Low-cost private schools are expanding across much of the developing world. In India and Pakistan, for instance, more than one quarter of children are now enrolled in low-cost private schools, and this proportion is growing quickly. In Lagos, Nigeria, a recent census estimated that 58% of children enrolled in school attended private schools. The proportion of children in private schools is generally higher in urban areas and in lower grades.

There is a growing body of evidence showing that children in these schools learn more than children in government schools. However, there is a wide variation in performance between schools, with some performing much better than government schools, and others performing worse. Private schools tend to do much better than government schools on some metrics (particularly, teacher attendance, English language learning) but the same or worse on others (for instance learning of local languages).

A number of programs have explored using vouchers or other government subsidies to enable more children to attend private schools.

There is good evidence that these interventions are effective. For instance, a recent randomized trial in India found that children provided with vouchers made greater progress in literacy and math than other children. A similar evaluation of a randomized voucher program in Sindh, Pakistan, found that enrolment rates in villages targeted by the program increased by 30%, along with increases in learning, gender equity, and school facilities.

More importantly, vouchers can be taken to scale. With 1.4 million children, the Punjab Education Foundation already funds the education of more children than the world’s 30 smallest countries combined (and more children than Austria, Bulgaria, Denmark, Estonia, Hungary, Mongolia and New Zealand). As is the case with most low-cost private schools, it does so at a lower cost per student than the government system. In Chile, vouchers fund more than 50% of children in school. Hong Kong, one of the world’s best performing (and most equitable) school systems, effectively uses a voucher style model to fund around 90% of children to attend privately owned schools (like the Punjab Education Foundation, parents in generally cannot top-up the value of the voucher). Voucher models have also been taken to scale in India, the Philippines and Uganda.

Vouchers therefore appear to be an intervention which:

  • Raises learning levels
  • Reduces the cost of schooling
  • Is scalable to reach large numbers of children

Despite that, vouchers have some serious limitations, which, at a minimum, mean that they need to be implemented carefully to have impact.

First, while learning levels in low-cost private schools are generally higher than those in government schools, they are not that much higher. Many voucher systems in the developing world (and low-cost private schools more broadly) still leave large numbers of children not learning. With good training, assessment and other interventions, this learning challenge can be addressed. This means that a voucher system has to be one part of a broader reform agenda, not a substitute for it.

Second, there is a risk of duplication of resources. In many cases, the government ends up funding competing schools, or continuing to fund empty government schools while at the same time paying for a private school next door. Proponents argue that this is a better situation than children being in government schools and not learning. Nonetheless, it raises serious questions about resources in otherwise resource constrained systems. The best voucher systems attempt to target the most vulnerable or underserved areas, but this is difficult to affect in practice.

Third, vouchers, and market-based schooling models in general, have a tendency towards inequality. In Chile, vouchers have benefited students from better-off families more than those from less-well-off families. Good design of the vouchers can ensure that they benefit those who need them most, but this is not guaranteed from the outset.

Fourth, as long as public schools are the dominant provider, a voucher program may prove a distraction from the main task of improving the public school system. Many opponents of vouchers argue that political and financial capital would be better deployed improving the public school system. The best reforms combine both approaches, though were financial and political capital are limited, this can be difficult.

Finally, as voucher systems scale, they begin to become subject to the same political economy challenges which are often the source of problems in the public school system. Good governance structures can mitigate this, but ultimately, as voucher programs scale, they will become further enmeshed in the political and bureaucratic structures which dominate the school system.



Andrabi, T., Das, J., Khwaja, A., Vishwanath, T. and Zajonc, T., Learning and Educational Achievements in Punjab Schools, 2007.
CFBT, Preliminary Study into Low Fee Private Schools and Education, 2011.
Dixon, P., Why the Denial? Low-Cost Private Schools in Developing Countries and Their Contributions to Education, 2012.
Härmä, J., Lagos Private School Census 2010-2011 Report, 2011.
Tooley, The Beautiful Tree, 2009.
World Bank, Using Low-Cost Private Schools to Fill the Education Gap: An Impact Evaluation of a Program in Pakistan, 2013.