*partie=titre Education Charges. A Tax on Human Development *partie=nil New research illustrates how education charges trap children in a cycle of illiteracy and poverty. Governments and international organizations must act immediately on their commitments to deliver free basic education for all. November 12 2001 Executive summary *partie=nil ‘No countries seriously committed to education for all will be thwarted in their achievements of this goal by lack of resources.’ G8 Communiqué, Okinawa, 2000. ‘How can I afford to send my children to school when I can’t afford to feed and clothe them? I know how important education is. But the school fees and books are beyond my means.’ Tatu Huseyni, 35-year old coffee farmer in Kilimanjaro Region of Tanzania, 15 December 2000. Half a century ago, the Universal Declaration of Human Rights established free basic education as a fundamental human right. Yet on current trends there will be 75 million children out of school in 2015. Governments and international organizations have paid lip-service to the idea that basic education should be free; they have done precious little to address it in reality. In the absence of sufficient public finance, the cost of education is being transferred to poor families as part of a creeping privatisation of education financing. Households face a bewildering array of education charges, from direct school fees to indirect costs for books, pencils, and uniforms. Parents consistently cite cost as the major factor in deciding to keep children out of school. Education charges also magnify other problems: for example, where a low value is already ascribed to girls’ education, gender inequalities are likely to be reinforced by high costs, with parents reluctant to pay for their daughters’ schooling. In almost every developing country it costs the poorest 40 per cent of the population over 10 per cent of annual income to send two children to primary school . The equivalent amount for the average family in the USA would be more than US$4000. Educating children, particularly girls, helps families earn more income, stay healthier, and ultimately break out of the cycle of poverty. That is why education is the foundation for human development. Delivering universal basic education would be a key step towards creating a global economy whose benefits are shared more widely. Yet millions of children are being consigned to a life of poverty because their parents are too poor to send them to school. Education charges amount to a tax on school attendance and human development, denying millions of children the chance to escape from the poverty trap. New research commissioned by Oxfam in Tanzania, Ghana, and Zambia illustrates how these charges are denying children a basic education. Education has become a privilege ordinary people simply cannot afford. In Ghana, sending just one child to primary school for a year can cost the equivalent of one month’s earnings from agricultural labour. In Zambia, between one-half and three-quarters of total education spending at primary level is now met directly by parents. When interviewed, parents said that education charges frequently force them to make impossible choices, such as whether to buy basic medicines or keep their children in school. In Tanzania, schools are often in poor repair; children sit on the floor in overcrowded classes, sharing the few books available. Poor families may not see the value of spending their meagre resources on such sub-standard education. In northern Tanzania the collapse of international coffee prices means the cost of education is too great for many households to meet. The evidence is indisputable. Success in achieving universal basic education depends on education becoming affordable to the poor, and this requires the abolition of education charges. When Malawi removed education fees in 1994 the school population almost doubled in one year, illustrating a huge pent-up demand for education. But abolishing user fees requires a major increase in financing as well as coherent strategies to improve the quality of education and manage the surge in demand that will inevitably follow. Many poor countries lack the resources to deliver such a strategy, and will require external help. The international community will need to improve on its grossly inadequate record. •Many developing countries are spending less then two per cent of GDP on primary education. Developing countries should produce national Education For All (EFA) plans by September 2002, and should cover half the additional cost of achieving free and universal basic education. These plans must be firmly rooted in wider poverty reduction strategies. •Rich countries have so far failed to deliver on their pledge that “no countries seriously committed to education for all will be thwarted in their achievements of this goal by lack of resources”. Inadequate debt relief is denying children an education. Of the first 23 countries to receive debt relief, over half will spend more on debt than on primary education in 2001. Less than two per cent of international aid goes to basic education. This is two per cent of a shrinking pot. Aid has tumbled to 0.22 per cent of donor GNP. •The World Bank’s financing of education fell from an average of US$2bn per year in the 1990s to US$1bn per year for the last two years. The IMF continues to set public-spending targets without reference to the need for governments to finance free basic education. Oxfam is part of the Global Campaign for Education, which is calling on developing-country governments, rich-country governments, and international institutions finally to deliver on the promise they made two years ago to launch a ‘global initiative’ which will achieve free basic education for all. The World Bank has an important role to play in developing a global initiative. It should: 1)Work with governments to ensure that all PRSPs include strategies to guarantee that every EFA goal is met, and to eliminate all charges for basic eduction within 3 years. 2)Work with governments to identify the financing gap for basic education within national EFA plans. 3)Publish, with Unesco, in April 2002 a summary of individual country ledgers tracking progress against EFA indicators, progress on national EFA plans, the extent of the basic eduction financing gap in individual countries, and the extent and nature of donor support. 4)Deliver deeper debt relief to countries with strong national EFA plans. *{Introduction} In many countries, education has become a privilege for which households must pay. However, mass poverty means that many parents cannot afford to send their children to school. Education charges take a variety of forms. They range from direct charges levied by national and district governments or schools, to indirect charges, such as spending on uniforms and pencils and informal levies imposed by teachers. Cost-sharing is a catch-all phrase that describes the distribution of education costs between parents and the public purse. In reality, though, it is a euphemism for the privatisation of education financing, and a tax on school attendance. There are many country-specific reasons for the high cost of education to poor households, but one near-universal factor is inadequate public finance. This briefing paper draws on evidence from household interviews carried out in three countries in sub-Saharan Africa – Tanzania, Zambia, and Ghana – which highlight the impact of education charges on poor households. In different ways, each country illustrates the depth of the crisis facing sub-Saharan Africa, the developing region that is furthest off-track for achieving the 2015 target of universal primary education. However, the devastating interaction between household poverty and rising education costs is apparent across the developing world, hampering efforts to accelerate growth and reduce poverty, and exacerbating problems such as child labour. Public financing is not the only issue, but it magnifies the other factors that are obstructing more rapid progress towards the 2015 target, such as gender discrimination, poor quality and inappropriate education, and unaccountable bureaucracies. In many developing countries, education infrastructure has been ravaged by years of under-investment. All too often, children are being taught in crumbling buildings lacking even the most rudimentary teaching materials. Where the quality of education is inadequate, or is perceived as generating few benefits, poor households are unlikely to spend their meagre resources financing it. It is overwhelmingly the children of the poor who account for those out of school, although there are also powerful country-specific sources of inequality based on gender, race, region, and rural-urban divides. Education charges interact with these other factors: for example, where a low value is ascribed to girls’ education, charges are likely to reinforce gender inequalities in levels of attendance and completion. In countries such as Brazil and Bolivia, more than one-quarter of adolescents aged 13-17 are in full-time work, the vast majority of them having dropped out of school. One of the reasons for high drop-out rates is the need for children from poor households to contribute to the family’s income. One review of nine Latin American countries found that without income from working children, poverty rates would rise by 10-20 per cent. But the imperative to generate household income denies children the educational opportunities that could provide an escape route from poverty. Education charges thus add to the existing opportunity costs (in the form of lost income and labour) for poor households in sending children to school. *partie=titre The cost of education: new evidence from case studies *partie=nil This section of the paper summarises some of the results of recent research in Tanzania, Zambia, and Ghana. Oxfam partners led the research in each country: in Tanzania, the lead agency was Maarifa ni Ufunguo, an education NGO based in Arusha; in Ghana, the research in the Northern Region was carried out by ISODEC; in Zambia, it was managed by the Jesuit Centre for Theological Reflection, Lusaka. The three countries are, in many ways, a microcosm of what has happened to education in Africa. All three countries made rapid progress after independence towards universal primary education, extending the education infrastructure. During the 1980s, progress first stalled and then went into reverse under the weight of economic collapse and debt. As the state’s capacity to maintain the education infrastructure declined, the quality of education also fell, and the burden of education financing was shifted to households. It is difficult to estimate with any precision the share of total spending on primary education now being met through out-of-pocket payments by households. In Tanzania, research by the World Bank suggests that just under one-third of all costs are met through household payments. In Zambia, the proportion is substantially higher: between one-half and three-quarters of total education spending at primary level is met directly by parents. This is one of the highest levels in the developing world. What is beyond dispute is that – as in other countries – the cost of sending children to school is exceptionally high in relation to the income of the poor. This simple fact has often been obscured by data suggesting that education costs represent a small share of average income. For poor households, as the summaries below suggest, education charges can take up a large proportion of income – in some cases, more than one-quarter of monthly income. Moreover, maintaining education payments is particularly difficult for poor households given the seasonality of income flows, vulnerability related to climatic conditions and poor health, and large average family sizes. It should be stressed that the problems posed by education charges are not restricted to Africa. In almost every developing country it costs the poorest 40 per cent of the population more than 10 per cent of annual income to send two children to primary school. The Public Report on Basic Education in India, which covered 234 villages in six states, found that the average cost of sending a child to school was equivalent to 40 days’ wages for a farm labourer. Research by Oxfam in Vietnam found that it cost households in the poorest fifth of the population one-quarter of their income to pay for one year’s primary education. Education charges increase inequalities across the developing world. At the school level, reliance on charging parents for non-salary items widens gaps in provision between rich and poor communities. The same is true at a household level. Education charges can leave parents with a choice between keeping children in school, or providing adequate food and meeting health costs. They can also result in household debt or distress sales of assets, thus increasing household vulnerability. *partie=titre Tanzania *partie=nil Since independence, Tanzania's education performance has resembled a roller-coaster ride. In 1961, well under half of primary school age children were in school. By 1976, the country was on the verge of universal primary education. But by the mid-1990s, the system was in free-fall, with enrolment rates below 70 per cent. Today, an estimated 1.7 million children are out of school, and the country has one of the lowest transition rates from primary to secondary education in the world. Research in Tanzania focused on the Kilimanjaro. Coffee is the mainstay of the local economy, with most households inter-cropping coffee trees with bananas, yams, cassava, and vegetables. Poor households supplement their income from coffee with agricultural labour, cutting grass for fodder or tending the animals of wealthier households. While this is a region with some of the best education indicators in Tanzania, there are extreme differences, and the situation is worsening. Coffee prices received by farmers have halved in the past two years, in line with international prices. This has had devastating consequences for their main source of income, and thus the principal source of education funding. Umati primary school, located on the lush coffee-growing foothills of Mount Kilimanjaro in Hai District, was one of the research sites for the Oxfam/Maarifa study. Its classrooms are made of crumbling mud, some with gaping holes in the roof. Children sit on the floor in overcrowded classes, sharing the small number of books available. Most lack a simple pencil or chalk slate. Teachers complain that they have to work without even the most rudimentary learning aids. Yet the children in school regard themselves as the lucky ones. Visit any village in Hai District on a school day and it's easy to see why. Many children are working in fields minding goats or cutting grass for fodder instead of sitting in classrooms. Their parents know the value of education. The Chagga people living in the area traditionally see education as a source not just of wealth, but also of status and opportunity. The problem is that many are unable to meet the costs of sending their children to school. Interviews with parents in the villages covered by the research indicate that sending a single child to school costs between US$8 and $16 per year, depending on the grade. For poor households this is a very large sum, equivalent to between one and two months’ agricultural wages, or about 40-60kg of coffee production at prevailing prices; the latter would represent the entire crop grown on a typical one-acre plot. Given that many poor households have three or more children eligible for primary education, the costs are prohibitive. The collapse in coffee prices during 2000 had a devastating impact on education in the area. Many parents reported having to keep their children at home. Teachers confirmed the scale of the problem. At Ngumeni primary school in Moshi Rural District, the head-teacher estimated that at least one-third of children were not in school. Non-payment of fees was identified as the main reason for non-attendance, by parents and teachers alike. The situation of Mzee Tingisha and Mama Regunda's household (see box) was not untypical in reflecting the efforts made by poor people to keep their children in school. In order to send children to school in Tanzania, parents must pay a bewildering array of charges. These range from direct school fees paid to the local education authority or to the school (for exams, repairs, sports days, and food) to indirect fees paid for books, pencils, and uniforms. Table 1 shows that the official government education fee accounts for a small proportion of total costs, while indirect fees represent more than half of the total. Uniforms account for almost half of the cost of sending a child to primary school. In their efforts to finance education, households reported being faced with invidious choices. In Uru North, Ernest Moki, a coffee farmer with seven children, told researchers that he had recently made urgently needed investments to repair his house, which was close to collapse. This had made it impossible for him to meet school charges. As a result, two of his children, both attending grade 3 of the local primary school, had been turned away. Other families said that they were frequently forced to withdraw children from school during July and August. This is the pre-harvest period, when poor households have depleted their food stocks and have limited income. Malaria and kwashiorkor are serious problems at this time, which means that poorer families face rising health costs at a time when they can least afford to meet them. During focus group discussions, parents in each of the three research sites said that they were often forced to choose between health and nutrition or education for their children. The effects of prohibitive education costs were reinforced by other factors. Parents complained that they did not know what schools were spending their money on. None of the head teachers interviewed were able to provide details of school income and expenditure. Levels of trust between parents on the one side, and teachers and district education officials on the other, were low. In Ngumeni, parents accused teachers of using children to work on their plots. In Umati, girls complained that they spent much of the time that they should have been learning collecting water, or working in the kitchen. Many children described brutal levels of physical punishment, including beating on the feet. Such problems highlight the pressing non-financial elements of education reform in Tanzania, including more accountability on the part of education providers to the local community. *partie=titre Ghana *partie=nil Education indicators in Ghana started to improve during the 1990s, but very slowly, and from a low base. Primary school enrolment has recovered to 85 per cent, although drop-out rates are high and quality indicators poor. Only three per cent of children leaving school have a solid grasp of mathematics. One recent study found that no teaching was taking place in one-fifth of schools visited during official hours. Moreover, there are extreme national inequalities, with the Northern Region in particular performing badly. The research in Ghana, undertaken with ISODEC, a local NGO, covered four sites in the districts of Tolon-Kumbungu and Savelugu in the Northern Region. This is one of the poorest parts of the country, where more than 70 per cent of people live below the poverty line. It also has some of the worst education indicators. One-third of the primary age cohort is not enrolled. There is a 10 per cent gap between enrolment rates for girls and boys, reflecting the low value attached to girls' education. Most people’s livelihoods are based on smallholder farming, the main crops being rice, beans, cassava, and maize. Almost all farming is rain-fed, which means that production fluctuates widely, and income flows are uneven and unpredictable. In the village of Manguli, lessons were taking place under trees and elephant-grass shelters, the school buildings having collapsed during the rainy season. Since lessons had to be taken outside, no teaching could take place during the rainy season between May and October. Parents expressed widespread discontent with the quality of education, echoing many of the concerns voiced in Tanzania. Schools lacked basic resources such as blackboards, textbooks, chalk, and paper. Teacher absenteeism was endemic. Unsurprisingly, educational outcomes were poor, and the perceived benefits of sending children to school were low: many parents were unwilling to incur the financial costs or the loss of labour involved. The interviews demonstrated the low value attributed to girls’ education and the importance of female labour, with fathers frequently describing the education of their daughters as a matter for their wives. In Gbanyamni, young girls of primary-school age reported walking up to four miles a day during the dry season to collect water. In terms of education charges, the household interviews revealed a state of chaos. In Gbanyamni and Manguli villages, most parents described school fees as low and collection as haphazard. However, both schools were perceived as having ceased to function in any meaningful sense as places of learning. In Tolon, parents reported that the head-teacher insisted on children wearing uniforms, which cost between US$3-5, depending on the grade. Poor households cited this as a reason for their children not attending school. In Nawani, where fees were collected, it cost over US$10 a year to send a child to school. This represents 10 per cent of average non-food spending for a household in the Northern Region, or one month's agricultural labour. In some cases children were only able to attend school by working outside school hours. Dora Kwame, a thirteen-year-old girl attending grade 4 of Nawuni School, pays for her own fees and uniform by selling boiled eggs in the local market. Interviews revealed that the ability of poor households to meet these charges was compromised by the cost of sickness episodes. During the rainy season, virtually every household reported episodes of malaria, with a single episode costing up to US$3 for treatment. Several households reported the choice they face between providing food, protecting health, or keeping children in school. Given that the malaria season in June and July coincides with the pre-harvest period, when household incomes are at their lowest, many families are forced to resort to distress sales. The case of Abena, a fourteen-year-old girl who dropped out of Nawuni School in grade 5 to care for her mother, is typical. *partie=titre Zambia *partie=nil After independence Zambia was widely seen as a model for others to follow in education. The country achieved major advances in school enrolment and literacy over a relatively short space of time. However, since the 1980s the picture has been one of almost continual decline. Today, more than half a million children – one-quarter of the primary age cohort – are outside the system. Parents cite four key concerns: education is unaffordable, the quality of provision is poor, little effective learning takes place, and primary schooling does little to advance employment prospects. Oxfam's research in Zambia concentrated on two sites: the Lundazi District of Eastern Province, bordering Malawi, and Mufulira, a peri-urban centre on the Copperbelt. Livelihood patterns vary widely. In Lundazi, most families work on small plots growing sorghum and maize for household consumption, and sunflowers, cotton, and groundnuts for cash. This is a rain-fed area that suffers periodic drought. Poverty rates are high, and literacy rates less than 40 per cent – the lowest in Zambia. In Mufulira, the collapse of the copper industry has left households dependent on a range of informal market activities, such as the production and sale of charcoal and dried fish. Income poverty rose by 15 per cent in the 1990s. While the problems faced by poor households in the two sites differed in their detail, cost was a common problem. In Mufulira, parents with children in Chibolya basic school estimated the average cost of sending one child to primary school as being 131,000 kwacha (US$37). This represented one-fifth of average household income on the Copperbelt. More than 10 per cent of children fail to sit exams in grade 7 of primary school, citing the reasons as being inability to pay. In Lundazi, the primary school enrolment rate is 60 per cent; less than half of children complete a full primary cycle. Here, too, focus group discussions highlighted cost as a major barrier. On average, rural households are spending US$2 per child per month to send a child to school. Given that more than three-quarters of the rural population are struggling to survive on less than US$1 a day, this is an enormous financial burden. *partie=titre Financing strategies for achieving free and universal basic education *partie=nil The case study findings from Tanzania, Ghana, and Zambia raise fundamental questions about the practice of charging for education. For many households, education has become an unaffordable luxury. Some children are being kept out of school altogether, while others are dropping out before completing their primary education. Parents are being forced to make choices about which child to send to school. The overall effect is devastating, not just for the children who are being denied an escape route from poverty, but for society at large. Mass exclusion from education is bad for economic growth, because of its impact on productivity and investment, it is bad for public health, and it is bad for democracy. In effect, charging for education amounts to a tax on human development. Evidence from a number of countries has confirmed the link between education charges and education demand. When Malawi's first democratically elected government removed education fees in 1994, the school population almost doubled in one year, as an additional 1.4 million children entered the primary school system. In Uganda, the withdrawal of fees at the start of the 1997/1998 school year resulted in an increased enrolment of nearly three million children. In both countries, the move to free education was accompanied by serious problems. Classrooms became seriously over-crowded, teachers and textbooks were in chronically short supply, and, initially, schools lost an important source of revenue. But the experience demonstrated the enormous pent-up demand for education among poor parents, and the devastating impact of education charges on their children's prospects. *partie=titre Moves towards free education *partie=nil In principle, most governments in sub-Saharan Africa and other developing regions are now committed to free education, at least at the primary level. The same is true of the World Bank, the IMF, and most Northern donors. In practice, however, little effort has been made to develop coherent financing strategies for making the transition to free education. The elimination of charges on parents has two key implications. First, it reduces revenue, especially at the school level. In the absence of compensatory transfers from government, this will diminish the resource envelope available for school maintenance and, crucially, teachers’ salaries. Second, abolishing charges can be expected to increase demand for education and the size of the primary school population, as in Malawi and Uganda. Other things being equal, a smaller resource envelope and a larger school population translates into lower levels of spending per student. In a worst-case scenario, the provision of free education can be an empty gesture. Parents, especially the poorest, are unlikely to send their children into education systems lacking even the most rudimentary requirements for effective learning. Even if they do, universal access to bad quality education will do little to enhance human development. It follows that the transition to free education needs to be part of a coherent strategy for raising the overall quality of education, improving service delivery, and transferring the burden of financing away from poor households to government. Countries such as Malawi and Uganda have been able to sustain progress towards universal education not just because they have removed fees, but because they have also dramatically increased their efforts to finance education through well-defined national education plans. Unfortunately, the same is not true of the three countries covered in this study. While Malawi and Uganda spend between three and four per cent of their GDP on primary education, Tanzania, Zambia, and Ghana spend less than two per cent. National commitments to 'free' education can be largely meaningless gestures, which owe more to populist politics than a commitment to education for all. If school fees are withdrawn in the absence of increased government financing, schools will inevitably resort to informal fees to compensate for the loss of revenue. There are some signs in Tanzania that this may be about to change. In early 2001, the Government of Tanzania adopted a Poverty Reduction Strategy Paper (PRSP), agreed with the World Bank and the IMF, committing it to phasing out education charges. The new policy is reflected in its Education Sector Development Programme (ESDP), which aims to achieve universal intake in grade 1 by 2005. Under the new policy, the abolition of fees will be offset by a US$6 grant per pupil provided to schools. However, the research carried out in Kilimanjaro suggests that this sum is inadequate, because the fee withdrawal proposal applies only to formal charges, and not to the wide range of informal costs faced by parents. Failure to address the latter will leave a substantial cost barrier to the education system still in place, undermining the Tanzanian Government's efforts to enrol another 2.5 million children in school by 2005. This is not the only problem. The ESDP in Tanzania has been costed at US$969m over the three-year period ending in 2004. The Tanzanian Government is meeting just over half of the cost, with an International Development Association loan from the World Bank covering another US$150m. This leaves a financing gap of US$92m per year to be met through donor contributions. The UK’s Department for International Development (DFID) has earmarked funds to cover two-thirds of the gap in the first year, but after that there is a large hole in the education-financing plan. If, as is distinctly possible, Tanzania's growth rate and/or revenue collection performance falls short of that projected by the World Bank, this hole will increase. The outlook for Zambia is worse. The Government’s education strategy aims to achieve free universal primary education by 2005. It includes a plan for constructing almost 13,000 classrooms at a cost of US$140m. However, the plan is woefully off-track, and undermined by chronic under-funding, weak implementation, and disputes between donors and government. More seriously, no attempt has been made to factor in the cost of transferring to public finances the burden currently borne by households. *partie=titre The global education financing gap *partie=nil The evidence is indisputable. Success in achieving universal basic education depends on education becoming affordable to the poor, and this requires the abolition of education charges. In turn, this requires that governments provide adequate financing, not just to get children into school, but to raise the quality of education once they get there. Almost two years have passed since the World Education Forum in Dakar. At that meeting, governments in developing countries agreed to draw up plans to meet the goal of ‘free and compulsory education of good quality’ by 2015. Rich countries pledged to ensure that no such plan would fail for want of finance. Two successive Group of Seven summits and two rounds of IMF-World Bank meetings have reaffirmed the promise made at Dakar to the world’s children. But the record on delivery is one of scandalous under-achievement. Fifteen countries in sub-Saharan Africa cut primary education spending in the 1990s, raising serious questions about political commitment to universal education. Many governments continue to skew education budgets towards higher-income groups. Even with increased national commitments, it is clear that many developing countries, especially in Africa, will be unable to finance universal primary education from domestic resources alone. However, aid flows fell in 2001 to their lowest level in a decade, stagnating at 0.22 per cent of donor GNP (table 2). Moreover, the share of aid allocated to basic education is exceptionally low. At the end of the 1990s it was equivalent to 1.7 per cent of total development assistance – less than US$1bn. Many donors, including Germany, France, Spain, and Italy, spend less than 20 per cent of their aid for education on basic education. Similar spending patterns by African governments would be regarded by the donor community as totally unacceptable. Debt relief is inadequate. Sixteen of the 23 countries now receiving debt relief under the Heavily Indebted Poor Countries (HIPC) Initiative are still spending more than 10 per cent of government revenue on debt servicing. Ten of these countries are spending more on debt repayments than on health and basic education combined. Zambia spends three times as much on debt servicing as it does on primary education. In Tanzania, the education-financing gap is approximately US$68m per year over the next three years, according to the Government’s strategy for achieving universal primary education. It will be spending more than twice this amount on debt servicing. The World Bank’s financing of education has fallen from an average of US$2bn per year in the 1990s to US$1bn per year for the last two years. Meanwhile, the IMF continues to set public-spending targets without reference to the need for governments to finance free basic education. According to the UN, the additional annual financing required to achieve universal basic education by 2015 is approximately US$9bn. But this does not take into account the need to end charges for basic education. Phasing out household financing for children currently in school over a three-year period would cost approximately US$500-600m per year, both in Africa and South Asia. The annual cost of extending free education to all children by 2015 might add another US$400-500m a year. The UN figure also excludes programmes that aim to increase demand for education. These are vital to create the incentives necessary to get children from poor households, especially girls, into the education system. They include school feeding programmes, bursaries for girls, incentives for working children, and special support for the disabled. In the Brazilian capital of Brasilia, for example, working families or the unemployed are guaranteed US$128 per month, provided that their children are in regular school attendance. The programme now extends to over 40,000 children, and has significantly reduced drop-out rates. In the Indian state of Andhra Pradesh, school feeding programmes have been shown to have similar effects, with additional benefits for child nutrition. And in Bangladesh and Pakistan, small bursaries targeted at girls have helped to address gender differences in enrolment. If the UN figure is adjusted to take into account the need to expand these kinds of programmes and eliminate education charges, the annual funding requirement for achieving universal education by 2015 might rise to US$12bn. While this figure may seem large, it is dwarfed by the US$109bn that would be released if the developed countries met their target of 0.7 per cent of GNP on aid. *partie=titre A ‘Global Initiative’ for free and universal basic education *partie=nil At the World Education Forum in April 2000, the international community agreed to launch ‘with immediate effect a global initiative aimed at developing strategies and mobilising the resources needed to provide effective support to national efforts’. Since Dakar, there has been no progress towards launching such an initiative. World leaders should act now on their commitment to a global initiative to deliver free and universal basic education for all. The ‘Global Initiative’ should have four key components. *partie=titre National Education Action Plans *partie=nil There can be no progress in achieving free and universal basic education unless developing countries make it a national priority. Governments are committed to producing National Education For All plans by September 2002. These should be clearly linked to national poverty reduction and development planning, and embedded within a medium-term expenditure framework. Countries must improve the effectiveness of spending and delivery. There should be a strong focus on quality, and on boosting education demand among girls, poor households, and marginalized groups. *partie=titre Public ledgers tracking progress *partie=nil UNESCO and the World Bank, along with other EFA stakeholders, should publish a regular global compilation of national EFA ledgers, starting in April 2002. Ledgers would track EFA progress in individual countries, using data from the UNESCO Institute. They would track the development and endorsement of national plans, country-level finance gaps using World Bank data and analysis, and the expansion of contributions towards filling the financing gap. *partie=titre A global financing framework for achieving free and universal basic education *partie=nil The international aid effort suffers from a combination of chronic under-financing, weak co-ordination, and inadequate targeting. International action is needed to close the financing gap for basic education, and to provide capacity building and technical support. The financing framework would co-ordinate increased bilateral and multilateral assistance. It would broker additional resources so that ‘no countries seriously committed to education for all will be thwarted in their achievement of this goal by lack of resources’. What is unique about the education sector is that donors have committed themselves to meeting the financing gap in full, where accompanied by a good plan. Rich countries should aim to cover around half the cost of achieving free and universal basic education. The IMF will need to ensure maximum flexibility in macro-economic targets so that the 2015 goals can be met. Donors must deliver deeper debt relief to countries where debt servicing is undermining education spending. *partie=titre An international effort to eliminate school fees for basic education *partie=nil National EFA plans should include the cost of getting all children into school, where such estimates are based not on existing costs per child, but the additional costs of eliminating all formal and non-formal fees for all children. This analysis should be an integral part of costing PRSPs, and for mobilising donor finance in support of PRSPs. *partie=titre What the World Bank should do *partie=nil The World Bank could play an important role (in close collaboration with UNESCO, governments, and civil society organisations) in developing a global initiative. It should: 1)Work with governments to ensure that all PRSPs include strategies to guarantee that every EFA goal is met, and to eliminate all charges for basic eduction within 3 years. 2)Work with governments to identify the financing gap for basic education within national EFA plans. 3)Publish, with Unesco, in April 2002 a summary of individual country ledgers tracking progress against EFA indicators, progress on national EFA plans, the extent of the basic eduction financing gap in individual countries, and the extent and nature of donor support. 4)Deliver deeper debt relief to countries with strong national EFA plans.