Extreme weather events and stock returns in the insurance industry

This thesis examines abnormal returns on insurance stocks listed on NYSE subsequent to an Atlantic hurricane making landfall in the U.S., using data from 2000 to 2018. We investigate if firm characteristics explain the abnormal return using panel data regression. We further elaborate by constructing high, medium, and low portfolios of stock sorted by various multiples, and testing if there is a difference in the abnormal return. To control for the extraordinary market conditions during the financial crisis, we conduct the analysis both with and without events occurring in this period. The goal of this thesis is to further add on to the existing research on extreme weather effects on the stock market by using panel data regression and multiples. We conclude that the insurance firms on the NYSE do not exhibit negative abnormal returns after hurricanes making landfall. Furthermore, we find the dividend yield, margin, and cash explains the abnormal returns and time dummy variables are significant in all events. When the financial crisis events are excluded, only time-dummy variables are significant. When testing between the constructed portfolios, we find no significant differences