Attitudes to Immigration

by Ian Preston

Centre for Research and Analysis of Migration, Dept of Economics, University College London

1. The economic effects of immigration to the UK - both its impact on the labour market and on public finances - have attracted considerable attention in the past month as a consequence of the release of a number of reports and official statistics.  The evidence that immigration is harmful to the UK-born is weak and indeed there are stronger empirical grounds for thinking immigration is on the whole economically beneficial, although it would be difficult to pick this fact up from much press coverage.  If there are any undesirable effects then they are probably largely distributional effects of a sort which could be addressed by combining a relaxed immigration policy with other interventions.
   
2. Nonetheless immigration is unpopular and this raises questions as to whether economics is at the root of this opposition.  Of course, fears about economic effects can exist even if they are unfounded but there are also other impacts which may generate hostility to immigration.  These social and cultural issues are delicate topics for discussion since they touch on discomforting questions of racial and ethnic prejudice.  Undoubtedly, unpleasant xenophobic motivations underlie some disquiet at the social effects of immigration but it is also true that not all such concern is so motivated.  Cultural change brought about by immigration can be seen as enriching but also as not. Loss of linguistic, religious or cultural homogeneity can be seen as regrettably undermining social solidarity and as aggravating social tensions.  Opinions differ on these issues and these differences are associated with differing views on immigration policy.
3. These issues are cross-national.  Countries across Europe differ greatly in their openness to immigration and also in the prevalence of related attitudes but the questions are universal.
4. The European Social Survey is a very large cross-national data source.  We at CReAM were lucky enough to be involved in the design of a module dealing with immigration in the first round of the survey in 2002, some of the questions from which have been retained in later rounds. This gave an opportunity to collect information on opinion about a wide range of effects and about the desirability of openness to immigration from different parts of the world.
   (The design team was admirably headed by Roger Jowell whose recent death is a source of sadness to anyone interested in serious study of opinion formation).
5. The data contains answers to questions on economic effects of immigration (jobs; wages; inequality; skill shortages) and sociocultural effects (linguistic, religious and cultural homogeneity; cultural change; social tensions; crime) and questions on preferred openness of immigration policy towards arrivals from richer and poorer countries inside and outside Europe.  There are also some rather broad questions on whether immigration is good or bad for the receiving country's economy and way of life; reassuringly, answers to economic questions do seem to be associated especially strongly with the former and answers to socio-cultural questions with the latter.  But which type of question is most strongly associated with opinion on what sort of immigration policy is most desirable?
6. Here at CReAM, disentangling the motives underlying individual attitudes to immigration has been part of the research agenda for more than a decade (see here or here, for example).  The latest paper, published this month in the Journal of the European Economic Association, addresses this question using data on the 22 European countries participating in the ESS.
7. The central finding is that both economic and social concerns are associated with hostility to immigration but the association with the latter is much stronger.  This is not to say that economic issues are unconnected with opposition to immigration but they appear to be connected much less strongly than other issues.  This finding is robust to different ways of treating the data.
8. This pattern is reasonably consistent across Europe.  The strength of concerns differs across countries as does the willingness to accept immigration but the ways in which they are related to each other is much more homogeneous.
9. It is important to emphasise that correlation is not causation.  The fact that those hostile to immigration are also concerned about its effects on employment, say, does not mean the latter opinion caused the former.  It is tempting to think of individuals as rationally considering the effects of immigration before weighing up their views on such matters to determine an opinion regarding immigration policy; it is quite possible however that attitudes towards immigrants held on other grounds encourage receptivity to views on effects that validate these opinions.  It is very plausible for instance to think someone concerned about preserving social homogeneity might be driven by that to oppose immigration and this opposition could bleed into a preparedness to accept the validity of secondary economic concerns.  Concerns about economic and social effects are themselves highly correlated suggesting some common source.
10. There are other reasons to be welcoming or not to immigration.  A government minister has argued that the "key insight" of the recent MAC report "is that the measure of a successful immigration policy is how it increases the wealth of the resident population".  This is not really a finding so much as the expression of a moral point of view and not one universally shared.  Some people would think the gains or losses to immigrants and to their home countries also matter.  Others would hold it to be unjust that rights of free movement internationally should depend on place of birth and others that richer countries have responsibilities, perhaps associated with inglorious colonial histories, towards residents of poorer countries.  Questions in the survey also tested these dimensions of opinion and analysis of the data suggests that such concerns are also correlated with preferences on immigration policy.
11. It is and has been recognised for some time that attitudes towards immigration vary in systematic ways with the socioeconomic and demographic characteristics of the opinion holder.  One common finding is that the less educated are more hostile to immigration and this fact has often been put forward as evidence for the primacy of economic concerns in driving such attitudes, given that such individuals are often thought to be more vulnerable to labour market competition from immigrants.  Education, however, is also strongly associated with a more generally liberal attitude to social questions, possibly because education either exposes individuals to broader perspectives on life or because it is more attractive to those who already have such views.   Analysis of the ESS data suggests that a much greater fraction of the difference in attitudes across education groups is accountable for by difference in social concerns than economic ones.  Older individuals also tend to be less liberal on immigration and this is more difficult to see as evidence for the primacy of economics since the elderly presumably suffer little labour market competition.  Again analysis suggests it is difference in social attitudes that are more strongly associated.