How do institutions affect the impact of natural disasters?

In this paper we study how differences in the quality of countries' institutions affect the impact of natural hazards in these countries. To do so, we first build a new data set that allows us to adequately control for countries' development and geological characteristics and, importantly, the physical intensity of the natural hazard. We then analyze our data using an output distance frontier model to assess two important aspects of the relation-ship between institutions and hazard impacts. First, the model allows us to estimate the trade-offs between different types of (negative) outcomes (e.g., deaths, affected, and damages). Second, it enables us to estimate the excess deaths, affected inhabitants and damages that countries, all else equal, suffer relative to the best performing countries. We can refer to this as the countries' (in)efficiency at managing natural hazards. Our results show that countries differ a lot in their disaster management efficiencies, with richer countries performing better than poorer countries. Richer countries also incur higher capital losses in exchange for fewer lives affected, controlling for their overall level of development and population density. For rich and poor countries we show that institutions of higher quality indeed correlate with higher disaster relief efficiencies. Most important are indicators of good governance and government effectiveness, whereas the de jure indicators are not informative. Our estimates suggest that a country with a 10%higher disaster relief efficiency will save one more life and protect four more people at the cost of $8 million in capital losses in an average intensity natural hazard.