Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
When weather shocks under global warming affect productivity in climate-sensitive sectors, migration represents a potential adaptation mechanism. We exploit exogenous variation in temperature and precipitation across Brazil and examine the response in state-to-state migration flows between 1981 and 2010. Accounting for time-varying migration costs using a novel road dataset constructed by digitizing historical maps, and addressing the endogeneity of the roads network, we find strong evidence that a reduction in travel cost is associated with larger migrant flows. We also find suggestive evidence of climate-induced poverty trap −-- states with warming temperatures exhibit a smaller increase in out-migration, particularly among individuals who were likely employed in the agricultural sector at their origin states. Interestingly, our results imply that migrants are generally not sensitive to the destination weather conditions, with the exception for migrants working in agriculture at the destination.