Who cares what the people think? Public attitudes and refugee protection in Europe

Abstract

This paper discusses why and how public attitudes should matter in regulating asylum and refugee protection in rich democracies, with a focus on Europe. Taking a realistic approach, I argue that public views constitute a soft feasibility constraint on effective and sustainable policies towards asylum seekers and refugees, and that a failure to take seriously and understand the attitudes of the host country’s population can have a very damaging effect on refugee protection and migrants’ rights in practice. Bringing together insights from political philosophy, the politics of asylum, and research on public attitudes, I develop my argument by discussing why ‘what the people think’ should matter in asylum and refugee polices; how public views can and should matter given the well-known challenges with measuring attitudes and policy preferences; and what the prevailing public views might mean for the reform of asylum and refugee policies in Europe.

What role should public views play in the making of public policies? Should public policy-makers pay attention to what people think and prefer when designing policies and, if so, how? This paper addresses these long-standing questions in the specific context of asylum and refugee policies in high-income democracies, with a focus on Europe. How to regulate asylum and refugee protection has become one of the most politicised policy issues in many countries, and new policies have often been justified by referring to “what the people want.” It is, therefore, important to consider not only what these public views are (e.g. Jeannet et al., 2021) but also what role they should play in actual asylum and refugee policy-making.

The rising importance of asylum and refugee protection in the national politics of many high-income countries stems, at least in part, from the increasing scale of the challenge during the past decade. The global numbers of asylum seekers (estimated at 4.1 million in 2020) and refugees (over 20 million) have doubled since 2010 (UNHCR, 2021). These numbers are not historically unprecedented (White, 2019), and it is important to recognise that the vast majority of the world’s refugees, an estimated 85 percent, are hosted by low- and middle income countries. Nevertheless, many high-income countries have in recent years experienced substantial and largely unanticipated increases in the number of asylum seekers asking for protection. The largest increases have been recorded in Europe where the number of annual first-time asylum claims rose from 338 thousand in 2013 to 1.2 million in both 2015 and 2016 (at the height of Europe’s so-called “refugee crisis”) before declining again to 620 thousand in 2017 and fewer than in 450 thousand in 2020 (Eurostat, 2022). The spike in the number of new asylum seekers in 2015 quickly led to the de-facto collapse of the EU’s common framework for asylum (the “Dublin system”) and a public perception that asylum and refugee protection were mishandled by governing institutions (Connor, 2018) and generally “out of control”. The rising numbers have also increased the salience of asylum and refugee issues among the public and contributed to the rise of anti-immigrant populist parties (Dennison and Geddes, 2019; Dennison, 2020). How and under what circumstances different types of exposure to refugees affect voting for right-wing parties has become an active area of research (e.g. Dinas et al., 2019; Gessler et al., 2019; Steinmayr, 2021; Rudolph and Wagner, 2022).

In response to these developments, European and national policymakers have been engaged in highly acrimonious and divisive debates about how to reform asylum and refugee policies. A wide range of new policies has been proposed, and in some countries already implemented, with contrasting implications for opportunities to apply for asylum in Europe; the rights of asylum seekers and recognized refugees such as access to the national welfare state and family reunification; the resettlement of already recognized refugees from first countries of asylum near conflict regions; assistance and cooperation with countries of origin and transit; responsibility-sharing across host countries in Europe; and, more generally, the desirable degree and modes of policy-harmonisation across EU countries (see, for example, Owen, 2020; Aleinikoff and Zamore, 2019; Hathaway, 2018; UNHCR, 2018; Lucke et al., 2018; Betts and Collier, 2017). Some reform advocates have called for a “paradigm change” in how asylum and refugee protection are regulated in Europe, suggesting new policies that would deviate from some of the existing protection principles and norms set out in the 1951 Geneva Convention (see, for example, Austrian Ministry of Interior and Danish Ministry of Integration, 2018). Calls for paradigm change have grown even louder in 2021, when the withdrawal of US and other international forces from Afghanistan led to a large increase in forced displacement, fuelling political concerns in many European countries about a rapid rise in the number of asylum applications of Afghan citizens. An increasingly common argument is that significant policy changes are needed to make Europe’s policies for refugee protection more aligned with public views and/or to restore the public’s trust in political institutions which, proponents of radical reform often argue, has been fatally undermined by large-scale immigration (e.g. Betts and Collier, 2018).

In this context of intensifying policy debates that make frequent reference to “what the people want”, this paper provides a normative discussion of the role that public views should play in the making of asylum and refugee policies. My normative starting point is that states have a moral duty to provide protection to refugees which includes the duty not to send people back to countries where they might face serious risks. As Matthew Gibney (2018) has recently pointed out, essentially all moral philosophers agree on this moral duty. Given this starting point, I argue that public views constitute a feasibility constraint on effective and sustainable policies towards asylum seekers and refugees, and that a failure to take seriously and understand the nature and characteristics of this constraint can have a very damaging effect on refugee protection and migrants’ rights in practice.

Building on philosophical work on political feasibility, I define a feasible asylum and refugee policy as one that can be brought about and is likely to be relatively stable over time (e.g. Cohen, 2009). Public views can be best understood as a “soft” rather than “hard” constraint on feasible policy-making. A soft feasibility constraint makes certain policies “comparatively less feasible” (Gilabert and Lawford-Smith, 2012) rather than ruling them out completely as a hard constraint would do. ² I thus follow theorists who have emphasised the importance of understanding political feasibility as a scalar rather than binary concept (e.g. Lawford-Smith, 2013). Soft constraints have probabilistic and dynamic elements in the sense that they are, at least to some extent, potentially malleable over time. This means that considerations of feasibility involve judgements about what actors can do in given contexts (also see Southwood, 2018) and about the extent to which these contexts can and are likely to change over time. Such judgements are of fundamental importance to political debates and decisions about policy reforms, and they need to be informed by the available evidence (Gilabert and Lawford-Smith, 2012). This point also applies to asylum and refugee policies where I argue that public views constitute an important soft feasibility constraint that needs to be recognised, “judged”, and debated (in terms of the available evidence) much more explicitly than has been the case so far.

There are very few studies that have discussed the role of public attitudes as feasibility constraints in asylum and refugee policies. Luara Ferracioli’s (2014) analysis of the appeal and danger of a new Refugee Convention is one of the few exceptions. She argues that negative public attitudes to asylum seekers in host countries are part of a soft feasibility constraint that makes unlikely the successful negotiation of a new Refugee Convention with enhanced rights protections. Ferracioli argues that hostile attitudes are often based on concerns about procedural and distributional fairness associated with the “irregular” arrival of asylum seekers and other migrants in high-income countries, and that a shift toward greater resettlement of already-recognised refugees (directly from regions of conflict) would alleviate public concerns in rich host countries and thereby reduce the soft feasibility constraint to negotiating a New Refugee Convention in the future.

This paper differs from the analysis in Ferracioli (2014) in at least two ways. First, rather than discussing the specific issue of negotiating a new Refugee Convention, I discuss broader questions of why and how public views should and can matter in normative debates about asylum and refugee protection in rich democracies, based on a critical discussion of the available evidence on the nature, characteristics, and determinants of “what the people think”. Second, my discussion of the implications of the public views constraint for asylum and refugee policy-making in Europe takes a more in-depth look at the available empirical evidence on people’s fundamental beliefs, attitudes and policy preferences.

The paper is structured around four questions: First, why, if at all, should public views matter in normative policy debates about asylum seekers and refugees? Second, why are democratic elections not enough as an expression of public views? Third, how can public views matter for policy-making in practice, given the well-known challenges with measuring what the people “really” want and questions about the stability of people’s views over time? And fourth, what do we know about public views on asylum and refugee policies in Europe, and what are the implications for policy-making?

To explore these issues, I draw on theories and insights from research on the political philosophy of immigration and asylum, the politics and determinants of public policies including migration and refugee policies, and the characteristics and determinants of public attitudes and policy preferences. One of the main contributions of the paper lies precisely in the bringing together of these different strands of theories and research, to discuss a core question that lies at the heart of contemporary research and debates on the future of asylum and refugee policy-making.

Before I begin the discussion, it is important to clarify my terminology. I use “public views” as an umbrella category that includes three elements: “values”, “attitudes”, and “opinions”. The distinction between values and attitudes is common and central to the research literature on public views (e.g. Bergman, 1998). Values are deep-seated and relatively stable beliefs, what Worcester (1997) has called the “deep tides of public mood”. Examples could include beliefs in “tradition”, “conformity”, or “universalism” (Schwartz, 1992). In contrast, public attitudes are generally understood as evaluations of specific objects, actions or policies. For example, people could view a particular car as “good” or “bad”, or a policy as desirable or undesirable. In the context of this paper, attitudes also include policy preferences, i.e. public views on whether certain policies should be supported and how they should be designed or implemented. People’s attitudes and policy preferences on specific issues are usually based on their more deep-seated values, but they can and typically are also informed by their interests and other contextual factors. Finally, the term “public opinion” refers to the most superficial and changeable views of the public, what Worcester (1997, p. 160) called “the ripples on the surface of the public’s consciousness—shallow and easily changed”. Given these distinctions, my paper discusses the role that people’s values, public attitudes and policy preferences – but not their much more volatile opinions – should play in policy-making.

Why public views should matter

The first and most basic question to be addressed in my analysis is whether and why public values, attitudes and policy preferences should matter at all in normative debates and policy-making on asylum and the protection of refugees. Having found myself in many discussions where some people argued that the answer is “obviously yes” and others “obviously no”, I suggest that the answer to this fundamental question is not as obvious as it may seem to some.

I argue that public views should matter primarily because, in practice, they constitute an important feasibility constraint on public policy-making on asylum and refugee protection in democracies. As I explain below, this justification is mainly grounded in a realistic approach that is concerned with the nature and characteristics of policy-making processes in practice. My argument is linked to, and can gather further normative support from justifications based on the need to protect democratic self-governance and institutions. ³

Public views as feasibility constraints on effective asylum and refugee policies

My case begins with the familiar and, I believe, largely uncontested argument that if the aim of our normative reflections is to inform policy-making in practice (i.e. in the world as it is today rather than how we would wish it to be), we need to take a realistic, rather than purely idealistic, approach that accepts certain realities and considers the feasibility and long-term sustainability of policies (e.g. Betts and Collier, 2017; Carens, 1996). This is not only the starting point of realists defending the right of states to restrict immigration (e.g. Miller, 2016) but is also accepted by scholars who have advocated open borders based on appeals to cosmopolitan principles. For example, Joseph Carens has written one of the best-known ethical cases for open borders (Carens, 1987) but, when discussing the ethics of immigration policies, he switches from ideal theory to an approach that accepts certain realities (Carens, 2013). This does not mean, of course, that there are not large differences in the degrees and types of “realism” accepted and advocated by different scholars working on the political philosophy of real-world migration and migration policies.

What, then, constitutes a minimally-sufficient degree of realism, to make the normative debate meaningful for real-world policy-making? A minimal list of realities that need to be accepted in a realistic normative debate about asylum and refugee policies would surely need to include the existence of nation states and their desire to exercise sovereign control over national borders. In addition, few would dispute the importance of recognising the national politics of asylum and refugee issues. As Mathew Gibney (1999) and many others have suggested, in addition to the inherent humanitarian dimensions of the issue, how to protect asylum seekers and refugees are highly (and inevitably) political questions in liberal democracies. This is why it is important to consider “ …the role politics plays in constraining the ability of states to bring about a morally defensive response to refugees and asylum seekers” (Gibney, 1999, p. 174).

A minimally realistic approach thus requires us to recognise and consider the moral and political challenge of how to balance the basic tension between the claims and protection of refugees on the one hand, and the claims and interests of citizens of the host country. Gibney, for example, proposes that liberal democracies should respond to this challenge by using the principle of “humanitarianism”, which suggests that there is a duty to admit and protect asylum seekers as long as the costs of doing so are low (Gibney, 1999). What exactly constitutes a “low cost” is an important point for discussion. An obvious example of a “high cost” that needs to be avoided is a political backlash among the host country’s population that would undermine the new asylum and refugee policy and potentially lead to a decline (not just to no change) in the protection of refugees in practice.

If we accept the necessity of a realistic approach that recognises the need to consider the claims of citizens and the national politics of asylum and refugee protection, does that mean policy-makers must take account of public views? Put differently, are public views a constitutive part of the feasibility constraints on the making of asylum and refugee policies in rich democracies? I suggest two inter-related reasons why even a minimally realistic approach must pay at least some attention to certain types of public views, namely, values and public attitudes and preferences. These reasons relate to the role of ideas in constraining and shaping feasible policies, and to the effects of cross-national variations of ideas and attitudes on transnational policy-making.

Research on the role of ideas in policy-making suggests that, to be politically feasible and sustainable, public policies need to be compatible with the public’s fundamental beliefs and values, and they need to pay attention to people’s preferences for the principles and rules that should govern policies in specific areas (e.g. Weir, 1992). The research literature often refers to the former as people’s fundamental “worldviews” (Goldstein and Keohane, 1993) or “public philosophies” (e.g. Campbell, 1998), while the latter are frequently described as “programmatic ideas” (e.g. Schmidt, 2008). People’s fundamental values and normative beliefs matter in policy-making because they constitute an important part of the ‘background ideas’ that constrain the spectrum of policies considered to be legitimate and acceptable (see e.g. Schmidt, 2016; Campbell, 2002). Background ideas thus influence the perceived ‘social appropriateness’ of particular policies and institutions in a given context and at particular points in time (see, e.g. March and Olsen, 2011 and 1989). Public philosophies are relatively rigid and tend to change only slowly over time. Public policies that are not consistent, or at least clearly inconsistent, with the prevailing background ideas are unlikely to be feasible or sustainable in practice.

Beyond considering the consistency of policies with people’s fundamental values and normative beliefs, feasible policy-making also requires policy-makers to pay attention to people’s views on the fundamental principles and rules for policy-making in a particular area. Such “programmatic ideas” are situated in the space between public philosophies and worldviews (general and rigid) and ‘policy ideas’ (more specific and changeable). Programmatic ideas matter because they are a reflection of both people’s underlying world-views and their material interests vis-à-vis specific policy issues. Consideration of material interests obviously matters to realistic policy-making given that any policy, even one with an overtly humanitarian dimension such as asylum and refugee policy, will generate costs and benefits for different groups at different points in time.

The establishment of the Geneva Convention, which institutionalised basic principles of refugee protection in international law, was a political act that was influenced and supported, at least in part, by prevailing interests as well as supportive social and normative “public philosophies” at the time. The continued sustainability of asylum and refugee policies based on the Geneva Convention thus also depends, at least in part, on the continuing prevalence of supportive ideational background conditions as well as policy preferences that reflect the basic protection principles enshrined in the convention. Knowledge about the structure and determinants of public preferences vis-à-vis asylum seekers and refugees can, for example, provide an important warning signal and “ground for caution” (compare Swift, 1999) about potential disconnects between national asylum and refugee policies on the one hand, and fundamental public values and preferences for principles of refugee protection on the other.

Consideration and understanding of the fundamental values and public preferences for asylum and refugee policies – and their variations – are also critical for the identification of feasible and sustainable policies at a supra-national level. There is a strong case for cross-country cooperation in the regulation of asylum and refugee protection (see, e.g. Luecke et al., 2018; Betts, 2011). The European Union has, or until recently had, one of the world’s most developed cross-national institutional frameworks for cooperating on asylum and refugee issues. Six years after this system effectively collapsed, EU Member States remain deeply divided about how to reform the EU’s common asylum policies. Some Member States see the solution to the immigration challenge in “more Europe” (i.e. through greater harmonisation and EU centralisation of the system) and greater solidarity among EU countries (e.g. through a voluntary or even mandatory system of distributing refugees across EU countries). In contrast, other EU countries are pursuing national or transnational policy responses that involve a few “like-minded” European states (see, for example, the joint measures by Austria and nine Balkan states in 2016 to help “close down” the Western Balkan route for irregular migration, and the recent joint proposal by Austria and Denmark to limit considerably the right to apply for asylum in Europe). These disagreements have further deepened divisions and raised profound questions about the future of the EU and its ability to find common ground on a fundamental and, arguably, existential policy challenge.

A key question in any supra-national governance system, and one that is at the heart of European debates, is whether there are tensions between the common policies on the one hand, and national institutions, normative preferences and fundamental normative beliefs on the other (see, for example, Ruhs and Palme, 2018; Scharpf, 2010). The existence and nature of such tensions is an important factor in the design of effective common policies. For example, in her analysis of the reluctance of the Polish Government to admit more asylum seekers from Syria, Weiner (2015) argues that “a real refugee is quite different in the Polish and Western imagination”. Understanding how and why “public philosophies” and public policy preferences vis-à-vis asylum and refugee protection vary across countries, and what role these differences play in the politics of asylum and refugee protection across EU countries, is of critical importance to the design of feasible and sustainable common policies.

To conclude my case for considering public views as feasibility constraints on asylum and refugee policies, I want to address a potential counter-argument, namely, that we know from existing research that ideas need interests to carry them forward (e.g. Weber, 1946), i.e. ideas affect policy-making only if they are taken up by particular interest groups, such as political parties. For example, it is possible to imagine a situation where public views would support a particular new policy, say the re-introduction of the death penalty, but there is no political party that is willing to support the new policy preferred by the public. In that case, the extent to which public views constitute a political feasibility constraint would be much reduced. However, in the specific case of asylum and refugee protection policies, this is a highly unlikely scenario. Immigration and refugee protection have become highly politicised issues that are polarizing public discourse of political parties (e.g. Hutter and Kriesi, 2021). In most rich democracies, we see different political parties advocating a wide range of policy positions on asylum and refugee protection, including highly open and restrictive policies. This means that there is a spectrum of interests ready to exploit any gaps between public views and current policies on asylum and refugee issues.

The need to protect democratic self-governance and institutions

I would like to clarify how my argument that we need to pay attention to public views because they act as important feasibility constraints on asylum and refugee protection policies in democracies is related to the (more common) normative claim that democratic states have a right to control immigration, and that they must pay attention to “the people” when doing so, because of “democratic self-determination”.

While self-determination has been frequently used to justify a state’s right to control immigration, there are many difficult questions related to group-agency including what is the nature of the state and the political community, who exactly is the collective agent who makes decisions, and which collective is the self-determining one (e.g. Joseph, 2021; Kukathas, 2021). Most political philosophers who cite democratic concerns to justify the state’s right to exclude agree that this right is not unlimited and that it is constrained by an imperative to protect fundamental human rights such as the right to life and freedom of conscience (what Christiano 2017 calls an “authority-limiting principle”). The admission of refugees who are fleeing from threats to life is, therefore, typically – and rightly – recognised as a special case within the ethics of immigration policy, raising special duties for potential host countries that override, or at least limit significantly, democratic concerns (e.g. Song, 2018).

There is, however, a specific way in which aspects of the idea of democratic self-determination can provide normative justification for paying attention to public attitudes even in the case of asylum and refugee policies. While I agree that refugees constitute a special case that generates special duties, democratic concerns matter in the narrow and specific sense that asylum and refugee protection policies should not undermine the core features and conditions of democratic self-government (see Bauböck, 2020) including states’ “abilities to carry out their core responsibilities [such as providing security, basic rights, and justice] in a democratically accountable way” (Christiano, 2017, p.254, parentheses mine). Paying attention to public views should, in my view, be part of the assessments states make when considering the potential consequences of their asylum and refugee policies for democratic self-governance and the delivery of core state responsibilities (to avoid, for example, implementing policies that lead to the rise of populist extremist parties that threaten democratic institutions).

Why democratic elections are not enough

One response to the argument that we should take account of what the people think when discussing and designing asylum and refugee policies is that this is best done through democratic elections whereby the public selects politicians who then use their own best judgement, perhaps with the help of technical experts, to decide on the most appropriate policy. In this section I explain why I disagree with this view.

Declining trust in political institutions and actors

The question of whether political representatives should be viewed as “trustees” or “delegates” is of course at the centre of long-standing debates about democracy and representative government (e.g. Pitkin, 1967). Rehfeld (2009, p. 214) refers to the “specification of the proper relationship between citizen preferences and the laws that govern them” as the “central normative problem of democracy”. It is possible to argue that the “proper specification” depends on the policy issue at stake, including on the extent to which it is defined by fundamental moral principles (which can be a contested issue in itself).

My argument for why policy-makers should pay attention to public attitudes even outside the context of elections is not based on a particular position in this debate, but on the empirical observation that there has been a long-term decline in people’s political trust, i.e. their confidence that their political institutions serve the best public interest. Research has shown that, although the degrees of trust in political institutions vary considerably across countries, over the past few decades there has been a long-term decline in people’s trust – and a rise in distrust ⁴ – in political institutions in many high-income democracies (on Europe, see e.g. Jeannet, 2020; Dustmann et al., 2017; and Foster and Frieden, 2017; for the US, e.g. Pew Research Centre, 2021; Hetherington, 2004). A recent review of the research literature on the determinants of political trust suggests that it is unlikely that its decline can be reversed quickly or easily (Citrin and Stoker, 2018). Given the central importance of political trust to democratic representation, and the link between distrust and political alienation (Jeannet, 2020), the decline of political trust and the rise of distrust mean that the making of feasible and sustainable policies cannot afford to ignore public attitudes and preferences based on the argument that democratically elected politicians “know best”. Even if a representative was elected because the public supported their general world-view, people’s programmatic ideas, e.g. their preferences for basic principles and rules of asylum and refugee policies, can change between elections, e.g. in response to major external events such as the unexpected inflow of a large number of asylum seekers. This means that policy-makers need to pay attention to these possible changes in public attitudes and preferences.

I am not suggesting that democratic processes cannot be reformed to re-engage citizens and increase their political trust. In particular, processes of deliberative democracy – which can be defined as “a format of government in which free and equal citizens (and their representatives), justify decisions in a process in which they give one another reasons that are mutually acceptable and generally accessible, with the aim of reaching conclusions that are binding in the present on all citizens but open to challenge in the future” (Gutmann and Thompson, 2004, p.7) – can increase citizens’ political trust (e.g. Boulianne, 2019). While there have been many experiments with different types of processes of deliberative democracy, in most countries the use of “deliberative mini-publics” (such as citizens’ juries/councils/assemblies) has remained relatively marginal to core policy decision-making (Smith and Setäla, 2018). While they constitute potentially promising ways of giving citizens credible opportunities to express their views and shape policies on specific issues, processes of deliberative democracy are not yet sufficiently institutionalised to play a major role in core policy-making processes relating to asylum seekers and refugees in rich democracies.

Evidence on the links between public attitudes and policy-making in practice

Whatever we may think about the precise underlying mechanisms, a considerable body of quantitative empirical research has shown that there is a link between public attitudes and policy-making in practice (for reviews of the relevant literature, see e.g. Wlezien and Soroka, 2016; Burstein, 2003; Manza and Cook, 2002). The positive relationship is based on two way causalities: public attitudes affect public policies and, at the same time, attitudes are also responsive to changes in policies (see e.g. Soroka and Wlezien, 2010). As can be expected, the strength of the link is contingent on a number of factors, especially the salience of the policy domain (i.e. in policy domains with higher salience, the link between public attitudes and policy making is stronger) and, to some extent, also political institutions (such as electoral systems) although the evidence on the mediating role of the latter is mixed. A recent study of the nexus between attitudes and policy in Europe found a positive relationship whose strength varies across policy domains but not across political institutions in different countries (Rasmussen, Reher, and Toshkov, 2019).

Although relatively few studies have provided systematic analysis of whether and how public views shape immigration policies, and even fewer have explored the link between public attitudes and asylum/refugee policies, the research that has been done so far suggests a significant relationship. For example, in his recent analysis of the restrictiveness of immigration policies (with a focus on internal controls after admission) and public attitudes in 25 European countries during 2002 and 2016, Böhmelt (2019) found evidence of a positive relationship, but only in countries with a large foreign born population i.e. when migration is likely to be more salient as a public policy issue.

The above-mentioned study by Rasmussen, Reher, and Toshkov (2019) included asylum seekers’ right to work as one of the policy issues analysed, and it found a positive although statistically insignificant relationship between public attitudes and policies across European countries. I am not aware of any systematic research that has included a more comprehensive analysis of the responsiveness of asylum and refugee policies (going beyond asylum seekers’ right to work) to public attitudes and ideas. However, as refugee protection has become a highly salient policy issue in many countries in recent years, it is reasonable to expect that a link between attitudes and policies exists also in this area of public policy.

Too misinformed, messy, and volatile? How public views can and should matter

My argument that public views should matter in the regulation of asylum and refugee protection in rich democracies requires that there are relatively stable and meaningful public values, attitudes, and policy preferences that can actually be measured and identified. Clearly, if “what the people think” was highly volatile, and/or if it was impossible to cut through people’s superficial opinions to identify their “real” values and preferences, public views could not constitute a feasibility constraint on asylum and refugee policies in any meaningful and practical way.

My aim in this section is to discuss four issues that are frequently raised to cast doubt on the idea that there are meaningful and identifiable public views and policy preferences that can and should be considered in public policy-making. These critiques relate to the role of (mis)information and narratives in shaping people’s views; the potential disconnect between individuals’ views and policy realities; polarization and conflicted views; and, more generally, the degree of stability of people’s beliefs, attitudes and preferences. While we clearly need to take a critical approach to the data in which each of these potential critiques needs to be taken seriously, I argue that none of these issues constitutes grounds for rejecting the idea that there are identifiable public views that act as political feasibility constraints on asylum and refugee policy-making in practice.

Information, media, and narratives: How much do they matter?

When formulating their views on a particular policy issue, most people clearly do not know and understand all the relevant policy realities and complexities. This is simply because people’s cognitive capacities, and the available information to them, are not only variable across individuals and policy issues but they are also, and always, limited in one way or another. People’s views on asylum and refugee policies are clearly no exception. Even among those of us who dedicate most of our professional lives to studying the issue, who can confidently say that that they know and understand all the relevant asylum and refugee laws and policies including the consequences of different policy designs? Given the high political salience of the issue in many countries, it is not surprising that migration has become a key target of disinformation campaigns (e.g. Butcher and Neidhart, 2020). Research has shown that most people have considerable misperceptions about the numbers and composition of migrants in their countries (e.g. Alesina et al., 2018) and a poor understanding of relevant policy frameworks. Public perceptions of the effects of migration often have little to do with realities (e.g. Blinder and Markaki, 2019; Ipsos Mori, 2014).

The inevitable gap between people’s understandings and policy complexities has been frequently used in support of the argument that public views should not matter at all in policy-making in democracies (e.g. Weissberg, 2001). ⁵ I disagree with this argument because the role of public attitudes as feasibility constraints on asylum and refugee policies does not depend, at least not fully, on the extent to which people’s views are based on facts, careful reflection and reasoning. Put differently, my argument why public views constrain the feasibility of policies does not require these views to be evidence-based and “rational”. Their presence is enough to matter, at least in a democratic setting (also see Runciman, 2018).

Nevertheless, the point about people’s limited capacities and (mis)information has important implications for how their public attitudes and preferences should matter for policy-making. Most fundamentally, it is important to ask whether and to what extent people’s preferences are shaped by the information available, and this includes assessing the role of (dis)information campaigns on what and how people think about a certain issue. Clearly, if it was the case that people’s views on a particular policy issue were highly dependent on their knowledge of the relevant facts, the provision of better information (e.g. through “myth busting” and information campaigns) could change public views and therefore reduce the extent to which they are feasibility constraints in practice. However, the effects of providing accurate information on people’s views on immigration appear to be relatively small (e.g. Hopkins et al., 2019). Studies that find large effects conclude that the magnitude of the impact declines considerably over time (e.g. Facchini et al., 2016). Furthermore, changes in attitudes because of better information might not translate into equivalent changes in policy preferences (e.g. Grigorieff et al., 2016).

Similarly, research suggests that the effects of media on attitudes and preferences may not be durable (e.g. Chong and Druckman, 2007). It is also important to keep in mind that media might not only affect but also be affected by public attitudes (Allen et al., 2019). Denison and Dražanová (2018) suggest that the primary impact of media is likely to be on the salience of the issue, rather than on attitudes themselves.

While information and media effects are likely to be limited and may not last very long, there is evidence in the existing research literature that politicians can and do affect public attitudes, at least to a certain extent, either by changing people’s socio-economic situations and/or by persuasion and framing (Matsubayashi, 2012). For example, Boswell and Hampshire (2017) show how policy-makers in Britain and Germany have been influenced by “public philosophies” but also been able to strategically mobilise and foreground different types of ideas in immigration policy-making. If policy-makers can influence public attitudes and preferences through particular frames and narratives, the extent to which public views constitute a constraint on policy-making is obviously reduced (compare Manza and Cook, 2002). The broader point here is that, while the evidence suggests significant stability in values and attitudes towards migrants (see the discussion further below), the political feasibility constraint created by public views is not fully exogenous and fixed, especially in the longer run. This is why, as stated at the outset of the paper, public views are a type of “soft” rather than “hard” constraint on asylum and refugee policy-making.

Individuals’ views vs policy realities: Who and what do people have in mind?

Another common critique of using data on public views for policy-making is that there can be a large gap between how the public formulate and express their views on the one hand, and the realities and processes of policy-making on the other. This gap arises for a range of reasons, some of which can be addressed, at least partly, by careful survey design. A first obvious but important issue relates to the (mis)match between the migrant ‘target group’ of the policies (in our case, asylum seekers and refugees) and who the public have in mind when expressing their views (Blinder, 2015). Public attitudes and policy preferences vis-à-vis migrants and immigration are likely to vary across different types of migrants such as labour migrants, family migrants, foreign students, and asylum seekers and refugees. This is partly because public and policy responses to the admission and integration of different categories of migrants involve differential degrees of humanitarian considerations. For example, while it is widely accepted that humanitarian considerations (should) loom large in asylum and refugee policies, they typically play no role in policy-making on labour immigration. To understand how public attitudes constrain asylum and refugee policies, we need fine-grained attitudes data on public views toward these specific groups and policies, rather than toward migrants and migration policies in general (Blinder, 2012).

A second problem is that people can have unrealistic expectations about what asylum and refugee policies are feasible in practice. Public views aside, the design and implementation of public policies is typically characterized by a range of constraints (e.g. domestic and international legal constraints), trade-offs (e.g. between the number of people protected and the standard of protection provided, assuming a limited budget available), and a range of financial and other risks (e.g. uncertainties about the short and long-term costs associated with different refugee policy options). These factors are not necessarily or even typically considered by the public when they formulate and express their policy preferences. While some consider this disconnect reason enough to conclude that policy-makers cannot – and indeed should not – look to the public when designing public policies (e.g. Weissberg, 2001), there are advances in survey methodologies and experimental methods that can reduce some of these problems. For example, potential costs and trade-offs associated with particular policy design can be built into surveys and interviews.

A third challenge with relating data on public attitudes and policy preferences to real-world policy choices stems from social desirability bias, i.e. when expressing views on sensitive topics some people may be reluctant to express their real attitudes and preferences because they want to present themselves in the best possible light and avoid, for example, coming across as being “politically incorrect”. This issue is likely to be important in the context of asylum and refugee policies, a policy field characterized by strong moral norms and where some people might be reluctant to express views that openly deviate from what is considered socially acceptable. The degree of social desirability bias is known to vary across individuals and there are a range of methodological strategies that can be adopted to reduce this problem (see e.g. Hanmer et al., 2014; Krumpal, 2013). For example, in survey experiments conjoint designs can be used to ask people about their support for multi-dimensional policy packages rather than posing direct questions about specific policy views which some people may be reluctant to disclose.

Polarization and conflicted views

A key issue in assessing how public views should influence policy-making, including on asylum and refugee issues, relates to the extent to which there is agreement or disagreement in basic beliefs, attitudes and policy preferences among the population, and to the degree to which these disagreements are linked to larger societal cleavages. There has been a large amount of research on the increasing and new forms of polarisation of politics and societies around the world (e.g. Klein 2020 on the US). In Europe, research has shown that there has been a significant change in the social basis of politics that takes the form of new cleavages between the winners and losers of globalisation (e.g. Kriesi et al., 2008; 2012). These new cleavages along pro- and anti-globalisatlion lines are closely associated with immigration (e.g. Hooghe and Marks, 2018; Ademmer and Stöhr, 2019). Attitudes to immigration, including asylum seekers and refugees, can therefore be expected to be shaped by, and potentially tightly linked to these larger cleavages in people’s values and identities.

A high degree of polarization in attitudes toward asylum and refugee protection does not mean, as some people argue, that public views cannot or should not inform policy-making on these issues. The extent to which attitudes and policy preferences are polarised, and whether and how these divisions are linked to larger cleavages in values and identities, are critical pieces of information for the design of effective and sustainable asylum and refugee policies. If views on asylum and refugee protection are highly divided, feasible policy-making needs to find a way of bridging at least some of these divides.

Similarly, it is important to recognise that, in addition to disagreements between different people, individuals may also be internally conflicted about certain policy issues. In other words, their attitudes and policy preferences may be characterised by ‘ambivalence’ which refers to the presence of multiple and frequently conflicting considerations about important issues (Zaller and Feldman, 1992, p.584). One potential explanation for such ambivalence may be a conflict among the fundamental values underlying people’s attitudes and policy preferences. Arguably, this is likely to be the case with asylum and refugee policies where the tension between control (or, more broadly, the national interest) and the protection of refugees (or minority groups more generally) is not only an issue for policy-makers but also a dilemma which most individuals face when forming and expressing their views on asylum and refugee policies (see Jeannet et al., 2021). Knowledge about the existence and nature of these internal conflicts, and understanding how people negotiate competing objectives when formulating their policy preferences, is critical to effective and sustainable policy-making.

Stability

Another common argument against taking account of public views in policy-making, both in general and on asylum and refugee issues in particular, is that people’s attitudes and preferences are not very stable, in the sense that they are not consistent over time. As discussed in the introduction, potential instability in beliefs and attitudes over time is in fact a key characteristic of a “soft” feasibility constraint, so pointing to potential change over time does not, in itself, undermine the importance of taking public views seriously as a constraint on feasible policy-making in a particular context. The focus on the softness of the constraint does, however, make it important to ask and assess critically whether, why and how quickly public views change over time.

Research suggests that there is a considerable degree of instability in individual’s attitudes across many issues, and that key factors causing instability include weak attitudes and external stimuli (Druckmann and Leeper, 2012). A key reason for weak attitudes is low salience of the issue in public debates. As mentioned earlier in the paper, migration and asylum have become highly politicised issues in many rich democracies, i.e. with high salience and significant degrees of polarization, which suggests that people are more likely to hold strong rather than weak attitudes and policy preferences. A recent study of attitudes in the US and Europe has found that “immigration attitudes are remarkably stable over time and robust to major economic and political shocks. Overall, these findings provide more support for theories emphasizing socialization and stable predispositions rather than information or environmental factors” (Kustov et al. 2021, p.1478).

In the context of asylum and refugee policies, a key question related to external events is whether and how public policy preferences have changed over the past few years when many high-income countries experienced unexpected spikes in inflows of asylum seekers. Have these events, and the political crises that ensued (especially in European countries), led to a shift in people’s policy preferences vis-à-vis asylum and refugee protection, or even in their fundamental values about solidarity and minority protection? If so, are these changes in policy preferences and/or fundamental values likely to be enduring? Debates and policy proposals that refer to public attitudes as a justification for fundamental policy reforms need to engage critically and explicitly with these questions.

Reforming asylum and refugee policies in Europe

The remainder of this paper draws on the considerations above to discuss critically the implications of public views for asylum and refugee policies in Europe. As discussed in the introduction, there is broad agreement among European countries that asylum and refugee policies in Europe need to be reformed, but there is considerable disagreement both across and within EU countries about how this should be done. While a wide range of policy aspects has been discussed, two of the most fundamental policy questions are: first, whether future policies should remain based on the principles of the Geneva Convention, with some countries wanting to limit the right to asylum in Europe (e.g. Austrian Ministry of Interior and Danish Ministry of Integration, 2018) and increasing evidence that some border enforcement measures supported by the EU include “illegal pushbacks” that violate the principle of non-refoulement (e.g. Human Rights Watch, 2021; Morales, 2019; Council of Europe, 2019); and second, whether and how European countries can increase their cooperation and improve their common policies. What do we know about public views on these issues, and how should they matter for policy-making?

Salience and polarization

As discussed earlier in this paper, the extent to which public attitudes matter, and should matter, for policy-making partly depends on the importance of the policy issue in people’s minds. Unfortunately, and perhaps somewhat surprisingly, to the best of my knowledge there are no systematic longitudinal and cross-country data on the salience of asylum and refugee issues among people in European countries. We do, however, have data on the salience of immigration in general. Notwithstanding the importance of distinguishing between immigration and the more specific issues of asylum and refugee protection, the high share of asylum seekers and refugees among new migrants coming to Europe in recent years means that public attitudes data on the salience of immigration can be expected to be heavily influenced by the importance people attach to asylum and refugee issues.

The sharp increase in the numbers of asylum seekers and other migrants in 2015-16 was associated with a considerable rise in the salience of immigration among the European public. According to data from Eurobarometer (also shown in Figures A1-A3 in the Online Appendix), when asked to identify the two key issues facing the EU, the share of EU citizens mentioning immigration increased considerably between 2013 (10%) and 2016 (almost 50%) and, despite a decline over the past four years, has remained relatively high in recent years (over 30% in 2019 and 25% in 2021). When measured in terms of key issues facing respondents’ countries (rather than the EU), the salience of immigration appears lower although it has followed a similar pattern over time: There was a sharp increase between 2014 (when 12% of people mentioned immigration as a key issue for their countries) and 2015 (28%, second only to “unemployment” mentioned by 33%), and a decline since then (falling to 17% in 2019 and 10% in 2021). As is well known, the share of people saying that immigration is a key issue facing themselves is much lower than the share of people saying that it is a key issue for their countries and the EU. In fact, it is striking that the share of people mentioning immigration as a key issue for themselves has remained very low and stable throughout 2009-21 (never exceeding 10%).

While the over-time trend in the salience of immigration has been similar in most European countries, there has been considerable variation across European countries (see Figure A4). For example, in November 2015, the share of people who considered immigration as a key issue for the country was 76% in Germany, 60% in Denmark, 56% in Austria, but only 17% in Poland, 11% in Ireland, and 9% in Spain. Portugal, Ireland, and Spain are interesting outlier cases where the salience of immigration has been very low and remained essentially unchanged during 2015-16. While there has been limited systematic research on the determinants of the salience of immigration in Europe in recent years, a recent study (Dennison, 2020) confirmed that the realities of immigration (i.e. the actual inflows and numbers of migrants) have played a key role. Although inflows have sharply declined since the spikes in 2015-16, there is continuing uncertainty, exploited by many political parties, about the scale and volatility of future flows. Many European countries are now grappling with long-term challenges with the integration of the new migrants who arrived a few years ago. Recent events in Afghanistan, where the take-over of the Taliban in September 2021 led to large-scale forced displacement, fuelled public debates in many European countries about the possibility of new large inflows of asylum seekers.

There is also evidence of rising polarization of public attitudes to immigration in many European countries (Heath and Richards, 2018; Ademmer and Stöhr, 2018) which has increased the politicisation of immigration and asylum issues (e.g. Pasetti and Garcés-Mascareñas, 2018). Rising numbers of refugees can and often do increase support for populist right-wing parties (e.g. Dinas et al., 2019; Steinmayr 2021; Rudolph and Wagner, 2022) which means that these parties have a strong interest in keeping immigration high on the public agenda. Radical right success can also affect the policy positions of mainstream political parties (e.g. Abdou-Chadi and Krause, 2020) including their responsiveness to public opinion on immigration (Vrânceanu, 2019).

Taken together, all these factors suggest that immigration in general, and asylum and refugee issues in particular, are likely to remain of high importance to Europeans in the coming years. As a consequence, public views are likely to be stronger constraints on the space of feasible and sustainable policies available to policy-makers. High degrees of polarization suggest that, to be effective and sustainable, policy-making on asylum and refugee issues needs to find ways of at least partially bridging the divides in views among the public. This has implications not only for policy outputs but also for policy processes such as the extent to which people with opposing views are included in debates and decision-making processes as well as political communication (e.g. Global Forum on Migration and Development 2019).

Basic values and public support for asylum and refugee protection

As discussed in section 2 of this paper, people’s basic human values constitute an important part of the “background ideas” that can constrain feasible policy-making. While people’s fundamental values tend to change only slowly, if at all, major external events could have an impact. So, what do we know about Europeans’ fundamental values and how they have changed over the past few years when many European countries experienced significant volatility in the inflows of asylum seekers and other migrants?

Schwartz (1992, 2012) distinguishes between ten human values which can be grouped into four categories: self-transcendence, conservation, self-enhancement, and openness to change. The former two dimensions are most relevant to asylum and refugee issues. Within the Schwartz classification, “conservation” is based on three human values: “security”, “conformity”, and “tradition”, while “self-transcendence” is based on “universalism” and “benevolence”. In the context of asylum and refugee issues, the two dimensions thus relate to ‘control / national interest’ and 'protection', respectively. Data from the European Social Survey (ESS) suggest that there has been very little change over time in Europeans’ self-identification with "conservation" and “self-transcendence” during 2002-2020 (see Figures A5 and A6 in the Appendix). Based on these data, there is no evidence that the spikes in inflows of asylum seekers and other migrants in 2015-16 have had a noticeable impact on Europeans’ fundamental values. This suggests that there has been no fundamental shift in the background ideas and values that underpin feasible policy-making on asylum and refugees.

The European Social Survey also includes questions about attitudes to immigration including a small number of questions about attitudes to asylum seekers and refugees. In 2002, 2014, 2016, the survey included the following question:

Some people come to this country and apply for refugee status on the grounds that they fear persecution in their own country. Using this card, please say how much you agree or disagree with the following statements. Firstly… “The government should be generous in judging people's applications for refugee status”.

Respondents were asked to indicate if they: disagree strongly (1); disagree (2); neither agree nor disagree (3); agree (4); or agree strongly (5) with this statement. The results for 2014-15 (ESS round 7) and 2016-17 (ESS round 8) are shown in Figure A7 in the Appendix. In most countries included in the ESS, fieldwork for round 7 was completed in early 2015, i.e. before the peak of the inflows of asylum seekers and other migrants, and round 8 did not start until late 2016, i.e. after the peak (Heath and Richards, 2018). The data show that in most countries attitudes became more negative during these few years. This is in contrast to stable or improving attitudes toward immigration in general during this period (Ademmer and Stöhr, 2018; Heath and Richards, 2018). However, a comparison of attitudes over a longer time period (since 2002 when ESS Round 1 was conducted) shows that attitudes to refugees improved considerably between 2002 and 2016-17, despite the moderate worsening of attitudes in many countries between 2014-15 and 2016-17 (see Figure A8). It is important to emphasise that even after the peak of inflows on 2015-16, in many countries attitudes to refugees remained, on average, relatively moderate.

It is also important to recognize that, despite common trends over time in many countries, there are some important cross-country variations that are likely to have consequences for common policy-making on asylum and refugees. For example, according to the ESS data, in 2016-17 public attitudes towards refugees in the Czech Republic, Estonia, Hungary, and the Netherlands were considerably worse than in the other European countries included in the survey. A Pew Research Centre survey of Europeans’ attitudes to “taking in refugees from countries where people are fleeing violence and war” confirms the existence of considerable cross-country variations, with national averages ranging from 86% support in Spain to 32% support in Hungary (Connor 2018 - and see Figure A9). Similarly, surveys carried out by the polling company Ipsos suggest considerable variation in the share of the population who agree that “We must close our borders to refugees entirely - we can't accept any at this time", ranging from 51% in Sweden to 33% in Britain in 2019 (2019 - also see Figure A10).

To consider whether and how to respond to public attitudes, it is important to understand its determinants. The rapidly growing research literature on the determinants of attitudes toward asylum seekers and refugees in Europe indicates that the number of asylum seekers and refugees, and especially the type of exposure of the host country’s population to them, are likely to matter. Based on the “contact hypothesis” (for a recent review, see Paluck et al., 2019), it is possible that more sustained contacts and meaningful interactions between local residents and asylum seekers will be associated with more positive attitudes among the local population than mere short term exposure. For example, in their analysis of the impacts of the refugee crisis in Greece, Hangartner et al. (2019) find that direct exposure to refugees increased local residents’ hostility toward asylum seekers and refugees. Importantly, the study examined the attitudes of residents of Greek islands where refugees were passing through rather than staying for prolonged periods of time. Similary, Rudolph and Wagner (2022) recently found that the presence of asylum seekers has substantial negative effects on public attitudes to asylum seekers in Austria. The authors argue that the limited (and thus lack of meaningful) contact between local residents and asylum seekers is likely to explain part of this negative effect. Another recent study of the Austrian case finds that exposure to refugees can increase support for the far right. However, the same study shows that contact and sustained interactions between local residents and asylum seekers decreased votes for the far right (Steinmayr 2021).

Existing research also suggests that a key factor shaping attitudes toward refugees in Europe is their perceived “deservingness”, i.e. the extent to which they are viewed as having fled from war and persecution rather than migrated for mainly economic reasons (e.g. Herzowicz and Raijman, 2020; Hager and Veit, 2019; Czymara et al., 2017; Bansak et al., 2016a). Perceived economic and cultural threats have been found to matter as well (e.g. Bansak et al., 2016a) although they appear to be less important than perceived deservingness. Hager and Veit (2019) suggest that perceived threats gain in importance as a negative driver of attitudes when the perceived deservingness of refugees is low, i.e. there are interaction effects between deservingness and perceived threats. From a policy perspective, this insight underscores the difficulty with, and potential dangers of blurring distinctions between economic migrants and refugees in policy-making. Bansak et al. (2016b, p. 2) suggest that “to ease social tensions and create greater acceptance of asylum seeks from all walks of life, policy-makers should speak clearly to their deservingness, vulnerability, and ability to contribute to their new homes”.

Most existing studies also find that the religion of asylum seekers matters. Europeans’ attitudes to Muslim asylum seekers appear to be worse than to Christian asylum seekers (e.g. Bansak et al., 2016a). The magnitude of this effect remains unclear, however (e.g. Hager and Veit, 2019). Taking account of individuals’ religion when assessing their claims for asylum would, of course, not be compatible with the Geneva Convention or with most people’s fundamental understandings of liberal democracy.

Policy preferences

In contrast to the rapidly growing research on the determinants of attitudes to asylum seekers and refugees in Europe, there are still relatively few analyses of Europeans’ preferences for different aspects of asylum and refugee policy. A key finding from the small existing research is that, while there is no evidence of widespread public support for highly restrictive policy features that completely abolish protections for asylum seekers and refugees, Europeans prefer more control-based asylum and refugee policies, including EU policy cooperation with non-EU countries of migrants’ transit and origin, that are not completely aligned with the rights-based policies of the current international refugee system (with the Geneva Convention at its core). The available research also suggests that there might be more public support for common European policy-making, including for redistributing asylum seekers across European countries, than is often assumed in public debates.

Jeannet et al. (2021) conducted a conjoint experiment with 12,000 respondents across eight European countries to study the nature and determinants of public support for multi-dimensional asylum and refugee policies. The results suggest that Europeans are generally committed to policies that provide protection to asylum-seekers and refugees but this commitment tends to be contingent upon the use of policy controls such as limits and conditions (e.g. annual limits on asylum applications; an income requirement for family reunification; and making financial assistance to non-EU countries conditional on their help with reducing the numbers of asylum seekers coming to Europe). This raises the important question whether policies that remain based on the current international framework for refugee protection (incl the Geneva Convention) are politically sustainable. The answer partly depends on the determinants of these policy preferences.

In a related study based on the same experiment, Jeannet et al. (2020) show that individuals’ trust in the political institutions of the European Union has a central role in the formation of their asylum and refugee policy preferences. As can be expected from the research literature on the relationship between political trust and support for public policy (e.g. Hetherington, 2004), individuals with lower levels of political trust in European institutions were found to be less supportive of asylum and refugee policies that provide expansive, unlimited, or unconditional protection and more supportive of policies with highly restrictive features. The study (Jeannet et al., 2020) also shows, however, that even politically distrusting individuals can systematically support policies that provide protection and assistance to refugees if there are limits or conditions on policy provision. The authors suggest that the use of limits and conditions can compensate for individuals’ lack of trust in generating supportive preferences for asylum and refugee protection. As political trust changes only slowly and has been in long-term decline in many European countries (e.g. Jeannet, 2020), it seems unlikely that efforts to re-build Europeans’ political trust will, on their own, be an effective policy response to these public policy preferences.

There is also emerging evidence on how European voters think about EU cooperation with non-EU countries (including migrants’ origin and transit countries) to regulate irregular migration and refugee protection. The EU’s New Pact on Migration and Asylum (2020) emphasises the need to strengthen such international migration policy-cooperation. Recent analysis of public preferences vis-à-vis the EU-Turkey migration deal struck in 2016 found that German and Greek voters are favourable to several core features of the current EU-Turkey cooperation regarding the return of irregular migrants, financial aid to refugees in Turkey, and the intensity of Turkish border controls (Vrânceanu et al., 2021). The study also finds evidence of public support for new modes of resettlement of refugees directly from Turkey to the EU, and of opposition among German and Greek voters to ‘pushbacks’ of irregular migrants from Greece to Turkey. The study indicates that European voters’ preferences appear to be driven by a combination of instrumental (national interest) and humanitarian considerations (thus echoing the findings in Jeannet et al., 2021) as well as a desire for reciprocity in EU-Turkey policy cooperation on migration.

With regard to common EU policy-making on asylum and refugee issues, the studies by Jeannet et al. (2021, 2020) find that, compared to the status quo of national governments assessing and deciding applications for protection in their countries, having a fully centralised EU asylum system that considers and decides on applications for asylum throughout Europe would significantly reduce public support for asylum and refugee policy. While this finding illustrates public resistance to complete centralisation of EU asylum policies, it does not necessarily mean that there is no support for different types of cooperation on asylum across EU countries. Indeed, in contrast to the popular impression that people across different EU countries are hopelessly divided on these issues, Jeannet et al. (2021) found that the pattern of public preferences (for protection based on limits and conditions) is remarkably similar in both the old and more recent EU Member States included in the study, including in Hungary and Poland, two countries whose governments have been openly hostile to more expansive asylum and refugee policies. The evidence thus suggests that there is room for more effective cooperation on asylum and refugee policies in Europe, based on policies that provide protection based on limits and conditions. While this obviously still leaves open the important question of what exactly those limits and conditions should be, it constitutes an important corrective to the idea, often heard in policy circles and public policy debates, that people in some European countries (e.g. in some East European states) have fundamentally different asylum and refugee policy preferences than people in other EU countries (e.g. some West European states) and that this justifies the case against a more harmonised European asylum system.

A key question in European asylum and refugee policy debates is whether and how asylum seekers should be distributed across EU countries. Before its collapse, the Dublin system assigned the responsibility of processing asylum applications to countries of first entry, i.e. to EU states on the outer borders of the EU. This system clearly disadvantaged EU countries such as Greece, Spain, and Italy. EU states have over the past few years been bitterly divided over how to reform the system. One of the European Commission’s first responses to the rapid increase in asylum seekers coming to Europe in 2015 was to call for greater responsibility-sharing across EU member states, specifically in the form of an EU-wide mechanism that would re-distribute and allocate refugees across member states using an agreed distribution key based mainly on the size of countries’ populations and economies (GDP). Although the relocation policy and mechanism were agreed in 2016, very few EU countries accepted their allocated share of refugees, with some openly opposing the policy, so that the policy quickly ended in failure.

A recent survey of 18,000 people across 15 European countries explored public preferences for different mechanisms for allocating asylum seekers across countries (Bansak et al., 2017). Despite the political debates and failures over this issue over the past few years, the study found that Europeans’ most preferred system for allocating asylum seekers is one based on proportional allocation (which the survey defined as based on GDP, population size, unemployment, and number of past asylum applications). This preference was found to be consistent across countries and people with different political orientations. Importantly, the public preference for such a proportional system declined but persisted even when people were told about the consequences of such a new system for the inflows of asylum seekers to their countries (which, for many countries, would increase). The authors concluded that “considerations of both con­sequences and fairness shape voters’ preferences over asylum allo­cation policy. Yet when the two collide, the norm of proportional equality overrides consequentialist preferences for most voters” (Bansak et al., 2017, p. 5).

A more recent paper confirms that numbers of asylum seekers, and a country’s past experience with immigration more generally, matter for people’s support for redistributing asylum seekers across European countries. Based on an analysis of Eurobarometer data for 2015, Heizmann and Ziller (2020) find that people who live in a country with a higher share of asylum seekers are more likely to support EU-wide redistribution of asylum seekers (their dependent variable captures support for the idea of re-distribution in general, without specifying the criteria used). In line with the conclusion about the central role of political trust in the analysis of Jeannet et al. (2020), Heizmann and Ziller (2020) confirm the importance of individuals’ evaluations of the EU which, their study finds, are related strongly to support for a common EU policy of redistributing asylum seekers across EU countries.

The role of media, political discourse, and information in shaping Europeans’ views on asylum and refugee protection

As discussed earlier in the paper, to assess how public attitudes and preferences should affect policy-making, it is important to consider the potential role and effects of the accuracy of information available to individuals and the framing of asylum and refugee issues by media and political leaders. Although there is a growing body of research that analyses the role of media in shaping public attitudes to migration and migrants in general (e.g. Strömbäck et al., 2021), there has, to the best of my knowledge, been very little systematic research on how Europeans’ attitudes to asylum seekers and refugees (specifically) have been influenced by media coverage and political discourse in recent years. As a consequence, there is no justification for claims, sometimes made by supporters of more generous refugee protection policies in Europe, that individuals’ concerns about asylum seekers and refugees are largely based on skewed media frames and political discourses. Of course, media and discourses are likely to play some role but I am not aware of evidence suggesting that they are the primary factor driving Europeans’ views in the specific area of asylum and refugee protection.

Most existing studies have been focused on how the media have framed asylum and refugee issues in recent years, and how the framing has changed over time. For example, Greussing and Boomgaarden (2017) showed that as the refugee crisis unfolded during 2015, Austrian media coverage evolved from employing multiple frames (including humanitarianism and victimization of refugees) in the early stages to a more narrow set of frames focused on border management and economic impacts in the later stages (also see Vezovnik, 2018; and Georgiou and Zaborowski, 2017, who reach similar conclusions based on cross-country analyses). Reviewing the existing literature, Consterdine (2018) concludes that while political discourse has largely mirrored media discourse of the “refugee crisis”, there have been considerable cross-country variations in political discourse around asylum and refugee issues, and “whether the media sets the agenda on such framing, or whether the media takes cues from political discourse (or vice versa) is contested” (p. 23).

With regard to the role of information, the high salience of the issue in recent years suggests that, on average, people’s general knowledge about asylum and refugee issues would have increased. This does not mean, of course, that there are not still many misperceptions and misunderstandings, and/or that they do not play a role in the formation of Europeans’ attitudes and policy preferences vis-à-vis asylum seekers and refugees. However, the existing research suggests that attitudes and policy preferences are not only (or perhaps even not mainly) driven by perceived impacts but, rather, by the principles and design of asylum and refugee policies. While “more and better information” may improve people’s understandings of the impacts of asylum seekers and refugees, it is, arguably, unlikely to have large effects on their preferences for particular principles and policy designs.

Conclusion

There has been a considerable amount of empirical research on the characteristics and determinants of public attitudes and public policies, and on how public attitudes and public policies influence each other in practice. At the same time, political theorists have long debated the normative question of whether and how public attitudes should influence policymaking. This paper has brought together and drawn on the insights from these different research literatures to discuss why and how public attitudes should matter in the specific context of policy-making on asylum and refugee protection in rich democracies.

I argued that public views should matter in policy-making on asylum and refugee policies because of the need for a realistic approach that recognizes the politics of asylum and refugee protection, and because public values and attitudes are an integral part of the feasibility constraints on effective and sustainable policy-making within such an approach. Paying attention to public views can also be justified by the need to ensure that policy-making on asylum and refugee protection does not undermine (e.g. through the rise of extremist anti-democratic parties) basic institutions of democratic self-governance including states’ abilities to carry out their basic responsibilities vis-a-vis their citizens.

Following recent philosophical work on political feasibility, I consider public views as a “soft” constraint on feasible policies, meaning that its nature and characteristics may, at least in principle, change over time. It is clearly necessary to take a critical approach to identifying what the people “really think” and to consider how and under what circumstances public views can constrain policy-making, and how this may change over time. As I have argued, such a critical approach needs to consider and engage with questions about the role of (mis)information and narratives in shaping people’s views; the potential disconnect between individuals’ views and policy realities; the extent of polarization and conflicted views; and, more generally, the degree of stability of people’s beliefs, attitudes and preferences vis-à-vis asylum and refugee protection. A misreading or misunderstanding of public views can have highly adverse effects on the moral and political foundations of polices towards asylum seekers and refugees and thus ultimately undermine refugee protection in practice.

Applying these considerations to the European context, I have suggested that public views have become and are, at least in the short to medium term, likely to remain an important feasibility constraint because of the rising degrees of politicisation of refugee and other migration issues in many European countries, driven in part by an increasing polarization of public attitudes and political discourse. I found no fundamental shift in the background ideas and values that underpin feasible policy-making on asylum and refugees in European countries in recent years, meaning there is no evidence of a major change in Europeans' basic values as a result of the 2015-16 “refugee crisis”. Average attitudes in most countries are fairly moderate (i.e. not strongly in favour or against generous refugee protection) although there are some important country differences. The perceived deservingness of refugees is a key determinant of attitudes, which points to the difficulties and potential dangers of blurring categories (e.g. refugees and economic migrants) in policy-debates and policy-making.

In contrast to claims made by many politicians in recent years, there is no evidence of widespread public support for highly restrictive measures that eliminate protections for asylum seekers and refugees. While some European politicians have in recent years called for a “zero-asylum” approach, I have found no evidence in existing research that public views are a constraint on the continuing existence of an asylum system in Europe. Instead, the available research on policy preferences shows that Europeans are committed to providing refugee protections but they have a preference for a greater use of limits and conditions than is currently the case. There is also emerging evidence suggesting that European voters are supportive of certain forms of EU cooperation with non-EU countries, aimed at reducing irregular migration and protecting refugees. The available research further suggests that common European policy-making on certain aspects of refugee protection, including proportional redistribution of asylum seekers across European countries, may be more acceptable to the public than to the many European policy-makers who have vigorously opposed it in recent years. The implication is that, as Bansak et al. (2017) have also observed, public views are unlikely to be the major obstacle to well-designed and more cooperative policy-making on asylum issues in Europe.

Overall, what do Europeans’ views and preferences for more controls (limits and conditions) in asylum and refugee policy mean for the feasibility of policy-making which continues to adhere to the core principles of the Geneva Convention, specifically European states’ duties to examine asylum applications and respect the principle of non-refoulment? As public views are a type of “soft constraint” any answer to this question necessarily involves making a range of judgements, including assessments of the likely malleability and future changes of public views in this area. These judgements need to be informed by the best available evidence some of which I have discussed in this paper. Although more research is clearly needed to understand better the nature of public views and their determinants, both supporters of the continuation of the status quo legal framework and advocates of “paradigm change” in asylum and refugee protection should, at the minimum, take note of these views and preferences and consider critically their stability and potential consequences.

My own judgement is that European publics’ views on asylum and refugee protection, as I have understood and explained them in this paper, do not suggest that European states’ adherence to the core principles of the Geneva Convention has become unfeasible. However, there is a need to consider and implement policy measures that introduce a greater degree of control over the number of asylum seekers and refugees, and the manner in which they reach and are protected in Europe and elsewhere. For example, there is a need for measures that help shift the balance of beneficiaries of protection away from asylum seekers toward greater numbers of already-recognized refugees who are resettled directly from conflict regions. Most asylum seekers applying for protection in Europe are forced to cross borders illegally and their numbers are highly variable year-on-year. In contrast, the resettlement process is much more regulated and provides host countries with much larger degrees of control. Ferracioli (2014) makes a similar argument in her discussion of the appeal and prospects of negotiating a new Refugee Convention which, she argues, can be ‘rendered’ more feasible by measures that reduce public concerns about procedural and distributive fairness.

Expanded resettlement is, however, unlikely to be feasible unless simultaneous measures are taken to introduce more control in the asylum system itself. This could include temporary limits on the number of new asylum applications that a country considers each year (a measure some European countries implemented in response to the spike in inflows in 2015-16) and greater cooperation with migrants’ transit and origin countries to provide refugees with protection outside the EU. These measures are sometimes criticised as leading to the ‘end of asylum’ in Europe. While there clearly is a danger of a “slippery slope”, I argue instead that, in light of my understanding of public views and preferences, more controls on the asylum process are necessary to protect the continuation of the right to apply for asylum in Europe.

I have argued in this paper that empirical social science research can and should play an important role in informing judgements about soft feasibility constraints on asylum and refugee policy-making. It can do so by investigating the nature and characteristics of such soft constraints, clarifying the key insights and limits of the available evidence, and thus also facilitating a critical discussion of policy actors’ claims about “what the people want”. Given its focus on the role of public views, my discussion has paid less attention to other important aspects of soft feasibility constraints in asylum and refugee policy-making. Examples include socio-economic institutions such as welfare states and political institutions such as electoral systems and their associated opportunity structures for small populist parties to make major electoral gains through polarising debates and politicising asylum and migration issues. How these institutional factors affect the feasibility of particular asylum and refugee policies, and how they interact with public views, is an important area for future research.