[1] Institutional Births and Child Mortality: Evidence from the Janani Suraksha Yojana in India (with Ontiq Dey and Syeda Tasnime Kabir) [Conference Presentation: 8th SANEM Annual Economists' Conference (SAEC) 2025, NEUDC 2024, ASHEcon 2024]
Abstract: The Janani Suraksha Yojana (JSY) is a conditional cash transfer program by the Government of India aimed at reducing maternal and child mortality by promoting facility-based deliveries. This study examines the impact of the JSY on facility birth and child mortality outcomes. Using a difference-in-differences approach, we compare outcomes between eligible and non-eligible women before and after the implementation of the JSY. Our findings indicate that the JSY significantly increased public facility deliveries and reduced infant and under-5 child mortality. Robustness checks confirm the validity of these results. We identify that the JSY improved child mortality outcomes through increased facility births and better access to antenatal and postnatal care. This study contributes to the debate on the efficacy of cash transfer programs by providing robust evidence that well-designed financial incentives can drive significant health improvements, especially in resource-constrained settings.
[2] Parental Involvement and Child Development: Evidence from Bangladesh (with Syeda Tasnime Kabir and Sourov Kumer Paul) (Under Review) [Conference Presentation: ASHEcon 2025, Texas Early-Career Health Economist Network (TEXHEN) Workshop 2024]
Abstract: This study estimates the causal effect of parental involvement on the mental health of primary school children. Using an instrumental variable approach that leverages local parental involvement norms, we find that increased parental involvement significantly reduces children’s cognitive, behavioral, and emotional difficulties. The protective effect of parental involvement varies by socio-economic background: stronger in joint families and among children of more educated mothers, but weaker when fathers live outside the household or when mothers married at an older age. These results are robust across multiple sensitivity tests addressing selection issues and potential violations of the exclusion restriction. We explore two key mechanisms: household hygiene practices and parenting style. Parental involvement is associated with better household hygiene conditions and more accommodating discipline strategies. This study underscores the importance of parental involvement for children’s mental health in resource-constrained settings.
[3] The Lasting Effects of Early Childhood Interventions: The National Vaccination Commando Program in Burkina Faso (with Richard Daramola, Harounan Kazianga, and Abdoul Karim Nchare Fogam) (R&R JDE) [Conference Presentation: Midwest International Economic Development Conference (MWIEDC) 2025, ASSA Annual Meeting 2024, The South East Exchange of Development Studies (SEEDS) 4th Annual Conference, Fifth SANEM-World Bank North America Discussion Forum 2024, Stanford Rosenkranz Symposium 2024, NEUDC 2022, ASHEcon 2022]
Abstract: This study evaluates the long-term impacts of the National Vaccination Commando Program, an early childhood health intervention in Burkina Faso, during the 1980s. Using a difference-in-differences approach, we find significant reductions in child mortality and improvements in educational attainment, including increased primary school completion rates. We also find significant positive effects on adult employment and agricultural productivity, yielding a substantial rate of return on the initial health intervention. These findings underscore the lasting benefits of early childhood health interventions in low-income countries.
[4] Remittances and Household Dependence: Evidence from Bangladesh (with Adesola Sunmoni and Estiaque Bari) (Under Review) [Conference Presentation: SEA 2024, ASSA Annual Meeting 2024, Seventh SANEM Annual Economists' Conference 2024]
Abstract: Remittances, as private transfers to households, can generate two opposing effects: the income effect and the liquidity effect. The income effect may lead to lower labor supply by recipient household members and reduce households’ non-remittance income. Conversely, the liquidity effect may enhance income-generating activities through capital accumulation and improved health productivity. It is theoretically ambiguous to determine which of these effects is relatively stronger. Consequently, this study empirically investigates how remittances affect the likelihood and magnitude of households’ non-remittance income in Bangladesh. To address endogeneity concerns, we employ instrumental variable and imperfect instrumental variable approaches. We find that remittances increase households’ non-remittance income at both the extensive and intensive margins. These findings are robust across different model specifications and relaxation of the instrumental variable exogeneity assumption. Exploring the potential channels of the effect, we find that remittances significantly enhance households’ human and physical capital accumulation, thereby promoting higher productivity and engagement in income-generating activities.
[5] Why Do Workers from Low-Income Backgrounds Earn Less as Adults? (with Eric Nielsen and David Slichter) [Conference Presentation: SOLE 2023]
Abstract: Do children raised in low-earning families earn less as adults because they are less skilled (a “productivity” explanation) or instead because they are worse at translating a given level of skill into earnings (an “inefficiency” explanation)? Differences in observable characteristics can account for most of the gap in earnings between people raised in high- vs. low-earning households, and we present evidence about the nature of the unexplained part. Dividing explanatory factors into those related to productivity vs. inefficiency, we estimate the overall earnings gap is roughly equally due to productivity and inefficiency, with the importance of productivity rising with age. Alternative classifications change the exact shares but preserve the conclusion that both productivity and inefficiency are quantitatively important.
[6] Lost in the Queue: Government Job Aspirations and Labor Market Outcomes (with Elisa Taveras) (Under Review) [Conference Presentation: SEA 2025 (Forthcoming), PacDev 2025; 8th SANEM Annual Economists' Conference (SAEC) 2025]
Abstract: We examine how the structure of public employment shapes labor market outcomes. In Bangladesh, college graduates can apply for elite government positions through the Bangladesh Civil Service examination only before turning 30, creating strong incentives to queue for these jobs. Using a sharp regression discontinuity design around this age cutoff, we find that ineligibility increases the probability of employment, with no change in labor force participation. Moreover, these newly employed college graduates work in sectors and occupations that are more likely to underutilize their skills. Overall, the results point to queuing while eligible and lower-quality matches once eligibility ends.
[1] Program to End Modern Slavery (PEMS) Bangladesh With Guy Grossman (UPenn), Carlos Schmidt-Padilla (UC Berkeley), Harsha Thirumurthy (UPenn), Catalina Udani (UPenn), Nudrat Faria, and Alyssa Heinze (UC Berkeley). Co-Principal Investigator.
B-PEMS is a five-year (2022-2027), $7.9 million collaborative project between Winrock International and PDRI-DevLab. The project is funded by the US Department of State’s Program to End Modern Slavery (PEMS). The project’s primary goal is to reduce the prevalence of Trafficking in Persons in Bangladeshi farming and fishing households vulnerable to climate change. Winrock International provides technical assistance to farmers, fishers, and fish farmers in climate-change affected regions to conduct climate-smart agriculture and aquaculture and to link them to skills training, job placement, and micro-enterprise development for alternative livelihoods. Using a randomized controlled trial (RCT) study design, we are evaluating the effect of Winrock’s community-level activities on the prevalence of climate change-induced human trafficking and associated risk factors.