Loss aversion in social image concerns

This paper explores whether loss aversion applies to social image concerns. In a simple model, we combine loss aversion in social image concerns and attitudes towards lying. We then test its predictions in a laboratory experiment. Subjects are first ranked publicly in a social image relevant domain, intelligence. This initial rank serves as within-subject reference point. After inducing an exogenous change in subjects' rank across treatments, subjects are offered scope for lying to improve their final rank. We find evidence for loss aversion in social image concerns. Subjects who face a loss in social image lie more than those experiencing gains if they sufficiently care about social image and have a reputation to lose. Individual-level analyses document a discontinuity in lying behavior when moving from rank losses to gains, indicating a kink in the value function for social image.