Investing in the Education Market: Strengthening Private Schools for the Rural Poor

October 3, 2016

Private schools that rely entirely on student fees for financing are increasingly popular in many low-income countries and parents often prefer these schools to government-run ones. In Pakistan, children in these schools tend to outperform students in government-run schools. But financial constraints can limit the growth of these private schools, whose fees are set low to attract poor students, especially if they cannot access formal credit markets. Researchers will seek to overcome these constraints by studying private financing models -- grants and loans -- to support these schools in Pakistan.

Research area: Education

Country: Pakistan

Evaluation Sample: About 2,000 schools in about 650 villages in Punjab, Pakistan

Timeline: 2012 - 2016

Intervention: Grants, micro-loans, private schools

Researchers: Tahir Andrabi, Pomona College; Asim Khwaja, Harvard University; Jishnu Das, World Bank

Partners: Centre for Economic Research in PakistanAman FoundationTameer Microfinance Bank Ltd.; Research Consultants; Pomona CollegeHarvard University


Policy Issue

So-called low-cost private schools are a growing and increasingly popular option in poor countries. These private schools usually spend less per student than government-run schools, holding down costs by paying their teachers lower salaries than in the government system. Although the teachers often are not as formally qualified as teachers in the government schools, students in these private schools tend to do as well or better than their counterparts in the other schools. One question is how to encourage these schools to expand beyond primary education, and how to encourage them to make further investments in the education they offer. This evaluation of a new financing mechanism for low-cost private schools in Pakistan will help policymakers understand their options for supporting these schools, allowing them to harness the power of the market rather than relying on greater public subsidies to the private sector.


Since 1980, the number of private schools in Pakistan has grown from about 3,000 to about 45,000. Nearly one-third of all primary school children in Pakistan country attend private schools, covering all income spectrums. A 2001 survey showed that about one in five of Pakistan’s poorest families sends their children to private village schools.

Children in these private schools tend to outperform those in public schools, while costs per student can be 20 to 50 percent lower than those in public schools, generally because these private schools hire less-qualified local teachers at lower wages. Despite these advantages, private school growth may be limited by the lack of financing possibilities. This evaluation assesses the benefits of different financing models for encouraging school expansion.

Photo: Caroline Suzman / World Bank

Intervention and Evaluation Details


The intervention centers on two financing approaches for private schools in Pakistan: a grant model and microloans. The program includes a pilot microloan intervention to allow researchers to better develop and target loan products.

Grants: Schools in the randomized evaluation were eligible for a direct cash grant equal to about $525 as long as they were willing to submit an investment plan for the money. In the experiment, all schools complied with this requirement. Furthermore, while schools were expected to use the grants for educational purposes, in practice, they could choose to use only a portion of the grant towards education.

Microloans (Pilot): Together with a major microfinance institute in Pakistan, the team has developed and piloted new financial products for schools that are consistent with the institutes and the government’s loan and regulatory policies. These products were developed and refined over a two-year pilot program offering various loan configurations to about 50 private schools to determine interest, financial need, repayment behavior and logistical requirements.  


This randomized control trial covers about 2,000 schools in about 650 villages across two districts in Punjab, Pakistan’s most populous province.

In the grants program, schools and villages are assigned to two treatment groups and one control group. In the high-intensity treatment group, comprising 228 schools in 75 villages, every private school in the village is offered a grant. In the low-intensity group, only one school in each of 114 villages is offered a grant, resulting in 114 schools with grant offers and 264 schools without. The control group comprises 248 schools in 77 villages. Of the schools invited to participate, 322 received the full grant.

The differences shown between the two treatment groups will help gauge individual and market-level effects. Specifically, the evaluation captures both the “credit constraint” effect (in low-intensity villages) and the “credit policy” effect (in high-intensity villages). In the first treatment, the team can ascertain whether private schools are credit constrained. If the grants had a positive return on investment in these villages, this proves that schools are credit constrained. Yet, no financial policy should treat schools differentially. Therefore, in the second village, the team can replicate the impact of a blanket financial policy by providing grants to all schools.

For the loan intervention, researchers will randomly assign village schools to either a control group, or to receive a new financial loan product or a new financial equity product. [The equity product requires schools to pay back the money depending on how well they do in terms of meeting their revenue needs.]

These financial products are being designed following the analysis of the pilot program. The total number of schools in the program will range between 1,500 and 1,800, depending on the results of a screening exercise to determine interest in the program. The screening exercise covers close to 6,000 schools.

Researchers will collect data through school questionnaires and child testing at the beginning, middle, and end of the interventions, in addition to student, school, and teacher surveys. Altogether, about 20,000 3rd and 4th graders will be tested in Urdu, math, English and civics. The microfinance institution will provide data on loan repayment, and other publically available information will be collected.  

Policy Impact

The design of the intervention will allow researchers to assess whether direct grants and loans can improve private school and student performance, as well as the impact of increasing resources to all private schools in a village, rather than to just one school. The results will help develop appropriate financial products for private school growth and improved student performance in a manner that can then be globally scaled. It will also deliver valuable information for designing microloans for private schools, informing topics such as whether interest subsidies are needed or whether risk can be mitigated through greater due diligence.