On Thin Ice : Risk, Return and Crash

This paper describes results from a new experiment studying determinants and effects of economic risk-taking. In each session four subjects choose three slots for ice fishing on their portion of a frozen lake. The farther out on the lake the higher are the returns but also the higher is the risk of a crash and a loss. This setup permits us to investigate how transparency and incentive structures – two issues intensively debated in policy circles – affect risk taking and vulnerability to crashes. Our results indicate that increasing transparency of choices raises output and reduces the frequency of crashes whereas introducing a steeper return-risk profile has the opposite effects. Bounded rationality in the sense of naïve risk taking does not explain crashes. Instead, behavioral traits of subjects regarding risky choices and their distribution across subjects help to understand economic performance and frequency of crises