Adverse and Advantageous Selection in the Laboratory

Asymmetric information plays an important role in markets and politics. When parties are asymmetrically informed and have misaligned preferences, they may be hurt by adverse selection. By contrast, if parties know that their preferences are aligned, they may benefit from advantageous selection. Using a laboratory experiment, we investigate the degree to which individuals account for adverse and advantageous selection. By comparing behavior in a game in which subjects are asymmetrically informed with behavior in a game where those same subjects are symmetrically uninformed, we find evidence that a significant fraction of subjects account for these selection effects. We find that removing strategic uncertainty significantly increases the fraction of subjects who account for selection effects. Across our treatments, we find that subjects account for adverse selection to a greater degree than they account for advantageous selection. In addition, we find that a sizable fraction of subjects who do not behave according to predictions are in fact able to understand selection effects but do not apply that knowledge.