Research

Presented at SUFE seminar, ZUEL seminar*, CUFE seminar, SWUFE seminar*, SOLE 2025*, CES China 2024, AMES 2024, SEHO 2024, CES North America 2024, SEA 2024, GLO 2024, CHLR 2024,  UCSD Workshop, Stanford SCCEI Young Researcher Workshop

Abstract: Women at different life stages may respond differently to economic shocks as they face varying trade-offs. This paper explores the role of public sector employment, characterized by gender equal wage setting and family-friendly childcare policies, in shaping the gender gap in labor market outcomes and women's key life decisions over the life cycle. We focus on the privatization of the state sector in urban China in the late 1990s, which introduced competition and caused massive layoffs. Using cross-prefecture variation in reform exposure, based on initial differences in state-owned enterprise employment shares, we employ a difference-in-differences design. We show that older women were more likely to be laid off and faced greater difficulty in finding re-employment than men. Middle-aged women were less likely than men to transition to the emerging private sector, while younger women prioritized human capital investment more than men and delayed marriage. Further mechanism analysis reveals that privatization increased the demand for high-skilled workers, reshaped household specialization, intensified childcare burdens on women, and reinforced traditional gender role attitudes.

Presented at UCSD Workshop, FDU seminar*, CESI 2023, SMU seminar*, RUC/GLO Conference 2023*, All CA Labor Economic Conference 2023 (Poster)  

Abstract: This paper examines the long-term impact of privatization on the next generation’s educational and labor market outcomes, as well as its implications for intergenerational mobility. We leverage the 1997–2000 privatization of State-Owned Enterprises (SOEs) in urban China—which resulted in over 35 million layoffs—as a natural experiment. Using individual survey data and prefecture-level employment data from statistical yearbooks, we implement a triple-difference design that exploits variation in reform intensity across regions, children's birth cohorts, and whether their parents were employed in SOEs before the reform. Our findings reveal that the reform had a negative impact on human capital accumulation for the next generation: individuals exposed to the reform attained significantly fewer years of schooling and earned lower incomes later in life. These effects were particularly pronounced for girls and children whose fathers were laid off. Additionally, we find suggestive evidence of a decline in intergenerational persistence in income and education among households affected by layoffs.

Presented at WEAI 2025, ASHEcon 2025*, All CA Labor Economic Conference 2025* (Poster) , UCSD Workshop 

Previously circulated under the title "Maternity Ward Crowding, Birth Outcomes, and Future Fertility and Healthcare Decisions: Evidence from California."

Abstract: The recent global trend of maternity ward closures is reshaping childbirth care delivery and intensifying capacity pressures on the remaining facilities. We study how overcrowding during a woman’s first childbirth influences both clinical practices and subsequent healthcare choices, using comprehensive administrative records on all California births between 1989 and 2017. Leveraging quasi-random, within-hospital variation in daily number of patients, we find that overcrowding reduces the intensity of medical interventions—such as C-sections, epidurals, inductions, and augmentations—consistent with efforts to relieve physician workload. The reduction in epidural use is concentrated among disadvantaged mothers, suggesting disparities in care access during high-demand periods. Despite these adjustments, we find no detectable adverse effects on immediate maternal or infant health. Looking beyond the initial birth, we show that overcrowding does not alter future fertility but significantly increases the likelihood that mothers switch hospitals for subsequent deliveries. These effects are non-linear and most concentrated in the busiest quartile. We find no systematic patterns in hospital selection, indicating that switching is driven primarily by negative first-birth experiences. Together, our findings highlight a critical trade-off: short-term operational adjustments during peak-demand days can preserve safety, but risk undermining patient experience and long-term patient loyalty. 

Presented at  All CA Labor Economic Conference 2024 (Poster), SUFE Seminar*, CES NA 2025, WEAI 2025*, AEDC2025*, ICCDS 2025*

Abstract: This paper examines the long-term impact of privatization on the next generation’s education, labor market outcomes, and intergenerational mobility. We study the 1997–2000 privatization of state-owned enterprises (SOEs) in urban China, which displaced more than 35 million workers. Using a triple-differences design that exploits variation in reform intensity across regions, children’s birth cohorts, and parents’ pre-reform employment status, we find that the reform reduced human capital accumulation and worsened long-run labor-market outcomes for the second generation. These adverse effects operate primarily through intergenerational transmission via parental layoffs, while rising returns to education during the reform period acted as a countervailing force. Furthermore, we show that the reform weakened intergenerational income persistence, which we attribute to the erosion of network-based job search following the reform.

Presented at SITE 2025, Young Scholars Conference 2024 (Fudan-UC Center), WEAI 2023, Stanford SCCEI Young Researcher Workshop, The UCSD Migration Research Conference: Interdisciplinary/Perspectives, UCSD Graduate Workshop - Applied Microeconomics, Summer School in Economics of Migration - UC Davis Global Migration Center  

Abstract: China’s unprecedented expansion of higher education, launched in 1999, increased annual college enrollment from 1 million to over 9.6 million by 2020. This paper traces the global ripple effects of that expansion by examining its impact on US graduate education. Combining administrative data from China’s college admissions system and the US SEVIS database, we leverage the centralized quota system in Chinese college recruitment for identification. We find that China’s college expansion explains roughly 27\% of the increase in Chinese graduate student enrollment in the US between 2003 and 2015. We then document downstream consequences for US universities: Chinese master students generate positive spillovers, driving the birth of new STEM master's programs, and increasing the number of other international and American master's students. Our findings highlight how domestic education policy in one country can reshape the academic and economic landscape of another.

China's healthcare system provides a unique context for studying the dynamics of patient-physician relationships, as patients are not assigned a primary care provider and have significant flexibility in choosing a doctor for each visit. Using administrative medical claims data from a city in China, this paper investigates the stickiness of patient-physician relationships and examines how patients respond when their regular doctors are on temporary leave.