Behind the Veil of Ignorance∗

This paper uses theory and experiments to study situations in which individuals make choices for themselves and for others in situations intended to resemble what Rawls and Harsanyi would describe as “behind the veil of ignorance.” Theory suggests a (surprisingly overlooked) decomposition of choices behind the veil of ignorance into choices in more familiar altruistic and risk environments. Our experimental design elicits many choices from each subject in each of the three environments. The rich data that results enable us to do meaningful revealed preference analysis and to test the predictions of theory. We find that: 1) individual choice behavior is consistent with utility maximization; 2) the predictions of the theoretical decomposition of choices are confirmed in the data; 3) choice behavior is extremely heterogeneous across individuals. PRELIMINARY Please do not quote or circulate ∗We are grateful to Ray Fisman and Daniel Markovitz for helpful conversations and to Neil Lessem for research assistance. Financial support was provided by the National Science Foundation under grants SES-0317752, SES-0518936 and SES-0617027, and by the UC Berkeley and UCLA Academic Senate Committees on Research. Any opinions,