We study the short-run, dynamic employment effects of natural disasters. We exploit monthly data for over 90 3-digits NAICS industries and 78 Puerto Rican counties over the period 1995-2017. Our exogenous measure of exposure to natural  disasters is computed using the maximum wind speed recorded in each county during each hurricane. Using panel local projections, we find that after the “average” hurricane, employment and wages fall by 1% on average. The effects peak after six months and disappear within two years. Across industries, we find substantial heterogeneity in the employment responses. This heterogeneity can be partly explained by input-output linkages.