This paper provides cross-country macro evidence on the effect of immigration on productivity. It combines panel data with instrumental variables to correct for OLS biases due to reverse causality. In particular, it introduces push-targeting instruments; they constitute a new approach to find...