Abstract
Experimental research in many countries shows that ethnic or racial minorities are discriminated against in rental housing markets. Such discrimination may reflect an aversion to minority tenants, but can also be due to the use of stereotypical images of minorities in the absence of clear signals that minority apartment seekers are suitable tenants. Existing experimental research has struggled to distinguish these two channels. In a nationwide field experiment in Switzerland (paired applications to over 5,700 proprietors), we find evidence of discrimination against apartment seekers with Turkish and Kosovar names (i.e., distant minorities), while individuals with names from neighbouring countries (i.e., close minorities) are not discriminated against. Requests with more information generally receive more invitations, but the effect of this increased invitation rate is similar across minority groups. We find that adding more information does not reduce discrimination as such, as many previous studies have been claiming, but that the kind of information provided matters: while offering additional information about employment status reduces discrimination for individuals with Kosovar and Turkish names, signalling naturalisation does not lead to less discrimination.