Early-Childhood Education in India
Introduction
Ganimian, Muralidharan and Walters (2024) compute the MVPF of expanding early childhood services in India by adding a pre-school instructor to government-run education centers.
The Government of India provides both pre-school education and health and nutrition services to 36 million 3- to 6-year-olds through the Integrated Child Development Services (ICDS) program. Services are provided through 1.35 million anganwadi centers (AWCs), each serving a catchment area of 400-800 people and typically staffed with one anganwadi worker (AWW) and one anganwadi helper. An AWC’s staff is usually overworked and spends limited time on pre-school instruction.
The authors partnered with the Indian state of Tamil Nadu and provided 160 randomly-selected centers within the state the opportunity to hire an extra early-childhood education (ECE) facilitator to focus on pre-school instructional tasks. The paper compares outcomes between children in treated centers with a control group of 160 AWCs in the same state and find that the intervention increased net instructional time, math and language test scores, and decreased stunting and severe malnutrition.
Pays for Itself
Net Cost
To compute the net costs of the program, the paper considers the direct implementation costs and the fiscal externalities caused by better education and nutrition.
The average implementation cost of the intervention was roughly Indian Rupees (INR) 74,000 per center. This figure includes a one-time training cost for facilitators and 18 months of salary and administrative costs.
In addition to the direct costs, the intervention also increased expected government revenue. Children in the treatment group had better education and health outcomes, and therefore have higher expected future earnings and expenditures. This translates into higher expected tax revenue from both income taxes and value-added taxes. The paper calculates the average future tax rate that would mean the intervention would pay for itself. To calculate this number, the paper divides the direct cost of the intervention by the estimated present discounted value of future earnings: (74,478 / 964,439) x 100 = 7.7%. The paper then argues that the government is likely to collect more than 7.7% of the increase in earnings: most of the extra income will be spent, and the Indian value-added tax (GST) covers the majority of transactions at a rate between 12 and 18%. If these assumptions hold, it would imply a negative net cost of the program.
Willingness to Pay
Willingness to pay is determined by the change to future income. The paper estimates the impact on future income from both improvements in education and health. In order to estimate the increase in income due to better education, the paper multiplies their estimated impact of the intervention on a composite measure of learning (0.11 standard deviations, p < 0.01) by the estimated effect of test scores on earnings from Kline and Walters (2016), who find that each standard-deviation increase in early-childhood test-scores is associated with a 0.13 increase in earnings as an adult. The overall effect of the intervention on earnings is therefore an increase of 0.11 x 0.13 x 100 = 1.4%. The paper applies this number to the predicted present discounted value of earnings for individuals in the sample, which they estimate to be roughly INR 3.6 million in the absence of the intervention. The resulting estimated earnings increase through the education channel is 0.014 x 3.6 million = INR 51,796 per treated child.
The intervention reached an average of 14 children per center per year. In addition, about one third of the intervention costs went towards treating a new cohort of children whose outcomes were not tracked. Thus, the predicted benefit of the intervention per center is INR 51,796 x 14 x 1.33 = INR 964,439 million.
MVPF Summary
The program has a positive WTP and negative net costs (as long as the government captures 7.7% or more of the increase in incomes through taxes). Therefore, the program has an infinite MVPF.
References
Ganimian, Alejandro, Karthik Muralidharan, and Christopher R. Walters (2024). “Augmenting State Capacity for Child Development: Experimental Evidence from India.” Journal of Political Economy, 132(5): 1565-1602. DOI: https://doi.org/10.1086/728109
Hoddinott, John, John Maluccio, Jere R. Behrman, Reynaldo Martorell, Paul Melgar, Agnes R. Quisumbing, Manuel Ramirez-Zea, Aryeh D. Stein, and Kathryn M. Yount (2011). “The Consequences of Early Childhood Growth Failure Over the Life Course.” International Food Policy Research Institute (IFPRI) Discussion Paper No. 01073.
Kline, Patrick and Christopher R. Walters (2016). “Evaluating Public Programs with Close Substitutes: The Case of Head Start.” The Quarterly Journal of Economics, 131(4), 1795-1848. DOI: https://doi.org/10.1093/qje/qjw027
Policy Features
- Category
- Education
- Sub-Category
- Child Education
- Beneficiary Type(s)
- Children and Youth, Children under five, Students
- Average Age
- 4
- Country of Implementation
- India
- Year of Implementation
- 2016
- Empirical Method
- RCT
- Research Type
- Primary
- Peer Reviewed
- No
- MVPF Publication Link
- doi.org/10.1086/728109