Public Attitudes to Immigration in the Aftermath of COVID-19

Authors: James Dennison, Alexander Kustov, and Andrew Geddes

Journal: International Migration Review (2023)

DOI: 10.1177/01979183221134272

Resources

Figures & Tables

Abstract

How has the Covid-19 pandemic affected public attitudes towards immigration? Long-term evidence in Europe and the United States suggests attitudes to immigration are relatively stable and, in some cases, becoming more favorable with high volatility instead in the perceived importance of the issue. However, theoretically a global pandemic could exacerbate people’s fears of outsiders or that migration may contribute to the disease. By contrast, attitudes could remain stable if their distal drivers prove to be robust enough to withstand the shock of COVID-19. We draw from Eurobarometer data from 2014 to 2021 across 28 European countries, weekly national survey data during the outbreak from the US and individual panel data from the UK and Germany to find little systematic change in immigration preferences and no country-level correlation between the observed changes and the outbreak’s severity. Instead, the perceived importance of immigration has consistently and significantly decreased. These findings suggest that, if COVID-19 is to have an impact on attitudes to migration, it is likely to emerge via longer-term means, such as early-life socialization and value change, rather than reactions to the immediate pandemic shock.

Keywords: Immigration Attitudes, COVID-19, Europe, United States, Longitudinal Data

Introduction

How has the COVID-19 pandemic affected public attitudes towards immigration? While evidence in Europe and the United States of long-term attitudinal trends regarding immigration suggests relative stability and, in some countries, greater favorability, we consider whether these long-term trends have been affected by the COVID-19 pandemic. In a 2020 report surveying the pandemic’s impact on migration and mobility, the International Organization for Migration identified the potential for the pandemic to be ‘weaponized’ against migrants, leading to scapegoating, discrimination, xenophobia, and violence (Guadagno 2020). Similarly, academic research has predicted that the pandemic will lead to more negative attitudes to immigration, reinforcing existing trends to border controls and security while fueling expression of hostility, discrimination, xenophobia, and racismWhile strict travel restrictions had been at least partially rolled back in Europe and North America by 2021 due to rising vaccination rates, some argue that this crisis will deal a permanent blow to international human mobility via a significant worsening of public attitudes (Yayboke, 2020). Commentators and politicians have made similar predictions with UN Secretary-General Antonio Guterres, for instance, claiming in May 2020 that the pandemic had ‘unleash[ed] a tsunami of hate and xenophobia, scapegoating and scaremongering’. These responses highlight two key issues that the following analysis explores: (i) how the pandemic has affected various types of attitudes to immigration; (ii) how the pandemic has affected the perceived importance of immigration relative to other political issues.

The article proceeds as follows. We first theoretically consider the likely effects of the Covid pandemic on immigration preferences and on the perceived importance of immigration as an issue to the public. To do so, we draw on Eurobarometer data from 2014 to July 2021 across 28 European countries and the weekly national survey during the outbreak from the US (Voter Study Group, VSG), complemented with individual panel data from the UK (British Election Study, BES) and Germany (German Longitudinal Election Study, GLES). While we find little evidence of systematic change in immigration preferences during the outbreak, we do find that the perceived importance of immigration as an issue has significantly decreased across most national contexts. Taken together, these findings suggest that, if COVID-19 has a lasting impact on attitudes to migration, it is likely to emerge via longer-term means such as early-life socialization and value change linked to negative effects on educational or mobility opportunities, rather than an immediate emotional reaction to a powerful exogenous shock from the pandemic. Moreover, our findings point to the durability of attitudes regarding exogenous changes and the importance of real-world events—which media and political actors can only mediate—on public issue salience.

Theoretical Expectations: Immigration Attitudes under a Pandemic

Theoretically, significant natural and economic shocks such as a global pandemic—and the widespread social restrictions in response—can have important and lasting consequences for a variety of political institutions and attitudes. Centuries later, the Black Death’s effect on political institutions is still visible today. Furthermore, there is some evidence from the recent Ebola epidemic that widespread contagious diseases can shape immigration attitudes under certain politicized conditions.

How should we expect immigration attitudes to change in the aftermath of COVID-19? It has been argued that exposure to a global pandemic can make people more anti-immigration, at least in the short term. Esses and Hamilton’s (2021) literature review mentions the ‘increased feelings of threat and competition, heightened uncertainty, lack of control, and a rise in authoritarianism’ among other possible psychological effects of COVID-19. Related to that, find that ‘[b]y promoting epistemic and existential motivational processes and activating people’s behavioral immune systems’, this pandemic may have made people more socially conservative. Moreover, social psychological theories of Right-Wing Authoritarianism and Social Dominance Orientation suggest that, amongst those with certain psychological orientations, the COVID-19 pandemic may trigger perceptions of a threat to ‘in-groups’ and thus greater and/or more vociferous opposition to immigration (Craig and Richeson 2014; Whitley 1999; Hartman et al. 2021; Pratto et al. 1994). Given this reasoning, we formulate the following baseline expectation:

H1: Immigration attitudes have become more negative during the COVID-19 outbreak.

It is also possible that immigration attitudes are stable and robust enough to withstand even the shock of a global pandemic. According to a recent comprehensive literature review and the longitudinal cross-national evidence, for instance, neither the 2008 economic recession nor the post-2015 refugee crisis were able to significantly shift public preferences regarding immigration. Moreover, it may be that with migration flows falling considerably, there are few new events that are likely to change public attitudes. In line with this, a recent study based on a two-wave survey in Germany was unable to detect any changes in xenophobic attitudes during COVID-19 in particular. Should attitudes to immigration continue to remain stable or even become more positive despite a third ‘crisis’ that, like the previous two, had been widely predicted to increase animosity to immigration, it would support theories that see attitudes to immigration as deeply rooted, a result of early life socialization and primarily changed at the aggregate level via generational replacement. Consequently, another plausible expectation is null systematic effects—that is, although immigration attitudes could go up or down in the aftermath of COVID-19 in different contexts, there is no systematic change in pre-COVID trends.

H0: Immigration attitudes have not systematically changed during the COVID-19 outbreak.

Finally, while the preponderance of theoretical expectations points to the likely negative impact of COVID-19 on immigration attitudes, there may also be reasons to expect more positive attitudes in the pandemic’s aftermath, particularly regarding the role of so-called “essential workers” which the pandemic has arguably highlighted in certain countries. However, we consider this possibility relatively less likely outside of some isolated cases.

Overall, whereas there have been predictions that the pandemic would increase negativity to migrants and immigration - with some socio-psychological theoretical grounding - there are also good empirical and theoretical reasons to expect that any effect of short-term contextual changes—even as grave as a global pandemic—may be muted compared to longer term and deeper socialization effects.

It is important, however, to differentiate between two distinct forms of immigration public opinion that the pandemic could affect: on the one hand, attitudes, perceptions and preferences to immigration and immigrants and, on the other, the importance or salience that the public believes the issue of immigration to have (Dennison, 2019). Attitudes have hitherto been found to be relatively slow-moving and resulting from deep-seated early-life sociological and psychological formative forces, public issue salience—being typically measured in relative terms of perceived importance compared to other political issues—is volatile. It has been shown to result from media and political actor’s agenda-setting and “real-world” external developments, with the pandemic being an obvious example of the latter, and, the literature increasingly suggesting that public issue salience is determined more by exogenous events and trends than by political and media cues, with the latter often following or mediating the agenda rather than setting it (Singer, 2011; Klüver and Sagarzazu, 2016; Dennison, 2020; Gilardi et al, 2021; Klüver, & Spoon, 2016; Spoon and Klüver, 2014). Moreover, whereas attitudes to immigration and immigrants remained generally stable or even became more positive during the ‘migration crisis’ in Europe after 2015, the salience of immigration soared, particularly in Western and Northern Europe, which has been identified as a cause of the rise—and, later, partial decline—of anti-immigration populist radical right parties (Dennison and Geddes, 2019; Dennison, 2020, Mendes and Dennison, 2021). Just as the salience of immigration typically fell during the 2008 financial crisis and, where applicable, the following Eurozone sovereign debt crises, the salience rose during the ‘migration crisis’. We expect the perceived importance of immigration is likely to decrease with the rise of health and related economic concerns due to the outbreak. Consequently, we also expect that, if COVID-19 had any significant impact on immigration attitudes, it would mostly be apparent in terms of the issue’s decreasing salience to many people across the world.

H2: Immigration issue salience has significantly declined during the COVID-19 outbreak.

Cases, Data and Methods

To test the above hypotheses, we choose the US and Europe as our cases as major immigration destination countries and regions. We use additional data from Germany and the UK that allows verification of our findings according to alternative data sources and other types of attitudes to immigration: specifically, perceived effects of immigration on culture and the economy and immigration policy preferences. In Europe and the US there has been a tendency towards border restrictions and controls that predates the pandemic, as well as growth in support for populist, right-wing anti-immigration political parties and movements. There were also divergent responses to the pandemic with variation in, for example, political leadership and imposition of restrictions between Europe and the USA. For example, in the US we saw an overt tendency from ex-President Trump and other political leaders to associate migrants and foreigners with the virus.

As the main source for our analysis, we rely on the Eurobarometer data from November 2014 to July 2020 across 27 European countries for the items on immigration policy preferences and the Eurobarometer data across 28 European countries from May 2005 to July 2021 for issue salience. For policy preferences, we use responses to two questions: “Does immigration of people from other EU member states give you a positive or negative feeling?” and “Does immigration of people from outside the EU give you a positive or negative feeling?”. Possible responses to both questions are on a five-point Likert scale from very positive to very negative, which we collapse into a net positivity measure at the country level. For issue salience, we use responses to “what do you think are the two most important issues affecting your country?” Respondents are offered around 14 responses, including immigration, which have changed only slightly over time, and not at all in the rounds immediately prior to and after Covid. We also provide some preliminary exploration of the pandemic’s possible heterogeneous impact by comparing the net change in attitudes across countries by the outbreak severity.

We then look at the unusually frequent nationally representative survey conducted as a part of the Voter Study Group in the United States from January 2 to June 19 of 2020. Finally, we complement our repeated cross-sectional analyses with two individual panel datasets (surveying the same individuals over time): the British (BES) and German (GLES) election studies. This selection of data sources both allows us to verify the identified trends and make use of their most robust panel character.

To summarize our results, we use simple graphic evidence indicating average public attitudes across survey waves by country. We also provide fixed effects panel data models for each of our data sources in the appendix. For the BES and GLES these are done at the individual level, acting as a robustness check against the ecological fallacy.

Analysis and Results

Repeated Cross-national Evidence from Eurobarometer

First, we rely on the Eurobarometer data to examine the change of public positivity toward EU and non-EU immigration across 27 European countries in the last seven years with a focus on the last two (pre-and post-outbreak) waves from November 2019 to July 2020 (see Figure 1). As can be seen, there is much diversity in terms of the attitudinal changes during 2020—while some countries have experienced a worsening of public immigration preferences (including the EU average)—other countries have seen an improvement in those preferences. We test this proposition with a fixed effects panel model in Table A1, which shows that the Covid period had no statistically significant effect on attitudes to immigrants from other EU countries and a small negative one to attitudes to immigrants from outside the EU.

Clearly, public response to the pandemic across countries can vary. In line with the theoretical reasoning behind H1, countries that have been particularly hit by the outbreak may display the biggest negative change in attitudes. Figure A1, however, for the EU shows that this is not the case—the average attitudinal changes toward either migrants from within the EU or non-EU immigration are unrelated to the outbreak’s severity across countries. More significantly, none of the observed changes in net attitudes appear to be systematic after accounting for countries’ pre-trends in attitudes (see Figure A1).

Figure 1 about here

Second, we repeat our analysis in the EU for immigration salience or whether people consider immigration one of the most important issues facing their country now (see Figure 2). Unlike the case of policy preferences, most European countries have experienced an unambiguous decline in immigration’s issue salience after the COVID-19 outbreak. Furthermore, while it is also true that immigration issue salience has been in decline for a few years prior to the outbreak across many countries, a ‘COVID-19 effect’ holds even after accounting for these pre-trends, as shown in Table A2, in which we again used a fixed effects panel model to show that the strong negative general effect of Covid on issue salience was statistically significant. However, it should also be noted that in the most recent round of the Eurobarometer, we can see salience beginning to slightly increase again.

Fine-grained Repeated Cross-national Evidence from the United States (VSG)

We can also take advantage of the unusually frequent representative US survey (VSG) and look at how the US public reacted to the outbreak and the pandemic’s progression (see Figure 3). As can be clearly seen, none of the available immigration preference items have moved much throughout the first half of 2020. US attitudes slightly worsened when a national emergency was declared on March 13, 2020, but these changes quickly dissipated within a week. Despite the significant pandemic toll and the numerous (even if temporary) restrictive changes to the actual immigration policy, US immigration attitudes were remarkably stable from the beginning (in January) to the end (in June) of the survey panel. In Figure A5 we also see favorability to various groups over time using the same data, showing that unfavourability to Asians remained the lowest of the four groups considered and declined during the period after Covid.

Longitudinal evidence from the United Kingdom and Germany (BES and GLES)

We can also consider the individual panel data from the UK and Germany where it is possible to see whether the same individuals changed their mind on immigration throughout the outbreak (see Figures A2, A3 and A4). As before, there is no evidence that immigration preferences have worsened during the pandemic. In fact, in both Germany and the UK there seems to be a slight positive change, albeit insignificant after accounting for the pre-trends in those countries. Table A3 uses fixed effects models to show no statistically significant effect of Covid in Germany, while Table A4 shows (1) a small positive effect on allowing immigrants to enter, (2) a small negative effect on the perceived effect of immigrants on the economy; and (3) no effect on the perceived effect of immigrants on culture.

Figure 2 about here

Figure 3 about here

Discussion

Public support for international mobility is often cited as being among the pandemic’s possible long-term costs. With the increased economic uncertainty due to a deadly pathogen, many have understandably expected a resurgence of ethnocentrism and anti-immigration sentiments around the world. Nonetheless, as we show in our research based on the best available cross-national and longitudinal public opinion data, immigration attitudes have not systematically become more negative in the aftermath of COVID-19. This conforms with longer term, generational trends that underlie trends towards greater positivity. It may be that any negative effect of the pandemic was offset by increasingly visible economic role of immigrants and their occupation in high skilled employment or work labelled as essential particularly during the lockdowns. We also find no evidence of a specific effect on the favorability felt towards Asians in the US nor of a differential effect according to pandemic severity in Europe. What did change is that issue salience sharply declined because of the importance of other economic and public health problems. It is entirely plausible that immigration could return to the forefront of the public agenda as and when the pandemic declines, as has already been evident in the US and we see some evidence of in our Eurobarometer data (Figure 2). Similarly, it may also be possible that the pandemic’s effects on immigration preferences will only become apparent in the long-term if decreased mobility leads to changes in generational socialization or relevant changes in societal norms that result from the pandemic.

Overall, these results are in line with the growing research demonstrating the remarkable stability of immigration preferences, as well as the volatility and the corresponding political importance of immigration salience and its exogenous, real-world determinants in line with recent findings that see agenda-setters as only having the power to mediate. Theoretically, we we argue that our findings support the notion that attitudes to immigration are slow-moving because their key determinants—early-life socialization, such as education, and psychological predispositions—such as values—are stubborn and either slow to change or largely fixed as individuals enter adulthood. This is likely to be particularly the case after a period of high politicization—such as in Europe and the United States of the last decade—in which individuals have had ample time to “make up their minds”. As such, even an unprecedented pandemic with clear migratory consequences is not enough to shift public opinion significantly. Furthermore, it may be that the negative effects of greater threat could be counterbalanced by greater recognition of migrant’s contributions, something future research may want to test. On salience, by contrast, the high observed volatility and negative effect of the pandemic likely reflect both media coverage and more fundamentally radically reduced migration flows as well as the ordinal nature of the “most important issue/problem” question that is just as much a reflection of other issues—health and the economy in this case—as immigration itself.

This research note is not without limitations. As of Summer 2022, COVID-19 is still present in many parts of the world, meaning that it is possible that our conclusions could require updating as more evidence becomes available, especially if the pandemic lasts significantly longer than expected. Evidence of the effects of previous pandemics highlight the potential for differential effects depending on how hard hit specific places or countries have been by the disease (Gingerich and Vogler 2021). To that end, future research can expand on our analysis by including more extensive data from our case countries and regions as well as other countries, as well as comparing with other periods or crisis events, as well as exploring potential heterogeneous impacts in more detail (including the consideration of policy responses across different countries). Moreover, although we provide fixed effects panel data models using the BES and GLES at the individual level in the appendices, acting as a robustness check against the ecological fallacy, these are two countries in which immigration has been highly politicized in recent years (Dennison and Geddes, 2019) making such attitudes likely to be strong and durable (e.g. Druckman et al, 2012); in countries in which immigration has been less politicized, such attitudes should be less strong and thus it is more plausible that there is an observable effect of COVID-19 (see Figure 1). Finally, although our results suggest that COVID-19 has had no systematic effect on public attitudes to immigration in Europe and the United States, this is not to say that no events can effect such attitudes. Indeed, terrorist attacks by immigrants or their descendants, visible refugee populations in public, and criminality and sexual attacks by immigrants—in short, events that are perceived to be directly related to or caused by immigration—having previously been shown to have some effect on public attitudes (Dhanani et al. 2021; Erisen and Kentmen-Cin 2017) albeit contingent on motivated social cognition (e.g., Jost et al. 2017; Pascilli et al. 2022).

References

Figure 1: Immigration positivity pre-and post-the outbreak of COVID-19

Notes: Net positivity measured as total positive responses minus total negative responses to the questions: “Does immigration of people from other EU member states / outside the EU give you a positive or negative feeling?” Source: Eurobarometer, vertical line represents emergence of COVID-19. November 2014-July 2020

Figure 2: Immigration Salience Pre-and Post the Outbreak of COVID-19

Notes: Responses to the question “what do you think are the two most important issues affecting your country?”, source: Eurobarometer; vertical line represents emergence of COVID-19, May 2005-June 2021

Figure 3: Immigration Preferences Pre-and Post the Outbreak of COVID-19 (United States)