Ethnic Segregation and Radical Right-Wing Voting in Dutch Cities

Abstract

Previous studies have linked anti-immigrant voting and other indications of ethnic animosities to ethnic segregation, yielding different results. In this study, we focus on the locally strongly diverging support for Geert Wilders’s Party for Freedom (Partij voor de Vrijheid [PVV]) in the Dutch national parliamentary elections of 2006 and 2010 to assess how it can be understood that the effect of ethnic segregation on anti-immigrant voting varies, and how this can be theoretically interpreted. Our analyses on 50 Dutch cities demonstrate that ethnic segregation leads to PVV voting, and that this positive effect is stronger in cities with a more tolerant cultural atmosphere and lower levels of unemployment. This positive effect is at odds with ethnic threat theory, and our contextualization informed by the cultural and economic conditions of cities enables empirically distinguishing between contact theory and concentration theory. Whereas both predict a positive effect, only contact theory is corroborated by our results.

Keywords

bohemian index ethnic contact ethnic segregation ethnic threat radical right-wing voting urban culture

Introduction

Issues concerning immigration and the integration of ethnic minorities gained a central role on the political stage in most European countries in recent decades (Achterberg 2006; Ignazi 2003). This is reflected in the rise of various radical right-wing parties, whose electoral support proves to be largely driven by discontents concerning immigration and ethnic diversity (Oesch 2008; Rydgren 2008; Van der Brug 2003). Unsurprisingly, such parties have therefore been frequently labeled as “anti-immigration parties” (see, for example, De Koster, Achterberg, and Van der Waal forthcoming; Rink, Phalet, and Swyngedouw 2009; Van der Brug and Fennema 2007; Van der Brug, Fennema, and Tillie 2000, 2005) and “movements of exclusion” (Rydgren 2005).

The electoral support for such anti-immigration parties has been linked to ethnic segregation in the scholarly literature (e.g., Biggs and Knauss 2011; Van der Waal, De Koster, and Achterberg 2011). Likewise, it has been found that ethnic segregation affects a wide range of other issues that are directly or indirectly related to ethnic animosities, such as the level of ethnocentrism (Rocha and Espino 2009), support for racial policies among the “white” population (Kinder and Mendelberg 1995), and police arrests of “blacks” (Stolzenberg, D’Alessio, and Eitle 2004). Most of such studies are informed by ethnic threat theory, which basically claims that confrontations between the majority population and ethnic minorities generate competition for scarce resources and value dominance, and hence lead to a negative stance toward the other group (cf. Esses, Jackson, and Bennett-AbuAyyash 2011).

From this theory, it follows that high levels of ethnic segregation are expected to be accompanied by relatively low levels of support for anti-immigrant parties, as high levels of segregation indicate that ethnic groups hardly confront each other. The results found in empirical research on this matter, however, vary greatly, even to the extent that some studies find a negative relationship between ethnic segregation and various indicators of resistance toward immigrants and ethnic minorities, most notably “blacks” (e.g., Kent and Jacobs 2005; Stolzenberg, D’Alessio, and Eitle 2004), whereas others report positive relationships (Biggs and Knauss 2011; Kinder and Mendelberg 1995).¹

In light of these mixed findings, it is important to note that two theories exist that suggest that the relationship between ethnic segregation and anti-immigrant voting or other expressions of ethnic animosities is positive and not negative. First, Allport’s (1954/1979) famous contact theory holds that interethnic contact diminishes racial prejudice and its corollaries under certain conditions. In his seminal work, The Nature of Prejudice, he, therefore, already predicted that desegregation in the United States “creates a condition where friendly contacts and accurate social perceptions can occur” (Allport 1954/1979, p. 260). Second, a line of reasoning that can be termed concentration theory, which can be traced back to the work of Hawley (1944), also predicts a positive relationship between ethnic segregation and anti-immigrant voting, albeit on the basis of a different theoretical rationale. This theory suggests that segregation “causes the group’s peculiarities to stand out in clear contrast to the traits of the majority, it emphasizes the uniformity within the group. Thus the group is exposed to categorical definitions of one kind and another” (Hawley 1944, p. 672).

The above gives rise to two questions. How can the varying effects of segregation on anti-immigrant voting and other indicators of ethnic animosities be understood? And in case of positive effects, is it possible to determine whether the interpretation offered by contact theory or the one proposed by concentration theory is valid? To make a contribution to answering these questions, we propose to contextualize the effect of ethnic segregation on anti-immigrant voting by means of the economic and cultural conditions of cities, as is outlined below.

The Dutch case provides an excellent opportunity for doing so, because of the substantial yet locally strongly diverging electoral support for Geert Wilders’s Party for Freedom (Partij voor de Vrijheid [PVV]) in the Dutch national parliamentary elections of 2006 and 2010. The radical right-wing PVV is widely considered to be the Dutch anti-immigrant party par excellence. International observers have noted that it “has been particularly vocal in proposing controversial policies and in resorting to racist or xenophobic discourse” (European Commission Against Racism and Intolerance [ECRI] 2008, p. 35), and numerous political scientists indicate that its most distinctive feature is Wilders’s focus on a strict and ethnocentric agenda regarding issues of immigration and integration (see, for example, De Lange and Art 2011; Otjes 2011; Van Kersbergen and Krouwel 2008). As such, the PVV is widely considered to be the political heir of the late Pim Fortuyn’s LPF (Lijst Pim Fortuyn [List Pim Fortuyn]), which made its entry in the Dutch multiparty system by putting issues of integration and immigration on the Dutch political agenda in 2002 (cf. De Lange and Art 2011; Otjes 2011; Pellikaan, De Lange, and Van der Meer 2007; Penninx 2006; Van Kersbergen and Krouwel 2008). In what follows, we aim to find out to what extent ethnic segregation affects support for the PVV in Dutch cities and how this can be interpreted. Before presenting our research, we elaborate on the theoretical notions guiding our inquiry and the hypotheses deducted from these.

Ethnic Segregation, Anti-ImmigrantVoting, and Urban Contexts

Three Theories on the Impact of Ethnic Segregation on Anti-Immigrant Voting

Most studies that focus on the impact of ethnic segregation on anti-immigrant voting and sentiments are informed by so-called “ethnic threat theory.” This theory sees resistance toward immigrants as resulting from competition between ethnic groups in the broadest sense: competition over economic resources, political power, beliefs and values, and so on.² In short, according to ethnic threat theory, confrontations with immigrants lead the native population to consider immigrants threatening to their own way of life. As such, the theory predicts that more interethnic contacts lead to anti-immigrant sentiments and support for anti-immigrant parties. Applied to the issue at hand, this theory resembles V. O. Key’s (1949/1984) famous work explaining the political behavior among whites in the South of the United States, which has been corroborated time and again in empirical research (e.g., Giles and Buckner 1993; Wright 1977): The highest support for “racist candidates” can be found in places with most opportunities for contacts with African-Americans. Thus, from ethnic threat theory, it follows that ethnic segregation is likely to diminish support for an anti-immigrant party, as natives are hardly confronted with immigrants in highly ethnically segregated contexts.

This theory contrasts with Allport’s (1954/1979) famous contact theory. Simply put, contact theory holds that familiarity breeds understanding. This implies that less hostility toward immigrants exists in contexts where the native population can come into contact with inhabitants of foreign descent. Taking into account the fact that “segregation severely restricts the operation of positive intergroup contact processes” (Pettigrew, Wagner, and Christ 2010, p. 637), this theory gives reason to expect a positive instead of a negative relationship between ethnic segregation and support for an anti-immigrant party. Whereas this suffices to distinguish it from ethnic threat theory, it is important to note that contact theory is more complex than this. As Allport and many others have emphasized, more is needed for friendly contacts to come into existence than mere spatial proximity: Contacts only induce mutual understanding under specific conditions (cf. Pettigrew 1998). Among the conditions that need to be met before residential proximity between immigrants and natives leads to contacts that reduce interethnic animosity is that natives should already have a certain level of receptivity toward immigrants. This suggests that low levels of ethnic segregation steer natives more strongly away from anti-immigrant voting in urban contexts characterized by specific conditions inducing positive or less negative views of immigrants. In other words, the positive effect of ethnic segregation on anti-immigrant voting predicted by contact theory is supposed to be stronger in urban contexts characterized by conditions making natives more receptive toward immigrants.

Such a contextualization of contact theory is exactly what is needed to empirically disentangle it from concentration theory. This third theory also predicts that ethnic segregation leads to anti-immigrant voting, which means that merely distinguishing between positive and negative effects is not helpful for doing so. Concentration theory basically claims that high levels of ethnic segregation spawn feelings of anxiety among the native population because segregated ethnic minority groups can be perceived as threatening monolithic wholes. In other words, strong concentrations of ethnic minorities—implied by high levels of ethnic segregation—may arouse negative feelings toward them among the majority population because they are perceived as an internally cohesive out-group, fueling fears of a strange, ominous “parallel society.” This line of reasoning is present in the works of Hawley (1944, pp. 668-69), who claims that segregation

has the effect of throwing into sharp focus the differences between the groups; in fact, it accentuates those differences through heightening the visibility of the minority population. In exaggerating the illusion of homogeneity and, at the same time, obscuring the reality of individual variation spatial segregation provides support for prevailing stereotypes . . . In short, spatial segregation keeps the interracial issue in the foreground of attention.

Hawley’s argument is echoed in various other studies on the subject (e.g., Esser 1986; Jiobu and Marshall 1971; Lieberson 1961; Taylor 1979) and the underlying mechanism should be distinguished from contact theory: “Hawley (1944) and Allport (1954) suggested that residential segregation increases ethnic prejudice. For Hawley, this occurs because segregation heightens the visibility of minority groups and facilitates stereotyping. For Allport, residential segregation limits chances for interpersonal contact” (Kunovich and Hodson 2002, p. 193).³

Because contact theory and concentration theory both predict a positive effect of ethnic segregation on anti-immigrant voting, they can be easily discerned from ethnic threat theory in empirical research. For empirically distinguishing between contact theory and concentration theory, however, more is needed. It is important to note that from a mere positive relationship between ethnic segregation and support for anti-immigrant parties, it cannot be determined whether contact theory or concentration theory is supported. A recent study reporting a positive relationship between ethnic segregation and support for the British National Party nevertheless interpreted it according to concentration theory, claiming that ethnic segregation “makes difference—butchers advertising halal meat, women wearing headscarves—more visible, and it makes assimilation seem more uncertain” (Biggs and Knauss 2011, p. 3). It is, however, an empirical question whether this really underlies the study’s finding, or whether the theoretical logic of Allport’s contact theory accounts for it instead. Differences between urban contexts provide a key to answering such questions. This is because concentration theory, contrary to contact theory, predicts that the positive effect of ethnic segregation on anti-immigrant voting is weaker in urban contexts characterized by specific conditions making natives more receptive toward immigrants. After all, natives living under conditions inducing positive or less negative views of immigrants will most likely find spatial concentrations of immigrants seem less threatening.

The Cultural and Economic Conditions of Cities

These abstract ideas on the importance of urban contexts for the effect of ethnic segregation on anti-immigrant voting can be incorporated in empirical research by identifying two specific types of urban contexts. First, research suggests that this effect is conditional on the cultural atmosphere of cities. In line with earlier studies demonstrating the existence of various regional cultures that affect political attitudes of regional populations in the United States (e.g., Elazar 1966; Lieske 1993), various scholars have claimed that some cities have a far more culturally tolerant atmosphere than others (Brown, Knopp, and Morrill 2005; Clark and Rempel 1997; Florida 2004, 2005; Sharp 1996, 2002). And recent research on American (Sharp and Joslyn 2008) and Dutch (Van der Waal and Houtman 2011) cities, indeed, demonstrates that this urban cultural atmosphere affects the racial tolerance (anti-immigrant sentiments) of whites (natives) in accordance to those claims. In Dutch cities with the most culturally tolerant atmosphere—as measured by Florida’s bohemian index—ethnocentrism proves substantially lower among both less- and high-educated natives than in cities with the least culturally tolerant atmosphere (Van der Waal and Houtman 2011). It is important to note that this is not a compositional effect but a contextual effect. In other words, it is not the sum of individual characteristics but a certain milieu or atmosphere that affects the ideological outlook of the urban population (cf. Clark and Harvey 2010; Deleon and Naff 2004; Sharp 2007). In cities with the most culturally tolerant atmosphere, the native population thus is far more receptive toward immigrants than natives in cities with the least tolerant cultural atmosphere are. This suggests that the impact of ethnic segregation on support for an anti-immigrant party will differ between the former and the latter cities.

The second urban condition that is likely to affect the impact of ethnic segregation on anti-immigrant voting is the economic opportunity structure. Some regions or cities fare less well in an economic sense than others. This might affect the resistance toward immigrants of the urban population because of “social closure” (Weber 1922/2006; cf. Roscigno, Garcia, and Bobbitt-Zeher 2007), which suggests that people will be less receptive toward immigrants if economic opportunities are scarce (cf. Blalock 1967; Dancygier 2010; Olzak 1992). This, too, suggests that the impact of ethnic segregation on support for anti-immigrant parties varies across cities, because natives in cities with scarce labor-market opportunities will be less receptive to immigrants than those in cities with abundant labor-market opportunities.

Three Sets of Hypotheses

Combining these insights on the cultural and economic conditions of cities with ethnic threat theory, contact theory, and concentration theory, respectively, leads to three sets of expectations that enable us to find out according to which of those three theories the effect of ethnic segregation on anti-immigrant voting can be interpreted. For each of these theories, we formulate hypotheses on the overall effect of segregation on anti-immigrant voting (Hypothesis 1), the moderating role of the urban cultural atmosphere (Hypothesis 2), and the moderating role of the urban opportunity structure (Hypothesis 3).

First, as low levels of ethnic segregation indicate high exposure toward immigrants, ethnic threat theory predicts that lower levels of ethnic segregation lead to higher levels of support for the PVV (Hypothesis 1a). Yet, this negative effect is likely to be affected by the cultural and economic conditions of cities. In Dutch cities with the most culturally tolerant atmosphere, the native population is most receptive toward immigrants (Van der Waal and Houtman 2011). It therefore seems likely that the negative effect of ethnic segregation on PVV voting predicted by ethnic threat theory will be least strong in cities with the most culturally tolerant atmosphere (Hypothesis 2a). Interestingly, the results of many so-called “context studies” in the United States point in this direction: Studies reporting empirical support for the ethnic threat theory primarily focus on the effect of living among many African-Americans on the ethnocentrism or support for racist candidates among “whites” in the South (Giles and Buckner 1993; Oliver and Wong 2003; Stein, Post, and Rinden 2000; Wright 1977), which is renowned for its culturally intolerant atmosphere (Kuklinski, Cobb, and Gilens 1997; Quillian 1996). In other words, if the theoretical logic of ethnic threat theory accounts for the effect of ethnic segregation on support for the PVV, this effect should not only be negative but also most strongly so in cities with the least culturally tolerant atmosphere. Next to these cultural conditions, economic conditions are likely to matter as well. As natives are most likely to be least receptive toward immigrants in Dutch cities offering least economic opportunities, it can be expected that the negative effect of ethnic segregation on PVV voting predicted by ethnic threat theory will be strongest in cities with the highest unemployment levels (Hypothesis 3a). Corroboration of these two additional hypotheses would offer additional evidence that a negative effect of ethnic segregation on PVV voting should be interpreted according to the central arguments of ethnic threat theory.

Second, contact theory stands in contrast to ethnic threat theory because it predicts a positive instead of a negative effect of segregation on PVV voting (Hypothesis 1b). Again, this effect should vary with cultural and economic urban characteristics. This theory states that more is needed for friendly interethnic contacts to come into existence than mere spatial proximity. One of the necessities is that natives are already to a certain extent receptive toward immigrants. This implies that the central mechanism in contact theory is especially salient in cities with the most culturally tolerant atmosphere. This is because interethnic contacts reduce racial prejudice most strongly in such tolerant cities. In cities with the least culturally tolerant atmosphere, however, the majority population is less open toward cultural others, and as a result interethnic contacts are less likely to lead to less anti-immigrant voting. Consequently, low levels of ethic segregation are less likely to result in low support for the PVV in cities with the least culturally tolerant atmosphere than in cities with the most culturally tolerant atmosphere. This suggests that the positive effect of ethnic segregation on PVV voting predicted by the contact theory is strongest in cities with the most tolerant cultural atmosphere (Hypothesis 2b).

In addition, contact theory implies that economic conditions of cities matter too. One would expect that interethnic contacts do not underlie openness toward immigrants in cities where the native population is not receptive to immigrants because of scarce economic resources. Two studies support this expectation. Hjerm (2009, p. 47) recently found that “the economic context matters in that anti-immigrant attitudes of people are strongest in poor municipalities with a large share of immigrants.” Branton and Jones (2005, p. 359, emphasis added) found in a study on the United States that

high socioeconomic context and highly diverse contexts are related to higher levels of support for racial social issues; however, contexts characterized by low socioeconomic [status] and high racial and ethnic diversity are associated with lower levels of support for such issues.

Both studies on racial attitudes support our suggestion that the positive relationship between ethnic segregation and support for anti-immigrant parties—a consequence of such attitudes—predicted by contact theory is affected by the economic fortunes of urbanites. More specifically, contact theory underlies the expectation that the positive relationship between ethic segregation and support for the PVV is least strong in cities with the highest unemployment levels (Hypothesis 3b).

To summarize the above, if the effect of ethnic segregation on anti-immigrant voting is to be understood by means of contact theory, the effect should be positive, and it should be most strongly so in cities with the most culturally tolerant atmosphere (Hypothesis 2b) and in cities with the lowest unemployment levels (Hypothesis 3b). Things are different, however, if one follows concentration theory. This theory, too, predicts a positive effect of ethnic segregation on anti-immigrant voting (Hypothesis 1b), but the impact of cultural and economic urban conditions as theorized by concentration theory differs from the predictions derived from contact theory. Combining concentration theory with the finding that the native population is more receptive toward immigrants in cities characterized by a tolerant cultural atmosphere, it is to be expected that in those cities high concentrations of ethnic minorities—as indicated by high levels of ethnic segregation—will be considered less threatening than in cities with the least culturally tolerant atmosphere. The positive effect of ethnic segregation on PVV voting predicted by concentration theory is therefore likely to be weaker in cities with a more tolerant cultural atmosphere (Hypothesis 2c). Moreover, if people are least receptive toward immigrants in cities with little economic opportunities, spatial concentrations of immigrants are considered most threatening by natives living in such cities. This implies that the positive effect of ethnic segregation on PVV voting that concentration theory predicts is strongest in cities with the highest unemployment levels (Hypothesis 3c). Needless to say, Hypotheses 2c and 3c stand in contrast to predictions informed by contact theory.

For reasons of clarity, we have depicted the hypotheses formulated above in two figures. Figure 1 shows the hypotheses formulated by combining insights on the varying levels of cultural tolerance across cities with ethnic threat theory (Hypothesis 2a), contact theory (Hypothesis 2b), and concentration theory (Hypothesis 2c), respectively. Ethnic threat theory predicts a negative effect of ethnic segregation on anti-immigrant voting that is weaker in cities that are more culturally tolerant, contact theory predicts a positive effect that is stronger in such cities, while concentration theory predicts a positive effect that is weaker in such cities. Figure 2 depicts the hypotheses formulated by taking urban economic conditions into account. Ethnic threat theory predicts a negative effect of ethnic segregation on anti-immigrant voting that is stronger in cities with higher levels of unemployment (Hypothesis 3a), contact theory predicts a positive effect that is weaker in such cities (Hypothesis 3b), while concentration theory predicts a positive effect that is stronger in such cities (Hypothesis 3c).

Figure 1.

Hypothesized interaction effects of ethnic segregation with the level of tolerance of the urban cultural atmosphere on the share of the population that supports the PVV.

Figure 2.

Hypothesized interaction effects of ethnic segregation with the urban economic opportunity structure on the share of the population that supports the PVV.

Data and Operationalization

We have constructed our own data set by combining data retrieved from three different sources. The first one is the atlas for municipalities (Atlas voor Gemeenten; www.atlasvoorgemeenten.nl), which contains an abundance of city-level data for the 50 largest Dutch municipalities in 2004 and 2008, including most of the independent variables needed for our analyses.⁴ The second source is the Statline service of Statistics Netherlands (Centraal Bureau voor de Statistiek; statline.cbs.nl/statweb/?LA=en), from which we retrieved the data for calculating our ethnic segregation measure for those 50 cities, and the unemployment levels in the 29 COROP (Coördinatie Commissie Regionaal OnderzoeksProgramma [Coordination Committee Regional Research Programme]) areas in which those cities are located. Such a COROP area is based on a nodal classification principle, and is primarily measured by labor-market relations such as commuting. As such, it is the most appropriate level to measure labor-market indicators such as unemployment level. The third data source is the elections database (databank verkiezingsuitslagen; www.verkiezingsuitslagen.nl), from which we retrieved data on the share of votes for the PVV in the Dutch national parliamentary elections of 2006 and 2010 in all 50 cities. Table 1 shows the descriptive statistics of all unstandardized variables discussed below, and Table 2 shows their correlations. All variables have been standardized in the analyses that follow.

Table 1.

Descriptive Statistics.

	Minimum	Maximum	M	SD
Votes for PVV 2006 (n = 50)	2.50	17.80	6.570	2.659
Votes for PVV 2010 (n = 50)	7.40	30.20	16.684	5.188
Ethnic segregation 2006 (n = 50)	0.13	1.00	0.394	0.184
Ethnic segregation 2010 (n = 50)	0.14	0.99	0.409	0.179
Unemployment 2006 (n = 29)	3.40	7.40	5.383	1.161
Unemployment 2010 (n = 29)	4.00	7.90	5.441	0.909
Bohemian index (n = 50)	0.07	2.00	0.398	0.318
Bohemian index (Amsterdam excluded) (n = 49)	0.07	1.04	0.365	0.221
Share higher educated (n = 50)	1.50	29.70	10.418	6.266

Source: Atlas for municipalities 2004 and 2008, elections database, and Statistics Netherlands (CBS) (own calculations).

Note: PVV = Partij voor de Vrijheid; CBS = Centraal Bureau voor de Statistiek.

Table 2.

Correlations.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)
Votes for PVV 2006 (1)	1.000	.904****	.001	.009	.209	.314**	−.420***	−.491****	−.395***
Votes for PVV 2010 (2)		1.000	−.123	−.118	.258*	.420***	−.507****	−.591****	−.527****
Ethnic segregation 2006 (3)			1.000	.992****	.107	−.023	.453****	.226	.311**
Ethnic segregation 2010 (4)				1.000	.092	−.023	.437***	.220	.309**
Unemployment 2006 (5)					1.000	.804****	−.067	−.170	−.202
Unemployment 2010 (6)						1.000	−.156	−.261	−.289**
Bohemian index (7)							1.000	1.000	.652****
Bohemian index (Amsterdam excluded) (8)								1.000	.695****
Share higher educated (9)									1.000

Source: Atlas for municipalities 2004 and 2008, elections database, and Statistics Netherlands (CBS) (own calculations).

Note: PVV = Partij voor de Vrijheid; CBS = Centraal Bureau voor de Statistiek.

p < .1. **p < .05. ***p < .01. ****p < .001 (two-sided tests).

Votes for PVV 2006 and 2010 measure the share of the population in each municipality that voted for the PVV in the national parliamentary elections of 2006 and 2010, respectively. Although the PVV received substantially more support in 2010 than in 2006, the geographical pattern of this support is very stable: The correlation between PVV support in 2006 and 2010 is .904 (n = 50, p < .0005). We will test our hypotheses for these elections separately.

Ethnic segregation is also measured for both election years (2006 and 2010). Most studies focusing on the consequences of ethnic segregation use Duncan and Duncan’s (1955) index of dissimilarity (D) (cf. Lieberson 1980; Massey and Denton 1988). There are several good reasons for this “Pax Duncana” (Massey and Denton 1988, p. 281), but for the issue at hand it is not the most appropriate indicator, as it is independent of group size (Lieberson 1980; Massey and Denton 1988; Robinson 1980). This has previously been noted by Lieberson (1980, p. 254):

The index of dissimilarity by itself is inadequate for some research because D is insensitive to the actual interaction potential between the groups, a factor that is affected by the relative number as well as the pattern of spatial segregation.

To overcome this problem, there is need for a segregation measure that is not independent of group size, and consequently indicates potential contacts instead of merely uneven distribution. Therefore, Lieberson created his isolation index, which he defined as the “average probability of interacting with some specified population based on the distribution of persons by subareas and the assumption that interaction is with someone in the same subarea” (Lieberson quoted in Robinson 1980, p. 308). He observed that “as one moves away from describing dissimilarity in spatial patterns and gets into the question of actual isolation of the groups, it is likely that P* [i.e., Lieberson’s index] will prove to be a more complete and adequate measure” (Robinson 1980, p. 309) than Duncan’s index of dissimilarity. Lieberson’s isolation index—which is also referred to as “exposure index”—is measured as follows: I = Σ [(x_i / X) × (x_i / t_i)]. Where x_i is the minority population of a neighborhood, X is the total minority population in the city, and t_i is the total population of the neighborhood. To calculate Lieberson’s isolation index for a given city, the scores of all its neighborhoods are summarized. Higher scores on this measure indicate higher levels of segregation—The rank order of cities did not change between those years considering the strong correlation between the indices of 2006 and 2010 (.992, n = 50, p < .0005).

Unemployment 2006 and 2010 are two indicators for the urban economic opportunity structure. They measure the share of the working population in a so-called “COROP” area that is looking for a job—the standard unemployment indicator in the Netherlands. Although the unemployment level in 2010 is higher than in 2006 - representing the economic decline in the Netherlands in that period—the correlation between the two is substantial: .755 (n = 29, p < .0005). This means that the economic opportunity structure was generally better prior to the elections of 2006 than prior to the elections of 2010, but that the differences between cities roughly remained intact. Clearly, irrespective of the general economic climate, some cities are characterized by more economic opportunities than others.

Bohemian index measures the extent to which cities have an atmosphere of cultural tolerance, for, according to Florida (2002, p. 64; 2005, pp. 113-28), a high concentration of bohemians in a city “indicate[s] an underlying openness to diversity.” It measures the share of the urban population involved in the production of culture and the arts, such as writers, designers, architects, composers, painters, sculptors, photographers, dancers, artists, and actors. According to Florida’s reasoning, it is not the sheer number of bohemians that makes a city’s atmosphere culturally tolerant; instead, bohemians are drawn toward cities with the most culturally tolerant atmosphere, and therefore their presence is considered a valid indicator for an urban context of cultural tolerance. The atlas for municipalities contains the bohemian index only for 2004. Contrary to the one of Florida, it has not been measured by means of census data on occupations. For many Dutch cities, the small share of urbanites employed in culture and in the arts does not allow to make reliable estimates on the basis of such data. Therefore, the bohemian index has been measured in the atlas for municipalities by means of data on members of the federation of artist associations (Federatie van Kunstenaarsverenigingen), which results in more reliable estimates. Due to outlier Amsterdam, this index is distributed very unequally. That is why our analyses will be conducted both with and without this outlier, as to make sure that our findings are not outlier-driven. Higher scores on bohemian index indicate a more culturally tolerant urban atmosphere.

Share higher educated is a control variable measuring the share of the urban population that is highly educated (over Level 4 in the UN International Standard Classification of Education [ISCED] Code). This control is included as to make sure that the bohemian index measures a city-level phenomenon instead of a compositional effect driven by the well-documented higher level of cultural tolerance of the higher educated (Achterberg 2006; Van der Waal and Houtman 2011). It is therefore measured for the same year as the bohemian index: 2004.

The bivariate correlations in Table 2 show that—besides the obvious high correlations between indicators that have been measured in 2006 and 2010—no independent variables are correlated very strongly. This indicates that despite the relatively small number of cases, multicollinearity problems are very unlikely.⁵

Results

The empirical analyses on the Dutch national parliamentary elections of 2006 can be found in Table 3 and those on the elections of 2010 in Table 4. Because the data are nested (we analyzed all 50 Dutch cities nested within all 29 Dutch COROP areas), there is need for multilevel modeling. Tables 3 and 4 therefore start with a null model. These indicate that 24.3% of the variation in support for the PVV exists at city level in 2006 (0.266 / (0.266 + 0.827)), while the variation at city level is 35.9% in 2010 (0.349 / (0.349 + 0.623)). The 75.7% (2006) and 64.1% (2010) of the variance that remain can consequently be attributed to differences between COROP areas. Model 1 of both tables shows roughly similar results: Controlled for the share of higher-educated urbanites, ethnic segregation is positively related to the share of the population that voted for the PVV.⁶ This direction is in line with both contact theory and concentration theory (corroborating Hypothesis 1b), and at odds with the prediction derived from ethnic threat theory (rejecting Hypothesis 1a). The same models show that while unemployment hardly affects PVV voting, the effect of bohemian index is more substantial. The latter hardly surprises as it means that there are fewer votes for the PVV in cities with a more culturally tolerant atmosphere. In itself, that relationship does of course not have any substantive meaning or can even be considered tautological. However, in the following models, we will assess whether the extent of tolerance of an urban cultural atmosphere affects the impact of ethnic segregation on support for the PVV, and that is why the main effect of bohemian index has already been modeled here.

Table 3.

PVV Voting of Dutch Urbanites in 2006 Explained by Ethnic Segregation (Multilevel Regression Analysis; Entries Are Standardized Coefficients; Estimation: Maximum Likelihood).

	Null Model	Model 1	Model 2a	Model 2b (Outlier Amsterdam Excluded)	Model 3	Model 4
	β	β	β	β	β	β
City level
Constant	−.108	−.089	−.062	−.057	−.095	−.074
Share higher educated		−.320**	−.226*	−.244*	−.310***	−.243**
Ethnic segregation		.131	.125	.111	.315***	.300**
Bohemian index		−.145	−.396**	−.231*	−.177*	−.361**
COROP level
Unemployment		−.011	.004	.005	−.059	−.046
Interactions
Ethnic segregation × Bohemian index			.234*	.069		.175
Ethnic segregation × Unemployment					−.263**	−.249**
Variance city level (n = 50)	.266	.160	.164	.173	.110	.112
Variance COROP level (n = 29)	.827	.749	.648	.678	.826	.778
Deviance	126.717	111.167	108.292	109.428	105.806	103.608
ΔDF	4	4	4	4	6

Source: Atlas for municipalities 2004 and 2008, elections database, and Statistics Netherlands (CBS) (own calculations).

Note: PVV = Partij voor de Vrijheid; COROP = Coordination Committee Regional Research Programme; ΔDF = degrees of freedom; CBS = Centraal Bureau voor de Statistiek.

p < .1. **p < .05. ***p < .01. ****p < .001 (two-sided tests).

Table 4.

PVV Voting of Dutch Urbanites in 2010 Explained by Ethnic Segregation (Multilevel Regression Analysis; Entries Are Standardized Coefficients; Estimation: Maximum Likelihood).

	Null Model	Model 1	Model 2a	Model 2b (Outlier Amsterdam Excluded)	Model 3	Model 4
	β	β	β	β	β	β
City level
Constant	−.120	−.100	−.097	−.054	−.125	−.114
Share higher educated		−.396***	−.338***	−.350***	−.415****	−.367***
Ethnic segregation		.150	.100	.106	.292**	.201
Bohemian index		−.374**	−.460***	−.312**	−.411***	−.462***
COROP level
Unemployment		.118	.130	.127	.027	.067
Interactions
Ethnic segregation ×Bohemian index			.149**	.115		.114
Ethnic segregation × Unemployment					−.194*	−.124
Variance city level (n = 50)	.349	.149	.147	.150	.112	.121
Variance COROP level (n = 29)	.623	.470	.404	.427	.571	.483
Deviance	125.826	96.922	93.036	94.645	94.614	92.242
ΔDF	4	4	4	4	6

Source: Atlas for municipalities 2004 and 2008, elections database, and Statistics Netherlands (CBS) (own calculations).

Note: PVV = Partij voor de Vrijheid; COROP = Coordination Committee Regional Research Programme; ΔDF = degrees of freedom; CBS = Centraal Bureau voor de Statistiek.

p < .1. **p < .05. ***p < .01. ****p < .001 (two-sided tests).

In Models 2a (all 50 cities) and 2b (outlier Amsterdam excluded) of Tables 3 and 4, the interaction effects of ethnic segregation with bohemian index have been added. Both resulting coefficients are positive, as was expected on the basis of contact theory (Hypothesis 2b), albeit that the effect proves to be largely driven by outlier Amsterdam in 2006. In 2010, however, the coefficient of the interaction effect of ethnic segregation with bohemian index decreases less strongly if that outlier is left out of the analyses. Notwithstanding these outlier-driven findings, the direction of the interaction effects of ethnic segregation with the bohemian index indicates that a positive effect of ethnic segregation on PVV voting needs to be interpreted according to contact theory instead of concentration theory (corroborating Hypothesis 2b). Following Brambor, Clark, and Golder (2006), we have depicted the interaction effects in 2006 and 2010 (Model 2a) in Figure 3 to illustrate this finding. It shows that the effect of ethnic segregation on PVV voting in cities scoring low on the bohemian index is nonexistent or weak, while it is substantially positive in cities with higher levels of cultural tolerance. In cities in which the native population is more receptive toward immigrants, ethnic segregation more strongly leads to PVV voting, or, put differently, low levels of ethnic segregation steer natives more strongly away from anti-immigrant voting. If, instead, the underlying rationale of concentration theory had accounted for the positive effect of ethnic segregation, it should have been stronger in cities with a less tolerant cultural climate. In those cities, highly visible concentrations of ethnic minorities are after all most likely to spawn anxieties toward a “dangerous” homogeneous out-group.

Figure 3.

The effect of ethnic segregation on votes for the PVV in the Dutch national parliamentary elections of 2006 (straight line) and 2010 (dotted line) by bohemian index.

Although these findings lend empirical support to contact theory instead of concentration theory, the question remains if this also holds when economic instead of cultural urban conditions are taken into account. Therefore, we have modeled the interaction effects of ethnic segregation with unemployment in the third model of both tables. In both 2006 and 2010, this yields negative coefficients, which again corroborates contact theory (Hypothesis 3b) instead of concentration theory. We have depicted these interaction effects in Figure 4, which both show that there is hardly an effect of ethnic segregation on PVV voting in cities with high unemployment levels, while it is positive, and quite strongly so, in cities with lower levels of unemployment.

Figure 4.

The effect of ethnic segregation on votes for the PVV in the Dutch national parliamentary elections of 2006 (straight line) and 2010 (dotted line) by unemployment level.

This demonstrates that ethnic segregation most strongly leads to PVV voting in cities in which the native population’s receptivity toward immigrants is hampered least because unemployment levels are lowest. In other words, low levels of ethnic segregation steer people more strongly away from PVV voting in cities inducing more positive stances toward immigrants due to favorable economic conditions.

All in all, the available evidence supports contact theory instead of ethnic threat theory or concentration theory. The positive effect of ethnic segregation on PVV voting contradicts ethnic threat theory, and the fact that this effect is stronger in cities in which the native population is more receptive toward immigrants due to cultural or economic conditions corroborates contact theory and contradicts concentration theory. To check the robustness of these results, both interaction effects are entered simultaneously into a fourth model for both years. Both models are in line with the results presented above. Although some effects decline somewhat in strength, their directions all remain in accordance with contact theory: The interactions between ethnic segregation and bohemian index are positive, and those between ethnic segregation and unemployment are negative. Again, the available evidence supports contact theory.

Conclusion and Debate

Previous studies have linked support for anti-immigrant parties and other indications of ethnic animosities to ethnic segregation. In this study, we have focused on the locally strongly diverging support for Geert Wilders’s PVV in the Dutch national parliamentary elections of 2006 and 2010 to answer two interrelated questions: “How can the varying effects of ethnic segregation on anti-immigrant voting be understood?” and “How can these effects be theoretically interpreted?” Our analyses on 50 Dutch cities demonstrate that ethnic segregation leads to PVV voting and that this positive effect is stronger in cities with a more tolerant cultural atmosphere and in cities with lower levels of unemployment. First, these findings may help understanding why previous studies on the effect of ethnic segregation on ethnic prejudice and anti-immigrant voting find scattered results. Our results demonstrate that local cultural and economic conditions can strongly moderate this effect, even to the extent that it does not exist in some contexts (in our case: cities with an intolerant cultural atmosphere or little economic opportunities), while it is very strong in other contexts (in our case: cities with a tolerant cultural atmosphere or ample economic opportunities). Second, the fact that we found a positive instead of a negative effect indicates that ethnic threat theory does not explain the relationship between segregation and anti-immigrant voting in Dutch cities. Moreover, our contextualization informed by the cultural and economic conditions of cities has enabled us to empirically disentangle between contact theory and concentration theory. Whereas both theories predict a positive effect, only contact theory is in line with our results. This means that if future studies reporting a positive impact of ethnic segregation on ethnic animosities or anti-immigrant voting incorporate local cultural and economic conditions, they will be able to test whether contact theory or concentration theory applies instead of simply assuming the validity of one of these two theories (cf. Biggs and Knauss 2011).

Having discussed our results and their relevance, the reader should note that by lack of suitable individual-level data on PVV voting we had to rely on aggregate-level data, while voting in the end is an individual-level phenomenon. Data on voting behavior of individuals nested within neighborhoods with different immigrant shares, which are nested in cities that differ in an economic and cultural sense would have been the best way to tackle the research problem at hand. Using such data would also make it possible to measure anti-immigrant voting more validly. For although the PVV is without doubt the Dutch anti-immigrant party par excellence, other sentiments among the electorate, such as political cynicism or general feelings of discontent might also partially account for PVV support (even though research suggests that such sentiments do not underlie the radical right-wing vote but rather result from it, see Van der Brug 2003). In short, if better data become available these should be used to perform stricter tests of our hypotheses. Yet, all our findings point in the same direction: The impact of ethnic segregation on support for the PVV depends on the economic and the cultural conditions of cities, and the way in which can only be interpreted by means of contact theory.

Our findings furthermore show that economic and cultural features of cities can not only be analytically distinguished (cf. De Koster et al. 2008) but also function independently from one another in explaining ethnocentrism (Van der Waal and Houtman 2011), and support for an anti-immigrant party. We would argue this is a valuable insight for the field of urban politics where most attention is usually paid to institutions and economics when it comes to explaining voting behavior and ethnocentrism (cf. Sharp 2007). All the more so as cultural phenomena are rising in salience for electoral behavior in Western countries (Achterberg 2006; Morrill, Knopp, and Brown 2007; Van der Waal and Achterberg 2006; Van der Waal, Achterberg, and Houtman 2007). Yet, one should keep in mind that our study and prior research on the impact of an urban atmosphere of cultural tolerance on the ideological outlook of citizens (Sharp and Joslyn 2008; Van der Waal and Houtman 2011) have been conducted with a newly developed indicator—the bohemian index—that might entail more than what is accounted for here.

One could, for instance, argue that the effect of the bohemian index does not so much stand for contextual effects but for compositional effects. Although this study controlled for compositional effects by including a measure of the share of the urban population that is highly educated, future research might shed more light on this issue, especially if it is conducted with data on individuals nested in cities. Furthermore, the bohemian index might measure something different or something more than an atmosphere of cultural tolerance. One explanation for our findings might be that the consumption patterns of bohemians create more job openings for lower-educated service workers and therefore more receptiveness toward immigrants at the bottom of the urban labor market. If so, the interaction effect of ethnic segregation with the bohemian index would be in line with an economic instead of a cultural contextualization (but note that it would still support contact theory instead of concentration theory). Yet, this seems unlikely considering that (1) we modeled an economic explanation by including the unemployment level, and (2) the share of bohemians in the urban population seems way too small for creating a substantial number of jobs in the service sector (0.07% through 2% or 0.07% through 1.04% without outlier Amsterdam). Further research would be helpful, though, for disentangling the potentially economic aspects of the bohemian index from the cultural phenomenon it is supposed to measure.

All that being said, the findings presented in this article seem promising for understanding the effect of ethnic segregation on ethnic animosities and support for anti-immigrant parties/racist candidates. All the more so if one takes into account that, for three interrelated reasons, the findings for the Dutch case are likely to be more manifest in most other Western countries. First, our findings result from a comparison of cities with relatively small differences, because the Netherlands is one of the smallest Western countries. This suggests that a comparison of cities in vast countries like the United States will result in even stronger findings. All the more so if one, second, takes into account that the level of ethnic segregation in Dutch cities is relatively low, and does not strongly differ between cities (Van der Laan Bouma-Doff 2007). Third, the Dutch welfare state and labor-market policies are quite centralistic, and therefore yield relatively small differences in economic fortunes among Dutch cities (Burgers and Musterd 2001). This makes the suggestion that comparisons of cities in most other Western countries will result in more outspoken findings even more plausible. Whether they do, and if so, whether they can help interpreting the scattered results of studies carried out thus far is an empirical question that calls for further research.

Another empirical question is whether the local cultural and economic conditions discerned here can shed light on two issues that are salient in Dutch debates: PVV support (1) can not only be found in large cities but also in so-called “new towns,” and (2) is higher in the Southern provinces than in the Northern ones (Van Gent and Musterd 2010). Those observations are commonly considered remarkable, as neither new towns nor the Southern provinces have high immigrant shares in their population. It might, however, be fruitful to incorporate insights on economic and cultural contexts when explaining such geographic differences in voting behavior. It might well be that remarkably high levels of support for radical right-wing parties prove part of a more encompassing conservative cultural climate or driven by poor regional economic conditions.

In addition to the relevance of our findings for scholarly debates, they suggest it is important to critically reflect on social mixing policies. In the Netherlands, both the political left and right assume, in accordance with contact theory, that neighborhoods with a mixed ethnic composition support interethnic contacts and consequently breed interethnic understanding (cf. Uitermark and Duyvendak 2008). As a consequence, the Netherlands has “probably the most ambitious and well-funded social mixing policy” (Uitermark 2003, p. 531). The results of this study, however, suggest that such general policies wrongly neglect the importance of local economic and cultural contexts that are important in shaping their outcomes: In cities that have high levels of unemployment or intolerant cultural atmospheres, desegregation is not very likely to result in inter-ethnic understanding.

Footnotes

Acknowledgements

The authors would like to thank the anonymous reviewers and the editors for their helpful and constructive comments and suggestions.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Jeroen van der Waal’s contribution to this research has been enabled by a grant from the Netherlands Organization for Scientific Research (Nederlandse Organisatie voor Wetenschappelijk Onderzoek [NWO]; Grant 07.68.117.00).

Notes

Author Biographies

Jeroen van der Waal (1974), political/urban sociologist, is an assistant professor at Erasmus University Rotterdam and member of the Centre for Rotterdam Cultural Sociology (CROCUS). He studies the impact of economic globalization on social inequality and ethnic relations in cities in the advanced economies. In addition, he is engaged in research on the impact of cultural change on value orientations and voting behavior. He has published in European Journal of Cultural Studies, British Journal of Criminology, International Political Science Review, Journal for the Scientific Study of Religion, Journal of European Social Policy, Politics & Society, Urban Affairs Review, and Urban Studies.

Willem de Koster (1984), cultural sociologist, is an assistant professor at Erasmus University Rotterdam and member of the CROCUS. His primary research interests are twofold: He studies online communities as well as the genesis and manifestations of contemporary cultural conflict in Western countries. He has published in journals such as European Journal of Cultural Studies, British Journal of Criminology, Information, Communication and Society, International Political Science Review, Journal of European Social Policy, Journal for the Scientific Study of Religion, and Symbolic Interaction. His book Paradoxes of Individualization has been published with Ashgate (with Houtman and Aupers).

Peter Achterberg (1977) is an associate professor at Erasmus University Rotterdam and member of the CROCUS. His research mainly focuses on the effects of processes of religious, political, and cultural change on public opinion. His recent research has been published in, among others, Public Understanding of Science, International Political Science Review, and Public Opinion Quarterly.

References

Achterberg

2006. Class voting in the new political culture—Economic, cultural and environmental voting in 20 Western countries. International Sociology 21 (2): 237-61.

Allport

G. W.

1954/1979. The nature of prejudice. New York: Doubleday.

Biggs

Knauss

2011. Explaining membership in the British National Party: A multilevel analysis of contact and threat. European Sociological Review 28 (5): 633-46.

Blalock

H. M.

1967. Towards a theory of minority group relations. New York: Wiley.

Brambor

Clark

W. R.

Golder

2006. Understanding interaction models: Improving empirical analyses. Political Analysis 14 (1): 63-82.

Branton

R. P.

Jones

B. S.

2005. Examining racial attitudes: The conditional relationship between diversity and the socio-economic environment. American Journal of Political Science 49 (2): 359-72.

Brown

Knopp

Morrill

2005. The culture wars and urban electoral politics: Sexuality, race, and class in Tacoma, Washington. Political Geography 24 (3): 267-91.

Burgers

Musterd

2001. Global and local determinants of social exclusion: Amsterdam versus Rotterdam. In Governing European cities: Social fragmentation, social exclusion and urban governance, edited by Andersen

H. T.

van Kempen

, 125-49. Aldershot: Ashgate.

Clark

T. N.

Harvey

2010. Urban politics. In Handbook of politics. State and society in global perspective, edited by Leicht

K. T.

Jenkins

J. C.

, 423-41. New York: Springer.

10.

Clark

T. N.

Rempel

1997. Citizen politics in post-industrial societies. Boulder: Westview Press.

11.

Dancygier

R. M.

2010. Immigration and conflict in Europe. Cambridge: Cambridge Univ. Press.

12.

De Koster

Achterberg

van der Waal

Forthcoming. The new right and the welfare state: On the electoral relevance of welfare chauvinism and welfare populism in the Netherlands. International Political Science Review 34 (1): 3-20.

13.

De Koster

Van der Waal

Achterberg

Houtman

2008. The rise of the Penal State: Neo-liberalization or new political culture? British Journal of Criminology 48 (6): 720-34.

14.

De Lange

Art

2011. Fortuyn versus wilders: An agency-based approach to radical right party building. West European Politics 34 (6): 1229-49.

15.

Deleon

R. E.

Naff

K. C.

2004. Identity politics and local political culture: Some comparative results from the social capital benchmark survey. Urban Affairs Review 39 (3): 689-719.

16.

Duncan

O. D.

Duncan

1955. A methodological analysis of segregation indices. American Sociological Review 20 (2): 210-217.

17.

Elazar

1966. American federalism: A view from the states. New York: Thomas Y. Crowell.

18.

Esser

1986. Social context and inter-ethnic relations: The case of migrant workers in West German Urban areas. European Sociological Review 2 (1): 30-51.

19.

Esses

V. M.

Jackson

L. M.

Bennett-AbuAyyash

2011. Intergroup competition. In The Sage handbook of prejudice, stereotyping and discrimination, edited by Dovidio

J. F.

Hewstone

Glick

Esses

, 225-40. London: SAGE.

20.

European Commission Against Racism and Intolerance. 2008. Third report on the Netherlands. Strasbourg, France: Council of Europe.

21.

Florida

2002. Bohemia and economic geography. Journal of Economic Geography 2 (1): 55-71.

22.

Florida

2004. The rise of the creative class, and how it’s transforming work, leisure, community and everyday life. Paperback ed. New York: Basic Books.

23.

Florida

2005. Cities and the creative class. New York: Routledge.

24.

Giles

M. W.

Buckner

M. A.

1993. David Duke and black threat: An old hypothesis revisited. Journal of Politics 55 (3): 702-13.

25.

Hawley

A. A.

1944. Dispersion versus segregation: Apropos of a solution of race problems. Papers of the Michigan Academy of Science, Arts, and Letters 30: 667-74

26.

Hjerm

2009. Anti-immigrant attitudes and cross-municipal variation in the proportion of immigrants. Acta Sociologica 52 (1): 47-62.

27.

Ignazi

2003. Extreme right parties in Western Europe. Oxford: Oxford Univ. Press.

28.

Jiobu

R. M.

Marshall

H. H.

1971. Urban structure and the differentiation between blacks and whites. American Sociological Review 36 (4): 638-49.

29.

Kent

S. L.

Jacobs

2005. Minority threat and police strength from 1980 to 2000: A fixed-effects analysis of nonlinear and interactive effects in large U.S. Cities. Criminology 43 (3): 731-60.

30.

Key

V. O.

1949/1984. Southern politics in state and nation. Knoxville: The Univ. of Tennessee Press.

31.

Kinder

D. R.

Mendelberg

1995. Cracks in American Apartheid: The political impact of prejudice among desegregated whites. Journal of Politics 57 (2): 402-24.

32.

Kuklinski

J. H.

Cobb

M. D.

Gilens

1997. Racial attitudes and the “New South.” Journal of Politics 59 (2): 323-49.

33.

Kunovich

R. M.

Hodson

2002. Ethnic diversity, segregation, and inequality: A structural model of ethnic prejudice in Bosnia and Croatia. The Sociological Quarterly 43 (2): 185-212.

34.

Lieberson

1961. The impact of residential segregation on ethnic assimilation. Social Forces 40 (1): 52-57.

35.

Lieberson

1980. A piece of the pie. Blacks and white immigrants since 1880. Berkeley: Univ. of California Press.

36.

Lieske

1993. Regional subcultures of the United States. Journal of Politics 55 (4): 888-913.

37.

Massey

D. S.

Denton

N. A.

1988. The dimensions of residential segregation. Social Forces 67 (2): 281-315.

38.

Morrill

Knopp

Brown

2007. Anomalies in red and blue: Exceptionalism in American electoral geography. Political Geography 26 (5): 525-53.

39.

Oesch

2008. Explaining workers’ support for right-wing populist parties in Western Europe: Evidence from Austria, Belgium, France, Norway, and Switzerland. International Political Science Review 29 (3): 349-73.

40.

Oliver

J. E.

Wong

2003. Intergroup prejudice in multiethnic settings. American Journal of Political Science 47 (4): 567-82.

41.

Olzak

1992. The dynamics of ethnic competition and conflict. Stanford: Stanford Univ. Press.

42.

Otjes

2011. The Fortuyn effect revisited: How did the LPF affect the Dutch parliamentary system? Acta Politica 46 (4): 400-24.

43.

Pellikaan

De Lange

S. L.

Van der Meer

2007. Fortuyn’s legacy: Party system change in the Netherlands. Comparative European Politics 5 (3): 282-302.

44.

Penninx

2006. Dutch immigrant integration policies before and after the Van Gogh Murder. Journal of International Migration and Integration 7 (2): 241-54.

45.

Pettigrew

T. F.

1998. Intergroup contact theory. Annual Review of Psychology 49:65-85.

46.

Pettigrew

T. F.

Wagner

Christ

2010. Population ratios and prejudice: Modelling both contact and threat effects. Journal of Ethnic and Migration Studies 36 (4): 635-50.

47.

Quillian

1996. Group threat and regional change in attitudes toward African Americans. American Journal of Sociology 102 (3): 816-60.

48.

Rink

Phalet

Swyngedouw

2009. The effects of immigrant population size, unemployment, and individual characteristics on voting for the Vlaams Blok in Flanders 1991-1999. European Sociological Review 25 (4): 411-24.

49.

Robinson

1980. Lieberson’s Isolation Index: A case study evaluation. Area 12 (4): 307-12.

50.

Rocha

R. R.

Espino

2009. Racial threat, residential segregation, and the policy attitudes of Anglos. Political Research Quarterly 62 (2): 415-26.

51.

Roscigno

V. J.

Garcia

L. M.

Bobbitt-Zeher

2007. Social closure and processes of race/sex employment discrimination. The ANNALS of the American Academy of Political and Social Science 609 (16): 16-48.

52.

Rydgren

ed 2005. Movements of exclusion: Radical right-wing populism in the Western World. Hauppauge, NY: Nova Science Publishers.

53.

Rydgren

2008. Immigration sceptics, xenophobes or racists? Radical right-wing voting in six West European countries. European Journal of Political Research 47 (6): 737-65.

54.

Sharp

E. B.

1996. Culture wars and local politics. Local government’s role in social conflict. Urban Affairs Review 31 (6): 738-58.

55.

Sharp

E. B.

2002. Culture, institutions, and urban officials’ responses to morality issues. Political Research Quarterly 55 (4): 861-83.

56.

Sharp

E. B.

2007. Revitalizing urban research: Can cultural explanation bring us back from the periphery? Urban Affairs Review 43 (1): 55-75.

57.

Sharp

E. B.

Joslyn

M. R.

2008. Culture, segregation, and tolerance in urban America. Social Science Quarterly 89 (3): 573-91.

58.

Stein

Post

S. S.

Rinden

A. L.

2000. Reconciling context and contact effects on racial attitudes. Political Research Quarterly 53 (2): 285-303.

59.

Stolzenberg

D’Alessio

S. J.

Eitle

2004. A multilevel test of racial threat theory. Criminology 42 (3): 673-98.

60.

Taylor

R. L.

1979. Black ethnicity and the persistence of ethnogenesis. American Journal of Sociology 84 (6): 1401-23.

61.

Uitermark

2003. “Social mixing” and the management of disadvantaged neighbourhoods: The Dutch policy of urban restructuring revisited. Urban Studies 40 (3): 531-49.

62.

Uitermark

Duyvendak

J. W.

2008. Civilising the city: Populism and revanchist urbanism in Rotterdam. Urban Studies 45 (7): 1485-503.

63.

Van der Brug

2003. How the LPF fuelled discontent: Empirical tests of explanations of LPF support. Acta Politica 38 (1): 89-106.

64.

Van der Brug

Fennema

2007. What causes people to vote for a radical-right party? A review of recent work. International Journal of Public Opinion Research 19 (4): 474-87.

65.

Van der Brug

Fennema

Tillie

2000. Anti-immigrant parties in Europe: Ideological or protest vote? European Journal of Political Research 37 (1): 77-102.

66.

Van der Brug

Fennema

Tillie

2005. Why some anti-immigrant parties fail and others succeed: A two-step model of aggregate electoral support. Comparative Political Studies 38 (5): 537-73.

67.

Van der Waal

Achterberg

2006. Stille revolutie, contra-revolutie of cultureel conflict? Veranderingen in de politieke cultuur en hun Invloed op het klassengebonden stemgedrag [Silent revolution, contra-revolution or cultural conflict? Political cultural change and its influence on class voting in Western countries (1956-1990)]. Res Publica 46 (4): 369-92.

68.

Van der Waal

Achterberg

Houtman

2007. Class is not dead—It has been buried alive: Class voting and cultural voting in postwar Western societies (1956-1990). Politics & Society 35 (3): 403-26.

69.

Van der Waal

de Koster

Achterberg

2011. Stedelijke context en steun voor de PVV: Interetnische nabijheid, economische kansen en cultureel klimaat in 50 Nederlandse steden [Urban context and support for the PVV: Inter-ethnic propinquity, economic opportunities and cultural atmosphere in 50 Dutch cities]. Res Publica 53 (2): 183-201.

70.

Van der Waal

Houtman

2011. Tolerance in the post-industrial city. Assessing the ethnocentrism of less-educated natives in 22 Dutch Cities. Urban Affairs Review 47 (5): 642-71.

71.

Van der Laan Bouma-Doff

2007. Confined contact: Residential segregation and ethnic bridges in the Netherlands. Urban Studies 44 (5/6): 997-1017.

72.

Van Gent

Musterd

2010. Isolement en angst: PVV in Haagse buurten bij de gemeenteraadsverkiezingen van 2010 [Isolation and Fear: PVV in The Hague’s Neighborhoods during the Municipal Elections in 2010]. Beleid en Maatschappij 37 (2): 140-53.

73.

Van Kersbergen

Krouwel

2008. A double-edged sword! The Dutch centre-right and the “foreigners issue.” Journal of European Public Policy 15 (3): 398-414.

74.

Weber

. 1922/2006. Wirtschaft und Gesellschaft [Economy and society]. Paderborn, Germany: Voltmedia.

75.

Wright

G. C.

1977. Contextual models of electoral behavior: The Southern Wallace vote. American Political Science Review 71 (2): 497-508.