The skills premium and the ethnic premium: a cross-national experiment on European attitudes to immigrants

We introduce a cross-national survey experimental design which combines three different sources of potential variation in migration attitudes – migrant characteristics, native respondent characteristics and national context. Our survey experiment tests the impact of randomly assigned cues for migrant skills (professional or unskilled) and origins (European or non-European) in twenty-one national contexts. In order to make the origins cue both contextually relevant while retaining comparability across contexts, we use the largest poorer European and non-European source countries for migrants in each country as our origins cue. We find that the ‘skills premium’ observed in previous research is universal – in all of our contexts native respondents are more willing to accept skilled professional migrants than unskilled migrants, but with considerable variation in effect size between countries. We find that a significant preference for European over non-European migrants exists in the majority of our countries, but varies a great deal between contexts, and is insignificant in several countries, suggesting that this ‘European premium’ is more contingent than the ‘skills premium’. The two treatment effects are unrelated at the national level, and show only a modest interaction at the individual level, suggesting preferences for high skills and European origins are largely independent of each other. We find no evidence that treatment effects are related to views about the overall impact of immigration at the national level or the individual level. However, some individual attitudes – in particular racial prejudice, social distance and support for anti-discrimination legislation – predict variation in responses to the experimental cues.