Hurricanes can cause significant economic loss in the eastern coastal United States. Forecasts of seasonal hurricane frequency and intensity are available, but the use of these forecasts in the reinsurance industry to mitigate the economic impacts of hurricane landfall is not apparent. If reinsurance utilized forecasts, they could conceivably increase the likelihood of being able to pay insurance companies in the event of a damaging hurricane strike. This would also benefit insurance companies and property owners by stabilizing the insurance industry as a whole.

We first analyzed the dependence of hurricane frequency and damage on El Niņo-Southern Oscillation (ENSO). Results indicated that La Niņa (El Niņo) years experience more (less) frequent, more (less) damaging hurricanes. We then built a simulation model of damage. Assuming knowledge of the ENSO state, the model generated annual total damage based on adjusted Poisson and lognormal distributions of annual hurricane number and storm damage, respectively. We obtained the 5% level of loss exceedance associated with all, La Niņa, Neutral, and El Niņo years. They showed 20-year period return levels of $33.7, 73.2, 30.1, and 9.9 billion, respectively (constant 1995$).

ENSO state appears to have a considerable effect on the distribution of economic damage. Through the model's inclusion of specific reinsurance contract options (and associated costs) and a complete hurricane forecast (i.e. not just the ENSO aspect of the forecast), the optimal use and forecast value to the industry could be quantified.