Do Risk Preferences Change? Evidence from Panel Data Before and After the Great East Japan Earthquake

We investigate whether experiencing a natural disaster the Great East Japan Earthquake in 2011 changes individualsrisk preferences. The novelty of our study is that we use panel data, and we can track the change in risk preference of the same individuals before and after the Earthquake. Previous studies use cross-section data collected after the negative shocks have occurred, and hence can be biased by unobserved individual heterogeneity. We \u85nd that people who experienced larger intensity of the Earthquake become more risk tolerant. Interestingly, all the results are driven by men. Further, we \u85nd corroborative evidence that men become more engaged in gambling and drinking if they were more exposed to the Earthquake. Finally, we compare the estimates from cross-section and panel speci\u85cations, and demonstrate that the estimate relying on cross-section data may be biased