artwork from the cover of this publication

NO.  33

Other Titles in this Series

Also available
Educating Children in Poor Countries
Arye L. Hillman
Eva Jenkner

©2004 International Monetary Fund

[Preface]  [Educating Children in Poor Countries ]  
[Educational attainment in poor countries]  
[Supply and demand shortfalls]  [Pros and cons of user payments
[Conclusions]  [Author Information]


Children are entitled to a free, quality basic education. Recognizing this entitlement, world leaders made the achievement of universal primary education by the year 2015 one of the Millennium Development Goals. In 2004, this goal appears to be out of reach for many poor countries. School attendance, especially for girls, is far from universal, and many children drop out of school before completing their primary education. Many children who do attend school receive an inadequate education because of poorly trained, underpaid teachers, overcrowded classrooms, and a lack of basic teaching tools such as textbooks, blackboards, and pens and paper.

The problem in many developing countries is that governments lack either the financial resources or the political will to meet their citizens' educational needs. In response, poor parents in some low income countries have organized and paid for their children's education themselves. It is true that school fees and other user payments are a heavy burden for some parents to bear. But, given the alternative—children receiving no education at all—such payments can represent a temporary, if less than ideal, solution to the problem.

Economic Issue No. 33 examines this critical issue. Michael Treadway prepared the text based on "User Payments for Basic Education in Low-Income Countries" (IMF Working Paper 02/182, November 2002), by Arye L. Hillman and Eva Jenkner, which is available free of charge at The working paper provides statistics on schooling trends, the theory underlying the equity-efficiency problem, case studies, and a full bibliography. Public Finance and Public Policy: Responsibilities and Limitations of Government (Cambridge, U.K., and New York: Cambridge University Press, September 2003), a textbook by Professor Hillman, provides a broader examination of education and other public policy choices.

Educating Children in Poor Countries

In an ideal world, primary education would be universal and publicly financed, and all children would be able to attend school regardless of their parents' ability or willingness to pay. The reason is simple: when any child fails to acquire the basic skills needed to function as a productive, responsible member of society, society as a whole—not to mention the individual child—loses. The cost of educating children is far outweighed by the cost of not educating them. Adults who lack basic skills have greater difficulty finding well-paying jobs and escaping poverty. Education for girls has particularly striking social benefits: incomes are higher and maternal and infant mortality rates are lower for educated women, who also have more personal freedom in making choices.

Despite considerable progress over the past two decades, however, school attendance in the world's poorest countries is by no means universal. According to the United Nations Development Program, about 113 million children worldwide were not enrolled in school at the end of 2003.

Because basic education is a recognized entitlement and society benefits when children are educated, the state should bear the cost, especially for poor children. In many poor countries, however, the state does not fulfill this obligation. The government may not have the resources to provide a free education for all, either because there is a large, untaxed shadow economy and the tax base is small, or because tax administration and collection are ineffective. And, in many countries (often the same ones), the state does a poor job with the resources it has. Funds are badly managed, and inefficiency or outright corruption may prevent resources from reaching schools. The political will to provide universal education may also be absent in undemocratic societies, if ruling elites fear that an educated population will be better equipped to challenge them. Although correcting these deficiencies is clearly a priority, it will take time. What can be done in the meantime to ensure that poor children in poor countries get an education?

A recent World Bank study found that payments by parents for basic education were widespread in 77 out of the 79 countries surveyed. User payments can take different forms. School fees may cover teachers' and administrators' salaries, materials such as pencils and textbooks, and school maintenance. Or parents may make payments in kind, for example, providing food for the teachers, assisting in the classroom, or contributing their labor for school construction or maintenance. It is important to examine the effect of such user payments on education in poor countries before deciding whether they should be continued, reformed, or prohibited.

Educational attainment in poor countries

The bulk of the world's poor live in East Asia and the Pacific, South Asia, and sub-Saharan Africa. School enrollment in these regions mirrors their economic performance.

In fast-growing East Asia, primary school enrollment was virtually universal (99 percent) by 1997, up from 86 percent in 1980, according to World Bank data. (These are net enrollment figures, defined as the percentage of children in the appropriate age group who are in school; gross enrollment rates include children who are older than is customary for their grade level and may thus exceed 100 percent.) South Asia lags far behind, with only 77 percent of children enrolled in 1997, but this represents a huge improvement from 1980, when net enrollment was only 64 percent. Sub-Saharan Africa is a different story. Although data for 1997 are not available, we know that the gross enrollment rate fell slightly between 1980, when it was estimated at 54 percent, and 1996. It is probably about 50 percent today.

Many children in poor countries drop out of school before graduating. In 1999, the completion rate—the percentage of children of graduating age who actually completed primary school that year—was 73 percent in developing countries as a group—81 percent in East Asia, compared with 50 percent in South Asia and sub-Saharan Africa. As discouraging as these figures are, they, too, represent an improvement: completion rates were lower in 1990.

But completion of primary school is no guarantee that children have acquired basic academic skills. Surveys in a number of low-income countries document that many adults who have received some schooling (five­six years or less) are functionally illiterate and innumerate.

Equally disheartening are the disparities in educational attainment between different groups within countries and regions. Sadly, but not surprisingly, in most low-income countries, children from poor families are much less likely to be in school than children from more affluent families, except in countries like Uzbekistan that have a strong legacy of universal education. In Pakistan, in the early 1990s, 86 percent of rich children aged 6­14 were in school, compared with 37 percent of poor children, making for a rich-poor gap of 49 percentage points; the gap was 52 percentage points in Senegal, and 63 percentage points in Morocco. The gap is narrower but still wide in countries like Bangladesh, Ghana, and Indonesia.

The disparity between sexes is even more dramatic. Girls figure disproportionately among the children who do not attend school in all low-income countries. The bias against girls is especially marked in South Asia and Africa; in many other countries, boys and girls attend school in roughly equal numbers, and, in some, the male-female ratio slightly favors girls.

Children with disabilities are particularly disadvantaged. It is estimated that only 5 percent of learning-disabled African children who need special education go to school, whereas 70 percent of them could attend if the schools had the right facilities. Even with better facilities, however, parents may send disabled children out to beg rather than enroll them in school.

Supply and demand shortfalls

Given the evident benefits of a basic education, why do so many children in so many countries fail to get one? According to economic theory, a lack or shortage of a good or service that is clearly desirable stems from either a failure of demand or a failure of supply. That is, there are demand-side and supply-side impediments.

Cost is one obvious reason why demand for education might be low, given that poor families must meet their essential needs—food and shelter—first. In addition to tuition, books, and school supplies, there may be expenses for transportation and clothing. Some parents might be able to afford an education for their children if they could spread the cost over several years, but many poor people in low-income countries do not have access to credit.

Demand for education may not be present because of the opportunity costs of educating children: parents may prefer that their children work to supplement household income, do household chores, or care for sick family members. In African countries afflicted by AIDS, children may stay out of school to care for sick parents or orphaned siblings. Opportunity costs make even free schooling unaffordable for some families.

Also affecting demand for education is its perceived value. Parents may not have enough information to assess the return on an investment in their children's education accurately, or they may see the return—sometimes correctly—as too low to justify the cost, perhaps because of the poor quality of the education available to them. They may believe that jobs in the local economy do not require academic skills or that getting a job depends more on personal connections than on skill. They may feel that more competitive urban job markets are too far from community and family to consider. Or they may simply be unaware of the opportunities that exist, especially if they are uneducated themselves. If they live in a community that devalues education or frowns upon educating girls, they may be reluctant to violate social norms.

Mexico's Progresa program, for example, has cut the cost of attending school by more than half, but enrollment has risen only 10 percent. Under the program, the central government makes payments directly to village officials, who distribute them to female household heads—conditional on their sending their children to school and taking them to clinics for preventive health care. The income transfers amount to negative user payments. In countries like Mexico, where weak demand for education is not attributable to cost, school attendance will not improve until cultural barriers and prejudices are overcome and the higher opportunity costs for girls than for boys are addressed.

There may be a failure of supply for the reasons laid out above: a government may lack sufficient resources to provide educational services or be administratively incapable of channeling resources to the schools that need them. Government-financed schools may exist in urban areas but not in rural areas or may vary greatly in quality. Or publicly financed schools may be a low priority if the country's economic and political elites send their children to private schools. Public resources may be diverted from primary education to institutions of higher education that serve the governing elite or shifted out of education altogether and into other projects—say, military buildup—favored by elites seeking to protect their interests.

Corruption is another reason poor children in poor countries may not have access to quality government-financed schools: government officials may shun spending on schools in favor of big-ticket items such as defense or road construction, for example, since funding for them is easier to divert and such projects are likelier to involve kickbacks.

Foreign donors—aid tends to account for a large share of poor countries' budgets—also seem to favor capital spending over recurring school expenses like salaries and textbooks.

Pros and cons of user payments

It is important to distinguish between the factors influencing the demand for and supply of education. User payments are less likely to increase school attendance when low enrollment is due to a failure of demand than when it is due to a failure of supply. Obviously, when the problem is a failure of supply because a government lacks the means or will to build or staff schools for children from poor families, user payments are one way parents can provide an education for their children. Of course, demand-side influences make user payments ineffective. If parents are unwilling or unable to put their children in school when education is free, they will not send their children to school when they are obliged to pay for it out of their own pockets.

There is, however, one qualification: user payments may increase the quality of education, motivating parents who were previously unwilling to pay for their children's education to do so. Studies have confirmed that user payments increase demand for schooling if they lead to academic and other improvements. For example, at first, school attendance declined in rural Mali when school fees were raised, but this decline was more than offset by the rise in attendance that followed the construction of new schools closer to rural communities and the realization by parents that the quality of the schools had improved.

User payments in practice
Chad, a Central African nation with an estimated gross national product per capita of only $215, is a classic example of a low-income country with weak public-revenue-generating capacity (total revenue is only 8 percent of gross domestic product), severe net underenrollment in school (about 50 percent), a low primary school completion rate (20 percent), and widespread illiteracy (over 60 percent). As dismal as these averages are, extreme regional and gender disparities mean that the situation is even worse for many rural children and for girls.

However, Chad has a long history of parental involvement in both financing and managing basic education. Community-managed schools were already in place during the colonial period. After independence, parents' associations responded to the country's political and economic instability and to the government's failure to provide even the most basic education for many rural children by taking full responsibility for managing rural schools. These associations not only took over schools that the government had abandoned but also built and operated new ones. Twenty percent of all primary school pupils in Chad are enrolled in community-run schools.

Parents also contribute to the government-run schools. The average annual cash contribution is only about $2, but parents also bear the costo of books and other supplies and contribute time as volunteers. Attendance is voluntary, but informal insurance mechanisms prevent children from being excluded even if their parents cannot pay the modest fee.

Parent-teacher associations hire and supervise more than half of Chad's primary school teachers, who teach in their own villages (centrally trained teachers are seldom willing to relocate to remote regions) after receiving minimal training. The government, with help from the World Bank, pays 80 percent of their salaries (which are only about one-third of what government-employed teachers receive), and the community covers the rest.

The case of Chad demonstrates the value that even desperately poor parents may attach to their children's education. Yet, despite their sacrifices and their cooperation, educational outcomes in Chad are poor: fewer than half of adults with six or more years of schooling can read with fluency. One can only do so much with scant resources and poorly trained teachers. The government has promised to increase its spending on education with resources freed up under the Heavily Indebted Poor Countries Initiative of the World Bank and the IMF and from anticipated future oil revenues. In the meantime, user payments remain the only realistic means of making educational ends meet. Without them, the children of Chad would be far worse off.

If demand is inadequate because parents are misinformed about the true value of education for their children, user payments may have a "demonstration effect": when user payments allow more children to go to school, gain skills, and ultimately find jobs, parents of other children may be persuaded that schooling is worthwhile.

Moreover, when parents are required to pay for their children's education, they are more likely to monitor teacher performance to be sure they are getting their money's worth, so a virtuous circle is created. A study in the Philippines found that primary schools that relied more heavily on local funding (both from municipal governments and from user payments to parent-teacher associations) tended to be more cost effective. By shifting education financing to the local level, user payments can improve school management. Schools that are accountable to the local community—and to parents, specifically—have a greater incentive to provide a high-quality education, even though corruption may exist at the local level as well as at the national level. Put simply, it is easier for parents to watch over local officials and to take collective action against them if need be. In the end, if payment is voluntary and schools do not perform satisfactorily, parents can take their children out of school, and school administrators will lose some of the funding out of which their own salaries are paid.

Malawi and Uganda
Moving beyond user payments

In the mid-1990s, Malawi replaced a system of user payments with free primary education. Although government spending on primary education rose sharply, quality declined as school enrollment surged by 60 percent (1 million new students), leading to overcrowding and a shortage of teachers. A delay in donor funding contributed to the country's failure to prepare adequately for increased enrollment. Teacher performance also appears to have deteriorated, because parents, relieved of the financial burden, felt less compelled to monitor the teachers. On the other hand, parents were still expected to contribute labor and materials to school construction and to buy school supplies and clothes; this, together with the opportunity cost of forgone child labor, left total costs high for some parents. The result was a rise in dropout rates. By 1999, the primary completion rate was only 50 percent. Gender biases persisted as well.

Uganda moved most of the way toward free universal primary education in 1997, when it waived tuition for up to four children per household. Families remained responsible for school supplies and contributions to construction, as in Malawi, and had to purchase uniforms and pay final examination fees as well. Uganda did better than Malawi, however, in preparing for the influx of new students. The government doubled the share of recurrent government spending targeted to primary education and used external aid to train new teachers, build classrooms, and purchase teaching materials. Even so, educational quality has fallen, with high pupil-teacher and pupil-classroom ratios and inadequate materials, and net enrollment has declined, from 85 percent in 1997 to 77 percent in 2000. Gender biases are still reported, as in Malawi.

The experiences of these two countries demonstrate that universal public education cannot be achieved simply by abolishing fees and opening classroom doors. Obstacles arise on both the demand and the supply sides. Education can still be costly for the poor, thus discouraging enrollment, especially when maintaining quality is a problem. Maintaining quality, in turn, is not just a matter of increasing spending; good planning, implementation, and monitoring are also necessary. Achieving the goal of universal school attendance in both countries will require measures to relieve poor parents of all education-related costs, perhaps through a system of income transfers.

User payments are a means of overcoming a political bias toward urban elites at the expense of the rural poor. They enable poor parents to pay directly for their own children's education instead of subsidizing, through their taxes, the education of rich children. While the ideal would be for the rich to subsidize education for the poor through progressive taxation, in reality, the opposite often occurs; user payments can prevent such an unfair outcome.

Educational reform based on user payments may even generate a virtuous political circle. As more people become better educated, they seek greater democratic participation for themselves and greater accountability from their government. It is not inconceivable that, in some countries, this is precisely why the existing regime opposes educational reform that would provide all children with a quality education—or any education at all.

In countries where public finance for schools is not available because of tax evasion, user payments may restore the link between money paid out and services received. That is, people may be prepared to pay for schools because they can see what they are getting for their money, whereas they may evade taxes because they perceive, correctly, that their taxes will be appropriated by corrupt officials and not reach the schools.

That said, user payments are not a panacea. Education that is publicly financed and free to users is both more equitable and more inclusive than education that is paid for by parents. If school attendance is compulsory, a required user payment is a regressive tax—that is, the burden of paying the tax is greatest for the households with the lowest incomes. In the simplest arrangement, the user payment is a fixed amount per child. Such a payment represents a larger share of a poor family's income than of a rich family's; as a share of disposable income, the burden on the poor family is even greater.

Although user payments allow parents to finance education for their children when quality publicly financed schooling is not available, voluntary school attendance (that is, attendance contingent on payment) may lead some parents not to send their children to school. In general, the higher the fee, the more likely it is that families will decide not to educate their children or be forced to make choices about which of their children to educate. Some children may therefore not receive an education. Such exclusion is unfair. The alternative may be even worse, however; in the absence of user payments in countries whose governments are unable or unwilling to provide free schooling, all poor children may be excluded from getting an education.

When children cannot go to school because their parents cannot afford user fees, not only is the situation unfair and socially unjust but it can also be economically inefficient. Many school costs—such as the physical infrastructure and teacher salaries—are fixed, and allowing an additional child to attend would not increase them. This is especially true in rural areas in low-income countries, where the one-room schoolhouse with one teacher and one chalkboard is a common sight. It costs little or nothing to put one more child into such a school. Of course, quality will decline after some point, as the teacher's attention is divided among more pupils.

In principle, a government could monitor the financing of schools through user payments and make the payments progressive by exempting poor families or basing payment on income. This, however, means involving the government bureaucracy—whose very inadequacies forced many parents to assume the responsibility for their children's schooling in the first place—in the school system. A small community usually does a better job of monitoring itself. Members tend to know about their neighbors' economic circumstances. The information they possess about each other can help keep a system of personalized user payments honest, deterring those who might, without cause, seek special treatment. Community ties can thus support a rudimentary insurance system, so that families suffering an economic hardship, such as the disability of an income earner, can continue, with their neighbors' help, to send their children to school, countering the exclusion from schooling that can result from voluntary user payments.

Because children are entitled to a free, quality, basic education that does not depend on their parents' willingness or ability to pay for their education, user financing of schools is a temporary solution until the government is able to exercise its responsibility. The success of user payments should not be interpreted by governments as an indication that they have been absolved of this responsibility.


User payments for basic education should never be more than a temporary solution: the ideal arrangement and the appropriate goal of education policy remain universal education financed by government out of public revenues. User payments are undesirable because they are a regressive tax when school attendance is compulsory. Voluntary user payments are undesirable because children are excluded from schooling if their parents are unable or unwilling to pay school fees.

Statistics show that primary school enrollment and completion rates are low in many low-income countries, where underfinanced budgets, weak administrative capacity in the government bureaucracy, and lack of political will—indeed, political opposition to broadening access to educational opportunities—can deny children of poor parents an education. Such a situation may continue for some time if parents wait for the government to initiate change. In these compromised circumstances, user payments may have a role to play.

The effectiveness of user payments will depend on whether the reasons children are not in school are demand-side or supply-side. If children are unschooled because of a lack of demand resulting from their parents' poverty or information problems or a social bias against educating girls, imposing user payments cannot be expected to increase school enrollment—unless the revenues from user payments improve school quality sufficiently to overcome demand-side impediments.

If, on the other hand, the problem is on the supply side—that is, the government's inability or lack of will to provide publicly financed schooling—user payments are a way of addressing the unsatisfied demand for education. In such situations, in fact, user payments may be the only means by which parents in low-income countries can provide their children with an education until governments become willing or able to assume responsibility for providing all children with the publicly financed, free, quality education to which they are entitled.

Author Information

Eva Jenkner has worked on expenditure policy issues and fiscal sector reform at the IMF, where she is an Economist in the Western Hemisphere Department, and has contributed chapters on fiscal policy to IMF reports on Hungary, Mexico, and the Middle East. She has a B.A. from Cambridge University and a Master's degree in Public Affairs from Princeton University's Woodrow Wilson School of Public and International Affairs.

Marc Quintyn
Arye L. Hillman holds the William Gittes Chair at Israel's Bar-Ilan University, where he is Professor of Economics. In 1995, he shared the Max Planck Prize in Economics for contributions to political economy. His research focuses on economic policy decisions and the roles of governments and markets in promoting social and economic improvement. His recent book on this theme is Public Finance and Public Policy: Responsibilities and Limitations of Government (Cambridge University Press, 2003). He is an editor of the European Journal of Political Economy and has a Ph.D. in Economics from the University of Pennsylvania. Michael Taylor