Scholarly article on topic '"White Flight" or Positive Contact? Local Diversity and Attitudes to Immigration in Britain'

"White Flight" or Positive Contact? Local Diversity and Attitudes to Immigration in Britain Academic research paper on "Political Science"

0
0
Share paper
Academic journal
Comparative Political Studies
OECD Field of science
Keywords
{""}

Academic research paper on topic ""White Flight" or Positive Contact? Local Diversity and Attitudes to Immigration in Britain"

Article

"White Flight" or Positive Contact? Local Diversity and Attitudes to Immigration in Britain

Comparative Political Studies 2015, Vol. 48(12) 1563-1590 © The Author(s) 201 5

Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/0010414015581684 cps.sagepub.com

Eric Kaufmann1 and Gareth Harris2

Abstract

Does the local presence of immigrant groups increase White hostility to immigration? Most research finds that diverse neighborhoods reduce White opposition to minorities and immigration. However, most studies at higher geographies find the reverse effect. We confirm this pattern for England and Wales for 2009-2012. Yet, contextual studies are open to selection bias, which is where this article makes its main contribution. Is White tolerance in diverse neighborhoods the result of a positive effect of inter-ethnic contact, or does it arise from White flight, with anti-immigrant Whites exiting diverse areas but remaining within wider geographies as radicalized opponents of immigration? We provide the first attempt we are aware of to track the opinions of in- and out-migrants, as well as stayers, from local areas over an extended period. We use 20 years of large-scale geocoded British longitudinal data and find only limited evidence of selection effects associated with White flight.

Keywords

elections, public opinion, and voting behavior, European politics, migration, race, ethnicity and politics, contextual effects, White flight

'Birkbeck College, University of London, UK

2Coventry University, UK

Corresponding Author:

Eric Kaufmann, Department of Politics, Birkbeck College, University of London, London, UK. Email: e.kaufmann@bbk.ac.uk

How does local racial context affect attitudes to immigration? Do diverse locales reduce hostility as local Whites gain firsthand experience of minorities and immigrants? Or have Whites in diverse locales self-selected themselves for toleration, with less tolerant individuals leaving the area. In a departure from previous literature on contextual effects in public opinion and voting, we use longitudinal data to test for selection effects associated with "White flight" so as to more clearly evaluate the contending claims of contact and threat theory for the case of England and Wales.

Theoretical Perspectives

Contact Theory

The debate between contact and threat theory is well established and sets the overarching framework for our analysis. Allport (1954) advances the argument that positive intergroup contacts improve groups' perceptions of each other, reducing prejudice and social distance. Since then, considerable experimental evidence has confirmed this view (e.g., Barlow, Louis, & Hewstone, 2009; Hewstone et al., 2005). In a meta-analysis of 713 studies, Pettigrew and Tropp (2006) found that contact tended in most cases to significantly reduce intergroup prejudice. The effect of contact on prejudice reduction operates via improved knowledge, greater empathy, and especially, a reduction in intergroup anxiety (Barlow et al., 2012; Pettigrew & Tropp, 2006).1

Threat Theory

Against the contact hypothesis, the threat hypothesis claims the presence of minorities activates a sense of anxiety among members of the ethnic majority. Diversity, far from incubating toleration, results in heightened conflict (Putnam, 2007). An extensive tradition of research on the racial threat hypothesis, beginning with the landmark studies of Key (1949) and Blalock (1957) revealed higher levels of White animus toward African Americans in the relatively diverse South, especially in counties with higher proportions of African Americans. Subsequent work generalized these findings to the Midwest and Northeast (Stein, Post, & Rinden, 2000).

Variation by Geographic Level

Since the work of Allport and Key, numerous studies (many summarized in Table 1) have been undertaken at various geographic scales, examining a series of dependent variables including attitudes to racial minorities,

Table 1. Meta-Analysis of Ethnic Context Literature.

Context Study

Prison cells, social Berg, 2009; Trulson & Marquart, 2002 -

networks

Block groups, Baybeck, 2006; Hjerm, 2009 +/-

municipalities Rink et al., 2009 Biggs & Knauss, 20'2; Lubbers et al., 2006; Rydgren & Ruth, 20'' -

Perceived neighborhood Kitschelt, '996; Mayer, '995; Robinson, '980; Sigelman & Welch, '993; Surace & Seeman, '967; Yancey, '999

Quillian, '995 +

Housing projects Ford, '973 -

U.S. census tracts Hajnal et al., 20''; Wilson, '979 Ha, 20'0; Oliver & Wong, 2003 +

British MSOAs Laurence, 20'3 -

Tract and locale Welch et al., 200' -

British wards Bowyer, 2008; Harris, 20'2 -

Zip codes Emerson et al., 200'; Gilliam et al., 2002 -

Precincts/boroughs Carsey, '995 -

British local authorities Biggs & Knauss, 20'2; Bowyer, 2008; Ford & Goodwin, 20'0; Harris, 20'2 +

French departements Kitschelt, '995; Mayer, '995 +

U.S. counties Blalock, '957; Branton & Jones, 2005; Eitle et al., 2002; Giles & Evans, '986; Glaser, '994; Hopkins, 20''; Key, '949; Reed, '972; Rocha et al., 20''; Stein, Post, & Rinden, 2000; Wright, '976 Hood & Morris, '997, 2000; Morris, 2000 +

Corzine et al., '983; Dixon, 2006; Newman +/-

& Johnson, 20'2

Campbell, Wong, & Citrin, 2006 =

Louisiana Parishes Giles & Buckner, '993; Giles & Hertz, '994; Voss, '996 +

Metropolitan areas Arp et al., '999; Dixon, 2006; Fossett & Kiecolt, '989; Jacobs & Carmichael, 200', 2002; Rocha & Espino, 2009; Taylor, '998 +

Ha, 20'0 +/-

Countries Lubbers & Scheepers, 200', 2002; Quillian, '995; Schneider, 2008; Semyonov et al., 2006; Semyonov & Glikman, 2009 Citrin & Sides, 2008 +

"+" denotes a positive relationship between minority/immigrant share and hostility to outgroups/ immigration (i.e., threat response); "-" denotes a negative (contact response); and "=" denotes a neutral relationship. MSOAs = middle layer superoutput areas.

multiculturalism policies, immigrants, and immigration. Some are aggregate analyses while others take the form of surveys that include contextual predictors. Most derive contextual data from the census, although several ask individuals to answer with reference to their perceived "neighbourhood." An increasing number of analyses are multi-level in nature. The methodology varies: Sometimes there are controls for socioeconomic and political characteristics of neighborhoods; certain works contain a richer variety of attitudi-nal items than others; some focus on attitudes to African Americans, others on Hispanics or Muslims, still others on immigrants or support for anti-immigrant parties.

From this research, a pattern—though far from uniform—emerges. A summary, by no means complete, of work in this area, appears in Table 1, which updates a similar exercise undertaken by Cho and Baer (2011).2 We located 24 studies at ward or tract level (population around 7,500) or below, and found that about three quarters link local diversity to reduced animosity toward minorities, immigrants, and immigration, whereas just 25% find that more diversity in the locale increases hostility. Work at larger geographic scales reveals a different relationship. Suddenly, the association between diversity and toleration reverses itself. At county level or above, for instance, diversity is associated with heightened White threat perceptions of minorities or immigrants in 84% of a sample of 44 papers (by no means exhaustive), which deploy contexts containing approximately 100,000 people or more.3

Contact and threat theorists differ in their explanation of the aforementioned pattern. Because Whites' opportunity for contact with minorities is greater if they encounter diversity in their local area, it should follow that diverse contexts up to the level of the neighborhood or ward will reduce opposition to minorities, immigrants, and immigration. At larger levels, opportunities for contact are reduced so threat effects may prevail. As Schlueter and Scheepers (2010) remark,

We suggest that, for social contexts smaller than those used in the present study [Dutch municipalities], the primary impact of minority group size will be an enhancement of opportunities for intergroup contact ... In contrast, for relatively large contexts . . . we consider outgroup size to be . . . associated with an enhancement of threat perceptions. (p. 293)

The principal reason threat theorists expect to find lower White threat responses to diversity at neighborhood level than at higher geographies is that threatened individuals can readily leave a locale but find it harder to vacate an entire metropolitan area. In this sense, they are acting to maintain ethnic boundaries in the face of immigration, realizing Michael Walzer's (1983)

dictum that "Neighborhoods can be open only if countries are at least potentially closed . . . The distinctiveness of cultures and groups depends upon closure and without it cannot be conceived as a stable feature of human life" (p. 39).

Might White flight account for the geography-specific empirical patterns noted by researchers? This is intimated by the results of the Multi-City Study of Urban Inequality (MCSUI), which uses neighborhood racial composition showcards to tap underlying residential proclivities. It finds White respondents who prefer more homogeneous neighborhoods in the four cities sampled—Detroit, Los Angeles, Boston, and Atlanta—to be significantly more likely to hold negative opinions of minority groups (Krysan, 2002). A replication of this study in the Netherlands renders a similar verdict, adding that Whites who are more hostile to minorities are significantly more likely to say they would leave an area with a large proportion of minorities (van Londen, 2012). At the subjective level, therefore, the link between White attitudes to minorities and White flight appears robust and generalizable across several Western nations. Which endorses threat theorists' claims that self-selection, that is, White flight, accounts for divergent findings between small and large-scale contexts.

British Research

No existing studies of British attitudes to immigration simultaneously utilize small and large-unit ethnic contextual variables. However, there is mounting evidence for a disjuncture between local and metro-level contexts in research on voting for British anti-immigration parties. The importance of local ethnic contexts as predictors of support for the anti-immigrant British National Party (BNP) has been demonstrated in several studies (Bowyer, 2008; Ford & Goodwin, 2010; Harris, 2012; John, Margetts, Rowland, & Weir, 2006). Multi-level analyses of the BNP demonstrate that support is positively correlated with diversity at the Local Authority (LA) level, a unit typically consisting of between 100,000 and 200,000 people.

By contrast, at ward level, with populations approximately a tenth as large as in LAs, diversity appears to exert a neutral or negative effect on BNP support. In neighborhoods above a threshold of 25% minority population, researchers find reduced White support for the BNP, whereas homogeneous wards and neighborhoods with few immigrants are more likely to vote for this anti-immigrant party (Biggs & Knauss, 2012; Bowyer, 2008).4 Others point to the combination of homogeneity-within-diversity: Heavily White British wards nested in diverse and changing LAs appear to offer fertile ground for the far right (Harris, 2012). Although only a small minority of White British

Table 2. Multi-Level Logistic Regression on Desire to Reduce Immigration (White U.K.-Born Only).

Income -.04'*** (.009)

No qualifications .376*** (.077)

Middle class (ref: upper class) .4'3*** (.067)

Lower supervisory class (ref: upper class) .70'*** (.096)

Working class (ref: upper class) .450*** (.07)

Private renter (ref: home owner) -.'47 (.076)

Age .003 (.002)

London (ref: all other regions) -.'97 (.'48)

Resident over '0 years (ref: less than '0 years) .''9* (.052)

Couple (ref: single) .'54** (.05)

Ward: minority share of population -.017*** (.003)

Ward: Carstairs Index of Multiple Deprivation .049*** (.0'')

Ward: population density -.007*** (.00')

LA: minority share of population .011** (.004)

Random effects

Variance intercept ward .'48*** (.037)

Variance intercept LA .049** (.02')

N '3,6'6

Table entries are logged odds with robust standard errors, clustered by ward (N = 1,709) and LA (N = 364). Pseudo R2 not available in MLwiN. Bold results refer to the effect of diverse contexts at different levels. LA = Local Authority. *p < .05. **p < .01. ***p < .001.

people who desire less immigration support the BNP, the salience of the immigration issue for BNP supporters suggests the contextual drivers of BNP support may hold for the wider swathe of public opinion as well.

It is not our intention to replicate such analyses, but it is worth noting that the same pattern holds for immigration opinion in England and Wales. For example, if we examine the combined 2009-2010 and 2010-2011 waves of the Citizenship Surveys (Office for National Statistics and Home Office, 2010, 2011), which asks around 16,000 White U.K.-born respondents in England and Wales about whether the level of immigration should be reduced or increased, we find precisely this relationship. Table 2 features a multi-level model of immigration opinion. This shows that, with controls for major individual and contextual-level variables, the share of ethnic minorities at ward level (population averaging 7,700) is associated with less White U.K.-born opposition to immigration while the share of minorities at LA (population 100,000-200,000) predicts heightened opposition.5

Does this pattern confirm contact theorists' claim that only lower geographic scales offer opportunities for inter-ethnic contact to reduce White threat perceptions, or does it furnish evidence that White flight has left a disproportionately tolerant, skewed sample of Whites in diverse wards?

Self-Selection and White Flight Threat Theory

Migration theory tells us that most moves take place over small distances (Crowder & South, 2008) for reasons of cost and information; thus, those who exit diverse neighborhoods typically take up residence within the wider metropolitan area, county, or, in Britain, LA. This, therefore, would account for a pattern of higher-level opposition and ward-level tolerance. Previous researchers have done the best they can with available data to guard against endogeneity. Where researchers have addressed self-selection, they have typically regressed place of residence on individual values such as conservatism or racial animus (e.g., Branton & Jones, 2005; Gay, 2006; Ha, 2010). Such work typically claims that Whites in diverse areas differ from Whites in homogeneous areas on the racial attitudes represented by the dependent variable but not on racial attitudes that are used as instruments for it.

The self-selection tests are, as some authors admit, less than ideal. Such tests use a restricted number of parameters and depend upon imperfect instruments for the dependent variable. Finally, the reliance on population stock data precludes any consideration of the magnitude and character of White population flows between geographic units that shape the composition of stocks—especially in diverse contexts where population turnover is often high. The ideal test for selective White flight would involve longitudinal data on attitudes. Krysan (2002) asks,

What are the mechanisms through which racial context operates on thoughts of mobility? Although the optimal approach for answering questions about motivations would be to use longitudinal data that measure both attitudes and behavior at the individual level, these data do not exist.

In a political science context, Abrajano and Hajnal (2009) remark that

most studies of contextual effects have been plagued by concerns about selection . . . Existing studies often try to control for various aspects of this selection but in the end few have been able to solve this fundamental problem.

Some scholars uncover significant self-selection effects. Oliver and Wong (2003), for example, account for endogeneity in a model of out-group stereotypes

by controlling for preferences in neighborhood ethnic composition. The parameter for White neighborhood preferences significantly predicted White attitudes, a similar association to that found in the MCSUI (Krysan, 2002). Although this did not eliminate the positive effect of local diversity on White attitudes toward outgroups, self-selection effects were considerably larger than contact effects.

Our longitudinal data take us a step further. They permit us to address, for what we believe to be the first time, the methodological difficulties of cross-sectional studies of attitudes toward outgroups, so as to better gauge the extent of self-selection over a 20-year period. Whites who remain in diverse neighborhoods are predicted to be more tolerant of minorities and immigration than those who depart. Thus, if White flight theory is accurate:

Hypothesis 1 (H1): White flight: White British residents who are more hostile to immigration will be more likely to leave diverse wards than White British people who are less opposed to immigration. Hypothesis 2 (H2): White avoidance: White British residents who are more hostile to immigration will tend to select less diverse wards to move to than White British people who are less opposed to immigration

Threat theory depends not only on a White flight effect but also on the volume of self-selection. Only significant replacement of anti-immigration Whites with pro-immigration Whites can account for the common finding that Whites in diverse neighborhoods, tracts, or wards are more tolerant of immigration than those in more homogeneous areas. Yet we know from a 1% sample of the census (ONS LS, 2001) that 6% to 7% of White British people leave diverse wards of England and Wales each year, that is, between 900 and 1,050 individuals in a large urban ward containing 15,000 Whites. Given this figure, it would take 14 to 16 years for the attitudinal composition of the White population stock of a diverse ward to match the White inflow.6 The volume of White population turnover in diverse neighborhoods in England and Wales is thus large enough for selective White migration to account for the relatively pro-immigration attitudes of White British people in wards with a large minority presence. However, should we find that Whites departing an area of minority concentration manifest similar or more liberal views on immigration compared with Whites who remain or enter, the White flight threat hypothesis fails and contact theory may offer the best explanation.

We use the longitudinal British Household Panel Survey (BHPS) to test for self-selection through White flight. The BHPS is an annual longitudinal study

of some 5,500 households containing 10,300 individuals in England and Wales, which began in 1991. Attrition of cases reached 11% in the transition from Waves 1 to 2, but since then, recontact rates have remained high, generally well above 95%. In 2009, the survey merged into the Understanding Society longitudinal survey (the U.K. Household Longitudinal Study [UKHLS], 2012), but the UKHLS contains an annual sample of approximately 40,000, much larger than the 10,000 in the BHPS. We use linearly interpolated 1991, 2001, and 2011 ward-level census data from the Office for National Statistics (ONS; 2013), which is attached to individual survey records in the Citizenship Survey and BHPS-UKHLS. We use a common 2001 ward geography to link census data across the three waves.7

Although similar to the Panel Study of Income Dynamics (PSID) in the United States, the BHPS contains modules covering a wider array of subjective measures, notably, modules on party support, political participation, and national identity. This permits a fuller examination of the cultural and political subjectivity of Whites who leave, enter, and remain in diverse areas, enabling us to generate a 20-year profile of White incomers to, out-migrants from, and stayers in, diverse wards. Our BHPS-UKHLS sample consists of 172,200 White U.K.-born person-years of data for 1991-2012 across 19 survey waves. Wave size varies between 6,684 and 10,218 for the 18 waves of the BHPS, with 25,277 for the 19th wave (the 1st wave of the UKHLS), which contains a subsample of 7,000 individuals linked to the BHPS.8

Method

To test for self-selection, we first test for H1, the "flight" of conservative Whites. Namely, do conservative Whites tend to disproportionately outmi-grate from diverse areas? Second, we examine H2, White "avoidance" of diverse destinations, by examining whether conservative Whites tend to move to Whiter areas than do liberal Whites. This is measured by the difference in the share of ethnic minorities between respondents' origin and destination wards when they move. The combined BHPS-UKHLS for 1991-2012 yields a total of 172,200 person-years of data on White U.K.-born British people, as shown in Table 3.

Of these, 156,779 (91% of the sample) did not move in the previous year. Even among movers, a further 2,953 individuals moved within ward in the previous year. This means only 7.2% of the sample moved between wards in the previous year. Note that this figure is in person-years; therefore, the share of individuals in the survey who moved at least once would be considerably higher, especially among individuals who survived the 19 waves of the survey (thereby leaving a 19-person-year footprint each in the data).

Table 3. Aggregate White Population Flows From BHPS-Understanding Society, 1991-2012 (Person-Years).

Change in diversity (relative)

White mover status Same Less More Total Share (%)

White stayer 153,567 1,534 1,678 156,779 91.0

White inter-ward mover 8,542 2,095 1,831 12,468 7.2

White intra-ward mover 2,879 33 41 2,953 1.7

Total 164,988 3,662 3,550 172,200 10.0

Source. BHPS (1991-2009); U.K. Household Longitudinal Study (UKHLS; 2012).

BHPS = British Household Panel Survey. Change in diversity is relative, adjusted for average

annual ethnic shift in dataset, based on change in diversity quintile.

Our BHPS-UKHLS figure compares well with census figures. The 2001 census finds that 40,614 individuals in a 1% sample of the census (526,458 persons) moved into their ward during the year 2000 to 2001 from another ward (ONS LS, 2001). This represents 7.7% of the sample, analogous to our 7.2% annual movers. Notice that this data set—arguably the second longest running longitudinal survey in the world—underpins our contribution to knowledge: Without a longitudinal structure to the data, we could not determine moves to and from diversity. In the absence of a large enough sample, we could not amass sufficient cases for meaningful analysis of White movers in diverse wards.

Dependent Variable

The dependent variable in the first set of models is the change in the percentage of ethnic minorities experienced by an individual when they move, that is, the difference in minority share between their ward of origin and ward of destination. This theoretically spans the range from -100 to +100, and ranges from +82 to -75 in our data. As the sample is restricted to movers, this taps ethnic changes brought about by destination choice rather than in situ ethnic change. The dependent variable in the second set of models is a dummy variable coded 1 for an inter-ward move in a given year and 0 for a non-move or intra-ward move.

Independent Variables

The BHPS-UKHLS permits us to observe the demographic, socioeconomic, and attitudinal characteristics of movers. In an ideal world, the data set would ask a question on attitudes to immigration. The longitudinal surveys do not,

but record a series of items that are well-known proxies for immigration attitudes—some in each wave, some only occasionally. These include modules on voting behavior and national identity, which have been found to directly predict immigration attitudes in the literature (Fetzer, 2000; Wright, Citrin, & Wand, 2012). We focus on Conservative party voting versus left-wing alternatives, and English national identity versus British or other, as proxy indicators for greater opposition to immigration. These choices are justified on the basis of cross-tabulations in Table 4. Moreover, English identity is a significant predictor of opposition to immigration in a version of the model presented in Table 2, even in the presence of numerous controls.9 Other attitudinal items in the BHPS were asked only in occasional years, and, given our interest is in movers to and from diverse wards, a small subsample, these variables cannot be used for panel analysis. However, we test these in cross-sectional models of mobility.10

Questions on party vote and support were asked across all waves, permitting us to conduct longitudinal data analysis using a dichotomous variable taking the value of 1 for Conservative party support and 0 for those backing the left-wing Liberals or Labour. In all, 36.6% of person-years in the sample supported the Conservatives, while 63.4% backed Labour or the Liberal Democrats. Another important proxy question for immigration attitudes, which is asked in all waves, is, "Please say which, if any, of the words on this card describes the way you think of yourself? Please choose as many or as few as apply." Choices included English, British, European, Irish, Northern Irish, Scottish, Welsh, Other, None. Those who selected "English" only were assigned a 1, all others a 0. Across all waves, 8.87% of U.K.-born Whites opted to describe themselves as English only, which represents 39.2% of those who answered the national identity question.11 English national identity is associated, in many surveys, with significantly greater opposition to immigration and, in Understanding Society, with elevated support for anti-immigration parties, even in the presence of a wide range of individual and contextual controls.12

This said, as Table 4 reveals, the immigration attitude difference—across a range of distinct questions—between White Conservative and Left (Labour, Liberal Democrat, Green) party voters in the British Election Study (BES) was just 13 points in 1997 when Tony Blair took office. It stood at 12 points in the 2010 British Social Attitudes Survey (BSA), although this widens to over 30 points for recalled 2005 vote in the 2015 BES. Among English identifiers, opposition to immigration was only 2 points higher than the White average in the 1997 BES and 3 points higher in the 2010-2011 Citizenship Survey and 2010 BSA, although nearly 20 points more elevated among those identifying as "strongly English" in the 2015 BES.

Table 4. Proxy Measures of White British Immigration Opinion.

"Black and Asian immigration good for GB" (BES 1997-2001, 1997 recalled vote) N Immigration enriches or undermines cultural life (BES 20IS, Wave 2, 2005 recalled vote) N Immigration should be increased or stay the same versus be reduced (BSA 2010) N

University-educated Left 77.2% 237 67.9% 4,971 50.0% 1 19

English identity 32.6%a 883 21.5%b 8,1 15 I2.83%a 204

Non-university educated 24.9% 317 16.1% 3,190 7.68% 43

Conservative voter

Working-class 19.8% 121 14.7% 1,758 6.70% 12

Conservative voter

Conservative voter 29.8% 446 21.6% 5,596 10.37% 75

Left party voter 43.1% 1,016 52.1% 9,787 22.97% 221

Middle-class Left voter 67.4% 270 67.5% 3,180 29.42% 163

UK.IP and BNP supporter n.a. 7.6% 563 n.a.

White British average 34.8% 1,947 40.9% 22,597 15.73% 41 1

Source. BES 1997-2001: Heath, Jowell, and Curtice (2002); BES 2015: Fieldhouse, Green, Evans, Schmitt, and van der Eijk (2014); BSA 2010: National Centre for Social Research (2012).

BES = British Election Study; BSA = British Social Attitudes; BNP = British National Party.

a. English identity or no English identity.

b. "Strongly English" on 7-point Englishness scale.

In view of the sometimes modest association between univariate proxies (Conservative voting, Englishness) and anti-immigration attitudes, we also include variables derived from the cross-tabulation of university education and voting, or class and voting. These offer stronger proxies for immigration opinion. Thus, university-educated left voters were more than twice as likely as the average respondent to support Black and Asian immigration in 1997, three times as likely to favor increased or current levels in the 2010 BSA, and nearly 30 points more likely than average to say immigration enriches rather than undermines Britain's cultural life. Middle-class Left voters are nearly as pro-immigration as university-educated Left voters. Working-class Conservative voters are less than half as positive on immigration as the average White British respondent. Finally, the 2015 BES offers a sufficient sample (N = 563) of populist right BNP and UKIP voters to see that virtually none (7.6%) are positive about the cultural effects of immigration. The crosstabu-lation in Table 4 forms the basis for our choice to use university-educated Left voter, working-class Conservative voter, and populist right voter as proxies for immigration attitudes in the analyses to follow.

Results

Results of models with univariate proxies are run first and displayed in Table 5. We use a single dummy variable for party vote taking the value of 1 for Conservative and 0 for a Left party (Labour, Liberal Democrat, Green). Likewise for English identity. Model 1 presents a linear regression of the change in ethnic minority share experienced by an individual who moves wards. This measures H2: whether movers seek or avoid diversity. Independent variables include demographic, economic, and contextual variables with robust standard errors clustered on wards. Only inter-ward movers are considered. To focus on variables of interest, most demographic and economic parameters are not shown. Models 1 and 3 include all respondents (both U.K.born Whites and other ethnic groups), while Models 2 and 4 are restricted to U.K.-born Whites.

Model 1 shows that inter-ward movers, whether White or minority, who originate in more diverse wards move to less diverse wards than those who leave from more homogeneous wards. This reflects the fact that those moving from diverse wards have fewer higher diversity options open to them than those in homogeneous wards. Yet, from Model 1, we also see that there is a negative coefficient for the White U.K.-born dummy variable. This means U.K.-born Whites who move wards move to significantly less diverse places than minorities, even with controls for individual class, income, education, age, housing tenure, and contextual characteristics—the deprivation,

Table 5. Models of White "Flight" and "Avoidance," Univariate Immigration Proxies, 1991-2012.

Model 1 Model 2 Model 3 Model 4

Linear regression Fixed-effects model Logistic regression of Fixed-effects conditional logit

model predicting predicting percentage move to Whiter ward model of move to Whiter ward

percentage point point increase in ward (1) vs. not (0), for those (1) vs. not (0), for those who

increase in ward minority share (inter- who lived in diverse wards lived in diverse wards (over

minority share (inter- ward movers only, (over 30% minorities) 30% minorities) before moving

ward movers only) White U.K.-born only) before moving (White U.K.-born only)

Lagged minority -0.61 (0.03)*** -0.16 (0.02)*** 0.02 (0.01)* 0.13 (0.05)**

population

White U.K.-born -2.94 (0.73)*** 0.24 (0.27)

Conservative voter 0.43 (0.25) -0.38 (0.74) -0.10 (0.26) 1.55 (1.12)

(ref: Left voter)

English identifier -0.33 (0.22) — -0.14 (0.20) —

R2 or pseudo R2 .64 .30 .67

Within 0.54

Between 0.21

N 4,973 4,950 4,701 986

Groups 2,963 201

Source. BHPS (1991-2009); U.K. Household Longitudinal Study (UKHLS; 2012).

The following controls were added to the models but not shown in the regression table: Carstairs index of deprivation (current and lagged), population density (current and lagged), percentage of private renters and change in percentage of private renters, age, highest qualification, income, social class, employment status, having dependent children, marital status, religious practice, and housing tenure. For Models 3 and 4, additional controls were added for change in job, and marital and employment status. BHPS = British Household Panel Survey. *p < .05. **p < .01. ***p < .001.

population density, and proportion of renters in origin and destination wards. This finding is in keeping with the White avoidance hypothesis but could be generated by minority behavior or by differences in information or social networks between groups rather than by White attitudes. Thus, we need to test whether tolerant Whites are less likely to move away from diversity than anti-immigrant Whites before we can confirm a White avoidance pattern that might explain the hypothesized contextual effects (Whites more tolerant in diverse wards) we saw in Table 2.

Model 1 reveals that U.K.-born Whites move to significantly Whiter wards than non-Whites or foreign-born Whites. Model 3, by contrast, which tests for White flight (H1), shows that while the coefficient is signed in the direction of White exit, it is not significant: White British are no more likely to leave diverse wards than other ethnic groups. Indeed, all ethnic groups tend to leave wards with high concentrations of minority groups, which comports with the existing U.K. urban studies literature (i.e., Catney & Simpson, 2010). So we have evidence for "ethnic avoidance" but not "White flight" if we define these terms by reference to unexplained ethnic residuals. But what of the deeper meaning of flight and avoidance, linked to subjective attitudes toward immigrants?

Here, Model 1 shows no significant association between measures of antiimmigration sentiment—Conservative voting and English identity—and moving to Whiter wards. Where Model 1 looked at differences both within and between individuals, Model 2 narrows the focus to examine differences over time within individuals. We possess 125,692 observations across 13,877 unique individuals yielding an average of 9.1 observations per individual. Longitudinal data permit us to monitor changes in attitudes before and after people move to different racial contexts. Conditioning on variables that are correlated with both mobility and attitudes provides an estimate of the average causal effect of a change in (proxy) immigration attitudes on the degree to which an inter-ward mover selects a different racial environment. Model 2 presents a fixed-effects generalized least squares (GLS) regression of ward ethnic composition on party support, for Whites only. If the White avoidance hypothesis is correct, we would expect lagged Conservative voting to predict that an individual will move to a Whiter area than a Left party supporter when they move. In other words, conservative Whites in diverse areas should move to less diverse neighborhoods than their liberal neighbors. The dummy variable for Conservative (vs. Liberal/Labour/Green) voting is, however, not statistically significant, again disconfirming H2.13

Models 3 and 4 examine White flight (H1), asking whether those who originate in diverse wards (defined as 30% minority or more) stay or move to Whiter areas. The dependent variable is a dummy taking the value of 1 for a

move to a non-diverse (<30% minority) ward and 0 for staying in, or moving to another, diverse ward. This model asks whether Whites, especially conservative Whites, are more likely to leave diverse wards for Whiter wards as compared with liberal Whites. Model 3 is a logistic regression model that includes minorities and Model 4 a conditional (fixed-effects) logistic regression model restricted to Whites.14

The results reveal a similar pattern to Models 1 and 2: Although Model 3 does not show that Whites are more likely to leave diverse wards than minorities, the coefficient is signed in the right direction. Model 3 finds no significant attitudinal effects. Model 4, which is restricted to Whites, shows likewise that conservative and liberal Whites are equally likely to leave diverse wards, disconfirming H1. The lack of association is striking: Across all models, coefficients for Conservative vote and English identity are in the hypothesized direction in half and in the wrong direction in half. Results are robust to the inclusion of non-voters and those who voted for parties other than those included in the party variables used here.15

To recap, first, while White movers select significantly Whiter destinations than ethnic minorities, anti-immigration and pro-immigration Whites (on our proxy measures) select wards of similar diversity to move to. Second, there is no selective flight of White conservatives out of diverse areas. It is worth adding that attitudinal variables measuring British patriotism, traditional gender roles, right-wing ideology, and negative attitudes to homosexuals, asked only in occasional years in the BHPS, were similarly uncorrelated with moving to Whiter wards.16 These results chime with those of Gallego, Buscha, Sturgis, and Oberski (2014) whose work with the BHPS finds that Conservative supporters in Britain tend not to self-select into Conservative-dominated constituencies (which tend to be somewhat Whiter than Labour ones)—all of which cast doubt on the argument that Whites who dislike immigration leave diverse areas while those who are tolerant of immigration move toward, or remain in, diverse areas.

Yet it could be argued that the proxy measures we deployed are too blunt to capture immigration opinion, the dependent variable in Table 2. In our discussion of predictors of immigration opinion, we noted in Table 4 that multivariate proxies were much stronger measures of immigration attitudes than univariate terms for party support or English identity. Indeed, we saw that among U.K.-born Whites, university-educated voters for left-wing parties are outstandingly pro-immigration while working-class Conservative voters are strongly anti-immigration. Moreover, it is only in interaction with diverse contexts that we would expect these measures to be activated. Interacting these multivariate proxy groups with minority share therefore affords us a sharper test of H1 and H2. For instance, H1 predicts that a

degree-holding Left party voter would be more likely to remain in a ward than others as the ward becomes more diverse. H2 holds that the same person, if they move, will choose a ward that is more diverse than that selected by another respondent.

Table 6 presents results of four models along the same lines as Table 5 but restricted to U.K.-born Whites and using multivariate proxies for immigration attitudes. These show, as expected, that the share of minorities in one's prior ward of residence (a year ago) is an important determinant of whether, and where, one moves. As in Table 5, residents of diverse wards tend to leave more often than residents of Whiter wards. Residents of diverse wards also tend to select Whiter wards because there are few other diverse wards to move to, and because diverse wards in Britain tend to be deprived and crowded.

Moving to tests of our main hypotheses, our more granular immigrationproxy measures tell a similar story to Table 5, in that pro- and anti-immigration Whites exhibit a similar propensity to leave diverse wards. However, we now find some weak effects from self-selection, in the hypothesized direction. The sign on Conservative voter is in the hypothesized (H2) direction in Model 1 and close to significance. The working-class Conservative interaction, our finer proxy for anti-immigration attitudes, as well as its interaction with minority share, is also in the hypothesized direction but well short of significance. Moreover, in Model 2, White working-class Conservative voters are significantly more likely to remain in diverse wards than others, which runs counter to what is predicted by H1. Results are robust to the exclusion of non-voters and those who voted for parties other than those included in the party variables used here.

In terms of White avoidance, the story is more complex: White working-class conservatives do not move to significantly Whiter wards than other White movers, but White left-voting degree holders—our proxy for proimmigration respondents—who originate in diverse wards do move to relatively diverse wards compared with White movers originating in the same areas. Likewise, English identifiers who move from diverse wards select significantly Whiter places to move to than other Whites who depart from analogous wards. This lends support to H2, but is the effect large enough to account for the 10-point difference in immigration opinion17 among Whites living in the least and most diverse wards in England and Wales?

We know there is no difference in "flight" from diverse wards between pro- and anti-immigration Whites, so White avoidance must do all the work to produce the requisite 10-point immigration tolerance gap between Whites in diverse and homogeneous wards observed in Table 2. We would expect anti-immigration Whites to select wards that average a full 10 points Whiter

Table 6. Models of White "Flight" and "Avoidance," Multivariate Immigration Proxies, 1991-2012.

Model I Model 2 Model 3 Model 4

GLS linear regression GLS logistic GLS linear regression GLS logistic

predicting increase in regression predicting increase in regression

ward minority share predicting move ward minority share predicting move

(restricted to U.K.- out of ward (White (restricted to U.K.- out of ward (White

born White movers) U.K.-born only) born White movers) U.K.-born only)

Share (%) minorities in ward (lag) -0.436*** (0.008) 0.015*** (0.001) -0.438*** (0.009) 0.014*** (0.001)

Conservative voter (lag) -0.445 (0.266) 0.033 (0.033)

Working class x Conservative voter (lag) -0.092 (0.624) 0.005 (0.071)

Working class x Conservative voter x % -0.002 (0.001) -0.000* (0.000)

minorities in ward (lag)

Left party voter (lag) 0.222 (0.229) 0.020 (0.029)

Degree x Left party voter (lag) -1.099 (0.601) 0.109 (0.075)

Degree x Left party voter x % minorities 0.064** (0.023) 0.001 (0.003)

in ward (lag)

English identifier (lag) 0.271 (0.41 1) -0.048 (0.053)

English identifier x % minorities in ward -0.1 14*** (0.030) 0.003 (0.004)

9,154 122,306 9,077 122,306

Groups 4,847 4,839

R2/log likelihood .385 -33,532.23 .383 -33,620.98

Source. BHPS (1991-2009); U.K. Household Longitudinal Study (2012).

The following controls were added to the models but not shown in the regression table: age, change in being single, change in being a renter, and lags of degree and income. Models 2 and 4 also include lagged mover. GLS = generalized least squares; BHPS = British Household Panel Survey. *p < .05. **p < .01. ***p < .001.

% minorities in individual's origin ward

Other -» Left voter with degree (prior to move)

Figure 1. Predicted change in % minorities due to move (Whites only).

than those chosen by pro-immigration Whites. This would generate a distinctive attitudinal inflow, producing a selectively liberal population stock which is 10 points more tolerant in diverse wards than in homogenous ones. In this case, self-selection would account for the disparate LA-ward contextual effects set out in Table 2.

Yet, as Figure 1 shows, the predicted change in the share of minorities in one's ward as a result of a move differs only modestly between pro- and antiimmigration Whites. Using left degree-holders as the pro-immigration attitude proxy yields a difference of just 2 points from other White respondents, while using English identifiers (not shown) produces a difference of just 0.3 points from other Whites. These differences, while significant, take us at most a fifth of the way toward H2, 8 points shy of what is required to account for the softening effects of diversity on immigration opinion recorded in Table 2.

Even this may overstate the effects since H1 is not borne out: White left degree-holders are not disproportionately represented among those choosing to stay in diverse areas. In response to concerns over the modifiable area unit problem (i.e., Wong et al., 2012), it is worth noting that a similar pattern holds

for moves at the higher geography of the Local Authority (LA): Whites move to Whiter LAs than minorities, but there is only a very modest difference in destination between liberal and conservative Whites—the coefficients for immigration proxies are even weaker than at ward level. This suggests heightened opposition to immigration in diverse LAs is also not the result of selection, thus leaving the door open to threat-based explanations.

The BHPS allows us to proxy immigration attitudes using party support— English national identity, education, class, and multivariate combinations thereof—but one might still argue these measures do not fully capture immigration attitudes. Therefore, we undertake two further robustness checks. First, there is little doubt that support for a populist right party (BNP or UKIP) is associated with opposition to immigration—in Table 4 we found that all but 7.6% of populist right supporters held negative attitudes toward immigration. Data on far right (mostly BNP) and UKIP voting comes from the first three waves of the Understanding Society (UKHLS) surveys of 2009-2013, with a sample of approximately 30,000 White British individuals per wave. The number of far right supporters is 490 in Wave 1, 337 in Wave 2, and 221 in Wave 3, a declining pattern corresponding to the waning of BNP vote share over 2009-2012. UKIP supporters numbered 393 in Wave 1, 364 in Wave 2, and 431 in Wave 3, a broadly rising pattern consonant with the general trend of growing UKIP support during 2009-2012. Although UKIP also trades on an anti-elite and anti-European message, concerns over immigration are central to its appeal (Ford & Goodwin, 2014). Combining UKIP and Far Right supporters furnishes 2,226 anti-immigration person-years of data out of a sample of some 90,000 White British person-years.

Table 7 presents results from two random-effects GLS models. We use the same analytical strategy used earlier with BHPS data on working-class Conservative voting, this time concentrating on UKHLS data on the BNP and UKIP. As in Table 6, the dependent variable in Model 1 is a dummy coded 1 for a move out and 0 for remaining in, or moving within, ward.18 In Model 2, which is restricted to movers, the dependent variable is the difference in minority share between origin and destination ward.

The story corroborates findings from Tables 5 and 6. Model 1 considers White flight. Here, the key parameter is the interaction between anti-immigration party voting and share of minorities in origin ward. Whites who support anti-immigration parties and live in diverse wards are not significantly more likely to leave their ward than Whites in identical wards who do not support such parties or Whites who support such parties but live in Whiter wards. This contradicts H1. Model 2 examines White avoidance. The main effect and interactions for anti-immigration party support are not significant. Thus, again we find that anti-immigration Whites (as indicated by populist

Table 7. Random Effects Models Predicting Mobility (White British Respondents Only), 2009-2012.

Model 1

Model 2

GLS logistic regression predicting move out of ward

GLS linear regression predicting increase in minority share in ward due to move

Mover (lag)

Minority population share in

ward (lag) UKIP/BNP supporter (lag) UKIP/BNP supporter x Minority

population share (lag) R2 N

Groups

0.933 (0.192)*** 0.003 (0.003)

-0.191 (0.277) 0.014 (0.019)

20,806 20,806

-0.419 (0.01)**

-0.226 (1.721) -0.146 (0.138)

.529 2,557 2,361

Source. U.K. Household Longitudinal Study (2012).

The following controls were added to Model 1 but not shown in the regression table: Carstairs index of deprivation in ward (lagged), age, highest qualification (lagged), marital status (lagged), and housing tenure (lagged). The following controls were added to Model 1 but not shown in the regression table: Carstairs index of deprivation in ward (lagged), change in ward Carstairs index of deprivation, change in ward population density, age, highest qualification (lagged), marital status (lagged), and housing tenure (lagged). BNP = British National Party.

*p < .05. **p < .01. ***p < .001.

right support) move to areas that are no Whiter than those selected by more liberal Whites, contrary to that predicted by H2. This said, it bears mentioning that all coefficients are signed in the hypothesized direction, hinting at small White flight and avoidance effects among those who strongly oppose immigration.

As a final robustness check, we commissioned a YouGov political tracker survey of 1,869 British adults that contains both immigration-proxy items from the BHPS-UKHLS such as party vote and national identity as well as direct questions on race and immigration.19 Of the 1,638 White British respondents, 906 claimed not to have moved wards over the past decade, 384 said they had moved wards, while the rest were unsure. Comparing White British stayers with those who reported they had moved to less or more diverse wards, Table 8 shows that the BHPS-UKHLS attitude proxies have a similar relationship to mobility as anti-immigration attitudes. Namely, although there are small attitudinal differences between Whites who said they

Table 8. Characteristics (in %) of White British Survey Sample, by Self-Reported Mobility Status, 2013.

Movers from diversity Movers to diversity Stayer

Immigration: Increase or same 23% 25% 17%

Immigration: reduce a little 16% 21% 19%

Immigration: reduce a lot 60% 54% 64%

N 146 89 906

English identity 43% 40% 50%

Conservative party 28% 25% 28%

N 148 91 927

Discomfort interracial 21% 14% 23%

marriage

N 139 83 903

Source. YouGov (2013).

N = 1,638 White British adults. Note that number of cases is slightly different for different groups of variables depending on response rate.

moved toward diversity and away, negative immigration attitudes are not associated with moving away from diversity. This is especially so when demographic and socioeconomic variables are included in a model predicting a move to/from diversity.20

Stayers, meanwhile, are more conservative on immigration than those leaving diversity. These findings echo work on the BNP, which finds support for the party to be strongest in wards with limited White inflows and outflows, that is, with a high proportion of stayers (Harris, 2012). At the 2008 Greater London Authority elections, when support for the BNP was at its peak, wards that received a large amount of "White flight" were actually less supportive of the BNP. All of which questions H1 and H2. Previous work finds that people's subjective assessment of their local diversity matters more for voting, participation, and attitudes than objective measures of diversity based on census boundaries. They misperceive levels of diversity at higher geographies more than at lower levels (Leighley, 2001; Wong et al., 2012). Crucially, this YouGov survey tests people's pseudoenvironment—their subjective perceptions of past and present neighborhood diversity—and uncovers a similar pattern to that observed with the objective census measures used in Tables 5 and 6. Likewise, in Table 2, when we substitute a subjective measure of area diversity for census ward minority share, we find that Whites who describe their area as more diverse are more tolerant of immigration.21 This gives us added confidence that selection effects, whether based on

objective or subjective measures of context and locale, are not driving the findings in contextual studies of voting and public opinion.

Discussion

Most contextual studies of voting and public opinion find that diversity in large geographic contexts such as county or metropolitan area predicts a threat response among Whites, while diversity at low levels of geography such as ward or tract is associated with more liberal attitudes and voting. Is this pattern best explained by the fact there are more opportunities for contact at lower rather than higher geographies? Or is it a by-product of "White flight": the self-selection of intolerant Whites out, and tolerant Whites in, to diverse locales.

Research on the effects of local diversity on voting and public opinion has been hampered by the lack of longitudinal data with which to test for self-selection through White flight from diversity, or partisan flight. The validity of work on the role of racial context in activating threat perceptions or facilitating positive contacts with minorities hinges on whether self-selection can be satisfactorily addressed. This analysis, which uses large-scale longitudinal data geocoded to low geographic levels, addresses this concern. It helps validate previous contextual work by demonstrating that self-selection is not a convincing explanation for positive White attitudes to immigration and diversity in diverse local contexts. Nor can it explain the relatively negative attitudes to immigration in diverse larger geographies such as Local Authorities.

We believe these findings may be generalized to other countries. Dutch replication of the MCSUI study suggests the relationship between White residential preferences and attitudes to minorities in the Netherlands and America are quite similar (van Londen, 2012). Moreover, European and American research shows that ethnic context has broadly similar effects on White immigration attitudes on both sides of the Atlantic (Semyonov & Glikman, 2009).

This article also reinforces the small but growing number of studies that find that self-selection does not account for partisan residential patterns or contextual effects. Gallego et al. (2014), analyzing the same longitundinal BHPS data used here, and Cho, Gimpel, and Hui (2013), using American voter registration data, likewise find no evidence that partisans tend to "sort" into strongly partisan areas, as intimated by work such as that of Bishop and Cushing (2008). In both the British and American case, socialization and compositional effects (i.e., Whites and families moving to Republican counties) seem to account for partisan segregation. Our findings likewise uncover only a weak effect from self-selection. Self-selection is arguably the major criticism of findings in the

wider literature on the contextual effects of voting and public opinion. This work should go some way toward allaying those concerns.

Acknowledgments

The authors wish to thank the Economic and Social Research Council (ESRC), Award Ref. ES/K003895/1, for supporting this research. They received useful feedback from seminars at Oxford University (sociology), University of Manchester (CCSR), and University of Texas (political science). The authors wish to thank Jonathan Laurence, Stephen Jivraj and Charles Breton for helpful comments on earlier drafts.

Declaration of Conflicting Interests

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

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

1. This said, negative contact in intimate contexts has the capacity to lead to rising threat perceptions (Barlow et al., 2012).

2. References for Table 1 are in Supplemental Table 1 at www.sneps.net/cps1sup-plemental.htm

3. Some of the exceptions entail situations where hostility is directed toward specific groups (i.e., Hispanics in America, Muslims in Europe, low-income minorities) but not others (i.e., East Asians, Afro-Caribbeans in Europe, high-income minorities). See, for example, Ha (2010) for East Asians, or Biggs and Knauss (2012) for Afro-Caribbeans.

4. Biggs and Knauss (2012) use Output Areas, with populations under 1,000, as their measure of locale, whereas Bowyer (2008) and Harris (2012) use wards. In all studies, local diversity is associated with lower, and LA diversity with higher, BNP support.

5. Full model specification and variable distributions available upon request.

6. For calculations, see Supplemental Table 2 at www.sneps.net/cps1supplemental. htm

7. Potential problems associated with changes in ward boundaries between the 1991 and 2001 censuses are mitigated by our use of GeoConvert software (http:// geoconvert.mimas.ac.uk/, accessed July 2, 2013). To arrive at common 20012011 wards, we build wards up from a common geography based on Middle Layer Super Output Areas.

8. Those born abroad, and subjects living in Scotland or Northern Ireland, are excluded, to most closely approximate the ethnic majority group in England and Wales.

9. See Supplemental Table 3, Model 4 at www.sneps.net/cps1supplemental.htm

10. See Supplemental Table 4, Model 4 at www.sneps.net/cps1supplemental.htm

11. This was just below the response for British, at 41.75%.

12. See Supplemental Table 3, Model 3 at www.sneps.net/cps1supplemental.htm

13. The annual sample size for the English identity question is small and variable, so it is omitted from fixed-effects models; however, this term is also not statistically significant.

14. Note that sample size drops in Model 4 because it is a conditional (fixed-effects) logit model.

15. This is done by running Conservative and Left party as separate variables, compared with the entire sample, which includes non-voters.

16. See Supplemental Table 4 at www.sneps.net/cps1supplemental.htm

17. The gross difference is actually 20 points, but 10 points are accounted for by age, education, marital status, and other compositional characteristics that differ between diverse and homogeneous wards.

18. A move to a ward of similar ethnic diversity is coded 0.

19. For details on YouGov survey methodology and questionnaire, see http:// d25d2506sfb94s.cloudfront.net/cumulus_uploads/document/kf5d231qce/ YG-Archive-ESRC-Demos-Birkbeck-results-300713.pdf

20. Models (available upon request) using YouGov data with move toward/away from diversity as the dependent variable find immigration and immigrationproxy attitudes not to be significant predictors of White mobility.

21. The Citizenship Survey question reads, "People in my local area have the same ethnic group as me," with responses on a four-item scale from "all the same" to "less than half." Coefficients are similar to the objective ward diversity measure, although standard errors are somewhat larger—which could be due to differences of unit.

References

Abrajano, M., & Hajnal, Z. (2009). Latino context and White partisan choices. La Jolla: University of California San Diego.

Allport, G. W. (1954). The nature of prejudice. Cambridge, MA: Addison-Wesley.

Barlow, F. K., Louis, W. R., & Hewstone, M. (2009). Rejected! Cognitions of rejection and intergroup anxiety as mediators of the impact of cross-group friendships on prejudice. British Journal of Social Psychology, 48, 389-405. doi:10.1348/01 4466608x387089

Barlow, F. K., Paolini, S., Pedersen, A., Hornsey, M. J., Radke, H. R. M., Harwood, J., . . . Sibley, C. G. (2012). The contact caveat: Negative contact predicts increased prejudice more than positive contact predicts reduced prejudice. Personality and Social Psychology Bulletin, 38, 1629-1643. doi:10.1177/0146167212457953

Biggs, M., & Knauss, S. (2012). Explaining membership in the British National Party: A multilevel analysis of contact and threat. European Sociological Review, 28, 633-646. doi:10.1093/esr/jcr031

Bishop, B., & Cushing, R. G. (2008). The big sort: Why the clustering of like-minded

America is tearing us apart. Boston, MA: Houghton Mifflin. Blalock, H. M. (1957). Per cent non-White and discrimination in the south. American

Sociological Review, 22, 677-682. Bowyer, B. (2008). Local context and extreme right support in England: The British National Party in the 2002 and 2003 local elections. Electoral Studies, 27, 611-620. Branton, R. P., & Jones, B. S. (2005). Reexamining racial attitudes: The conditional relationship between diversity and socioeconomic environment. American Journal of Political Science, 49, 359-372. doi:10.1111/j.0092-5853.2005.00128.x British Household Panel Survey. (1991-2009). Waves 1-18: Special licence access, census area statistics wards (UK Data Archive SN#6135). Colchester, UK: Institute for Social and Economic Research, University of Essex. Campbell, A. L., Wong, C., & Citrin, J. (2006). "Racial threat," partisan climate, and direct democracy: Contextual effects in three California initiatives. Political Behavior, 28, 129-150. doi:10.1007/s11109-006-9005-6 Catney, G., & Simpson, L. (2010). Settlement area migration in England and Wales: Assessing evidence for a social gradient. Transactions of the Institute of British Geographers, 35, 571-584. Cho, W. K. T., & Baer, N. (2011). Environmental determinants of racial attitudes redux: The critical decisions related to operationalizing context. American Politics Research, 39, 414-436. doi:10.1177/1532673x10377167 Cho, W. K. T., Gimpel, J. G., & Hui, I. S. (2013). Voter migration and the geographic sorting of the American electorate. Annals of the Association of American Geographers, 103, 856-870. doi:10.1080/00045608.2012.720229 Crowder, K., & South, S. J. (2008). Spatial dynamics of White flight: The effects of local and extralocal racial conditions on neighborhood out-migration. American Sociological Review, 73, 792-812. Fetzer, J. (2000). Public attitudes toward immigration in the United States, France,

and Germany. Cambridge, UK: Cambridge University Press. Fieldhouse, E., Green, J., Evans, G., Schmitt, H., & van der Eijk, C. (2014). British Election Study Internet Panel: Waves 1-2. Retrieved from www.britishelection study.com/data-objects/panel-study-data/ Ford, R., & Goodwin, M. J. (2010). Angry White men: Individual and contextual

predictors of support for the British National Party. Political Studies, 58, 1-25. Ford, R., & Goodwin, M. J. (2014). Revolt on the right: Explaining support for the

radical right in Britain. London, England: Routledge. Gallego, A., Buscha, F., Sturgis, P., & Oberski, D. (2014). Places and preferences: A longitudinal analysis of self-selection and contextual effects. British Journal of Political Science. Advance online publication. doi:10.1017/S0007123414000337 Gay, C. (2006). Seeing difference: The effect of economic disparity on Black attitudes toward Latinos. American Journal of Political Science, 50, 982-997. doi:10.1111/ j.1540-5907.2006.00228.x Ha, S. E. (2010). The consequences of multiracial contexts on public attitudes toward immigration. Political Research Quarterly, 63, 29-42. doi:10.1177/ 1065912908325255

Harris, G. (2012). The rise and fall of the British National Party: The demand for extreme right politics in the UK (Unpublished doctoral thesis). Birkbeck College, University of London, UK.

Heath, A., Jowell, R., & Curtice, J. K. (2002, July). British Election Panel Study, 1997-2001; Waves 1 to 8 [Computer file] (4th ed.). Colchester: UK Data Archive [distributor]. Doi:10.5255/UKDA-SN-4028-1

Hewstone, M., Cairns, E., Voci, A., Paolini, S., McLernon, F., Crisp, R., . . . Craig, J. (2005). Intergroup contact in a divided society: Challenging segregation in Northern Ireland. In D. Abrams, J. M. Marques, & M. A. Hogg (Eds.), The social psychology of inclusion and exclusion (pp. 265-292). Philadelphia, PA: Psychology Press.

John, P., Margetts, H., Rowland, D., & Weir, S. (2006). The BNP: The roots of its appeal. London, UK: Democratic Audit.

Key, V. O. (1949). Southern politics in state and nation (1st ed.). New York, NY: Alfred A. Knopf.

Krysan, M. (2002). Whites who say they'd flee: Who are they, and why would they leave? Demography, 39, 675-696.

Leighley, J. (2001). Strength in numbers? The political mobilization of racial and ethnic minorities. Princeton, NJ: Princeton University Press.

National Centre for Social Research. (2012, February). British Social Attitudes Survey, 2010 [Computer file]. Colchester: UK Data Archive [distributor]. doi:10.5255/ UKDA-SN-6969-1

Office for National Statistics. (2013). Census of England and Wales 2011 (Table CT0010 ethnic group (write-in responses), wards in England and Wales). Available from www.ons.gov.uk

Office for National Statistics and Home Office. (2010). Communities group, home office citizenship survey, 2009-2010 (UK Data Archive Study No. 673). London, UK: Home Office.

Office for National Statistics and Home Office. (2011). Communities group, home office citizenship survey, 2010-2011 (UK Data Archive Study No. 7111). London, UK: Home Office.

Oliver, J. E., & Wong, J. (2003). Intergroup prejudice in multiethnic settings. American Journal of Political Science, 47, 567-582. doi:10.2307/3186119

ONS LS. (2001). Office ofNational Statics. The permission of the Office for National Statistics to use the Longitudinal Study is gratefully acknowledged, as is the help provided by staff of the Centre for Longitudinal Study Information & User Support (CeLSIUS). CeLSIUS is supported by the ESRC Census of Population Programme (Award Ref: RES-348-25-0004). The authors alone are responsible for the interpretation of the data. Census output is Crown copyright and is reproduced with the permission of the Controller of HMSO and the Queen's Printer for Scotland.

Pettigrew, T. F., & Tropp, L. R. (2006). A meta-analytic test of intergroup contact theory. Journal of Personality and Social Psychology, 90, 751-783. doi:10.1037/0022-3514.90.5.751

Putnam, R. (2007). E Pluribus Unum: Diversity and community in the twenty-first

century. Scandinavian Political Studies, 30, 137-174. Quillian, L. (1995). Prejudice as a response to perceived group threat—Population composition and anti-immigrant and racial prejudice in Europe. American Sociological Review, 60, 586-611. doi:10.2307/2096296 Schlueter, E., & Scheepers, P. (2010). The relationship between outgroup size and anti-outgroup attitudes: A theoretical synthesis and empirical test of group threat- and intergroup contact theory. Social Science Research, 39, 285-295. doi:10.1016/j.ssresearch.2009.07.006 Semyonov, M., & Glikman, A. (2009). Ethnic residential segregation, social contacts, and anti-minority attitudes in European societies. European Sociological Review, 25, 693-708. doi:10.1093/esr/jcn075 Stein, R. M., Post, S. S., & Rinden, A. L. (2000). Reconciling context and contact effects on racial attitudes. Political Research Quarterly, 53, 285-303. doi:10.2307/449282

U.K. Household Longitudinal Study. (2012). Waves 1-2, 2009-11 (Vol. SN#6614).

Colchester: UK Data Archive. van Londen, M. (2012). Exclusion of ethnic minorities in the Netherlands: The effects of individual and situational characteristics on opposition to ethnic policy and ethnically mixed neighbourhoods (Doctoral thesis). Radboud University, Nijmegen, The Netherlands. Walzer, M. (1983). Spheres of justice: A defense of pluralism and equality. Oxford, UK: Blackwell.

Wong, C., Bowers, J., Williams, T., & Simmons, K. D. (2012). Bringing the person back in: Boundaries, perceptions, and the measurement of racial context. Journal of Politics, 74, 1153-1170. doi:10.1017/s0022381612000552 Wright, M., Citrin, J., & Wand, J. (2012). Alternative measures of American national identity: Implications for the civic-ethnic distinction. Political Psychology, 33, 469-482.

YouGov. (2013). YouGov/ESRC, Demos andBirkbecksurvey results. Retrieved from http://d25d2506sfb94s.cloudfront.net/cumulus_uploads/document/kf5d231qce/ YG-Archive-ESRC-Demos-Birkbeck-results-300713.pdf

Author Biographies

Eric Kaufmann is a professor of politics at Birkbeck College, University of London. He is the author of Shall the Religious Inherit the Earth (Profile 2010), The Rise and Fall of Anglo-America (Harvard 2004), and The Orange Order (Oxford 2007), and editor of Political Demography (Oxford 2012) and Rethinking Ethnicity (Routledge 2004). He is an editor of the journal Nations and Nationalism.

Gareth Harris is a research associate, Centre for Trust, Peace and Social Relations. He is co-author, with Eric Kaufmann, of Changing Places: The White British Response to Ethnic Change (London: Demos, 2014).