How COVID-19 affects Insurance Demand?: Multiattribute Utility and Sign of Cross-derivative

The COVID-19 pandemic has affected our daily life with respect to both wealth level and health status simultaneously. Many decision analyses under risks, which have multiattributed outcomes, employ multiattribute utility function. The sign of cross-derivative of the utility function plays an important role because it captures trade-offs between attributes. This study aims to estimate the sign of the cross-derivative using a Japanese representative survey on the intention to purchase two types of insurance, those are income protection insurance and death insurance, before and during the COVID-19 pandemic. The COVID-19 pandemic is regarded as a natural experiment because exogenous shocks change risks, such as mortality probability. To estimate the preference, we construct an economic model of the demand for income protection insurance and death insurance under the COVID-19 pandemic. We also examine how the sign of the cross-derivative affects the optimal demand for insurance when the government increases spending such as lockdowns against the infection. Finally, we estimate the sign of the cross-derivative from a large representative survey of the intention to purchase both types of insurance in Japan. We find that preferences exhibit correlation aversion.