Other Work

Data on Health and Socioeconomic Characteristics for Indian Districts in 2019

In this collaborative project, Nishtha Kochhar and I are using geocoded data from various sources to map socioeconomic and health data to the latest Indian district boundaries. These data are available here, and we describe the data sources and included variables here.

Other Work

Scaling Gender-Inclusive Social Audits: Preliminary Insights from Process Monitoring and a Randomized Controlled Trial (with Akash Bhatt, Simone Schaner, Aruj Shukla, Natalie Theys, and Charity Troyer Moore)

We study the recent rapid expansion of Bihar's social audits initiative, enabled through employing local self-help group-affiliated women as auditors, as a measure to improve access and delivery of social protection under India's national workfare scheme. We combine process monitoring data collected through qualitative reports and survey interviews and a causal evaluation utilizing administrative data from a randomized controlled trial across the state to paint a rich picture of both the challenges and opportunities of the initiative. The rapid scaling of audits marks a major achievement for Bihar, especially given staffing and resource constraints. Investing in staff, boosting training, and responding to female auditors' unique needs would significantly strengthen the program implementation. The randomized controlled trial finds that audits lead to a significant, sustained reduction in administrative work claims, a result that could reflect some combination of reduced corruption, a goal of audits, or declines in work provision, an unintended consequence. Looking ahead, we aim to collect the additional survey data needed to differentiate between these possibilities and provide further direction for program improvement.

The Timing of Menarche as an Instrument for the Timing of Marriage: A Short Note

In this note, I revisit the use of the timing of menarche as an instrument the timing of marriage to study the effect of the latter on a host of policy-relevant outcomes, including education. I adopt two complementary approaches to understand the validity of this identification strategy in the Indian context. First, I discuss existing evidence to argue that menarche is likely to affect experiences of a young girl's life other than just the timing of her marriage. Second, I use the methodology proposed by Conley, Hansen, and Rossi (2012) to show that the two-stage least squares estimates that use the timing of menarche to instrument for the timing of marriage is not stable.

Do Financial Incentives Improve the Quantity and Quality of Maternal and Child Health Services in Nigeria (with Eeshani Kandpal, Benjamin Loevinsohn, Elina Pradhan, Christel Vermeersch, Gyorgy Fritsche, Opeyemi Fadeyibi, Kevin McGee, Emmanuel Meribole and Wu Zeng)

Nigeria has among the world’s highest rates of maternal and infant mortality rates despite having adequate physical infrastructure and human resources. Nigeria States Health Investment Project (NSHIP), a results based financing (RBF) intervention, was designed as response to the limited accountability of health care workers in Nigeria, typically seen as one of the primary causes behind Nigeria’s under performance on health front. RBF links financing to agreed-upon improvements in key indicators of quality and quantity of health. We find significant improvements in access and utilization of care, along with some improvements in structural and process quality of care. Importantly, NSHIP increased skilled birth attendance rate by fifteen percent and institutional delivery rate by thirteen percent over the baseline average. The increase in institutional delivery rate was entirely driven by an increase in the uptake of public health care that displaced the demand for private health care. The impacts on skilled birth attendance and institutional delivery rates were exclusively driven by the increased uptake among those households who were in the middle of the wealth distribution. This suggests that the constraints in the health care sector do not stem only from limited capacity of the health facilities, but also from the financial constraints of the potential patients. These results demonstrate that NSHIP could leverage RBF to effectively utilize additional financial resources and improved accessibility and quality of health care offered by the existing infrastructure.

Evaluating Tamil Nadu Empowerment and Poverty Alleviation (Pudhu Vaazhvu) Project, Additional Financing Blocks (with Nishtha Kochhar and Smriti Sakhamuri)

Participatory projects have become an increasingly popular anti-poverty and sustainable development tool over the last two decades. The Pudhu Vaazhvu Project (PVP) was implemented by the Rural Development and Panchayati Raj Department of the Government of Tamil Nadu, with assistance from the World Bank. PVP was rolled out in 2 phases. In 2006, the first phase covered sixteen districts and seventy sub-districts. In 2012, PVP was extended to another four sub-districts within existing districts and forty-six sub-districts within the remaining ten districts of Tamil Nadu. This paper uses a weighted differences-in-difference design to compare the second phase of the PVP to the Tamil Nadu State Rural Livelihoods Mission (TNSLRM), Government of India's livelihood-focused participatory project. Both projects are women-centric antipoverty programs and have the same institutional structure. Our results suggest that PVP has higher first-order impacts with better access to credit and savings than TNSRLM. We see no differences in impacts on economic welfare, women’s empowerment, or political participation. Further, PVP's first-order impacts decay with time.