Sectoral Shocks and Gender Differences in Field-of-Study Choice: Evidence from the Dot-Com Collapse

This paper studies how sector-specific shocks shape gender differences in field-of-study choice. I examine the collapse of the dot-com bubble as an exogenous shock that increased uncertainty in the Information Technology (IT) sector. I develop a theoretical model of field choice under uncertainty in which students differ in risk aversion by gender. When sectoral risk rises, more risk-averse individuals are disproportionately deterred from entering that field, leading to lower female participation, stronger positive selection among those who remain, and reallocation toward close substitute fields. Using Swedish administrative data and a difference-in-differences design, I find that the gender gap in IT graduation widens by about 4.2 percentage points following the bust, a relative increase of roughly 50 percent. At the labor market entry margin, a negative and statistically significant gender gap emerges for later cohorts, with women less likely than men to enter IT employment, conditional on holding an IT degree. Women who select into IT become more positively selected, with their average GPA rank rising by about three percentile points relative to men. Cross-field evidence shows that women shift toward engineering, a close substitute but with lower exposure to the shock. These findings provide evidence consistent with a mechanism in which sector-specific shocks amplify gender segregation by altering both participation and the allocation of talent across fields. More broadly, the results suggest that gender differences in field-of-study choice evolve with economic conditions: sectoral downturns may reinforce existing disparities by diverting relatively high-ability women away from more volatile but promising fields.

Property Rights, the Intra-Couple Wealth Gap, and Family Outcomes: Evidence from China

This paper documents intra-couple housing wealth gaps in China and examines their consequences for household decisions. I exploit the 2011 Marriage Law reform, which replaced equal division at divorce with a registered-owner rule, as a natural experiment. Using spouse-level panel data and a difference-in-differences design, I find that the reform increased the husband–wife legal share gap by 28.3 percentage points and generated an average housing wealth gap of CNY 63,667, approximately ten times the wives' annual income. The wealth gap persisted and widened as housing prices rose, even as legal share gaps narrowed over time. In the second stage, 2SLS estimates show that a one-unit increase in the husband’s relative ownership share reduces divorce probability by 6.75 percentage points, raises the wife's probability of employment by 6.8 percentage points, and increases her weekday housework by approximately one hour. Heterogeneity analysis reveals that the reform lowers divorce probability regardless of which spouse holds the title; female employment rises when the husband’s relative ownership increases but no significant change when the wife's does; and wives' housework burden rises in both cases. Standardized comparisons show that wealth gap effects exceed share gap effects across all outcomes, indicating that the economic magnitude of housing inequality, not just formal ownership status, amplifies bargaining consequences. The findings show how a formally gender-neutral legal reform can reinforce intra-household inequality when layered onto pre-existing gendered ownership norms.

You Are the Elite Now: Admission Effects of an Excellence Initiative in the Chinese Higher Education System

with Tianze Liu

This paper examines how university applicants respond to government policies that label specific programs as centers of excellence. Such policies, increasingly common worldwide, aim to enhance human capital and global competitiveness by directing students toward high-priority fields. We study this question in the context of China’s excellence initiative, which designates selected university-discipline units as First-Class Disciplines (FCDs). Using a difference-in-differences design, we show that designation attracts higher-ranking applicants, raising admission competition by about 1.5 percentiles in exam rankings, equivalent to surpassing roughly 1,800 additional peers per province. Importantly, we also find strong spillover effects: non-designated disciplines within the same universities become more competitive, suggesting that quality signals extend beyond the targeted programs. By increasing competition in designated fields, this policy not only reallocates top students toward strategic areas but also improves the supply of skilled students in priority fields. Our findings provide the first evidence that university-discipline-level excellence initiatives can shape talent allocation, amplify institutional reputation, and ultimately influence the future composition of a country’s human capital