Abstract
I use remittances in Colombia to study the effects of an exogenous increase in household welfare on different outcome variables. I first use a matched difference-in-difference strategy to investigate the average effect of an exogenous increase in Colombian remittances on: labor supply (extensive and intensive margin), entrepreneurship, and education. I identify how these effects vary across the income distribution depending on the amount of the transfer as a proportion of labor income and if this relative amount matters in order to trigger an effect. Estimates suggest that when the relative amount of the transfer corresponds to the third quartile of the distribution, the likelihood of entrepreneurship and school enrollment significantly increase. People do not stop working but reduce the number of hours worked.
Afterwards, trying to examine distributional impacts, which usually are hidden when finding an average treatment effect, I deploy a matched quantile difference-in-difference approach. I investigate the heterogeneous effects of the exogenous increase in Colombian remittances on the entire distribution of labor income and job search duration. I find that the effects are higher at the bottom of the labor income distribution, and start decreasing with some noise. This is also true for both, women and men. In addition, a negative statistically significant effect is found at the top of the job search duration distribution. I find that these effects are mainly driven by women.