ESS
Western Europe  
Germany University education (negative), trust (negative) 
France University education (negative), trust (negative) 
UK University education (negative), trust (negative) 
Sweden University education (negative), trust (negative) 
Eastern Europe  
Poland Trust (negative) 
Czech Republic Trust (negative) 
Hungary Trust (negative) 
Slovenia Ethnic minority (negative), trust (negative) 

Table 7 presents a summary of the implications for each explanation, in both regions. All in all, the first hypothesis (socioeconomic position) received no support in either region; the second hypothesis (education) received support in Western Europe but not in Eastern Europe; and, finally, the third hypothesis (trust) received support in both regions.

TABLE 7.

Summary of implications for existing theories of welfare chauvinism

Socioeconomic position No support No support 
Cultural capital Support No support 
Trust Support Support 
Socioeconomic position No support No support 
Cultural capital Support No support 
Trust Support Support 

CONCLUSION

This article examined the strange case of welfare chauvinism in Eastern Europe. Countries in Eastern Europe tend to have higher shares of people who adopt welfare chauvinism, in comparison with Western Europe. This seems to be the case despite the fact that there is very little immigration into the region. The welfare state in Eastern Europe is not very generous either, and so it cannot be assumed to be a magnet for immigrants. So, why do East Europeans seem to be welfare chauvinists?

The analysis presented in this article examined two large survey projects—the European Social Survey and the Life in Transition Survey—in order to locate the most robust individual-level determinants of welfare chauvinism. All in all, the analysis points to certain similarities across and differences between Western and Eastern attitudes. In Western Europe, higher education is negatively associated with welfare chauvinism. In Eastern Europe, higher education does not seem to have an effect. However, in both regions interpersonal trust is negatively associated with welfare chauvinism.

There are several implications of this analysis. First, the fact that higher education does not weaken welfare chauvinism in Eastern Europe means that even this rather imperfect conduit of chauvinism reduction does not work. If we were to ask if western values can be exported east (Kymlicka, 2009), the answer would have to be: not through education, it seems. In Eastern Europe, it would appear that higher education does not by itself make people more accepting of diversity. In terms of policy prescriptions, it does not seem that a higher share of university graduates, for instance, would necessarily lead to a smaller share of welfare chauvinists.

Second, the finding regarding trust suggests that at least one conduit of chauvinism reduction works. Boosting trust may be a way to reduce welfare chauvinism. However, a note of caution is in order. As the analysis showed, effect sizes were much smaller in Eastern Europe than in Western Europe. This means that, although the mechanism works, it works less effectively. In addition, the share of people who are trusting of others is much lower in Eastern Europe: around 25–30% in Poland, Czech Republic, Slovenia, and Hungary. By contrast, the share of people who trust others increases to about 42% in Germany, 33% in France, 48% in UK, and 65% in Sweden.12 Once again, the impact that trust can have on attitudes is bound to be more limited in Eastern Europe.

Although the finding on trust is robust, it is difficult to say what policy prescriptions flow from it. One way of boosting trust is through a solidaristic and universalistic welfare state. The fact that Sweden has the lowest incidence of welfare chauvinism among all countries included in this analysis points to the same conclusion: a strong welfare state is a way to build trust, which then dampens welfare chauvinism (see also Crepaz & Damron, 2009). Such systems are better able to integrate immigrants, in comparison with less-generous welfare states built around the principle of means testing. In addition, the solidary character of the universalistic welfare state avoids the segmentation across groups that the liberal welfare state establishes and/or amplifies.

This study, naturally, has certain limitations. Although an analysis of data from existing large comparative survey projects can provide insights into the micro-level determinants of welfare chauvinism, it is less attuned to the impact of contextual variables that vary cross-nationally: aggregate immigration levels, welfare state generosity, and countrywide inequality, to name a few. The number of countries required for such an analysis is just not large enough. Even if one were to include all EU members, there would be fewer than 30 data points, which does not provide sufficient statistical power (however, for one such contribution, see van der Waal et al., 2013).

Additionally, the very thing that makes the case of Eastern European welfare chauvinism unusual—i.e., the fact that there is essentially no immigration—may also suggest a hypothesis for additional research. As contact theory suggests, contact between groups may be able to lessen animosity (see the seminal contributions in Williams, 1947; Allport, 1964). In other words, as East Europeans gain experience with immigration, as they meet and befriend immigrants, they may turn away from welfare chauvinism. The evidence for Western Europe seems to suggest that, all in all, this mechanism does indeed work (see the overview in McLaren, 2003). However, the impact of contact is difficult to gauge with general-purpose surveys such as ESS and LITS. More targeted research strategies will be required.

A final note of caution is in order as well. The data seem to be more unstable in the case of Eastern Europe. The large swings in the overall percentages, as shown in Table 1, indicate that public attitudes toward welfare chauvinism may not have reached a steady state. Therefore, welfare chauvinist views in the region may yet undergo significant changes. Researchers should be particularly sensitive to this volatility.

NOTES

1.

Slovakia and the Baltic countries are comparable. However, Slovakia was not included in the 2016 round of the European Social Survey and the appropriate question was not asked in 2010. The Baltic countries, on the other hand, were included in 2016, but they may be a little too remote as immigrant destinations. The other countries of Eastern Europe are significantly behind in terms of GDP per capita and thus of much less interest to potential immigrants: Croatia, Romania, Bulgaria, Serbia, Montenegro, North Macedonia, Bosnia and Herzegovina, Albania, Moldova, and Ukraine.

2.

Examples can readily be found in the media (e.g., Divisek, 2017; Traynor, 2015; Levy Gale, 2016).

3.

Figures based on European Commission data (see Eurostat, 2016).

4.

European Commission figures reported in Malnick (2013).

5.

See the EBRD website for access to the data (EBRD, 2020).

6.

Italy was included in the 2016 LITS wave as well but, unfortunately, the 2014 ESS Italian sample was not asked the appropriate question on welfare chauvinism. The most recent Italian sample that was asked this question was from 2004, and this was judged to be not recent enough (for more on the data, see EBRD, 2020; ESS, 2020).

7.

See the ESS website for access to the data (ESS, 2020).

8.

Data from the eighth round (2016) could not be used since it did not include the question on welfare chauvinism, which most resembles the one included in LITS.

9.

As mentioned, Slovakia would also be a possible case, but unfortunately, ESS was not administered in this country in 2016 or 2014. Going further back, the appropriate question on welfare chauvinism was not included in the fifth wave (2010). Therefore, Slovakia was excluded.

10.

The questionnaire for the 2010 round of the LITS includes a question on ethnicity, but it is absent from the data set. The work-around solution was to code the mother tongue of the respondent and their parents. If at least two of the three spoke something other than the dominant language in a given country (e.g., something other than French in France), the person would be coded as belonging to an ethnic minority (for the original data, see EBRD, 2020).

11.

For instance, in the 2016 round of the LITS, more than 60% of respondents in Germany failed to specify how much monthly income they have. In the Czech Republic and Slovenia, it was similar. In Poland, it was more than 70%. This constraint should be kept in mind when making conclusions about the impact of income on welfare chauvinism. For the 2010 round of the LITS, an additional problem is that respondents were not asked to cite their monthly income but to place themselves in the distribution of income in their country (from 1 to 10). The same approach is used in the ESS. With this approach, there is less missing data, usually in the single digits, but it introduces a source of bias that is difficult to gauge. Therefore, the results for the income variable should be taken with a grain of salt (for the original data, see EBRD, 2020; ESS, 2020).

12.

Calculations based on the 2014 round of the ESS (for the original data, see ESS, 2020).

REFERENCES

REFERENCES
Allport
,
G. W.
(
1964
)
The nature of prejudice
.
Albany, NY
,
Doubleday
.
Angrist
,
J. D.
&
Pischke
,
J. S.
(
2009
)
Mostly harmless econometrics: An empiricist's guide
.
Princeton, NJ
,
Princeton University Press
.
Bansak
,
C.
,
Simpson
,
N. B.
&
Zavodny
,
M.
(
2015
)
Economics of immigration
.
London and New York
,
Routledge
.
Banting
,
K.
&
Kymlicka
,
W.
(
2010
)
Multiculturalism and the welfare state: Recognition and redistribution in contemporary democracies
.
Oxford
,
Oxford University Press
.
Ceobanu
,
A. M.
&
Escandell
,
X.
(
2008
)
East is west? National feelings and anti-immigrant sentiment in Europe
.
Social Science Research
.
37
(
4
),
1147
1170
.
Collier
,
P.
(
2015
)
Exodus: How migration is changing our world
.
Oxford
,
Oxford University Press
.
Crepaz
,
M. M. L.
(
2010
)
Trust beyond borders: Immigration, the welfare state and identity in modern societies
.
Ann Arbor
,
University of Michigan Press
.
Crepaz
,
M. M. L.
&
Damron
,
R.
(
2009
)
Constructing tolerance: How the welfare state shapes attitudes about immigrants
.
Comparative Political Studies
.
42
(
3
),
437
463
.
Divisek
,
M.
,
2017
.
How (the European) Trump won a second term
.
Politico.eu
. Available from: https://www.politico.eu/article/milos-zeman-czech-republic-how-the-european-donald-trump-won-a-second-term/ [Accessed
15
July
2020
].
EBRD (European Bank for Reconstruction and Development)
. (
2020
)
Life in Transition Survey (LITS)
. EBRD. Available from: http://www.ebrd.com/what-we-do/economic-research-and-data/data/lits.html [Accessed
15
July
2020
].
ESS (European Social Survey)
. (
2020
)
Countries by round (year). European Social Survey
. Available from: http://www.europeansocialsurvey.org/data/country_index.html [Accessed
15
July
2020
].
Ervasti
,
H.
&
Hjerm
,
M.
(
2012
) Immigration, trust and support for the welfare state. In:
Ervasti
,
H.
,
Goul Andersen
,
J.
,
Fridberg
,
T.
&
Ringdal
,
K.
(eds.)
The future of the welfare state: Social policy attitudes and social capital in Europe
.
Cheltenham, UK
,
Edward Elgar
, pp.
153
171
.
Eurostat
. (
2016
)
Migration and migrant population statistics
.
Eurostat
. Available from: http://ec.europa.eu/eurostat/statistics-explained/index.php/Migration_and_migrant_population_statistics [Accessed
15
July
2020
].
Goul Andersen
,
J.
&
Bjorklund
,
T.
(
1990
)
Structural changes and new cleavages: The progress parties in Denmark and Norway
.
Acta Sociologica
.
33
(
3
),
195
217
.
Guiso
,
L.
,
Herrera
,
H.
,
Morelli
,
M.
&
Sonno
,
T.
(
2017
)
Demand and supply of populism
. Innocenzo Gasparini Institute for Economic Research Working Paper 610.
Houtman
,
D.
(
2001
) Class, culture and conservatism: Reassessing education as a variable in political sociology. In:
Clark
,
T.
&
Lipset
,
S. M.
(eds.)
The breakdown of class politics: A debate on post-industrial stratification
.
Washington, DC
,
Woodrow Wilson Center
, pp.
161
196
.
Inglehart
,
R.F.
&
Norris
,
P.
(
2016
)
Trump, Brexit and the rise of populism: Economic have-nots and cultural backlash
.
Harvard Kennedy School Working Paper Series
.
Kitschelt
,
H.
(
1997
)
The radical right in Western Europe: A comparative analysis
.
Ann Arbor
,
University of Michigan Press
.
Kouvo
,
A.
,
Kankainen
,
T.
&
Niemela
,
M.
(
2012
) Welfare benefits and generalized trust in Finland and Europe. In:
Ervasti
,
H.
,
Goul Andersen
,
J.
,
Fridberg
,
T.
&
Ringdal
,
K.
(eds.)
The future of the welfare state: Social policy attitudes and social capital in Europe
.
Cheltenham, UK
,
Edward Elgar
, pp.
195
213
.
Kymlicka
,
W.
(
2009
)
Can liberal pluralism be exported? Western political theory and ethnic relations in Eastern Europe
.
Oxford
,
Oxford University Press
.
Lamont
,
M.
(
1987
)
Cultural capital and the liberal political attitudes of professionals: Comment on Brint
.
American Journal of Sociology
.
92
(
6
),
1501
1506
.
Levy Gale
,
S.
(
2016
)
Hungarian prime minister calls migrants entering Europe a “poison.”
Independent
.
27
July
. Available from: http://www.independent.co.uk/news/world/europe/refugee-crisis-hungary-migrants-europe-prime-minister-viktor-orban-asylum-seekers-a7157871.html [Accessed
15
July
2020
].
Lipset
,
S. M.
(
2001
)
The breakdown of class politics: A debate on post-industrial stratification
.
Washington, DC
,
Woodrow Wilson Center
, pp.
161
196
.
Malchow-Moller
,
N.
,
Munch
,
J. R.
,
Schroll
,
S.
&
Skaksen
,
J. R.
(
2008
)
Attitudes towards immigration: Perceived consequences and economic self-interest
.
Economics Letters
.
100
(
2
),
254
257
.
Malnick
,
E.
,
2013
.
Benefits in Europe: Country by country
.
Telegraph
.
19
October
. Available from: https://www.telegraph.co.uk/news/worldnews/europe/10391238/Benefits-in-Europe-country-by-country.html [Accessed
15
July
2020
].
McLaren
,
L. M.
(
2003
)
Anti-immigrant prejudice in Europe: Contact, threat perception, and preferences for the exclusion of migrants
.
Social Forces
.
81
(
3
),
909
936
.
Mewes
,
J.
&
Mau
,
S.
(
2013
)
Globalization, socio-economic status and welfare chauvinism: European perspectives on attitudes toward the exclusion of immigrants
.
International Journal of Comparative Sociology
.
54
(
3
),
228
245
.
Mudde
,
C.
(
2005
)
Racist extremism in central and eastern Europe
.
East European Politics and Societies
.
19
(
2
),
161
184
.
Putnam
,
R. D.
(
2006
)
Making democracy work
.
Princeton, NJ
,
Princeton University Press
.
Rothstein
,
B.
&
Uslaner
,
E.
(
2005
)
All for all: Equality, corruption, and social trust
.
World Politics
.
58
(
1
),
41
72
.
Scheve
,
K. F.
&
Slaughter
,
M. J.
(
2001
)
Labor market competition and individual preferences over immigration policy
.
Review of Economics and Statistics
.
83
(
1
),
133
145
.
Stubager
,
R.
(
2008
).
Education effects on authoritarian-libertarian values: A question of socialization
.
British Journal of Sociology
.
59
(
2
),
327
350
.
Traynor
,
I.
(
2015
)
Hungary PM: Bring back death penalty and build work camps for immigrants
.
Guardian
.
29
April
. Available from: https://www.theguardian.com/world/2015/apr/29/hungary-pm-death-penalty-work-camps-for-immigrants-viktor-orban [Accessed
6
June
2020
].
Uslaner
,
E.
(
2003
)
The moral foundations of trust
.
Cambridge
,
Cambridge University Press
.
van der Waal
,
J.
,
Achterberg
,
P.
,
Houtman
,
D.
,
de Koster
,
W.
&
Manevska
,
K.
(
2010
)
“Some are more equal than others”: Economic egalitarianism and welfare chauvinism in the Netherlands
.
Journal of European Public Policy
.
20
(
4
),
350
363
.
van der Waal
,
J.
,
de Koster
,
W.
&
van Oorschot
,
W.
(
2013
)
Three worlds of welfare chauvinism? How welfare regimes affect support for distributing welfare to immigrants in Europe
.
Journal of Comparative Policy Analysis: Research and Practice
.
15
(
2
),
164
181
.
van Oorshot
,
W.
(
2006
)
Making the difference in social Europe: Deservingness perceptions among citizens of European welfare states
.
Journal of European Social Policy
.
16
(
1
),
23
42
.
Wilkinson
,
R.
&
Pickett
,
K.
(
2010
)
The spirit level: Why equality is better for everyone
.
London
,
Penguin
.
Williams
,
R. M.
(
1947
)
The reduction of intergroup tension
.
New York
,
Social Science Research Council
.