1 What Are Public Attitudes Towards and the Politics of Immigration?

Identifying and describing attitudes to immigration, let alone explaining them, is not a simple matter. First, human attitudes in general are abstract and so any measurement of them is bound to be highly qualified and contingent on theoretical assumptions and methodological approaches. Second, immigration is a broad topic. Attitudes to immigration alone can be divided into attitudes towards immigrants, towards immigration policy, the perceived effects of immigration, or towards how important immigration is as an issue. Each of these can be divided by immigrant group as the most obvious qualifier. In this chapter we follow the political science literature in conceptualising attitudes as “people’s orientations toward objects” (Druckman & Lupia, 2000, p. 4). Below, we outline the major scholarly works explaining attitudes to immigration. We also sketch out existing research on the politics of immigration and the effects of attitudes to immigration on democratic politics.

2 Key Theories Explaining Attitudes to Immigration

A vast literature has been devoted to explaining variation in attitudes to immigration, particularly between individuals in host populations in developed western countries. Here we outline six strands: economics, socialisation, psychology, attitudinal embeddedness, cueing, and context and contact (for other useful reviews see Berg, 2015; Dennison & Dražanová, 2018; Hainmueller & Hopkins, 2014).

2.1 Economic Interests

Reflecting broader trends in scholarly work on political attitudes and behaviour, the use of economic factors to explain variation in attitudes to immigration is one of the longest standing, most developed and—currently—increasingly contested theoretical strands. In particular, evidence of the causal effect of actual economic indicators, such as income or employment, at the individual level is mixed; for example, Espenshade and Calhoun (1993) find no evidence to support this hypothesis, though recent studies have found evidence that labour market competition does affect attitudes (Huber & Oberdabernig, 2015; Pardos-Prado & Xena, 2019; Polavieja, 2016). The effect of psychological perceptions of economic threat has received greater support; Burns and Gimpel (2000) and Espenshade and Hempstead (1996) find that a pessimistic personal economic outlook leads to greater negativity.

At the national level, however, it has been shown repeatedly that economic downturns and rising unemployment rates increase anti-immigration sentiment (Ruist, 2016). Furthermore, Magni (2020) shows that inequality leads to decreasing support for access to welfare for immigrants. It seems that any negative effect of economic downturns on attitudes to immigration is primarily in sociotropic rather than pocketbook terms, i.e. individuals are more concerned about the potential effect on their fellow citizens than themselves, and when they are concerned about themselves it is in terms of perceived economics rather than actual economics. Somewhat tautologically, many works have shown that belief that immigration is bad for the economy or that immigrants take finite resources lead to opposition to immigration, though this may simply be a post facto justification (Fussell, 2014). However, there is evidence that some immigrant groups are seen as likely to contribute and are thus more likely to receive public support (Alba et al., 2005).

2.2 Socialisation

Other studies have suggested that attitudes to immigration are the result of one’s socialising experiences early in life. Importantly, McLaren et al. (2020; see also García-Faroldi, 2017; Kauff et al., 2013) show that being socialised in a more heterogenous society creates more pro-immigration attitudes. Individuals socialised in countries with strong ethnic, rather than civic or multicultural, identities have been shown to be less supportive of immigration (Van Assche et al., 2017; Levanon & Lewin-Epstein, 2010; Hiers et al., 2017; at the individual level, see McAllister, 2018), as have those in which there is a strong collective rather than individualist culture (Meeusen & Kern, 2016; Shin & Dovidio, 2016).

Education has been repeatedly shown to be positively associated with attitudes to immigration, particularly tertiary education, and to explain shifts in generational patterns (McLaren & Paterson, 2020). Jackman and Muha (1984) and Janus (2010) argue that education has an indoctrinating effect which leads individuals to support certain normative ideologies, in this case leading to pro-immigration views while attending university, with its focus on a ‘universal’, rather than national, outlook. Inversely, Hainmueller and Hiscox (2007) and Mayda (2006) argue that less education leads to less tolerance for diversity which leads to anti-immigration attitudes.

Other important socialising experiences include living or being born abroad, white collar-work or belonging to an ethnic minority, which lead to a cosmopolitan worldview and, thus, pro-immigration attitudes (Haubert & Fussell, 2006). Interestingly, less integrated Latinos have been shown to be more pro-immigration (Branton, 2007). Finally, being religious and taking part in religious activities have been argued to increase empathy or induce universalistic values and thus lead to support for immigration (Knoll, 2009; however, see Helbling & Traunmüller, 2016).

2.3 Psychological

Other studies have suggested that attitudes to immigration result from fundamental psychological predispositions, such as personality types, values and identities. The “Big Five Personality Types” have been shown to predict different types of attitudes to immigration (Dinesen et al., 2016). Individuals that value (defined as their long-term and deep-seated motivational goals) tradition, conformity, and security oppose immigration whereas those who value universalism are supportive (e.g. Davidov et al., 2008). Values shared by conservatives and progressives—such as benevolence—are not likely to divide individuals. Similarly, so-called “right wing authoritarian” predispositions—valuing order and unambiguity above all—have been shown to increase anti-immigration attitudes (Cohrs & Stelzl, 2010). Consciousness of in-group and out-group social identities are often shown to be associated with immigration attitudes (Fussell, 2014). Brewer (1999) showed that in-group favouritism was more important than out-group prejudice, with the former leading to a desire to see one’s group’s interests furthered. Lower societal trust is associated with anti-immigration attitudes (MacDonald, 2020).

2.4 Cueing

A common finding in the public opinion research is that individuals tend to take cues from trusted sources of information, such as political elites, in order to form opinions on a wide range of issues (e.g. Zaller, 1992). While individuals’ views may to some extent be influenced by cues from the overall elite stance (Sanders & Toka, 2013), much of the literature explores the extent to which individuals take cues from the party they identify most closely with (e.g. Brader & Tucker, 2012). Several studies find this to be the case with respect to the immigration issue too, the impact of party cues being larger among the more highly educated individuals (Hellwig & Kweon, 2016; Vrânceanu & Lachat, 2021). There is, however, variation in the strength of cueing effects. Harteveld et al. (2017) suggest that political parties at the extremes of the political spectrum have a higher capacity to cue their supporters. Since parties at the extremes are likely to adopt very distant positions from one another, the authors suggest that the cueing effect may contribute to mass polarisation. Along similar lines, Arndt (2016) corroborates that the Danish public opinion polarisation on cultural issues, including immigration, occurred in response to elite polarisation.

2.5 Contact and Context

Both contact theory and group threat theory predict that greater interaction with immigrants will affect attitudes to immigration, but with opposite theorised effects. Similarly, both, though particularly the former, have been studied extensively (Gravelle, 2016; Wilson-Daily et al., 2018). Contact theory theorises that individuals hold misconceptions about immigrants and that contact lessens those misconceptions and thus makes individuals more pro-immigration, as first outlined by Allport (1954; see also, e.g. Ha, 2010; Berg, 2009). Nevertheless, these findings suffer from two methodological weaknesses: contact tends to be either measured through the ethnic composition of the individual’s neighbourhood, which fails to actually measure contact and more pro-immigration individuals are likely to be more willing to have contact with immigrants to start with. These weaknesses have to some extent been overcome by experimental studies (Hewstone et al., 2005), which support contact theory’s supposed mechanisms of improved knowledge, greater empathy, and especially, a reduction in intergroup anxiety (Barlow et al., 2012; Pettigrew & Tropp, 2006).

By contrast, group threat theory expects contact with immigrants to increase the sense of threat felt by non-immigrants, who then become more opposed to immigration, with the effect sometimes contingent on the size of the majority group (e.g. Berg, 2014; McLaren, 2003). Two syntheses have been put forward, first, regarding the level of intimacy of the contact (Fetzer, 2000; Kaufmann, 2014) and, second, regarding who the immigrants are (Ha, 2010). Moreover, despite the vast literature, the effect sizes in either direction are usually considerably smaller than those related to socialisation and psychology and, theoretically, should be less persistent.

In policy terms, Schlueter and Davidov (2013) show that European countries that actively pursue immigrant integration policies foster lower levels of feelings of group threat amongst their citizens. Messing and Ságvári (2018) argue that perceptions of state capacity in general affect attitudes to immigration. Terrorist attacks had been shown to affect attitudes to immigration (Legewie, 2013), but more recent evidence suggests that attitudes to immigration have become sufficiently embedded in Europe that short term events and attacks are unlikely to affect them further (Brouard et al., 2018).

At the personal level, Jackson et al. (2001) show that having a family and children leads to greater anti-immigration views, as individuals become more concerned and cautious about major societal changes. A lack of feeling of safety in one’s neighbourhood has also been shown to lead to anti-immigration views (Chandler & Tsai, 2001).

2.6 Attitudinal Embeddedness

One of the reasons for the high interest in contact theory and group threat theory is that they are intuitively only applicable to attitudes to immigration. However, attitudes to immigration are to a large extent formed by similar forces that determine attitudes to other prevalent political issues, which, as a result, they correlate strongly with and together determine placements within broader attitudinal sets such as ‘left-right’ or ‘authoritarian-libertarian’ (de Vries et al., 2013). Owing to cognitive dissonance, this embeddedness limits the flexibility that individuals might have over such attitudes. Indeed, the correlation between immigration attitudes and broader political attitudes has increased over time (Semyonov et al., 2006).

3 Politics of Immigration

Having reviewed various theoretical accounts for attitudes to immigration, it is relevant to ask to what extent political parties articulate and respond to public preferences. Moreover, which political entrepreneurs mobilise public views on immigration and what are the main patterns of party competition around this issue? We review below several findings from research focusing mostly on European countries.

3.1 Responsiveness

Political representation is the cornerstone of democratic functioning and political parties have a key role in this process (e.g. Dalton, 2017, p. 610). The immigration issue has gained growing political attention in European countries in recent decades (Green-Pedersen & Otjes, 2019) and it has been highly salient in recent national elections and referenda (e.g. Aardal & Bergh, 2018; Hobolt, 2016). There is thus increasing scholarly attention to how responsive political parties are to voters’ preferences on this issue. O’Grady and Abou-Chadi (2019) find at best limited evidence of party responsiveness to short-term changes in the general public opinion. By contrast, Dalton (2017) suggests that political parties tend to be responsive to their own supporters’ views on cultural issues (including immigration), although he identifies a representation gap illustrated by lower congruence between parties’ and citizens’ policy positions, notably among leftist parties (see also Brady et al., 2020; Costello et al., 2012). In a recent study covering 17 European countries, Vrânceanu (2019) finds that mainstream parties tend to be more responsive to the average voter when they face competition from strong radical right competitors. This highlights the role of ‘issue entrepreneurs’ (De Vries & Hobolt, 2012) in enhancing the responsiveness of mainstream parties to the general electorate.

3.2 Support for Radical Right and Other Party Families

Radical right parties (RRPs hereafter) represent the party family that has arguably benefitted most from mobilising public anti-immigrant sentiment (Kriesi et al., 2006). As Ivarsflaten (2008, p. 3) argues, “only the appeal on the immigration issue unites all successful populist right parties”. Research consistently shows that holding anti-immigration views increases the likelihood of voting for an RRP (Kriesi & Schulte-Cloos, 2020; Lubbers et al., 2002; Rydgren, 2008). This is especially so when anti-immigrant parties are evaluated by voters on policy considerations (van der Brug et al., 2005). A string of recent studies focusing on single countries confirm the importance of anti-immigration attitudes for the success of RRPs and extreme-right parties. Focusing on Greece, Dinas et al. (2019) document that exposure to refugees in the context of the 2015 refugee crisis increased the support for the extreme right party Golden Dawn. Similarly, in Germany, the Alternative for Germany (AfD) shifted radically in an anti-immigration direction by 2017, increasingly attracting voters with strong anti-immigrant views (Arzheimer & Berning, 2019). With reference to the same country, Neuner and Wratil (2020) suggest that the combination of anti-immigration and people-centric appeals is particularly attractive to voters. Dennison and Geddes (2019; see also Dennison, 2019; Mendes & Dennison, 2020) also show that the vote share of RRPs in Western Europe increases as the public issue salience of immigration rises.

However, several studies dispute the idea that immigration is mobilised by RRPs alone (e.g. Alonso & da Fonseca, 2011). There is cross-country variation in the extent to which centre-right and centre-left parties used the immigration issue for purposes of electoral competition (Odmalm & Super, 2014). Pardos-Pardo (2015) suggests that centre-parties can benefit from mobilising anti-immigrant sentiment when party competition occurs in a unidimensional space, i.e. when the economic and cultural dimensions of party competition overlap. Moreover, Downes and Loveless (2018) show that in the period following the 2008 economic crisis non-incumbent centre-right parties gained electorally from emphasising the immigration issue. By contrast, more recent studies suggest that Social Democratic parties fail to attract votes when they adopt tough positions on immigration, as this strategy is especially likely to alienate highly educated voters and socio-cultural and self-employed professionals (Abou-Chadi & Wagner, 2020). Instead, mainstream left parties may win votes by combining liberal sociocultural positions (on immigration, among other issues) with investment-oriented economic positions (Abou-Chadi & Wagner, 2019).

In line with the issue voting literature, which examines how voting decisions are based on voters’ issue preferences (e.g. Hobolt & Rodon, 2020, p. 228), the empirical evidence indicates that proximity between voters’ positions on immigration and parties’ stances on this issue matters for vote choice, including when examining mainstream parties only or countries without RRPs at the time of the study (Pardos-Pardo, 2012; Brady et al., 2020). The effect of issue proximity on party support seems to be moderated by voter polarisation (Han, 2018), or by issue constraint and how immigration fits underlying cleavages (Pardos-Pardo, 2012). Note that perceptions of proximity may be endogenous to party affect (Dinas et al., 2016). Finally, the political supply may be scarce for voters who are economically left-wing but hold anti-immigration attitudes (Van der Brug & van Spanje, 2009). The vote choice of citizens in this group should thus be influenced by the relative personal importance of the two issues, economy and immigration (Lefkofridi et al., 2014).

3.3 Party Competition

Research on party competition documents how RRPs can benefit from the strategies that mainstream parties adopt on their main issue dimension, immigration (e.g. Meguid, 2005). On the one hand, studies exploring how the adoption by mainstream parties of accommodative strategies, that is, convergence toward the hard-line policy positions of RRPs, affects the electoral success of RRPs come with mixed findings. Arzheimer (2009) finds no effect, although RRPs do benefit when their competitors talk more about immigration. By contrast, Arzheimer and Carter (2006) find that the probability to vote for an RRP increases when the mainstream right competitor adopts tough positions on RRP’s core issues, which is in line with a legitimisation effect. Down and Han (2019) find a similar effect, but only among voters that did not consider RRPs to be the most competent on immigration. On the other hand, Meguid (2005) suggests that niche parties, such as RRPs, lose electoral support when both mainstream-left and right competitors ignore the immigration issue or converge toward the position of the RRP (see also Dahlström and Sundell (2012) who also show that the behaviour of the mainstream left matters to a higher extent), and win votes when at least one of their competitors adopts an adversarial strategy.

There is also vast research on the extent to which mainstream parties adopt accommodative strategies in response to RRPs’ electoral success, which would be indicative of a contagion effect. Van Spanje (2010) finds that electoral pressures exerted by anti-immigration parties generate incentives for other parties to adopt restrictive positions on immigration (see also Abou-Chadi, 2016; Abou-Chadi & Krause, 2018). Han (2015) documents that, while contagion affects mainstream right parties unconditionally, mainstream left parties are affected only when their supporters become more negative about multiculturalism or immigration, or when they had suffered electoral losses in the previous election. RRPs’ welfare chauvinistic stances may impact as well, although differentially, the positions of mainstream competitors on multiculturalism and welfare (Schumacher & van Kersbergen, 2016). Finally, large parties tend to adopt more restrictive stances also when issue saliency at the party-system level increases (Abou-Chadi et al., 2020). Interestingly, whereas issue attention tends to increase in countries with stronger RRPs and higher shares of foreign-born population, the potential to become a top issue on the political agenda depends on the coalition incentives facing centre-right parties (Green-Pedersen & Otjes, 2019).

However, other studies suggest that the effect of RRPs on mainstream parties’ issue saliency and position-taking may have been exaggerated (e.g. Dancygier & Margalit, 2019). In terms of position-taking, these parties have at times adopted restrictive immigration stances before being challenged by radical right competitors (Alonso & da Fonseca, 2011; see also Mudde, 2013). Furthermore, according to Bale et al. (2010), the response of centre-left (social democratic) parties to the RRP challenge depends on additional factors such as the strategic behaviour of centre-right and left-wing parties. Moreover, parties’ strategies can also be affected by the extent to which the immigration issue aligns with the dominant societal fault line in a given country (Odmalm & Super, 2014). Finally, political parties may actually decide to blur their issue stances, particularly when it comes to issues that they do not primarily mobilise on and in a context of voter polarisation or divided partisan base (Han, 2018).

3.4 Politicisation

Grande et al. (2019) define issue politicisation as a combination of issue salience and polarisation. Focusing on six Western European countries, the authors document growing politicisation of the immigration issue after the 2000s, mostly due to growing party polarisation and issue entrepreneurial strategies on the part of RRPs. This resonates with Dancygier and Margalit’s (2019, p. 28) claim that “if polarisation around immigration has occurred, it has likely been driven by parties located on the farther ends of the ideological spectrum” (but see Alonso & da Fonseca, 2011, p. 880). Researchers have also analysed contexts where party polarisation is driven by the first-time entry into Parliament of RRPs, to assess how this affects voter polarisation. Bischof and Wagner (2019) employ a range of methods to show that the first-time entry into Parliament of a RRP generates voter polarisation on the left-right dimension (see also Castanho Silva, 2018). The mechanisms responsible for this effect are legitimation among supporters of radical right parties and backlash among citizens considering that RRPs violate social norms. Issue politicisation may also depend on the coalition incentives of mainstream right parties (Green-Pedersen & Krogstrup, 2008).

4 Future Avenues for Research

As already made clear in this review, there remain considerable debates and shortcomings in the literature related to attitudes to immigration. In terms of explanations for attitudes, there are relatively few comprehensive models that seek to explain variation in toto, with a strong preference instead for the testing of singular causal mechanisms. As such, we know relatively little about the respective importance of factors, their causal relationships to each other, or their respective positions in the ‘funnel of causality’. The vast majority of studies consider ‘western advanced democracies’ despite immigration being an important political issue in every region of the world. Also typical, though decreasingly so, is the reliance on relatively naïve methods. Moreover, while existing research sheds some light on the political representation of the average voter’s, or of partisan constituencies’ preferences, future research should explore in greater detail the potential contextual influences such as the dimensionality of the political space or the degree of public polarisation (e.g. Ezrow et al., 2014). Future studies should also seek to explore potential drivers and political consequences of mass polarisation specifically on the immigration issue.