Is the veil of ignorance only a concept about risk? An experiment

We implement the Rawlsian veil of ignorance in the laboratory. Our experimental design allows separating the effects of risk and social preferences behind the veil of ignorance. Subjects prefer more equal distributions behind than in front of the veil of ignorance, but only a minority acts according to maximin preferences. Men prefer more equal allocations mostly for insurance purposes, women also due to social preferences for equality. Our results contrast the Utilitarian's claim that behind the veil of ignorance maximin preferences necessarily imply infinite risk aversion. They are compatible with any degree of risk aversion as long as social preferences for equality are sufficiently strong.