Empirical causal analysis of flood risk factors on US flood insurance payouts:Implications for solvency and risk reduction

This paper presents a regression model that quantifies the causal relationship between flood risk factors and the flood insurance payout in the U.S. The flood risk factors that have been considered in this research are flood exposure, infrastructure vulnerability, social vulnerability, and the number of mobile homes. Historical data for the annual flood insurance payout, flood risk factors, and other control variables were collected for six years between 2016 and 2021 and used in a Mixed Effects Regression model to derive the empirical relationships. The regression model expressed the natural logarithm of the annual flood insurance payout in a county based on the flood risk factors and control variables. The paper presents the regression coefficients that quantify the causal influence. It has been found that all four flood risk factors have statistically significant positive influence on the flood insurance payout in a county. However, the extent of the influence is different for different flood risk factors. Among them, flood exposure has the highest influence on the flood insurance payout, which is followed by the number of mobile homes, infrastructure vulnerability, and social vulnerability. Since the federal flood insurance program in the U.S. has a large debt to the U.S. treasury, the government should plan for effective risk reduction that can reduce the flood insurance payout in future to keep the program solvent. The outcomes of this research are expected to facilitate that decision-making process by providing the empirical relationship between flood risk factors and flood insurance payout.