Financing Education for All by 2015: Simulations for 33 African Countries Alain Mingat Ramahatra Rakotomalala Jee-Peng Tan Africa Region The World Bank © June 2002 Human Development Sector Africa Region The World Bank The views expressed herein are those of the authors and do not necessarily reflect the opinion or policies of the World Bank or any of its affiliated organization. Cover photo by P. Waeles, courtesy of UNESCO Photobank. Cover design by Word Express, Inc. Contents Forward ................................................................................................................................ v Acknowledgments............................................................................................................. vii Abstract ............................................................................................................................... ix 1 Introduction ......................................................................................................................... 1 2 Approaches to EFA Costing............................................................................................... 3 3 Operationalizing the EFA Concept .................................................................................... 5 4 The Simulation Model ......................................................................................................... 7 5 Initial Conditions and Policy Targets toward EFA by 2015 .......................................... 11 6 Simulation Results ............................................................................................................ 19 7 Conclusion ......................................................................................................................... 27 Annexes Annex 1. Comparison with Estimates from Other EFA Costing Estimates: Supplementary Figure and Tables .................................................................................................................................. 29 Annex 2. Supplementary Figure and Tables ............................................................................................. 31 Annex 3. The Simulation Model ............................................................................................................... 39 Annex 4. Incorporating AIDS-related Costs in the Simulations .............................................................. 41 Annex 5. Africa Region Aggregate EFA Costing Simulation Results and Detailed Country-specific Estimates, 33 Countries ............................................................................................................ 43 List of Tables Table 5.1. Range, means and targets for parameters affecting enrollments in publicly funded primary schools, 1999/2000 and 2015 ................................................................................................... 12 Table 5.2. Range, means and targets for parameters affecting service delivery costs in publicly funded primary schools, 1999/2000 and 2015 ..................................................................................... 13 Table 5.3. Sample range and means of variables affecting domestic resources mobilized for primary education, circa 1999/2000 ....................................................................................................... 16 Table 5.4. ndicative targets for simulation parameters affecting the volume of domestic resources mobilized for primary education............................................................................................... 17 Table 6.1. Assumptions underlying six alternative simulation scenarios in the EFA costing exercise...... 20 Table 6.2. EFA 2015 financing gap under alternative policy measures, Niger ......................................... 21 Table 6.3. EFA 2015 cost estimates and sources of financing under "good practice" policies, Niger ....... 22 Table 6.4. Annual EFA 2015 financing gaps in Sub-Saharan Africa and the rest of the world, 2001­2015 ................................................................................................................................ 24 List of Figures Figure 3.1. Primary school completion rates and gross enrollment ratios across countries, circa 1999 ...... 6 Figure 4.1. Stylized dynamics of selected indicators of student flow, 2000­2015 ........................................ 7 Figure 4.2. Stylized dynamics of primary school enrollments, 2000­2015 .................................................. 8 Figure 6.1. Average annual shortfall in financing to attain EFA by 2015 in 33 African countries under alternative simulation assumptions, 2001­2015 ...................................................................... 23 Figure 6.2. Trends in the shortfall in financing to attain EFA by 2015 in Africa by cost component, 2001­2015 ................................................................................................................................ 25 List of Annex Tables Table A1.1. Alternative estimates of the cost of achieving EFA by 2015 in Sub-Saharan Africa and elsewhere in the world............................................................................................................... 30 Table A2.1. Selected indicators of primary education, low-income countries in Africa and other regions, circa 1999 .................................................................................................................................. 32 Table A2.2. Simulations of the average annual volume of domestic resources mobilized for primary education and shortfall in financing to attain EFA by 2015 in 33 African countries .............. 34 Table A2.3. Aggregate annual costs and shortfall in financing to attain EFA by 2015 in 33 African countries under various simulation scenarios, 2001­2015 ...................................................... 36 Table A3.1. Country example illustrating the simulation model and the 2015 target parameters under various simulation scenarios, Angola ....................................................................................... 39 Tables A5.1­33.1. EFA 2015 Financing Gap under Alternative Policy Measures .......................................... 44 Tables A5.1­33.2. EFA 2015 Cost Estimates and Sources of Financing under "Good Practice" Policies ...... 44 Annex Figure Figure A2.1. Primary school completion rates and gross enrollment ratios across African countries, circa 1999 .................................................................................................................................. 31 Foreword S ince the Dakar World Education Forum in the recipient countries. I am pleased that already April 2000, there has been a worldwide re- there is progress toward securing the needed re- surgence in the recognition of the sources following the recent Spring 2002 indispensable role education plays in social and Development Committee meetings of the World economic development. The political commitment Bank and the International Monetary Fund with to act has also been gaining momentum, both finance and development ministers at which the within countries and in the international arena. results of this paper, along with those for non-Af- Most African governments have embraced Educa- rican countries, were discussed. tion For All (EFA) by 2015 as an integral part of Increased donor funding, welcome as it is, will their poverty reduction strategy. It is a goal to which not be enough to reach the EFA goal, however. donors have now expressed a strong commitment. Countries must also prioritize education in their While most of the action to achieve EFA neces- own budget allocations, and perhaps even more sarily rests with the countries themselves, the importantly, reform their education systems to industrialized countries also increasingly recognize improve the quality and efficiency of service de- that they have a responsibility to facilitate progress livery to reach under-served populations, including toward that goal by helping to finance it. This is girls, children living in poverty, and children from the pledge they made in Dakar when they promised minority groups. Improving the quality of the that no country with a "credible" EFA plan would learning process will be critical in this regard, given be unable to implement it for lack of resources. that EFA is more meaningfully defined in terms of What then is the volume of additional resources primary school completion rather than in terms required, and what is the best way to use the extra of enrollment as such. The new emphasis is ap- resources to catalyze progress toward the EFA goal? propriate given that of the 79 developing countries These issues are the focus of this paper as they worldwide that have attained a gross enrollment pertain to 33 of Africa's most impoverished nations rate of 100 percent, only in 27 is the primary where EFA remains an elusive goal. Taking 2015 completion rate close to 100 percent. In the re- as the target date for ensuring that all children in maining countries--many of them in Africa--grade these countries will be able to complete a full cycle repetition and premature dropping out continue of primary schooling, the paper estimates that in to frustrate the educational aspirations of far too the aggregate, and under the assumption that coun- many children. In short, countries must develop tries implement successfully a set of basic education and implement a "credible education development reforms (i.e., a "credible plan"), additional donor plan." funding will be required at the average rate of some Reforming education systems in the desired US$2.1 billion a year for the next 15 years. Two- direction will not be easy. It will require address- thirds of this funding would be for defraying the ing many issues which generally are politically very recurrent cost of service delivery. sensitive. Yet, given the increased national and These results carry important implications, not international priority for education, the opportu- only for the volume of resources to be mobilized nity to address such issues effectively may be greater but also for the modalities for resource transfer to at this time than at any other time in the past two V decades. History tells us that strong political will Uganda and Tanzania have made significant on the part of governments is essential, and this progress toward EFA following the 1990 Jomtien must be seen in the pattern of budget allocations World Education Conference. While the road and the willingness to implement reforms to es- ahead remains arduous for many countries on the tablish credible education programs characterized continent, the prospects for further progress re- by fiscal sustainability, sound technical quality, and mains excellent in light of the renewed focus on political and social viability in the national con- EFA as a key development goal in the new millen- text. Success will also depend on creative nium. It is my hope that the publication of this collaboration between the government and key study on the external funding requirements to at- stakeholders in the country, particularly teachers tain that goal would stimulate broad discussion and parents. Legitimate concerns that teachers may and deepen the understanding among all concerned have about their working conditions need to be with extending education to every African child addressed, but ways must also be found to build regarding what it will take to realize this goal. mutual accountabilities for achieving results. Governments also need to affirm the valuable role that non-government organizations can play in service delivery. Birger Fredriksen EFA is a challenging task, but it is by no means Senior Education Advisor an impossible task. Thanks to very deliberate Human Development Department policies during the 1990s, African countries such Africa Region as Mauritania, the Gambia, Guinea, Malawi, VI Acknowledgments W e would like to acknowledge with thanks Without the help of these friends and colleagues, the contribution of many people at the our paper would certainly have been much more World Bank and elsewhere--too numer- limited in scope and therefore much less useful ous to name one by one--whose generosity with for discussions on how countries can aspire to their time and personal collection of education the goal of Education for All by 2015. We are statistics and other data have made it possible also extremely grateful for the generous finan- for us to compile the country-specific data needed cial support provided by the Norwegian for the simulations as well as for checking the Government through the Norwegian Education internal consistency of the database itself. Trust Fund. VII Abstract E ducation for All (EFA) by 2015 is one of eight spending on nonteacher inputs, and so on) for all Millennium Development Goals to which 189 33 countries by 2015, based on the patterns ob- countries committed themselves in 2000 as part served in low-income countries that have already of the effort to reduce poverty and improve the wel- achieved EFA or are close to achieving it. The im- fare of their populations. In some countries, reaching plicit assumption is that countries would reform these goals will require more resources than their their education sector policies as needed to ensure governments can mobilize from the domestic that resources are used to offer quality services in a economy. This paper estimates that for 33 of Africa's cost-efficient manner. This policy-based approach poorest countries, the aggregate gap is about $2.1 distinguishes this paper from other attempts to es- billion annually between 2001 and 2015 (includ- timate the cost of achieving EFA by 2015. Moreover, ing the AIDS-related incremental costs of school our estimate of $2.1 billion annually (in 2000 prices, services and support for the schooling of maternal or about $1.8 billion in 1995 prices) is much lower and double orphans). The estimate amounts to about than the $7 billion to $15 billion (in 1995 prices) 28 percent of the total cost of attaining EFA by 2015 proposed elsewhere. One reason is that other stud- in these countries, which implies that the bulk of ies typically include many more countries than the the financial burden of achieving EFA by 2015 will 33 in our sample; another is that our estimate re- continue to be borne by the countries themselves. fers to the external funding gap, whereas estimates About two-thirds of the shortfall in funding is for in other studies typically refer to the total cost of recurrent costs, a result with important implications achieving EFA by 2015. The main reason for the for the choice of mechanisms to channel donor difference, however, is that the assumed cost of support to the recipient countries. The estimate is service delivery in the other studies--under status based on country-specific simulations that set the quo conditions--tends to be much higher than that same indicative targets for service delivery (such implied by the policy-based simulations for our set as pupil-teacher ratios, teacher salaries, share of of countries. IX 1 Introduction A t the Dakar Education for All (EFA) Forum ited set of key variables relevant to the exercise.2 in April 2000, World Bank President James For the sake of simplicity, we set aside at this stage D. Wolfensohn committed the Bank's help the physical feasibility of the EFA-by-2015 objec- to any country with a viable and sustainable plan tive. We focus instead on the size of the financing for achieving EFA but without the external re- gap under alternative policy scenarios in each coun- sources to implement it.1 The onus therefore is on try, in particular policies affecting the volume of each country to propose a credible plan and on domestic resources mobilized for the purpose, and the Bank to help mobilize the resources needed to those affecting the efficiency of service delivery finance the plans. Together with the international and its focus on student learning. As such, the community's pledges, the Bank's commitment to results of this costing exercise are designed to re- EFA has been formalized in the United Nations spond to the dual challenge contained in the World Millennium Declaration in September 2000. The Bank's commitment at the Dakar EFA Forum: that Millennium Development Goals (MDGs), the each country present a credible plan for achieving agreed road map for implementing the Declara- EFA, and that the international community close tion, includes EFA as an explicit objective: "that the financing gap to reach that goal in the frame- by 2015, children everywhere, boys and girls alike, work of such a plan. will be able to complete a full course of primary schooling." A related MDG seeks to "eliminate Notes gender disparity in primary and secondary educa- tion, preferably by 2005, and to all levels of 1. For more information, see resources at . nal resource needs for achieving EFA by 2015 in 2. Despite being low-income countries with populations 33 of the poorest countries in Africa. These are exceeding 1 million people, Somalia and Liberia were the countries with populations of at least a mil- excluded from the costing exercise for lack of data; lion people for which it has been possible to gather Zimbabwe was excluded because the country has al- reasonably reliable and complete data on a lim- ready attained the EFA goal. 1 2 Approaches to EFA Costing S everal estimates of the cost of achieving EFA 4. The model recognizes that in order to achieve already exist, including those made by such the EFA goal, target setting must be matched organizations as UNESCO and UNICEF.1 by efforts to improve the learning environment These exercises put the aggregate additional cost at the classroom level, and by financial and other to attain EFA in all countries in the range of US$7 interventions to attract the out-of-school popu- billion to US$15 billion annually. In the World lation to enroll. Accordingly, the cost estimates Bank, a recent estimate put the cost of achieving include explicit provision of resources to pay universal primary education by 2015 at between for school inputs to complement the services US$10 billion and US$15 billion annually (in 1995 of teachers, as well as to support demand-side prices).2 Although the calculations differ in the financing where needed to boost the incentives details, they typically involve the equivalent of mul- for schooling. tiplying the estimated out-of-school population by 5. The model leaves room for policy reform to the estimated per-pupil cost of service delivery. improve the delivery of primary school services In this paper we simulate the cost of achieving over the simulation period. Thus, as enrollments EFA by 2015 using a somewhat different approach expand toward the EFA goal by 2015, the out- from previous attempts. Its key distinguishing fea- of-school children would enter a system with a tures are as follows: gradually improving delivery mechanism. This approach contrasts with past EFA costing ex- 1. The simulations are country-specific and the cost ercises where policies to improve the operation of EFA 2015 for the African continent is the sum of the system were absent, so that the target total of the estimates for the 33 African coun- population would receive services as they tries in our analysis. Country-specific initial were currently provided by the unreformed conditions in each of the countries are there- system. fore incorporated in the calculations. 6. The model relies on a set of indicative targets 2. The simulations define the target variable in regarding service delivery and domestic resource terms of the primary school completion rate-- mobilization--applied uniformly across coun- that is, the share of each school-age cohort that tries--to define what is meant by a credible plan completes five or six years of the first cycle of for achieving EFA. schooling.3 7. The model makes a distinction between the re- 3. The model puts the resource needs of EFA in current and capital costs of EFA, and includes the broader context of the education sector by separate estimates for both components. explicitly recognizing the need to set aside an 8. Finally, the data used for the exercise are the adequate share of overall resources for latest available when the costing work was postprimary levels of education. started in December 2001; thus, for most of the 3 4 APPROACHES TO EFA COSTING indicators and for most of the countries, the data tion Primaire Universelle: Combien? UNESCO Insti- refer to 1999 or 2000. Standard databases such tute of Statistics Working Document (Montreal, 2001, as the World Bank's Africa Live Database (LDB), processed). For Sub-Saharan Africa, the latter paper Statistical Information Management and Analy- estimated the additional cost to achieve EFA at be- sis (SIMA) and EdStats, and the IMF's tween US$2.9 and US$3.4 billion (in 1995 prices) annually during 1998­2015. Government Finance Statistics (GFS), were used 2. See Shatayanan Devarajan, Margaret J. Miller, and Eric to compile the data on overall government fi- V. Swanson, "Goals for Development: History, Pros- nance, debt service, and population sizes (which pects and Costs," World Bank Policy Research are in turn based on data from the United Na- Working Paper 2819 (Washington, DC, March 2002); tions Population Division). see also Annex 1 below for further details and for a comparison of the estimates for African countries in Education-related data were compiled from ad the Devarajan et al. paper and those from the costing hoc sources, including the statistical abstracts pub- exercise presented below. lished by the education ministries in each country, 3. In some countries, the primary cycle is longer than and supplemented by other government reports, six years, while in others it is shorter. Where it is longer, recent World Bank country reports on education, the calculations are prorated to the first six years of and unpublished data from the UNESCO Institute the cycle; where it is shorter, no adjustment is made so as to avoid the complication of making simulations of Statistics (UIS). For some countries, the infor- for two clearly separate cycles of schooling. mation was so sparse that in-country contacts at 4. For example, recurrent spending on teacher salaries the personal or agency level were mobilized to is by definition the product of average teacher sala- gather the desired information. The data were ries and the reported number of public sector teachers. checked for internal consistency by ensuring that As another example, it can also be shown that total obvious structural relationships were preserved public spending on primary education as a percent- among the simulation variables.4 age of the GDP should be equal to the product of the The foregoing approach has also been applied following components: (teacher salaries as a multiple to 14 non-African countries that have yet to attain of the per capita GDP divided by the pupil-teacher the EFA goal as of 2000. The results for these coun- ratio) x (1 ­ proportion of pupils in privately financed tries are reported in detail elsewhere, and are used schools) x (1 + spending on inputs other than teach- in this paper to serve as a reference, where appro- ers as a multiple of spending on teacher salaries)x priate, for discussing the results for Africa.5 (total enrollments relative to the school-age popula- tion, i.e. the gross enrollment ratio) x (school age population as a proportion of the total population). Notes The data for the two sides of the equation would ob- viously need to be consistent with this structural 1. Enrique Delmonica, Santosh Mehrotra, and Jan relationship among the several variables. Vandermoortele, "Is EFA Affordable? Estimating the 5. World Bank, "Achieving Education for All by 2015: Global Minimum Cost of Education for All," UNICEF Simulation Results for 47 Low-Income Countries," Staff Working Paper EPP-01-001 (New York, 2001); Human Development Network, World Bank (Wash- and Mathieu Brossard and Luc Gacougnolle, Educa- ington DC, 2002, processed). 3 Operationalizing the EFA Concept A s indicated above, we use the primary school to overestimate the financing gap to reach EFA by completion rate (CR) to operationalize the 2015. EFA concept in this costing exercise. The As an indicator for operationalizing the EFA variable reflects the combined effect of entry rates concept, the primary school completion rate, as to first grade in the cohort, as well as the share of calculated in the manner explained above, remains entrants who go on to complete five or six years superior to the gross or net enrollment ratios (GER of the cycle. One way to compute the CR is to first or NER) that costing exercises in the past have estimate the entry and survival rates separately and typically used. One advantage of the GER or NER then multiply them to obtain the desired statistic. is that the data needed to compute them are rela- A more direct method is to compute the rate as tively easy to collect.2 Despite this advantage, the ratio of the number of nonrepeaters in grade 6 neither indicator is adequate for the following rea- (or in grade 5 if the cycle lasts only five years) to, sons: (1) they pertain to the cross-sectional typically, the number of 12-year-olds in the popu- coverage of primary schooling at a given time; (2) lation, 12 being the sum of the official age at entry they reflect averages across all grades in the whole to grade 1 (usually 7 years old) and the five years cycle; and (3) the presence of grade repetition and between grades 1 and 6).1The calculation is easily overage enrollments further reduces their ability applied to the data on hand, but suffers from the to capture the extent to which children in fact com- usual flaws of using cross-sectional data to cap- plete the full course of primary schooling. These ture a flow phenomenon involving cohorts whose problems explain why a high GER or NER does behavior may have been evolving. Therefore, in not necessarily imply that a large proportion of countries where primary school enrollments have children finish a full course of primary schooling, been rising rapidly in the recent past, the comple- a fact confirmed in Figure 3.1 by the data for the tion rate thus computed would not give the most sample of 47 countries worldwide.3 Thus, where up-to-date picture of the primary school comple- the GER or NER is mentioned below, the main tion rate among children currently making their purpose is to relate this paper to analyses elsewhere way through the system. While adjustments to the that rely on these indicators. calculation can be made to improve the estimates, In Figure 3.1, the diagonal line represents per- we chose to make no adjustments so as to keep fect one-to-one mapping between the GER and the the task manageable within the available time, and CR, but as can be seen, very few countries in the to maintain uniform treatment across countries. sample fall on that line. The countries can be sepa- Because the approach tends to underestimate rated into four groups according to values of these completion rates in 2000, it correspondingly tends indicators and the relation between them: 5 6 OPERATIONALIZING THE EFA CONCEPT Figure 3.1 performers. Note that because the CR is a continu- Primary school completion rates and gross ous variable, some countries in Group 2 are in fact enrollment ratios across countries, circa 1999 close neighbors of those in Group 1. Using similar criteria, we can also identify the high EFA achiev- ers in the African sample (see Annex Figure A2.1). 100 Here the CR threshold of 70 percent would have Group 1 90 yielded too few observations, so we extended the (%) 80 sample by lowering the threshold to 55 percent. 70 cohort For the world sample, the high EFA achievers in- 60 age clude the following 10 countries: Bangladesh, in 50 Group 3 rate 40 Bolivia, The Gambia, India, Indonesia, Lesotho, Group 2 30 Uganda, Vietnam, Zambia and Zimbabwe.5 Among 20 the 33 African countries included in the EFA cost- Completion 10 ing exercise, the high achievers are: The Gambia, 0 0 10 20 30 40 50 60 70 80 90 100 110 120 130 Ghana, Kenya, Lesotho, Nigeria, Uganda, and Zam- Gross enrollment ratio (%) bia. Note: Both axes refer to five or six years of primary schooling; see text for more details. Data refer to 47 countries worldwide. See also Annex 2, Figure Notes A2.1 for a similar graph based on data for the 33 African countries in this costing exercise. Source: Base on data in Table A2.1 1. For countries with a five-year cycle, the relevant popu- lation would be the 11-year-olds. Note that because the simulations are country-specific, the relevant age Group 1. The GER is high (at least 85 percent) cohort also varies across countries according to the official entry age to grade 1. and the CR is at least 70 percent. 2. The gross enrollment ratio is defined as the ratio of Group 2. The GER is high (at least 70 percent) the total number of pupils in a cycle of schooling to but the CR is below 45 percent. the population of children in the official age range for that cycle. The net enrollment ratio is computed Group 3. The GER and CR are both low. the same way, except that the numerator excludes pupils who are outside the official age range for that Group 4. The GER and CR are in the middle cycle. range or the GER is an outlier (that 3. The ratios in systems where primary schooling lasts is, all the remaining countries). more than six years have been adjusted to reflect the ratio corresponding to the first six years of the cycle. Countries in Group 1 can be considered as high 4. We know this because the gap between the CR and the GER is to a large extent accounted for by the extent EFA achievers: their relatively high CRs imply that of grade repetition in the system. they are making good progress toward the EFA 2015 5. Other low-income countries, such as China and Sri goal; and the fact that the GER is also high implies Lanka, could also have been included, but are not that they are doing so without major efficiency losses because of lack of readily available data at the time of arising from widespread grade repetition in the this exercise. Note that although Zimbabwe is included primary cycle.4 The cutoff threshold of 70 percent here to help describe the characteristics of the top for the CR is somewhat arbitrary and is chosen sim- EFA achievers, it is not among the 33 African coun- ply to identify a sufficiently large group of top tries for which the cost of EFA by 2015 was simulated. 4 The Simulation Model W e begin by clarifying the behavior of some six years of the cycle if it lasts longer than that). of the key EFA variables over the course Figure 4.1 shows how the student flow profile in a of the simulation period (that is, 2001­ hypothetical country would shift over time as it 2015). This is followed by a brief overview of the attains the EFA goal. In panel (a) the profiles at structure of the simulation model itself. More de- four points in time are shown. In the initial year tailed information on the model can be found in (2000) 75 percent of the school-age population Annex 3. enter first grade, and 40 percent of that popula- tion complete sixth grade. By 2005, these rates rise The Stylized Dynamics to 90 percent and 65 percent, respectively; by 2010, of Some Key Variables the entry rate reaches 100 percent, while the completion rate climbs to 85 percent. By 2015, both Achieving EFA 2015 implies that all children in the entry and completion rates reach 100 percent, an age cohort enter the system and remain in it signifying attainment of EFA. In panel (b) of the until they have completed the primary cycle (or figure, the same information is presented differ- Figure 4.1 Stylized Dynamics of Selected indicators of Student Flow, 2000­2015 Panel (a) Panel (b) 100 2015 100 2010 80 80 (%) 2005 Intake cohort 60 60 age in Completion rate 40 2000 Percent 40 Repetition 20 20 Enrollment 0 0 1 2 3 4 5 6 2000 2005 2010 2015 Grade Source: Authors' construction. 7 8 THE SIMULATION MODEL ently to emphasize that in order to attain EFA by Overview of the Simulation Model 2015, countries must necessarily have achieved uni- versal entry to first grade by 2010. Panel (b) also The model has a simple and lean structure that shows that the repetition rate would need to de- seeks to minimize the data needs while incorpo- cline over the period (if they are high to begin with rating key policy variables relevant to projecting in the initial year), given that extensive grade rep- the external financing gap to attain the EFA goal. etition tends to encourage dropping out and is For each country, estimation of the external financ- therefore incompatible with the EFA goal. ing gap involves four steps, as elaborated below. The profiles of student flow in Figure 4.1 are The first step is to estimate the recurrent cost of applied to projections of the school-age population achieving EFA by 2015. Two sets of variables are to obtain estimates of total primary school enroll- relevant to the calculation. The first pertains to the ments over the course of the simulation period. projected number of children served by publicly Figure 4.2, panel (a), shows typical patterns in the funded schools. This number depends on: (a) the growth of the school-age population, as well as the projected size of the school-age population over corresponding number of primary school pupils and the simulation period; (b) the intake rate to first nonrepeaters among them. Notice that as the year grade; (c) the share of first-graders who complete approaches 2015, the gap between total enrollments the primary cycle; (d) the prevalence of grade rep- and the number of nonrepeaters narrows, reflect- etition (which implies that extra places must be ing the assumed decline in repetition rates. Panel financed beyond the number based only on intake (b) of Figure 4.2 shows the distribution of enroll- and survival rates); and (e) the share of pupils ments between publicly and privately financed enrolled in publicly funded schools. The second schools. The share of pupils in the public sector is set of variables relates to the cost of service deliv- expected to grow because as schooling extends to ery in the publicly funded sector. The estimate the whole population of school-age children, the depends on the following factors: (a) the ratio of increase would increasingly comprise children from pupils to teachers in publicly funded schools (which the poorest families or those from the most deprived allows the number of teachers in such schools to circumstances (for example, orphans). It is unlikely be estimated); (b) the average salaries of public that this population would be willing or able to sector teachers (which is multiplied by the num- afford to enroll in private schools. Hence, the simu- ber of public sector teachers to obtain the aggregate lation model assumes that the share of enrollments salary bill); and (c) the cost of complementary in the privately financed sector would fall as coun- school inputs to promote student learning as well tries approach EFA by 2015. as to finance top-up spending targeted to the most Figure 4.2 Stylized dynamics of primary school enrollments, 2000­2015 Panel (a) Panel (b) School-age population pupils Enrolled pupils of of All pupils population % in private Number Pupils in Non-repeaters Number public schools schools 2000 2005 2010 2015 2000 2005 2010 2015 Source: Authors' construction. FINANCING EDUCATION FOR ALL BY 2015: SIMULATIONS FOR 33 AFRICAN COUNTRIES 9 disadvantaged children who might otherwise not ers are assigned to teach all subjects to the same attend school (for example, those from the poor- group of pupils). The unit cost to construct and est families or hard-to-reach communities). furnish classrooms ranges widely across African Because of the high prevalence of AIDS in many countries, and is set at what we consider to be a African countries, a separate calculation is made reasonable level corresponding to the middle of to incorporate the epidemic's impact on the cost of the range for the countries for which the relevant extending primary education to all children. The data are available. impact is mediated through three channels: (a) The third step establishes the volume of resources changes in the size of the school-age cohort; (b) that the country itself mobilizes to reach the goal increased teacher absenteeism which effectively of EFA by 2015. The amount mobilized depends translates into higher costs as more teachers would on the behavior of the following factors over the be needed beyond the number determined solely course of 2001­2015: (a) the projected growth rate by the average pupil-teacher ratio; and (c) increased of the country's GDP; (b) the size of government incidence of orphanhood as parents die of AIDS, revenues relative to the GDP; (c) the share of gov- widening the need for public subsidies to encour- ernment revenues devoted to education; and (d) age school attendance by the affected children. The the share of public spending on education that is population projections used in the simulations al- set aside specifically for primary education. ready incorporate the effects of AIDS, so no separate The fourth step in the calculation is to deter- adjustment is made. The second effect is incorpo- mine the resource gap for external financing to rated by relying on projections made by UNAIDS support progress toward EFA by 2015. The calcu- on the prevalence of the disease in each country; lation simply involves subtracting the result from and findings by researchers at London's Imperial steps 1 and 2 from those in step 3. The algorithm College on the infection rates among teachers and is arranged to reflect the assumption that domes- the impact on teacher absenteeism. The impact of tic resources would first be used to finance the AIDS on orphanhood relies on UNAIDS projections recurrent costs of service delivery, and any surplus of the number of maternal and double orphans and would be applied toward the capital costs. In coun- an assumed support of US$50 per orphan.1 tries where domestic resources are insufficient to The second step in the simulation involves esti- cover the recurrent costs, the assumption is that mation of the capital costs of building and external resources would be available to help fi- furnishing classrooms to accommodate the grow- nance the shortfall in financing for both the ing number of children enrolled in the publicly recurrent and capital costs. funded sector. The calculation involves determin- ing the number of new classrooms implied by the Note growth in enrollments, assuming that the pupil- classroom ratio is the same as the pupil-teacher 1. For more details on the AIDS-related part of the simu- ratio (which is equivalent to assuming that teach- lation model, see Annex 4. 5 Initial Conditions and Policy Targets toward EFA by 2015 T he model assumes that, for all countries in Number Enrolled in Publicly Funded the sample, the policy variables affecting ser- Primary Schools vice delivery converge by 2015 to a single set of standard indicative benchmarks that are Table 5.1 shows the variables relevant to this part consistent with what we believe to be good prac- of the exercise. The first four columns show the tice policies; for the policy variables affecting summary statistics for these variables for 1999/ domestic resource mobilization, we used three 2000--for the 33 African countries, as well as for sets of indicative targets that take into account the high EFA achievers in the African and world stylized patterns across countries at different samples. The last column shows the 2015 indica- income levels. In defining good practice, we have tive targets for four policy variables: the entry rate been guided in large part by the characteristic to first grade, the primary school completion rate arrangements in low-income countries that have in the age cohort, the share of repeaters in total come closest to achieving EFA (that is, Group 1 enrollments, and the share of enrollments in the countries in figure 3.1). The use of indicative tar- private sector. Consider first the school-age popu- gets has the obvious advantage of simplicity, and lation relative to the size of the total population. serves the immediate purpose of generating in- Although not treated as a policy variable in the dicative estimates of the aggregate external simulations, the demographic burden has impor- financing gap to achieve EFA by 2015 in Africa. tant implications for education finance. Across In more detailed work on country-specific simu- African countries, the burden ranges between 13.5 lations, it can obviously be refined to take into and 17.7 percent, which implies that all else being account the special circumstances of individual the same, the country with a burden of 17.7 per- countries. cent would require about 31 percent more resources Below we present the initial and target values to achieve the same schooling coverage as the coun- for all the variables relevant to the simulations. try with a burden of 13.5 percent. At an average of The discussion is grouped under four rubrics cor- 16.1 percent, the demographic burden for the 33 responding to the sequence of calculation: African countries is comparable to the average for estimation of the number of pupils in publicly the high EFA achievers in the African sample, and funded schools, the cost of services in publicly larger than the average of 14.3 percent among the funded schools, the cost of classroom construc- high achievers in the world sample. tion, and the volume of domestic resources The next two parameters in Table 5.1 are the available to finance primary schooling. intake rate to first grade and the primary school 11 12 INITIAL CONDITIONS AND POLICY TARGETS TOWARD EFA BY 2015 Table 5.1 Range, Means, and Targets for Parameters Affecting Enrollments in Publicly Funded Primary Schools, 1999/2000 and 2015 Sample mean in 1999/2000 High EFA achievers a Range in Africa Simulation variables 1999/2000 sample Africa World Target for 2015 School-age population as % of total population 13.5­17.7 16.1 16.5 14.3 -- Entry rate to grade 1 (%) 37­100 81 94 103 100 b Primary school completion rate in the school-age cohort (%) 20­85 43 71 82 100 Share of repeaters in total enrollment (%) 1­36 18 9 10 10 Private enrollments as % of total 0­36 8 5 5 10 Memo item: Gross enrollment ratio (%) 32­117 77 90 103 110 Note: The gray cells denote variables that do not enter the simulations as variables embodying policy choices. -- Indicates not applicable. The size of the school- age population in 2015 relative to the total population reflects population projections by the United Nation's Population Division; and the projected gross enroll- ment ratio in 2015 is the value implied by the assumed rates of intake into grade 1 and survival to grade 5, as well as the rate of grade repetition. a. The high EFA achievers in the African sample refers to the seven countries identified as Group 1 in Annex Figure A2.1 (The Gambia, Ghana, Kenya, Lesotho, Nigeria, Uganda, and Zambia). Recall that in these countries, the primary school completion rate in the school-age cohort is among the highest in the sample, and the gross enrollment ratio is also high, indicating that grade repetition is under reasonable control. The high achievers in the world sample include the following countries: Bangladesh, Bolivia, The Gambia, India, Indonesia, Lesotho, Uganda, Vietnam, Zambia, and Zimbabwe. b. Achieved no later than 2010. Source: Various as described in the text; see Appendix Table A2.1 for country-specific data. completion rate in the school-age cohort. The co- would be implied by the entry and survival rates hort survival rate to the end of the primary cycle, alone. As the table shows, the share of repeaters in although obviously also relevant, is not shown primary education in 1999/2000 ranged from a low separately because it is implied once the values of of about 1 percent, to an astonishing 36 percent, the other two variables are fixed. The entry rate to for an average of 18.2 percent regionwide in Af- first grade currently ranges from only 37 percent rica, and 7.8 percent among the high achieving to 100 percent across the 33 African countries, and countries in the African sample. High repeater rates the completion rate goes from only 20 to 85 per- are incompatible with the goal of raising the cent (which implies that survival rates in most completion rate to 100 percent by 2015 for rea- African countries are well short of the desired 100 sons explained earlier. Thus, taking as a guide the percent). Among the high EFA achievers in Africa, current share among the high EFA achievers world- first grade intake rates and primary school comple- wide, we set a target of 10 percent by 2015 for this tion rates are, respectively, 94 and 72 percent. These parameter in the simulations. Consistent with this levels are comparable to, but admittedly still be- assumption and the target of 100 percent for the low, the levels among the high achievers worldwide. entry and completion rates, the implied gross en- By definition, the EFA goal as specified in the rollment ratio in 2015 would be 110 percent. Millennium Declaration implies that the intake rate The share of enrollments in privately financed would reach 100 percent by 2010 at the latest, and schools is the last variable that feeds into the cal- the completion rate, by 2015. These then are the culation of the number of children who would be targets for these two parameters in the simulations served by publicly funded schools. In 1999/2000, reported below. the private sector enrolled between 0 and 36 per- The next variable--repeaters as a share of total cent of primary school children across African enrollments--captures the efficiency of student countries. The share averaged 8.0 percent for the flow in the education system. The greater the share, whole sample, compared with 5.0 percent among the more total enrollments are inflated beyond what the high EFA achievers in the African sample and FINANCING EDUCATION FOR ALL BY 2015: SIMULATIONS FOR 33 AFRICAN COUNTRIES 13 5 percent in the world sample. To leave room for disadvantaged circumstances. Rather than itemiz- diversity in service delivery, we set the 2015 target ing the individual amounts for these inputs, the for the share of enrollments in privately financed model specifies a lump sum relative to the aggre- schools at 10 percent in countries with a share that gate bill for recurrent spending. This approach has currently exceeds this threshold; for the remain- the advantage of simplifying data collection--not ing countries, the share is maintained at the current a small advantage considering the large number level for the duration of the simulation period. of countries included in the sample. It also is con- sistent with the focus on broad expenditure Cost of Delivering Publicly Funded Primary categories in this costing exercise, and the recog- Schooling nition that country conditions would play a key role in identifying the best way to use the resources Teacher salaries are the single most important cost within the overall envelope set aside for inputs item. The variable enters the simulation model as other than teachers. the average annual salary of teachers expressed as The relevant summary statistics and target val- a multiple of per capita GDP. The total salary bill ues of the three policy parameters appear in Table is computed by multiplying the average salary by 5.2. Consider first the data on teacher salaries. the total number of teachers in publicly funded Across the 33 African countries, average salaries primary schools. The number of teachers is in turn range between 1.5 and 9.6 times the per capita GDP, calculated by dividing total enrollments in such illustrating the enormous diversity across coun- schools--obtained from the calculations described tries. Salaries relative to the per capita GDP are in the previous section--by the simulation's sec- generally higher in Africa than in other regions, ond policy parameter, the pupil-teacher ratio.1 The but the averages for the high EFA achievers in Africa third and final parameter for this part of the simu- and in the world are largely comparable. For the lation is the lump sum set aside to cover other costs simulations, the 2015 target for this parameter is besides teacher salaries. Such costs include those set at 3.5 times the per capita GDP, a round figure for administration (systemwide and at the school that is consistent with teacher salary levels among level), support services (for example, inspectorates, the high EFA-achieving countries. This indicative examinations, and student evaluation), textbooks, target is applicable mainly to countries with a per and other pedagogical supplies, as well as subsi- capita GDP in the US$300 to US$700 range, given dies or other welfare services that may be needed that most high EFA achievers are in this income to encourage school attendance by children from range. In the poorest countries, the target may not Table 5.2 Range, Means and Targets for Parameters Affecting Service Delivery Costs in Publicly Funded Primary Schools, 1999/2000 and 2015 Sample mean in 1999/2000 High EFA achievers a Range in Africa Simulation variables 1999/2000 sample Africa World Target for 2015 Average annual teacher salary as multiple of per capita GDP 1.5­9.6 4.4 3.8 3.6 3.5 Pupil-teacher ratio 24­79 48 39 40 40 Nonteacher salary expenditure as % of total recurrent spending 4­45 24 17 26 33 a. The high EFA achievers in the African sample refer to the seven countries identified as Group 1 in Annex Figure A.21 (The Gambia, Ghana, Kenya, Lesotho, Nigeria, Uganda, and Zambia). Recall that in these countries, the primary school completion rate in the school-age cohort is among the highest in the sample, and the gross enrollment ratio is also high, indicating that grade repetition is under reasonable control. The high achievers in the world sample include the following countries: Bangladesh, Bolivia, The Gambia, India, Indonesia, Lesotho, Uganda, Vietnam, Zambia, and Zimbabwe. Source: Various as described in the text; see Table A2.1 for country-specific data. 14 INITIAL CONDITIONS AND POLICY TARGETS TOWARD EFA BY 2015 translate into a "living wage" for teachers, and it going from 24 pupils per teacher to 79, with a re- may not be consistent with the greater scarcity of gional average of 47.8. Among the high EFA educated labor in such economies, a possibility achievers, the average is 38.5 for the Africa sample, suggested by the fact that teacher salaries relative and 41.1 for the world sample. The simulations to the per capita GDP typically fall as the per capita assume that this policy parameter would move to GDP rises. For the purpose of computing the glo- 40 pupils per teacher by 2015, which again is a bal financing gap, however, we applied a uniform round figure consistent with the ratios in the high- indicative target across all countries, leaving re- achieving countries. Unlike the case of teacher finements for follow-up work on each country. salaries, however, no distinction is made regard- The dynamics of the teacher salary variable in ing the starting value of this variable in 1999/2000. the simulations depend on whether or not initial This is because teaching is manageable at this ra- salaries are above or below the target value of 3.5 tio, and the move toward the target of 40 pupils times the per capita GDP. For countries below the per teacher by 2015 should be reasonably feasible target, the simulations assume that the average to implement. salaries for teachers taken as a group would move Again as with teacher salaries, initial conditions up to 3.5 times the per capita GDP, and the rise would affect how we interpret the move toward would take effect immediately. Because the rise is the intended target. In countries where the ratio is likely to improve the quality of services--by help- currently above the target value of 40 pupils per ing to lower teacher absenteeism, attracting better teacher, the move would amount to a quality-en- qualified teachers, and stimulating greater account- hancing measure (by virtue of the implied reduction ability for teaching effectiveness--the assumption in classroom congestion), while in countries cur- corresponds largely to a measure to improve the rently below the target, it would mostly represent quality of services. a measure to improve the efficiency of service de- For countries above the target level for teacher livery (by virtue of the implied decline in unit salaries, the simulations assume that incumbent costs). teachers would continue to be paid on their cur- Finally, we turn to the amount of spending on rent salary scale, while new recruits would be paid inputs other than teachers. The volume of such at 3.5 times the per capita GDP. Over time, as the spending in 1999/2000 ranged from 4 to 45 per- share of teachers on the old scale diminishes, the cent of recurrent spending on primary education. overall average would approach the intended tar- The average for the 33 African countries was 24 get level by 2015. For such countries, the move to percent, compared with 17 percent among the high the target level of teacher salaries is more appro- EFA achievers in the African sample, and 26 per- priately viewed as a measure to improve the cent among the high achievers in the world sample. cost-efficiency of services. This is because in most As the table indicates, the simulations assume a African countries where teacher salaries are cur- 2015 indicative target for this parameter of one- rently on the high side, the number of applicants third of total recurrent spending on primary tends to exceed the number of teaching posts, education. Spending on inputs other than teach- sometimes by large margins. Well-qualified can- ers tends to be on the low side across African didates would thus still be available to fill teaching countries, even in the high-achieving countries. positions at the target salary level of 3.5 times the The results are twofold: (1) well-known deficien- per capita GDP, implying that the cost savings cies in the adequacy of support to ensure would probably not come at the expense of the instructional effectiveness in the classroom in most quality of services. Instead, they would free up public school systems; and (2) limited availabil- resources to finance additional teaching posts, ity--or indeed absence--of provision for thereby relieving congestion in the classrooms demand-side financing to encourage school atten- (while incidentally also producing the social ex- dance by children from poor families. While the ternality of reducing unemployment among the target of 33 percent is admittedly a rough guess at pool of qualified candidates for teaching jobs). what would be the appropriate level of spending Consider now the pupil-teacher ratio. As with on nonteacher inputs in different settings, it rep- teacher salaries, the range is also strikingly wide-- resents a generous, though not yet an excessive, FINANCING EDUCATION FOR ALL BY 2015: SIMULATIONS FOR 33 AFRICAN COUNTRIES 15 allocation to support effective learning environ- Domestic Resources Mobilized ments in publicly funded schools. for Primary Education As with the other two policy parameters in this part of the simulation exercise, how one interprets The relevant data appear in table 5.3. Consider first moving to 33 percent by 2015 depends on initial the level of public spending on primary education conditions: reaching it from a current allocation as a share of the GDP. In 1999/2000, this variable that is smaller than the target may be viewed largely ranged from 0.6 to 3.2 percent of GDP across the as a quality-enhancing measure (since it would 33 African countries, for an average of 1.4 percent allow improvements in the availability of inputs for the whole sample, and 1.9 percent and 1.8 per- to complement teachers' time and effort); reach- cent, respectively, among the high EFA achievers ing it from the opposite direction may be described in the African and world samples. The wide gap in most instances as a cost-efficiency measure, since across countries arises from differences in the over- the cut can probably be absorbed by reducing all volume of public revenues, the priority placed spending on back-line administration and other on education in the allocation of public spending, overheads without adverse effects on the quality and within education, the priority placed on pri- of services at the classroom level. mary education.2 As Table 5.3 indicates, government revenues as Cost of Constructing and Furnishing a share of GDP lie in the 8 to 26 percent range Classrooms (excluding Angola, Eritrea, Lesotho, and Nigeria where the range is 34 to 56 percent, reflecting the The capital cost of EFA 2015 depends on three unique public finance circumstances of those coun- variables: (1) the increase in enrollments in pub- tries). The corresponding average was 15.4 percent licly funded schools; (2) the pupil-classroom ratio for the full African sample (again without the out- (which would be the same as the pupil-teacher ratio, liers), 18.8 percent among Africa's high EFA on the assumption that each teacher is responsible achievers, and 20.7 percent for the high achievers for one section of children); and (3) the cost of worldwide. In the simulations, we apply three dif- constructing and furnishing each classroom. The ferent values for the 2015 target volume of first two variables fall out of the calculations al- government revenues relative to the GDP, depend- ready completed in the steps described above, and ing on the country's level of income: can be used directly to compute the incremental stock of classrooms needed to accommodate the · 14 percent for countries with a per capita new enrollments implied by the EFA 2015 goal. GDP below US$300 in 2000; Construction costs tend to be higher in Africa than · 16 percent for countries with a per capita in other low-income settings, perhaps reflecting the GDP between US$300 and US$600 in 2000; nature of markets in the business, and there is wide and variation across countries, ranging from US$4,000 · 18 percent for countries with a per capita per classroom for community-built structures to GDP above US$600 in 2000. as much as US$17,000 apiece for prefabricated structures built from imported materials. Because The motivation for differentiating across coun- community-built schools typically rely on unpaid tries at different income levels is to recognize the labor from parents and other community members, fact that richer countries generally enjoy larger tax as well as in-kind contributions, the actual cost bases than poorer countries. The range of 14 to 18 may be higher than the US$4,000 at the low end of percent is broadly consistent with the patterns the scale. Such schools tend to require more main- across countries worldwide, and responds to the tenance and are also generally less durable. For the principle that the richer the country, the more simulations, we use a benchmark of US$8,000 per domestic resources it should be expected to mobi- classroom--again a rough guess, but arguably a lize to finance the country's public services such sensible norm that is sufficiently generous with- as primary schooling. out being excessive to the point of crowding out For some countries in each income range, gov- community participation in school construction. ernment revenues as a share of the GDP in 1999/ 16 INITIAL CONDITIONS AND POLICY TARGETS TOWARD EFA BY 2015 Table 5.3 Sample Range and Means of Variables Affecting Domestic Resources Mobilized for Primary Education, circa 1999/2000 Sample mean High EFA achievers a Range in Africa Simulation variables Africa Sample sample Africa World Government revenues (excl. grants) as % of GDP b 8­26 15.4 18.8 20.7 Public recurrent spending on education as % of government revenues c 4­33 17.8 19.3 18.5 Primary education as % of public recurrent education spending 35­66 49.0 44.4 44.1 Recurrent spending on primary education as % of GDP 0.6­3.2 1.4 1.9 1.8 a.The high EFA achievers in the African sample refer to the seven countries identified as Group 1 in Annex Figure A2.1 (The Gambia, Ghana, Kenya, Lesotho, Nigeria, Uganda, and Zambia). Recall that in these countries, the primary school completion rate in the school-age cohort is among the highest in the sample, and the gross enrollment ratio is also high, indicating that grade repetition is under reasonable control. The high achievers in the world sample include the following countries: Bangladesh, Bolivia, The Gambia, India, Indonesia, Lesotho, Uganda, Vietnam, Zambia, and Zimbabwe. b. Range and means in the first two columns exclude data for Eritrea, Lesotho, Nigeria, and Angola where government revenues as a share of GDP are exception- ally high, ranging between 34 and 56 percent; the mean for the "best" in the Africa sample excludes data for Lesotho and Nigeria for the same reason. c. Denominator refers to revenues net of grants; in Africa, the amount is close to total public spending net of debt servicing. Source: Various as described in the text; see Table A2.1 for country-specific data. 2000 already exceed the benchmark of 18 percent; share rising to 20 percent if currently below these include Angola, Republic of the Congo, 20 percent, or falling to 26 percent if cur- Eritrea, Kenya, Lesotho, Malawi, Mauritania, Ni- rently above 26 percent; or geria, Senegal, and Zambia. As an alternative · a uniform rate of 20 percent in all countries simulation scenario, we assume that, for these coun- by 2015. tries, the 2015 target would remain the same as in the initial year, while for the other countries the One obvious implication of the first assump- target would be those indicated above. tion is that the more a country places priority on Turning now to the share of public spending de- education in the allocation of its own public spend- voted to education services, Table 5.3 shows that the ing, the smaller the external financing gap, and range goes from 4 to 33 percent among the 33 Afri- therefore the more the country is penalized in the can countries, for an average of 17.8 percent for distribution of donor support for countries' the sample. Among Africa's high EFA achievers, the progress toward the EFA 2015 goal. The second range is between 20 and 28 percent and the aver- option, however, applies a common standard across age, about 19.3 percent. Among the high achievers the board, and is therefore free of what would ap- in the world sample, the share ranges from 9 to 30 pear to be a perverse bias against countries that percent, for an average of 18.5 percent. In choosing are themselves making a strong effort toward the the 2015 target value for the variable, an important EFA goal. These implications are useful to keep in consideration is that it should represent a sustain- mind when interpreting the simulation results able claim on public spending by the education presented below. sector in the context of legitimate claims that other The last simulation variable for this part of the sectors have on the public purse. Taking account calculations pertains to the share of primary edu- of the patterns of spending across countries, we cation in overall recurrent public spending on chose the following two alternative 2015 targets for education. As Table 5.3 shows, the variable varies the share of public spending allocated to education: widely across African countries, ranging from 35 percent at the low end, to 66 percent at the high · the current share persisting to 2015 if it is in end. The average for the region is 49.0 percent, the 20 to 26 percent range; otherwise the while it is 44.4 percent for the high EFA achievers FINANCING EDUCATION FOR ALL BY 2015: SIMULATIONS FOR 33 AFRICAN COUNTRIES 17 Table 5.4 Indicative Targets for Simulation Parameters Affecting the Volume of Domestic Resources Mobilized for Primary Education Targets for 2015 under 3 alternative scenarios Simulation variables Scenario 1 Scenario 2 Scenario 3 Government revenues as % of GDP 14­18 14­18 unless current share exceeds 18%, in which case the target is the same as at present All public spending on education as % of government revenues 20­26 20 20 Primary education as % of public recurrent education spending 50 Note: Shaded cells indicate that the target for 2015 remains unchanged from the values under scenario 1. Source: See text discussion. in Africa, and 44.1 percent for the high achievers consistent with current spending patterns in a large worldwide. All the high-achieving countries in number of countries. Only one value is used in an Africa have six-year cycles, while in three coun- effort to keep the number of alternative scenarios tries in the world sample--Bangladesh, India, and to a manageable and meaningful minimum. Vietnam--primary schooling lasts only five years. To summarize, Table 5.4 shows the target val- For these countries, the equivalent allocation for ues of the variables affecting the volume of domestic a six-year cycle would be 20 percent higher, as- resources mobilized towards the goal of EFA by suming that the extra year costs the same as the 2015. average year in the current system. Making this adjustment would raise the average allocation for Notes primary education in the high EFA group in the world sample to 46.3 percent. 1. Alternatively, the total salary bill calculation can be In an EFA costing exercise, it is tempting to aim estimated by first computing the cost of teachers per for maximum allocation for primary education. Yet pupil (that is, dividing average salaries by the pupil- depleting the other levels of resources would be teacher ratio), and then multiplying the result by the counterproductive: an education system does not total number of pupils in publicly funded schools. consist of primary education alone, and postprimary 2. As the simulation model in Annex 3 indicates, the growth rate of the GDP also enters into the simula- levels of education are the source of the new teach- tions. However, because teacher salaries, the main ers who will be needed for the growing school cost of service delivery, is expressed as a multiple of population as countries move toward the EFA goal. the per capita GDP, changes in the rate of growth of In the simulations, we set the target allocation for the GDP has only a modest impact on the simulation primary education at 50 percent by 2015, a level results. For our purpose, we apply a 5 percent an- that is both slightly above the 46.3 percent charac- nual growth rate across all countries in the costing teristic of high EFA-achieving countries and broadly exercise. 6 Simulation Results W hat, then, is the shortfall in funding to The simulations set uniform 2015 targets for these achieve Education for All by 2015 in the variables in all 33 countries, a justifiable approach 33 African countries in this study? While because these indicative targets correspond to a multiplicity of scenarios can be simulated, we physical arrangements that are both meaningful to focus on six sets of assumptions. Below we sum- educators and characteristic of the arrangements marize these simulation scenarios; present the in countries that have come closest to achieving results for one country, Niger, as an example of EFA without significant efficiency losses.1 Move- the country-specific results that are reported in ment toward the indicative targets can typically be Annex 5; and summarize the aggregate estimate described as leading to improvements in either the of the funding shortfall for the 33 countries. quality of the learning environment or the efficiency of service delivery. Thus, in Table 6.1, these vari- The Simulation Scenarios ables are grouped under quality or efficiency measures. With regard to teacher salary, movement The target is to ensure that, by 2015, all chil- toward the indicative target can interpreted either dren would have access to a complete cycle of way, depending on whether or initial salary levels primary schooling (lasting at least five years; and are below or above the norm. if the cycle is longer, six years). The cost of reaching With regard to the parameters affecting domes- this goal is simulated under the six simulation tic resource mobilization, again there are four assumptions shown in Table 6.1: (1) with un- variables that matter: (1) the share of government changed policies in service delivery and domestic revenues in GDP; (2) the share of revenues spent resource mobilization (the status quo scenario); on education; (3) the share of primary education (2) with only quality-enhancing measures in service in total government spending on education; and delivery (scenario A); (3) with both quality- and (4) the share of pupils in privately financed pri- efficiency-enhancing measures in service deliv- mary schools. Although three sets of assumptions ery (scenario A + B), corresponding to what we are used, we believe scenario A + B + C2 responds call the "good practice" scenario; and (4) with best to the expectation in the donor community "good practice" service delivery combined with that recipients of external funding toward the EFA three alternative assumptions on the domestic 2015 goal would present a credible plan toward effort to mobilize resources toward the EFA 2015 this end--that is, those countries making a good goal (scenarios A + B, combined with C1, C2, or faith effort to prioritize their own public spending C3). toward the goal and ensuring that services are fo- To recapitulate, four variables are used to de- cused on student learning and delivered in a scribe service delivery: (1) the pupil-teacher ratio; cost-efficient manner. Further, as explained before, (2) the share of recurrent spending on inputs other in terms of its implications for the incidence of ag- than teachers; (3) the average level of teacher pay; gregate external support toward the EFA goal, the and (4) the share of repeaters in total enrollments. scenario also avoids penalizing countries that cur- 19 20 SIMULATION RESULTS Table 6.1 Assumptions underlying six alternative simulation scenarios in the EFA costing exercise Policy measures and the relevant simulation variables A. Quality measures B. Efficiency measures C. Financing measures Average Government revenues Primary annual education % of teacher share of spending salary as recurrent % of Pupil- on non- multiple of Repeaters spending on pupils in Simulation teacher teacher per capita as % of As % % for education private scenario ratio inputs a GDP enrollments of GDP education (%) schools Status quo A only 40 33.3 A + B 40 33.3 3.5 10 A + B + C1 40 33.3 3.5 10 14­18 20­26 50 10 A + B + C2 20 A + B +C3a 14­18 unless current share exceeds 18%, in which case the target is the same as at present 20 Note: A shaded cell indicates that the variable remains unchanged at its initial value in 1999/2000; a blank cell indicates that the variable has the same value as those specified in the simulation scenario A + B + C1. a. The simulation results for this scenario are the same as those for A + B + C2 in all countries except for Angola, Republic of Congo, Eritrea, Kenya, Lesotho, Malawi, Mauritania, Nigeria, Senegal, and Zambia where government revenues as a share of the GDP currently exceed 18 percent. Source: Authors' contruction. rently allocate more than the average share of pub- publicly funded services, the relevant data are as lic revenues for primary education. follows: government revenues average 9.1 percent of GDP, and 31.5 percent of the revenues are de- Overview of Results voted to education. Of total recurrent spending on with Niger as an Example education, 62.0 percent goes to primary education. Currently 4.0 percent of the primary school pu- The simulation results under each of the six simu- pils attend privately funded schools. lation scenarios for the 33 countries appear in As Table 6.2 shows, continuation of the above Annex 5. As a example, consider the results for policies implies that Niger would face an estimated Niger in Table 6.2. The first row shows the initial annual shortfall of US$135 million if it sets itself values of the key simulation variables. With re- the goal of attaining EFA by 2015. If the country gard to the variables related to quality and efficiency embarks on policies to improve the quality of ser- of service delivery, the data are as follows: the pupil- vices--specifically by increasing spending on teacher ratio averages 37 in publicly financed nonteacher salary inputs so that it reaches the in- schools; 25.9 percent of recurrent public spend- dicative benchmark of one-third of total recurrent ing on education currently goes to pay for inputs spending on primary schooling services--the short- other than teachers, public sector teachers are paid fall would rise to US$146 million, assuming that at the rate of 9.6 times the per capita GDP, and all other conditions remain unchanged. If Niger repeaters account for 13.0 percent of all pupils. combines the quality-enhancing measures with With regard to variables affecting the financing of policies to improve the efficiency of service deliv- FINANCING EDUCATION FOR ALL BY 2015: SIMULATIONS FOR 33 AFRICAN COUNTRIES 21 Table 6.2 EFA 2015 financing gap under alternative policy measures, Niger A. Quality measures B. Efficiency measures C. Financing measures Primary education share of Average recurrent % of Annual Pupil- Nonteacher annual Repeaters spending on pupils in shortfall Policy teacher salary teacher as % of As % % for education private in financing, scenario ratio spendinga salaryb enrollments of GDP education (%) schools 2001-2015 c Status quo 37 25.9 9.6 13.0 9.1 31.5 62.0 4.0 135 A only 37 33.3 9.6 146 A + B 40 33.3 4.2 10.0 52 + C1 40 33.3 4.3 10.0 14.0 26.0 50.0 10.0 46 B practice": + C2 20.0 53 A C3 20.0 53 "Good Note: Shaded cells denote that the simulation parameter remains unchanged from the status quo; blank cells indicate that values are the same as those in scenario C1. Except for Angola, Eritrea, Lesotho, and Nigeria, the assumptions for C2 and C3 are identical, but are shown here to maintain a uniform format with the country-specific tables in Annex 5. a. As a percentage of total recurrent spending on primary education. b. As a multiple of the per capita GDP. c. In millions of 2000 US$; refers to the difference between total cost of service delivery (ignoring AIDS-related costs) and the resources domestically mobilized. Source: ery--specifically through three changes: (1) mov- C2) shows the impact for Niger if the share of re- ing the average pupil-teacher ratio to the indicative current spending on education is fixed instead at standard of 40 pupils per teacher by 2015; (2) re- a flat rate of 20 percent. Assuming that all else cruiting new teachers at an average salary of 3.5 remains the same as in the (A+ B + C1) scenario, times the per capita GDP (while keeping the sala- the shortfall in funding for Niger would rise to ries of incumbent teacher at current levels); and US$53 million annually, an increase of US$7 mil- (3) reducing grade repetition to 10 percent by lion from the previous result. As explained 2015--the annual financing gap would drop to elsewhere, the third simulation scenario, (A + B + US$52 million. C3), is relevant only to countries where govern- What would the shortfall in financing be if the ment revenues currently exceed 18 percent of GDP. "good practice" policies in service delivery were Since this is not the case in Niger, the scenario is combined with alternative scenarios of resource the same as scenario (A + B + C2). Although it is mobilization for primary education? In scenario redundant in the table, it is nonetheless shown as (A+ B+ C1), the financing arrangements are as a separate line here in order to maintain a uni- follows: (1) government revenues (net of grants) form format for the 33 country-specific tables in rise to 14.0 percent of GDP by 2015; (2) the share Annex 5. of revenues devoted to education falls to 26 per- cent, the top end of the indicative band; (3) the The Recurrent and Capital Cost Composition of share of recurrent spending on education allocated Niger's Financing Gap to primary schooling falls to 50 percent, and (4) the share of pupils in privately funded schools rises Table 6.3 shows details of the simulation results to 10 percent. Under these conditions, the country's for Niger under "good practice" service delivery external financing gap to reach EFA by 2015 would policies combined with alternative education fi- fall to US$46 million annually. Scenario (A + B + nance scenarios. As before, C2 and C3 are identical 22 SIMULATION RESULTS Table 6.3 EFA 2015 cost estimates and sources of financing under "good practice" policies, niger (millions of 2000 US$) Cost item, Financing sources period during 2001­2015 & Cost of EFA 2015 Domestic resources Gap for external financing simulation scenarios Domestic A+ B +(C1, resources C2 or C3) a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total School C1 796 1,078 403 1,481 796 0 796 282 403 685 services C2 684 1,078 403 1,481 684 0 684 394 403 797 Cumulative, 2001­2015C3 C1 53 72 27 99 53 0 53 19 27 46 Annual C2 46 72 27 99 46 0 46 26 27 53 C3 AIDS- C1 3 3 0 0 3 3 related C2 costs Annual C3 Both C1 74 27 101 53 0 53 22 27 49 items C2 74 27 101 46 0 46 29 27 56 Annual C3 Note: Shaded cells denote not applicable. For specific policies under "good practices," see Table 6.2 for measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. (A + B) refer to simulation assumptions that correspond to "good practice" policies in service delivery, and C1, C2, and C3 refer to alternative assumptions about domestic resource mobilization. Except for Angola, Republic of Congo, Eritrea, Kenya, Lesotho, Malawi, Mauritania, Nigeria, Senegal, and Zambia, C2 and C3 are identical. Source: Authors' simulations. in the case of Niger but separate rows are shown mestic resources for primary education, against the to facilitate reading of the results in Annex 5. The projected annual cost of US$99 million (made up first block in the table relating to school services of US$72 million in recurrent costs, and US$27 refers to cumulative amounts over the 15-year million in capital costs) needed to reach EFA by period of the simulation exercise. Under scenario 2015. The financing gap thus consists of US$19 (A + B + C1), the country mobilizes a cumulative million for recurrent spending, and US$27 million total of US$796 million, just over half of the esti- for capital costs, for an annual total US$46 mil- mated cumulative cost of US$1,481 million to lion. Under the second scenario, the gap rises to achieve EFA by 2015. The domestic resources US$26 million and US$27 million, respectively, for would cover only 73.8 percent of the cumulative a total of US$53 million a year. In Niger, the AIDS- recurrent costs, so external financing would be related costs amount to US$3 million annually. needed to make up the shortfall in recurrent costs Thus, inclusion of this cost component in the to- (US$282 million), as well as cover the total amount tal raises Niger's financing gap to US$49 million of capital costs (US$403 million), for a global annually under scenario (A + B + C1) and US$56 amount of US$685 over the 15-year period. With million under scenario (A + B + C2). the smaller effort for domestic resource mobiliza- tion under scenario (A + B + C2), the external gap Africa's Aggregate Shortfall is about 16 percent larger, US$797 million instead in Funding toward EFA by 2015 of US$685 million. The annualized cost estimates appear in the next Applying the same procedure for Niger to the data block of figures in Table 6.3. In scenario (A + B + for the other 32 African countries, we obtain the C2), Niger raises US$53 million annually in do- country-specific results reported in Annex 5. Be- FINANCING EDUCATION FOR ALL BY 2015: SIMULATIONS FOR 33 AFRICAN COUNTRIES 23 low we present the aggregated estimates for the salaries to 3.5 times the per capita GDP where it 33 countries and their composition in terms of currently exceeds this level--the shortfall in financ- recurrent and capital costs, based on the more ing would drop to US$2.7 billion a year. These detailed summary results reported in Appendix results imply that fully three-fourths of the cost of Tables A2.2 and A2.3. quality improvements can in fact be financed with improvements in the efficiency of service delivery. Aggregate shortfall in funding In the remaining three scenarios, the quality- under the six scenarios and efficiency-enhancing measures are combined with alternative assumptions about domestic re- The results appear in figure 6.1, showing the aver- source mobilization. The shortfall in funding under age annual shortfall between 2001 and 2015. Under scenarios A + B + C1 and A + B + C2 are highly the status quo scenario, the 33 countries would comparable, about US$2.0 billion and US$2.1 bil- fall short by a total of US$2.3 billion annually if lion annually, respectively. Under scenario A + B + they all aim to achieve EFA by 2015. If in the pro- C3, the shortfall in funding falls to US$1.6 billion cess of expanding coverage, these countries also a year between 2001 and 2015, reflecting the ef- put in place measures to improve the quality of fect of the results in 10 countries where government service delivery--mainly by keeping pupil-teacher revenues as a share of GDP persist at their current ratios at no more than 40 on average, allocating at high levels beyond the upper threshold of 18 per- least a third of recurrent spending on primary cent in the other two scenarios. As argued above, schooling for inputs other than teachers, and rais- we believe that scenarios C1 and C2 are probably ing teacher salaries to 3.5 times the per capita where more sustainable in the long run, and that C2 is it is currently below this level--the shortfall in more equitable in terms of its implications for the financing would rise to US$4.1 billion a year, an distribution of donor resources across countries. increase of 80 percent over the estimate under the status quo scenario. Composition of the shortfall in funding In the third simulation scenario, where the qual- between recurrent and capital costs ity-enhancing measures are combined with policies to improve efficiency in service delivery--mainly Table 6.4 summarizes the results for the Africa by reducing grade repetition and lowering teacher region, as well as similar estimates for 14 non-Af- rican low-income countries that, like the countries in Africa, have yet to achieve Education for All as Figure 6.1 of 2000. The simulations assume "good practice" Average annual shortfall in financing to attain EFA by policies in service delivery, combined with one of 2015 in 33 African countries under alternative the three alternative options (C1, C2, or C3) re- simulation assumptions, 2001­2015 garding domestic resource mobilization. As indicated before, measures to enhance quality of service delivery and efficiency are targeted toward 5,000 the same indicative benchmarks by 2015 in all 4,000 countries included in the calculation. Consider first the size of the financing gap for US$ 3,000 the delivery of school services to attain EFA by 2000 of 2015. For the region as a whole, the estimated gap 2,000 ranges from US$1,153 million to US$1,482 mil- Millions lion a year, a difference of some 29 percent between 1,000 the two figures. If external funding of this magni- tude were made available, it would defray between 0 Status A: A + B: 18 and 28 percent of the total cost of service de- quo Quality Quality & A + B + Changes in own resource mobilization livery toward EFA by 2015 in Africa. The estimates improvement efficiency C1 C2 C3 only improvement for Africa are between 3.8 and 4.5 times that of Simulation scenario the corresponding aggregate for the 14 non-Afri- 24 SIMULATION RESULTS Table 6.4 Annual EFA 2015 financing gaps in Sub-Saharan Africa and the rest of the world, 2001­2015 (millions of 2000 US$) Sub-Saharan Africa Rest of the world Cost component and "good practice" scenario A + B + Recurrent cost Capital cost Total Total School services C1 749 733 1,482 339 C2 865 728 1,594 354 C3 568 585 1,153 300 AIDS-related costs C1 556 0 556 -- C2 559 0 559 -- C3 432 0 432 -- Both components C1 1,304 733 2,037 -- C2 1,424 728 2,152 -- C3 1,000 585 1,585 -- -- Not applicable. Source: See text and Annex Tables A5.1­33.1 and A5.1­33.2. can countries that have yet to reach the EFA goal. cise, the AIDS-related costs refer only to two items: The size of the estimate for the region implies that (1) the costs of compensating for increased absen- external funding to support progress toward EFA teeism by teachers stricken by the disease; and (2) by 2015 must necessarily focus overwhelmingly subsidies to support school attendance by the large on the countries in Africa. number of orphans created largely as a result of The second noteworthy feature in Table 6.4 is the disease. The simulations point to an external that external funding will be needed in roughly funding gap of between US$432 and US$559 mil- equal proportion to fund recurrent and capital lion a year to cover the AIDS-related costs alone. costs. This result raises important issues about the Equivalent to between 65 and 76 percent of the mechanisms for channeling donor funding to the financing gap for the recurrent cost of school ser- recipient countries. Such funding has typically vices, this cost is obviously substantial. Including focused on capital investments (for example, the estimate would raise the aggregate external through donor-supported education sector financing gap for the Africa region to between projects) and substantial effort is made to ensure US$1,580 and US$2,032 million annually. that funds are properly processed and spent to achieve project-defined goals. If donors were to Trends in the shortfall in funding extend their support to include recurrent cost fi- between 2001 and 2015 nancing on the scale implied by the cost simulations, the question is whether the current The foregoing results pertain to the average an- arrangements for processing donor funding would nual shortfall during the period. Clearly the remain adequate to ensure accountability for the shortfall will be smaller in earlier than in later years, use of the resources. Clearly, it would become more by virtue of the fact that enrollments will be grow- important than ever to monitor closely the progress ing over time as the EFA goal is attained.2 Figure of EFA-related outcome indicators, including en- 6.2 documents the trends in the annual shortfall try and survival rates in primary schooling; and in funding during the period. The shortfall in capi- the implementation of reforms to improve the tal spending is uneven from year to year, reflecting quality of learning environments and the efficiency the lumpiness of spending; it has therefore been of service delivery. evened out in the graph, averaging around US$720 Finally, Table 6.4 also shows the magnitude of million annually over the simulation period. The the external financing gap associated with the AIDS gap in recurrent spending shows a steady increase epidemic. Recall that for the purpose of this exer- over the period, reflecting increases both for the FINANCING EDUCATION FOR ALL BY 2015: SIMULATIONS FOR 33 AFRICAN COUNTRIES 25 Figure 6.2 recurrent spending overtakes that for capital spend- Trends in the shortfall in financing to attain EFA ing after 2005. The time is therefore fast by 2015 in Africa by cost component, 2001­2015 approaching to reevaluate the effectiveness of cur- rent arrangements--typically via project-based 5,000 mechanisms--for resource transfer from donor Total capital & recurrent spending countries to support progress toward the EFA goal 4,000 School services & in low-income countries. AIDS-related costs US$) 3,000 shortfall Notes 2000 School services of 2,000 1. Recall from Figure 3.1 that in moving toward EFA by Financing (Millions Classroom construction 2015, countries in Groups 2 and 3 would effectively 1,000 move into the same position as countries in Group 1. Thus the service delivery arrangements that have 0 enabled the latter countries to approach or achieve 2001 2005 2010 2015 EFA provide a reasonable benchmark for the other countries to emulate. 2. Recall from Figure 6.1 that the cost of quality-en- delivery of school services and for AIDS-related hancing interventions can largely be financed by spending. The shortfall in total recurrent spend- improving the efficiency of service delivery. Thus, the ing rises from US$529 million a year in 2001, to increasing trends in the financing gap are indeed US$3,154 million a year in 2015. A key feature in driven in large part by the growth in enrollments as the graph is that the external financing gap for countries approach the EFA goal. 7 Conclusion E ducation for All (EFA) by 2015 is among the strate to the constituencies in donor nations that eight explicit Millennium Development Goals the countries themselves are pulling a fair weight defined by the international community to with their own resources, and that the extra amount help implement its commitment to a fuller vision in donor funding would indeed catalyze tangible of development, one that focuses more on outcomes progress toward education for all, rather than be than on inputs alone. EFA occupies a special place wasted in an ineffective delivery system. among these Goals because it is both an objective The simulations produce cost estimates that seek in itself as well as the means for accelerating to reflect the quid pro quo in a mature partnership progress toward the other MDGs. Given the im- between rich and poor countries designed to give portance of achieving EFA, the World Bank has all children the chance to complete at least five committed itself to ensuring that no country with years of primary schooling. Because at least half if a credible and sustainable plan to achieve the goal not more of the aggregate external funding gap would be thwarted from implementing the plan would be for recurrent rather than capital costs, it for lack of external funding. In this paper, we have is important for donors to recognize the possible attempted to estimate the magnitude of external limitations in the traditional mechanisms (such as funding needed to help 33 of Africa's poorest coun- education projects) for channeling resources to tries attain the EFA goal by 2015. In total, the recipient countries. In recent years, the World Bank financing gap ranges between US$1.6 and US$2.1 has introduced new operational instruments (in- billion annually, including financing to defray the cluding, for example, poverty reduction support extra costs imposed on the education system by credits) that make it possible to provide funding the AIDS epidemic. for budget support against progress toward spe- The estimate is an aggregation of country- cific social goals, including those in the education specific simulations that take into account demo- sector. Clearly such instruments could play an in- graphic trends in each country and initial conditions creasing role if donor funding for recurrent cost regarding arrangements for service delivery and financing materializes on the scale estimated in this public funding for primary education. It incorpo- costing exercise. Grant funding also warrant con- rates the influence of policy targets affecting the sideration, at least in some country settings, as a quality of services, the efficiency of service deliv- way to help defray part of the cost of EFA; the is- ery, and the volume of resources that countries sue here is to determine the right balance between themselves mobilize for primary schooling. The grants and loans as alternative mechanisms for re- underlying philosophy in this approach is that while source transfer from the donor countries. lack of resources is undeniably an obstacle to achiev- For their part, when recipient countries accept ing EFA by 2015 in most African countries, poor external funding to support their efforts toward policies also play a role. Thus, to make a persua- the EFA goal, they also implicitly accept a greater sive and credible argument for increased funding participation by donor countries in ensuring that for primary education, it is important to demon- the extra resources are indeed used to provide ser- 27 28 CONCLUSION vices that support student learning in a cost- the country may be not be able--for political efficient manner. Focusing on results would be the economy and other reasons--to realign public poli- hallmark of the partnership between donors and cies in the desired direction, both in terms of the recipient countries, and both parties must obvi- volume of domestic resources mobilized for edu- ously commit the effort and energy to track cation, and the implementation of interventions to outcomes as well as implement the reforms needed improve service delivery. Third, the supply of man- to achieve Education for All by 2015. agement skills may be too thin to assure efficient In conclusion, it bears emphasizing that the main administration of what will inevitably grow to be focus of this paper has been to evaluate the vol- large systems of education as the EFA goal is ume of external funding needed to facilitate the reached. The evidence suggests that even at their achievement of EFA by 2015 in 33 of Africa's poor- current size, the education systems in most Afri- est countries. Issues relating to implementation can countries could be better managed, for example, have not been dealt with in any detail. An immedi- to ensure fewer leakages as resources travel from ate next step concerns the setting up of appropriate the center to the periphery, and that teachers are institutional arrangements for managing the trans- efficiently deployed across schools, and are ad- fer of resources from the donor community to the equately supported and motivated to offer effective recipient countries. Even more challenging is the teaching in the classroom. The international com- task faced by the countries themselves in using the munity would be wise to recognize that these external resources to transform their EFA plans into potential constraints may pose major obstacles to realities on the ground. Three constraints pose po- achieving the EFA goal. It is thus essential to match tential problems in this regard. First, a beneficiary increases in donor funding for EFA with a com- country's capacity to absorb the extra funds may mitment to work alongside the recipient countries be low because of weak budgeting and expendi- to develop and implement viable sector develop- ture management systems, a problem that cannot ment plans, so that the extra resources are used be ignored in light of the pattern of poor disburse- with maximum effectiveness to achieve Education ments in many donor-financed projects. Second, for All. Annex 1 Comparison with Other EFA Costing Estimates O ur estimates of the external financing gap, to note that the Devarajan et al. estimate refers to US$2.1 billion annually (in 2000 prices, or the total incremental cost of achieving EFA by 2015, about US$1.8 billion in 1995 prices) at its whereas our estimate refers to the external financ- highest, is well below the range of US$7­15 bil- ing gap. Since countries themselves would match lion estimated elsewhere, and the range of donor resources by about US$4.0 billion a year in US$10­15 billion (in 1995 prices) in Devarajan et extra spending, the total incremental spending in al.1 One reason for the difference may be that our our simulation amounts to US$6.1 billion annu- estimates refer to only 33 countries in Africa, ally (or about US$5.3 billion in 1995 prices), whereas the Devarajan et al. estimate is for 171 compared with the Devarajan et al. estimate of countries worldwide (with per capita GDP below US$2.1 billion a year. One reason for the smaller US$10,000 in 1995), 48 of which are in Sub- cost in Devarajan et al. is that the authors assume Saharan Africa. If the samples were made compa- that population sizes remain unchanged over time; rable--by focusing only on the estimate for this assumption implies that costs would be un- Sub-Saharan Africa--would there still be large dis- derestimated, but the effect is partly offset by their crepancies? assumption of an instantaneous shift to 100 per- Part of the answer can be found in Table A1.1 cent enrollment. which shows the results of alternative estimates, The upper bound of US$4.94 billion a year in all in 1995 prices, in the Devarajan et al. paper for Devarajan et al. comes close to our estimate of Sub-Saharan Africa and the rest of the world, rang- US$5.3 billion a year, but it is based on a level of ing in total from US$8.6 billion annually to US$27.6 spending per child that is almost certainly too high billion annually.2 For the countries in Sub-Saharan for the countries in our sample--$110.60 per child Africa, the estimate ranges from US$1.27 billion compared with the median per capita GDP of to US$4.94 billion a year. US$413 in the sample of Sub-Saharan African coun- At first sight, the lower bound estimate, based tries. Moreover, since the sum is estimated as the on country-specific unit spending (averaging about total cost of achieving EFA, it tells us nothing about US$48.20 per child), is comparable to the estimate the volume of external funding needed to support of US$2.1 billion (equivalent to about US$1.8 bil- countries toward that goal. lion in 1995 prices) annually under our preferred As indicated in chapter two, UNESCO and scenario, A + B + C2 (where the simulation pa- UNICEF have also proposed estimates of the cost rameters imply a unit spending of US$40.60 per of achiving EFA by 2015. What are we to make of child). Although both the unit spending and ag- this plethora of estimates? Like the Devarajan et gregate estimate appear comparable, it is important al. calculations, these other exercises rely on the 29 30 COMPARRISON WITH OTHER EFA COSTING ESTIMATES Table A1.1 Alternative estimates of the cost of achieving EFA by 2015 in Sub-Saharan Africa and elsewhere in the world (billions of 1995 US$ a year, unless otherwise indicated) Alternative assumptions for level of spending per out-of-school child Median specific to countries with Number of Flat rate for Region-specific "better" and "worse" Country-specific 13% of per Country grouping countries all countries median institutionsa averages capita GDP Countries in Sub-Saharan Africab (assumed unit cost 4.94 2.63 2.15 1.27 per child) 48 (US$110.60) ($58.80) -- (US$48.20) (US$28.50) All other countries 123 6.46 5.97 -- 8.25 26.33 Total 171 11.4 8.6 14.9 10.4 27.6 -- Not available. a. Countries with "better" institutions are those rated above the median on the World Bank's Country Policy and Institutional Assessment (CPIA) index (averaged over 1995­99), while those with "worse" institutions are those rated below the median. As African countries span the two groups, it has not been possible to calculate the estimate for the region under this scenario. b. The aggregate estimate is obtained by multiplying the assumed spending per child multiplied by the estimated 44.6 million out-of-school children in the region. Sources: Shantanayan Devarajan, Margaret J. Miller, and Eric V. Swanson, "Goals for Development: History, Prospects and Costs," World Bank Policy Research Working Paper 2819 (Washington, DC, March 2002); supplemented by Dean Filmer, "Costing the Goal of Universal Primary enrollments by 2015: Back of the (Big) Envelope Calculations," World Bank (Washington, DC, 2002, processed). simple method of applying a range of unit spend- sustained over the next 15 years--would meet the ing per child to the population of out-of-school aggregate shortfall in financing faced by these coun- children to arrive at estimates of the cost of achiev- tries as they seek to achieve that goal under good ing universal primary schooling. While the method practice policy regimes. is disarmingly simple, it blurs the critical distinc- tion between the total cost of achieving EFA and Notes the gap in external financing faced by low-income countries in achieving that goal. The "unit cost" 1. Shatayanan Devarajan, Margaret J. Miller, and Eric V. approach yields unreliable estimates to the extent Swanson, "Goals for Development: History, Prospects that it ignores the dynamics of population and and Costs," World Bank Policy Research Working enrollment growth. Further, because "unit costs" Paper 2819 (Washington, DC, March 2002). are not explicitly linked to underlying service de- 2. Dean Filmer, "Costing the Goal of Universal Primary livery policies that affect them--including policies Enrollments by 2015: Back of the (Big) Envelope on teacher pay, pupil-teacher ratio, and share of Calculations," World Bank (Washington, DC, 2002, processed), a background to the Devarajan et al. pa- spending on nonteacher inputs--the calculations per, also shows an estimate of US$30 billion annually may have limited use for policy dialogue. Thus, on the assumption that countries spend 13 percent even as the campaign to mobilize donor funding of per capita GDP per school-age child; and a figure to achieve universal primary schooling by 2015 of US$320 billion annually based on regression analy- gets under way on the basis of previously an- sis of the relation between spending and the net nounced estimates of US$10 to US$15 billion a enrollment ratio. Both figures are considered too high year, it bears remembering that for the 33 African to be taken seriously; we therefore also dropped these countries in our study, an external funding level from our discussion here. of just US$2.1 billion (in 2000 prices) annually-- Annex 2 Supplementary Figure and Tables Figure A2.1 Primary school completion rates and gross enrollment ratios across African countries, circa 1999 1.0 Group 1 Uganda 0.8 Zambia Gambia cohort* Nigeria Lesotho Ghana Kenya Malawi age Togo in Tanzania 0.6 rate Mauritan Sudan Congo Rep. Burundi Cameroon Group 2 Congo Group 3 Democratic Senegal CóteD'ivoire Completion Rep. Mozambique Benin 0.4 Sierra Leone Eritrea Guinea Guinea-B Angola Rwanda Madagascar Burkina Chad Ethiopia Niger Mali 0.2 Central African Rep. 0.3 0.5 0.7 0.9 1.2 Gross enrollment ratio 31 32 SUPPLEMENTARY FIGURE AND TABLES Table A2.1 Selected Indicators of Primary Education, Low-Income Countries in Africa and Other Regions, circa 1999 Govt. current Govt. current spending on education revenue As percentage Primary GER Completion (excl.grants current govt. (5 or 6years) rate in age as percentage revenues, Percentage Country Year of data (%) cohort (%) of GDP excl. grants of GDP Angola 2000 79.2 29 55.7 4.3 2.4 Bangladesha 2000 112.2 70 12.8 9.6 3.0 Benin 1998 86.4 39 15.3 16.5 2.5 Bolivia 2000 113 77 21.2 25 6.3 Burkina Faso 1998 45.2 25 14.7 17.1 2.5 Burundi 1998 59.7 43 17.4 20.4 3.6 Cambodia 2000 134 61 11.5 15.0 1.7 Cameroon 1999 82 43 15.5 10.8 1.7 C.A.R. 2000 45.3 19 9.6 12.5 1.2 Chad 2000 71.2 19 8.0 20.9 1.7 Congo, Dem. Rep. 2000 60.2 40 10.6 3.2 0.3 Congo, Rep. 2000 84.2 44 26.7 8.6 2.3 Côte d'Ivoire 1999 77.3 40 16.5 21.5 3.5 Eritreaa 1999 52.9 35 34.6 8.0 2.8 Ethiopia 1999 54.6 24 17.8 15.0 2.7 The Gambia 2000 87.7 70 18.5 16.6 3.1 Ghana 1999 79 64 21.8 17.6 3.8 Guinea 2000 62.2 34 11.1 18.1 2.0 Guinea-Bissau 2000 69.7 31 19.6 9.8 1.9 Haiti 1997 112 9.3 17.0 2.0 Honduras 2000 105.4 67 18.1 18.0 3.4 Indiaa 1999 100.6 76 21.2 17.4 2.8 Indonesia 2000 106.3 91 18.4 8.9 2.3 Kenya 1999 91.3 58 24.2 26.2 6.3 Lesotho 2000 102.7 77 35.9 22.2 8.0 Madagascara 1998 89.5 26 10.6 18.8 2.0 Malawi 1999 117 50 18.1 19.8 3.6 Mali 1998 48.9 23 16.8 13.7 2.3 Mauritania 1998 88.4 46 26.5 13.7 3.6 Mongolia 1999 91.5 82 29.2 20.7 7.4 Mozambiquea 1998 78.5 36 11.3 18.1 2.0 Nepala 1998 113.3 57 10.4 11.1 2.6 Nicaragua 2000 100.8 65 30.2 17.4 6.5 Niger 1998 31.2 20 9.1 31.5 2.9 Nigeria 2000 85.1 67 46.1 9.9 4.6 Pakistana 2000 67.1 59 16.7 8.3 1.8 Rwanda 2000 101.1 33 9.8 32.6 3.2 Senegal 2000 69.5 41 18.1 18.6 3.4 Sierra Leone 2000 63.9 37 11.4 30.4 3.5 Sudan 2000 61.1 35 11.1 16.2 1.8 Tanzania 1999 66.4 59 10.9 16.4 1.8 Togo 1999 115 63 14.9 25.6 3.8 Uganda 2000 102 82 10.8 30.1 3.2 Vietnama 1998 109.5 90 22.3 14.7 3.8 Yemen 1998 74 33.5 13.0 5.1 Zambia 1998 84.9 80 18.8 12.3 2.3 Zimbabwe 1997 112 103 27.4 28.3 7.1 b All countries 83.9 51.8 19.1 17.1 3.24 47.4 Africa 76.6 45.1 19.0 17.8 3.04 49.2 World's "Best" 103.1 81.6 20.7 18.5 4.2 44.1 Averages Africa's "Best" 90.4 71.1 25.2 19.3 4.5 44.4 Blank cells indicate data not available. C.A.R. Central African Republic. a. Countries where the primary school cycle lasts five years; the level spending on primary education for these countries is as reported for the five-year cycle. b. World's and Africa's "best" refers to countries with high primary school completion rate in age cohort and gross enrollment ratios; countries are listed in text Table 5.1. Source: Compiled by authors from various sources explained on pages 2-3. FINANCING EDUCATION FOR ALL BY 2015: SIMULATIONS FOR 33 AFRICAN COUNTRIES 33 Table A 2.1 (continued) Govt. current spending on primary education Govt. current Teacher Percentage Percentage of pupils As percentage spending per salary as of recurrent of current pupil as multiple of spending on Pupil- spending on As percentage percentage of per capita inputs other teacher In private education of GDP per capita GDP GDP than teachers ratio schools Repeaters 41.6 1.0 7.8 1.5 19.0 24.4 6.0 25 41.6 1.2 9.4 5.3 70.3 30 62.6 1.6 11.6 4.6 26.4 54.0 10.8 25 34.4 2.2 11.7 2.6 25.4 3 64.0 1.6 23.6 8.0 30.7 48.9 10.8 17.7 35.5 1.3 12.4 5.3 22.1 55.1 0 27.5 51.0 0.9 3.6 1.7 53.3 16.6 66.3 1.1 9.5 3.4 32.5 64.6 19.0 25.9 52.4 0.6 8.7 4.9 28.5 78.9 3.3 32.8 65.5 1.1 10.1 4.8 34.2 72.0 8.8 24.6 65.1 0.2 2.4 0.9 10.3 42.2 10.0 15 36.6 0.8 7 3.4 20.3 61.0 15.2 31.1 49.0 1.7 16 5.7 22.5 46.0 11.6 24.7 53.6 1.5 22.2 7.7 29.6 49.2 10.1 19.4 46.2 1.2 14 6.8 20.5 61.3 5.0 12.8 51.7 1.6 13.2 3.7 24.9 37.0 8.5 10.6 37.2 1.4 12.7 3.6 17.7 34.1 18.0 5 37.2 0.8 8.4 2.7 34.7 48.9 16.1 23.3 35.0 0.7 6.7 1.6 34.3 37.4 8.5 27.1 38.7 0.8 16.2 6.8 46.3 17 29.7 1.0 7.9 2.1 31.7 8 32.1 0.9 10.2 3.4 43.0 6.7 59.3 1.4 10.3 1.6 19.9 5.9 44.2 2.8 17.6 5.3 4.2 31.4 2.2 14.2 40.2 3.2 20.8 6.6 29.9 45.3 0 18.3 54.7 1.1 10.8 3.3 42.4 53.7 22.0 33 49.2 1.8 8.8 4.0 14.0 52.8 2.0 14.7 42.1 1.0 14.3 6.1 31.1 61.0 21.2 17.9 49.0 1.8 13.1 5.1 18.2 48.0 1.8 16 33.6 2.5 15.1 3.9 30.7 0.9 46.4 1.0 7.9 3.2 26.1 54.4 0.0 23.7 53.2 1.4 7.1 2.3 35.9 29.9 48.7 3.2 13.4 1.8 35.9 12 62.0 1.8 35.5 9.6 25.9 36.5 4.0 13 41.0 1.9 13.8 4.9 9.1 39.0 1.0 1 51.8 0.9 13.1 2.5 26.7 3.9 44.7 1.4 9.1 4.0 8.6 52.0 0.8 36.1 43.9 1.5 14.2 4.9 36.6 54.7 10.7 13.6 51.3 1.8 16.4 4.3 33.1 39.6 0 9.3 50.5 0.9 10.3 2.2 22.5 27.5 0 1.2 63.0 1.1 10 3.6 11.2 40.0 0 3.2 48.3 1.8 13.2 4.5 25.2 45.5 35.6 27 53.2 1.7 9.8 2.9 26.2 40.9 2.0 9.8 38.7 1.5 7.1 1.2 30.4 3.5 43.8 2.2 34.7 6.0 21.0 7.0 43.2 1.0 6.9 2.7 21.7 50.0 1.6 6.2 46.1 3.3 19.4 6.1 25.0 39.0 11.0 2 1.47 12.7 4.11 24.1 44.6 8.2 16.0 1.44 12.9 4.5 24.1 47.6 8.2 17.7 1.8 11.9 3.6 25.5 40.1 4.6 9.6 1.9 13.5 4.2 19.1 39.7 4.8 9.3 34 SUPPLEMENTARY FIGURE AND TABLES Table A2.2 Simulations of the Average Annual Volume of Domestic Resources Mobilized for Primary Education and Shortfall in Financing to Attain EFA by 2015 in 33 African Countries Simulation scenario A+B+C1 Simulation scenario A+B+C2 Domestic Shortfall in financing Domestic Shortfall in resources School resources financing Country mobilized services AIDS-related Capital costs Total mobilized School services Nigeria 1,275 92 113 122 328 1,275 92 Ethiopia 141 53 58 133 243 141 58 Congo, Dem. Rep. 54 43 34 103 180 54 43 Sudan 205 85 3 20 107 205 85 Kenya 368 61 38 0 100 316 113 Tanzania 164 52 43 28 123 164 52 Ghana 100 7 12 14 21 100 7 Mozambique 80 23 17 14 54 80 23 Cameroon 175 27 21 28 76 175 27 Côte d'Ivoire 321 26 30 16 72 307 40 Madagascar 58 12 0 21 33 58 12 Uganda 168 27 47 14 88 146 49 Angola 253 36 5 5 46 253 36 Niger 53 19 3 27 46 46 26 Burkina Faso 51 25 22 23 71 51 25 Mali 45 23 4 28 51 45 23 Senegal 103 33 3 19 55 103 33 Zambia 41 16 29 9 54 41 16 Chad 29 10 5 19 34 28 11 Malawi 33 9 20 9 39 33 9 Rwanda 45 5 14 11 30 39 11 Guinea 53 23 2 12 38 53 23 Burundi 13 5 10 11 26 12 6 Benin 52 11 1 10 21 52 11 Togo 35 0 8 7 15 31 5 Sierra Leone 17 2 4 6 11 15 Eritrea 13 3 1 6 11 13 3 Lesotho 38 6 3 1 7 36 8 C.A.R. 15 3 4 7 14 15 3 Congo, Rep. 70 3 3 6 11 70 3 Mauritania 23 4 0 3 8 23 4 Guinea-Bissau 4 2 0 1 4 4 2 The Gambia 9 2 1 1 4 9 2 Total 4,104 749 556 733 2037 3,993 865 C.A.R. Central African Republic. Note: See text Table 6.1 and related text for explanation of the assumptions underlying the three simulation scenarios. Source: Authors' simulations. FINANCING EDUCATION FOR ALL BY 2015: SIMULATIONS FOR 33 AFRICAN COUNTRIES 35 Table A2.2 (continued) Simulation scenario A+B+C3 Shortfall in financing Domestic Shortfall in financing resources School Capital AIDS-related Capital costs Total mobilized services AIDS-related costs Total 113 122 328 2,131 0 0 0 0 60 128 245 164 30 58 133 220 34 103 180 54 43 34 103 180 3 20 105 205 85 3 20 105 38 0 152 397 32 38 0 70 43 28 123 164 52 43 28 123 12 14 33 113 0 12 8 20 17 14 54 80 23 17 14 54 21 28 76 175 27 21 28 76 30 16 86 307 40 30 16 86 0 21 33 58 12 0 21 33 47 14 110 146 49 47 14 110 5 5 46 479 0 0 0 0 3 27 56 46 26 3 27 56 22 23 71 52 23 22 23 69 4 28 55 51 17 4 28 49 3 19 55 111 26 3 19 44 29 9 54 50 8 29 9 45 5 19 30 28 11 5 19 30 20 9 39 38 4 20 9 33 14 11 36 39 11 14 11 36 2 12 38 53 23 2 12 38 10 11 26 14 4 10 11 24 2 10 22 52 11 1 10 21 8 7 20 32 3 8 7 19 4 6 10 15 4 4 6 14 1 6 11 22 0 1 1 2 3 1 12 51 0 0 0 0 4 7 14 15 3 4 7 14 3 6 11 89 0 0 0 0 0 3 8 31 0 0 0 0 0 1 4 5 1 0 1 3 1 1 4 11 0 1 1 2 559 728 2,152 5,278 568 432 585 1,585 36 SUPPLEMENTARY FIGURE AND TABLES Table A2.3 Aggregate Annual Costs and Shortfall in Financing to Attain EFA by 2015 in 33 African Countries under Various Simulation Scenarios, 2001­2015 (millions of 2000 US$) EFA costs, domestic resources mobilized, and shortfall in financing Year and simulation scenarioa 2001 2002 2003 2004 2005 2006 2007 Status Quo School services 2,609.6 2,816.3 3,040.0 3,278.1 3,536.5 3,812.5 4,107.2 of AIDS-related 270.0 302.8 338.1 374.8 415.9 459.7 506.4 2015 A Capital investments 764.4 764.4 764.4 764.4 764.4 764.4 764.4 Costs EF Total 3,644.0 3,883.5 4,142.5 4,417.2 4,716.8 5,036.7 5,377.9 Domestic resources mobilized 2,521.7 2,649.1 2,782.9 2,923.5 3,071.2 3,226.4 3,389.4 Shortfall in financing 1,122.2 1,234.4 1,359.6 1,493.7 1,645.6 1,810.3 1,988.6 Quality-enhancing measures only (scenario A) School services 2,963.7 3,267.9 3,603.5 3,968.3 4,371.0 4,810.2 5,289.3 of AIDS-related 269.8 302.8 338.2 375.1 416.4 460.5 507.4 2015 A Capital investments 958.5 958.5 958.5 958.5 958.5 958.5 958.5 Costs EF Total 4,192.0 4,529.1 4,900.1 5,301.9 5,745.8 6,229.1 6,755.1 Domestic resources mobilized 2,521.7 2,649.1 2,782.9 2,923.5 3,071.2 3,226.4 3,389.4 Shortfall in financing 1,670.2 1,880.0 2,117.2 2,378.4 2,674.6 3,002.8 3,365.7 Quality- and efficiency-enhancing measures (scenario A + B) School services 2,865.1 3,075.1 3,303.2 3,545.8 3,813.2 4,100.9 4,410.6 of AIDS-related 268.2 299.8 333.5 367.3 406.0 447.0 490.3 2015 A Costs Capital investments 790.3 790.3 790.3 790.3 790.3 790.3 790.3 EF Total 3,923.6 4,165.2 4,426.9 4,703.4 5,009.5 5,338.2 5,691.2 Domestic resources mobilized 2,521.8 2,649.1 2,783.0 2,923.5 3,071.2 3,226.4 3,389.5 Shortfall in financing 1,401.8 1,516.0 1,643.9 1,779.9 1,938.2 2,111.7 2,301.8 Quality- and efficiency-enhancing measures with change in domestic resource mobilization (scenario A + B + C1) School services 2,844.7 3,042.5 3,259.2 3,492.0 3,747.7 4,022.2 4,317.1 of AIDS-related 266.0 295.2 327.3 361.6 400.6 442.0 485.9 2015 A Costs Capital investments 729.8 729.8 729.8 729.8 729.8 729.8 729.8 EF Total 3,840.5 4,067.6 4,316.3 4,583.3 4,878.1 5,194.0 5,532.8 Domestic resources mobilized 2,602.7 2,801.7 3,007.0 3,218.0 3,433.9 3,654.1 3,877.5 Shortfall in financing 1,237.8 1,265.9 1,309.3 1,365.4 1,444.1 1,539.9 1,655.3 Quality- and efficiency-enhancing measures with change in domestic resource mobilization (scenario A + B + C2) School services 2848.6 3051.1 3270.9 3503.0 3757.8 4031.3 4324.7 of AIDS-related 268.1 299.6 333.2 367.2 405.8 446.7 490.0 2015 A Costs Capital investments 725.3 725.3 725.3 725.3 725.3 725.3 725.3 EF Total 3842.0 4076.0 4329.4 4595.4 4888.9 5203.3 5540.0 Domestic resources mobilized 2588.0 2777.6 2972.6 3172.4 3376.2 3583.1 3792.2 Shortfall in financing 1254.0 1298.4 1356.8 1423.1 1512.7 1620.1 1747.8 a. See text Table 6.1 for details of simulation assumptions corresponding to A, B and C1 and C2. Source: Author's simulations. FINANCING EDUCATION FOR ALL BY 2015: SIMULATIONS FOR 33 AFRICAN COUNTRIES 37 Year and simulation scenarioa 2008 2009 2010 2011 2012 2013 2014 2015 Average Status Quo 4,421.6 4,757.1 5,114.9 5,458.1 5,822.8 6,210.1 6,621.4 7,058.2 4,577.6 555.9 608.5 664.3 718.1 774.6 834.0 896.3 961.7 578.7 764.4 764.4 764.4 764.4 764.4 764.4 764.4 764.4 764.4 5,742.0 6,130.1 6,543.7 6,940.7 7,361.8 7,808.4 8,282.1 8,784.3 5,920.8 3,560.7 3,740.6 3,929.7 4,128.3 4,337.0 4,556.3 4,786.7 5,028.7 3,642.2 2,181.3 2,389.4 2,614.0 2,812.3 3,024.8 3,252.1 3,495.4 3,755.5 2,278.6 Quality-enhancing measures only (scenario A) 5,811.8 6,381.8 7,003.4 7,626.4 8,303.4 9,039.2 9,839.4 10,710.0 6,199.3 557.2 610.2 666.3 720.3 777.1 836.7 899.4 965.1 580.2 958.5 958.5 958.5 958.5 958.5 958.5 958.5 958.5 958.5 7,327.5 7,950.4 8,628.1 9,305.2 10,038.9 10,834.4 11,697.2 12,633.5 7,737.9 3,560.7 3,740.6 3,929.7 4,128.3 4,337.0 4,556.3 4,786.7 5,028.7 3,642.2 3,766.9 4,209.8 4,698.4 5,176.8 5,701.9 6,278.1 6,910.5 7,604.8 4,095.7 Quality- and efficiency-enhancing measures (scenario A + B) 4,744.1 5,103.2 5,490.1 5,898.2 6,339.1 6,816.0 7,332.1 7,891.2 4,981.9 536.1 584.4 635.3 686.7 740.6 797.3 856.9 919.4 557.9 790.3 790.3 790.3 790.3 790.3 790.3 790.3 790.3 790.3 6,070.5 6,477.9 6,915.7 7,375.1 7,870.0 8,403.6 8,979.3 9,600.8 6,330.1 3,560.8 3,740.7 3,929.8 4,128.5 4,337.2 4,556.5 4,786.9 5,029.0 3,642.3 2,509.7 2,737.2 2,985.9 3,246.6 3,532.8 3,847.1 4,192.4 4,571.8 2,687.8 Quality- and efficiency-enhancing measures with change in domestic resource mobilization (scenario A + B + C1) 4,633.8 4,974.3 5,340.3 5,722.9 6,135.6 6,581.0 7,062.3 7,582.6 4,850.5 532.3 581.4 633.2 684.4 738.2 794.8 854.3 916.7 554.3 729.8 729.8 729.8 729.8 729.8 729.8 729.8 729.8 729.8 5,896.0 6,285.5 6,703.3 7,137.1 7,603.6 8,105.7 8,646.4 9,229.2 6,134.6 4,102.8 4,328.9 4,554.1 4,776.6 4,994.4 5,205.2 5,406.2 5,594.5 4,103.9 1,793.2 1,956.6 2,149.2 2,360.5 2,609.2 2,900.5 3,240.2 3,634.7 2,030.8 Quality- and efficiency-enhancing measures with change in domestic resource mobilization (scenario A + B + C2) 4639.8 4978.1 5341.6 5724.1 6136.5 6581.7 7062.6 7582.6 4855.6 535.7 584.0 634.8 686.1 740.0 796.7 856.1 918.6 557.5 725.3 725.3 725.3 725.3 725.3 725.3 725.3 725.3 725.3 5900.8 6287.4 6701.7 7135.5 7601.9 8103.7 8644.1 9226.5 6138.4 4002.2 4211.6 4418.8 4622.1 4819.1 5007.7 5184.9 5347.7 3991.7 1898.7 2075.8 2282.9 2513.4 2782.7 3096.0 3459.2 3878.8 2146.7 Annex 3 The Simulation Model Table A 3.1 Country Example Illustrating the Simulation Model and the 2015 Target Parameters under Various Scenarios Targets for 2015 under alternative simulation scenariosa A= A + B = A + B + domestic Base quality quality + resource mobilization measures year Status measure efficiency Angola data quob only measures C1 C2 C3 2000 2015 2015 2015 2015 2015 2015 GDP (millions of LCU), 2000 & annual growth rate, 88,642 5.0% 2001­2015 GDP per capita (LCU) 6,970 Exchange rate (LCU/US$) 10 Total population (thousands), 2002 and annual 12,717 2.7% growth rate, 2000­2015 School-age population as % of total population 17.2% 16.8% School age population (in thousands) 2,182 Total public domestic revenue, excl. grants 49,373 (millions of LCU) Public domestic revenue (excl. grants) as % of GDP 55.7% 55.7% 18.0% 18.0% Recurrent spending on education as share 4.3% 4.3% 20.0% 20.0% 20.0% of government revenue Public spending on primary educ. as % of total public 41.6% 41.6% 50.0% 50.0% 50.0% spending on educ. Total public recurrent spending on education 2,122 (millions of LCU) Total recurrent spending on education as % of GDP 2.39% Total domestic public resources for primary education 883 (millions of LCU) Number of pupils enrolled in primary education 1,727,740 (5 or 6 years) Repeaters as a % of total primary school enrollments 25.0% 25.0% 10.0% 10.0% 10.0% 10.0% Target year for % of repeaters to fall to 10% 2010 Completion rate (%) 28.9% 100.0% Intake rate (%) 96.0% 100.0% Target year for intake rate to reach 100% 2010 Gross enrollment ratio (%) (memo item) 79.2% 125% 110% 110% 110% Share of pupils in private schools (%) 6.0% 6.0% 10.0% 10.0% 10.0% Number of pupils in public primary education 1,624,076 (Table continues on the following page.) 39 40 THE SIMULATION MODEL Table A 3.1 (continued) Targets for 2015 under alternative simulation scenariosa A= A + B = A + B + domestic Base quality quality + resource mobilization year Status measure efficiency measures Angola data quob only measures C1 C2 C3 2000 2015 2015 2015 2015 2015 2015 Number of teachers in public primary schools 66,560 (grade 1 to 5 or 6) Attrition rate of teachers (% per annum) 2.5% Number of teachers in pay category 1 66,560 (<=3.5 times per capita GDP) Number of teachers in pay category 2 0 (>3.5 times per capita GDP) Pupil-teacher ratio in public primary education 24.4 24.4 40.0 40.0 40.0 40.0 (average grade 1 to 5 or 6) Number of pupils in private primary schools 103,664 Average annual teacher salary as a multiple of per capita GDP Pay category 1 1.5 1.5 3.5 3.5 3.5 3.5 3.5 Pay category 2 1.5 1.5 3.5 3.5 3.5 3.5 3.5 Total teacher salary bill (millions of LCU) HIVS (% increase to the teacher salary bill) 0.1% 0.0% Spending on inputs other than teachers 23.5% 23.5% 50.0% 50.0% 50.0% 50.0% 50.0% (%of teacher salary bill) Public subsidy for private schools (millions of LCU) Public subsidy per pupil in private schools (LCU) 0 Maternal and double orphans as % of population 1.2% 2.2% Subsidies per maternal and double orphan (US$) 100 Cost per furnished classroom, incl. latrines 8 80 60 60 60 60 60 (thousands of US$, base year; LCU, 2015) Number of teachers per classroom 1.0 a. A cell is left blank where its 2015 target value is the same as in the status quo; see text Table 6.1 for more details. b. Cells containing a value in the status quo column denote simulation parameters for which a value is entered in each simulation. Source: Authors' simulations. Annex 4 Incorporating AIDS-relatedCosts intheSimulations V irtually all African countries today are affected tioned above. First, the projections of the school- by the AIDS epidemic, even though its se- age population used in our simulations are those verity is uneven across the continent. In some supplied by the United Nations Population Divi- countries, the adverse effects on the education sys- sion, and these have been updated by tem are already being felt, and these effects can only demographers using information from the latest be expected to worsen and become more widespread demographic and health surveys. Second, we in- between now and 2015. In the context of this EFA corporated the impact of the epidemic on teacher costing exercise, we have incorporated the effects absenteeism by making use of new findings by of the epidemic by recognizing three channels researchers at Imperial College, UK. That research through which it can affect the cost estimates: suggests that the rate of AIDS infection among teachers is the same as in the adult population, · Size of school-age population. The school-age and that the sickness evolves over a period of 10 population is likely to be smaller in high- years, during which time the infected person is prevalence countries than in the absence of likely to be on sick leave for total of 260 days. the disease. For estimates of the infection rate in the adult · Teacher absenteeism.This is likely to be higher population, we relied on the estimate made by in the presence of the epidemic, and teach- UNAIDS for 2000­2015. Finally, with regard to ers are more likely to die while still in service, the rate of orphanhood, we used the estimates pro- both problems implying an increased need vided by UNAIDS of the number of orphans in for teacher recruitment and training. 1999 in 10 countries and projected the number · Number of orphans. The share of orphans in to 2015 in these countries using a simulation model the school-age population is likely to rise in developed by researchers at Imperial College. We the presence of the epidemic; in light of re- then used the results to extrapolate the estimates search findings by Subbarao et al.1 that to the other African countries, differentiating the maternal and double orphans are at high risk extrapolations according to differences in the of either not enrolling or of dropping out of patterns across subregion (western Africa, east- school, achieving the EFA goal implies that ern Africa, and southern Africa). To obtain the additional funding to encourage schooling total cost of support to orphans in each country, by such children would be needed. we apply a fixed rate of US$50 (in 2000 prices) per year per maternal or double orphan to the The cost simulations presented in this paper estimated population of such children in the incorporate all three AIDS-related effects men- school-age population. 41 42 INCORPORATING AIDS-RELATED COSTS IN THE SIMULATIONS Note 1. Subbarao, K., Angel Mattimore, and Kathrin Plangemann (2001). Social Protection of Africa's Orphans and Other Vulnerable Children. Africa Re- gion Human Development Working Paper Series. Africa Region, World Bank, Washington D.C. Annex 5 Africa Region Aggregate EFA Costing Simulation Results and Detailed Estimates for 33 Countries 43 44 AFRICA REGION AGGREGATE EFA COSTING SIMULATION RESULTS AND DETAILED COUNTRY-SPECIFIC ESTIMATES AFRICA Table A5.1 EFA 2015 financing gap under alternative policy measures A: Quality measures B: Efficiency measures C: Financing measures (%) Average teacher Government revenues Primary Nonteacher salaries Repeaters education % of pupils Annual Policy Pupil-teacher salary in per capita as % of As % of % for recurrent in private financing scenario Ratio spendinga GDP unitb enrollments GDP education spendingc schools gapd Status quo 48 24.1% 4.4 18.4% 18.7% 17.5% 49.3% 8.1% 1,707 A only 38 33.6% 4.7 3,534 A + B 40 33.3% 3.8 9.0% 2,136 C1 40 33.3% 3.8 9.0% 15.0% 21.2% 49.3% 10.0% 1,482 a + B practice": + C2 15.0% 20.0% 1,594 A "Best C3 20.0% 20.0% 1,153 Note: Shade cells denote not applicable. For specific policies under "best practices" see previous table for measures in the three policy domains--quality, efficiency and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. Refers to "best practice" scenarios (A+B+C1), (A+B+C2) or (A+B+C3). Except for Angola, Eritrea, Lesotho, and Nigeria, C2 and C3 are identical. b.As a multiple of the per capita GDP. c. As a percentage of total recurrent education spending d. In millions of 2000 US$; refers to the difference between total cost of service delivery ignoring AIDS-related costs) and the resources domestically AFRICA Table A5.2 EFA 2015 cost estimates and sources of financing under "best practice" policies (millions of 2000 US$) Cost item, period during Domestic Financing sources 2001-2015 & simulation resources Cost of EFA 2015 Domestic resources Gap for external financing scenario A + B + a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total C1 61,558 72,786 10,996 83,782 61,556 2 61,558 11,231 10,993 22,224 y C2 59,876 72,857 10,923 83,781 59,876 0 59,876 12,981 10,923 23,904 deliver Cumulative, 2001-2015 vice C3 79,168 72,786 10,996 83,782 64,268 2,220 66,488 8,518 8,776 17,294 ser C1 4,104 4,852 733 5,585 4,104 0 4,104 749 733 1,482 C2 3,992 4.857 728 5,585 3,992 0 3,992 865 728 1,594 Education Annual C3 5,278 4,852 733 5,585 4,285 148 4,433 568 585 1,513 C1 556 556 0 0 556 556 C2 559 559 0 0 559 559 costs Annual AIDS-related C3 556 556 124 0 432 432 C1 5,408 733 6,141 4,104 0 4,104 1,304 733 2,037 items C2 5,416 728 6,144 3,992 0 3,992 1,424 728 2,152 Annual Both C3 5,408 733 6,141 4,408 148 4,556 1,000 585 1,585 a. See previous table for specific policy measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education--under the "best practice" scenario. FINANCING EDUCATION FOR ALL BY 2015: SIMULATIONS FOR 33 AFRICAN COUNTRIES 45 ANGOLA Table A5.1.1 EFA 2015 Financing Gap Under Alternative Policy Measures A: Quality measures B: Efficiency measures C: Financing measures (%) Average teacher Government revenues Primary Nonteacher salaries Repeaters education % of pupils Annual Policy Pupil-teacher salary in per capita as % of As % of % for recurrent in private financing scenario Ratio spendinga GDP unitb enrollments GDP education spendingc schools gapd Status quo 24 19.0% 1.5 25.0% 55.7% 4.3% 41.6% 6% 87 A only 24 33.3% 3.5 354 A + B 40 33.3% 3.5 10.0% 170 C1 40 33.3% 3.5 10.0% 18.0% 20.0% 50.0% 10.0% 41 a + B practice": + C2 18.0% 20.0% 41 A "Best C3 55.7% 20.0% -185 Note: Shaded cells denote not applicable. For specific policies under "best practices," see previous table for measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. Refers to "best practice" scenarios (A + B + C1), (A + B + C2), or (A + B + C3). Except for Angola, Eritrea, Lesotho, and Nigeria, C2 and C3 are identical. b. As a multiple of the per capita GDP. c. As a percentage of total recurrent education spending. d. In millions of 2000 US$; refers to the difference between total cost of service delivery (ignoring AIDS-related costs) and the resources domestically. ANGOLA Table A5.1.2 EFA 2015 Cost Estimates and Sources of Financing Under "Best Practice" Policies (millions of 2000 US$) Cost item, period during Domestic Financing sources 2001-2015 & simulation resources Cost of EFA 2015 Domestic resources Gap for external financing scenario A + B + a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total C1 3,793 4,337 75 4,412 3,793 0 3,793 544 75 619 y C2 3,793 4,337 75 4,412 3,793 0 3,793 544 75 619 deliver Cumulative, 2001-2015 vice C3 7,191 4,337 75 4,412 4,337 75 4,412 0 0 -2,779 ser C1 253 289 5 294 253 0 253 36 5 41 C2 253 289 5 294 253 0 253 36 5 41 Education Annual C3 479 289 5 294 289 5 294 0 0 0 C1 5 5 0 0 5 5 C2 5 5 0 0 5 5 costs Annual AIDS-related C3 5 5 5 0 0 0 C1 294 5 299 253 0 253 41 5 46 items C2 294 5 299 253 0 253 41 5 46 Annual Both C3 294 5 299 294 5 299 0 0 0 a. See previous table for specific policy measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education--under the "best practice" scenario. 46 AFRICA REGION AGGREGATE EFA COSTING SIMULATION RESULTS AND DETAILED COUNTRY-SPECIFIC ESTIMATES BENIN Table A5.2.1 EFA 2015 Financing Gap under Alternative Policy Measures A: Quality measures B: Efficiency measures C: Financing measures (%) Average teacher Government revenues Primary Nonteacher salaries Repeaters education % of pupils Annual Policy Pupil-teacher salary in per capita as % of As % of % for recurrent in private financing scenario Ratio spendinga GDP unitb enrollments GDP education spendingc schools gapd Status quo 54 26.4% 4.6 25.0% 15.3% 16.5% 62.6% 10.8% 16 A only 40 33.3% 4.6 39 A + B 40 33.3% 3.9 10.0% 20 C1 40 33.3% 3.8 10.0% 16.0% 20.0% 50.0% 10.0% 20 a + B practice": + C2 16.0% 20.0% 20 A "Best C3 16.0% 20.0% 20 Note: Shaded cells denote not applicable. For specific policies under "best practices," see previous table for measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. Refers to "best practice" scenarios (A + B + C1), (A + B + C2), or (A + B + C3). Except for Angola, Eritrea, Lesotho, and Nigeria, C2 and C3 are identical. b. As a multiple of the per capita GDP. c. As a percentage of total recurrent education spending. d. In millions of 2000 US$; refers to the difference between total cost of service delivery (ignoring AIDS-related costs) and the resources domestically. BENIN Table A5.2.2 EFA 2015 Cost Estimates and Sources of Financing Under "Best Practice" Policies (millions of 2000 US$) Cost item, period during Domestic Financing sources 2001-2015 & simulation resources Cost of EFA 2015 Domestic resources Gap for external financing scenario A + B + a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total C1 778 936 145 1,081 778 0 778 158 145 303 y C2 778 936 145 1,081 778 0 778 158 145 303 deliver Cumulative, 2001-2015 vice C3 778 936 145 1,081 778 0 778 158 145 303 ser C1 52 62 10 72 52 0 52 11 10 20 C2 52 62 10 72 52 0 52 11 10 20 Education Annual C3 52 62 10 72 52 0 52 11 10 20 C1 1 1 0 0 1 1 C2 2 2 0 0 2 2 costs Annual AIDS-related C3 1 1 0 0 1 1 C1 63 10 73 52 0 52 11 10 21 items C2 64 10 74 52 0 52 12 10 22 Annual Both C3 63 10 73 52 0 52 11 10 21 a. See previous table for specific policy measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education--under the "best practice" scenario. FINANCING EDUCATION FOR ALL BY 2015: SIMULATIONS FOR 33 AFRICAN COUNTRIES 47 BURKINA FASO Table A5.3.1 EFA 2015 Financing Gap under Alternative Policy Measures A: Quality measures B: Efficiency measures C: Financing measures (%) Average teacher Government revenues Primary Nonteacher salaries Repeaters education % of pupils Annual Policy Pupil-teacher salary in per capita as % of As % of % for recurrent in private financing scenario Ratio spendinga GDP unitb enrollments GDP education spendingc schools gapd Status quo 49 30.7% 8.0 17.7% 14.7% 17.1% 64.0% 10.8% 79 A only 40 33.3% 8.0 106 A + B 40 33.3% 4.2 10.0% 44 C1 40 33.3% 4.2 10.0% 14.0% 20.0% 50.0% 10.0% 48 a + B practice": + C2 14.0% 20.0% 48 A "Best C3 18.8% 20.0% 47 Note: Shaded cells denote not applicable. For specific policies under "best practices," see previous table for measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. Refers to "best practice" scenarios (A + B + C1), (A + B + C2), or (A + B + C3). Except for Angola, Eritrea, Lesotho, and Nigeria, C2 and C3 are identical. b. As a multiple of the per capita GDP. c. As a percentage of total recurrent education spending. d. In millions of 2000 US$; refers to the difference between total cost of service delivery (ignoring AIDS-related costs) and the resources domestically. BURKINA FASO Table A5.3.2 EFA 2015 Cost Estimates and Sources of Financing Under "Best Practice" Policies (millions of 2000 US$) Cost item, period during Domestic Financing sources 2001-2015 & simulation resources Cost of EFA 2015 Domestic resources Gap for external financing scenario A + B + a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total C1 764 1,138 351 1,489 764 0 764 374 351 725 y C2 764 1,138 351 1,489 764 0 764 374 351 725 deliver Cumulative, 2001-2015 vice C3 787 1,138 351 1,489 787 0 787 351 351 703 ser C1 51 76 23 99 51 0 51 25 23 48 C2 51 76 23 99 51 0 51 25 23 48 Education Annual C3 52 76 23 99 52 0 52 23 23 47 C1 22 22 0 0 22 22 C2 22 22 0 0 22 22 costs Annual AIDS-related C3 22 22 0 0 22 22 C1 98 23 122 51 0 51 47 23 71 items C2 98 23 122 51 0 51 47 23 71 Annual Both C3 98 23 122 52 0 52 46 23 69 a. See previous table for specific policy measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education--under the "best practice" scenario. 48 AFRICA REGION AGGREGATE EFA COSTING SIMULATION RESULTS AND DETAILED COUNTRY-SPECIFIC ESTIMATES BURUNDI Table A5.4.1 EFA 2015 Financing Gap under Alternative Policy Measures A: Quality measures B: Efficiency measures C: Financing measures (%) Average teacher Government revenues Primary Nonteacher salaries Repeaters education % of pupils Annual Policy Pupil-teacher salary in per capita as % of As % of % for recurrent in private financing scenario Ratio spendinga GDP unitb enrollments GDP education spendingc schools gapd Status quo 55 22.1% 5.3 27.5% 17.4% 20.4% 35.5% 0.0% 18 A only 40 33.3% 5.3 33 A + B 40 33.3% 3.9 10.0% 21 C1 40 33.3% 3.9 10.0% 14.0% 20.4% 50.0% 10.0% 16 a + B practice": + C2 14.0% 20.0% 16 A "Best C3 17.4% 20.0% 15 Note: Shaded cells denote not applicable. For specific policies under "best practices," see previous table for measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. Refers to "best practice" scenarios (A + B + C1), (A + B + C2), or (A + B + C3). Except for Angola, Eritrea, Lesotho, and Nigeria, C2 and C3 are identical. b. As a multiple of the per capita GDP. c. As a percentage of total recurrent education spending. d. In millions of 2000 US$; refers to the difference between total cost of service delivery (ignoring AIDS-related costs) and the resources domestically. BURUNDI Table A5.4.2 EFA 2015 Cost Estimates and Sources of Financing Under "Best Practice" Policies (millions of 2000 US$) Cost item, period during Domestic Financing sources 2001-2015 & simulation resources Cost of EFA 2015 Domestic resources Gap for external financing scenario A + B + a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total C1 189 270 163 433 189 0 189 81 163 244 y C2 187 270 163 433 187 0 187 84 163 247 deliver Cumulative, 2001-2015 vice C3 214 270 163 433 214 0 214 56 163 219 ser C1 13 18 11 29 13 0 13 5 11 16 C2 12 18 11 29 12 0 12 6 11 16 Education Annual C3 14 18 11 29 14 0 14 4 11 15 C1 10 10 0 0 10 10 C2 10 10 0 0 10 10 costs Annual AIDS-related C3 10 10 0 0 10 10 C1 28 11 39 13 0 13 15 11 26 items C2 28 11 39 12 0 12 15 11 26 Annual Both C3 28 11 39 14 0 14 14 11 24 a. See previous table for specific policy measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education--under the "best practice" scenario. FINANCING EDUCATION FOR ALL BY 2015: SIMULATIONS FOR 33 AFRICAN COUNTRIES 49 CAMEROON Table A5.5.1 EFA 2015 Financing Gap under Alternative Policy Measures A: Quality measures B: Efficiency measures C: Financing measures (%) Average teacher Government revenues Primary Nonteacher salaries Repeaters education % of pupils Annual Policy Pupil-teacher salary in per capita as % of As % of % for recurrent in private financing scenario Ratio spendinga GDP unitb enrollments GDP education spendingc schools gapd Status quo 65 32.5% 3.4 25.9% 15.5% 10.8% 66.3% 19.0% 24 A only 40 33.3% 3.5 100 A + B 40 33.3% 3.6 10.0% 82 C1 40 33.3% 3.6 10.0% 16.0% 20.0% 50.0% 10.0% 55 a + B practice": + C2 16.0% 20.0% 55 A "Best C3 16.0% 20.0% 55 Note: Shaded cells denote not applicable. For specific policies under "best practices," see previous table for measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. Refers to "best practice" scenarios (A + B + C1), (A + B + C2), or (A + B + C3). Except for Angola, Eritrea, Lesotho, and Nigeria, C2 and C3 are identical. b. As a multiple of the per capita GDP. c. As a percentage of total recurrent education spending. d. In millions of 2000 US$; refers to the difference between total cost of service delivery (ignoring AIDS-related costs) and the resources domestically. CAMEROON Table A5.5.2 EFA 2015 Cost Estimates and Sources of Financing Under "Best Practice" Policies (millions of 2000 US$) Cost item, period during Domestic Financing sources 2001-2015 & simulation resources Cost of EFA 2015 Domestic resources Gap for external financing scenario A + B + a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total C1 2,625 3,029 420 3,450 2,625 0 2,625 405 420 825 y C2 2,625 3,029 420 3,450 2,625 0 2,625 405 420 825 deliver Cumulative, 2001-2015 vice C3 2,625 3,029 420 3,450 2,625 0 2,625 405 420 825 ser C1 175 202 28 230 175 0 175 27 28 55 C2 175 202 28 230 175 0 175 27 28 55 Education Annual C3 175 202 28 230 175 0 175 27 28 55 C1 21 21 0 0 21 21 C2 21 21 0 0 21 21 costs Annual AIDS-related C3 21 21 0 0 21 21 C1 223 28 251 175 0 175 48 28 76 items C2 223 28 251 175 0 175 48 28 76 Annual Both C3 223 28 251 175 0 175 48 28 76 a. See previous table for specific policy measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education--under the "best practice" scenario. 50 AFRICA REGION AGGREGATE EFA COSTING SIMULATION RESULTS AND DETAILED COUNTRY-SPECIFIC ESTIMATES CENTRAL AFRICAN REPUBLIC Table A5.6.1 EFA 2015 Financing Gap under Alternative Policy Measures A: Quality measures B: Efficiency measures C: Financing measures (%) Average teacher Government revenues Primary Nonteacher salaries Repeaters education % of pupils Annual Policy Pupil-teacher salary in per capita as % of As % of % for recurrent in private financing scenario Ratio spendinga GDP unitb enrollments GDP education spendingc schools gapd Status quo 79 28.5% 4.9 32.8% 9.6% 12.5% 52.4% 3.3% 14 A only 40 33.3% 4.9 32 A + B 40 33.3% 3.7 10.0% 19 C1 40 33.3% 3.5 10.0% 14.0% 20.0% 50.0% 10.0% 10 a + B practice": + C2 14.0% 20.0% 10 A "Best C3 14.0% 20.0% 10 Note: Shaded cells denote not applicable. For specific policies under "best practices," see previous table for measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. Refers to "best practice" scenarios (A + B + C1), (A + B + C2), or (A + B + C3). Except for Angola, Eritrea, Lesotho, and Nigeria, C2 and C3 are identical. b. As a multiple of the per capita GDP. c. As a percentage of total recurrent education spending. d. In millions of 2000 US$; refers to the difference between total cost of service delivery (ignoring AIDS-related costs) and the resources domestically. CENTRAL AFRICAN REPUBLIC Table A5.6.2 EFA 2015 Cost Estimates and Sources of Financing Under "Best Practice" Policies (millions of 2000 US$) Cost item, period during Domestic Financing sources 2001-2015 & simulation resources Cost of EFA 2015 Domestic resources Gap for external financing scenario A + B + a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total C1 232 273 103 376 232 0 232 41 103 143 y C2 232 273 103 376 232 0 232 41 103 143 deliver Cumulative, 2001-2015 vice C3 232 273 103 376 232 0 232 41 103 143 ser C1 15 18 7 25 15 0 15 3 7 10 C2 15 18 7 25 15 0 15 3 7 10 Education Annual C3 15 18 7 25 15 0 15 3 7 10 C1 4 4 0 0 4 4 C2 4 4 0 0 4 4 costs Annual AIDS-related C3 4 4 0 0 4 4 C1 22 7 29 15 0 15 7 7 14 items C2 22 7 29 15 0 15 7 7 14 Annual Both C3 22 7 29 15 0 15 7 7 14 a. See previous table for specific policy measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education--under the "best practice" scenario. FINANCING EDUCATION FOR ALL BY 2015: SIMULATIONS FOR 33 AFRICAN COUNTRIES 51 CHAD Table A5.7.1 EFA 2015 Financing Gap under Alternative Policy Measures A: Quality measures B: Efficiency measures C: Financing measures (%) Average teacher Government revenues Primary Nonteacher salaries Repeaters education % of pupils Annual Policy Pupil-teacher salary in per capita as % of As % of % for recurrent in private financing scenario Ratio spendinga GDP unitb enrollments GDP education spendingc schools gapd Status quo 72 34.2% 4.8 24.6% 8.0% 20.9% 65.5% 8.8% 21 A only 40 33.3% 4.8 48 A + B 40 33.3% 3.9 10.0% 36 C1 40 33.3% 3.9 10.0% 14.0% 20.9% 50.0% 10.0% 30 a + B practice": + C2 14.0% 20.0% 30 A "Best C3 14.0% 20.0% 30 Note: Shaded cells denote not applicable. For specific policies under "best practices," see previous table for measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. Refers to "best practice" scenarios (A + B + C1), (A + B + C2), or (A + B + C3). Except for Angola, Eritrea, Lesotho, and Nigeria, C2 and C3 are identical. b. As a multiple of the per capita GDP. c. As a percentage of total recurrent education spending. d. In millions of 2000 US$; refers to the difference between total cost of service delivery (ignoring AIDS-related costs) and the resources domestically. CHAD Table A5.7.2 EFA 2015 Cost Estimates and Sources of Financing Under "Best Practice" Policies (millions of 2000 US$) Cost item, period during Domestic Financing sources 2001-2015 & simulation resources Cost of EFA 2015 Domestic resources Gap for external financing scenario A + B + a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total C1 430 584 290 873 430 0 430 154 290 444 y C2 418 584 290 873 418 0 418 165 290 455 deliver Cumulative, 2001-2015 vice C3 418 584 290 873 418 0 418 165 290 455 ser C1 29 39 19 58 29 0 29 10 19 30 C2 28 39 19 58 28 0 28 11 19 30 Education Annual C3 28 39 19 58 28 0 28 11 19 30 C1 5 5 0 0 5 5 C2 5 5 0 0 5 5 costs Annual AIDS-related C3 5 5 0 0 5 5 C1 44 19 63 29 0 29 15 19 34 items C2 44 19 63 28 0 28 16 19 35 Annual Both C3 44 19 63 28 0 28 16 19 35 a. See previous table for specific policy measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education--under the "best practice" scenario. 52 AFRICA REGION AGGREGATE EFA COSTING SIMULATION RESULTS AND DETAILED COUNTRY-SPECIFIC ESTIMATES CONGO DEMOCATIC REPUBLIC Table A5.8.1 EFA 2015 Financing Gap under Alternative Policy Measures A: Quality measures B: Efficiency measures C: Financing measures (%) Average teacher Government revenues Primary Nonteacher salaries Repeaters education % of pupils Annual Policy Pupil-teacher salary in per capita as % of As % of % for recurrent in private financing scenario Ratio spendinga GDP unitb enrollments GDP education spendingc schools gapd Status quo 42 10.3% 0.9 15.0% 10.6% 3.2% 65.1% 10.0% 109 A only 40 33.3% 3.5 198 A + B 40 33.3% 3.5 10.0% 187 C1 40 33.3% 3.5 10.0% 14.0% 20.0% 50.0% 10.0% 146 a + B practice": + C2 14.0% 20.0% 146 A "Best C3 14.0% 20.0% 146 Note: Shaded cells denote not applicable. For specific policies under "best practices," see previous table for measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. Refers to "best practice" scenarios (A + B + C1), (A + B + C2), or (A + B + C3). Except for Angola, Eritrea, Lesotho, and Nigeria, C2 and C3 are identical. b. As a multiple of the per capita GDP. c. As a percentage of total recurrent education spending. d. In millions of 2000 US$; refers to the difference between total cost of service delivery (ignoring AIDS-related costs) and the resources domestically. CONGO DEMOCATIC REPUBLIC Table A5.8.2 EFA 2015 Cost Estimates and Sources of Financing Under "Best Practice" Policies (millions of 2000 US$) Cost item, period during Domestic Financing sources 2001-2015 & simulation resources Cost of EFA 2015 Domestic resources Gap for external financing scenario A + B + a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total C1 814 1,456 1,542 2,998 814 0 814 642 1,542 2,185 y C2 814 1,456 1,542 2,998 814 0 814 642 1,542 2,185 deliver Cumulative, 2001-2015 vice C3 814 1,456 1,542 2,998 814 0 814 642 1,542 2,185 ser C1 54 97 103 200 54 0 54 43 103 146 C2 54 97 103 200 54 0 54 43 103 146 Education Annual C3 54 97 103 200 54 0 54 43 103 146 C1 34 34 0 0 34 34 C2 34 34 0 0 34 34 costs Annual AIDS-related C3 34 34 0 0 34 34 C1 131 103 234 54 0 54 77 103 180 items C2 131 103 234 54 0 54 77 103 180 Annual Both C3 131 103 234 54 0 54 77 103 180 a. See previous table for specific policy measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education--under the "best practice" scenario. FINANCING EDUCATION FOR ALL BY 2015: SIMULATIONS FOR 33 AFRICAN COUNTRIES 53 CONGO REPUBLIC Table A5.9.1 EFA 2015 Financing Gap under Alternative Policy Measures A: Quality measures B: Efficiency measures C: Financing measures (%) Average teacher Government revenues Primary Nonteacher salaries Repeaters education % of pupils Annual Policy Pupil-teacher salary in per capita as % of As % of % for recurrent in private financing scenario Ratio spendinga GDP unitb enrollments GDP education spendingc schools gapd Status quo 61 20.3% 3.4 31.1% 26.7% 8.6% 36.6% 15.2% 17 A only 40 33.3% 3.5 47 A + B 40 33.3% 3.5 10.0% 34 C1 40 33.3% 3.5 10.0% 18.0% 20.0% 50.0% 10.0% 9 a + B practice": + C2 18.0% 20.0% 9 A "Best C3 26.7% 20.0% -11 Note: Shaded cells denote not applicable. For specific policies under "best practices," see previous table for measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. Refers to "best practice" scenarios (A + B + C1), (A + B + C2), or (A + B + C3). Except for Angola, Eritrea, Lesotho, and Nigeria, C2 and C3 are identical. b. As a multiple of the per capita GDP. c. As a percentage of total recurrent education spending. d. In millions of 2000 US$; refers to the difference between total cost of service delivery (ignoring AIDS-related costs) and the resources domestically. CONGO REPUBLIC Table A5.9.2 EFA 2015 Cost Estimates and Sources of Financing Under "Best Practice" Policies (millions of 2000 US$) Cost item, period during Domestic Financing sources 2001-2015 & simulation resources Cost of EFA 2015 Domestic resources Gap for external financing scenario A + B + a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total C1 1,046 1,086 89 1,175 1,046 0 1,046 40 89 129 y C2 1,046 1,086 89 1,175 1,046 0 1,046 40 89 129 deliver Cumulative, 2001-2015 vice C3 1,342 1,086 89 1,175 1,086 89 1,175 0 0 -167 ser C1 70 72 6 78 70 0 70 3 6 9 C2 70 72 6 78 70 0 70 3 6 9 Education Annual C3 89 72 6 78 72 6 78 0 0 0 C1 3 3 0 0 3 3 C2 3 3 0 0 3 3 costs Annual AIDS-related C3 3 3 3 0 0 0 C1 75 6 81 70 0 70 5 6 11 items C2 75 6 81 70 0 70 5 6 11 Annual Both C3 75 6 81 75 6 81 0 0 0 a. See previous table for specific policy measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education--under the "best practice" scenario. 54 AFRICA REGION AGGREGATE EFA COSTING SIMULATION RESULTS AND DETAILED COUNTRY-SPECIFIC ESTIMATES CÔTE D'IVOIRE Table A5.10.1 EFA 2015 Financing Gap under Alternative Policy Measures A: Quality measures B: Efficiency measures C: Financing measures (%) Average teacher Government revenues Primary Nonteacher salaries Repeaters education % of pupils Annual Policy Pupil-teacher salary in per capita as % of As % of % for recurrent in private financing scenario Ratio spendinga GDP unitb enrollments GDP education spendingc schools gapd Status quo 46 22.5% 5.7 24.7% 16.5% 21.5% 49.0% 11.6% 102 A only 40 33.3% 5.7 187 A + B 40 33.3% 4.3 10.0% 60 C1 40 33.3% 4.3 10.0% 18.0% 21.5% 50.0% 10.0% 42 a + B practice": + C2 18.0% 20.0% 56 A "Best C3 18.0% 20.0% 56 Note: Shaded cells denote not applicable. For specific policies under "best practices," see previous table for measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. Refers to "best practice" scenarios (A + B + C1), (A + B + C2), or (A + B + C3). Except for Angola, Eritrea, Lesotho, and Nigeria, C2 and C3 are identical. b. As a multiple of the per capita GDP. c. As a percentage of total recurrent education spending. d. In millions of 2000 US$; refers to the difference between total cost of service delivery (ignoring AIDS-related costs) and the resources domestically. CÔTE D'IVOIRE Table A5.10.2 EFA 2015 Cost Estimates and Sources of Financing Under "Best Practice" Policies (millions of 2000 US$) Cost item, period during Domestic Financing sources 2001-2015 & simulation resources Cost of EFA 2015 Domestic resources Gap for external financing scenario A + B + a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total C1 4,815 5,205 237 5,442 4,815 0 4,815 391 237 628 y C2 4,603 5,205 237 5,442 4,603 0 4,603 603 237 840 deliver Cumulative, 2001-2015 vice C3 4,603 5,205 237 5,442 4,603 0 4,603 603 237 840 ser C1 321 347 16 363 321 0 321 26 16 42 C2 307 347 16 363 307 0 307 40 16 56 Education Annual C3 307 347 16 363 307 0 307 40 16 56 C1 30 30 0 0 30 30 C2 30 30 0 0 30 30 costs Annual AIDS-related C3 30 30 0 0 30 30 C1 377 16 393 321 0 321 56 16 72 items C2 377 16 393 307 0 307 70 16 86 Annual Both C3 377 16 393 307 0 307 70 16 86 a. See previous table for specific policy measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education--under the "best practice" scenario. FINANCING EDUCATION FOR ALL BY 2015: SIMULATIONS FOR 33 AFRICAN COUNTRIES 55 ERITREA Table A5.11.1 EFA 2015 Financing Gap under Alternative Policy Measures A: Quality measures B: Efficiency measures C: Financing measures (%) Average teacher Government revenues Primary Nonteacher salaries Repeaters education % of pupils Annual Policy Pupil-teacher salary in per capita as % of As % of % for recurrent in private financing scenario Ratio spendinga GDP unitb enrollments GDP education spendingc schools gapd Status quo 49 29.6% 7.7 19.4% 34.6% 8.0% 53.6% 10.1% 16 A only 40 33.3% 7.7 22 A + B 40 33.3% 4.3 10.0% 10 C1 40 33.3% 4.3 10.0% 14.0% 20.0% 42.0% 10.0% 10 a + B practice": + C2 14.0% 20.0% 10 A "Best C3 34.6% 20.0% 1 Note: Shaded cells denote not applicable. For specific policies under "best practices," see previous table for measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. Refers to "best practice" scenarios (A + B + C1), (A + B + C2), or (A + B + C3). Except for Angola, Eritrea, Lesotho, and Nigeria, C2 and C3 are identical. b. As a multiple of the per capita GDP. c. As a percentage of total recurrent education spending. d. In millions of 2000 US$; refers to the difference between total cost of service delivery (ignoring AIDS-related costs) and the resources domestically. ERITREA Table A5.11.2 EFA 2015 Cost Estimates and Sources of Financing Under "Best Practice" Policies (millions of 2000 US$) Cost item, period during Domestic Financing sources 2001-2015 & simulation resources Cost of EFA 2015 Domestic resources Gap for external financing scenario A + B + a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total C1 198 250 94 344 198 0 198 52 94 146 y C2 198 250 94 344 198 0 198 52 94 146 deliver Cumulative, 2001-2015 vice C3 325 250 94 344 250 75 325 0 19 19 ser C1 13 17 6 23 13 0 13 3 6 10 C2 13 17 6 23 13 0 13 3 6 10 Education Annual C3 22 17 6 23 17 5 22 0 1 1 C1 1 1 0 0 1 1 C2 1 1 0 0 1 1 costs Annual AIDS-related C3 1 1 0 0 1 1 C1 18 6 24 13 0 13 4 6 11 items C2 18 6 24 13 0 13 4 6 11 Annual Both C3 18 6 24 17 5 22 1 1 2 a. See previous table for specific policy measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education--under the "best practice" scenario. 56 AFRICA REGION AGGREGATE EFA COSTING SIMULATION RESULTS AND DETAILED COUNTRY-SPECIFIC ESTIMATES ETHIOPIA Table A5.12.1 EFA 2015 Financing Gap under Alternative Policy Measures A: Quality measures B: Efficiency measures C: Financing measures (%) Average teacher Government revenues Primary Nonteacher salaries Repeaters education % of pupils Annual Policy Pupil-teacher salary in per capita as % of As % of % for recurrent in private financing scenario Ratio spendinga GDP unitb enrollments GDP education spendingc schools gapd Status quo 61 20.5% 6.8 12.8% 17.8% 15.0% 46.2% 5.0% 159 A only 40 33.3% 6.8 332 A + B 40 33.3% 4.0 10.0% 214 C1 40 33.3% 4.1 10.0% 14.0% 20.0% 50.0% 10.0% 186 a + B practice": + C2 14.0% 20.0% 186 A "Best C3 17.8% 20.0% 163 Note: Shaded cells denote not applicable. For specific policies under "best practices," see previous table for measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. Refers to "best practice" scenarios (A + B + C1), (A + B + C2), or (A + B + C3). Except for Angola, Eritrea, Lesotho, and Nigeria, C2 and C3 are identical. b. As a multiple of the per capita GDP. c. As a percentage of total recurrent education spending. d. In millions of 2000 US$; refers to the difference between total cost of service delivery (ignoring AIDS-related costs) and the resources domestically. ETHIOPIA Table A5.12.2 EFA 2015 Cost Estimates and Sources of Financing Under "Best Practice" Policies (millions of 2000 US$) Cost item, period during Domestic Financing sources 2001-2015 & simulation resources Cost of EFA 2015 Domestic resources Gap for external financing scenario A + B + a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total C1 2,113 2,908 1,989 4,897 2,113 0 2,113 795 1,989 2,784 y C2 2,113 2,979 1,917 4,896 2,113 0 2,113 867 1,917 2,783 deliver Cumulative, 2001-2015 vice C3 2,454 2,908 1,989 4,897 2,454 0 2,454 455 1,989 2,444 ser C1 141 194 133 326 141 0 141 53 133 186 C2 141 199 128 326 141 0 141 58 128 186 Education Annual C3 164 194 133 326 164 0 164 30 133 163 C1 58 58 0 0 58 58 C2 60 60 0 0 60 60 costs Annual AIDS-related C3 58 58 0 0 58 58 C1 251 133 384 141 0 141 111 133 243 items C2 258 128 386 141 0 141 118 128 245 Annual Both C3 251 133 384 164 0 164 88 133 220 a. See previous table for specific policy measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education--under the "best practice" scenario. FINANCING EDUCATION FOR ALL BY 2015: SIMULATIONS FOR 33 AFRICAN COUNTRIES 57 THE GAMBIA Table A5.13.1 EFA 2015 Financing Gap under Alternative Policy Measures A: Quality measures B: Efficiency measures C: Financing measures (%) Average teacher Government revenues Primary Nonteacher salaries Repeaters education % of pupils Annual Policy Pupil-teacher salary in per capita as % of As % of % for recurrent in private financing scenario Ratio spendinga GDP unitb enrollments GDP education spendingc schools gapd Status quo 37 24.9% 3.7 10.6% 18.5% 16.6% 51.7% 8.5% 3 A only 37 33.3% 3.7 4 A + B 40 33.3% 3.6 10.0% 3 C1 40 33.3% 3.6 10.0% 14.0% 20.0% 50.0% 10.0% 3 a + B practice": + C2 14.0% 20.0% 3 A "Best C3 18.5% 20.0% 2 Note: Shaded cells denote not applicable. For specific policies under "best practices," see previous table for measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. Refers to "best practice" scenarios (A + B + C1), (A + B + C2), or (A + B + C3). Except for Angola, Eritrea, Lesotho, and Nigeria, C2 and C3 are identical. b. As a multiple of the per capita GDP. c. As a percentage of total recurrent education spending. d. In millions of 2000 US$; refers to the difference between total cost of service delivery (ignoring AIDS-related costs) and the resources domestically. THE GAMBIA Table A5.13.2 EFA 2015 Cost Estimates and Sources of Financing Under "Best Practice" Policies (millions of 2000 US$) Cost item, period during Domestic Financing sources 2001-2015 & simulation resources Cost of EFA 2015 Domestic resources Gap for external financing scenario A + B + a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total C1 140 170 17 187 140 0 140 31 17 47 y C2 140 170 17 187 140 0 140 31 17 47 deliver Cumulative, 2001-2015 vice C3 164 170 17 187 164 0 164 7 17 23 ser C1 9 11 1 12 9 0 9 2 1 3 C2 9 11 1 12 9 0 9 2 1 3 Education Annual C3 11 11 1 12 11 0 11 0 1 2 C1 1 1 0 0 1 1 C2 1 1 0 0 1 1 costs Annual AIDS-related C3 1 1 0 0 1 1 C1 12 1 13 9 0 9 3 1 4 items C2 12 1 13 9 0 9 3 1 4 Annual Both C3 12 1 13 11 0 11 1 1 2 a. See previous table for specific policy measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education--under the "best practice" scenario. 58 AFRICA REGION AGGREGATE EFA COSTING SIMULATION RESULTS AND DETAILED COUNTRY-SPECIFIC ESTIMATES GHANA Table A5.14.1 EFA 2015 Financing Gap under Alternative Policy Measures A: Quality measures B: Efficiency measures C: Financing measures (%) Average teacher Government revenues Primary Nonteacher salaries Repeaters education % of pupils Annual Policy Pupil-teacher salary in per capita as % of As % of % for recurrent in private financing scenario Ratio spendinga GDP unitb enrollments GDP education spendingc schools gapd Status quo 34 17.7% 3.6 5.0% 21.8% 17.6% 37.2% 18.0% 29 A only 34 33.3% 3.6 44 A + B 40 33.3% 3.5 5.0% 25 C1 40 33.3% 3.5 5.0% 18.0% 20.0% 50.0% 10.0% 21 a + B practice": + C2 18.0% 20.0% 21 A "Best C3 21.8% 20.0% 8 Note: Shaded cells denote not applicable. For specific policies under "best practices," see previous table for measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. Refers to "best practice" scenarios (A + B + C1), (A + B + C2), or (A + B + C3). Except for Angola, Eritrea, Lesotho, and Nigeria, C2 and C3 are identical. b. As a multiple of the per capita GDP. c. As a percentage of total recurrent education spending. d. In millions of 2000 US$; refers to the difference between total cost of service delivery (ignoring AIDS-related costs) and the resources domestically. GHANA Table A5.14.2 EFA 2015 Cost Estimates and Sources of Financing Under "Best Practice" Policies (millions of 2000 US$) Cost item, period during Domestic Financing sources 2001-2015 & simulation resources Cost of EFA 2015 Domestic resources Gap for external financing scenario A + B + a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total C1 1,499 1,604 207 1,812 1,499 0 1,499 105 207 312 y C2 1,499 1,604 207 1,812 1,499 0 1,499 105 207 312 deliver Cumulative, 2001-2015 vice C3 1,692 1,604 207 1,812 1,604 88 1,692 0 120 120 ser C1 100 107 14 121 100 0 100 7 14 21 C2 100 107 14 121 100 0 100 7 14 21 Education Annual C3 113 107 14 121 107 6 113 0 8 8 C1 12 12 0 0 12 12 C2 12 12 0 0 12 12 costs Annual AIDS-related C3 12 12 0 0 12 12 C1 119 14 133 100 0 100 19 14 33 items C2 119 14 133 100 0 100 19 14 33 Annual Both C3 119 14 133 107 6 113 12 8 20 a. See previous table for specific policy measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education--under the "best practice" scenario. FINANCING EDUCATION FOR ALL BY 2015: SIMULATIONS FOR 33 AFRICAN COUNTRIES 59 GUINEA-BUISSAU Table A5.16.1 EFA 2015 Financing Gap under Alternative Policy Measures A: Quality measures B: Efficiency measures C: Financing measures (%) Average teacher Government revenues Primary Nonteacher salaries Repeaters education % of pupils Annual Policy Pupil-teacher salary in per capita as % of As % of % for recurrent in private financing scenario Ratio spendinga GDP unitb enrollments GDP education spendingc schools gapd Status quo 37 34.3% 1.6 27.1% 19.6% 9.8% 35.0% 8.5% 3 A only 37 34.3% 3.5 7 A + B 40 33.3% 3.5 10.0% 5 C1 40 33.3% 3.5 10.0% 14.0% 20.0% 50.0% 10.0% 3 a + B practice": + C2 14.0% 20.0% 3 A "Best C3 14.0% 20.0% 2 Note: Shaded cells denote not applicable. For specific policies under "best practices," see previous table for measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. Refers to "best practice" scenarios (A + B + C1), (A + B + C2), or (A + B + C3). Except for Angola, Eritrea, Lesotho, and Nigeria, C2 and C3 are identical. b. As a multiple of the per capita GDP. c. As a percentage of total recurrent education spending. d. In millions of 2000 US$; refers to the difference between total cost of service delivery (ignoring AIDS-related costs) and the resources domestically. GUINEA-BUISSAU Table A5.16.2 EFA 2015 Cost Estimates and Sources of Financing Under "Best Practice" Policies (millions of 2000 US$) Cost item, period during Domestic Financing sources 2001-2015 & simulation resources Cost of EFA 2015 Domestic resources Gap for external financing scenario A + B + a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total C1 55 87 18 105 55 0 55 32 18 50 y C2 55 87 18 105 55 0 55 32 18 50 deliver Cumulative, 2001-2015 vice C3 68 87 18 105 68 0 68 19 18 37 ser C1 4 6 1 7 4 0 4 2 1 3 C2 4 6 1 7 4 0 4 2 1 3 Education Annual C3 5 6 1 7 5 0 5 1 1 2 C1 0 0 0 0 0 0 C2 0 0 0 0 0 0 costs Annual AIDS-related C3 0 0 0 0 0 0 C1 6 1 7 4 0 4 3 1 4 items C2 6 1 7 4 0 4 3 1 4 Annual Both C3 6 1 7 5 0 5 2 1 3 a. See previous table for specific policy measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education--under the "best practice" scenario. 60 AFRICA REGION AGGREGATE EFA COSTING SIMULATION RESULTS AND DETAILED COUNTRY-SPECIFIC ESTIMATES KENYA Table A5.17.1 EFA 2015 Financing Gap under Alternative Policy Measures A: Quality measures B: Efficiency measures C: Financing measures (%) Average teacher Government revenues Primary Nonteacher salaries Repeaters education % of pupils Annual Policy Pupil-teacher salary in per capita as % of As % of % for recurrent in private financing scenario Ratio spendinga GDP unitb enrollments GDP education spendingc schools gapd Status quo 31 4.2% 5.3 14.2% 24.2% 26.2% 44.2% 2.2% 53 A only 31 33.3% 5.3 178 A + B 40 33.3% 4.7 10.0% 16 C1 40 33.3% 4.8 10.0% 16.0% 26.2% 50.0% 10.0% 61 a + B practice": + C2 16.0% 20.0% 113 A "Best C3 24.2% 20.0% 32 Note: Shaded cells denote not applicable. For specific policies under "best practices," see previous table for measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. Refers to "best practice" scenarios (A + B + C1), (A + B + C2), or (A + B + C3). Except for Angola, Eritrea, Lesotho, and Nigeria, C2 and C3 are identical. b. As a multiple of the per capita GDP. c. As a percentage of total recurrent education spending. d. In millions of 2000 US$; refers to the difference between total cost of service delivery (ignoring AIDS-related costs) and the resources domestically. KENYA Table A5.17.2 EFA 2015 Cost Estimates and Sources of Financing Under "Best Practice" Policies (millions of 2000 US$) Cost item, period during Domestic Financing sources 2001-2015 & simulation resources Cost of EFA 2015 Domestic resources Gap for external financing scenario A + B + a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total C1 5,517 6,436 0 6,436 5,517 0 5,517 919 0 919 y C2 4,734 6,436 0 6,436 4,734 0 4,734 1,702 0 1,702 deliver Cumulative, 2001-2015 vice C3 5,959 6,436 0 6,436 5,959 0 5,959 477 0 477 ser C1 368 429 0 429 368 0 368 61 0 61 C2 316 429 0 429 316 0 316 113 0 113 Education Annual C3 397 429 0 429 397 0 397 32 0 32 C1 38 0 38 0 0 38 38 C2 38 0 38 0 0 38 38 costs Annual AIDS-related C3 38 0 38 0 0 38 38 C1 4670 0 467 368 0 368 99 0 99 items C2 4670 0 467 316 0 316 152 0 152 Annual Both C3 467 0 467 397 0 397 70 0 70 a. See previous table for specific policy measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education--under the "best practice" scenario. FINANCING EDUCATION FOR ALL BY 2015: SIMULATIONS FOR 33 AFRICAN COUNTRIES 61 LESOTHO Table A5.18.1 EFA 2015 Financing Gap under Alternative Policy Measures A: Quality measures B: Efficiency measures C: Financing measures (%) Average teacher Government revenues Primary Nonteacher salaries Repeaters education % of pupils Annual Policy Pupil-teacher salary in per capita as % of As % of % for recurrent in private financing scenario Ratio spendinga GDP unitb enrollments GDP education spendingc schools gapd Status quo 45 29.9% 6.6 18.3% 35.9% 22.2% 40.2% 0.0% 5 A only 40 33.3% 6.6 12 A + B 40 33.3% 5.0 10.0% 0 C1 40 33.3% 5.2 10.0% 18.0% 22.2% 50.0% 10.0% 7 a + B practice": + C2 18.0% 20.0% 9 A "Best C3 35.9% 20.0% -6 Note: Shaded cells denote not applicable. For specific policies under "best practices," see previous table for measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. Refers to "best practice" scenarios (A + B + C1), (A + B + C2), or (A + B + C3). Except for Angola, Eritrea, Lesotho, and Nigeria, C2 and C3 are identical. b. As a multiple of the per capita GDP. c. As a percentage of total recurrent education spending. d. In millions of 2000 US$; refers to the difference between total cost of service delivery (ignoring AIDS-related costs) and the resources domestically. LESOTHO Table A5.18.2 EFA 2015 Cost Estimates and Sources of Financing Under "Best Practice" Policies (millions of 2000 US$) Cost item, period during Domestic Financing sources 2001-2015 & simulation resources Cost of EFA 2015 Domestic resources Gap for external financing scenario A + B + a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total C1 565 659 15 674 565 0 565 94 15 108 y C2 534 659 15 674 534 0 534 125 15 140 deliver Cumulative, 2001-2015 vice C3 763 659 15 674 659 15 674 0 0 -90 ser C1 38 44 1 45 38 0 38 6 1 7 C2 36 44 1 45 36 0 36 8 1 9 Education Annual C3 51 44 1 45 44 1 45 0 0 0 C1 3 3 0 0 3 3 C2 3 3 0 0 3 3 costs Annual AIDS-related C3 3 3 3 0 0 0 C1 46 1 47 38 0 38 9 1 10 items C2 46 1 47 36 0 36 11 1 12 Annual Both C3 46 1 47 46 1 47 0 0 0 a. See previous table for specific policy measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education--under the "best practice" scenario. 62 AFRICA REGION AGGREGATE EFA COSTING SIMULATION RESULTS AND DETAILED COUNTRY-SPECIFIC ESTIMATES MADAGASCAR Table A5.19.1 EFA 2015 Financing Gap under Alternative Policy Measures A: Quality measures B: Efficiency measures C: Financing measures (%) Average teacher Government revenues Primary Nonteacher salaries Repeaters education % of pupils Annual Policy Pupil-teacher salary in per capita as % of As % of % for recurrent in private financing scenario Ratio spendinga GDP unitb enrollments GDP education spendingc schools gapd Status quo 54 42.4% 3.3 33.0% 10.6% 18.8% 54.7% 22.0% 23 A only 40 42.4% 3.5 50 A + B 40 33.3% 3.5 10.0% 25 C1 40 33.3% 3.5 10.0% 14.0% 20.0% 42.0 10.0% 33 a + B practice": + C2 14.0% 20.0% 33 A "Best C3 14.0% 20.0% 33 Note: Shaded cells denote not applicable. For specific policies under "best practices," see previous table for measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. Refers to "best practice" scenarios (A + B + C1), (A + B + C2), or (A + B + C3). Except for Angola, Eritrea, Lesotho, and Nigeria, C2 and C3 are identical. b. As a multiple of the per capita GDP. c. As a percentage of total recurrent education spending. d. In millions of 2000 US$; refers to the difference between total cost of service delivery (ignoring AIDS-related costs) and the resources domestically. MADAGASCAR Table A5.19.2 EFA 2015 Cost Estimates and Sources of Financing Under "Best Practice" Policies (millions of 2000 US$) Cost item, period during Domestic Financing sources 2001-2015 & simulation resources Cost of EFA 2015 Domestic resources Gap for external financing scenario A + B + a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total C1 874 1,053 314 1,367 874 0 874 180 314 494 y C2 874 1,053 314 1,367 874 0 874 180 314 494 deliver Cumulative, 2001-2015 vice C3 874 1,053 314 1,367 874 0 874 180 314 494 ser C1 58 70 21 91 58 0 58 12 21 33 C2 58 70 21 91 58 0 58 12 21 33 Education Annual C3 58 70 21 91 58 0 58 12 21 33 C1 0 0 0 0 0 0 C2 0 0 0 0 0 0 costs Annual AIDS-related C3 0 0 0 0 0 0 C1 70 21 91 58 0 58 12 21 33 items C2 70 21 91 58 0 58 12 21 33 Annual Both C3 70 21 91 58 0 58 12 21 33 a. See previous table for specific policy measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education--under the "best practice" scenario. FINANCING EDUCATION FOR ALL BY 2015: SIMULATIONS FOR 33 AFRICAN COUNTRIES 63 MALAWI Table A5.20.1 EFA 2015 Financing Gap under Alternative Policy Measures A: Quality measures B: Efficiency measures C: Financing measures (%) Average teacher Government revenues Primary Nonteacher salaries Repeaters education % of pupils Annual Policy Pupil-teacher salary in per capita as % of As % of % for recurrent in private financing scenario Ratio spendinga GDP unitb enrollments GDP education spendingc schools gapd Status quo 53 14.0% 4.0 14.7% 18.1% 19.8% 49.2% 2.0% 2 A only 40 33.3% 4.0 24 A + B 40 33.3% 3.7 10.0% 19 C1 40 33.3% 3.7 10.0% 14.0% 20.0% 50.0% 10.0% 19 a + B practice": + C2 14.0% 20.0% 19 A "Best C3 18.1% 20.0% 13 Note: Shaded cells denote not applicable. For specific policies under "best practices," see previous table for measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. Refers to "best practice" scenarios (A + B + C1), (A + B + C2), or (A + B + C3). Except for Angola, Eritrea, Lesotho, and Nigeria, C2 and C3 are identical. b. As a multiple of the per capita GDP. c. As a percentage of total recurrent education spending. d. In millions of 2000 US$; refers to the difference between total cost of service delivery (ignoring AIDS-related costs) and the resources domestically. MALAWI Table A5.20.2 EFA 2015 Cost Estimates and Sources of Financing Under "Best Practice" Policies (millions of 2000 US$) Cost item, period during Domestic Financing sources 2001-2015 & simulation resources Cost of EFA 2015 Domestic resources Gap for external financing scenario A + B + a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total C1 491 633 136 769 491 0 491 142 136 278 y C2 491 633 136 769 491 0 491 142 136 278 deliver Cumulative, 2001-2015 vice C3 571 633 136 769 571 0 571 62 136 198 ser C1 33 42 9 51 33 0 33 9 9 19 C2 33 42 9 51 33 0 33 9 9 19 Education Annual C3 38 42 9 51 38 0 38 4 9 13 C1 20 20 0 0 20 20 C2 20 20 0 0 20 20 costs Annual AIDS-related C3 20 20 0 0 20 20 C1 62 9 71 33 0 33 30 9 39 items C2 62 9 71 33 0 33 30 9 39 Annual Both C3 62 9 71 38 0 38 24 9 33 a. See previous table for specific policy measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education--under the "best practice" scenario. 64 AFRICA REGION AGGREGATE EFA COSTING SIMULATION RESULTS AND DETAILED COUNTRY-SPECIFIC ESTIMATES MALI Table A5.21.1 EFA 2015 Financing Gap under Alternative Policy Measures A: Quality measures B: Efficiency measures C: Financing measures (%) Average teacher Government revenues Primary Nonteacher salaries Repeaters education % of pupils Annual Policy Pupil-teacher salary in per capita as % of As % of % for recurrent in private financing scenario Ratio spendinga GDP unitb enrollments GDP education spendingc schools gapd Status quo 61 31.1% 6.1 17.9% 16.8% 13.7% 42.1% 21.2% 54 A only 40 33.3% 6.1 91 A + B 40 33.3% 3.8 10.0% 53 C1 40 33.3% 3.8 10.0% 14.0% 20.0% 50.0% 10.0% 51 a + B practice": + C2 14.0% 20.0% 51 A "Best C3 16.8% 20.0% 45 Note: Shaded cells denote not applicable. For specific policies under "best practices," see previous table for measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. Refers to "best practice" scenarios (A + B + C1), (A + B + C2), or (A + B + C3). Except for Angola, Eritrea, Lesotho, and Nigeria, C2 and C3 are identical. b. As a multiple of the per capita GDP. c. As a percentage of total recurrent education spending. d. In millions of 2000 US$; refers to the difference between total cost of service delivery (ignoring AIDS-related costs) and the resources domestically. MALI Table A5.21.2 EFA 2015 Cost Estimates and Sources of Financing Under "Best Practice" Policies (millions of 2000 US$) Cost item, period during Domestic Financing sources 2001-2015 & simulation resources Cost of EFA 2015 Domestic resources Gap for external financing scenario A + B + a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total C1 676 1,014 427 1,442 676 0 676 339 427 766 y C2 676 1,014 427 1,442 676 0 676 339 427 766 deliver Cumulative, 2001-2015 vice C3 762 1,014 427 1,442 762 0 762 252 427 679 ser C1 45 68 28 96 45 0 45 23 28 51 C2 45 68 28 96 45 0 45 23 28 51 Education Annual C3 51 68 28 96 51 0 51 17 28 45 C1 4 4 0 0 4 4 C2 4 4 0 0 4 4 costs Annual AIDS-related C3 4 4 0 0 4 4 C1 71 28 100 45 0 45 26 28 55 items C2 71 28 100 45 0 45 26 28 55 Annual Both C3 71 28 100 51 0 51 20 28 49 a. See previous table for specific policy measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education--under the "best practice" scenario. FINANCING EDUCATION FOR ALL BY 2015: SIMULATIONS FOR 33 AFRICAN COUNTRIES 65 MAURITANIA Table A5.22.1 EFA 2015 Financing Gap under Alternative Policy Measures A: Quality measures B: Efficiency measures C: Financing measures (%) Average teacher Government revenues Primary Nonteacher salaries Repeaters education % of pupils Annual Policy Pupil-teacher salary in per capita as % of As % of % for recurrent in private financing scenario Ratio spendinga GDP unitb enrollments GDP education spendingc schools gapd Status quo 48 18.2% 5.1 16.0% 26.5% 13.7% 49.0% 1.8% 6 A only 40 33.3% 5.1 16 A + B 40 33.3% 4.0 10.0% 8 C1 40 33.3% 4.0 10.0% 16.0% 20.0% 50.0% 10.0% 7 a + B practice": + C2 16.0% 20.0% 7 A "Best C3 26.5% 20.0% -1 Note: Shaded cells denote not applicable. For specific policies under "best practices," see previous table for measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. Refers to "best practice" scenarios (A + B + C1), (A + B + C2), or (A + B + C3). Except for Angola, Eritrea, Lesotho, and Nigeria, C2 and C3 are identical. b. As a multiple of the per capita GDP. c. As a percentage of total recurrent education spending. d. In millions of 2000 US$; refers to the difference between total cost of service delivery (ignoring AIDS-related costs) and the resources domestically. MAURITANIA Table A5.22.2 EFA 2015 Cost Estimates and Sources of Financing Under "Best Practice" Policies (millions of 2000 US$) Cost item, period during Domestic Financing sources 2001-2015 & simulation resources Cost of EFA 2015 Domestic resources Gap for external financing scenario A + B + a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total C1 343 406 48 454 343 0 343 63 48 111 y C2 343 406 48 454 343 0 343 63 48 111 deliver Cumulative, 2001-2015 vice C3 466 406 48 454 406 48 454 0 0 -12 ser C1 23 27 3 30 23 0 23 4 3 7 C2 23 27 3 30 23 0 23 4 3 7 Education Annual C3 31 27 3 30 27 3 30 0 0 0 C1 0 0 0 0 0 0 C2 0 0 0 0 0 0 costs Annual AIDS-related C3 0 0 0 0 0 0 C1 27 3 30 23 0 23 4 3 8 items C2 27 3 30 23 0 23 4 3 8 Annual Both C3 27 3 30 27 3 30 0 0 0 a. See previous table for specific policy measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education--under the "best practice" scenario. 66 AFRICA REGION AGGREGATE EFA COSTING SIMULATION RESULTS AND DETAILED COUNTRY-SPECIFIC ESTIMATES MOZAMBIQUE Table A5.23.1 EFA 2015 Financing Gap under Alternative Policy Measures A: Quality measures B: Efficiency measures C: Financing measures (%) Average teacher Government revenues Primary Nonteacher salaries Repeaters education % of pupils Annual Policy Pupil-teacher salary in per capita as % of As % of % for recurrent in private financing scenario Ratio spendinga GDP unitb enrollments GDP education spendingc schools gapd Status quo 54 26.1% 3.2 23.7% 11.3% 18.1% 46.4% 0.0% 27 A only 40 33.3% 3.5 75 A + B 40 33.3% 3.5 10.0% 59 C1 40 33.3% 3.5 10.0% 14.0% 20.0% 42.0% 10.0% 37 a + B practice": + C2 14.0% 20.0% 37 A "Best C3 14.0% 20.0% 37 Note: Shaded cells denote not applicable. For specific policies under "best practices," see previous table for measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. Refers to "best practice" scenarios (A + B + C1), (A + B + C2), or (A + B + C3). Except for Angola, Eritrea, Lesotho, and Nigeria, C2 and C3 are identical. b. As a multiple of the per capita GDP. c. As a percentage of total recurrent education spending. d. In millions of 2000 US$; refers to the difference between total cost of service delivery (ignoring AIDS-related costs) and the resources domestically. MOZAMBIQUE Table A5.23.2 EFA 2015 Cost Estimates and Sources of Financing Under "Best Practice" Policies (millions of 2000 US$) Cost item, period during Domestic Financing sources 2001-2015 & simulation resources Cost of EFA 2015 Domestic resources Gap for external financing scenario A + B + a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total C1 1,206 1,552 209 1,761 1,206 0 1,206 346 209 555 y C2 1,206 1,552 209 1,761 1,206 0 1,206 346 209 555 deliver Cumulative, 2001-2015 vice C3 1,206 1,552 209 1,761 1,206 0 1,206 346 209 555 ser C1 80 103 14 117 80 0 80 23 14 37 C2 80 103 14 117 80 0 80 23 14 37 Education Annual C3 80 103 14 117 80 0 80 23 14 37 C1 17 17 0 0 17 17 C2 17 17 0 0 17 17 costs Annual AIDS-related C3 17 17 0 0 17 17 C1 120 14 134 80 0 80 40 14 54 items C2 120 14 134 80 0 80 40 14 54 Annual Both C3 120 14 134 80 0 80 40 14 54 a. See previous table for specific policy measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education--under the "best practice" scenario. FINANCING EDUCATION FOR ALL BY 2015: SIMULATIONS FOR 33 AFRICAN COUNTRIES 67 NIGER Table A5.24.1 EFA 2015 Financing Gap under Alternative Policy Measures A: Quality measures B: Efficiency measures C: Financing measures (%) Average teacher Government revenues Primary Nonteacher salaries Repeaters education % of pupils Annual Policy Pupil-teacher salary in per capita as % of As % of % for recurrent in private financing scenario Ratio spendinga GDP unitb enrollments GDP education spendingc schools gapd Status quo 37 25.9% 9.6 13.0% 9.1% 31.5% 62.0% 4.0% 135 A only 37 33.3% 9.6 146 A + B 40 33.3% 4.2 10.0% 52 C1 40 33.3% 4.3 10.0% 14.0% 26.0% 50.0% 10.0% 46 a + B practice": + C2 14.0% 20.0% 53 A "Best C3 14.0% 20.0% 53 Note: Shaded cells denote not applicable. For specific policies under "best practices," see previous table for measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. Refers to "best practice" scenarios (A + B + C1), (A + B + C2), or (A + B + C3). Except for Angola, Eritrea, Lesotho, and Nigeria, C2 and C3 are identical. b. As a multiple of the per capita GDP. c. As a percentage of total recurrent education spending. d. In millions of 2000 US$; refers to the difference between total cost of service delivery (ignoring AIDS-related costs) and the resources domestically. NIGER Table A5.24.2 EFA 2015 Cost Estimates and Sources of Financing Under "Best Practice" Policies (millions of 2000 US$) Cost item, period during Domestic Financing sources 2001-2015 & simulation resources Cost of EFA 2015 Domestic resources Gap for external financing scenario A + B + a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total C1 796 1,078 403 1,481 796 0 796 282 403 685 y C2 684 1,078 403 1,481 684 0 684 394 403 797 deliver Cumulative, 2001-2015 vice C3 684 1,078 403 1,481 684 0 684 394 403 797 ser C1 53 72 27 99 53 0 53 19 27 46 C2 46 72 27 99 46 0 46 26 27 53 Education Annual C3 46 72 27 99 46 0 46 26 27 53 C1 3 3 0 0 3 3 C2 3 3 0 0 3 3 costs Annual AIDS-related C3 3 3 0 0 3 3 C1 74 27 101 53 0 53 21 27 48 items C2 74 27 101 46 0 46 29 27 56 Annual Both C3 74 27 101 46 0 46 29 27 56 a. See previous table for specific policy measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education--under the "best practice" scenario. 68 AFRICA REGION AGGREGATE EFA COSTING SIMULATION RESULTS AND DETAILED COUNTRY-SPECIFIC ESTIMATES NIGERIA Table A5.25.1 EFA 2015 Financing Gap under Alternative Policy Measures A: Quality measures B: Efficiency measures C: Financing measures (%) Average teacher Government revenues Primary Nonteacher salaries Repeaters education % of pupils Annual Policy Pupil-teacher salary in per capita as % of As % of % for recurrent in private financing scenario Ratio spendinga GDP unitb enrollments GDP education spendingc schools gapd Status quo 39 9.1% 4.9 1.0% 46.1% 9.9% 41.0% 1.0% 352 A only 39 33.3% 4.9 654 A + B 40 33.3% 4.1 1.0% 425 C1 40 33.3% 4.1 1.0% 18.0% 20.0% 50.0% 10.0% 214 a + B practice": + C2 18.0% 20.0% 214 A "Best C3 46.1% 20.0% -642 Note: Shaded cells denote not applicable. For specific policies under "best practices," see previous table for measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. Refers to "best practice" scenarios (A + B + C1), (A + B + C2), or (A + B + C3). Except for Angola, Eritrea, Lesotho, and Nigeria, C2 and C3 are identical. b. As a multiple of the per capita GDP. c. As a percentage of total recurrent education spending. d. In millions of 2000 US$; refers to the difference between total cost of service delivery (ignoring AIDS-related costs) and the resources domestically. NIGERIA Table A5.25.2 EFA 2015 Cost Estimates and Sources of Financing Under "Best Practice" Policies (millions of 2000 US$) Cost item, period during Domestic Financing sources 2001-2015 & simulation resources Cost of EFA 2015 Domestic resources Gap for external financing scenario A + B + a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total C1 19,123 20,502 1,831 22,333 19,123 0 19,123 1,380 1,831 3,210 y C2 19,123 20,502 1,831 22,333 19,123 0 19,123 1,380 1,831 3,210 deliver Cumulative, 2001-2015 vice C3 31,966 20,502 1,831 22,333 20,502 1,831 22,333 0 0 -9,633 ser C1 1,275 1,367 122 1,489 1,275 0 1,275 92 122 214 C2 1,275 1,367 122 1,489 1,275 0 1,275 92 122 214 Education Annual C3 2,131 1,367 122 1,489 1,367 122 1,489 0 0 0 C1 113 113 0 0 113 113 C2 113 113 0 0 113 113 costs Annual AIDS-related C3 113 113 113 0 0 0 C1 1,480 122 1,602 1,275 0 1,275 205 122 327 items C2 1,480 122 1,602 1,275 0 1,275 205 122 327 Annual Both C3 1,480 122 1,602 1,480 122 1,602 0 0 0 a. See previous table for specific policy measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education--under the "best practice" scenario. FINANCING EDUCATION FOR ALL BY 2015: SIMULATIONS FOR 33 AFRICAN COUNTRIES 69 RWANDA Table A5.26.1 EFA 2015 Financing Gap under Alternative Policy Measures A: Quality measures B: Efficiency measures C: Financing measures (%) Average teacher Government revenues Primary Nonteacher salaries Repeaters education % of pupils Annual Policy Pupil-teacher salary in per capita as % of As % of % for recurrent in private financing scenario Ratio spendinga GDP unitb enrollments GDP education spendingc schools gapd Status quo 52 8.6% 4.0 36.1% 9.8% 32.6% 44.7% 0.8% 18 A only 40 33.3% 4.0 47 A + B 40 33.3% 3.7 10.0% 27 C1 40 33.3% 3.7 10.0% 14.0% 26.0% 50.0% 10.0% 15 a + B practice": + C2 14.0% 20.0% 21 A "Best C3 14.0% 20.0% 21 Note: Shaded cells denote not applicable. For specific policies under "best practices," see previous table for measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. Refers to "best practice" scenarios (A + B + C1), (A + B + C2), or (A + B + C3). Except for Angola, Eritrea, Lesotho, and Nigeria, C2 and C3 are identical. b. As a multiple of the per capita GDP. c. As a percentage of total recurrent education spending. d. In millions of 2000 US$; refers to the difference between total cost of service delivery (ignoring AIDS-related costs) and the resources domestically. RWANDA Table A5.26.2 EFA 2015 Cost Estimates and Sources of Financing Under "Best Practice" Policies (millions of 2000 US$) Cost item, period during Domestic Financing sources 2001-2015 & simulation resources Cost of EFA 2015 Domestic resources Gap for external financing scenario A + B + a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total C1 682 739 165 904 682 0 682 58 165 222 y C2 591 739 165 904 591 0 591 148 165 313 deliver Cumulative, 2001-2015 vice C3 591 739 165 904 591 0 591 148 165 313 ser C1 45 49 11 60 45 0 45 4 11 15 C2 39 49 11 60 39 0 39 10 11 21 Education Annual C3 39 49 11 60 39 0 39 10 11 21 C1 14 14 0 0 14 14 C2 14 14 0 0 14 14 costs Annual AIDS-related C3 14 14 0 0 14 14 C1 63 11 74 45 0 45 18 11 29 items C2 63 11 74 39 0 39 24 11 35 Annual Both C3 63 11 74 39 0 39 24 11 35 a. See previous table for specific policy measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education--under the "best practice" scenario. 70 AFRICA REGION AGGREGATE EFA COSTING SIMULATION RESULTS AND DETAILED COUNTRY-SPECIFIC ESTIMATES SENEGAL Table A5.27.1 EFA 2015 Financing Gap under Alternative Policy Measures A: Quality measures B: Efficiency measures C: Financing measures (%) Average teacher Government revenues Primary Nonteacher salaries Repeaters education % of pupils Annual Policy Pupil-teacher salary in per capita as % of As % of % for recurrent in private financing scenario Ratio spendinga GDP unitb enrollments GDP education spendingc schools gapd Status quo 55 36.6% 4.9 13.6% 18.1% 18.6% 43.9% 10.7% 48 A only 40 33.3% 4.9 80 A + B 40 33.3% 4.0 10.0% 56 C1 40 33.3% 4.0 10.0% 16.0% 20.0% 50.0% 10.0% 52 a + B practice": + C2 16.0% 20.0% 52 A "Best C3 18.1% 20.0% 44 Note: Shaded cells denote not applicable. For specific policies under "best practices," see previous table for measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. Refers to "best practice" scenarios (A + B + C1), (A + B + C2), or (A + B + C3). Except for Angola, Eritrea, Lesotho, and Nigeria, C2 and C3 are identical. b. As a multiple of the per capita GDP. c. As a percentage of total recurrent education spending. d. In millions of 2000 US$; refers to the difference between total cost of service delivery (ignoring AIDS-related costs) and the resources domestically. SENEGAL Table A5.27.2 EFA 2015 Cost Estimates and Sources of Financing Under "Best Practice" Policies (millions of 2000 US$) Cost item, period during Domestic Financing sources 2001-2015 & simulation resources Cost of EFA 2015 Domestic resources Gap for external financing scenario A + B + a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total C1 1,543 2,043 282 2,324 1,543 0 1,543 500 282 782 y C2 1,543 2,043 282 2,324 1,543 0 1,543 500 282 782 deliver Cumulative, 2001-2015 vice C3 1,658 2,043 282 2,324 1,658 0 1,658 384 282 666 ser C1 103 136 19 155 103 0 103 33 19 52 C2 103 136 19 155 103 0 103 33 19 52 Education Annual C3 111 136 19 155 111 0 111 26 19 44 C1 3 3 0 0 3 3 C2 3 3 0 0 3 3 costs Annual AIDS-related C3 3 3 0 0 3 3 C1 139 19 158 103 0 103 37 19 55 items C2 139 19 158 103 0 103 37 19 55 Annual Both C3 139 19 158 111 0 111 29 19 48 a. See previous table for specific policy measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education--under the "best practice" scenario. FINANCING EDUCATION FOR ALL BY 2015: SIMULATIONS FOR 33 AFRICAN COUNTRIES 71 SIERRA LEONE Table A5.28.1 EFA 2015 Financing Gap under Alternative Policy Measures A: Quality measures B: Efficiency measures C: Financing measures (%) Average teacher Government revenues Primary Nonteacher salaries Repeaters education % of pupils Annual Policy Pupil-teacher salary in per capita as % of As % of % for recurrent in private financing scenario Ratio spendinga GDP unitb enrollments GDP education spendingc schools gapd Status quo 40 33.1% 4.3 9.3% 11.4% 30.4% 51.3% 0.0% 13 A only 40 33.3% 4.3 13 A + B 40 33.3% 3.8 9.3% 11 C1 40 33.3% 3.8 9.3% 14.0% 26.0% 50.0% 10.0% 8 a + B practice": + C2 14.0% 20.0% 10 A "Best C3 14.0% 20.0% 10 Note: Shaded cells denote not applicable. For specific policies under "best practices," see previous table for measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. Refers to "best practice" scenarios (A + B + C1), (A + B + C2), or (A + B + C3). Except for Angola, Eritrea, Lesotho, and Nigeria, C2 and C3 are identical. b. As a multiple of the per capita GDP. c. As a percentage of total recurrent education spending. d. In millions of 2000 US$; refers to the difference between total cost of service delivery (ignoring AIDS-related costs) and the resources domestically. SIERRA LEONE Table A5.28.2 EFA 2015 Cost Estimates and Sources of Financing Under "Best Practice" Policies (millions of 2000 US$) Cost item, period during Domestic Financing sources 2001-2015 & simulation resources Cost of EFA 2015 Domestic resources Gap for external financing scenario A + B + a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total C1 261 288 88 376 261 0 261 27 88 115 y C2 227 288 88 376 227 0 227 62 88 150 deliver Cumulative, 2001-2015 vice C3 227 288 88 376 227 0 227 62 88 150 ser C1 17 19 6 25 17 0 17 2 6 8 C2 15 19 6 25 15 0 15 4 6 10 Education Annual C3 15 19 6 25 15 0 15 4 6 10 C1 4 4 0 0 4 4 C2 4 4 0 0 4 4 costs Annual AIDS-related C3 4 4 0 0 4 4 C1 23 6 29 17 0 17 5 6 11 items C2 23 6 29 15 0 15 8 6 14 Annual Both C3 23 6 29 15 0 15 8 6 14 a. See previous table for specific policy measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education--under the "best practice" scenario. 72 AFRICA REGION AGGREGATE EFA COSTING SIMULATION RESULTS AND DETAILED COUNTRY-SPECIFIC ESTIMATES SUDAN Table A5.29.1 EFA 2015 Financing Gap under Alternative Policy Measures A: Quality measures B: Efficiency measures C: Financing measures (%) Average teacher Government revenues Primary Nonteacher salaries Repeaters education % of pupils Annual Policy Pupil-teacher salary in per capita as % of As % of % for recurrent in private financing scenario Ratio spendinga GDP unitb enrollments GDP education spendingc schools gapd Status quo 28 22.5% 2.2 1.2% 11.1% 16.2% 50.5% 0.0% 131 A only 28 33.3% 3.5 301 A + B 40 33.3% 3.5 1.2% 180 C1 40 33.3% 3.5 1.2% 14.0% 20.0% 50.0% 10.0% 105 a + B practice": + C2 14.0% 20.0% 105 A "Best C3 14.0% 20.0% 105 Note: Shaded cells denote not applicable. For specific policies under "best practices," see previous table for measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. Refers to "best practice" scenarios (A + B + C1), (A + B + C2), or (A + B + C3). Except for Angola, Eritrea, Lesotho, and Nigeria, C2 and C3 are identical. b. As a multiple of the per capita GDP. c. As a percentage of total recurrent education spending. d. In millions of 2000 US$; refers to the difference between total cost of service delivery (ignoring AIDS-related costs) and the resources domestically. SUDAN Table A5.29.2 EFA 2015 Cost Estimates and Sources of Financing Under "Best Practice" Policies (millions of 2000 US$) Cost item, period during Domestic Financing sources 2001-2015 & simulation resources Cost of EFA 2015 Domestic resources Gap for external financing scenario A + B + a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total C1 3,078 4,349 297 4,646 3,078 0 3,078 1,271 297 1,568 y C2 3,078 4,349 297 4,646 3,078 0 3,078 1,271 297 1,568 deliver Cumulative, 2001-2015 vice C3 3,078 4,349 297 4,646 3,078 0 3,078 1,271 297 1,568 ser C1 205 290 20 310 205 0 205 85 20 105 C2 205 290 20 310 205 0 205 85 20 105 Education Annual C3 205 290 20 310 205 0 205 85 20 105 C1 3 3 0 0 3 3 C2 3 3 0 0 3 3 costs Annual AIDS-related C3 3 3 0 0 3 3 C1 292 20 312 205 0 205 87 20 107 items C2 292 20 312 205 0 205 87 20 107 Annual Both C3 292 20 312 205 0 205 87 20 107 a. See previous table for specific policy measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education--under the "best practice" scenario. FINANCING EDUCATION FOR ALL BY 2015: SIMULATIONS FOR 33 AFRICAN COUNTRIES 73 TANZANIA Table A5.30.1 EFA 2015 Financing Gap under Alternative Policy Measures A: Quality measures B: Efficiency measures C: Financing measures (%) Average teacher Government revenues Primary Nonteacher salaries Repeaters education % of pupils Annual Policy Pupil-teacher salary in per capita as % of As % of % for recurrent in private financing scenario Ratio spendinga GDP unitb enrollments GDP education spendingc schools gapd Status quo 40 11.2% 3.6 3.2% 10.9% 16.4% 63.0% 0.0% 87 A only 40 33.3% 3.6 128 A + B 40 33.3% 3.5 3.2% 125 C1 40 33.3% 3.5 3.2% 14.0% 20.0% 50.0% 10.0% 80 a + B practice": + C2 14.0% 20.0% 80 A "Best C3 14.0% 20.0% 80 Note: Shaded cells denote not applicable. For specific policies under "best practices," see previous table for measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. Refers to "best practice" scenarios (A + B + C1), (A + B + C2), or (A + B + C3). Except for Angola, Eritrea, Lesotho, and Nigeria, C2 and C3 are identical. b. As a multiple of the per capita GDP. c. As a percentage of total recurrent education spending. d. In millions of 2000 US$; refers to the difference between total cost of service delivery (ignoring AIDS-related costs) and the resources domestically. TANZANIA Table A5.30.2 EFA 2015 Cost Estimates and Sources of Financing Under "Best Practice" Policies (millions of 2000 US$) Cost item, period during Domestic Financing sources 2001-2015 & simulation resources Cost of EFA 2015 Domestic resources Gap for external financing scenario A + B + a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total C1 2,454 3,234 413 3,646 2,454 0 2,454 780 413 1,193 y C2 2,454 3,234 413 3,646 2,454 0 2,454 780 413 1,193 deliver Cumulative, 2001-2015 vice C3 2,454 3,234 413 3,646 2,454 0 2,454 780 413 1,193 ser C1 164 216 28 243 164 0 164 52 28 80 C2 164 216 28 243 164 0 164 52 28 80 Education Annual C3 164 216 28 243 164 0 164 52 28 80 C1 43 43 0 0 43 43 C2 43 43 0 0 43 43 costs Annual AIDS-related C3 43 43 0 0 43 43 C1 259 28 286 164 0 164 95 28 123 items C2 259 28 286 164 0 164 95 28 123 Annual Both C3 259 28 286 164 0 164 95 28 123 a. See previous table for specific policy measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education--under the "best practice" scenario. 74 AFRICA REGION AGGREGATE EFA COSTING SIMULATION RESULTS AND DETAILED COUNTRY-SPECIFIC ESTIMATES TOGO Table A5.31.1 EFA 2015 Financing Gap under Alternative Policy Measures A: Quality measures B: Efficiency measures C: Financing measures (%) Average teacher Government revenues Primary Nonteacher salaries Repeaters education % of pupils Annual Policy Pupil-teacher salary in per capita as % of As % of % for recurrent in private financing scenario Ratio spendinga GDP unitb enrollments GDP education spendingc schools gapd Status quo 45 25.2% 4.5 27.0% 14.9% 25.6% 48.3% 35.6% 3 A only 40 33.3% 4.5 10 A + B 40 33.3% 3.7 10.0% -2 C1 40 33.3% 3.6 10.0% 14.0% 25.6% 50.0% 10.0% 7 a + B practice": + C2 14.0% 20.0% 12 A "Best C3 14.9% 20.0% 11 Note: Shaded cells denote not applicable. For specific policies under "best practices," see previous table for measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. Refers to "best practice" scenarios (A + B + C1), (A + B + C2), or (A + B + C3). Except for Angola, Eritrea, Lesotho, and Nigeria, C2 and C3 are identical. b. As a multiple of the per capita GDP. c. As a percentage of total recurrent education spending. d. In millions of 2000 US$; refers to the difference between total cost of service delivery (ignoring AIDS-related costs) and the resources domestically. TOGO Table A5.31.2 EFA 2015 Cost Estimates and Sources of Financing Under "Best Practice" Policies (millions of 2000 US$) Cost item, period during Domestic Financing sources 2001-2015 & simulation resources Cost of EFA 2015 Domestic resources Gap for external financing scenario A + B + a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total C1 531 529 107 636 529 2 531 0 105 105 y C2 460 529 107 636 460 0 460 69 107 176 deliver Cumulative, 2001-2015 vice C3 478 529 107 636 478 0 478 51 107 158 ser C1 35 35 7 42 35 0 35 0 7 7 C2 31 35 7 42 31 0 31 5 7 12 Education Annual C3 32 35 7 42 32 0 32 3 7 11 C1 8 8 0 0 8 8 C2 8 8 0 0 8 8 costs Annual AIDS-related C3 8 8 0 0 8 8 C1 43 7 50 35 0 35 8 7 15 items C2 43 7 50 31 0 31 13 7 20 Annual Both C3 43 7 50 32 0 32 12 7 19 a. See previous table for specific policy measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education--under the "best practice" scenario. FINANCING EDUCATION FOR ALL BY 2015: SIMULATIONS FOR 33 AFRICAN COUNTRIES 75 UGANDA Table A5.32.1 EFA 2015 Financing Gap under Alternative Policy Measures A: Quality measures B: Efficiency measures C: Financing measures (%) Average teacher Government revenues Primary Nonteacher salaries Repeaters education % of pupils Annual Policy Pupil-teacher salary in per capita as % of As % of % for recurrent in private financing scenario Ratio spendinga GDP unitb enrollments GDP education spendingc schools gapd Status quo 41 26.2% 2.9 9.8% 10.8% 30.1% 53.2% 2.0% 14 A only 40 33.3% 3.5 63 A + B 40 33.3% 3.5 9.8% 63 C1 40 33.3% 3.5 9.8% 14.0% 26.0% 50.0% 10.0% 41 a + B practice": + C2 14.0% 20.0% 63 A "Best C3 14.0% 20.0% 63 Note: Shaded cells denote not applicable. For specific policies under "best practices," see previous table for measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. Refers to "best practice" scenarios (A + B + C1), (A + B + C2), or (A + B + C3). Except for Angola, Eritrea, Lesotho, and Nigeria, C2 and C3 are identical. b. As a multiple of the per capita GDP. c. As a percentage of total recurrent education spending. d. In millions of 2000 US$; refers to the difference between total cost of service delivery (ignoring AIDS-related costs) and the resources domestically. UGANDA Table A5.32.2 EFA 2015 Cost Estimates and Sources of Financing Under "Best Practice" Policies (millions of 2000 US$) Cost item, period during Domestic Financing sources 2001-2015 & simulation resources Cost of EFA 2015 Domestic resources Gap for external financing scenario A + B + a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total C1 2,523 2,930 210 3,140 2,523 0 2,523 407 210 617 y C2 2,189 2,930 210 3,140 2,189 0 2,189 741 210 951 deliver Cumulative, 2001-2015 vice C3 2,189 2,930 210 3,140 2,189 0 2,189 741 210 951 ser C1 168 195 14 209 168 0 168 27 14 41 C2 146 195 14 209 146 0 146 49 14 63 Education Annual C3 146 195 14 209 146 0 146 49 14 63 C1 47 47 0 0 47 47 C2 47 47 0 0 47 47 costs Annual AIDS-related C3 47 47 0 0 47 47 C1 242 14 256 168 0 168 74 14 88 items C2 242 14 256 146 0 146 96 14 110 Annual Both C3 242 14 256 146 0 146 96 14 110 a. See previous table for specific policy measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education--under the "best practice" scenario. 76 AFRICA REGION AGGREGATE EFA COSTING SIMULATION RESULTS AND DETAILED COUNTRY-SPECIFIC ESTIMATES ZAMBIA Table A5.33.1 EFA 2015 Financing Gap under Alternative Policy Measures A: Quality measures B: Efficiency measures C: Financing measures (%) Average teacher Government revenues Primary Nonteacher salaries Repeaters education % of pupils Annual Policy Pupil-teacher salary in per capita as % of As % of % for recurrent in private financing scenario Ratio spendinga GDP unitb enrollments GDP education spendingc schools gapd Status quo 50 21.7% 2.7 6.2% 18.8% 12.3% 43.2% 1.6% 10 A only 40 33.3% 3.5 40 A + B 40 33.3% 3.5 6.2% 40 C1 40 33.3% 3.5 6.2% 14.0% 20.0% 50.0% 10.0% 25 a + B practice": + C2 14.0% 20.0% 25 A "Best C3 18.8% 20.0% 17 Note: Shaded cells denote not applicable. For specific policies under "best practices," see previous table for measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education. Scenarios C1, C2, and C3 correspond to those indicated in the bottom panel of that table. a. Refers to "best practice" scenarios (A + B + C1), (A + B + C2), or (A + B + C3). Except for Angola, Eritrea, Lesotho, and Nigeria, C2 and C3 are identical. b. As a multiple of the per capita GDP. c. As a percentage of total recurrent education spending. d. In millions of 2000 US$; refers to the difference between total cost of service delivery (ignoring AIDS-related costs) and the resources domestically. ZAMBIA Table A5.33.2 EFA 2015 Cost Estimates and Sources of Financing Under "Best Practice" Policies (millions of 2000 US$) Cost item, period during Domestic Financing sources 2001-2015 & simulation resources Cost of EFA 2015 Domestic resources Gap for external financing scenario A + B + a mobilized Recurrent Capital Total Recurrent Capital Total Recurrent Capital Total C1 619 866 131 996 619 0 619 247 131 378 y C2 619 866 131 996 619 0 619 247 131 378 deliver Cumulative, 2001-2015 vice C3 747 866 131 996 747 0 747 119 131 249 ser C1 41 58 9 66 41 0 41 16 9 25 C2 41 58 9 66 41 0 41 16 9 25 Education Annual C3 50 58 9 66 50 0 50 8 9 17 C1 29 29 0 0 29 29 C2 29 29 0 0 29 29 costs Annual AIDS-related C3 29 29 0 0 29 29 C1 86 9 95 41 0 41 45 9 54 items C2 86 9 95 41 0 41 45 9 54 Annual Both C3 86 9 95 50 0 50 37 9 45 a. See previous table for specific policy measures in the three policy domains--quality, efficiency, and domestic fiscal effort for primary education--under the "best practice" scenario.