Research

Abstract: As more women work into midlife, aging-related health transitions may shape economic security approaching retirement. This paper studies the economic outcomes of midlife women following the 2002 Women's Health Initiative (WHI) announcement—a major negative shock to beliefs about hormone replacement therapy (HRT). Combining predicted menopausal status from SWAN with CPS data from 1996–2008, I estimate event-study models tracing labor-supply responses to the resulting decline in HRT use. Post-menopausal women experienced a significant, delayed decline in labor-force participation beginning three years after the announcement, reaching an 11–20% decline relative to baseline by the sixth year. Among women who remain employed, I also find declines in wage income and hours worked. The effect is concentrated among married women, consistent with spousal income buffering the need to remain employed among single women. These results show how the menopause transition, compounded by a health information shock, can materially reshape women's economic lives during a critical juncture of their careers.

Abstract: Do restrictions on public funding create unintended reliance on social assistance? In this paper, we study the 1976 Hyde Amendment, which barred federal funding for abortion. Using county-level data and an event-study design, we show that reduced federal funding for abortion increased fertility among young women by 2%. These effects were concentrated among non-white women, who subsequently experienced greater welfare participation in states with larger abortion funding declines. The consequences extend into the next generation: non-white girls born after Hyde were more likely to rely on public assistance in adulthood. Abortion funding restrictions reinforce long-run economic inequality across generations.

Abstract: Maternal exposure to air pollution during pregnancy has serious health consequences for both mothers and infants, particularly by disrupting fetal development during critical gestational phases. This paper estimates the causal impact of air pollution on newborn health in India by linking satellite-derived PM2.5 measures with geocoded Demographic and Health Survey (DHS) data from 2019–2021. To address endogeneity concerns such as household reliance on biomass fuels or crop residue burning, which are unobserved in survey data, I employ an instrumental variable strategy that leverages biomass burning events in neighboring areas as an exogenous source of variation in local pollution. The results show that greater gestational exposure to PM2.5 significantly increases the likelihood of adverse birth outcomes. A one-standard-deviation increase in PM2.5 raises the probability of low birth weight by 5.4 percentage points and increases the likelihood of preterm birth by 3 percentage points. Low birth weight and preterm birth are among the strongest predictors of cognitive deficits, stunted development, and reduced human capital accumulation—consequences that are particularly severe in developing-country settings where compensatory health investments are limited. Heterogeneity analyses show that these adverse effects are far larger among mothers with lower education. These findings contribute to the literature on pollution and health in the developing world by offering new evidence that addresses endogeneity using a context-specific instrument.

Github Repository