Emigration and human capital: Who leaves, who comes back and what difference does it make

This thesis studies the characteristics of international migration in order to estimate some effects of emigration on labor market outcomes in the origin country. I extend the standard model by Borjas (1987) incorporating returning migration as a sequential choice. Under relatively weak conditions, characteristics that make individuals more likely to migrate make also migrants less likely to come back.; I use the Mexican and the American census to compare migrants with non migrants. I show that the usage of the two censuses is a valuable technique to study self selection on those flows. Underreporting of immigrants in the American census, although important in terms of magnitude, does not create a large bias when estimating the distribution of characteristics of the actual stock/flow of migrants.; Using the available data for the Mexican case I show some evidence that returning migration is a sequential decision for more than 50% of all returnees. The observed selection of returning migrants compared to the selection of migrants is exactly as the previous model predicted.; I then study several effects of emigration on different labor market outcomes. First, I analyze the effect of emigration on labor market outcomes of returning migrants. On the one hand, I show that not many of them changed their labor status after their migration. On the other hand, returning migrants earn a wage differential of 10% respect to non migrants, but this could be attributed to pre migration differences instead of human capital gains. Second, I analyze the human capital loss that emigration produces in Mexico. I non-parametrically estimate the Mexican human capital of emigrants using returning migrants' wages in order to partially account for differences in unobserved characteristics. I show evidence that those non returning migrants are mainly drawn from the middle part of the distribution of human capital. This evidence is consistent with previous findings by Hanson and Chiquiar (2002) that only considered differences in observed characteristics. I finally expose the importance of getting this estimation right if we want to estimate the effects of emigration on wages of those left behind in the short run.