Comment on Dahlberg, Edmark and Lundqvist (2012)

In this brief comment, we return to the question raised in Dahlberg, Edmark and Lundqvist (2012) concerning a causal relationship between ethnic diversity and preferences for redistribution. A re-analysis of their study indicates that the results are based on a severe sample selection bias and an invalid instrumental variable. Correcting for either of these problems reveals that there is no relationship between ethnic diversity and preferences for redistribution. In fact, our re-analysis shows that it is only possible to find a statistically negative significant relationship by combining a highly selected sample with the invalid instrument.