Research

Under Review, Journal of Public Economics.

This paper develops a framework for how nonlinear tax schedules—focusing on the Earned Income Tax Credit (EITC)—shape labor supply responses to wage shocks. We introduce a commuting-zone average treatment effect (CZATE), an equilibrium local-market estimand for how changes in EITC generosity affect commuting-zone employment, and an envelope semi-elasticity that maps generosity into the wage–employment relationship. The empirical estimates reproduce canonical individual-level employment responses to the EITC and are robust to contemporaneous welfare reform. Using local booms, China trade shocks, and minimum-wage hikes, we find that a $100 increase in EITC generosity rotates the wage–employment slope by about 0.05–0.2 percentage points per 1% wage change, with larger rotations in economic downturns. The results imply that the cyclical consequences of EITC expansions depend on where eligible workers lie on the schedule, making phase-in versus phase-out exposure central to in-work credit design.

Revise and Resubmit, Social Indicators Research.

This paper examines how increases in wives' market value, measured through wages, shape intra-household tradeoffs in Japan. Using a Difference-in-Differences strategy, I exploit minimum wage hikes to study their effects on wives' relative income and time allocation. Evidence from the Japanese Household Panel Survey shows that higher minimum wages reduce the gender wage gap by encouraging wives—particularly those with young children—to enter the labor force and extend their working hours. Wives without young children respond by increasing employment while reducing time devoted to home production, whereas those with young children primarily adjust by sacrificing leisure. By contrast, husbands exhibit limited responses to wage changes. These findings highlight that narrowing the gender wage gap, while promoting women's labor market participation, can also intensify their household burden. The results underscore the need for policies that address not only income disparities but also inequalities in unpaid work and time use within households.

with Hao Li. Under Review, Journal of Health Economics.

How does income support after widowhood affect health over the life cycle? To answer this question, we study Social Security Survivors Benefits, which become available to widowed individuals at age 60 in the United States. Using a fuzzy regression discontinuity design with Health and Retirement Study data, we find that eligibility reduces self-reported good health and increases diabetes diagnoses in the short run, with effects concentrated among financially dependent widows. These effects are not explained by changes in medical utilization but coincide with increased food expenditures, suggesting a consumption-based mechanism. To examine longer-run consequences, we exploit the 1965 reform that lowered the eligibility age from 65 to 60 and use cohort mortality data to compare cohorts with different exposure durations. We find that expanded eligibility improves survival to age 70. Descriptive cause-of-death patterns indicate larger mortality declines from cardiovascular disease, tuberculosis, and suicide in high-widow-prevalence states. Together, the results suggest that income support can generate short-run health risks while improving longer-run survival.

with Victoria Prowse.

Over the past several decades, import competition has reshaped employment in advanced economies. Initial waves of manufacturing imports displaced large numbers of less-educated workers. More recently, the rise of high-skill–intensive imports has raised concerns about the vulnerability of educated labor. The labor market consequences of globalization thus depend not only on the overall scale of imports but also on the skill composition of traded goods and the ability of local labor markets to transmit and amplify these shocks. This paper revisits the “China shock” through a new lens by distinguishing between local exposure to low-skill–intensive and high-skill–intensive imports. Even when direct employment losses in manufacturing are comparable, broader adjustments diverge: some local economies absorb displaced workers into non-manufacturing industries, while others experience persistently weaker re-employment prospects. Patterns of adjustment also vary across demographic groups and skill levels, underscoring the role of both worker characteristics and local task requirements in shaping outcomes. By highlighting these composition effects within import shocks, the analysis reconciles mixed evidence on regional recovery and underscores the importance of sector-specific retraining and placement policies.

with Kevin Mumford.

Between 2009 and 2022, the share of U.S. high school graduates enrolling in college immediately fell from 66% to 57%. Using a state-year panel from IPEDS and NCES CCD, we document this decline and estimate an interactive fixed-effects model that absorbs unobserved common shocks while controlling for observable, time-varying factors. We find that demographic age structure, in-state and out-of-state tuition, expected returns to college, K–12 quality measures, and political preferences (Democratic presidential vote share) all significantly predict immediate college-going. We further show that the expansion of the Common Application reshaped sorting across institution types: applications and enrollments reallocate toward colleges better matched to students' academic profiles. A consistent explanation is recruiter behavior—employers update beliefs about the distribution of talent across institutions and increasingly avoid lower-quality colleges, reducing the labor-market return to attending those schools. Together, the results highlight how prices and perceived payoffs, school quality, political climate, and application technology jointly drive the recent decline in immediate college enrollment and reallocate students across the postsecondary landscape.

with Anthony Yim.

with Kevin Mumford and Ben Zou.