In-group bias in preferences for redistribution: a survey experiment in Italy

Abstract

Using a new survey and experimental data, we investigate how information on inequality and immigration affect preferences for redistribution in Italy. Our randomized treatments show that preferences for redistribution are often inelastic to information. However, we find that provision of information on poverty statistics related to the native-immigrant composition of poverty reduces economic in-group bias by affecting exclusionary redistributive preferences: respondents are less likely to support policies which exclude immigrants from access to the welfare state once they learn that immigrants are less represented among the poor and natives are not as poor as they used to believe. Finally, we find some evidence of in-group bias by investigating the presence of heterogeneous treatment effects across groups.