Preferences for the resolution of risk and ambiguity

Abstract

Generalized recursive utility models often imply that agents have a preference over the timing of uncertainty resolution. Laboratory elicitations of subjects’ preferences generally provide direct evidence in support of this implication, but only in the domain of risk. We provide the first experimental examination of uncertainty resolution with respect to ambiguity, in addition to risk. The modal subject exhibits a preference for both early resolution of risk and ambiguity, but with only a minimal willingness to pay to realize either over late resolution. While preferences in both domains are positively correlated, the strength of that correlation varies based on ambiguity attitudes. Among ten commonly used representative recursive utility models, we identify the models that most efficiently explain observed subject preferences under two alternative assumptions: treating a subject’s token willingness to pay as either a true preference or indifference.