Judd Boomhower (UCSD)

Abstract:

In the face of mounting climate risk, well-functioning property insurance markets can provide households and businesses with protection from financial losses and incentives to reduce the risk of loss. To efficiently achieve these goals, markets need (1) a detailed understanding of property-level risk from weather disasters, and (2) the ability to set competitive premia that reflect this assessed risk. We consider the important case of California, where wildfire property losses now average almost $10 billion per year. We provide new descriptive evidence of insurance market responses to this escalating wildfire risk in terms of prices, availability of coverage, innovation in underwriting tools, and market structure. We develop a model of a market where insurers differ in their ability to accurately assess and price on granular, parcel-level information on wildfire risk. We use the model, together with microdata on observed premia and wildfire risk scores, to analyze how the use of probabilistic catastrophe models, together with regulations that limit how these models can be used by insurers to set rates, is affecting insurance availability, markups, and the distribution of insurance prices across customer groups.