The Long-Run Effects of STEM-Hours in High School: Evidence From Dutch Administrative Data

We analyze the short- and long-run effects of a policy change in Dutch secondary schools, which aimed at increasing the fraction of STEM graduates overall and in particular among previously underrepresented groups. Mandatory STEM hours were reduced in the STEM field, which is a prerequisite for enrolling in a STEM major at university. Hours decreased more strongly in the academic track (required for enrollment in research universities) than in the technical track (required for universities of applied sciences). Employing a difference-in-difference approach with Dutch administrative data, we find that the policy led to a significant increase in the take-up of the STEM field in high school, especially for women. In the longer-run, however, enrollment in STEM majors at university did not increase. Instead, after the policy change previously underrepresented groups, such as women and individuals from low-income families, were even relatively less likely to pursue a STEM degree. The decrease of women graduating from STEM was primarily driven by women with STEM parents, suggesting that it was due to negative signals about their preparedness for a STEM major.