Pricing of Private Education in Urban India
- Enrollment and attendance
- Student learning
- Pricing and fees
In India, there is a particularly large market for private tutoring services, and yet, little is known about how prices influence demand, attendance, and performance among low-income households. Researchers conducted a randomized evaluation to measure the demand for private tutoring services, estimate how pricing can influence take-up and enrollment in these services, and examine how enrollment and performance differs by gender and age. Researchers found that demand for tutoring services decreases with increased prices, but that higher initial prices could help identify students who are more likely to utilize the services.
In India and other developing countries, a growing number of fee-charging private schools are catering to the poor. This growth is taking place in spite of increasing government spending on public education and near-universal access to free public schooling. Many parents use scarce resources to send their children to private schools, and some policymakers have proposed giving government-funded vouchers to disadvantaged students who want to enroll in private schools. In India, there is a particularly large market for private tutoring services, and yet, little is known about how prices influence demand, attendance, and performance among low-income households. In addition, it remains unclear whether these private services may impact genders and age groups differently.
Context of the evaluation
India has the largest educational system in the world, with over 200 million children. Over 96 percent of primary-school-age children are enrolled in school, but less than 40 percent of these children are able to read at a second-grade level. Faced with a low-quality public school system, more and more parents choose to send their children to private schools. An estimated 28 percent of students in rural areas and 65 percent of students in urban areas attend private schools.
This study took place in Delhi, India’s largest metropolitan area and home to a very active private market for education services for the urban poor. Researchers partnered with Pratham, a large education NGO in India that runs regular group-based, after-school tutoring services for children between the ages of 3 and 14 in the slum areas in Delhi. The target population of this evaluation consisted of children from low-income households aged 9 to 14. Given the gender bias prevalent in educational attainment in India especially among older children, researchers explored how pricing policy could affect children based on their age and gender.
Details of the intervention
Researchers conducted a randomized evaluation to measure the demand for private tutoring services, estimate how and whether pricing can influence take-up and attendance of these services, and examine how enrollment and performance differs by gender and age.
The sample for this study consists of 5,437 children and their parents that were chosen randomly from households in the neighborhoods of Pratham’s tuition centers. Roughly 16 percent of the sample (891 students) consisted of households who already pay for tutoring at Pratham, and the other 84 percent (4547 students) consisted of households that had not previously paid for or attended Pratham classes. Households were offered tutoring services at prices ranging from 0 to 250 rupees per month (around US$4) through a two-part pricing design.
Households were first offered tutoring services for their children at randomly assigned prices (the “first price”). If a household accepted the offer price, the household received a randomly assigned discount of up to the amount of the offer price, resulting in a lower price paid (the “second price”).
Researchers measured the impact of different pricing schemes on enrollment and attendance by varying the first price and the second price. Because take-up was based on the offer price, this helped identify how much different households valued the tutoring services. However, households were required to pay the second price, which was varied randomly. This helped identify whether prices screen out those who are less likely to use these services, and whether paying higher prices could increase attendance.
Children were tracked throughout the school year, and researchers collected data on baseline characteristics of the households, take-up, enrollment and attendance in the classes, and child test scores.
Results and policy lessons
Project ongoing, results forthcoming.