We examine the impact of natural disasters on the likelihood that a government is removed from office using panel data for 156 countries over the period 1975–2010. Employing a conditional logit model, we find that the occurrence of natural disasters, the number of natural disasters and disaster-related losses increase the chances that a government will be replaced. The magnitudes of these effects differ widely across natural disaster types, but are robust to the inclusion of economic and political variables and to model specifications. Overall, these findings are consistent across our sample of OECD and non-OECD countries.