Seojung Oh - Research

Job Market Paper

The Role of Intermediaries in Administering In-kind Transfers: Evidence from SNAP and Food Retailers (Awarded Dean's Summer Graduate Fellowship in 2022)

Abstract. Many in-kind transfers are provided not as direct government transfers but rather as entitlements that allow recipients to access specific goods from private providers. This structure makes these providers key intermediaries, whose participation decisions can substantially impact the efficiency, access, and equity of in-kind transfer programs. Despite this significance, relatively little attention has been paid to how firms respond to these programs and the effects on benefit recipients, especially within the context of nutrition assistance. This study examines grocery retailers’ responses to changes in the Supplemental Nutrition Assistance Program (SNAP) following the adoption of Electronic Benefit Transfer (EBT) technology and explores subsequent implications for program efficiency and equity. Leveraging the staggered rollout of EBT across counties, I find that the transition to EBT led to a 9.6% decrease in the number of SNAP-authorized retailers, primarily driven by declines in the number of small SNAP stores. Evidence suggests that these exits among small stores were due to the initial costs associated with EBT setup. Further analyses indicate that, while EBT improved program efficiency by reducing administrative costs and increasing SNAP participation, these efficiency gains were largely concentrated in areas with relatively good access to large SNAP retailers.

Journal Publications

New Evidence on the Cycle in the Women, Infants, and Children Program: What Happens When Benefits Expire (with Marianne Bitler, Jason Cook, and Paige Rowberry). Forum article commissioned by the National Tax Journal (2024). 77(1): 175-197.

Abstract. We provide evidence of a benefit redemption cycle in the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC). Using novel administrative data on item-level redemptions, we show that unlike SNAP, WIC redemptions peak at both the beginning and the end of the month. Beginning-of-the-month excess redemptions are concentrated among more fully redeemed items such as infant formula, while end-of-the-month excess redemptions are concentrated among less fully redeemed items such as infant meats. We document that a substantial share of beneficiaries go at least one month without redeeming anything and discuss how administrative burdens may drive the WIC cycle.

Unintended Health Benefits of Adopting Preventive Behaviors during Virus Outbreak (with Sok Chul Hong and Eunju Lee). Health Economics (2023). 32(2): 324-342.

Abstract. This study investigates whether changes in risk perception play a critical role in improving preventive behaviors and health outcomes by examining the 2009 H1N1 influenza (or swine flu) pandemic in Korea. We employ a difference-in-differences estimation strategy by comparing the differential effects of the H1N1 outbreak on the confirmed cases of diseases that can be prevented by preventive behaviors (e.g., intestinal infections) and the cases of diseases that cannot (e.g., injuries). Using unique administrative data from South Korea’s National Health Insurance Service (NHIS), we find that the exogenous increase in health risk reduced the incidence of intestinal infections compared to the injuries during the H1N1 influenza outbreak. The reduction was the most substantial among children under five years of age, with a 25.4% decline in cases of intestinal infections relative to injuries. Our findings are robust across various alternative specifications. We provide suggestive evidence that active adoption of preventive behaviors is one of the channels underlying the unexpected decrease in diarrhea cases. The effects, however, faded away shortly after the end of the pandemic and did not last in the long run.

Other publications

“Economic Costs of the Opportunity Gap" (with Marianne Bitler), as a commissioned paper for the Committee on Exploring the Opportunity Gap for Young Children from Birth to Age 8, National Academies of Sciences, Engineering, and Medicine, Washington, DC: The National Academies Press, 2022

Working papers

Parental Gender Norms and College Major Choices (with Chulhee Lee and Hyewon Kim), Revise and Resubmit at Economics of Education Review.

Abstract. Why do women sort into “female” majors? This paper investigates whether parental gender norms affect children’s choice of college major. For identification, we leverage the regional variation in sex ratio at birth (SRB) in each parent’s birthplace in South Korea, which reflects sex-selective abortion practices, as a proxy for the region’s degree of gender norm conservativeness. We find that having fathers with more traditional gender norms increases the likelihood that daughters enroll in female-dominated majors such as education or home economics. However, this relationship is disappears for more recent cohorts who entered college after 2000. No significant effect is observed for sons, and only a weak influence of maternal gender norms is detected. We offer two possible explanations for the diminished influence of parental gender norms among younger generations: the growing labor-market disadvantages associated with female- dominated majors and the convergence in parental preferences regarding sons’ and daughters’ occupations. Our findings suggest that traditional gender stereotypes can play an important role in shaping gender-segregation in educational choices through cultural transmission.

Effects of the Welfare Time Limit on Welfare Use and Labor Market Outcomes: New Evidence from Welfare Reform Experiments

Abstract. The introduction of welfare time limits was arguably one of the most fundamental and striking changes of the 1996 welfare reform. This study reexamines how this policy change influenced welfare participation and labor supply among single mothers using experimental data from two welfare “waiver” experiments: Florida’s Family Transition Program (FTP) and Connecticut’s Jobs First Program (JFP), with a particular focus on JFP data, which has not been previously explored. I follow the identification strategy of Grogger and Michalopoulos (2003), which leverages the age of the youngest child in each household to isolate the effect of the time limit from other policy components of the 1996 welfare reform. My replication of the FTP data yields similar results with prior studies, indicating that the time limit policy reduced welfare use. However, results from the JFP data reveal novel and contrasting findings: Single mothers exposed to welfare time limits did not reduce their welfare use and may have even increased it. Importantly, more people in Connecticut began to combine both welfare and work after the time limit was introduced. I find suggestive evidence that Connecticut’s unusually generous earned income disregard policy might be the reason behind these different results. This new evidence demonstrates that the effects of welfare time limits are context-dependent and should be understood within the broader context of other policies.

The Gendered Effect of Manufacturing Decline on Infant Health: Lessons from Import Competition from China (with Chang Jae (CJ) Lee; draft upon request)

Abstract. While the negative consequences of rising Chinese import competition on U.S. manufacturing employment and low-skilled workers are well-documented, the spillover impacts on subsequent generations remain relatively unexplored. In this paper, we examine the influence of the decline in manufacturing employment due to increased Chinese imports on infant health outcomes. Infant health at birth is strongly linked to socioeconomic status and can have long-term effects into adulthood, playing an important role in reducing economic disparities. By analyzing restricted-access birth data from 1990 to 2007 and employing a shift-share instrumental variable approach, we identify exposure to import shocks across U.S. local labor markets. Our initial analysis finds no overall effect of import shocks on infant health, despite their significant impact on manufacturing employment. However, a gender-specific analysis reveals a nuanced picture. Exposure to import shocks predominantly affecting male manufacturing employment is associated with a significant decline in average birth weight (11.9-gram reduction). Conversely, exposure to shocks primarily affecting female manufacturing employment is linked to improvements in average birth weight (10.3-gram increase). These patterns are consistent in other measures of infant health, including the incidence of low birth weight. We find no evidence that these effects are driven by changes in fertility or migration patterns. Instead, our findings align with existing research suggesting that male job loss has a significant negative impact on family income, which negatively affects child outcomes, while female labor exit can have positive effects on child health due to changes in maternal health and increased time with children. These results underscore the importance of gender-specific dynamics in shaping child health outcomes.

Does Unemployment Insurance Attenuate the Scarring Effect? (with Wonsik Ko; draft upon request))

Abstract. While a large body of research has highlighted the substantial and persistent earnings losses suffered by displaced workers, referred to as the "scarring effect," policy interventions aimed at alleviating these losses remain largely unexplored. This paper bridges this gap by evaluating the long-term effects of unemployment insurance (UI) on post-displacement outcomes. Leveraging cross-state variations in UI policy over time, we find that a 13-week extension in maximum UI duration reduces long-term earnings losses by 10-15% over a 10-year period. This reduction is primarily driven by an increase in hours worked, with minimal impact on wage rates. Conversely, UI replacement rates have a small and largely insignificant effect on earnings, wage rates, and labor supply. These patterns are consistent across different educational levels, racial backgrounds, and economic conditions at the time of displacement. To explore the mechanisms, we provide suggestive evidence that the non-pecuniary aspects of a job improve with longer maximum duration, and explain a substantial portion of the effect of maximum duration on labor supply.

Work in Progress

SNAP EBT, SNAP Participation, and Infant Health (with Marianne Bitler, C.J. (Chang Jae) Lee, and Leah Shiferaw)

Abstract. This paper examines the effects of the adoption of the Electronic Benefit Transfer (EBT) system in the Supplemental Nutrition Assistance Program (SNAP) on SNAP participation rates and infant health at birth. Replacing paper coupons with EBT cards lowered transaction costs and stigma associated with SNAP use, potentially in- creasing participation among eligible families. While prior research has documented mixed effects of EBT on SNAP caseloads, downstream health impacts remain under- studied. Leveraging newly collected county-level EBT rollout data and the staggered adoption framework, we find that EBT adoption increased overall SNAP participation, especially among less disadvantaged groups who are more likely to churn, suggesting that reduced administrative burden may have improved program accessibility. We then examine whether these changes translated into better infant health outcomes. Preliminary evidence suggests some improvements in birth outcomes, particularly for infants born to Black mothers. Given that infant birth outcomes are strong predictors of later-life outcomes, our findings suggest the potential long-term benefits of reducing administrative barriers in safety net programs.

Effects of EBT on Program Enrollment: Evidence from a Mountain State (with Marianne Bitler, Jason Cook, and Caitlin Dutta)

Abstract. This paper investigates how a change in the benefit disbursement system in an in-kind transfer program from paper vouchers to an Electronic Benefit Transfer (EBT) system affects program enrollment and attachment, focusing on the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC). We leverage the rollout of the WIC EBT system (called eWIC) in the state of Utah to identify these effects. Specifically, we employ a regression discontinuity (RD) analysis that uses the distance to the end date of a participant's WIC certification period as a running variable and the county's eWIC transition date as the discontinuity. Rich administrative data from the state of Utah's WIC program allow us cleanly identify each individual's certification period as well as whether the person recertified successfully for WIC, terminated WIC enrollment early, and if so, the reason for the early termination. While we find a strong first-stage effect, the results indicate that the transition to eWIC had little impact on WIC enrollment or attachment to the WIC program. Our findings suggest that although eWIC may have improved some aspects of the WIC experience, WIC program participation and much of its administrative burden remained unchanged. The eWIC transition in Utah, at least in the short run, did not appear to provide recipients with additional incentives to remain on the program large enough to offset the existing administrative burden.