Natural hazard risk and life satisfaction : empirical evidence for U.S. hurricanes

The ongoing process of climate change goes along with an increase in the frequency and severity of various sorts of natural disasters. While the existing literature has almost exclusively focused on studying the direct effects resulting from different types of disasters, the effect of changing disaster risk so far has largely been neglected. In this paper we study the effect of hurricane risk on individual self-reported life satisfaction. In order to do so we combine geo-referenced survey data from the U.S. Gallup Daily Survey and hurricane data for the United States. Using Willoughby's (2006) wind field model we construct time-varying indicators of hurricane risk on the zip-code-level for the period of 2010 to 2018, based on historical hurricane data. We then study whether the time-varying hurricane risk indicators affect self-reported life satisfaction in a two-way fixed effects model. Our findings indicate that regions with comparatively high hurricane risk report significantly lower levels of life satisfaction than their counterparts in less hurricane-prone regions, even after controlling for zip-code-specific, time-specific and individual-specific differences. Thus, the impacts of natural disasters on life satisfaction tend to be underestimated when focusing on the direct effects exclusively.