Research

  • Working Papers

    Marrying Young: The Surprising Effect of Education

    with Madhuri Agarwal and Sayli Javadekar (Under Review)

    We study the impact of an increase in women's education on their age at marriage. Exploiting a regression discontinuity for a large-scale school construction program in India, we find that the program increases women's education by 0.4-1 year, but decreases their age at marriage by 0.75-1.55 years for a one-year increase in education. Next, we find that post-marriage, educated women experience less domestic violence, better health care access, lower fertility levels, and a wealthy match in the marriage market. Finally, we find a 50 percent increase in women's wages leads to a delay in age at marriage by 0.1 years.

    Conferences: 16th Annual Conference on Economic Growth and Development, ISI Delhi (2021); Pacific Conference for Development Economics, San Francisco, US, (2022 scheduled); Economics of Education Seminar, Nova SBE (2021).

    Micro-credit to women and temporary migration

    In this paper, I measure the impact of reducing credit market imperfections at the place of origin by providing loans to female members of the migrant household. I use a unique dataset provided by Shram Sarathi, a credit organisation in Rajasthan, India. Using past positive rainfall shock as an instrumental variable for gender of the applicant, this paper finds a positive impact of loans to female members of migrant household on the household well being indicators such as consumption, migration income, health expense, social expense and savings.

    Criminally accused politician and crime against women: Evidence from India

    with S. Maji

    A significant proportion of members of legislative assemblies in India have pending criminal cases against them. This study examines the first-order effect of electing a criminally accused politician (CAP) on the law and order of the constituency. Regression discontinuity estimates show that electing a CAP leads to a substantial increase in crime against women, homicide, and riots and affray. Further evidence shows that there is a fall in public trust in the police force and a significant increase in the number of pending criminal cases in the court. Electing a CAP weakens the state capacity through reporting and prosecution effect.

    Network size and temporary migration

    In this paper, I study the impact of network size on temporary migration. Network size reflects the extent of the informal risk-sharing options available to a rural household. Temporary migration is another form of self-insurance in the rural economy. I particularly focus on how the impact of network size varies depending on the endowment of the network, and show that the channel is via credit constraints for the low endowed network. I use the Indian Human Development Survey 2011-12 data to construct a measure of network size based on the caste of the households. To establish causality, I use historical caste-wise population shares from the 1961 census as an instrument of network size and low permanent migration regions. I find that there is a decrease in temporary migration in highly-endowed upper caste networks and an increase in temporary migration for the low-endowed lower caste networks. Further, I investigate the economic mechanisms using exogenous variation in network income due to negative rainfall shock. I show that an increase in the network size reduces the credit constraints of the low-endowment lower caste households via the income of the network.

  • Work in Progress

    Domestic violence and education: Evidence from school construction program (with M. Agarwal, K. Bergonzoli and S. Maji)

    Understanding the mechanism for better adolescent mental health: A meta analysis (with S. Roy and M. Avenando)

    Candidate selection and political representation: Impact of large scale education program (with M. Agarwal, and S. Maji)