The purpose of this reply is twofold. First, we discuss the major point raised by Stark and Lukasz (Review of Development Economics 17, no. 1 (2013):156–62), i.e. the fact that a framework which explicitly considers asymmetric information is correct and would imply a reversal of our finding. Although, we acknowledge that the mechanism highlighted by the authors is an alternative explanation to return decisions, we argue that the suggested framework is unsuitable in the specific context analyzed in our paper (as well as most real‐world situations). Instead, the assumptions underlying our simple theoretical model are strictly linked to data availability in order to perform a sensible empirical analysis. Second, we present a slightly different version of the model proposed in the original article that overcomes possible inconsistencies on the saving behavior of the migrants. Although all the computations are shown in one of the articles cited in our published paper, we now prefer to show them fully in this issue of the Review. The conclusions of our theoretical model do not change. Hence, we conclude that the empirical evidence of the original article—which is the main contribution of our work—is supported by a robust framework.