Evidence from eight waves of the European Social Survey
- Survey data from 13 EU member states show that Europeans assess the impact of immigration on their country and its economy and culture in a more positive light than they did in 2002 – even in the aftermath of the 2015 refugee inflow. Hungary is a prominent exception.
- Public opinion varies little across these EU member states on the impact of immigration. On average, people are rather agnostic about its overall impact on their country, and its economy and culture.
- At the same time, public opinion within member states (at least in Germany and Hungary) has become more polarized: many people hold stronger views today than they did in 2002.
- Our findings imply that policy-makers have room to develop well-balanced solutions for immigration and asylum policy.
Europe’s migration and asylum policy continues to cause political and economic conflict. Migration issues dominate the European agenda and create divisions within and across political parties in EU member states. How do Europeans view the impact of immigration on their countries, especially in the aftermath of the refugee crisis?
This Policy Brief1 provides an overview based on eight waves of the European Social Survey (ESS),2 including the most recent data of 2016-2017 covering attitudes after the 2015 refugee crisis. The data suggest that when asked about the impact of immigration – irrespective of its different types (e.g. EU- or non-EU migrants) and channels – on their country, the economy and their culture, people across Europe have on average evaluated immigration in more positive terms since 2002-03. Moreover, we do not see a strong increase in polarization when comparing those EU member states that have been regularly surveyed since 2002-2003 on an aggregate level. Yet, when comparing Germany and Hungary, we find that polarization within states has increased. We show that the variation in attitudes in these two countries is strongly associated with educational attainment. In addition, the refugee crisis has left its mark on both countries in different ways: in Hungary, it is associated with a fundamental change of attitudes towards the negative, but not in Germany. We conclude with a discussion of the policy implications of these findings.
On average Europeans have become more accepting of migrants since 2002
On average, Europeans have become slightly more accepting of migrants, compared to the situation in 2002 (see Heath and Richards, 2016, for similar results using ESS data until wave 7). This trend has not been reversed in the aftermath of the refugee crisis. Since its first survey wave of 2002-2003, the European Social Survey has asked Europeans regularly how they judge the impact of immigration on their country in general, on its cultural life and whether they consider immigration to be good or bad for the national economy. These questions are obviously quite broad: they do not cover specific policies, nor do they differentiate between types of immigration. Instead, they seek to ascertain general attitudes on the part of Europeans towards immigrants moving into their country or their neighborhood.3 On average, throughout the EU member states surveyed in all eight ESS waves, there has been a trend towards a slightly greater acceptance of immigration. Respondents increasingly say that immigration makes their country a better place in which to live, enriches its cultural life and makes their economies better off (Figure 1).
On average, there is also no large variation between European countries on these issues. The level of acceptance does not differ strongly across most of the countries surveyed here. Most of the numeric answers to the three questions shown in Figure 1 range from 4-6, which means that the citizens in most EU countries surveyed here are on aggregate quite agnostic concerning the effect of immigration on their country as a whole – at a value close to 5 on a scale that ranges from 0 (worse) to 10 (better). This finding may not be surprising for those who expect the various economic and societal costs and benefits of immigration to hinge on a whole battery of factors.4
Nevertheless, Gallup data from 2016 have suggested that Europe is the most polarized region in the world with regard to the acceptance of immigration (Ray, Pugliese, and Esipova, 2017). The average answers to the ESS question about the impact of immigration on the country show differences in levels. People in Sweden more often consider immigration to make their country a better place to live in than people in Portugal. Furthermore, there are some changes over time. In a majority of countries, among them Germany, there was a small average increase in acceptance. The opposite can be found in others, such as Hungary. The fact that Europe is home to some of those ‘outliers’, however, does not necessarily warrant special or renewed attention, as the “polarization” warning may suggest. First, the aggregate variation between countries is still not overly large (with most countries being either slightly more positive or slightly more negative than the average ‘agnostic’ countries), which may still indicate some common ground to generate common policies. More importantly, however, the ESS data show that this degree of polarization is not necessarily a new phenomenon. With the exception of Hungary (towards less acceptance) and Portugal (towards more acceptance), there have been few drastic changes in attitudinal patterns throughout the last decade on this aggregate level.
Polarization of attitudes towards immigration increases within states
What has changed throughout the last decade is the polarization of the issue of immigration within EU member states. At least, this change is evident if we compare the responses in the survey within Germany and Hungary across different years (Figure 2). While the largest group of respondents always found that immigrants neither make a country a better or worse place in which to live per se (indicated by the highest bar at a value of 5), this group has become smaller in both Germany and Hungary, suggesting that when compared to 2002-2003, more people express a rather positive or negative – as opposed to neutral – opinion about immigration in 2016-2017. The group viewing migration as having an overall positive impact has slightly grown in Germany since 2002-2003, whereas the group viewing it as having a negative influence has grown in Hungary. Despite the rise of a vocal anti-immigrant party in Germany in the aftermath of the 2015 refugee crisis, respondents’ views have not changed dramatically even when comparing the two surveys from 2014-2015 and 2016-2017 (there is a slight tendency towards more negative views, which may yet also be the result of different weightings of the survey results). In Hungary, by contrast, the growth in the number of people who view migration as very detrimental is substantial and visible to the casual observer.
What are the driving forces behind some of these within-country developments? To study this question more systematically, we estimate simple linear regression models5 to assess, for instance, the “2015” effect on different attitudinal items, controlling for several individual factors, such as age, gender, education and migratory background. We exploit the fact that we have data since 2002 and that two survey rounds were conducted during (2014-2015) or after (2016-2017) the peak of the refugee crisis.
Figures 3 and 4 show the results, with dark grey dots representing results for Germany and light grey diamonds those for Hungary. The dots and diamonds represent coefficients that tell us how a change in each variable listed to the right of the figure (such as age or gender, but also the year) changes the attitudinal variable (such as the expected impact on the economy in Figure 3.1). The coefficients are country-specific, which means that they do not provide information on the attitude that an average Hungarian has towards immigration, compared to a German individual. Instead, they inform us about the change within a country relative to a so-called baseline level that serves as a comparison. Imagine that we had asked a person in Hungary to rate the impact of immigration on the economy on a scale from 0 (bad) to 10 (good) in 2002-2003 and that person had said “5” at the time. The coefficient of about -1 then suggests that on average, if we had asked a person who was exactly the same with regard to age, education, gender and migrant background in Hungary in 2016-2017, the answer would have been “4”. Note that the baseline survey was conducted in 2002-2003 in the aftermath of the September 11th terrorist attacks in the United States and may thus be associated with a low level of acceptance of migration.
Education as the dividing line in Europe?
Figures 3 and 4 suggest that while age and gender do not account for a large amount of variation, education is an important factor in both Hungary and Germany. The tail of data points in Figures 3 and 4 associated with the education variables suggests that the more education an individual are has acquired, the more s/he thinks that immigration is good for the country, its economy and its cultural life. This finding is well established in the literature as well (Ceobanu and Escandell, 2010). The ESS data additionally suggests that there is a certain level of education (completion of upper-secondary school) above which attitudes become markedly more positive. The very clear-cut education effect is less pronounced when people are asked questions that are often used to study prejudice and discrimination such as whether they would mind a migrant marrying into their family (Figure 4.1). Also, answers to the question whether governments should judge refugee applications generously (4.2) show little variation across education levels. While it is still disputed what exactly the education effect measures, some authors have suggested that educational differences and the educational ‘revolution’ lie at the heart of the new political and economic dividing lines in Europe (see Bornschier, 2018 for an overview of some of these debates).
There are a variety of other country-specific factors, such as the media environment and the framing by political actors that shape public attitudes towards immigration, that may explain why there are such big differences between Germany and Hungary in their reaction to the crisis. (see Ademmer, Benček and Stöhr 2017, and Ademmer and Stöhr 2018 for a more detailed overview of the extensive literature on this subject.) Figures 3 and 4 also suggest that first-generation migrants do not differ significantly from their non-migrant peers in terms of their attitudes towards immigration. First-generation migrants are those individuals with a direct migration experience. Second-generation migrants are those individuals who are not also first-generation migrants, but whose parents have migration experience. Interestingly, unlike their parents, second-generation migrants have significantly different attitudes towards immigration than the average individual in Germany or Hungary: they are on average less skeptical towards immigration across all items shown here. The category ‘diaspora born abroad’ includes individuals who are first-generation migrants (and so have had a migration experience), but whose parents originated from the country where the individual currently lives. This may be a person who was born in France to two German parents and now lives in Germany.
The refugee crisis: A brief setback in Germany, a game changer in Hungary
How did attitudes in Germany and Hungary change over the refugee crisis? There is little evidence that the crisis dramatically changed attitudes in Germany (for current views also see SVR 2018 [in German]). On the contrary, quite noticeable changes had occurred prior to the onset of crisis: Germans had come to see immigration in increasingly favorable ways since 2004-2005 (Figure 3). This figure also suggests that in 2016-2017 – after the height of the refugee crisis – the acceptance of migrants was higher than it had been in 2002-2003.
In addition, the people surveyed in Germany display differentiated attitudes and the crisis has had a similarly differentiated impact on them. For some items, the refugee crisis was seemingly only a brief ‘shock’: Germans assessed the impact of immigration on the economy (Figure 3.1) slightly worse in 2014-2015 than in the preceding and succeeding waves. The acceptance of ethnically different immigrants (Figure 3.2) increased in 2014-2015 – associated with the widespread ‘Willkommenskultur’ at the time – and then it lapsed back to the pre-crisis level. Other items changed slightly, but not to levels previously unknown within the German public: from 2012-2013 onwards, people were more critical of the impact of immigration on Germany’s cultural life (Figure 3.3) than they were in 2002-2003, but this has been the case in all survey waves since 2004. People were also becoming more skeptical regarding the overall impact of immigration on the country (Figure 3.4), but they were still assessing the impact of immigration in more favorable terms than they did in 2002-2003.
In Hungary, however, the refugee crisis has apparently been a game changer. Until 2012-2013, public opinion in Hungary and Germany had moved in similar directions on a variety of items (note that we are not considering differences in levels). In Hungary, however, we see a dramatic change in attitudes in 2014-2015 and 2016-2017 in all items shown in Figure 3. For example, when asked in 2012-2013 whether they were in favor of allowing many ethically different immigrants into Hungary, Hungarians on average also replied more positively to this statement than in 2002-2003, but this pattern had been reversed by 2016-2017.
The data shown here suggest that Europeans have not become increasingly negative on average when asked about immigration in general – without differentiating between different types and channels of immigration. On the contrary, the responses indicate that aggregate public opinion has become more, rather than less accepting of immigration since 2002-2003. Also, polarization across EU member states has not increased substantially over time: for the time period surveyed, the ESS data suggest that Hungary is the odd man out, rather than representing an overall European trend. Drawing on Germany and Hungary as examples, we have, however, shown evidence of growing polarization within states: people in these two countries hold stronger views about immigration than they did in 2002.
Attitudes towards immigration may, of course, differ depending on the various types and channels in question. Immigrants from other EU member states may be viewed differently than third-country nationals, for instance. While we have shown that the same trends that are outlined above emerge when people are asked whether many or only a few immigrants should be admitted from poorer countries outside Europe, the data do not allow us to conclude what respondents would say specifically about people who come with different motives such as work, family reunion, or asylum. Given the evidence presented in the scientific literature, we are confident that most European citizens are able to differentiate between groups of migrants, especially when these differences are underlined by the careful use of precise terms in the survey instruments deployed.
Further research is required before we can draw reliable conclusions on the exact policy preferences of European citizens. The ESS data shown here, however, strongly suggest that a significant number of citizens would support migration policies that are not purely aimed at restricting inflows of migrants, also from poorer countries to the EU.
This stratification of public opinion would give policy-makers substantial room for maneuver to find common European policies and to deal effectively with the recent crisis (examples of such policy proposals can be found here). Indeed, other research suggests that a majority of Europeans would support common European solutions or combined national/European solutions (Sik and Szeitl, 2016).
In the current political climate it is important to stress that most European voters are not critical of immigration per se and, we hypothesize, might even welcome legal migrant labor from outside the EU. The existing literature shows that asylum seekers who have a claim that is seen as legitimate and labor migrants who can contribute to the economy are widely accepted. Under the current policy, which is highly restrictive with respect to all immigration from outside the EU apart from high-skilled migrants and specific occupations, many prospective migrants only have the option of entering as asylum seekers. By trying their luck, they sometimes blur the distinction between different the motivations behind migration. Restrictive labor migration policy may thus even reduce public acceptance of the asylum system.
1 The responsibility for the contents of this publication rests with the authors, not the Institute or the Mercator Dialogue on Asylum and Migration (MEDAM). Any comments should be sent directly to the authors.
2 European Social Survey Round 1-8 Data (2002-2016) NSD – Norwegian Centre for Research Data, Norway – Data Archive and distributor of ESS data for ESS ERIC.
3 Respondents then tend to reply based on the kind of people they associated with the terms used in specific survey questions, which can change over time with the visibility of different groups in their neighborhood or through the media (Blinder, 2015). When asked in more detail about individual groups, e.g. with different reasons for migration, the responses show that people’s attitudes to these differ (Hainmueller and Hiscox, 2015; Bansak et al., 2016). Yet, the trends we found hold for immigration in general, for immigrants from richer and poorer countries of origin as well as for origins within and outside of the EU.
4 Using alternative survey questions, such as the question whether one should admit many or only a few immigrants from poorer countries outside Europe, underlines this pattern. On this specific question, differences in opinion between countries are more pronounced, but the trends over time are very similar.
5 The results shown here are based on OLS estimations. Ordered logit estimates are usually considered to produce more reliable results for ordered categorical variables but they are more difficult to interpret for the reader. Since these alternative estimations leave our OLS-results largely unaltered, we decided to include OLS estimates, which are more intuitively interpretable.
Ademmer, E., D. Benček, and T. Stöhr. 2017. “Determinants of Public Attitudes towards Immigrants and Immigration.” In 2017 MEDAM Assessment Report on Asylum and Migration Policies in Europe: Sharing Responsibility for Refugees and Expanding Legal Immigration, MEDAM (Mercator Dialogue on Asylum and Migration), 51-56. Kiel: Kiel Institute for the World Economy.
Ademmer, E., and T. Stöhr. 2018. “The Impact of Security-Related Events on Attitudes towards Migration.” In 2018 MEDAM Assessment Report on Asylum and Migration Policies in Europe: Flexible Solidarity – A Comprehensive Strategy for Asylum and Immigration in the EU, MEDAM (Mercator Dialogue on Asylum and Migration), 62-68. Kiel: Kiel Institute for the World Economy.
Bansak, K., Hainmueller, J., and D. Hangartner. 2016. “How economic, humanitarian, and religious concerns shape European attitudes towards asylum seekers.” Science 354(6309), https://doi.org/10.1126/science.aag2147.
Blinder, S. 2013. “Imagined Immigration: The Impact of Different Meanings of ‘Immigrants’ in Public Opinion and Policy Debates in Britain.” Political Studies 63: 80-100.
Bornschier, S. 2018. “Globalization, Cleavages, and the Radical Right.” In The Oxford Handbook of the Radical Right, edited by J. Rydgren. Oxford: Oxford University Press.
Ceobanu, A. M., and X. Escandell. 2010. “Comparative Analyses of Public Attitudes Toward Immigrants and Immigration Using Multinational Survey Data: A Review of Theories and Research.” Annual Review of Sociology 36: 309–28.
Hainmueller, J. and D. J. Hiscox. 2015. “The Hidden American Immigration Consensus”, American Journal of Political Science 59(3): 529–548.
Heath, A., and L. Richards. 2016. “Attitudes towards Immigration and Their Antecedents: Topline Results from Round 7 of the European Social Survey.” ESS Topline Results Series No.7.
Hooghe, L., and G. Marks. 2017. “Cleavage Theory Meets Europe’s Crises: Lipset, Rokkan, and the Transnational Cleavage.” Journal of European Public Policy 0 (0): 1–27. https://doi.org/10.1080/13501763.2017.1310279.
Ray, J., A. Pugliese and N. Esipova. 2017. “Gallup: EU Most Divided in World on Acceptance of Migrants.” Gallup, Washington, DC, September 6. https://news.gallup.com/poll/217841/divided-world-acceptance-migrants.aspx?version=print.
Rodrik, D. 2018. “Populism and the Economics of Globalization.” Journal of International Business Policy 1: 12–33.
Sik, E. and B. Szeitl. 2016. “Some Characteristics of the European Public Opinion on Migrants and Asylum Seekers.” In The Social Aspects of the 2015 Migration Crisis in Hungary, edited by B. Simonovits and A. Bernát. http://old.tarki.hu/hu/news/2016/kitekint/20160330_refugees.pdf.
SVR (Sachverständigenrat deutscher Stiftungen für Integration und Migration). 2018. Stabiles Klima in der Integrationsrepublik Deutschland. SVR-Integrationsbarometer 2018. Berlin: SVR.