Education Policy in Developing Countries

Paul Glewwe, Paul Glewwe

• Publisher: University of Chicago Press
• Year: 2013
• ISBN:
  9780226078854
• Topic: Pedagogía
• Subtopic: Educación general

Almost any economist will agree that education plays a key role in determining a country's economic growth and standard of living, but what we know about education policy in developing countries is remarkably incomplete and scattered over decades and across publications. Education Policy in Developing Countries rights this wrong, taking stock of twenty years of research to assess what we actually know—and what we still need to learn—about effective education policy in the places that need it the most.Surveying many aspects of education—from administrative structures to the availability of health care to parent and student incentives—the contributors synthesize an impressive diversity of data, paying special attention to the gross imbalances in educational achievement that still exist between developed and developing countries. They draw out clear implications for governmental policy at a variety of levels, conscious of economic realities such as budget constraints, and point to crucial areas where future research is needed. Offering a wealth of insights into one of the best investments a nation can make, Education Policy in Developing Countries is an essential contribution to this most urgent field.

1

Overview of Education Issues in Developing Countries

Paul Glewwe (Department of Applied Economics, University of Minnesota)

During this century, education, skills, and other knowledge have become crucial determinants of a person’s and a nation’s productivity. One can even call the twentieth century the Age of Human Capital in the sense that the primary determinant of a country’s standard of living is how well it succeeds in developing and utilizing the skills, knowledge, health, and habits of its population.
Gary Becker (1995)

Most, if not almost all, economists agree with Gary Becker on the importance of human capital in determining a country’s standard of living, and that formal education is a large, and perhaps the largest, component of human capital. This consensus reflects the fact that economists and other researchers have accumulated a vast amount of evidence that education increases workers’ productivity and thus increases their incomes. They have also shown that education leads to improvements in health and many other types of nonmonetary benefits.

While economists and other researchers may lament that many of their research findings are routinely ignored by policymakers, this does not appear to be the case for education. International organizations fully endorse the importance of education for economic and social development. For example, two of the eight Millennium Development Goals (MDGs) adopted at the United Nations Millennium Summit in September 2000 focus on education. Even more important, policymakers in developing countries also generally agree that there are important benefits from investment in human capital. One consequence of this consensus is that policymakers have greatly increased their funding of education; governments in developing countries now spend about $700 billion each year on education, and parents’ expenditures on their children’s education are likely to be on the same order of magnitude.¹

Have these investments in education increased the stock of human capital in developing countries? There has certainly been substantial progress in terms of increases in school enrollment rates and completion rates. For example, the World Bank (2012) estimates that 87% of children in developing countries finish primary school, and the gross enrollment rate for secondary school in these countries in 2010 was on average 64%, which is a large improvement over the rate of 41% in 1980.

Yet it is still the case that 13% of children in developing countries do not finish primary school, and over one-third do not enroll in secondary school. Even more worrisome is the large amount of evidence that students in developing countries learn far less than do students in developed countries. In an international comparison conducted in 2009, 58.1% of U.S. fifteen-year-old students attained a literacy score of Level 3 or higher, where Level 3 corresponds to “capable of reading tasks of moderate complexity” (OECD 2010, p. 51). In contrast, the corresponding figures for fifteen-year-old students in many developing countries were much lower: 23.3% for Brazil, 12.2% for Indonesia, 20.1% for Jordan, and 13.1% for Peru. Results for mathematics Level 3 proficiency, which is defined as being able to “execute clearly described procedures, including those that require sequential decisions,” reveal an even larger gap: 52.2% for the United States, yet only 11.9% for Brazil, 6.4% for Indonesia, 11.9% for Jordan, and 9.5% for Peru.

While spending even more money may increase enrollment and learning to some extent, most developing countries face serious budget constraints that will make it difficult for them to devote significantly larger amounts of money to education. This raises the question of whether current spending could be allocated more efficiently, and more generally whether education policies could be improved in ways that increase both enrollment and learning at little additional cost. Economists and other researchers have conducted a large amount of research on education in developing countries in the last two decades, but their findings are scattered in many different academic journals and other types of publications.

The importance of education in determining countries’ and individuals’ standards of living, combined with the low levels of learning in many developing countries despite the hundreds of billions of dollars devoted each year to education in those countries, underscores the urgent need to find policies that will lead to better education outcomes in those countries. Fortunately, there has been a large increase in research on education in developing countries in the last two decades, which presents an unprecedented opportunity to assess how this research can be used to improve education policies in developing countries. In response to this opportunity, this volume has three goals: to take stock of what this recent research has found; to present the implications of this research for education policies in developing countries; and, finally, to set priorities for future research on education in those countries. These goals are accomplished in the remaining chapters of this book. This chapter lays out the broad issues and highlights some of the most important findings in the chapters that follow.

Broadly speaking, the factors that determine how many years children are enrolled in school, and how much they learn while they are in school, can be divided into child and family characteristics, and school and teacher characteristics. Child and family characteristics are often difficult to change through government policies, though some policies, such as those aimed at improving child health, can have important effects. In contrast, since most children in developing countries are enrolled in publicly operated schools, government policies can have direct (and indirect) impacts on school and teacher characteristics. Thus a reasonable place to begin when reviewing the evidence on the impact of government policies on students’ education outcomes is to focus on school and teacher characteristics. Indeed, while students’ characteristics and backgrounds can have important effects on how much they learn, careful research has shown that schools and teachers can also make a substantial difference.²

Almost every parent would like his or her child to attend a “high-quality” school, but it is not necessarily clear which schools are of high quality. By definition, a “high-quality school” is a school whose students are more likely to achieve or exceed the learning goals set by the educational system, compared to similar students in other schools. Similarly, a “low-quality school” is one whose students are less likely to attain those goals, again compared to similar students in other schools. But what makes some schools (and teachers) “high quality” and others “low quality”? The role of basic school and teacher characteristics in determining students’ educational outcomes is reviewed in chapter 2, by Glewwe, Hanushek, Humpage and Ravina. This chapter systematically reviews the research done in the last two decades to assess what has been learned about the causal impact of basic school and teacher characteristics—such as class size (student-teacher ratio), teacher education, availability of textbooks and desks, teacher training, and pedagogical methods used—on students’ educational attainment (years of schooling) and learning. To the extent that certain teacher or school characteristics have strong positive impacts on students’ educational outcomes, funding for education should be reallocated toward policies that promote those characteristics and away from interventions that focus on characteristics that appear to have little or no effect.

The findings of this chapter are sobering. The studies that are deemed to be of the highest quality yield only a few unambiguous results, and those results are not particularly surprising. The clearest findings are that having a fully functioning school—one with better quality roofs, walls or floors, with desks, tables, and chairs, and with a school library—appears conducive to student learning. These findings are little more than common sense, and even the impacts of these attributes may not be causal; perhaps the true causal factor is an interest in, and a commitment to, providing a quality education. On the personnel side, the most consistent results are positive impacts of having teachers with greater knowledge of the subjects they teach, having a longer school day, and providing tutoring. An additional, and again unsurprising, finding is that it matters whether the teacher shows up for work; teacher absence has a clear negative effect on learning. Again, these findings regarding teachers offer little more than what common sense would predict.

One immediate implication of these research findings is that countries that are interested in improving student outcomes should ensure that these common-sense solutions to increase their students’ performance are in fact being implemented. Remarkably, many countries around the world fail to provide a basic institutional structure that promotes student achievement. Yet these findings do not imply that common sense can serve as a reliable guide for education policies, because many other policies that may also appear to be common sense, such as reducing class size, are not supported by recent research.

Perhaps most important, the analysis in chapter 2 suggests that progress in improving students’ educational outcomes in developing countries will require going beyond a narrow focus on basic school and teacher characteristics. Several possible directions could be pursued, and they can be divided into three broad types: policies that alter student characteristics before they begin primary school (and perhaps while they are in school), policies that are designed to alter student and parent behavior, and policies that attempt to change the way that schools are operated in terms of both the management structure and the incentives faced by teachers and school administrators. These three types of policies are systematically reviewed in chapters 3–7.

Consider first policies that attempt to change student characteristics before they even enroll in primary school. The two main avenues to change child characteristics in the first years of life are early childhood development programs, especially preschools, and child health and nutrition programs. The former are examined in chapter 3, and the latter are reviewed in chapter 4.

In developed countries, most children attend preschool before entering primary school, and preschools are becoming increasingly common in developing countries. The potential role that preschools could play in raising students’ progress in primary and secondary school in developing countries is examined by Behrman, Engle and Fernald in chapter 3. They find evidence from a small number of high-quality studies that preschools can have strong, positive effects on children’s long-run educational and income-earning outcomes.

Yet preschools can vary in many dimensions, and the evidence to date is insufficient to determine what aspects of preschools are most important for boosting student outcomes and outcomes during adulthood. In addition, there are many unanswered questions regarding how preschool services should be provided. For example, should developing country governments establish a nationwide system of preschools, or should they offer subsidies that families can use to enroll their children in privately operated preschools? Chapter 3 concludes with priorities for future research, including specific recommendations for the types of research methods and data-collection efforts that are most promising.

Any comprehensive analysis of education in developing countries must address the relationship between students’ educational progress and their health and nutritional status. The need for a comprehensive analysis of this relationship reflects two fundamental facts: (1) Many children in developing countries suffer from malnutrition and other conditions of poor health; and (2) there is strong evidence that malnutrition and poor health, both during the first years of life and while in school, can have negative impacts on students’ educational outcomes. This analysis is provided by Alderman and Bleakley in chapter 4.

After a conceptual overview of the impact of poor health on education outcomes in developing countries, Alderman and Bleakley focus on two specific dimensions of poor health: malnutrition and parasitic infections. They highlight the findings that improvements in these two dimensions have the potential not only to increase years of schooling but also to increase learning per year of school, the latter of which is consistent with an emphasis on improving the quality of education. Interventions that reduce early childhood malnutrition and parasitic infections have economic returns that greatly exceed their costs. Alderman and Bleakley also point out that both of these dimensions are characterized by the presence of externalities; diarrheal infections that lead to early childhood malnutrition and parasitic infections that affect children during their school-going years are easily spread from one child to another, which implies that private investments in child health are below socially optimal levels. In addition, public interventions to improve child health lead not only to increased economic efficiency by addressing the above-mentioned externalities but also reduce inequality, since efforts to improve child health tend to benefit the poor more than they benefit higher-income households.

A final point regarding the impact of child health on schooling outcomes is that some successful interventions will require cooperation among two separate government ministries, the Ministry of Health and the Ministry of Education, that have had relatively little interaction in the past in many countries. This could lead to administrative conflicts; for example, some activities of the Ministry of Education that are not particularly effective in improving students’ outcomes should have their funding reduced, and those funds may better contribute to education outcomes if they are used to fund health programs administered by the Ministry of Health.³ This is consistent with the more general point made above that parts of the institutional structure in many developing countries impede more effective policies and thus inhibit better results.

Policies that attempt to improve students’ education outcomes by changing teacher and school characteristics are, in effect, policies that focus on the supply side of schooling. Yet there may be policies that are more effective by focusing on the demand side. This leads to the second type of policies that go beyond attempts to change school and teacher characteristics: policies that change the incentives faced by students and parents. The potential for such policies is explored by Behrman, Parker, and Todd in chapter 5. This chapter is particularly timely because such incentive programs, especially conditional cash-transfer programs, have become much more common in developing countries in the past one to two decades (see World Bank 2009).

Programs that provide incentives for students and their parents can take several forms. Chapter 5 reviews the evidence on four types of incentive programs: (1) Conditional cash-transfer programs, which provide parents monthly payments conditional on their children attending school regularly; (2) payments to students based on academic performance, such as scores on exams; (3) school-voucher programs that provide funds that parents can use to enroll their children in either public or private schools; and (4) “food-for-education” programs that provide children with meals at school or supply their families with staple foods to be consumed at home. The authors find that many of these programs lead to increases in school enrollment, although there is less evidence on whether they lead to increased student learning as measured by performance on academic tests.

While incentive programs seem to be a promising avenue for increasing students’ educational outcomes, further research is needed to understand the circumstances that make these programs particularly effective. Given the variation in country and education system characteristics across developing countries, it is not clear that a program that worked well in one country will also be effective in another country with very different characteristics. Another important issue is cost-benefit ratios. Some incentive programs, such as conditional cash-transfer programs, are quite expensive, and there may be other policies that can raise students’ educational progress at a much lower cost. On the other hand, when assessing the costs of those programs it is important to note that, from the perspective of society as a whole, transfers are simply a redistribution of resources from one group of households (taxpayers) to another (program participants) and thus these transfers should not be counted as a cost of the program to society as a whole (although raising taxes to pay for any program does entail a social cost, namely the deadweight cost of raising government revenue).

The third and final type of policies that go beyond attempts to change basic school and teacher characteristics is those that focus on the ways that schools, and more generally school systems, are organized. Such policies focus on the supply side of the education sector, but also on how that supply can be made more responsive to the demand for education. These types of policies can be divided into those that focus on how public schools are managed, and those that foster competition between (and among) both public and private schools. These two types of policies are examined in chapters 6 and 7 respectively.

Many observers argue that the focus of supply-side efforts pertaining to public schools should not be on their basic characteristics; instead it may be...