Thinking Globally about Attitudes to Immigration: Concerns about Social Con ﬂ ict, Economic Competition and Cultural Threat

This article looks globally at the motivations behind attitudes to immigration. Such motivations have been typically conceptualised by academics either in terms of the ‘ economic competition ’ or ‘ cultural threat ’ that immigrants are perceived to pose to the individual or their ‘ in-group ’ . We propose and test a third possibility: that support for or opposition to immigration is determined by one ’ s perceptions of immigration ’ seffectsonsocialcon ﬂ ict.Using the 2017 – 2020WorldValuesSurvey (WVS)forforty-nine countries, we show that: in most countries, globally, citizens are more likely to agree than disagree that immigration leads to social con ﬂ ict; levels of concern about the effects of immigration on social con ﬂ ict arehigherthanthoseregardingunemploymentorcultureinsixteen — disproportionatelyeconomically developed — countries; concern about social con ﬂ ict is conceptually and distributionally distinct; belief that immigration leads to social con ﬂ ict predicts immigration policy preferences; but, uniquely, is positively predicted by higher education. Our ﬁ ndings highlight the importance of institutional con ﬂ ict resolution capacity, including those related to integration, for the politics of migration.

WHILE THERE HAS been a significant increase in scholarly explanations for variation in attitudes to immigration in recent years, this has tended to focus on a relatively small group of major destination countries-typically, the USA, (Western) Europe and Australia. 1 Furthermore, explanations for support or opposition to immigration have tended to coalesce around two theoretical strands: those that consider the effect of greater immigration on (in both cases, perceived or real) 'economic competition' or 'cultural threat'. In this article, the geographic focus is broadened to consider the drivers of attitudes globally. In doing so, we identify a novel and neglected third factor in attitudinal formation: the perceived effect of immigration on intergroup societal conflict. To do so, we use data from the 2017-2020 World Values Survey (WVS) to demonstrate both the distribution and distinctiveness of the perception that immigration leads to social conflict.
Moreover, we theoretically link our insights to broader political outlooks, values and, as a result, attitudes to other issues. Substantively, our findings matter most obviously because migration will remain one of the world's most important political challenges of the twenty-first century, having important economic consequences and raising profound legal-and rights-based questions for millions far beyond the typically studied 'Western' democracies. Public attitudes are a crucial component of this challenge because they increasingly constitute the major parameter for policy makers and engender highly charged and complex political questions of identity, values and community. Our findings will be of practical use to advocacy organisations, governments, communicators, politicians and policy makers aiming to create a sustainable migration policy framework, communicate on migration matters effectively and design policies-notably in terms of integration, in addition to economic or cultural 1 1 policy, as recently suggested-that can generate public consent. 2 Finally, the global focus of our analysis is timely for three reasons. First, the relative scarcity of such analyses inevitably limits the confidence we can have in the generalisability of extant conceptual and theoretical frameworks. Second, there are increasingly high quality data sources that facilitate the development of analyses that are both comparative and global in focus. Although we use the WVS, regional barometers also offer the potential for cross country and time series analyses. Third, our analysis is pertinent for contemporary policy initiatives at the global level, such as the Global Compact on Migration (GCM) and the Global Compact on Refugees (GCR). GCM Objective 17, for example, identifies the need to 'promote evidence-based public discourse to shape perceptions of migration'. Identified tasks include tackling discrimination and hostility, providing migrants with means of redress and educating media professionals about migration. Strikingly absent from the GCM and GCR are considerations of how exactly perceptions are formed, distributed and, therefore, might be changed. An enhanced understanding of the factors that shape attitudes and perceptions is important to any discussion of 'evidence based public discourse' and its effects, not least because many of the factors that shape political attitudes are deeply rooted in value orientations, making them somewhat resistant to change and, in some cases, upheld by cognitive biases that can undermine or invert the intended effects of interventions. 3 The article continues as follows. First, we review the literature on attitudes to immigration, noting the tendency to theorise opposition to immigration as the result of either (perceived or real) 'economic competition' or 'cultural threat'. We then offer our theoretical proposition regarding the effect of a perception that immigration leads to social conflict and offer reasons as to why it is plausible and distinct. We then provide an overview of our data source, the WVS (2017-2020), before presenting the distribution of citizens agreeing that 'immigration leads to social conflict' across forty-nine countries and then demonstrating its distinctiveness from cultural anxieties. Finally, we present analyses at both the national and individual levels to show that the determinants of 'social conflict anxiety' are partially distinct.

Explanations for variation in attitudes to immigration
An extensive literature considers the determinants of support for or opposition to immigration. Co-variates considered range from deepseated and early-life socialisation factors, such as societal heterogeneity and education, to intermediary psychological drivers, such as personal values, to shorter-term explanations regarding cueing and contact with immigrants. 4 However, 2 E. G. T. Green, E. P. Visintin, O. Sarrasin and M. Hewstone, 'When integration policies shape the impact of intergroup contact on threat perceptions: a multilevel study across 20 European countries', Journal of Ethnic and Migration Studies, vol. 46, no. 3, 2020, pp. 631-648.  in terms of explaining the direct motivation for why some individuals oppose immigration, explanations have typically fallen into two, not necessarily mutually exclusive, strands of 'economic competition' and 'cultural threat'. On the one hand, the economic competition theory sees opposition to immigration as the result of real or perceived negative effects of immigration on individual material wellbeing or the material wellbeing of one's group. In this sense, it draws directly from 'realistic conflict theory', and more broadly 'group conflict theory', which see over-time increases in hostility to immigration as resulting from an increase in the size of the out-group and worsening economic conditions.
On the other hand, the cultural threat approach argues that individuals oppose immigration because they fear it will lead to change in their society's cultural traditions, religion, language, and so on. 5 To a lesser extent, this approach is also based on realistic conflict theory in that individuals oppose immigration to preserve the cultural hegemony of their 'in-group', that is, the majority, 'native' population, and as such their social status, with material implications. However, often the mechanism is left simply in the emotional terms of 'anxiety' or 'fear', with culture treated as a basket of non-economic preferences from which wellbeing is derived. In general, scholars have tended to find greater evidence in support of the 'cultural threat' strand-though recent studies cited above go some way to reaffirm the importance of economic considerations-and, together, have also been considered as explanations for radical right or populist voting, again with 'cultural threat' being given greater support.
We argue that there are compelling reasons to believe that those opposed to immigration may also be motivated by a third consideration-the perception that immigration increases the risk of social conflict-and that this motivation is distinct from both economic and cultural concerns. We should first note that the three motivations are not mutually exclusive, either in the capacity for individuals to hold them, or as correct explanations. However, as we show, the effect on opposition to immigration of the widespread perception that immigration leads to social conflict is distinct in important ways. First, individuals need not be motivated by a desire to 'win' conflict between their 'in-group' and 'out-groups' (immigrants) in terms of either economic wellbeing, cultural preferences or social status, but instead simply prioritise the avoidance of social conflict in sociotropic terms, which they see immigration as potentially causing. According to this mechanism, it is conceivable that individuals who stand to gain economically by immigration may oppose it and vice versa, as has been repeatedly found, and it is also conceivable that those whose culture deviates from the majority-group's culture (including immigrants and their descendants) may oppose immigration. As such, whereas the 'economic competition' and 'cultural threat' schools envisage immigration preferences as derived from group membership in a context of widely held group identities, conflict between those groups, and the pursuit of selfinterest, our theoretical proposition is derived more from psychological findings regarding the value basis of immigration preferences, notably in terms of the negative effect on support for immigration of valuing security, conformity and order, either in personal or political terms.
There are at least three reasons to think that individual immigration preferences may be affected by the perceived risk of social conflict. First, people with conservative political value orientations have repeatedly been shown to be more opposed to immigration. 6 philosophical concern of conservatism is the maintenance of societal order, which is deemed essential to wellbeing and progress, as well as fragile and difficult to reimpose once broken. As such, enduring societal arrangements-including its demographic composition-are particularly valued by conservatives because of their, arguably, tested functionality, in terms of social conflict resolution and, where necessary, collective action in relatively harmonious terms. Whereas socialists and liberals are more likely to conceive immigration as a means of, respectively, achieving greater human equality or enhancing personal freedom, conservatives are more likely to be wary of the risks of such a fundamental change to society as its demographic composition. This is not only because immigration may lead to undesired economic or cultural changes, but also, more profoundly, because of a perception that increased societal heterogeneity and complexity resulting from immigration makes collective action and resolving potential social conflicts more difficult. Indeed, as the conclusion to this article discusses, some of the most famous interventions against immigration by conservative politicians have been squarely focussed on predictions of intergroup, potentially bloody, social conflict, rather than on more immediate issues of economic competition and cultural threat. However, the association between the above political values and attitudes to immigration may be limited outside of those-essentially 'Western'-political cultures in which values are expressed most frequently in the above terms. This leads us to a second reason.
A number of personal, rather than political, values-'stable motivational constructs or beliefs about desirable end states that transcend specific situations and guide the selection or evaluation of behaviours and events'-have been associated with attitudes to immigration in a manner that suggest a concern about social conflict motivating immigra-tion policy preferences. 7 Notably, these include four of Schwartz's ten universal, basic human values-with valuing tradition, conformity and security increasing opposition to immigration and valuing universalism reducing it. 8 The motivations behind security, conformity and tradition values are, respectively: safety and stability of society; restraint of actions that may upset others or violate social expectations and norms; commitment to customs that culture or religion provide. Whereas the 'cultural threat' theory of attitudes to immigration aligns with these findings-particularly in terms of valuing tradition-'economic competition' fits less obviously. Furthermore, our proposal that individuals opposed to immigration may be concerned about any disruption to social harmony and potential for social conflict fits well with findings regarding the effects of valuing security and conformity.
Third, while there is generally weak evidence to support the notion that immigration has adverse economic effects on native citizens and the effect on culture being 'weakened' or 'enriched' is perhaps too subjective to be easily testable, there is a large literature presenting evidence that greater ethnic fractionalisation in societies can lead to social conflict and hinder state consolidation, democracy and development. 9 The causal mechanism in most of this S. Huntingdon, 'Conservatism as ideology', American Political Science Review, vol. 51, no. 2. 1957, pp. 454-473. On conservatism and preference for extant arrangements, see E. Burke and L. G. Mitchell, Reflections on the Revolution in France, Oxford, Oxford University Press, 2009; R. A. Nisbet, 'Conservatism and sociology', American Journal of Sociology, vol. 58, no. 2, 1942, pp. 167-175. 7 T. Brosch and D. Sander, 'Neurocognitive mechanisms underlying value-based decision-making: from core values to economic value', Frontiers in Human Neuroscience, vol. 7, 2013, p. 398. literature supposes that greater ethnic fractionalisation leads to weaker social cohesion, which we might also expect to result from immigration. As such, there may be an empirical as well as ideological justification for a belief that immigration leads to social conflict, albeit a contingent and contested one.
Global data on attitudes to immigration Prominent social surveys have often asked respondents about the extent to which they believe that immigration has certain effects. For example, the European Social Survey, which is one of the most utilised source of data for understanding attitudes to immigration, has asked respondents across Europe every two years about the effects of immigration on the economy, culture and quality of life; while two special immigration modules also asked about its effects on jobs, government accounts and crime. However, to our knowledge, until recently, no social survey has measured belief that immigration leads to social conflict. That has changed since the WVS Wave 7 (2017-2020), for the first time, included questions on the perceived effects of immigration, asking respondents whether they agree or disagree that, in their country, immigration: 'strengthens cultural diversity'; 'increases unemployment'; and 'leads to social conflict'. The survey also asks respondents about other, either similar or less theoretically relevant, perceived effects. At the time of writing, the data from forty-nine countries from every continent had been released, each based on random probability sampling. 10 This data complements growing non-'Western' data on attitudes to immigration from regional barometers, such as the Afrobarometer, Arab Barometer and Latin Barometer, as well as from global private polling firms, notably Gallup.
The prevalence of the perception that immigration leads to social conflict Figure 1 provides an initial overview of how the forty-nine countries vary in the extent to which their citizens agree or disagree with the proposition that immigration leads to social conflict. Several trends are apparent. First, there is extremely high variation between countries in the proportions that agree and disagree. Second, as shown by 'total', more citizens in the world-to the extent that the WVS is representative-agree that immigration leads to social conflict in their country (49 per cent) than disagree (28 per cent). Third, there are some trends between continents: seven of the eight European countries are more in agreement with the proposition than the weighted global average and are more likely to agree than not agree, the only exception being tiny Andorra. By contrast, the three majority English speaking countries of the USA, Australia and New Zealand are less in agreement than the average. Seven of the eleven Latin American countries have a higher proportion in agreement than the global average, the three sub-Saharan African (Ethiopia, Nigeria, and Zimbabwe) and two South Asian countries (Bangladesh and Pakistan) are relatively ambivalent, while the twelve Central and West Asian and North African countries vary significantly with some found at either extreme. Overall, citizens are more likely to agree than disagree that immigration leads to social conflict in their country in thirty-eight of the forty-nine countries surveyed. The reverse is only true in the following countries (in declining order of net agreement): Indonesia, China, South Korea, the Philippines, Iran, Vietnam, Tajikistan, New Zealand, Andorra, Nigeria, and Zimbabwe. Regionally, this highlights the relative lack of agreement by citizens in East Asia (though see Hong Kong and Taiwan).
Moreover, we note the prevalence of the three broadly theorised perceptions. Agreement that immigration leads to social conflict is more common than agreement that it increases unemployment and disagreement effects of ethnic fractionalisation, see W. Merkel and B. Weiffen, 'Does heterogeneity hinder democracy? ', Comparative Sociology, vol. 11, 2012, pp. 387-421;A. Alesina and E. La Ferrara, 'Participation in heterogeneous communities', Quarterly Journal of Economics, vol. 115, no. 3, 2020, pp. 847-904;M. Bleaney and A. Dimico, 'Ethnic diversity and conflict', Journal of Institutional Economics, vol. 13, no. 2, 2017, pp. 357-378. 10 that it strengthens cultural diversity, respectively, in sixteen of the sampled countries: Andorra, Australia, Chile, Colombia, Germany, Hong Kong, Japan, Malaysia, Mexico, New Zealand, Nicaragua, Puerto Rico, Russia, Serbia, Thailand and the USA. These countries disproportionately have developed economies. In only four countries are citizens more likely to disagree that immigration strengthens cultural diversity than agree with the other two statements, all of which are in Asia: Indonesia, Iraq, the Philippines and South Korea. In the remaining twenty-eight countries, citizens are more likely to believe that immigration has deleterious consequences for unemployment than for social conflict or culture. In one country-Guatemalaan equal number of respondents agreed with the statements about social conflict and unemployment.
The distinctiveness of the perception that immigration leads to social conflict With the high levels of global net agreement that immigration leads to social conflict demonstrated, we now consider the plausibility of our central theoretical proposition: that such a perception represents a distinctive third motivation for opposition to immigration, beyond cultural 'anxieties'. We do this in two steps. First, we demonstrate the perception's distinctiveness and, second, we demonstrate the extent to which agreement with the statement predicts preferred immigration policies.
To demonstrate the distinctiveness of perceptions of social conflict, we compare country-level agreement that immigration 'leads to social conflict' with agreement that it 'strengthens cultural diversity'. We do so for three reasons. First, it may be that individuals hold unidimensional views regarding immigration and so simply respond to the same extent either positively or negatively, regardless of the question. In this case, any empirically demonstrated relationship between perceptions and policy attitudes would not necessarily support our theoretical proposition-though it should be noted that previous studies have found attitudes to immigration to be multidimensional in the UK and the Netherlands. 11 Second, as mentioned above, the 'cultural threat' theory of immigration attitudes has been treated to some extent as a basket of non-economic 'anxieties' and 'fears', ranging from explicitly cultural issues such as language and traditions, to less obviously cultural factors such as values. In this case, one might argue that our proposition regarding fear of social conflict resulting from immigration simply falls into this broader extant theoretical strand, rather than being a third, distinct-though complementarystrand. Third, whereas the 'economic competition' theory is typically egocentric, in that it usually supposes that individuals oppose immigration for labour market or taxation reasons, the 'cultural threat' approach is arguably more sociotropic, in that it is usually concerned with maintaining the extant cultural order of the 'indigenous' majority (although this assumes that the opponent to immigration is a member of this group). In this sense, it is arguably closer to our theoretical proposition about fear of social conflict, despite the latter not being about culture, social status or other forms of relative self-or group-interest.
In Figure 2, below, we present the countrylevel correlation between the net percentage agreeing-defined as the percentage agreeing minus the percentage disagreeing-that immigration 'strengthens cultural diversity' and net percentage agreeing that immigration 'leads to social conflict'. We can see that there is essentially no correlation. Furthermore, in many countries there are large differences in the response rates, for example, whereas the net agreement amongst Germans that immigration 'strengthens cultural diversity' is 15 percentage points, the same figure for agreement that immigration 'leads to social conflict' is 75 percentage points. We can also see that in a large majority of countries, citizens simultaneously believe that immigration 'leads to social conflict' and 'strengthens cultural diversity', despite the former being a negative perception and the latter being positive. This strongly suggests that these are distinct concepts.
The predictive power of the perception that immigration leads to social conflict We now consider the predictive power on preferred immigration policy of agreement that immigration causes social conflict. After the battery of questions about perceived effects on immigration, the WVS (2017-2020) asks the following question: 'How about people from other countries coming here to work? Which one of the following do you think the government should do? (1) Let anyone come who wants to; (2) let people come as long as there are jobs available; (3) place strict limits on the numbers of foreigners who can come here; (4) prohibit people coming here from other countries.' Notably, both the question and one of the four responses explicitly mentions 'work' and 'jobs', arguably priming respondents to think in economic terms. Below, in Figure 3, we see the responses by country.
When the above immigration policy preferences at the individual level are predicted by, respectively, the three perceptions about the effects of immigration using separate multilogistic models with country fixed effects, each of the three perceptions have statistically significant effects and each of their respective three models explain a similar amount of variance in policy preferences, with R 2 s between 0.084 and 0.088. Again, it should be emphasised that both the preferred policy question and one of its responses explicitly mentioned 'work' and 'jobs'.

The predictors of the perception that immigration leads to social conflict
We now consider which individual-level variables predict agreement, rather than disagreement, with each of the three perceived effects of immigration, as shown in Table 1 Figure 2: National level correlation between net percentage agreeing that immigration 'strengthens cultural diversity' and net percentage agreeing that immigration 'leads to social conflict' positively associated with more negativity to the effects of immigration in every case. 12 However, whereas more education has a positive effect on belief that immigration strengthens culture, as shown elsewhere, it has non-linear effects on belief that immigration increases unemployment and, most notably, has positive effects on the belief that  immigration leads to social conflict. More predictably, individual income has a negative relationship with agreement that immigration leads to social conflict and increases unemployment. Also, surprisingly, being female is associated with greater belief that immigration increases unemployment and leads to social conflict.
The positive relationship between education and belief that immigration leads to social conflict possibly reflects the societywide focus of this form of concern-rather than protecting economic interests, which are less likely to be salient to the well-educated, or cultural traditionalism, which are also less common. It may also be that the positive relationship reflects the, arguably, better-supported empirical case for this concern, as described above, or that greater education increases one's awareness of conflicts, including those resulting from immigration, or increases the detachment from social ingroups needed to take an arguably more sophisticated, nuanced, or long-term critique of immigration. Regardless, the finding could provide an explanation for anti-immigration sentiment in more educated segments of society which, though less common, remain underexplored in the literature. 13 It may also reflect distinct dynamics globally in the formation of attitudes to immigration, compared to those in 'Western' countries, a possibility supported by the weak relationship between education and the perception that immigration increases unemployment.

Discussion
The primary motivations for opposition to immigration may not be limited to perceptions of 'economic competition' and 'cultural threat', Table 1: Three logistic regression models predicting agreement, rather than disagreement, that immigration (1) strengthens cultural diversity, (2) increases unemployment and (3)  Notes: Data source, World Values Survey, 2017-2020 unless otherwise stated. Weighted logistic models. Dependent variable treated as binary (1 = agree; 0 = disagree). Guatemala, Kyrgyzstan and China excluded from models for lack of variance/data. ***p < 0.001 **p < 0.01 *p < 0.05