30108 USER FEES IN PRIMARY EDUCATION THE WORLD BANK July 2004 Raja Bentaouet Kattan Nicholas Burnett EDUCATION SECTOR HUMAN DEVELOPMENT NETWORK WORLD BANK USER FEES IN PRIMARY EDUCATION CONTENTS Foreword ......................................................................................................................................... 3 Executive Summary ........................................................................................................................ 4 PART 1: THE GROWING POLICY SALIENCE OF FEES......................................................... 6 PART 2: FEES IN DEVELOPING COUNTRIES......................................................................... 9 PART 3: THE WORLD BANK AND PRIMARY EDUCATION FEES.................................... 20 PART 4: RECOMMENDATIONS............................................................................................... 24 References..................................................................................................................................... 68 Boxes: Box 1: Community Schools in Africa ........................................................................................ 11 Box 2: Household Education Expenditures in Africa ................................................................ 15 Box 3: Income and Enrollment .................................................................................................. 17 Box 4: Mexico OPORTUNIDADES ......................................................................................... 18 Box 5: Targeted Education Scholarships in The Gambia .......................................................... 22 Tables: Table 1: Summary Fee Survey Results......................................................................................... 10 Table 2: Household Expenditure on Public Primary Education and Breakdown of Household Payments......................................................................................................................... 14 Table 3: Household Expenditures on Primary Education as a Percentage of Household Spending ........................................................................................................................................ 15 Table 4: Kyrgyz Republic: Annual Household Spending on Education by Poverty Group......... 15 Table 5: Fee Mitigating Measures ................................................................................................ 20 Annexes: Annex 1—World Bank Current Statement on User Fees ............................................................. 28 Annex 2—User Fees and Primary Education ............................................................................... 31 Annex 3—User Fee Survey .......................................................................................................... 38 Annex 4—Summary of Survey Results........................................................................................ 39 Annex 5—The Case of Malawi: What Can Go Wrong with the Elimination of User Fees? ....... 54 Annex 6—The Case of Uganda: What Needs to be in Place to Support Elimination of User Fees? ........................................................................................................................................ 58 Annex 7—The Case of Kenya ...................................................................................................... 61 Annex 8—The Case of Tanzania.................................................................................................. 64 Annex 9—China- The Mounting Expenses of Education ............................................................ 65 Annex 10—HIPC Figure on Public Spending .............................................................................. 67 2 Foreword User fees in primary education are pervasive and a serious obstacle to enrollment and completion for millions of children around the world. This has been evidenced by the huge surges in enrollment that have accompanied fee removal in countries such as Uganda and East Timor, where surges have at times threatened to overwhelm these countries’ educational systems. The elimination of fees, which the World Bank supports, therefore requires care, if countries’ educational systems are not to be overwhelmed and if quality is not to suffer. Fees often ensure the provision of quality inputs, such as books, for those who do enroll. Opportunities to eliminate inefficiencies in educational spending must be grasped if expenditure is not to grow out of hand. External assistance is also available, notably in the form of HIPC debt-relief funds and the Fast Track Initiative Catalytic Fund, as well as bilateral funding, to help countries handle the elimination of fees. Different countries have managed the elimination of fees in different ways. Yet too many countries still maintain fees. This paper was originally drafted some two years ago by Raja Bentaouet-Kattan and Nicholas Burnett to help the World Bank understand the extent of the user fee issue in primary education. We are publishing it now in order to make its findings more widely available, especially the survey information that was collected specifically for the paper. At the same time, we are actively discussing the World Bank’s own policy stance which we expect to substantially modify along the broad lines that the paper suggests. Ruth Kagia 3 Executive Summary Fees, in the sense of direct household payments, are common in primary education around the world, and represent perhaps 20 percent of all education spending, and as much as 30 percent in Africa and even 40 percent in the former Commonwealth of Independent States (CIS) and East Asian socialist countries. Tuition fees are less common than fees for books, PTA and community contributions and compulsory uniforms. Fees represent a significant proportion of household spending, although this proportion varies widely from country to country. It is universally the case that the proportion is highest for the poorest households. Too little is known about the determinants of enrollment and learning, but fees are certainly obstacles in many, but not all, cases. Universal primary completion is a top World Bank priority, expressed in the Bank’s commitment to the Millennium Development Goals. The Bank has made abundantly clear in its policy statements that it does not support user fees for tuition in primary education and has in recent years actively supported fee abolition in countries, mainly in Africa, in which fees appear to represent an obstacle to enrollment. Important lessons have been learned from these fee abolition efforts, of which the most important is that, in cases where fees are an effective contribution to school access or quality, there is a need to ensure replacement of fees with revenues of equal effectiveness and size, if fees are reduced or abolished. There has not, however, been any systematic analysis of the determinants of enrollment in Bank client countries to determine whether fees represent an obstacle to access or completion. Based upon available information, there are a number of low enrollment countries in which fees might be impeding enrollments, as well as relatively high-fee countries where fees may not be impeding enrollment or completion (but may be impeding effective poverty reduction). While the Bank has not supported tuition fees, it has supported various other types of fees in primary education, including textbook fees, along with programs to offset or waive the fees for the poor and other target groups such as girls. Some of this past support is now under reconsideration. It is likely, for instance, that the Africa region of the Bank will soon move formally to support the free provision of textbooks. Very little systematic information is available about the effectiveness of fee waiver schemes in primary education. There are a large number of different “fees” that private households sometimes have to pay for publicly provided primary education, including tuition fees, textbook fees or costs and/or rental payments, compulsory uniforms, PTA dues, and various special fees such as exam fees, contributions to district education boards, and the like. In many countries, private tutoring, although not strictly compulsory, often adds to the household costs of primary education. This paper addresses all these fees and distinguishes between fees that are charged both formally and informally. • Indirect costs can be an even greater obstacle to school attendance than fees. • In many situations, fees cannot simply be abolished. Where fees make a contribution to the quality of schooling, simple abolition of fees could have undesired consequences. Adequate measures must therefore be in place to provide equivalent revenues to finance 4 the expenditure previously covered through fee revenue. There are four principal sources of such replacement revenues: o Countries may increase expenditure on education by switching spending from other sectors or by increasing revenues. o Improving the efficiency of education spending, particularly the balance between different educational levels and the balance between salaries and other expenditures. Expenditure shifts do take time, however, and must be well prepared politically, sometimes being as difficult to implement as the introduction or augmentation of fees. o Using HIPC debt relief funds on a temporary basis to close the financing gap in many heavily indebted countries over the next few years. o Using funds from the FTI Catalytic Fund on a temporary basis represents an extraordinary opportunity to provide temporary financing to FTI countries working towards the elimination of fees. Finally, based on the review, the paper recommends that the World Bank should: 1. Develop an explicit policy on user fees based on the premise that no child should be excluded from primary education on the basis of inability to pay and work actively with governments to find alternatives to existing user fee programs, focusing mainly on the replacement of lost revenues from fees. 2. Develop a clearer and shorter public policy statement on user fees in primary education. 3. Review the determinants of enrollment in all countries that are unlikely to achieve the education Millennium Development Goals. 4. Examine the effectiveness and impact of textbook charging schemes it has supported and clarify its position. 5. Analyze the effectiveness of targeted fee waiver schemes in primary education. 5 User Fees in Primary Education User fees in primary education have taken on increased salience in recent years. This has prompted the World Bank to consider further the position it should adopt toward fees in public primary education among its client countries. This paper reviews the issues and proposes actions by the World Bank. Part 1 discusses the definition of fees, considers why fees have become so salient a policy issue, and reviews the positions of other donors. Part 2 summarizes the situation in Bank client countries, in terms of the prevalence and utilization of fees, their share in household spending, their impact on enrollment and learning, and measures used to mitigate some of their negative effects. Part 3 presents the World Bank’s current position on fees and reviews the treatment of primary education fees in its country programs. Part 4 summarizes briefly and makes recommendations for the Bank. PART 1: THE GROWING POLICY SALIENCE OF FEES Definitions. General discussion of user fees is often explicitly or implicitly about tuition fees. In practice, however, there are a large number of different “fees” that private households sometimes have to pay for publicly provided primary education, including tuition fees, textbook fees or costs and/or rental payments, compulsory uniforms, Parent Teacher Association (PTA) dues, and various special fees such as exam fees, community contributions to district education boards, and the like. This paper refers to all these fees. We exclude from consideration, however, fees that are not to do with education but rather with student welfare (such as board and lodging charges at boarding schools, and charges for school meals). There can be three additional economic costs to primary education. First, households often incur transport costs in sending their children to school. In Cambodia, for instance, these costs account for some 21 percent of household spending on education. Second, in many countries, especially but not only in Asia, it is customary to send children not only to public school but also to private tutors (often the very public school teachers) outside school hours.1 Though not strictly compulsory, such fees often add to the household costs of primary education, either because parents feel that they must provide such tutoring to supplement formal schooling or because there is an “element of blackmail involved because at least some teachers provide only the bare minimum during school hours and reserve the real teaching for their private classes.”2 Third, poorer households often incur indirect economic costs by sending to school children who would otherwise contribute to the household economy, by working for income, working in farming or through such tasks as collecting water and firewood or looking after younger children. These indirect costs are often more critical than direct fees in keeping children out of school and must be addressed in any policy work to increase enrollments. 1 See, for instance, Mark Bray, Counting the Full Cost: Parental and Community Financing of Education in East Asia, World Bank, 1996, pp. 14-17. See also Mark Bray, The Shadow Education System: Private Tutoring and is Implications for Planners, UNESCO: IIEP, 1999; and Mark Bray, “Government and Household Financing of Education: Finding Appropriate Balances”, paper presented to the International Conference on Economics of Education, Beijing University, 16-19 May 2001. 2 Bray (1996), p. 17. 6 Increased Salience. User fees have become increasingly salient in recent years as a policy issue in efforts to achieve universal primary education for four principal reasons: • Progress toward achieving universal primary education and the focus on enrolling the remaining out-of-school population, who tend to be poor; • Growing inequality and, in some regions such as Africa, declining incomes; • Increased attention to the Millennium Development Goals; and • Dramatic jumps in enrollment in countries that have abolished fees. Following the Jomtien World Conference on Education For All in 1990, many countries have made considerable progress toward achieving universal primary education. Between 1990- 97, for instance, the primary gross enrollment rate increased by 27% in Bangladesh, 65% in Malawi and 18% in Morocco.3 Countries’ very success in increasing enrollments, however, has led to increased attention to two key aspects of primary education: (a) enrolling the remaining 10-20 percent of the school age population, who tend to be the poorest children; and (b) ensuring that those in school benefit by learning. User fees in the form of tuition payments, compulsory uniforms and other payments, are, in some countries, one factor keeping the poorest children out of school and making it hard for them to stay in school for the 5-6 years that are essential to achieving functional literacy. Fees in the form of charges for textbooks can also impede learning even for those who can stay in school. It is important to stress, however, that fees are not necessarily the only factor impeding school enrollment and retention or the availability of books to students. Moreover, in some countries, fees are not a factor at all in preventing access, as signaled by the fact that the enrollment rate is at 100% even in the presence of fees, though they may still have a regressive impact. The greater is inequality, the harder it is for the poor to pay fees. Since 1990, poverty rates have declined in the developing world, particularly in East Asia and notably in China.4 Except in East Asia, however, population growth has meant that the number of poor people has increased even though the proportion of people in poverty has declined. Worse, in the countries of the former Soviet Union, the proportion of poor people has more than tripled. Despite some recent growth, the economies of Sub-Saharan Africa have declined since 1990 and the region continues to have the highest proportion of poor people. In many parts of the world, however, and particularly in China, India and the former Soviet Union, inequality in household consumption has increased; these countries join those of Africa where inequality has changed little and remained very high. Through general endorsement of the Millennium Development Goals (MDGs), the international community has recently reaffirmed its widespread commitment and support for universalizing access to basic education. Two of the seven MDGs focus specifically on education and directly support the education targets initially established at Jomtien and reaffirmed at the World Education Forum held in Dakar, Senegal, in April 2000. The MDGs aim at ensuring that all children are enrolled in primary education by 2015, and that gender disparities in primary and secondary education are eliminated by 2005. Providing universal basic education has proven to 3 Development Committee, Education for Dynamic Economies: Accelerating Progress Towards Education for All, Document DC2001-0025, September 18, 2001. 4 This paragraph is based on World Bank, World Development Indicators 2001. 7 be considerably more challenging than was initially anticipated. Despite general progress, primary or basic school enrollment rates have actually declined over the past 5-10 years in some 27 countries. The challenge of enrolling all children in primary school by 2015 remains daunting; indeed, it seems likely that around 88 countries will not meet the goal of universal primary completion by 2015 unless a major and sustained focus is given to education. These countries are concentrated in three regions: Sub-Saharan Africa, South Asia and the Middle East and North Africa.5 In addition, gross enrollment rates have declined in the Europe and Central Asia region, from 94 percent in basic education in 1989 to 91 percent in 1997. Four prominent African countries have removed fees at the primary level with dramatic results, suggesting that the payment of fees is a major obstacle to enrollment in Africa. Malawi dissolved fees in 1994 and as a result, enrollments increased approximately 51 percent; Uganda abolished fees in 1996 and experienced approximately 70 percent increase in enrollments. In 2001, Tanzania abolished fees, which resulted in a rapid increase in the net primary enrollment rate from 57 percent to 85 percent by 2002. In the most recent case of Kenya, the elimination of fees resulted in 1.2 million additional students entering the school system. In all cases, increased enrollments have resulted in concerns for deteriorating quality and increased demand for secondary education. The situation in Malawi was particularly acute; the increase in enrollments resulted in the following figures: pupils:permanent classroom, 119:1; pupils:desk, 38:1; pupils:chair, 48:1; pupils:textbook, 24:1; and pupils:teacher, 62:1. Tanzania’s planned elimination of user fees resulted in a less severe decline in quality due to a strong government commitment to UPE, HIPC funding to help close the financing gap, and grants to fund quality inputs. These included, for example, a US $10 capitation grant per pupil to meet non-salary recurrent costs and a development grant budget of US$3500 per classroom for capital expenditures, including teacher houses, sanitation and the provision of clean water. Donor Positions. Several aid agencies have adopted positions as a result of the increased salience of the issue of user fees in primary education. The British Department for International Development (DFID), for instance, generally supports the position that general taxation and other forms of government revenue are more effective, efficient and equitable methods of financing basic social services than are cost sharing mechanisms. In addition, “Where costs have to be incurred by parents and guardians there is scope for measures which lessen the burden on poor people. These include reducing and/or staggering the direct costs of education, non- discretionary exemption schemes and flexibility in the provision of schooling to mitigate seasonal demands on child labor and the requirements of the daily household economy.”6 DFID’s position is essentially an endorsement of the 1997 Addis Ababa Consensus on principles of cost sharing in education and health, adopted by ministers and senior government officials from 17 Sub-Saharan countries, at a conference attended also by NGOs, bilateral donors and multilateral agencies.7 UNICEF has organized a campaign to eliminate all primary education user fees and costs in Africa. USAID’s position is that it will not support the introduction of fees at primary level. Where fees are already in place, USAID believes that: “to the extent possible, public funding for 5 This paragraph draws on Development Committee, ibid. 6 DFID, Learning Opportunities for All: A Policy Framework for Education, 1999. 7 UNICEF, Addis Ababa Consensus on Principles of Cost Sharing in Education and Health, June 1998. 8 basic education should replace user fees– including tuition as well as school uniforms, textbooks, and educational materials. Where user fees for basic education are in place, governments should be encouraged to take deliberate steps to replace them with adequate public funding, by shifting current public spending away from lower-priority uses, developing appropriate new revenue sources, or a combination of both. In the meantime, government should not be encouraged simply to drop school fees before securing adequate sources of public funding to replace them.”8 The Asian Development Bank’s new education policy paper (2002) makes it clear that ADB does not favor user fees for primary education. In fact it believes that basic education (ages 6-14 or grades 1-9) is a human right and that it should be free of charge. It also recognizes that fees other than "official" tuition fees may be charged (books, uniforms, etc.) and that associated costs of schooling (transportation, lunch, etc.) can prevent poor children from remaining in school even if they enroll. The policy paper states that special efforts should be made to support poor children in meeting these costs, including targeted subsidies, scholarships, and special interventions.9 The African Development Bank is studying the issue of user fees but has not as yet taken an official position. PART 2: FEES IN DEVELOPING COUNTRIES It is difficult to obtain a complete picture of fees in primary education, and particularly difficult to develop a quantitative estimate. User fees are often not reported in official school censuses or financial returns from education systems. Not all household surveys ask about spending on education, and those that do rarely distinguish between levels of education, between clothing and uniforms, or between formal schooling and supplementary private tutoring. Official statistics do not usually record the number of students receiving private tutoring or the costs of this tutoring. Very few countries compile data on the contribution of fees more generally to the public sector, even when the fees are commonplace, often because these fees may be formally unconstitutional, as in the Middle East and North Africa, or because they may be technically unofficial, as in most of the CIS countries. In order to overcome this relative lack of published information, a survey was conducted of Bank education task team leaders, enquiring about the prevalence of fees in primary education in the countries for which they hold responsibility (and also about the positions the Bank has adopted in its dialogue with these countries).10 Annex 2 presents the country-by-country results, Annex 3 is an example of the questionnaire sent to task team leaders, and Annex 4 presents the overall results in summary fashion. Data were collected by this means for 79 Bank client countries, out of a total universe of some 145 countries.11 Five fee categories are distinguished 8 Email dated April 11, 2001 from Don Sillers (USAID) to Tom Molony (ActionAid). 9 The ADB policy paper may be found at www.adb.org/Development/policies. 10 A preliminary survey was conducted in October 2000 and a full survey in October-November 2001. Where no response was received in 2001, 2000 data were used if available. 11 Survey responses were relatively comprehensive for East Asia and the Pacific, Latin America and the Caribbean, the Middle East and North Africa and South Asia. With 27 responses, however, only slightly over half the countries in Africa are included. In addition, there are almost no responses from Europe and Central Asia. Where possible, we have supplemented missing survey data with information from other sources, including meetings with groups of staff members from the various regions. 9 in the data: tuition, textbook charges, compulsory uniforms, PTA/community contributions, and school-based activity fees (exam fees, etc.). The data refer to public schools. Excluded from the data are countries in which most children attend fee-paying private schools, such as Haiti and Somalia. These are situations that reflect the absence of public schools, rather than any parental choice to avoid them, and should be borne in mind when considering the possible role of fees in restricting access and limiting learning. Appropriate public policy in such situations is not in the realm of fees at public schools, however. Prevalence of Fees. Fees are very common, being found in 97 percent of the countries surveyed; they are also unofficial in about a third of the cases in which they exist (Table 1). Of the 79 countries surveyed, only Algeria and Uruguay do not have fees of any type. Most countries (69 out of 77) have several different fees; the eight countries with only one fee have only PTA or community contributions, the most prevalent type of fee. These PTA dues still exist, interestingly, even in the widely publicized cases of Malawi and Uganda that abolished tuition fees but not all fees. Most of the 69 countries with several fees have two or three types of fees. Seven countries have all five types and another four have all types except tuition. Textbook charges and compulsory uniforms characterize about half the countries. Tuition fees are collected in almost 40 percent of countries.12 Table 1: Summary Fee Survey Results Percentage of Fee- Countries Percentage of Countries with charging Countries with Fees Countries in Survey Unofficial Fees with Unofficial Fees At least one fee 77 97 Tuition fees 30 38 11 37 Textbook charges 37 47 12 32 Compulsory uniforms 39 49 0 0 PTA/community 56 71 22 39 contributions Activity Fees 34 43 13 38 Source: Annex 4. The type of fee charged varies considerably from region to region.13 In Africa, community contributions and PTA dues are the most common type of fee (81% of countries surveyed). Other fees are less common but nonetheless significant: tuition (41%), textbooks (37%), uniforms (48%) and other activity fees (41%). Of increasing significance in low enrollment countries in Africa are also community schools, financed by parents when the state is unable to provide public education, as in several West African countries, notably Chad and Mali. These community schools are not exactly public and not exactly private. While financed mainly by parents (not necessarily through identical fees for all families), they may also receive some assistance from the public sector, such as the provision of materials or even a teacher (Box 1). 12 Data in this paragraph are drawn from Annex 4 if not presented in Table 1. 13 These two paragraphs are derived from Annexes 2 and 4. 10 Box 1: Community Schools in Africa Nonprofit community schools have increasingly become critical vehicles in some Central and West African countries for parents to meet their aspirations to educate their children when the state has made inadequate or no provision for primary schooling in a particular area. Community schools were often considered illegal in the past, though they are now generally recognized, with names such as “écoles clandestines.” More than 35 percent of primary school children in Togo attend such schools, with even higher proportions in Chad, where local communities have played a significant direct role in financing and operating schools since the education system deteriorated due to the 1979-82 war and continuing economic weakness. Community schools charge fees, sometimes the same fee for each child and sometimes differential fees according to family income circumstances. Payment of these fees can play an important role in building local ownership of school-related activities. The experiences of Chad, Togo and Mali, however, illustrate that community-based approaches are hard to sustain in times of economic downturn. They also raise serious equity questions, i.e. is it fair to ask parents in some of the poorest communities in the country to pay for schooling when their much better-off urban counterparts have it provided at no charge. Many community schools offer incomplete primary cycles, having found it difficult to finance teachers in sufficient numbers to cover all of the primary grades. The World Bank is working with governments to find ways to ensure that community schools receive needed support and that the burden on parents is alleviated. A recent example is Mali, where community schools were most common in cotton-growing parts of the country and grew over 8 years to enroll over 200,000 students, about a quarter of total enrollments in the country. When the price of cotton declined, however, many communities were unable to pay teacher salaries and many classrooms were closed. With World Bank support, HIPC funds were channeled in part to payment of community school teacher salaries, beginning in 2001. Most community school teachers in the country are now being paid through this mechanism, relieving a real financial burden that parents were supporting and once again allowing the expansion of the school system. Community schools are now officially recognized and supported in Mali, and have full access to the range of services provided to public schools, including provision of free textbooks, in-service teacher education, and pedagogic support services. Source: World Bank, Community Support for Basic Education in Sub-Saharan Africa, 2001. In the transitional economies of both the Commonwealth of Independent States (CIS) and East Asia, tuition fees are common. Educational systems in Europe and Central Asia (ECA) have responded to the decline in real expenditures and rising input costs that have characterized their economies’ transitions by shifting part of the costs formerly paid by the public sector on households. Worse, these fees are often collected unofficially. The fees can represent a very high proportion of education expenditures. The picture is much more varied among the former socialist countries in terms of fees other than tuition – fees for books are common in all such countries while compulsory uniforms are much more common in East Asia (all the former socialist countries) than in the CIS (in which Azerbaijan is the only country in the survey that apparently mandates uniforms). PTA and community contributions are required in almost all these countries – 80% of those in Europe and Central Asia (ECA) and all the former socialist East Asian countries. 11 There are essentially no tuition fees in Latin America. The survey reports that only Colombia has fees but Ecuador, although not included in the survey, also has tuition fees. Textbook fees are also relatively rare (29% of countries surveyed), but compulsory uniforms (76%) and PTA or community contributions (65%) are the norm. The Middle East and North Africa region is characterized by a contrast, in most countries, between the constitutional right to a free education (often at all levels) and the reality of fees. Tuition fees are collected in Egypt, Jordan, Tunisia and Yemen, though they are technically unofficial in Egypt. Textbook fees are charged in half the countries, as are PTA and community contributions. Uniforms are relatively rare (only in Egypt and Lebanon) as are other school-based fees. Fees are the norm in South Asia, of which private tutoring characterizes many of the fees. Schools in both India and Pakistan charge tuition fees for primary education, as do Bhutan and Nepal. Schools in Sri Lanka and Bangladesh do not charge tuition. Over half the countries also have textbook fees and PTA and community contributions, while uniforms are not generally required except in Pakistan. Given that fees are normal in much of South Asia, the most striking trend in recent years has been parents transferring their children from public schools to private ones, especially in urban India and urban Pakistan. At the private schools, the fees are about the same as at the public ones and the parents believe that quality is higher. In East Asia and the Pacific (other than the former socialist countries), private tutoring is pervasive, as are textbook fees (70% of countries), uniforms (80%), PTA and community contributions (80%) and other activity fees (70%). Tuition fees are less common but do exist in Indonesia, Papua New Guinea and the Solomon Islands, China and Vietnam. Utilization of Fees. Fees can account for a very high proportion of total educational expenditure on primary education, i.e. of combined public and private spending. The proportion ranges between 8 percent in Indonesia and 80 percent in Cambodia for 12 countries for which we have data (Table 2).14 The highest proportion of household spending on educational fees is generally for textbooks and uniforms, rather than for tuition, although data is insufficient for most countries. These estimates are consistent with those made by others. Experts working on Africa often use 30 percent as a rough rule of thumb for the household share of primary education spending, consistent with the three African countries included in Table 2. In East Asia, private spending accounts for about 20 percent of total spending in general and over 50 percent in both Cambodia and Vietnam.15 In addition, while data are unavailable, most observers of the primary education systems of the former Soviet Union consider that private payments are holding the system together, and may account for as much as 50 percent of all expenditure, especially in countries like Georgia in which public spending on education is as low as 1% of GDP. The proportions will likely be lower in Latin America and the Middle East, but comparable in South 14 Note, however, that the data in Table 2 cover also household costs for private tutoring and for transport, which we do not include in our strict definition of fees – they are shown in the table in order to remind of their existence and to indicate that fees are only one of the costs of school attendance for the household. 15 Bray (1996), p. 32. Bray’s estimates for the four countries shown in our table are very similar. In addition, he estimates the proportions for China at about 20%, Lao PDR at about 25%, Mongolia at about 15%, Myanmar at about 25%, and Thailand at about 15%. 12 Asia to those in Africa and East Asia. It is important to note that all these estimates are just for private spending on education at public schools – the proportions would increase significantly if private spending on private schools were also included. Globally, therefore, it is reasonable to estimate that household expenditures account for about 20 percent of total spending on public primary education, with higher proportions in Africa (30 percent) and in former socialist countries (approximately 40 percent). These are high proportions which imply that elimination or reduction of fees must be approached with appropriate attention to the impact on government revenues and expenditure. An important aspect of fees is where and how they are determined, collected and held. In most countries with fees, the fees are collected at the school level, especially textbook fees (generally, though the fees from the sale of books are normally sent elsewhere in the education system), PTA and community contributions (always), and activity fees (usually). Uniforms tend to be mandated nationally. Tuition fees offer a more mixed picture. In China and Egypt for instance, they are collected at the school but not used by the school; however, in most other cases, they are collected and used by the school. These factors make a potentially major contribution to how effective fees are as a source of revenue and therefore, have an impact on how carefully one needs to consider some form of substitution, were fees to be eliminated. Where and how these fees are collected and used, conditions the type of substitutes one could use. For example, simply providing schools with some of the inputs they might have purchased with their own free income, might not always be perceived by schools to be as efficient as purchasing their own inputs with their fee income. 13 Table 2: Household Expenditure on Public Primary Education and Breakdown of Household Payments Country Household Percent Shares Expenditure as percentage of Textbooks Total Public and and Private School Tuition Uniforms Transport Other Private Spending Learning Tutoring Contributions on Primary materials Education Cambodia 80 15 29 32 3 21 China 21 Ghana 30 India* 43 13 76 8 2 1 Indonesia 8 11 16 21 1 4 47 Mauritania 32 Nepal 16 Philippines 30 Tanzania 37 Thailand 14 37 38 20 5 Vietnam 44 4 29 15 13 39 Zambia 50 to 75 Sources: Mark Bray, Counting the Full Cost: Parental and Community Financing of Education in East Asia, World Bank, 1996. Mark Bray, The Shadow Education System : Private Tutoring and Its Implication for Planners UNESCO,1999. Mark Bray, “Government and Household Financing of Education: Finding Appropriate Balances”, paper presented to the International Conference on Economics of Education, Beijing University, 16-19 May 2001; Oxfam Briefing Paper “Education Charges – A tax on human development” (2001); World Bank: Nepal Public Expenditure Review (2000) * Average number, actual numbers varying from state to state. Impact of Fees: Household Spending. The burden on households as a result of fee payments is significant (Table 3). In Zambia, for instance, it amounts to some 16 percent of spending on items other than food, with similar numbers in Ghana (Box 2). The greater the burden of education spending, the poorer the family (Table 3). In the Kyrgyz Republic, for instance, extremely poor households spend 17 percent of their consumption on education while non-poor households spend 13 percent (Table 4). 14 Table 3: Household Expenditures on Primary Education as a Percentage of Household Spending All Households Poorest Quintile Europe and Central Asia Bulgaria 24 36 Macedonia 13 17 East Asia China 19 29 Indonesia 17 17 Mongolia 6 8 Thailand 16 47 Vietnam 12 22 South Asia Nepal 16 29 Source: World Bank survey; Mark Bray, Counting the Full Cost: Parental and Community Financing of Education in East Asia, World Bank, 1996. Table 4: Kyrgyz Republic: Annual Household Spending on Education by Poverty Group Type of Expenditure Category of Household Extremely Poor Poor Non-Poor National Tuition 21 53 375 204 Books, uniforms, fees, tutors 193 228 356 288 School repair, classroom supplies 36 43 64 53 Meals, transport, other 43 82 350 207 Total 292 406 1145 752 Education as share of annual consumption (%) 16.9 14.1 13.1 13.6 Source: World Bank, Update on Poverty in the Kyrygyz Republic, 1999. Box 2: Household Education Expenditures in Africa • In Ethiopia, a 1995 survey found that annual per capita education expenditure accounted for 4.5 percent of total per capita expenditure. The average annual cost for a rural household of sending one child to primary school was 58 birr, roughly equivalent to US$9. Fees accounted for half of this expenditure. • In Ghana, recent Living Standards Survey data show that mean household cash expenditure on education accounts for 18 percent of non-food expenditures in Accra, and 12 percent in rural areas. Per student annual average expenditure on primary education is 122,000 cedis, equivalent to 7 percent of mean annual household expenditure for the poorest quintile. Fees and PTA contributions account for almost one-third of this sum. • In Nigeria, extensive household questionnaires in 1997 revealed that annual household costs per primary student ranged from 1000-2400 naira. N100-150 is the daily rate of rural consumption. • In Tanzania, 1995 data showed that an average 5 percent of household non-food expenditure is allocated to education. In the case of primary education, households met 37 percent of all recurrent costs, public and private. • In Zambia, recent survey data shows that an average of 16 percent of household non-food expenditures are allocated to education. Source: World Bank, Community Support for Basic Education in Sub-Saharan Africa, 2001. 15 Impact of Fees: Enrollment. It is important not to assume that less-than-universal enrollment is automatically the result of fees, when fees exist. Low enrollment may also reflect a lack of supply (no schools in the area), the opportunity costs of attending school, the perceived low returns from schooling in the labor market, or other standard factors that limit access, such as the distance to school and, for girls, the existence of female teachers, and separate toilets, for example. There is a serious absence of current literature examining these different possible causes of low enrollment among the countries that have not achieved UPE. In part, this literature gap reflects the considerable progress made in increasing primary enrollments around the world, which make it a less pressing global issue, even if it remains of high priority for Africa and South Asia. In part, it also reflects the lack of usable data in many of the poorest countries that have low primary enrollment rates. On the other hand, the fact that a country has reached UPE even though schools charge fees, does not mean fees cease to be a cause of concern. Though in such a case fees may not limit access or attendance, they may still be an equity concern. Among major determinants of enrollment identified in recent studies are household income, schooling cost, presence or absence of schools, community involvement, transportation, education quality and relevance. For girls, additional factors that contribute to enrollment are the presence of female teachers, the availability of proper features such as latrines, daycare facilities for younger siblings and water points, and the flexibility of the school calendar. Not all studies use the same list of possible determinants, of course, and methodologies differ. Fees can be similar to prices in their effect on individual behavior, and thus influence enrollment through both a price and an income effect. Studies of the income effect are summarized in Box 3. 16 Box 3: Income and Enrollment Household income is found to be a significant correlate of school enrollment in many countries including India, Mexico, Senegal and Zambia. For example, in Senegal the enrollment of 6 to 14 year olds from the poorest households is 52 percentage points lower than for those from the richest households. In Zambia there is a 36 percentage point difference in the enrollment rate of children from the richest and poorest households. In India, children from the wealthiest households are over 30 percentage points more likely to be in school than those from the poorest households, although this gap varies considerably across states. In a further six countries, school costs are significant in explaining enrollment (Bangladesh- for girls, Egypt, Ghana, Indonesia, Mexico, China and Pakistan). For example, in Egypt, lack of access to schools does not appear to be a concern to the vast majority of children, however, the high private cost of schooling has been cited as a major barrier to entering and staying in school. In Ghana, more than 50% among the poor claimed that they never attended school because of financial reasons. In only one country are school costs found to be insignificant (Malaysia) in explaining enrollment. Female disadvantage in enrollment can be found in many countries, primarily those in Western and Central Africa, North Africa and South Asia. Some of these countries have large gender gaps in enrollment. For example, in India, there is a 16.6 percent percentage point difference between the school enrollment of girls and boys aged 6 to 14. In Benin, the enrollment rate of boys aged 6 to 14 is 63 percent higher than the enrollment rate of girls. In some countries (Niger, Egypt, Morocco, India, and Pakistan) where there is a high degree of female disadvantage in enrollment on average, wealth interacts with gender to exacerbate gaps in educational enrollment among the poor. This difference can be quite large. For example, in India, while 80 percent of girls from households in the top 20 percent complete grade 8, only 9.5 percent of girls from the poorest 40 percent complete grade 8. Sources: Deon Filmer and Lant Pritchett, “ The Effect of Household Wealth on Educational Attainment ,” Policy Research Working Paper, The World Bank, 1998; Deon Filmer, “The Structure of Social Disparities in Education: Gender and Wealth”, Policy Research Working Paper, The World Bank, 2001. The direct impact of fees on enrollment can be assessed by calculating the price elasticity of demand for primary education. Estimates in the 1980s showed elasticities on average to be less than 1 for education at all levels16; there have been very few estimates in the 1990s other than Lavy for Ghana (1996) and Sipahimalani for India (1999).17 Generally, however the estimates that do exist do not examine the elasticities for different levels of education or for different household income groups or for the two sexes. There are exceptions; Gertler and Glewwe, for instance, did in the 1980s estimate that the price elasticity of demand for the poorest quartile of the rural population in Peru was 2-3 times larger than for the highest income quartile.18 It has also been established that the price elasticity is consistently higher for girls than for boys in several countries in Africa, Asia and Latin America.19 Though the evidence is limited, it is nonetheless convincing that enrollments will be lower for the poor and for girls as the household cost of education rises. As a result, reductions in the costs of schooling for the poor and for girls will promote their enrollment, attendance and retention. These reductions can include both the reduction/abolition of fees and the provision of incentives designed to offset 16 See, for instance, Emmanuel Jimenez, Pricing Policy in the Social Sectors, World Bank, 1987, pp. 80-81. 17 Cited in World Bank, Engendering Development, 2001, p. 166. 18 See for instance, Paul Gertler and Paul Glewwe, “ The Willingness to Pay for Education in Developing Countries: Evidence from Rural Peru”. 19 World Bank, Engendering Development, 2001, pp. 165-169. 17 some of the opportunity costs. In some countries where indirect costs are high, for instance, cash transfers have proven to be successful in improving girls’ school attendance (Box 4). Box 4: Mexico OPORTUNIDADES OPORTUNIDADES, formerly known as PROGRESA, is a major conditional cash transfer program in Mexico aimed at developing the human capital of poor households. The program provides monetary transfers to families contingent upon children’s regular school attendance. The transfers are provided to families and the benefit levels are intended to offset the opportunity costs of sending children to school. Transfers increase with the grade level in school—recognizing that the opportunity cost of children’s time increases as they grow older. Rigorous impact evaluations of the program indicate that it has significantly increased the enrollment of children, particularly girls and especially at the secondary school level. The results indicate that the children will have an average of 0.7 years of extra schooling because of OPORTUNIDADES. Schools in CIS countries have generally increased fees in recent years and enrollments appear to have declined as a result, though the formal causation has not been established. In Europe and Central Asia, as already noted, gross basic education enrollments fell during the 1990s, with the reductions largely concentrated (and hence proportionately much greater) in Western CIS, the Caucasus and Central Asia. In Georgia, for instance, basic education enrollment rates fell by 15% from 1989-96. Evidence varies but indicates that the decline in enrollment and in school attendance, at least in Bulgaria, Russia, Azerbaijan, Armenia, Tajikistan, the Kyrgyz Republic, Moldova, Romania and Hungary, is concentrated among the poor and minorities (especially Roma).20 It is perhaps to be expected that the two regions in which user fees appear to have the greatest negative impact on enrollment are Africa and Europe and Central Asia (ECA). These two regions have experienced prolonged economic decline in the last two decades. Inequality is also very high in Africa and increasing very rapidly in ECA. It seems likely that fees would pose a greater problem for parents when there is limited economic growth. Under such circumstances, governments find it harder to raise revenues and so are more likely to resort to fees, formally or informally, while parents simultaneously will find it harder to pay fees out of reduced or stagnant household incomes. In many cases, it is argued, the only means to improve the availability of resources directly to the school and the teacher is to permit the local generation of fees that are locally utilized. Impact of Fees: Learning. The impact of fees on learning or quality is much harder to disentangle. Demand for schooling tends to be more sensitive to quality for poorer households and for girls.21 Given certain elasticity ranges, some argue that there is scope to increase fees 20 World Bank, Making Transition Work For Everyone, 2000, pp.228 – 235. See also, Peter S. Heller and Christian Keller, Social Sector Reform in Transition Countries, IMF Working Paper WP/01/35. 2001 21 See, for instance, World Bank, Engendering Development, 2001, pp. 171-172. 18 without lowering enrollment if the increased revenue is used to cross- subsidize the poor.22 Again, however, such an argument fails to distinguish the different groups of actual and potential students, especially by income group and by sex. This presents a complex analytical dilemma. Does the evidence that demand for education by the poor and by girls is relatively more sensitive to price and relatively more sensitive to quality mean that there is a trade-off between price and quality? There is no clear empirical answer, but it is important to note that the existence of fees does not, in and of itself, assure quality (though it may sometimes contribute to it when the fees are used to provide books and other learning materials) while it almost certainly does deter enrollments by the poor and by girls. As the world now moves toward universal primary education, it is particularly important to examine this possible trade-off in specific country circumstances. Yet there appears to be no recent study that does this.23 Fees are often the main source of qualitative inputs. In India, for example, textbooks and learning materials account for 76% of the fees collected. One area where a systematic review has been conducted is textbook charges – sales and rentals. None of the textbook sales schemes in Africa supported by the Bank seems to have been successful in reaching the poor.24 Even with a subsidy on the price, for instance, most rural primary students in Togo could not afford books (a full set of which represented more than twice average annual family spending on education). In addition, with the exception of Lesotho, none of the textbook rental schemes have been successful, either because the poor could not pay the rental fees (Burkina Faso and Swaziland) or because the schemes were not financially sustainable (The Gambia). Offsetting Fees. Many countries recognize that fees can be a burden on the poor and may deter enrollments and learning. As a result, there is a wide variety of measures in place around the world to mitigate the effects of fees. These can include reducing or eliminating tuition fees for the poor, offering scholarships to attend school, providing free textbooks and other learning materials, and providing free uniforms (Table 5). In addition, some countries provide targeted subsidies to help offset some of the other costs of attending school, such as transport costs and bursaries and stipends to enable poor families to overcome the indirect opportunity costs (in terms of lost contributions to the household economy) of attending school. 22 See, for instance, Nancy Birdsall, “ Demand for Primary Schooling in rural Mali: Should User Fees be Increased?”, Population, Health, and Nutrition Department Technical note, World Bank, 1987. 23 An analytical study of what has happened in Malawi following the abolition of fees could be useful in this regard. 24 Sakhevar Diop, World Bank Support for Provision of Textbooks in sub-Saharan Africa, draft, 2001. 19 Table 5: Fee Mitigating Measures Mitigating Measure Country Examples Tuition Related Interventions Reduce/eliminate tuition fees for the poor and China, Djibouti, Ethiopia, Jordan, Lesotho disadvantaged Scholarships, Stipends Often accompanied by performance related Bangladesh, Brazil, Ethiopia, The Gambia, Ghana, measures25 Guatemala, India, Indonesia, Malawi, Mauritania, Morocco, Mozambique, Nepal, Nicaragua, Panama, Pakistan, Senegal Targeted Bursaries China, Colombia, India, Indonesia, Mexico, Nicaragua Tanzania, Zambia Textbook-related interventions Replace textbook rental fee with book loan scheme Gambia, Vietnam Provide free textbooks to grades 1 to 7 Mali, Nigeria, Rwanda, Senegal Provide free textbooks to target groups (poor, girls, Armenia, China, Chile, Ethiopia, Guinea, India, etc.) Malaysia, Morocco, Nepal, Tajikistan, Turkey Provide book allowance to poor families Bulgaria Learning Materials Provide other learning materials (stationery, etc.) Bangladesh, India, Mozambique Uniforms Provide free uniforms India Transport-related interventions Provide bicycles for poor rural students to get to Thailand school Sources: User Fee Survey; Project Appraisal Documents. PART 3: THE WORLD BANK AND PRIMARY EDUCATION FEES Policy Position. At the level of broad policy, the World Bank has long argued for the importance of basic education and the priority that basic education should take in Bank support for education. Accordingly, general World Bank education policy documents have never supported tuition fees at the primary level because such fees could limit demand for primary education. Bank policy, in line with its support of the Millennium Development Goals, is to maximize primary enrollments and learning.26 The current overall Bank policy document for education, Education Sector Strategy (1999), emphasizes the need for universal basic education but does not go into the fee issue; although, some of its parallel regional strategy papers, notably those from the African and the Latin American and Caribbean regions, clearly call for free basic education. The MNA and EAP strategies do not discuss fees while the ECA strategy warns against growing over-reliance on fees in the region.27 25 For example, “girls in deprived regions who excel in science and math “ or scholarships are “contingent on regular school attendance and successful grade completion”. 26 The Bank’s current external position, posted on its external website and reproduced at Annex 1, states that the Bank “does not support user fees for primary education and for basic health services for poor people”. 27 There is no strategy paper for South Asia. 20 The Bank’s stance on fees other than for tuition has been less definitive. It has recognized that there may be a role for fees for books and materials at primary level to improve quality, when public resources are inadequate. It has also recognized that there may be a role for community contributions to primary school construction and running costs.28 In addition, Priorities and Strategies for Education (1995) went beyond the discussion of tuition fees and other household payments to argue for targeted stipends for households that cannot afford to enroll their children because of obstacles such as the direct costs of fees or the indirect costs of the contribution of children to the household economy. This concept also forms part of the Bank’s current external position (Annex 1). In addition to broad policy positions, the Bank also has detailed operational policies and guidelines. These do not take positions on user fees in education, at primary or indeed any level, with the exception of textbooks. The new Operational Guidelines on Textbooks (2001) include several principles, one of which is “assurance that cost will not be an obstacle for poor students’ access to textbooks and reading materials.” The textbook guidelines do not, however, suggest any ways in which this access might be achieved. Country Program Practice. The survey of task managers conducted for this report examined what the Bank is doing with regard to primary education user fees in specific countries. There are only a few cases in which the World Bank has supported fees of any type at the primary level in recent years. Recent support for textbook fees has been limited to the ECA region (Bosnia Herzegovina, Latvia and Macedonia), although the Bank encouraged such fees in many African and Asian countries during the 1990s. In total, 64 of the Bank’s 89 education projects in 40 African countries between 1985-2000 included textbook components, with 43 encouraging some form of fee charging (18 by book sales, 23 by rentals, and 2 by a refundable deposit).29 There are no cases where the Bank has encouraged compulsory school uniforms. In six countries the Bank has worked through its dialogue to encourage community and PTA contributions (Bosnia Herzegovina, Comoros, Madagascar, Mexico, Pakistan, and Tajikistan), usually to encourage decentralized educational administration. There is only one country, Comoros, in which the Bank has advocated specialized school-based fees. There are many more cases in which the Bank has worked and continues to work to reduce, eliminate or mitigate primary school fees. The Bank has supported the elimination of tuition fees in Uganda, Malawi, Lesotho, Cameroon, Tanzania and Kenya and is currently working with Benin, China, Guinea-Bissau, Jordan, Senegal, and Vietnam to systematically eliminate fees. In a further 15 countries, the Bank is working with clients through Bank projects to help offset tuition fees for those who cannot afford them, usually in a targeted manner. In Indonesia, Nicaragua and Zambia, for example, Bank projects include components to provide bursaries to disadvantaged children. The same is done for poor girls in Bangladesh, The Gambia (Box 5), Nepal and Pakistan, though some of these are confined to secondary school students 28 This paragraph derives from a review of Bank education strategy documents since 1980, including Education Sector Policy Paper (April 1980), Primary Education Policy Paper (1990), Priorities and Strategies for Education (1995) and Education Sector Strategy (1999). 29 Diop, op.cit. 21 only. Particularly in African HIPC countries, the Bank is working to see that revenue lost when fees are eliminated is replaced (at least partially) from debt relief funds, as is the case in Tanzania, Benin and Guinea-Bissau. Similarly, with regard to textbook fees, the Bank recently successfully argued for its elimination in Senegal, is currently working for their elimination in three countries (Mauritania, Nigeria, and Vietnam), and has included measures in projects in 20 countries designed to make fees affordable to the poor, such as fee waivers in Cape Verde and Guinea and free textbooks to targeted groups in Nepal (all girls in grades 4-5, boys in remote districts in same grades), Tajikistan (poor) and Turkey (rural students). More broadly, following a review of the impact of textbook fees in Africa, the Africa region of the Bank is currently moving toward adopting a policy that all primary school textbooks should be provided free of charge to students. Box 5: Targeted Education Scholarships in The Gambia One of the key objectives of Third Education Sector Program (TESP) in The Gambia is to increase the enrollment, retention and performance of girls attending Grades 1-12. Currently, girls' enrollment and completion rates are significantly lower than those of boys. In 1996, the overall Gross Enrollment Ratio (GER) stood at 70 percent. The GER for boys exceeded that of girls - 79 percent versus 61 percent. Research has shown that the overriding reason for these low enrollment and completion rates is poverty. Lowering the cost of schooling relative to parents' incomes is thus key to better education outcomes for girls in The Gambia. To help overcome this poverty barrier, the TESP will see the introduction of a scholarship scheme - a common demand-side financing mechanism - for girls at the upper basic and secondary levels. Key elements of the scheme include: full scholarships for tuition, books and examination fees will be awarded to one-third of girls in upper basic and secondary schools with low enrollment in the most deprived regions of The Gambia; and full scholarships for tuition and examination fees will be awarded to ten percent of girls in upper basic and secondary schools in less deprived regions who are excelling in science, mathematics and technology. A trust fund to finance the scholarship scheme will be administered by the Social Welfare Department. The scheme will be evaluated at various stages and could be extended to needy and meritorious students in less deprived regions. There are also three countries in which the Bank is working through its dialogue to eliminate compulsory uniforms (Mexico, Tanzania and Vietnam) and another eight countries in which Bank projects include measures to make compulsory uniforms affordable to those with insufficient means. In Turkey, for instance, a Bank project provides school uniforms and meals to poor urban and rural students. The Bank is also actively trying to assist with the elimination of community and Parent Teacher Associations (PTA) contributions in three countries where they are burdensome (Tanzania, Uganda and Vietnam) and its projects include mitigating measures in a further 18 countries. In the same three countries plus Cameroon, the Bank is 22 encouraging the abolition of other school-based activity fees while it includes measures to offset them for the poor in projects in 12 countries. Lessons Learned. The World Bank is actively involved in attempts to eliminate primary school user fees and to offset them for the poor. Annexes 5 and 6 detail the experiences of Uganda and Malawi. Some important lessons have been learned from these attempts: • Fee abolition alone is not enough. It must be part of a broad government commitment to achieving universal primary enrollment, with many complementary measures and strong political leadership from the highest levels. • Indirect costs can be an even greater obstacle than fees. Where appropriate, therefore, the Bank has now gone further in 30 countries to introduce stipends, targeted vouchers, bursaries and other demand-side financing mechanisms to encourage the enrollment of the poor and of girls at all levels of education (Table 5), offsetting the indirect opportunity costs.30 These mechanisms essentially involve payments to families to enroll their children. Those programs that have been subject to rigorous review have all been shown to have achieved considerable success in increasing enrollments of the poor (Guatemala’s Eduque a la Nina, Brazil’s Bolsa Escola, and Mexico’s OPORTUNIDADES programs). • Fees cannot simply be abolished without consideration of whether, and how, they should be replaced by an alternative source of income. When fees contribute to operational effectiveness, or even perceived quality; simply abolishing fees without replacement can result in a serious deterioration of the education system, as the case of Malawi regrettably shows (Annex 5). Uganda prepared better, or at least adapted more quickly to what was needed, increasing the recurrent budget for primary education from 9% of total education spending in 1996 to 19% by 1999, and a reduction in the subsidies for higher education (Annex 6). Adequate measures must therefore be in place to provide equivalent revenues to finance the expenditure previously covered through fee revenue. • There are four principal sources of such replacement revenues: o Countries may increase expenditure on education by switching spending from other sectors or by increasing revenues. o Improving the efficiency of education spending, particularly the balance between different educational levels and the balance between salaries and other expenditures. Public spending on education is generally very inefficient, as the figure in Annex 10 shows. Expenditure shifts do take time, however, and must be well prepared politically, sometimes being as difficult to implement as the introduction or augmentation of fees. 30 Not all these 30 cases include primary education, but many do. See also Table 5 above on mitigating measures. 23 o Using Heavily Indebted Poor Countries (HIPC) debt relief funds on a temporary basis to close the financing gap in many heavily indebted countries over the next few years. In practice, as this funding is necessarily limited in its availability in the future, any use of HIPC funds should be accompanied by measures to replace revenues on a longer term basis. The HIPC initiative, with its focus on improved budgetary management and tracking of debt service relief has given IDA borrowers the ability to increase education expenditures and mobilize additional support, which can be allocated to the education sector. A combination of HIPC funds and a start on improving efficiency is the most likely route, and indeed has already been adopted through the PRSP process in several countries. In the Bank’s HIPC dialogue in Africa, for instance, increased efficiency in public spending characterizes 12 out of 18 recent cases, either by reallocating public spending or by redeploying teachers. The HIPC Unit (2003) suggested that governments have indicated that they would use 40 percent of relief for education. o Using funds from the Fast Track Initiative (FTI) Catalytic Fund, which was established in 2003 to support countries that have at present a limited level of donor support for education, represents an extraordinary opportunity to provide temporary financing to FTI countries working towards the elimination of fees. The fund provides transitional assistance to countries and enables them to scale up implementation of their sector programs on a temporary basis . Clear lessons have been learned about fee abolition and reduction; however very little is known about fee-offsetting measures, as listed in Table 5. Do these measures effectively target the poor and other groups as intended? What do they cost to administer? To what extent can targeting criteria developed for purposes outside the education sector be applied? Ignorance about these questions in education stands in stark contrast to the health sector, where there has been much work on fee waiver schemes. Though the Bank has been involved in the abolition of fees, it has not conducted a systematic review of countries with fees to see if the fees represent an obstacle to enrollment or learning. There are several countries with low enrollments and relatively high tuition fees in which, though we lack empirical evidence, it may well be the case that fees impede enrollment. These countries include, Bhutan, Cambodia, Ethiopia, India, Lao PDR, Mali, Nepal, Pakistan, Papua New Guinea, Rwanda and Togo. In addition, the Bank has not adopted any very clear policies towards fees, official and unofficial, in the CIS countries and China, probably because these countries’ primary enrollment rates, though falling, remain relatively high. PART 4: RECOMMENDATIONS Fees, in the sense of direct household payments, are common in primary education around the world, and represent perhaps 20 percent of all education spending, and as much as 30 percent in Africa and even 40 percent in the former CIS and East Asian socialist countries. 24 Tuition fees are less common than fees for books, PTA and community contributions and compulsory uniforms. Fees represent a significant proportion of household spending, although this proportion varies widely from country to country. It is universally the case that the proportion is highest for the poorest households. Too little is known about the determinants of enrollment and learning, but fees are certainly obstacles in many, but not all, cases. Universal primary completion is a top World Bank priority, expressed in the Bank’s commitment to the Millennium Development Goals. The Bank has made abundantly clear in its policy statements that it does not support user fees for tuition in primary education and has in recent years actively supported fee abolition in countries, mainly in Africa, in which fees appear to represent an obstacle to enrollment. Important lessons have been learned from these fee abolition efforts, of which the most important is that, in cases where fees are an effective contribution to school access or quality, there is a need to ensure replacement of fees with revenues of equal effectiveness and size, if fees are reduced or abolished. There has not, however, been any systematic analysis of the determinants of enrollment in Bank client countries to determine whether fees represent an obstacle to access or completion; on the face of it, there are a number of low enrollment countries in which fees might be impeding enrollments, as well as relatively high-fee countries where fees may not be impeding enrollment or completion (but may be impeding effective poverty reduction). While the Bank has not supported tuition fees, it has supported various other types of fees in primary education, including textbook fees, along with programs to offset or waive the fees for the poor and other target groups such as girls. Some of this past support is now under reconsideration; it is likely, for instance, that the Africa region of the Bank will soon move formally to support the free provision of textbooks. Almost nothing systematic is known about the effectiveness of fee waiver schemes in primary education. Several implications follow directly from the above summary: 1. The World Bank should develop an explicit policy on user fees in primary education, taking account of their impact on enrollment and learning and of their major importance in total educational spending. This policy should be based on the premise that no child should be excluded from primary education on the basis of inability to pay. It should assume that the Bank will work actively with governments to find alternatives to existing fees, paying particular attention to revenue replacement. . 2. The World Bank should develop a clearer and shorter public position on user fees in primary education. The current position (Annex 1), while reasonable, is cautiously worded and hard to summarize and commit to memory. A clearer position should note the Bank’s active opposition to fees, rather than its current passive “does not support” statement. It should also reiterate the Bank’s opposition to other obstacles (principally indirect costs) to enrollment and learning at the primary level. 3. The World Bank should review the determinants of enrollment in all countries that are unlikely to achieve the education Millennium Development Goals. In those cases where fees impede enrollment, the Bank should then move quickly to help governments abolish the fees or, at a minimum, provide effective fee waivers to those unable to attend school because of the fees. In such cases, abolition of fees should be accompanied by 25 steps to replace the lost revenues – HIPC, FTI Catalytic Fund and Bank financing can be used, at least as a transitional step. Such a review will also identify countries in which indirect costs are an obstacle. In these cases, the Bank should extend its practice of supporting demand-side financing mechanisms aimed at out-of-school children. While the EFA countries should be the initial priority for such reviews, a subsequent step should be to examine the determinants of enrollment in higher enrollment countries with increasing inequality and rapidly declining enrollments, notably the IDA-eligible CIS countries in the ECA region. 4. The World Bank should examine the effectiveness of textbook charging schemes it has supported and of accompanying fee mitigation measures for those unable to pay for books, following the lead of the Africa region. This would seem to be particularly urgent in the ECA region but applies more generally also. Following such a review, the Operational Guidelines on Textbooks should likely be modified to include a more explicit statement on charging, and should also likely be accompanied by some lessons of experience about textbook fee mitigation measures. 5. The World Bank should analyze the effectiveness of targeted fee waiver schemes in primary education. This work would be analogous to that already done in health. 26 List of Annexes 1. Current external position. 2. Survey results (detailed country-by-country). 3. Survey instrument. 4. Summary of survey results. 5. The case of Malawi. 6. The case of Uganda. 7. The case of Kenya. 8. The case of Tanzania. 9. The case of China. 10. HIPC figure on public spending in education. 27 Annex 1—World Bank Current Statement on User Fees The World Bank and User Fees The Bank does not support user fees for primary education and for basic health services for poor people. Basic Education The World Bank does not support user fees for primary education. Not only is education a major route out of poverty, but primary education benefits society at large by creating a numerate and literate population. User fees on primary education can be a financial burden on poor people, forcing their children to leave school, trapping families in poverty, and hurting their local and national communities. Where governments do levy user fees, the Bank helps governments to reduce the burden on poor people by recommending, and providing finance for targeted subsidies as is the case in Bangladesh Female Secondary School Assistance Program. The FSSAP has introduced mitigating measures to offset school fees by providing monetary incentives for girls to reduce direct cost of schooling and to encourage participation. Since 1996, over 925,000 girls have received stipends. In some cases, these subsidies go beyond eliminating user charges for poor people and can provide a positive incentive for parents to send their children to school (examples: Brazil’s Bolsa Escola and Mexico’s OPORTUNIDADES). For example, Turkey illustrates a case where education is free of tuition fees but with high private costs for textbook fees, uniforms and PTA contributions. Turkey Basic Education Project (FY 98) addresses these financial constraints through the provision of material incentives to poor families through free school meals, free textbooks and free uniforms. While the World Bank strongly believes that governments should not charge user fees for education, local communities, however, sometimes do levy such charges (often paid in-kind through labor services) which can build community support for local schools and subsequently improve the quality of, and expand the access to, education. Given that such local charges can help sustain education services, the Bank believes they must be carefully designed to ensure that poor parents keep their children in school. For example, in Mozambique the running costs budget of schools is inadequate and thus, the responsibility is placed on the parents. Under the ESSP program, the World Bank is providing funds for “direct grants to schools” to ease the pressure on parents and to ensure that parents/communities are paid for their labor, if a new school is being build in the community. The World Bank has also worked with governments to plan for a phased elimination of user fees as in the case of Lesotho. Under ESDP II, The Bank is supporting the phased elimination of user fees one year at a time for six years and in the meantime providing bursaries for disadvantaged children. The World Bank has worked with governments to expand access to basic education, including the removal of financial and other barriers, especially for the poor. The World Bank is also working to document and help eliminate unauthorized or unofficial charges for basic education services (example: Vietnam Education Finance Study). 28 Several countries have already eliminated school fees at the primary school level. The case of Uganda illustrates what needs to be in place to support elimination of user fees. The government’s broad consultations with the World Bank and other external donors on the development of an investment of and policy framework for education to close the financing gap resulted in a comprehensive and systematic response by government , increased external resources, improved sector management, and transparency at the school level. In countries at the transitional stage, the World Bank is assisting governments to reform their policies. This is particularly evident in countries benefiting from the Highly Indebted Poor Countries (HIPC) Initiative and countries preparing Poverty Reduction Strategy Papers. Tanzania illustrates a case where the World Bank has worked with the government to reverse its past policies on charging primary school user fees. Cost-sharing had been introduced in Tanzania as a means of generating money at the community level to make up for the shortfall in government spending and to generate community “ownership” and accountability. However, from the beginning it was evident that for the poorest in any community, the burden of paying for education was considerable. This is indicated in the latest PRSP document, which states the intention to abolish primary school fees, starting in the fiscal year 2001/2002. The president of Tanzania announced in April of this year (April 2, 2001) that there would be “no fees/levies for school operations. However, communities can raise matching contributions in cash or kind for development projects”. This is also stated in the PRSP. The president stated that the prospect of this funding has improved as a result of Tanzania’s recent designation as a HIPC country by the World Bank and IMF. A World Bank project (The Human Development Project), currently at the PCD stage, is pushing for the “abolition of legal and illegal school levies” and includes a component that will attempt to identify the amount of household expenditure that goes to education as well as identify the type and amount of fees schools charge pupils. The World Bank has always advocated free basic education—because of high social returns (increased productivity, fertility and mortality reduction, etc.). In fact, Priorities and Strategies for Education goes even further to suggest that targeted subsidies be given to children from poor households (p. 10), although community participation is not ruled out in a system that protects the poor. Public investment in education is justified on high social returns, meaning that without public subsidies, individuals would not invest in education at socially optimal rates. Therefore, user fees, to the extent that they would act as a deterrent to enrollment, are socially inefficient. Health Services The Bank supports the provision of free basic health services to poor people. In the case of immunization, maternal and child care, and certain programs for TB, HIV/AIDS, STI, and malaria that have large benefits for the community and vulnerable groups, the Bank discourages user fees and helps countries to finance these activities using other measures. For example, the Bank advised China to stop charging people for TB diagnosis and treatment. With fees for these services, many low-income people failed to enter treatment or dropped out early. The Bank financed a large-scale project in China to support the implementation of the WHO recommended 29 treatment approaches for TB, with the elimination of all user fees for TB treatment. These changes have resulted in dramatic increases in the numbers of tuberculosis cases cured. In many low-income countries where taxation revenues, and therefore public resources for the health sector, are very limited, many households—including poor households—contribute premiums towards risk-sharing arrangements such as community financing and other insurance schemes that can protect them from the impoverishing effects of infrequent, but costly illnesses. The Bank works actively with countries to find ways to subsidize the premiums of poor populations so that financial protection under such insurance arrangements will expand to cover larger segments of the populations. In the absence of such insurance arrangements, in very low- income communities where the governments’ resources are extremely limited, well-designed and implemented user fees can mobilize additional resources from better-off groups that can in turn be used to improve services for poorer groups. Such cost-sharing schemes can play a critical role in helping ensure essential services are available. Where countries choose this option, the Bank works closely with policymakers to make sure that poor people face zero or minimal costs. The Bank also works with countries to identify and reduce other barriers to access to basic services, such as unofficial fees charged by officials, language difficulties, and long distances to services. 30 Annex 2—User Fees and Primary Education 31 32 33 34 35 36 Notes: S=School; O=Other; E=Eliminate; MA=Mitigate All; MS=Mitigate Some; P=Promote; **October 2000 response 37 Annex 3—User Fee Survey TYPE OF FEE 1.School tuition 2. Charges for 3. Purchase of 4. Financial 5. Other fees textbook mandatory contributions payments, (first-time (purchase or uniforms or schools (often including required registration fees rental) material for called “voluntary”) monetary and annual fees) mandatory required from contributions, uniforms Community or unofficial Parent-Teacher examination fees, Associations, supplementary including payment payments for of teacher teachers, etc.) salaries for (Please specify) community schools, etc. (Please specify) 1. Are these fees implemented? Y = Yes; N = No 2 Are these fees legal (that is, in accordance with national laws or constitution) Y = Yes; N = No. 3. At what level are the fees collected and by whom? S= School; O= Other (please specify) 4. What is the average amount of fees collected per student? 5.What is the estimated revenue from fees collected and what share of public spending on education does the revenue contribute? 6. How does the World Bank treat these fees in its dialogue and/or projects? E = works with government to Eliminate fees. MA = introduces strategies to Mitigate effects of fees on All students. MS = introduces strategies to Mitigate effects of fees on Some students (e.g. girls, the poor). P = Promotes or supports fee. Please provide examples. 7. We are trying to get a comprehensive picture of the primary education fee situation and the Bank’s role. It would be greatly appreciated if you could provide us with any information that might be of use and interest and is not covered by any of the questions so far. Please send us or refer us to any studies conducted in the country on User Fees whether by the World Bank or other partners, including NGOs. Thank you very much for your assistance and time. 38 Annex 4—Summary of Survey Results Results of User Fee Survey: Summary Findings of User Fees* A survey in a matrix form was sent to all Task Team Leaders of active Basic education projects in October 2001 (See Annex 3 for Sample E-mail User Fee Survey and Annex 2 for Survey Results detailed country by country in form of a matrix. Data Coverage: • We have data for 79 Bank client countries with IBRD/IDA-supported basic education projects. • Out of the 79 countries, almost all (97%)—77 countries—have some type of fee with the exception of Uruguay and Algeria that have no type of fees at all—here education is completely free. • Countries generally have several fees with the exception of the following countries: eight out of the 77 countries (10%), do not have other fees but require community/PTA contributions (Burkina Faso, Malawi, Senegal, Uganda, Zambia, Bangladesh, Bolivia, and Brazil). • 69 of the 77 countries (90%) that have fees have more than one fee; for example, the following countries have all five types of fees (tuition, textbook, uniforms, financial contributions, and other school-based fees)—Mauritius, Tanzania, Togo, Egypt, China, Vietnam, and Pakistan. The following countries have all the other fees, except tuition—Cambodia, Lao PDR, Malaysia, Trinidad and Tobago. The rest of the countries—58 out of the 77 (75%)—have between two and three types of fees. Tuition Fees: • For these 79 countries, tuition fees are implemented in 30 countries (38%), though only 19 countries (63%) have legal tuition fees. In 11 countries fees are implemented illegally—Benin, Ethiopia, Indonesia, Vietnam, India, Nepal, Colombia, Bosnia, Latvia, Russia, Egypt. • World Bank projects have introduced mitigating measures (either generalized or targeted) in 15 (50%) of these cases. • The World Bank is working to eliminate fees in eight of these countries—Benin, Guinea Bissau, Lesotho, Tanzania, China, Vietnam, and Jordan. In the past, The World Bank worked with governments to eliminate fees in Uganda and Malawi. 39 • With the exception of Solomon Islands, the World Bank does NOT promote tuition fees in any of the countries. Textbook Fees: • Textbook fees are implemented in 37 countries (47%), though only in 25 countries (68%) are these fees legally allowed. • Where fees are implemented, World Bank projects have introduced mitigating measures (either generalized or targeted) in 20 (54%) of these cases. • The World Bank promotes textbook fees in three countries—Latvia, Macedonia, Bosnia (all ECA countries). Note: In the past textbook fees were introduced by the World Bank in many African countries. • The World Bank is working to eliminate charges for textbooks in five countries— Lesotho, Mauritania, Nigeria, Tanzania, Vietnam—and has recently eliminated textbook charges in Senegal. Uniform Fees: • Purchase of mandatory uniforms is implemented in 39 (49%) of these cases. In 25 (64%) of these countries, fees for uniforms are legally allowed. • Where fees are implemented, World Bank projects have introduced mitigating measures (either generalized or targeted) in eight (21%) of these cases. • The WB does NOT promote purchase of mandatory uniforms in any of these countries and is working to eliminate fees in three of these countries—Tanzania, Vietnam, Mexico. PTA/ Community Contributions: • PTA/community contributions are common in 56 (71%) of the countries surveyed. In 34 (61%) of these countries, it is legally allowed. • World Bank projects have introduced mitigating measures (especially targeted to the poor) in 18 (32%) of these cases. • The World Bank is working to eliminate these fees in three of these cases —Tanzania, Uganda, Vietnam. • The World Bank promotes community contributions in six of these cases—Comoros, Madagascar, Pakistan, Mexico, Bosnia, Tajikistan. 40 Other School-based Fees: • Other school-based activity fees are collected in 34 (43%) of the countries. In 21 (64%) of these countries, it is legally allowed. • The World Bank has introduced mitigating measures in 12 (35%) of these cases. • With the exception of Comoros, the World Bank does NOT promote tuition fees in any of the countries and is working to eliminate these fees in four—Cameroon, Tanzania, Uganda, Vietnam. • The Bank is non-committal in 17 of these countries where fees are implemented. NOTE: The cases where the World Bank has been involved in eliminating fees have almost ALL been in Africa with the exception of Jordan; Mexico, where they are working to eliminate mandatory uniforms; and Vietnam, where the Bank is currently discussing elimination of fees in their dialogue with the government. 41 AFRICA Data Coverage: • We have data for 27 bank client countries from the Africa Region with IBRD/IDA- supported basic education projects. Tuition Fees: • For these 27 countries, tuition fees are implemented in 11 countries (41%)—Benin, Comoros, Eritrea, Ethiopia, Lesotho, Madagascar, Mali, Mauritius, Rwanda, Tanzania, Togo. With the exception of Benin, and Ethiopia, all fees are collected legally. • Where fees are imposed, World Bank projects have introduced mitigating measures (either generalized or targeted) in six (55%) of these cases—Comoros, Eritrea, Madagascar, Mauritius Mozambique, Togo. The World Bank is working to eliminate fees in four of these countries (45%)—Benin, Guinea Bissau, Lesotho, Tanzania. In the past, the World Bank also worked with the governments to eliminate fees in Uganda and Malawi. • The World bank does NOT promote tuition fees in any of the countries. Textbook Fees: • Textbook fees are implemented in 10 countries (37%)—Cameroon, Cape Verde, Comoros, Lesotho, Madagascar, Mauritania, Mauritius, Nigeria, Tanzania, Togo—and in all 10 countries, these fees are legally allowed. • Where fees are imposed, World Bank projects have introduced mitigating measures (either generalized or targeted) in seven ( 70%) of these cases—Cameroon, Cape Verde, Comoros, Madagascar, Mauritius, Mozambique, Togo. • WB does NOT promote textbook fees in any off these cases. In three (30%) countries, the World Bank is working to eliminate charges for textbooks—Lesotho, Mauritania, Tanzania. The World Bank also recently eliminated textbook charges in Senegal. Uniform Fees: • Purchase of mandatory uniforms is implemented in 12 (48%) of these cases—Benin, Cape Verde, Eritrea, Gambia, Ghana, Lesotho, Mali, Mauritius, Niger, Nigeria, Tanzania, Togo. • In 10 (83%) of these countries fees for uniforms are legally allowed—Cape Verde, Eritrea, Gambia, Ghana, Lesotho, Mali, Mauritius, Nigeria, Tanzania, Togo. 42 • Where fees are imposed, World Bank projects have introduced mitigating measures (either generalized or targeted) in two countries (16%)—Mauritius, Togo. • The WB does NOT promote purchase of mandatory uniforms in any of these countries and is working to eliminate fees in Tanzania. PTA/Community Contributions: • PTA/ Community contributions are common in 22 (81%) of the countries surveyed— Benin, Burkina Faso, Cape Verde, Cameroon, Chad, Comoros, Gambia, Ghana, Lesotho, Madagascar, Malawi, Mali, Mauritania, Mauritius, Mozambique, Niger, Nigeria, Senegal, Tanzania, Togo, Uganda, Zambia. In 12 (55%) of these countries, it is legally allowed—Cape Verde, Cameroon, Chad, Comoros, Lesotho, Madagascar, Mali, Mozambique, Nigeria, Tanzania, Togo, Zambia. • Where fees are imposed, World Bank projects have introduced mitigating measures (either generalized or targeted) in 10 (45%) of these cases. • The World Bank is working to eliminate these fees in two of these cases—Tanzania, Uganda. • The World Bank promotes community contributions in Madagascar. Other School-based Fees: • Other school-based activity fees are collected in 12 (41%) of the 50 countries— Cameroon, Comoros, Eritrea, Ghana, Lesotho, Madagascar, Mauritania, Mauritius, Rwanda, Tanzania, Togo, Uganda. In nine (75%) of these countries, it is legally allowed—Comoros, Eritrea, Ghana, Lesotho, Madagascar, Mauritius, Rwanda, Tanzania, Togo. • The World Bank has introduced mitigating measures in four (30%) of these cases— Eritrea, Madagascar, Mauritius, Togo. • With the exception of Comoros, the World Bank does NOT promote these types of fees and is working to eliminate these fees in Cameroon, Tanzania and Uganda. 43 EAST ASIA AND THE PACIFIC Data Coverage: • We have data for 10 bank client countries with IBRD/IDA-supported basic education projects. Tuition Fees: • For these 10 countries, tuition fees are implemented in five countries (50%)—China, Indonesia, Papua New Guinea, Solomon Islands, Vietnam. In three (60%) countries— China, Papua New Guinea, Solomon Islands—fees are collected legally. • Where fees are imposed, World Bank projects have introduced mitigating measures (either generalized or targeted) in one of these cases—Indonesia. • The World Bank is working to eliminate fees in two of these countries—China, Vietnam. • With the exception of Solomon Islands , the World Bank does NOT promote tuition fees in any of the countries. Textbook Fees: • Textbook fees are imposed in seven countries (70%)—Cambodia, China, Indonesia, Lao PDR, Malaysia, Thailand, Vietnam—though only four countries (57%) have legal fees— China, Lao, Malaysia, Vietnam. • Where fees are imposed, World Bank projects have introduced mitigating measures (either generalized or targeted) in two cases—Cambodia, Lao PDR. • The World Bank does NOT promote fees in any off these cases. • The World Bank is working to eliminate charges for textbooks in Vietnam. Uniform Fees: • Purchase of mandatory uniforms is implemented in eight (80%) of these cases— Cambodia, China, Lao PDR, Malaysia, Papua New Guinea, Philippines, Thailand, Vietnam. In five (63%) of these countries, fees for uniforms are legally allowed—China, Lao PDR, Malaysia, Papua New Guinea, Vietnam. • The World Bank introduced mitigating measures (either generalized or targeted) in two countries (25%) of these cases—Cambodia and China. • The WB does NOT promote purchase of mandatory uniforms in any of these countries and is working to eliminate uniform fees in Vietnam. 44 PTA/Community Contributions: • PTA/ Community contributions are common in eight (80%) of the countries surveyed— Cambodia, China, Lao PDR, Malaysia, Papua New Guinea, Philippines, Solomon Islands, Vietnam. In five (63%) of these countries, it is legally allowed—Lao, Malaysia, Papua New Guinea, Philippines, Solomon Islands. • Where fees are imposed, World Bank projects have introduced mitigating measures (either generalized or targeted) in two (25%) of these countries—Cambodia, the Philippines. • The World Bank does NOT promote community contributions in any of the countries. • The WB is working to eliminate these fees in Vietnam. Other School-based Fees: • Other school-based activity fees are collected in seven countries—Cambodia, China, Lao PDR, Papua New Guinea, Philippines, Solomon Islands, Vietnam—although it is legally allowed in only three (43%) countries—Lao PDR, Papua New Guinea , Solomon Islands. • The World Bank has introduced mitigating measures in two of these cases—Cambodia, Philippines. • The World Bank does NOT promote these types of fees in any of these countries and is working to eliminate these fees in Vietnam. 45 EUROPE AND CENTRAL ASIA Data Coverage: • We have data for 12 bank client countries with IBRD/IDA-supported basic education projects. Tuition Fees: • For these 12 countries, Tuition fees are implemented in five (42%) countries—Armenia, Bosnia, Latvia, Russia, Tajikistan—though only two (40%) countries have legal tuition fees—Armenia, Tajikistan. • The World Bank has not addressed the issue of tuition fees in any of these countries. Textbook Fees: • Textbook fees are imposed in eight (67%) countries—Armenia, Azerbaijan, Bosnia, Latvia, Macedonia, Moldova, Tajikistan, Turkey—and are all legally allowed. • Where fees are imposed, World Bank projects have introduced mitigating measures (either generalized or targeted) in three (38%) countries—Armenia, Tajikistan, Turkey. • The World Bank promotes textbook fees in three (38%) of these countries—Bosnia, Latvia, Macedonia. Uniform Fees: • Purchase of mandatory uniforms is implemented in only two (17%) countries— Azerbaijan, Turkey—and in both cases, it is legally allowed. • Where fees are imposed, World Bank projects have introduced mitigating measures (either generalized or targeted) in one country—Turkey. • The World Bank does NOT promote uniform fees in any of these countries. PTA/Community Contributions: • PTA/community contributions are common in eight (80%) of the countries—Armenia, Azerbaijan, Bulgaria, Latvia, Moldova, Tajikistan, Romania, Turkey—and these fees are legally allowed in seven (88%) of these countries—Armenia, Azerbaijan, Bulgaria, Latvia, Tajikistan, Romania, Turkey. • Where fees are imposed, World Bank projects have introduced mitigating measures (either generalized or targeted) in three (38%) countries—Armenia, Latvia, Turkey. 46 • The World Bank promotes community contributions in two (25%) of these cases— Bosnia, Tajikistan. Other School-based Fees: • Other school-based activity fees are collected in five (42%) countries—Armenia, Latvia, Moldova, Tajikistan, Turkey—and in four (80%) of these countries it is legally allowed—Armenia, Latvia, Tajikistan, Turkey. • The World Bank has introduced mitigating measures in one of these cases—Turkey. • The World Bank does NOT promote these types of fees in any of these countries. 47 LATIN AMERICA AND THE CARRIBEAN Data Coverage: • We have data for 17 bank client countries with IBRD/IDA-supported basic education projects. Tuition Fees: • For these 17 countries, only one country—Colombia—implements tuition fees, and it is not legally allowed. • World Bank projects have introduced mitigating measures addressing indirect costs in five countries—Brazil, Bolivia, Colombia, Mexico, Trinidad and Tobago. • The World Bank has not addressed the issue of user fees in any of these countries since there are no tuition fees. Textbook Fees: • Textbook fees are implemented in five (29%) countries—Dominica, Grenada, Nicaragua, Peru, Trinidad and Tobago—and these fees are legally allowed, except in Peru. • Where fees are imposed, World Bank projects have introduced mitigating measures in two (60%) of these cases—Dominica, Trinidad and Tobago. • The World Bank does NOT promote these fees in any of these countries. Uniform Fees: • Purchase of mandatory uniform is implemented in 13 (76%) of these cases—Colombia, Dominica, Dominican Republic, El Salvador, Grenada, Guatemala, Honduras, Mexico, Nicaragua, Panama, Paraguay, Peru, Trinidad and Tobago. In six (57%) cases, these fees are legally allowed—Dominica, Dominican Republic, Grenada, Honduras, Nicaragua, Trinidad and Tobago. • Where fees are imposed, World Bank projects have introduced mitigating measures (either generalized or targeted) in one country—Mexico. • The World Bank does NOT promote these fees in any of these countries. PTA/Community Contributions: • PTA/Community contributions are common in 11 (65% ) of the countries surveyed— Bolivia, Brazil, Colombia, Dominican Republic, Guatemala, Mexico, Nicaragua, 48 Panama, Paraguay, Peru, Trinidad and Tobago. In four (36%) of these countries, it is legally allowed—Colombia, Dominican Republic, Peru, Trinidad and Tobago. • Where fees are imposed, World Bank projects have introduced mitigating measures (either generalized or targeted) in five (45%) of the countries—Colombia, Dominican Republic, Nicaragua, Peru, Trinidad and Tobago. • The World Bank promotes community contributions in one of these cases—Mexico. Other School-based Fees: • Other school-based activity fees are collected in five (29%) countries—Bolivia, Dominican Republic, Paraguay, Peru, Trinidad and Tobago. In two (40%) of these countries—Dominican Republic, Trinidad and Tobago—it is legally allowed. • The World Bank does NOT promote fees in any of these countries. 49 MIDDLE EAST AND NORTH AFRICA Data Coverage: • We have data for six bank client countries with IBRD/IDA-supported basic education projects. Tuition Fees: • For these six countries, tuition fees are implemented in four (32%) countries—Egypt, Jordan, Tunisia, Yemen—though only three (75%) countries have legal tuition fees— Jordan, Tunisia, Yemen.**** • Where fees are imposed, World Bank projects have introduced mitigating measures in one country—Yemen. • The World Bank is working to eliminate fees in one country—Jordan. • The World Bank does NOT promote tuition fees in any of the countries. Textbook Fees: • Textbook fees are imposed in three (50%) countries—Egypt, Lebanon, Tunisia—and in two (67%) countries, these fees are legally allowed—Lebanon, Tunisia. • Where fees are imposed, World Bank projects have introduced mitigating measures (either generalized or targeted) in one country—Lebanon. • The World Bank does NOT promote fees in any off these cases. Uniform Fees: • Purchase of mandatory uniforms is implemented in two (33%) of these cases—Egypt, Lebanon—and in one of these countries fees for uniforms are legally allowed—Lebanon. • Where fees are imposed, World Bank projects have introduced mitigating measures (either generalized or targeted) in two countries—Egypt, Lebanon. • The WB does NOT promote purchase of mandatory uniforms in any of these countries. PTA/Community Contributions: • PTA/community contributions are common in three (50%) of these countries—Egypt, Lebanon, Tunisia—and in all three countries, it is legally allowed. • World Bank projects have introduced mitigating measures (especially targeted to the poor) in one of these cases—Lebanon. 50 • The World Bank does NOT promote community contributions in any of these cases. Other School-based Fees: • Other school-based activity fees are collected in two (33%) countries—Egypt, Yemen— and in these fees are legally allowed in Egypt. • The World Bank has not addressed these types of fees in any of these countries. Note: *** All countries in MENA with the exception of Tunisia and Lebanon have constitutional and legal provisions for education: “Education is a right. State-provided free instruction at all levels.” 51 SOUTH ASIA Data Coverage: • We have data for seven bank client countries with IBRD/IDA-supported basic education projects. Tuition Fees: • For these seven countries, tuition fees are implemented in four (57%)countries—Bhutan, India, Nepal, Pakistan—and two (50%) have legal tuition fees—Bhutan, Pakistan. • Where fees are imposed, World Bank projects have introduced mitigating measures (either generalized or targeted) in one of these cases—Pakistan. Textbook Fees: • Textbook fees are imposed in four (57%) countries—India, Maldives, Nepal, Pakistan— and these fees are legally allowed except in India. • Where fees are imposed, World Bank projects have introduced mitigating measures (either generalized or targeted) in two (50%) of these cases—India, Pakistan. • WB does NOT promote fees in any off these cases. Uniform Fees: • Purchase of mandatory uniforms is implemented in three (43%) of these cases—Bhutan, Maldives, Pakistan—and these fees are legally allowed except in Bhutan. • The WB does NOT promote purchase of mandatory uniforms in any of these countries. PTA/Community Contributions: • PTA/community contributions are common in four (57%) of the countries surveyed— Bangladesh, Bhutan, Nepal, Pakistan—and these fees are legally allowed except in Bangladesh. • The World Bank promotes community contributions in one of these cases—Pakistan. Other School-based Fees: • Other school-based activity fees are collected in three (43%) of the countries— Bangladesh, Nepal, Pakistan—and these fees are legally allowed except in Bangladesh. 52 • Where fees are imposed, World Bank projects have introduced mitigating measures (either generalized or targeted) in one of these cases—Bangladesh. • The World Bank does not promote these types of fees in any of these countries. NOTE: There are no cases in South Asia where the World Bank has eliminated any type of fees. 53 Annex 5—The Case of Malawi: What Can Go Wrong with the Elimination of User Fees? “Simply making education free does not guarantee that everybody will go to school.” (Dr. Joseph Chimombo, Africa Eye News, 2/6/01) Like most low income countries, Malawi’s education system still faces significant challenges relating to poor quality, low internal efficiency inadequate infrastructure and inequality. Education opportunities are inequitably distributed by poverty (dropout rates are higher among the poorest), gender (dropout rates are higher among girls, female teachers, and thus effective role models for girls are mostly in urban centers), geographical regions and (enrolments higher in the north, lower in the south and center, pupil:teacher ratio higher in rural areas), disability (no systematic manner of catering for the disabled). In the 1980’s less than half of primary school-aged children were in school. By 1988/89 GER was 66.9% (59.4% female) and NER was 50.6% (48.3% female). In June of 1994, the newly elected democratic government announced that primary school fees will be abolished effective from October 1994. The declaration of the Free Primary Education (FPE) policy was part of the government’s effort to improve access to education, and part of the larger national strategy to alleviate poverty through economic growth and investment in human capital. As a result of the FPE policy, enrollments jumped by 51%—from nearly 1.9 (48% females) million in 1993/94 to nearly 3 million (47% females) pupils in 1994/5, including former dropouts who were encouraged to reenter the system [Source: Government of Malawi, (2000) Public Expenditure Review]. The GER reached 134.1 (127 females; 141.2 males) relative to an NER of 95.7 (95.7 females; 95.6 males) [Source: E. Kadzamira, M. Chibwana, M. Chatsika, and J. Khozi, (1999), Gender and Primary Schooling in Malawi. Lilongwe: Ministry of Education and Sussex: Institute of Development Studies]. This rise in enrolments necessitated a 75% increase to the cadre of serving teachers. To meet this demand about 20,000 untrained teachers were recruited (75% junior secondary school leavers and 25% senior secondary school leavers) and deployed after a three-week induction course. Their qualification and certification were to be attained through in-service training. Simultaneously, about 4000 retired teachers were recalled to service [Source: The World Bank, (1995) Staff Appraisal Report, Malawi Primary Education Project. Washington D.C. The World Bank]. The Teacher:pupil ratio hovers between 1:130 and 1:160. The declaration of free primary education in 1994 further undermined popular commitment to the school committees as parents believed and expected that that the government would now take over the full cost of primary education, and school maintenance would fall to district and central government. 54 In June of 2000, the government acknowledged the commitment of communities to the provision of education stating explicitly that “Local communities and parents are increasingly playing a role in educational finance especially with regard to sharing in the cost of buildings and their maintenance, transport to schools, food, uniforms, learning materials and even extra-curricular activities. Community participation is particularly significant at the primary level. More than 75% of Malawi’s primary schools have been built with the support of local communities. Primary school maintenance has largely been a responsibility of communities” [Source: Ministry of Education, Sports and Culture (2000) Education Sector Policy & Investment Framework (PIF) p. 4]. Bank supported projects have also encouraged community contribution particularly to infrastructure development. For instance, from 1996, communities have constructed 41% of all new classrooms through the Malawi Social Action Funds (MASAF) which requires a 20% community contribution from communities [Source: Government of Malawi, (2000) Public Expenditure Review]. Through another Bank project the Primary Education Project communities have contributed bricks and labor to the construction of over 400 classrooms (this effort is still ongoing and numbers are expected to increase). Although primary education is said to be free, “parents do still incur expenses for exercise books and writing materials, in order to supplement state supplies, which are often inadequate. In addition, there are other school expenses that parents are required to meet, such as sports fees, transport, and water fund. Frequently, parents are also called upon to contribute to school construction and maintenance by providing labor, materials or, sometimes even cash. For poorer households these extra expenses can be prohibitive, particularly if cash payments are required.” The Government of Malawi notes that “Without doubt, the most significant educational achievement of the past five years has been the introduction of free primary education (FPE) in 1994….Although this policy was not strategically planned and implemented, and despite the insufficient budgetary provision for its implementation, it was widely welcomed by the Malawians and the donor international communities as a step in the right direction” [ Source: Ministry of Education, Sports and Culture (2000) Education Sector Policy & Investment Framework (PIF) pp. 6-7]. Public recurrent spending on education more than tripled. Primary recurrent spending surged by 5.3 times in real terms between 1990/1 and 1994/5 and its share has increased from 45 to 71 percent of total recurrent spending. The bulk of the recent increases in primary education spending has been spent on teacher salaries. In 1994/5 recurrent spending on salaries consumed 97 percent of all primary recurrent expenditures. Although still high, this proportion has fallen to 89% on average. Of the remaining 11%, about 5% goes toward teaching and learning materials, 2% to rehabilitation of infrastructure and 4% to district administration, school advisory services, and the inspectorate. Government’s contribution to teaching and learning materials is less than half the UNESCO recommended norm of US$ 5.00. The highest recorded unit expenditure on rehabilitation was US$ 1:10, also below the norm which is 2% of the construction cost per annum. [Source: Government of Malawi, (2000) Public Expenditure Review]. 55 After the rapid surge in enrollment without a proportional increase in budgetary allocations for non-wage items, quality indicators deteriorated even further. Because of many years of under- funding for infrastructure and supplies, Malawi’s public primary schools are among the most poorly equipped in Africa. According to 1997 estimates, the ratio of pupils to permanent classroom was 119:1, pupils per desk 38:1, pupils per chair, 48:1, pupil per textbook (English, Math and Chichewa), 24:1, pupil : teacher 62:1. The percentage qualified teachers (a good part of whom are under-trained) was 51% [Source: Ministry of Education, Sports and Culture (2000) Education Sector Policy & Investment Framework (PIF)]. Thus, the quality of education which was already severely compromised by large class sizes even before the surge in primary enrollment, has deteriorated further. A popular though sometimes meaningless indicator ‘performance on PSLEs’ shows that although the pass rate dipped to 62% in 1996/96, it has since come up to over 80% in recent years. Dropout rates for grade 1 to 7 have declined from 24% in 1994/95 to 17% in 1999. Within the same period, repetition rate (for grades 1-8) has decreased from 27% in 1994/95 to 14%. [Source: Government of Malawi, (2000) Public Expenditure Review]. Although public spending has increased to 7.5% of the GDP in 1994/5, there continues to be a dearth of even the minimal number of essential teaching materials and human resources to educate children. Malawi’s education system relies heavily on donor support, particularly for infrastructure development and provision of resources. Estimates for 1999/00 show that above 90% of the development expenditure in education in general and in primary education specifically comes from loans and bilateral donations [Source: Government of Malawi, (2000) Public Expenditure Review]. Despite the improvements in enrollment, there are still serious problems and severe inequalities in public primary education. Policy changes have increased the availability of public funds at the primary level, (between 1993 and 2000, recurrent expenditure on education increased by 41% ) in absolute as well as relative terms, but public primary spending per student are the lowest; and the distribution of enrollments across socio-economic groups shows that the poor students have lower enrollments than richer students. Enrollment patterns between 1990/91 and 1997/98 show that the increase in enrolment rates has favored the poor. However, the dropout rates are higher among the poorer than the richer groups. For instance, for 1997/98, grade 1 to 4 gross enrolment rates were 156, 159, 160, 150, 138 for the first, second, third, fourth, and fifth quintiles, respectively. For grades 5 to 8 the rates were 61, 75, 78, 88, 97 for the same quintiles [Source: Government of Malawi, (2000) Public Expenditure Review]. The internal efficiency of education is so low mainly because of high dropout and repetition rates. It is estimated that it takes on average, 13 years for a child to complete an eight-year cycle of primary school. Nearly half of the children who enter primary school dropout before Standard 5, and are not functionally literate or numerate. 56 In 1994, over one million pupils enrolled in Grade One but the numbers dropped by 41% to 590, 167 in Grade Two the following year. Only 289, 251 pupils made it to Grade Five in 1999—a 71 % drop. Only 18% of all children enrolled in primary school complete the full eight years of primary education. Between 1998 and 1999 survival rate increased to about 30% (although still very high). Most disconcerting is that about a third of the learners drop out between first and second grade (even way before most can read and write). For those who complete the cycle, the average length of study for the 8 year cycle has somewhat improved from 9.0 yr. in 93/94 to 8.9 yr. in 1999. For dropouts, the average length of study has actually decreased from 3.0 yr. to 2.6 yr. within the same period. Bearing in mind that dropping out is more prevalent among the poor, this means that they leave school after less than the 5 years required to attain sustainable literacy. The average number of ‘pupil years’ per completer has also declined from 25.3 yr. to 15.6 yr. (albeit still unacceptably high and very costly). Out of the 1.3 million students who entered grade1 in 1994 , only 300,000 completed the cycle in 2002. SACMEQ results show that only 30% achieved minimum comprehension in Grade 5 and only 1% achieved adequate level. The Malawi example indicates that although a larger increase in budget share is being allocated to primary education, completion and progression rates remain low. The priority now is to restructure the composition of budgetary allocations within the sector to continue improving equity, to raise the quality of service delivery and to raise the internal efficiency and cost- effectiveness of the system. 57 Annex 6—The Case of Uganda: What Needs to be in Place to Support Elimination of User Fees? UGANDA Government policy priority since the early 1990s, reaffirmed by Government White Paper, that the state shall promote free and compulsory basic education. This paper outlined the government’s long-term vision for education reform and set the stage for major policy reforms. At the time, private costs to schooling were too high for most Ugandans to support. Study in mid-1990s showed that parents financed 60% of direct costs of schooling. In 1995 almost 60% of school-aged children were out of school. President Museveni announced a policy of UPE in late 1996, abolishing user fees for up to four children from each family (two of whom should be girls). All orphans would be exempt from fees. Textbooks are now provided free if charge to pupils and uniforms are a school board decision, but no child can be excluded from school for lack of uniform. The schools are forbidden to require parental contributions except in urban areas where they may ask for contributions from families to pay for electricity and municipal water. Schools may not exclude pupils for nonpayment (information provided by Paud Murphy). This policy led to more than a 70% increase in total enrollment overnight, from 3.1 million to 5.3. GER went from 73% to 124%. This created an education emergency that may have resulted in a backlash against the UPE policy and an unintended reduction in participation of the poor. There were concerns about deteriorating quality. Government initiated broad consultations with external donors and other stakeholders on the development of an investment and policy framework for education to close the financing gap. Resulted in highly productive donor coordination which stabilized the system and created conditions where educational quality could be meaningfully addressed. Given the projected increase in enrollment and the fact that primary education was already under funded before UPE, it was reckoned the budget would roughly require doubling over three years. This, at the exchange rate of time, translated into 284 million of which the Bank, with its 155 million, was financing 55%. Macro context: Poverty Eradication Action Plan (strengthening the poverty focus of public sector spending); Third Structural Adjustment Credit. Resulting in more efficient implementation of ongoing education sector operations. 58 High-risk rapid response: Education Sector Adjustment Credit (ESAC) providing a $75 million IDA grant within the context of HIPC and an IDA credit of $80 million. For budgetary support against an agreed action plan developed wholly by national teams. Goals: to clarify and disseminate the UPE policy; increase resources available and efficiency of their use; protect quality of teaching and learning; strengthen sector management. Resource availability and use: Government shifted more resources to primary education by reducing subsidies at tertiary level and increased the recurrent budget for primary education from 9% in 1996 to 19% in 1999. Learning materials were added as a recurrent expenditure line item in budget for the first time. Liberalized textbooks: To achieve higher textbook to pupil ratios, new policies for the provision, distribution and utilization of textbooks were adopted, allowing for greater choice among titles for schools. Teacher education: Extended the Teacher Development and Management System to all districts. Sector management: Strengthened teacher management capacity (timely payment, improved recruitment, eliminating ghost teachers, monitoring system for funds allocated to districts and schools. Increased transparency at school level: Government clarified roles of parental financial and in- kind contributions in the context of UPE policy to ensure that their participation would continue. Introduced a capitation grant system which school committee manages, for purchase of materials other than textbooks or teacher salaries. Grant is calculated centrally and released as a conditional block grant to districts, who in turn, release all funds to schools on the basis of enrollment. The ministry released guidelines to schools for allocation of funds by expenditure category. The amounts received from the district office are posted publicly in the school. Some schools publicly display expenditures, but anyone can ask to see the records of how the money is spent. Regular audits show increasing evidence that the funds do reach the schools and are utilized for the purposes intended (information provided by Paud Murphy, 1999). Enrollment results: Primary enrollment 1992 1999 Top income quintile, total 82% 85% Bottom income quintile, total 50% 84% Girls aged 9-12 59% 89% Boys aged 9-12 67% 91% Financial resources invested by external agencies in the period 1997-2000 total over US$373 million, using average annual exchange rates. The Recent Education Sector Adjustment Credit has provided the funding as part of HIPC for these recent changes in education. Half of this adjustment operation included a $75 million grant 59 which amounted to 65% of total expenditure. (73% of the total education budget is earmarked for basic education.) In conclusion, implementation of the Program has stayed mostly on track thanks to massive political commitment and sustained budget support. Education was the principal electoral platform of President Museveni in 1996, and the share of Education in the budget rose from 22% in 1995 to 31% in 1999. Yet, the number of teachers in basic education cannot keep up with the enormous increase in enrollments. Learning achievement in many schools remains very low and primary completion rate is only 61%. Budget increases to fund more teachers , build more classrooms, and ensure adequate instructional materials are a high priority. Resources released through the HIPC initiative will be used for these purposes. 60 Annex 7—The Case of Kenya 'Too Many Brains' Pack Kenya's Free Schools Lack of Teachers, Inadequate Funding Hamper Efforts to Fulfill President's Campaign Vow By Emily Wax Washington Post Foreign Service Thursday, October 9, 2003; Page A24 HOMA BAY, Kenya -- The tree was sprawling, its trunk thick, its shade cool. But those weren't the reasons why a cherub-faced teacher with oversized glasses and a weakness for teaching "Romeo and Juliet" in Swahili decided to move his class of 150 fifth-grade students outside. "We are so packed," said Ernest Otieno, the erudite head teacher of Asumbi Primary School, who sat after the lesson in his closet-sized dark office with no phone, electricity or windows in this rural school about 175 miles northwest of the capital, Nairobi. "We have too many brains and not enough buildings or books." Kenya is trying once again to institute free primary education. When the new president, Mwai Kibaki, announced in January that starting immediately school through grade 8 would be free, 1.5 million new pupils showed up in just the first week. Some were barefoot, some orphaned by HIV-AIDS and others just desperate to join more privileged friends. With them came 30-year- old parents enrolling in first-grade classes to learn to read and write. "The birthright of a child is good health and free primary education," Kibaki said during last year's presidential campaign. Kenya's previous two presidents, Jomo Kenyatta and Daniel arap Moi, also abolished fees for several months as part of campaign promises. Each time, the fees were reintroduced when the government said it could not fund universal education. Primary schools were charging between $20 and $350 a year per child, depending on the area of the country. There was a separate fee of around $30 for a uniform and books. Many families could afford to send only one child. Other nations in sub-Saharan Africa have launched similar efforts to make education free. Uganda's program started in 1996, but the schools are overcrowded and parents are being asked to pay for some elements, such as lunch. Tanzania made primary school free in 2001 and has been struggling to fund the program. Ghana and Ivory Coast have model universal public education systems that have existed since independence. "Can we at least start to train people with enough skills to read instructions and become a mechanic?" asked Otieno, a thoughtful man who carries a worn briefcase of woven straw. "Or can we dream of teaching so well that our students could write literature or perform a heart operation? We want to do it all. But we aren't ready just yet." For the past nine months, teachers, government officials, parents and students have struggled to set up public schools that do not have enough teachers and restrooms, let alone books. The 61 budget so far has been about half of what is needed, school officials report. The government has promised to double it next year. "I like my teacher," whispered a shy Kephas Goga, 16, who was just starting Asumbi Primary School in a class of 106 students. "But I have never had the delight to meet him. There are so many of us stepping in line to inquire after the lessons that we can't all try." Some teachers use megaphones as they give lessons, more like professors at state universities. At Cheleta Primary School in Nairobi, there are two teachers for 213 first-grade students. Kenya has 175,000 public primary school teachers but needs 60,000 more, according to the Kenyan National Union of Teachers. "What you notice mainly is that there are so many children that learning has turned into the lecture method. But, you know, at primary level, each child has to have some attention," said Joseph Chirchir, acting chairman of the teachers union. "And the lack of teachers is causing problems. Sometimes the students come prepared with books and stationery but there is no teacher." Lunch programs are also a problem, because the few schools that operate them do not have enough meals for the number of hungry children now arriving. With a stunning number of children orphaned by HIV-AIDS at schools such as Asumbi Primary, where there are 181 orphans among the 954 students, these programs are needed to make sure students have the energy to learn. In other schools, children as young as 13 attend only for a few hours because they must work in the fields. Teachers in Kenya's rural central regions found that many teenage orphans in first or second grade did not stop working just because school fees were abolished. "They now just troop back to the working fields after school and before," said Justus Muthoka, the central province children's officer. "It's a vicious cycle, because they may be able to learn how to read, but they also have to earn their daily bread. They are about to collapse." In his Nairobi office, Stephen Karaba, senior deputy director of primary education, said that building a proper school system eventually will help every level of society. "I think the government will have to continue to pour a lot of money in to get things up to where they should be," Karaba said. "Of course, when you try anything new, there will be clashes and problems. But I don't think anything happened that we didn't imagine coming. It was best to get it started when there was euphoria around the new president and then worry about the problems as we go." The 12 teachers at Cheleta Primary School in suburban Nairobi agree, but they also saw enrollment double to 874 students. 62 In the fourth-grade class, the room was so packed with desks and chairs that students had to climb over the furniture to reach a seat. The teacher had a tiny aisle in front of the blackboard where he could stand. He could not even walk to the back of the classroom if there was a problem. Philomena Mrurigi, a third-grade teacher, attempted to conduct a spelling lesson with 136 students. "Spell 'chair,' " she yelled out in English. Primary education is conducted in both English and Swahili. A sea of hands flew in the air. Because Mrurigi didn't know her students' names, she had to wind her way through the maze of desks and tap a student on the shoulder if she wanted an answer. Headmaster Stanley Mukururo, an enthusiastic man dressed in a dark blue suit and tie, said his biggest problems include a teacher shortage and hungry students. For some problems, he has found creative solutions. To feed students, teachers have offered bananas growing behind the school to those who answer questions correctly. He also encourages teachers with rousing speeches: "Not everyone will be sick or need hospital care," he said on a recent day. "But everyone needs education." Special correspondent Candice Miranda contributed to this report. 63 Annex 8—The Case of Tanzania Cost sharing was introduced in the mid-1990s. Fees were treated as a general sales tax on primary education. Fees were particularly onerous for the poor and although children could not officially be turned away from school for not paying the fee, de facto, they were turned away and no action was taken against this practice. The process leading to the elimination of fees on primary education was the result of the PRSP process, civil society activism and the support of the Government of Tanzania (GoT) and the donor community towards providing UPE. In the late 1990s, the GoT produced an Education Country Status Report, which provided the basis for an Education Sector Development Program (ESDP). The report found the education sector to be in a critical situation, with over 3 million eligible children out-of-school, low and declining enrollment and literacy rates, and high out-of-pocket expenses for parents. Donor positions were also evolving. In the fall of 2000, as the GoT was considering the elimination of user fees for primary education in the context of the PRSP process, the donor community was considering whether they would support and help such a decision, which would be accompanied by a sector-wide approach to education reform. Overall, the World Bank and other bilateral donors were supportive, having been strongly influenced by the Uganda experience, where the elimination of user fees had resulted in a rapid rise in enrollment rates. Ultimately, the elimination of user fees was included as an action for HIPC completion and as part of the final PRSP document. This measure was only financially-feasible with donor support. The GoT worked with the World Bank, other donors and the NGO community toward the preparation of the Primary Education Development Program, which was to be supported by a $150 million World Bank project loan. The PEDP also included a commitment by a number of bilateral donors to integrate their individual projects into the PEDP. Following the official abolition of user fees in 2001, Tanzania experienced a rapid increase in the net primary enrollment rate, from 57 percent to 85 percent by 2002. Tanzania’s planned elimination of user fees resulted in a less severe decline in quality as a result of strong government commitment to UPE, HIPC funding to close the financing gap, a US $10 capitation grant per pupil to meet non-salary recurrent costs and a development grant budgeted of US$3500 per classroom for capital expenditures, including teacher houses, sanitation and the provision of clean water. 64 Annex 9—China- The Mounting Expenses of Education A Crumbling Promise in China: Access to School Many urban families cannot afford the mounting expenses By Jiang Xueqin The Christian Science Monitor, January 22, 2002 JILIN, CHINA - In cold Jilin city in China’s rugged northeast, a thin boy named Li Zhonggang darts around a dusty alley. His mother scolds him from nearby as she helps an elderly couple repair a window for small change. It’s early afternoon, and as other 10-year-olds recite passages from textbooks, Li is simply wandering the streets. He is one face of a disturbing trend: city kids so destitute that they cannot go to school. One of socialist China’s proudest and most remarkable achievements was to provide basic education for most of its children, and even today the Ministry of Education’s guarantee of nine years of education to China’s 200 million children is considered its most important task. Although Chinese officials acknowledge that many rural children do not have access to schooling, they say that all city kids can go to school. A few years ago that may have been true, but not anymore. With the onslaught of the market economy, China’s iron rice bowl - its socialist guarantees of employment, housing, schooling, and medical care - is shattered. Beijing has been reducing education subsidies during the past few years, even as schools enroll more and more students. Today, China spends less than 3 percent of GDP on education, meaning that it is trying to educate one-quarter of the world’s students with 1 percent of the world’s education budget. Fifty students must cram into a tiny classroom. In many parts of urban China, schools are decrepit and dilapidated, and some cannot even pay teachers. Earlier this year, school officials in Jiangxi Province were found forcing students to make firecrackers in the school basement. To compensate for reduced government support, schools continually raise tuition, which the government endorses in an attempt to increase consumer spending. Beijingers now spend a total of $8,000 on a child’s K-12 education, according to a recent survey. In a time of rising urban unemployment, fewer and fewer families can afford this amount. Some, like the Li family, cannot afford even basic schooling. The Lis live in an industrial part of the country. Once called a worker’s paradise, the region’s outdated money-losing factories earn it the moniker “China’s rust-belt.” Statistics are kept secret, but an estimated one-third to one-half of Jilin’s workers are now unemployed. Li Ming, the boy’s father, is one of the hundreds of thousands of roaming unemployed. His family was doing well until the state-owned factory where he worked went bankrupt in 1996. Older employees retired early, but Mr. Li was in his early thirties and found himself at the mercy of the market. 65 Every morning, he goes to stand in one of the city’s “labor markets,” a street corner where gruff middle-age men wait for someone to employ them for the day. If he manages to find jobs, he can make $25 a month, just enough for food. That’s why young Li can’t go to school. “I can’t ask friends and relatives for money, because we can never pay them back,” says his mother, Li Hua. She has taught him some mathematics, but she herself has only an elementary-level education. The boy has yet to learn to read or write. With tears gently streaming down her face, his mother adds: “He can’t go to school, so he can’t find a job when he grows up. I feel guilty.” Liao Haimin, a professor specializing in K-12 education at Northeastern Normal University in Changchun, Jilin Province’s capital, says that parents are asked to contribute toward their children’s schooling because government funds are limited. “It’s not that much,” he says, “but there are a few instances when families can’t afford the tuition.” Some parents, of course, don’t agree that it’s “not that much.” “Schooling costs $75 a year, but miscellaneous expenses—textbooks, uniforms, trips—double that,” says Wang Dexun, a laid-off worker in Jilin. “That’s two months of paid wages for an employed worker—what about us laid- off workers?” Mr. Wang believes that the costs are ratcheted up because schools know that Chinese families value education greatly and will pay anything for their only child to attend. “Today, kids are really worried,” he continues. “On one hand, if they don’t study hard, they can’t get into university and get employed. On the other hand, if they do study hard, their parents may not be able to afford the high costs of senior high school and university.” The issue is sensitive in China, and there is no official recognition of this growing problem. But a visit to another “rust-belt” city, Shenyang, also illustrates the dire circumstances families face. There, two laid-off brothers are finding it difficult to send their children to school. Sun Baopu makes only $25 a month selling cookies on the streets, but he managed to scrape enough money together to send his 7-year-old son to first grade this year. His unemployed younger brother, Sun Baozhong, however, leaves his 7-year-old daughter at home. “If I find a job next year, then I can send my daughter to school,” he says. “If I can’t, then she’ll just stay at home.” 66 Annex 10—HIPC Figure on Public Spending 67 References Asian Development Bank. 2002. Education Policy Paper. Birdsall, Nancy. 1987. Demand for Primary Schooling in Rural Mali: Should User Fees be Increased?” Population, Health and Nutrition Department Technical Note. Washington DC: The World Bank. Bray, Mark. 1996. “Counting the Full Cost: Parental and Community Financing of Education in East Asia.” Washington DC: The World Bank. Bray, Mark. 2001. Government and Household Financing of Education: Finding Appropriate Balances. Paper presented to the International Conference on Economics of Education, Beijing University. Bray, Mark. 1999. “The Shadow Education System: Private Tutoring and its Implications for Planners.” UNESCO: IIEP. Community Support for Basic Education in Sub-Saharan Africa. 2001. Washington DC: The World Bank. Development Committee. Education for Dynamic Economies: Accelerating Progress Towards Education for All. Document DC2001-0025, September 18, 2001. Washington DC: The World Bank. DFID. 1999. Learning Opportunities for All: A Policy Framework for Education. Diop, Sakhevar. Draft 2001. World Bank Support for Provision of Textbooks in Sub-Saharan Africa. Washington DC: The World Bank. Education Sector Policy Paper. April 1980. Washington DC: The World Bank. Education Sector Strategy Paper. 1999. Washington DC: The World Bank. Filmer, Deon. 2001. “The Structure of Social Disparities in Education: Gender and Wealth.” Policy Research Working Paper. Washington DC: The World Bank. Filmer, Deon and Pritchett, Lant. 1998. “The Effect of Household Wealth on Educational Attainment.” Policy Research Working Paper. Washington DC: The World Bank.. Heller, Peter S. and Keller, Christian. Social Sector Reform in Transition Countries. IMF Working Paper WP/01/35.2001. Hinchliffe, Keith. 2003. “Draft Notes on the Impact of the HIPC Initiative on Education and Health Public Expenditures in African Countries.” Washington DC: The World Bank. 68 I Terme, Rosa Alonso. 2002. “The Elimination of User Fees for Primary Education in Tanzania.” Washington DC: The World Bank. Jiminez, Emmanuel. 1987. Pricing Policies in the Social Sectors. Washington DC: The World Bank. Primary Education Policy Paper. 1990. Washington DC: The World Bank. Priorities and Strategies for Education. 1995. Washington DC: The World Bank. UNICEF. 1998. Addis Ababa Consensus on Principles of Cost Sharing in Education and Health. Update on the Kyrygz Republic. 1999. Washington DC: The World Bank. World Bank. Engendering Development. 2001. World Bank. Making Transition Work for Everyone. 2000. World Bank Policy Research Report. 2001. “Engendering Development.” Washington DC: The World Bank. World Development Indicators 2001. The World Bank. 69