Working Papers
Maternal Working Hours and Child's Cognitive Development in India: Evidence from Bunching Designs. (Job Market Paper) (Featured at the Economic Misfit Job Market Paper Series)
When mothers work longer hours, does it benefit children through increased income, or hurt them by reducing the time they spend with their mothers? In this paper, I study the effect of maternal working hours on children’s cognitive outcomes in India. I employ a bunching design for identification and model selection, exploiting the fact that 41% of mothers in my data work zero hours. I find that the number of hours a mother works has little effect on children’s reading, mathematics, and writing outcomes. These estimates suggest a near-zero net trade-off between income gains and time costs.
Gender of the firstborn and Mental health in later lives (with Ajinkya Keskar and Sulagna Mookerjee) (draft in progress)
This paper examines the effect of the first-born child’s gender on parents’ mental health in later life. Relying on the exogenous variation in first-child gender, and on the observation that sex selection is generally absent at first birth in India, I find that having a first-born son has limited average effects on parents’ later-life mental health, measured using the CES-D scale. However, the results reveal important heterogeneity by parent gender: fathers with a first-born son exhibit worse mental health outcomes, while the effects for mothers are small and statistically limited. These findings suggest that the traditional preference for sons may be weakening amid social change and modernization, potentially because daughters are increasingly playing a greater role in providing care and emotional support to aging parents.
Financial Shock and Children's Education Outcomes
This paper examines whether prior access to formal banking infrastructure mitigated the effects of India’s 2016 demonetization on children’s schooling and learning outcomes. Demonetization abruptly invalidated approximately 86 percent of the currency in circulation, generating a severe but temporary liquidity shortage that may have constrained households’ educational investments. I exploit a branch-expansion policy that classified districts as underbanked when their population-per-bank-branch ratio exceeded a national cutoff. Combining district-level banking data with repeated cross-sectional data from the Annual Status of Education Report, I first use a local-linear regression discontinuity design to establish that eligible districts experienced a substantial increase in bank-branch expansion. I then employ a difference-in-discontinuities framework to compare changes in educational outcomes across districts located near either side of the eligibility threshold. Although the policy significantly increased banking access, I find limited evidence that additional branches produced differential changes in school choice, private-tuition participation, tuition expenditure, or children’s reading and arithmetic performance following demonetization. One possible explanation is that learning outcomes are cumulative and respond primarily to sustained changes in educational inputs. Households in districts with fewer branches may also have relied on informal borrowing, shopkeeper credit, delayed payments, and family or community support, thereby reducing differences in effective liquidity constraints across districts. Overall, the findings indicate no detectable medium-run differential effect of banking access on children’s educational outcomes.
Work in Progress
Air Pollution and Cognition (with Shiv Hastawala)