The paper studies the impact of unemployment benefits on immigration. A sample of 19 European countries observed over the period 1993-2008 is used to test the hypothesis that unemployment benefit spending (UBS) is correlated with immigration flows from EU and non-EU origins. While OLS estimates reveal the existence of a moderate correlation for non-EU immigrants only, IV and GMM techniques used to address endogeneity issues yield, respectively, a much smaller and an essentially zero causal impact of UBS on immigration. All estimates for immigrants from EU origins indicate that flows within the EU are not related to unemployment benefit generosity. This suggests that the so-called "welfare migration" debate is misguided and not based on empirical evidence.