Measuring risk and time preferences during the emergence of the Covid-19 crisis

We measure risk and time preferences, and examine the relationship with the disposition effect. Firstly, to estimate preferences, we use and compare several well-known quantitative and qualitative methods. Our main specification is an extended version of the Convex Time Budget method of Andreoni and Sprenger (2012, AER). We elicit present bias, loss aversion, probability weighting and the discount rate in a large field experiment by means of a survey. We compare how risk and time preferences evolve during the Covid-19 crisis. Secondly, we test the disposition effect during the Covid-19 crisis – the tendency of investors to hold losing investments too long and sell winning investments too early – by a simple experiment based on realization utility from Barberis and Xiong (2012, JFE). Because of our simultaneous measurements, we are able to investigate whether individual trading behavior behaves in line with the structurally estimated preferences and its theoretical predictions from realization utility. Finally, we relate the estimated preferences to attitudes for insurance products and personal background characteristics.