Willingness to Purchase Insurance for Low Probability, Low Loss Natural Hazard Risk

Previous research has found that consumers respond to low probability, high loss risks in unpredictable ways. Some consumers ignore the low probability event while others overestimate the risk (McClelland, Schulze, and Coursey, J. Risk and Uncertainty, 1993). Experimental research has found that consumers respond in predictable ways to low probability, high loss risks (Ganderton et al., J. Risk and Uncertainty, 2000). This paper considers consumer response to low probability, low loss events. The application is to the willingness to pay for insurance against the costs of a hurricane evacuation. Household level hurricane evacuation costs are not catastrophic, relative to the other insured costs of natural hazards, but may act as a constraint on households deciding whether to evacuate. The perceived aggregate costs of hurricane evacuation (e.g., “one million dollars per mile”) may act as a constraint upon emergency managers who issue evacuation orders. These constraints may lead to unnecessary exposure to the risk of riding out a hurricane. Policy makers and insurance companies are considering ways to minimize these costs, including evacuation insurance. The data is from a telephone survey of residents of southeastern North Carolina, a region that has been recently exposed to several hurricanes. Respondents are faced with a hypothetical hurricane evacuation insurance product and asked whether they would be willing to purchase the product. The varying pricet of the product is randomly assigned. We examine whether respondents are willing to purchase evacuation insurance and the effects of perceived and objective risk, expected loss amounts and price of the insurance.