Research

JOB MARKET PAPER

Effects of a Conditional Cash Transfer Program Targeting Young Girls: Evidence from India [Most recent draft]

Abstract: This paper evaluates the effects of an ongoing conditional cash transfer (CCT) program in India targeting young girls that provides incentives in the form of deferred cash payments, with eligibility at birth and the largest payment coming at age 21. Introduced in 2007 by the Indian state of Madhya Pradesh, the program Ladli Laxmi Yojana (LLY) requires parents to enroll their first and second born daughters in the program at birth and benefits are conditional on school completion and adoption of family planning by parents. Using difference-in-differences framework, I exploit variation in exposure to LLY by birth cohort, state, and birth order to estimate the causal impact of the program on children's health and educational outcomes, and fertility behavior of parents. I find that the program significantly increased the likelihood of registering the births of daughters. However, the sex ratio does not change, suggesting no change in female mortality. I find evidence of families having children faster and in turn reducing the birth spacing. However, parents are moving toward lower eventual family sizes as I find an increase in the likelihood of parents adopting sterilization. Finally, the results using math and reading test score data show some evidence of parents increasing investments in their daughters' education. These results are consistent with the persistence of traditional son-preference norms in developing countries, despite policies explicitly intended to close gender gaps.

WORKING PAPERS

Impact of Diverse-by-Design Charter Schools on Student Outcomes [Working title] with Sarah Cordes

Abstract: This paper evaluates the impact of diverse-by-design (DBD) charter schools on student outcomes. We use administrative student-level data from New York and city and Denver to assess the impact of studying in DBD charter schools on student achievements. The identification of the study resides on random assignment of students through the lotteries.To address concerns about selection, we employ a variety of empirical strategies. In NYC, we link administrative data with the lottery records provided by participating DBD charters to conduct a standard lottery-analysis. In Denver, which employs a central enrollment mechanism, we use school preference data to employ a propensity score matching procedure developed by Abdulkadiroglu, Angrist, Narita, and Pathak (2017), which is designed to more fully exploit the random variation embedded in the assignment mechanism. Finally, in California, we rely on standard regression models to compare the outcomes of sample schools with outcomes of traditional public and other charter schools located within a 5 mile radius. Preliminary results from Denver and California suggest that DBD students attend more racially diverse schools, perform better in English Language arts, and are less likely to be suspended. In Denver, we find that the reduction in suspension is driven almost entirely by black students. Possible explanations for these findings include the use of small group instruction for students struggling with literacy and the use of restorative justice practices. This work stands to contribute both to the literature on charter school impacts and the literature on the effects of integration more broadly.

Spill-over Effects of Conditional Cash Transfer Program on Health and education [Draft available upon request]

Abstract: This paper examines the spillover effects on health and education outcomes of children living in households with children eligible for a conditional cash transfer program. In particular, I look at the Ladli Laxmi Yojana and its effects on non-eligible siblings of eligible girls. Using a difference-in-differences framework, I exploit variation in state, time and eligibility to estimate the effects of a CCT program on changes in intra-household investments in children's health and education, given that some children are eligible to receive cash transfers and some are not. Looking at the effects on ineligible older siblings, I find no effect on the schooling outcomes of both the older ineligible brother and sister. However, I find a negative effect on the reading and math skills of the older ineligible siblings. These results show some evidence of parents re-allocating resources away from the ineligible siblings towards the eligible daughters.

WORK IN PROGRESS

Effects of Air Pollution on Health and Education: Evidence from India