Understanding the Two Components of Risk Attitudes: An Experimental Analysis

Cumulative Prospect Theory (PT) introduced the weighting of probabilities as an additional component to capture risk attitudes. However, this addition would be a less significant challenge to expected utility theory (EU) if utility curvature and probability weighting showed strong positive correlation. In that case the utility curvature in EU alone, while not properly describing risky behavior in general, would still capture most of the variance of individual risk aversion. This study provides experimental evidence that such a strong and positive correlation does not exist. Although most individuals exhibit concave utility and convex probability weighting, the two components show no strong positive correlation. JEL classification: C91, D81 * We thank Peter Wakker for his detailed and valuable comments. We also thank Werner Güth, Michel Robe, Rupert Sausgruber, Christoph Vanberg, an anonymous associate editor, and two anonymous reviewers for their helpful comments and suggestions. We also thank Han Bleichrodt for sharing some of his experimental data with us. This paper was presented at the APESA meeting 2009 in Haifa and at the SMYE 2009 in Istanbul. Comments and suggestions of participants are gratefully acknowledged.