Eeshani Kandpal

Measuring the Effect of a Community-Level Program on Women's Empowerment Outcomes: Evidence from India

This paper uses primary data from rural north India to show that participation in a community-level female empowerment program significantly increases access to employment, physical mobility, and political participation. The program provides support groups, literacy camps, adult education classes, and vocational training for rural women in several states of India; the data are from Uttarakhand. The paper uses instrumental variables and truncation-corrected matching on primary data to disentangle the programs mechanisms, separately considering its effect on women who work, and those who do not work but whose reservation wage is increased by participation. The analysis also finds significant spillover effects on non-participants relative to women in untreated districts. It finds consistent estimates for average treatment and intent to treat effects

Does Involvement of Local NGOs Enhance Public Service Delivery? Cautionary Evidence from a Malaria-Prevention Evaluation in India

Using data from an experimental supportive intervention to Indias malaria control program, this paper studies the impact of leveraging local non-state capacity to promote mosquito net usage and recommended fever care-seeking patterns. The supportive activities were conducted simultaneously by three nongovernmental organizations in two endemic districts in the state of Orissa. The study finds that program impact varied significantly by location. Examining three potential sources of this variation (differential population characteristics, differential health worker characteristics, and differential implementer characteristics), the analysis provides evidence that both population and nongovernmental organization characteristics significantly affected the success of the program. The paper discusses these findings as they relate to the external validity of development policy evaluations and, specifically, for the ability of the health system to benefit from limited non-state capacity in under-resourced areas.

Does involvement of local NGOs enhance public service delivery? Cautionary evidence from a malaria-prevention program in India: Does involvement of local NGOs enhance public service delivery? Cautionary evidence from a malaria-prevention program in India

Partnerships between government and non-state actors that aim to enhance the quality or efficiency of service delivery are increasingly common in todays development policy landscape. We investigate the impacts of such an approach using data from an experimental supportive intervention to Indias malaria control program that leveraged local non-state capacity in order to promote mosquito net usage and recommended fever care-seeking patterns. The supportive activities were conducted simultaneously by 3 NGOs, contracted out by the Indian government, in 2 endemic districts in the state of Odisha. We find that program impact significantly varied by location. Examining 3 potential sources of this variation (differential population characteristics, differential health worker characteristics, and differential implementer characteristics), we provide evidence that both population and NGO characteristics significantly affected the success of the program. Specifically, the results suggest that the quality and effort of the local implementer played a key role in the differential effectiveness. We discuss these findings as they relate to the external validity of development policy evaluations and, specifically, for the ability of health and other service delivery systems to benefit from limited non-state capacity in underresourced areas.

A Conditional Cash Transfer Program in the Philippines Reduces Severe Stunting

BACKGROUND Pantawid, a conditional cash transfer (CCT) program in the Philippines, provided grants conditioned on health-related behaviors for children aged 0-5 y and schooling for those aged 10-14 y. OBJECTIVE We investigated whether Pantawid improved anthropometric measurements in children aged 6-36 mo. METHODS We estimated cross-sectional intention-to-treat effects using a 2011 cluster-randomized trial across 130 villages-65 treated and 65 control-with data collected after 31 mo of implementation. Anthropometry characteristics were measured for 241 children in treated areas and 244 children in control areas. Health service use for children aged 6-36 mo and dietary intake for those aged 6-60 mo also were measured. Outcome variables were height-for-age z scores (HAZs) and weight-for-age z scores (WAZs), stunting, severe stunting, underweight, and severely underweight. Impact also was assessed on perinatal care, institutional delivery, presence of skilled birth attendant, breastfeeding practices, immunization, growth monitoring and deworming, care-seeking, and childrens intake of protein-rich foods. RESULTS Pantawid was associated with a significant reduction in severe stunting [<-3 SD from WHO standards for healthy children; β = -10.2 percentage points (95% CI -18.8, -1.6 percentage points); P = 0.020] as well as a marginally significant increase in HAZs [β = 0.284 SDs (95% CI -0.033, 0.602 SDs); P = 0.08]. WAZs, stunting, underweight, and severely underweight status did not change. Concomitantly, several measures of health-seeking behavior increased significantly. CONCLUSIONS To our knowledge, Pantawid is one of few CCT programs worldwide that significantly reduced severe stunting in children aged 6-36 mo; changes in key parenting practices, including childrens intake of protein-rich foods and care-seeking behavior, were concurrent.

General equilibrium effects of targeted cash transfers : nutrition impacts on non-beneficiary children

Based on a randomized evaluation, the paper shows that a household-targeted Philippine cash transfer program significantly raised the local price of key foods relevant for child nutritional status. This shift in prices increased stunting among young nonbeneficiary children by 34 percent (11 percentage points). Price and stunting effects increase in program saturation; at median saturation, the village income shock is 15 percent. These effects persist 2.5 years after program introduction. The authors confirm the price patterns in their experimental sample against price information from nationally-representative household expenditure surveys across the 6-year rollout of the program. Failing to consider such general equilibrium effects may overstate the net benefits of targeted cash transfers. In areas where targeting of social programs covers a large proportion of the population, offering the program on a universal basis may avoid such long-lasting negative impacts at moderate additional cost.

Child Schooling and Child Work in the Presence of a Partial Education Subsidy

ABSTRACT:Could a partial subsidy for child education increase child labor? Using data from the randomized evaluation of a conditional cash transfer program (CCT) in the Philippines, we find that children who were neither in school nor work in the absence of the program not only increased school participation but also increased work for pay. We show suggestive evidence that because the cash transfer only provided a partial schooling subsidy children worked to cover the shortfall in schooling fees. Our findings contribute to the increasing evidence that the design of CCTs, in this case transfer size, matters considerably in terms of achieving program goals.