Can Narratives of White Identity Reduce Opposition to Immigration and Support for Hard Brexit? A Survey Experiment

Abstract

Britain’s vote to leave the European Union highlights the importance of White majority opposition to immigration. This article presents the results of a survey experiment examining whether priming an open form of ethno-nationalism based on immigrant assimilation reduces hostility to immigration and support for right-wing populism in Britain. Results show that drawing attention to the idea that assimilation leaves the ethnic majority unchanged significantly reduces hostility to immigration and support for Hard Brexit in the UK. Treatment effects are strongest among UK Independence Party, Brexit and White working-class voters. This is arguably the first example of an experimental treatment leading to more liberal immigration policy preferences.

Keywords

Brexit immigration experiment nationalism majority ethnicity

The politics of immigration has become increasingly prominent in the West. On 23 June 2016, in a vote driven by opposition to European Union (EU) immigration (Evans and Mellon, 2016), Britain voted to leave the EU. Research has consistently shown that populist right voting is predicted by immigration attitudes, which in turn are associated with cultural motivations, notably ethnic nationalism (Citrin and Sides, 2008; Hainmueller and Hopkins, 2014; Heath and Tilley, 2005). Building on this, I suggest ethnic nationalists are not monolithic, but fall into two groups, open and closed, with the former amenable to changing their immigration policy preferences. This is demonstrated through a survey experiment which primes open ethnic nationalism to shift immigration attitudes in a liberal direction. Past experiments have primed nationalist sentiment to heighten opposition to immigration (i.e. Sniderman et al., 2004), but none I am aware of have succeeded in reducing it. In addition to contributing to scholarship, the findings inform policy by suggesting new forms of political communication which speak to the cultural anxieties of populist right voters.

The article presents the results of a survey experiment conducted in Britain 2 months after the Referendum on EU membership in the midst of a debate over whether the country should opt for a ‘Soft Brexit’, accepting freedom of movement from the EU in exchange for access to the Single Market, or ‘Hard Brexit’, restricting freedom of movement at the expense of losing market access. These positions are operationalised through the use of a Contingent Valuation (CV) approach which asks respondents to make trade-offs between personal income and immigration levels. Respondents are allocated into three groups, with one treatment group reading a vignette about how immigrants tend to assimilate into the ethnic majority. The study examines the effect of this ‘assimilation treatment’ on the immigration policy preferences of respondents, focusing in particular on culturally conservative White British voters.

National Identity and Immigration Policy Preferences

This article contributes to scholarship on the relationship between national identity and immigration policy preferences. A considerable body of work locates an association between ‘ethnic’ or exclusive definitions of the nation and opposition to immigration (Citrin et al., 1990, 2001; Citrin and Wright, 2009; Coenders, 2001; Heath and Tilley, 2005; Pehrson et al., 2009; Schildkraut, 2014; Wong, 2010). A smaller number of studies seek to establish causal links between national identity and attitudes to immigration using survey experiments which prime national identity. Few of these have been able to manipulate anti-immigration policy preferences, as opposed to perceptions of immigrants or the number of immigrants. One study does however find that attitudes to minorities can be improved by priming civic nationalism (Charnysh et al., 2015). Louis et al. (2013: 131) uncover an indirect relationship among Australian and Canadian respondents between reading a prime on national identification and opposing immigration, but no unmediated effect. Wright and Citrin (2010) find that White American respondents respond less negatively to images of Latino protests when the protestors wave American rather than Mexican flags, though this does not prompt individuals to accept higher immigration levels. Breton (2015) shows in the Canadian case that reminding respondents of their national identity also has no significant effect on immigration attitudes. An important exception is the classic study by Sniderman et al. (2004: 44–45), who report that priming Dutch rather than personal identity increases anti-immigration sentiment – but this is true only among those with low prior hostility to immigration.

The aforementioned experimental studies treat national identity as an ordinal variable, but as Breton (2015: 375) notes, national identity is unlikely to be unidimensional when multiple traditions (i.e. Smith, 1997) of nationhood coexist. Recent theoretical and qualitative advances point to the importance of peer-to-peer emergence ‘from below’ rather than elite diffusion in constructing national identity (Fox, 2014). If national identity is in large measure a complex system emerging from the disparate constructions of a pluralistic mass public, the content of this identity will vary between individuals and groups (Kaufmann, 2016). Thus, a nation may be ‘ethnic’ and ‘civic’ at once, with competing conceptions of nationhood within the population (Hutchinson, 2005).

Thus, I focus on the more ethnic conception of the nation held by the conservative section of the White British majority, which differs in its symbolic-affective content from the civic or multicultural variants held by many minorities or White liberals. A national story of rising diversity may appeal to high-identifying ethnic minorities and White liberals with a weak sense of linked fate with other Whites, but would be predicted to alienate authoritarian or conservative Whites who identify strongly with their racial group (Feldman and Stenner, 1997; Jardina, 2014; Stenner, 2005: 762).

The UK Context

Immigration has been among the most important three issues in Britain since 2002 according to the quarterly Ipsos-Mori Issues Index and an annual survey of Members of Parliament asking about the leading concerns of their constituents. In addition, the immigration issue rose steadily in prominence from the late 1990s, reflecting both higher net migration levels – particularly from Eastern Europe – and growing media coverage of immigration (Duffy et al., 2015: 5). Multiculturalism, initially embraced by Tony Blair’s Labour government in 1997, soon fell out of favour among both centre-left and centre-right elites in response to a series of White-Asian riots in the so-called ‘mill towns’ of Oldham and Burnley in 2001. Politicians responded by talking tough on immigration and seeking to replace a national identity based on multiculturalism with one emphasising integration into a shared set of British values. The high-profile terrorist attacks of 11 September 2001 (US) and 7 July 2005 in Britain, intensified the integrationist mood. Left-wing politicians such as David Blunkett and Gordon Brown, as well as Conservatives such as David Cameron, downplayed ethnic difference and emphasised a shared civic British national identity (Goodhart, 2013: 224; Modood, 2013: 10–11).

Immigration levels remained at historically high levels under the Cameron government after 2010, notwithstanding his pledge to reduce net migration from around 300,000 to the ‘tens of thousands’ per year.¹ Meanwhile, despite the shift from multiculturalist to civic nationalist narratives emphasising the need to integrate immigrants, conservative public opinion was not placated. The immigration question eventually prompted the country to leave the EU: not only was it the leading issue among Leave voters, but those motivated by hostility to immigration turned out to vote in the Referendum in significantly higher numbers than pro-immigration voters (Clarke et al., 2017: Chapter 7).

Could it be that emphasising civic integration as a means of legitimating immigration was insufficient to allay concerns? Might the conservative swath of British public opinion be more invested in an ethnically inflected sense of national identity rather than the state-defined nationhood championed by Brown and Cameron? This seems to have been the case (i.e. Heath and Tilley, 2005), with the official version of national identity contested by both ethno-nationalists and multiculturalists. In Britain in 2013, for instance, 51% of the population answered that having British ancestry is an important aspect of being ‘truly British’ (Park et al., 2014). Among British Whites, the figure is 60%, versus only 18% for non-White; 73% of British Whites who want immigration reduced a lot say ancestry is important for being truly British compared to 36% of Whites who are comfortable with current or higher levels (NatCen Social Research, 2016). Would the 73% have responded more favourably to a narrative of assimilation which sought to reassure them that their ethnic boundaries were secure despite immigration? This is one question this article seeks to address.

Open Ethnic Nationalism and Immigration Preferences

It is vital to unpack the idea of national identity to concentrate explicitly on dominant ethnic groups, which in most Western countries consist of a White ethnic majority which considers itself to have founded the nation-state (Kaufmann, 2004). Conservative dominant group members assert a proprietary claim to national membership which may shape their attitudes to the boundaries of ‘we’ and, by extension, towards immigration. A number of studies prime American dominant ethnicity. For instance, Brader et al. (2008) and Hainmueller and Hopkins (2015) find that White respondents reduce their opposition to immigrants when immigrant profiles are European rather than non-European.

Yet, even majority ethnicity contains its ‘open’ and ‘closed’ variants: ethnic boundaries vary in exclusivity. In some locales, such as northern India or Northern Ireland, ethnic boundaries are tight, with strong sanctions against intermarriage. Elsewhere – notably in the Caribbean or Latin America where typologies of race are fluid, or in Sub-Saharan Africa where minorities often assimilate into dominant groups such as the Wolof, Baganda or Kikuyu – boundaries are more permeable (Wimmer, 2007). The definition of who is included may also vary over time and place (Barth, 1969; Brubaker, 2006). For Richard Alba, the boundary between Hispanic and White Americans is more permeable than that between European ethnic majorities and Muslim groups (Alba, 2005). Ethnicity is defined by a belief in common ancestry (Francis, 1976: 2–9; Smith, 1991), but those who trace shared ancestry along one line may nevertheless embody considerable admixture on other lines through intermarriage. As a result, ethnic groups may differ internally in ‘racial’ appearance (i.e. Jews, Uighurs, Native Americans or Pashtuns).

Accordingly, we may speak of open and closed forms of ethnic majoritarianism, with open ethnic majorities secure about their capacity to assimilate immigrants, while closed majorities insist on exclusion. A classic example of an ethnic group which oscillated between open and closed self-conceptions is American White Anglo-Saxon Protestants (WASPs). As John Higham notes of the period from 1865 to 1890:

the Anglo-Saxonists were pro rather than con. During an age of confidence almost no race-thinker directly challenged a tolerant and eclectic attitude toward other European groups. Instead, Anglo-Saxon and cosmopolitan nationalisms merged in a happy belief that the Anglo-Saxon has a marvellous capacity for assimilating kindred races, absorbing their valuable qualities, yet remaining essentially unchanged (Higham, 1955: 33).

Only when this confidence flagged, Higham writes, did immigration restriction rise up the political agenda. Note that open ethnic majoritarianism is distinct from civic nationalism in that the former seeks to preserve an ethnic majority, or descent community, rather than accept a multi-ethnic form of nationhood based solely on state institutions and values. For open ethnic nationalists, majority ethnicity remains a key component of nationhood, but its boundaries are maintained partly through marital and ‘identificational’ assimilation (Gordon, 1964).

Priming Open Ethnic Nationalism

The strategy of this article is to variously prime civic nationalism, the contemporary centrist narrative and open ethnic nationalism, to see whether this distinction produces disparate effects on immigration attitudes and right-wing populism. A narrative of growing ethnic diversity coupled with economic strength should be endorsed by civic, but not by ethnic, nationalists – whether of the open or closed variety. Alternatively, a vignette about successful assimilation into the ethnic majority should reassure open but not closed ethnic nationalists. Vignettes draw the attention of White respondents to processes of continuity – maintaining boundaries through assimilation – or, alternatively, change – immigration causing group decline. From a distance, a river is timeless, but up close it is constantly changing: you cannot put your foot in the same water twice. Directing an ethno-nationalist individual to step back and appreciate ethnic continuity through assimilation should reassure while drawing their focus to fast-flowing changes and White decline should prompt insecurity.

A psychological study which partly embodies this approach, by Craig and Richeson (2014), finds that priming White Americans about the impending demise of the White majority moves opinion in a more conservative direction across both ‘race-related’ and race-neutral issues. Major et al. (2016) show that high-identifying Whites who read an article about America losing its White majority in 2042 become more opposed to immigration and more supportive of Donald Trump. Low-identifying Whites had the opposite response, becoming less supportive of Trump when primed.

In addition to testing threat responses to a diversification story, a replication of previous work, this article tries to determine whether a reassuring narrative of assimilation may have the opposing effect among White conservatives. As noted, I could find no studies showing a liberalising effect on conservative White immigration policy preferences. This may be because immigration preferences represent hard-to-shift ‘strong attitudes’ which lead respondents to avoid or ignore information which fails to confirm prior opinions (Druckman and Leeper, 2012: 877). In line with this ‘motivated reasoning’ perspective, several studies which test whether correcting respondents’ misperceptions about the actual number of immigrants affects their immigration policy preferences report no significant effects (Hopkins et al., 2016; Lawrence and Sides, 2014).

Hypotheses

The literature on ethnic boundaries and the cultural sources of immigration opinion leads to the following expectations:

H1. Those reading a diversification story should become more opposed to immigration because it primes anxiety about group decline. Immigration should increase in salience and support for right-wing populism should rise.

H2. Those reading the assimilation story should become more relaxed about immigration because they perceive that majority ethnic boundaries can be maintained through assimilating immigrants. The salience of immigration and right-wing populist views should decrease.

Given the literature on authoritarianism, conservatism and open ethnic nationalism, we should also expect that

H3. Conservatives, authoritarians and ethnic nationalists will respond more strongly to both primes in the directions given in H1 and H2.

It could be the case that conservatives and authoritarians are closed rather than open ethnic nationalists, in which case,

H4. The diversity treatment will increase conservative and authoritarian anti-immigration sentiment, salience and populist right support but the assimilation treatment will have no effect.

Data and Method

Data come from a survey fielded by the survey firm YouGov which took place on 16–17 August 2016 and consisted of 1677 adults, of which 1485 were from the White British ethnic majority. Respondents are selected from YouGov’s online panel and weighted according to age, education, region, class and voting criteria listed in Appendix 4. Full question wording is provided in Appendix 3.

Independent Variables

Participants were first asked a series of standard items, including demographics, party identification, past voting behaviour (including EU referendum vote in Britain), and self-placement on a left-right scale. This was followed by two items tapping authoritarianism, and one measuring conservatism. The first is a relatively unobtrusive authoritarianism question, part of a longstanding four-item childrearing authoritarianism scale (Feldman and Stenner, 1997; Perez and Hetherington, 2014; Stenner, 2005): ‘Please tell me which one you think is more important for a child: 1) To be considerate; 2) To be well-behaved’. The second is more explicit, drawing on another long-used item deployed by the survey firm Cultural Dynamics (Rose, 2011): ‘How well does this describe the REAL you?: “I believe that sex crimes, such as rape and attacks on children, deserve more than mere imprisonment. I think that such criminals ought to be publicly whipped, or worse”’. Answers range from 1 = not at all like me to 6 = very much like me. A third item focuses on conservatism, as distinct from authoritarianism (Stenner, 2005), asking respondents to indicate their views on the statement ‘Things in Britain were better in the past’. Response categories run from 1 = strongly agree to 5 = strongly disagree.

Treatments

Thereafter, respondents are randomly allocated into three groups: two treatments and a control condition. Treatment groups read one of the two vignettes in Appendix 3 before answering three questions on immigration. The vignettes are designed to be realistic alternative national identity narratives which political parties or media commentators could potentially adopt. As such, treatments are designed to convey a holistic, positive message rather than merely maximising the manipulation of the outcome variable or deploying a conjoint strategy. The two narratives tell very different stories about the relationship between immigration and majority ethnic boundaries. The first treatment, the ‘diversity prime’, stresses change, urging ethnic majority respondents to embrace it, accept their group’s ethnic decline, and focus instead on the ethnically neutral civic nation-state as the embodiment of their collective identity:

Britain is changing, becoming increasingly diverse. The 2011 census shows that White British people are already a minority in four British cities, including London. Over a quarter of births in England and Wales are to foreign-born mothers. Young Britons are also much more diverse than older Britons. Just 4.5 percent of those older than 65 are nonwhite but more than 20 percent of those under 25 are. Minorities’ younger average age, somewhat higher birth rate and continued immigration mean that late this century, according to Professor David Coleman of Oxford University, White British people will be in the minority nationwide. We should embrace our diversity, which gives Britain an advantage in the global economy. Together, we can build a stronger, more inclusive Britain.

The second, ‘assimilation prime’, stresses continuity over change, seeking to reassure White respondents that immigration will leave the boundaries and size of the majority group unaffected. This is intended to prime an open ethnic nationalism rather than either civic nationalism or exclusive ethnic nationalism:

Immigration has risen and fallen over time, but, like the English language, Britain’s culture is only superficially affected by foreign influence. According to Professor Eric Kaufmann of the University of London, a large share of the children of European immigrants have become White British. Historians tell us that French, Irish, Jews and pre-war black immigrants largely melted into the white majority. Those of mixed race, who share common ancestors with White British people, are growing faster than all minority groups and 8 in 10 of them marry whites. In the long run, today’s minorities will be absorbed into the majority and foreign identities will fade, as they have for public figures with immigrant ancestors like Boris Johnson or Peter Mandelson. Britain shapes its migrants, migration doesn’t shape Britain.

Notice that the treatments are based on the view that national identity is a complex system in which different actors perceive the nation differently. This means testing for disparate impacts of the vignettes across a fine-grained set of group characteristics. Given the need for a realistic set of national/policy narratives, and the importance of achieving adequate degrees of freedom to test small groups (i.e. UK Independence Party (UKIP) voters, tabloid readers), this article does not adopt a conjoint technique nor does it test fragments of the narratives – though this is a potential route for future researchers to follow. Randomisation tests in Appendix 2 show that treatment groups do not differ from control groups across key demographic or attitudinal characteristics.²

Dependent Variables

The first outcome measure is a standard policy preference item which asks, ‘Do you think the number of immigrants coming to Britain nowadays should be increased, reduced or should it remain the same?’ Answers are arrayed on a 1–5 Likert scale from ‘increase a lot’ to ‘reduce a lot’ with ‘don’t know’ responses removed. The second, measuring the salience of immigration, reads, ‘What do you see as the most important issue facing Britain today?’ Respondents select from 10 options. One of these is immigration. Those selecting immigration as their most important concern are coded 1 and all others 0 to create an immigration salience dummy.

To get at more current expressions of anti-immigration populism, British respondents are asked a CV question designed to probe their willingness to trade access to the economic benefits of the European Single Market against openness to the free movement of people from the EU to Britain. As one is currently viewed by the EU as a condition for the other, there is a debate between proponents of ‘Hard Brexit’, who prefer border control and sharply reduced EU immigration even if this means losing Single Market access – and thus revenue, and ‘Soft Brexit’, whereby access to the Single Market is retained in exchange for permitting the free movement of people from the EU. This is, for example, the relationship Norway has with the EU.

Rather than a more abstract policy question, this item is designed to probe people’s willingness to make sacrifices in the event of a costly Brexit. This said, the question pertains to immigration and thus is expected to overlap with previous immigration questions given that nearly half the UK inflow stems from the EU. The CV question reads:

Roughly 185,000 more people entered Britain last year from the EU than went the other way. Imagine there was a cost to reduce the inflow. How much would you be willing to pay to reduce the number of Europeans entering Britain?

Respondents are asked what proportion of their income they would be willing to pay, from nothing, in which case numbers remain at 185,000, to 5%, which would reduce numbers to zero. Each additional percentage of income foregone reduces EU migration by 35,000.

For all questions, a ‘don’t know’ option is provided. For the authoritarianism, conservatism and immigration questions, the order of responses is randomised to prevent response-order bias. Unless indicated, I use logistic regression to analyse immigration salience and ordered logistic regression for the immigration attitudes and CV measures.

Results

I begin by examining the effect of the assimilation and diversity primes on immigration opinion, salience and the CV (EU immigration trade-off) question in Britain. The focus is on the White British ethnic majority as the vignettes are designed around the assumption that conservative and authoritarian members of the ethnic majority are concerned about maintaining symbolic boundaries and numerical preponderance. Figure 1 shows the percentage in each treatment group who favour reduced immigration as opposed to current or higher levels. Those reading the story about ethnic assimilation are six points less likely than those reading the diversity vignette and four points less likely than those in the control group to want lower levels of immigration. A chi-squared test of the three conditions cross-tabulated with immigration attitudes is significant at the p < 0.05 level. However, further comparisons reveal that the only statistically significant difference is between those receiving the assimilation prime and the rest.

Figure 1.

Percentage in Each Treatment Group Saying Immigration Should Be Reduced (White British Respondents; N = 1353).

When we expand the binary immigration reduction variable into its full 5-point form, from ‘increase a lot’ (1) to ‘reduce a lot’ (5), the White British mean score on the immigration levels question is 4.06 for the control group, 4.09 for the diversity prime and 3.99 for the assimilation prime. Across the 5-point immigration levels variable, the difference between those receiving the assimilation and diversity primes is strongly significant (p = 0.001) as is the difference between the assimilation treatment group and the control group (p = 0.006). This appears to be one of the first cases of where an experimental treatment has resulted in a significant liberal shift in Whites’ immigration policy attitudes (for discussion, see Hopkins et al., 2016).

This offers weak confirmation of H1 regarding the diversity prime, but strong confirmation for H2 – that assimilation primes will shift attitudes in a liberal direction – thereby refuting the claim that strong beliefs cannot be experimentally manipulated. Yet, the effect does not significantly vary between White liberals, conservatives, libertarians or authoritarians. This thereby fails to disprove the null hypothesis for H3 and H4.

I next consider the question of Hard versus Soft Brexit through the CV measure for Hard Brexit support. The Hard Brexit proxy question is phrased as a trade-off between European immigration and personal prosperity, with each 1% of income ‘purchasing’ a decline in European immigration of 35,000. Of the White British respondents who answered the question, 60% would rather accept the then-current level of 185,000 EU (net) immigrants per year than sacrifice any of their income to reduce EU immigration. Figure 2 reveals that over 80% of White British individuals in the survey would prefer to accept current migration levels than to pay 5% of their income. In all, 60% would not pay anything to reduce numbers, rising to 62% when including minorities, who were more likely to vote to remain in the EU. The sample is skewed 56–44 towards Remain compared to the actual 48–52 result, thus understating the population-level willingness to pay to reduce immigration. Yet even with Brexit weighting, a majority is unwilling to pay anything to lower EU immigration.

Figure 2.

Willingness to Pay 5% of Income to cut EU Migration (White British Respondents; N = 982).

Past research using the CV method suggests the share willing to pay should be considered an upper bound, with participants often unwilling to contribute as much in reality as they state on surveys (Loomis and White, 1996). Even taking into account the underrepresentation of Leave voters in the survey, this indicates that the centre of gravity of British opinion leans closer to the Soft Brexit position.³ Even among White British respondents who voted Leave, nearly 30% indicated they would not be willing to sacrifice any of their income to reduce European immigration. Again, this suggests a considerable ‘soft’ component to the Brexit vote.

At the other end of the spectrum, however, 26% of the sample indicated they would contribute at least 3% of their income to reduce European migration, rising to 54% among those voting Leave. Indeed, 35% of Leave voters said they were willing to pay 5% of their income to cut EU immigration to zero. Coding responses from 1, unwilling to pay anything, through 6, willing to pay 5%, yields an ordinal variable for analysis. Given the non-normally distributed nature of this measure, I have also transformed it into a dummy variable, with the willingness to pay a full 5% of income coded 1 and other responses 0.

I find White British respondents who read the assimilation prime become about a half standard deviation less likely to endorse the hard Brexit proxy than those in the control or diversity treatment groups. As Figure 3 shows, the share willing to pay 5% falls from 18.8% in the control group to 11.8% with the assimilation treatment, which is significant in a chi-squared test at the p < 0.05 level. This again confirms H2 – regarding the predicted response to the assimilation vignette – while H1 and H4 (which predict the diversity prime will enhance opposition to immigration) are refuted.

Figure 3.

Willingness to Pay 5% for No EU Migrants, by Subgroup.

However, most liberal-minded people are already unwilling to pay any of their income to reduce EU migration. Indeed, as Figure 3 shows, it is conservative and authoritarian respondents who respond most to the assimilation prime by softening their willingness to pay to halt EU migration. Thus, UKIP, authoritarian or lower status (C2, D and E occupations) respondents are more affected by the assimilation prime than the more liberal, libertarian or elite group. This backs H3, contrary to the results for immigration levels in Figure 2 where there were no differences between liberals and conservatives.

For instance, of the 145 White British people who voted for the UKIP in 2015, a majority – 55.3% – of the control group are willing to pay 5%, but just 32.4% of the assimilation treatment group are. Thus, the model in Figure 4 shows that among White Britons, UKIP voters are predicted to be 25 points more likely than non-UKIP voters to pay 5% to reduce EU immigration to zero. Within the assimilation treatment group, UKIP and non-UKIP White voters differ by just 11 points compared to a 33-point UKIP/non-UKIP divide on the 5% question within the other treatment groups. The interaction between the assimilation treatment and UKIP voting on this Hard Brexit measure is significant and negative at the p < 0.05 level. This suggests that priming open ethnic nationalism brings UKIP voters more in line with the mainstream of White British opinion.

Figure 4.

Assimilation Prime and Hard Brexit Support (UKIP vs Non-UKIP; N = 1485).

For Brexit voters, 40.9% of the control group are willing to pay 5% to end EU immigration, but this falls to 26.5% for the assimilation treatment group. Among the 332 White British respondents in the C2, D and E census social grades (skilled and unskilled working class and unemployed), the assimilation vignette reduces the willingness to pay 5% of income from 31.2% to 13.9%, which is significant at the p < 0.01 level. These are dramatic effects, especially when set in the context of previous survey work on immigration attitudes (i.e. Druckman and Leeper, 2012; Hopkins et al., 2016). The policy takeaway is that those trying to convince the conservative section of the British public of the merits of trading away reduced migration for greater access to EU markets can benefit from making the case that European immigrants are likely to assimilate and become ‘us’, now as in the past.

These results offer partial support for the view that primes are operating in the expected – different – directions, insofar as those reading the assimilation prime respond differently from those reading the diversity prime. However, the diversity vignette does not produce a significant increase in the Hard Brexit proxy, even among conservatives and authoritarians. One possibility is that respondents have already internalised a narrative of demographic decline, hence the diversity vignette adds little to their perceptions. Indeed, the British public believes that EU immigrants make up 15% of Britain’s population. Leave voters estimate this figure to be as high as 20%, compared to the actual figure of 5% (Duffy and Menon, 2016).⁴

Moving to the salience of immigration, our final dependent variable, it is readily apparent that those who cite immigration as their top concern are drawn almost entirely (97%) from among those who want less immigration. We saw that the assimilation prime reduces opposition to immigration, but does it similarly lower the salience of this issue among immigration opponents? Only weakly. Among White British in the sample who want immigration reduced a lot, the share citing immigration as their most important concern is 45% for the diversity treatment, 40% for the assimilation treatment and 42% for the control group. Results are in the right direction but not statistically significant. One possibility is that ceiling effects are limiting the impact of the treatments in raising awareness of immigration among anti-immigration respondents because half of anti-immigration White Britons already rate it as their top issue.

Relevance for Public Policy

Models of reported Brexit vote, shown in Table 1, confirm findings from the academic literature which claim that values and cultural issues are more important than personal economic circumstances in accounting for opposition to immigration and support for the populist right (Mudde, 2007; Sniderman et al., 2004; Hainmueller and Hopkins, 2014, 2015; Inglehart and Norris, 2016).⁵

Table 1.

Models of Reported Brexit Vote, White British Respondents.

	Model 1: Past Vote	Model 2: Issues and Attitudes	Model 3: Composite
Age	0.021*** (0.005)	0.000 (0.007)	0.019*** (0.005)
Education	−0.103*** (0.019)	−0.249* (0.100)	−0.070*** (0.020)
Income (personal)	0.012 (0.021)	−0.067** (0.025)	0.011 (0.030)
Female	0.353 (0.232)	0.047 (0.139)	0.341 (0.249)
2015 vote (ref.: green party)
Conservative	1.931*** (0.411)		0.781 (0.527)
Labour	0.703^‡ (0.385)		−0.173 (0.628)
Liberal democrat	0.809 (0.639)		0.027 (0.833)
Plaid Cymru	1.586*** (0.411)		−0.386 (0.535)
UK Independence Party	3.849*** (0.349)		2.066*** (0.540)
Don’t know	1.345** (0.533)		2.382* (1.048)
Whip sex criminals publicly		−0.105 (0.066)	0.140 (0.089)
Left-right self-placement		0.427*** (0.068)	0.084^‡ (0.044)
Things in Britain better in the past		0.361** (0.118)	0.413** (0.135)
Reduce immigration a lot		1.223*** (0.186)	0.985** (0.354)
Immigration top issue		1.094** (0.330)	0.900* (0.463)
Percentage of professionals among White British (ward)	−0.143 (2.119)		−0.002 (0.007)
White British average education (local authority)	−0.840 (0.671)		1.697 (2.047)
Percentage of White British (ward)	1.345* (0.533)		−0.226 (1.040)
Constant	0.420 (1.468)		−3.716 (2.461)
N	673	728	505
Pseudo R²	0.235	0.336	0.380

Source: Own data (YouGov) 2016.

Restricted to White British respondents. Dummy variable for reported Leave vote. Logistic regressions with standard errors clustered on UK region. Listwise deletion due to failure to match some wards and loss of ‘don’t know’ responses on values questions. White British average education not available at ward level.

‡

p < 0.1; *p < 0.05; **p < 0.01; ***p < 0.001.

In view of the association between cultural attitudes and populist right support, economic policies, such as Britain’s Controlling Migration Fund – designed to provide resources to districts such as Boston in Lincolnshire which are coping with a large influx (and which delivered the strongest Leave vote) – may prove ineffective (Department for Communities and Local Government (DCLG), 2016). After all, only a handful of the 335 districts of England and Wales have experienced large-scale immigration like Boston. Likewise, Centre-Left parties may have difficulty addressing the immigration question with economic policy proposals. This article therefore suggests that political communications which reframe the relationship between immigration and national identity stand a better chance of addressing the cultural grievances driving the populist right.

Conclusion

This article finds that priming open ethnic nationalism through an assimilation treatment reduces White British opposition to immigration and lowers White conservatives’ willingness to pay for Hard Brexit. The novelty of the assimilation message in Britain and its absence of liberal connotations may help to account for its ability to shift conservative opinion. This suggests that changing the way immigrants are perceived in relation to the ethno-national ‘we’ is important for addressing the concerns of conservative Whites which underpin right-wing populism. This work adds a new perspective to the growing literature on the relationship between national identity and immigration attitudes. It advances our understanding of public opinion on immigration and the sources of right-wing populism. Future work could employ a conjoint method to see which parts of the assimilation narrative contribute most to the effects noticed here. In addition, the study might be replicated in cases outside Britain.

How might the results of this survey experiment inform the policy process? The vignettes are designed as realistic forms of political communication that policymakers, politicians or the media may adopt. The narrative treatments show that priming a more open form of ethnic nationalism is an effective strategy for reassuring conservative White voters. This means the assimilation narrative has the potential to bring conservative British opinion towards the centre ground, reducing the political divisions that produced the Brexit vote. Liberals, Soft Brexit campaigners and free-market conservatives in Britain should think seriously about changing their political communications with conservative White British audiences. The current Soft Brexit narrative, which champions the merits of diversity and economic openness as key features of a civic British nationalism, is, as Stenner (2005: 331) might have predicted, failing to carry authoritarian and conservative audiences. A more promising approach could be to reserve this account for liberal or minority constituencies while developing an assimilationist ‘things remain the same’ appeal to conservative ‘Middle Britain’. When communicating with the nation more broadly, a ‘constructively ambiguous’⁶ form of national imagining, which provides room for liberals and conservatives to read their chosen narrative into the script, is preferable.

By subtly altering the boundaries of ‘us’ to include immigrants who wish to assimilate, this strategy ministers to authoritarian concerns (Stenner, 2005: 328). Against this, populist right parties such as the UK Independence Party will no doubt wish to keep the diversity narrative at the centre of their messaging in order to maximise ethno-nationalist concerns.

Footnotes

Acknowledgements

The development of this article benefited from comments received at a Nuffield College Oxford Political Science Seminar.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: This work was supported by the Economic and Social Research Council (ESRC), award ref# ES/K003895/1.

Supplementary Information

Additional supplementary information may be found with the online version of this article.

Appendix 1. Summary Statistics for Variables

Appendix 2. Randomisation Verification

Appendix 3. Questionnaire

Appendix 4. YouGov Weighting Data

Notes

Author Biography

Eric Kaufmann is Professor of Politics at Birkbeck College, University of London. He is author of Changing Places: Mapping the White British Response to Ethnic Change (Demos, 2014), Shall the Religious Inherit the Earth (Profile, 2010), The Rise and Fall of Anglo-America (Harvard, 2004) and two other books. He is co-editor, among others, of Political Demography (Oxford, 2012) and editor of Rethinking Ethnicity: Majority Groups and Dominant Minorities (Routledge, 2004). He is an editor of the journal Nations and Nationalism.

References

Alba

(2005) Bright vs. Blurred Boundaries: Second-Generation Assimilation and Exclusion in France, Germany, and the United States. Ethnic and Racial Studies 28 (1): 20–49.

Barth

(1998) Ethnic Groups and Boundaries: The Social Organization of Culture Difference. Waveland Press.

Brader

Valentino

Suahy

(2008) What Triggers Public Opposition to Immigration? Anxiety, Group Cues, and Immigration Threat. American Journal of Political Science 52 (4): 959–978.

Breton

(2015) Making National Identity Salient: Impact on Attitudes toward Immigration and Multiculturalism. Canadian Journal of Political Science 48 (2): 357–381.

Brubaker

(2006) Nationalist Politics and Everyday Ethnicity in a Transylvanian Town. Princeton, NJ: Princeton University Press.

Charnysh

Lucas

Singh

(2015) The Ties That Bind. Comparative Political Studies 48 (3): 267–300.

Citrin

Sides

(2008) Immigration and the Imagined Community in Europe and the United States. Political Studies 56: 33–56.

Citrin

Wright

(2009) Defining the Circle of We: American Identity and Immigration Policy. The Forum 17 (3).

Citrin

Reingold

Green

(1990) American Identity and the Politics of Ethnic Change. The Journal of Politics 52 (4): 1124–1154.

10.

Citrin

Wong

Duff

(2001) The Meaning of American National Identity. In: Ashmore

Jussim

Wilder

(eds) Social Identity, Intergroup Conflict, and Conflict Reduction. New York: Oxford University Press, pp.71–100.

11.

Clarke

et al . (2017) Brexit: Why Britain Voted to Leave the European Union. Cambridge University Press.

12.

Coenders

MTA

(2001) Nationalistic Attitudes and Ethnic Exclusionism in a Comparative Perspective: An Empirical Study of Attitudes toward the Country and Ethnic Immigrants in 22 Countries. Nijmegen: Radboud University.

13.

Craig

Richeson

(2014) More Diverse yet Less Tolerant? How the Increasingly Diverse Racial Landscape Affects White Americans’ Racial Attitudes. Personality and Social Psychology Bulletin 40 (6): 750–761.

14.

Department for Communities and Local Government (DCLG) (2016) Controlling Migration Fund: Mitigating the Impacts of Immigration on Local Communities. Prospectus, Department for Communities and Local Government, London, November.

15.

Druckman

Leeper

(2012) Learning More from Political Communication Experiments: Pretreatment and Its Effects. American Journal of Political Science 56 (4): 875–896.

16.

Duffy

Menon

(2016) The Perils of Perception and the EU: Public Misperceptions about the EU and How It Affects Life in the UK. London: Ipsos-MORI. Available at: https://www.ipsos-mori.com/researchpublications/researcharchive/3742/the-perils-of-perception-and-the-eu.aspx (accessed 12 February 2017).

17.

Duffy

et al . (2015) Shifting Ground: Changing Attitudes to Immigration in the Long Term and During Election Campaigns. IMSR Institute. London: Ipsos MORI.

18.

Evans

Mellon

(2016) How Immigration Became a Eurosceptic Issue. In: London School of Economics Brexit Blog. Available at: http://blogs.lse.ac.uk/brexit/2016/01/05/how-immigration-became-a-eurosceptic-issue/ (accessed 5 January 2017).

19.

Feldman

Stenner

(1997) Perceived Threat and Authoritarianism. Political Psychology 18 (4): 741–770.

20.

Fox

(2014) National Holiday Commemorations: The View from below. In: Tsang

Woods

(eds) The Cultural Politics of Nationalism and Nation-Building. Abingdon; New York: Routledge, pp.38–52.

21.

Francis

(1976) Interethnic Relations: An Essay in Sociological Theory. New York: Elsevier Scientific.

22.

Goodhart

(2013) The British Dream: Successes and Failures of Post-War Immigration. London: Atlantic Books.

23.

Gordon

(1964) Assimilation in American Life: The Role of Race, Religion and National Origins. New York: Oxford University Press.

24.

Hainmueller

Hopkins

(2014) Public Attitudes toward Immigration. Annual Review of Political Science 17: 225–249.

25.

Hainmueller

Hopkins

(2015) The Hidden American Immigration Consensus: A Conjoint Analysis of Attitudes Toward Immigrants. American Journal of Political Science 59 (3): 529–548.

26.

Heath

Tilley

(2005) National Identity and Attitudes towards Migrants: Findings from the ISSP. International Journal on Multicultural Societies 7 (2): 119–132.

27.

Higham

(1955) Strangers in the Land: Patterns of American Nativism, 1860–1925. New Brunswick, NJ: Rutgers University Press.

28.

Hopkins

Sides

Citrin

(2016) The Muted Consequences of Correct Information about Immigration, 18 June. Available at: https://ssrn.com/abstract=2798622 (accessed 18 February 2017).

29.

Hutchinson

(2005) Nations as Zones of Conflict. London: SAGE.

30.

Inglehart

Norris

(2016) Trump, Brexit, and the rise of Populism: Economic Have-Nots and Cultural Backlash. HKS Working Paper No. RWP16-026. Harvard Kennedy School.

31.

Jardina

(2014) Demise of Dominance: Group Threat and the New Relevance of White Identity for American Politics. PhD Dissertation, University of Michigan, Ann Arbor, MI.

32.

Kaufmann

(2004) Dominant ethnicity: from Background to foreground. In: Kaufmann

(ed.) Rethinking Ethnicity: Majority Groups and Dominant Minorities. London: Routledge, pp.1–14.

33.

Kaufmann

(2016) Complexity and Nationalism. Nations and Nationalism 23 (1): 6–25.

34.

Lawrence

Sides

(2014) The Consequences of Political Innumeracy. Research & Politics. Epub ahead of print 13 August. DOI: 10.1177/2053168014545414.

35.

Loomis

White

(1996) Economic Benefits of Rare and Endangered Species: Summary and Meta-Analysis. Ecological Economics 18 (3): 197–206.

36.

Louis

Esses

Lalonde

(2013) National Identification. Perceived Threat, and Dehumanization as Antecedents of Negative Attitudes toward Immigrants in Australia and Canada. Journal of Applied Social Psychology 43 (S2): E156–E165.

37.

Major

Blodorn

Blascovich

(2016) The Threat of Increasing Diversity: Why Many White Americans Support Trump in the 2016 Presidential Election. Group Processes & Intergroup Relations. Epub ahead of print 20 October. DOI: 10.1177/1368430216677304.

38.

Modood

(2013) Multiculturalism. Cambridge: Polity Press.

39.

Mudde

(2007) Populist Radical Right Parties in Europe. Cambridge; New York: Cambridge University Press.

40.

NatCen Social Research (2016) British Social Attitudes Survey, 2014 (Data collection). 2nd edn. UK Data Service. SN: 7809. Available at: https://dx-doi-org.web.bisu.edu.cn/10.5255/UKDA-SN-7809-2 (accessed 18 January 2017).

41.

Park

Curtice

Bryson

(eds) (2014) British Social Attitudes: The 31st Report. London: NatCen Social Research. Available at: http://bsa-31.natcen.ac.uk/ (accessed 18 January 2017).

42.

Pehrson

Vignoles

Brown

(2009) National Identification and Anti-Immigrant Prejudice: Individual and Contextual Effects of National Definitions. Social Psychology Quarterly 72 (1): 24–38.

43.

Perez

Hetherington

(2014) Authoritarianism in Black and White: Testing the Cross-Racial Validity of the Child Rearing Scale. Political Analysis 22 (3): 398–412.

44.

Powell

(2008) What I Learned in Belfast. Prospect, 24 May. Available at: https://www.prospectmagazine.co.uk/magazine/whatilearnedinbelfast.

45.

Rose

(2011) What Makes People Tick: The Three Hidden Worlds of Settlers, Prospectors and Pioneers. Leicester: Matador.

46.

Schildkraut

(2014) Boundaries of American Identity: Evolving Understandings of ‘Us’. Annual Review of Political Science 17: 441–460.

47.

Smith

(1991) National Identity. London: Penguin Books.

48.

Smith

(1997) Civic Ideals: Conflicting Visions of Citizenship in U.S. History. New Haven, CT: Yale University Press.

49.

Sniderman

Hagendoor

Prior

(2004) Predisposing Factors and Situational Triggers: Exclusionary Reactions to Immigrant Minorities. American Political Science Review 98 (1): 35–49.

50.

Stenner

(2005) The Authoritarian Dynamic. Cambridge: Cambridge University Press.

51.

Wimmer

(2007) The Making and Unmaking of Ethnic Boundaries: Toward a Comparative Theory. New Haven, CT: Yale Center for Comparative Research.

52.

Wong

(2010) Boundaries of Obligation in American Politics: Geographic, National, and Racial Communities. Cambridge; New York: Cambridge University Press.

53.

Wright

Citrin

(2010) Saved by the Stars and Stripes? Images of Protest, Salience of Threat, and Immigration Attitudes. American Politics Research 39: 323–343.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.04 MB