Competitive high-ranking positions still are occupied largely by men, and women in academia remain scarce in engineering and sciences.

Competitive high-ranking positions still are occupied largely by men, and women in academia remain scarce in engineering and sciences. Suggested explanations for this fact focus mostly on discrimination and differences in abilities or preferences (in terms of work hours or field of study). Niederle and Vesterlund explore an additional factor, namely that women and men may differ in their selection into competitive environments. In a laboratory experiment, the authors examine an environment in which women and men perform equally well under both a noncompetitive piece rate and a competitive tournament scheme. Participants then are asked to choose the incentive scheme for their next performance. The authors fred that twice as many men as women choose the tournament over the piece rate. This gender gap in tournament entry cannot be explained by performance before or after the entry decision has been made. While men are more optimistic about their relative performance than women, this difference can only explain a small share of the gender gap in tournament entry. Finally, the authors show that gender differences exist even when participants simply decide how to be paid for a past performance. They use this decision as a control for non-tournament specific gender differences (such as risk aversion, feedback aversion, general overconfidence), and fred a large residual gender effect when participants select tournament compensation for a future performance.