Leverages variation in the timing of women's colleges' transitions to coeducation to study how exposure to a gendered social environment affects women's human capital investments.
We leverage variation in the timing of U.S. women's colleges' transitions to coeducation to study how exposure to male peers affects women's human capital investments. Our event-study analyses of newly collected historical data find a 3:0 percentage-point (30%) decline in the share of women majoring in STEM after ten years of coeducation. We find little evidence that coeducation altered course offerings, class sizes, competitiveness of STEM classrooms, or the composition of the female student body. Extrapolation of our main estimate suggests that exposure to male peers in college can explain 36% of the contemporary gender gap in STEM.