The risk elicitation puzzle revisited: Across-methods (in)consistency?

• DOI: 10.1007/s10683-020-09674-8
• Publication Date: 2020-09-12T07:02:37Z
• Authors: Felix Holzmeister; Matthias Stefan
• Keywords: Time consistency

With the rise of experimental research in social sciences, numerous methods to elicit and classify people's risk attitudes laboratory have evolved. However, evidence suggests that towards may vary considerably when measured with different methods. Based on a within-subject design using four widespread preference elicitation tasks, we find indeed give varying estimates individual aggregate level preferences. Conducting simulation exercises obtain benchmarks for subjects' behavior, observed heterogeneity across is qualitatively similar arising from independent random draws choice distributions experiment. Our study, however, provides subjects are surprisingly well aware variation riskiness their choices. We argue this calls into question common interpretation revealed preferences as being inconsistent.