72364 Working Paper 5 for Afghanistan Public Expenditure Review 2010 Second Generation of Public Expenditure Reforms Education Sector Co-produced with the UK Department for International Development 1 Education Sector1 Key Findings  The education sector has expanded significantly since 2001. A six-fold increase in school enrollment has, however, put significant pressure on government finances and engendered a high degree of donor dependence. It is essential that the government make the best use of available resources by identifying and targeting its highest priorities with the greatest efficiency.  The Ministry of Education (MoE) has made significant progress since the 2005 Public Financial Management (PFM) Review, formulating the National Education Strategic Plan (NESP) and enhancing the roles of Provincial Education Departments (PEDs). By contrast, the Ministry of Higher Education (MoHE) has made limited progress, having only just begun to develop a national strategic framework.  Despite progress to date, many challenges remain including a shortage of competent teachers; a lack of school buildings and the poor quality of existing infrastructure; the low capacity of administrators, and insufficient financial resources. Donor funding to higher education has been considerably less than to basic education, and the security situation negatively affects the sector’s development.  Sustaining progress to date will require strong leadership, avoiding an over-reliance on technical assistance and narrowing the financing gap. Poor data quantity/quality is also an obstacle for monitoring, evaluation and analysis.  PFM in MoE/MoHE is key to improving the education sector. MoE has implemented PFM reform, reorganized its structure and established an internal audit unit. Pilot program budgeting in MoE is perceived as one of the best cases, though its benefits have not been well observed. The teacher salary payment system is being improved. Teacher registration has been completed and some are already receiving salaries through their bank accounts. However, the procurement system is not yet conducive to smooth execution of the core development budget and the capacity of PEDs remains underdeveloped. Auditing and reporting are still rudimentary, which will make results-based budget formulation difficult.  Expenditures on the education sector have constantly increased in absolute values, but are gradually decreasing as a share of GDP as the value of external budget expenditures has remained constant. About 60 percent of all education expenditures are external budget expenditures, and more than 80 percent of the core budget is allocated to operating expenditures, specifically teacher salaries. Regarding regional equity, the allocation of operating expenditures is generally in line with student-to-teacher ratios that do not vary greatly between provinces. The core development budget however, does not have a significant relationship to the number of students, although the external budget has some offsetting impacts. Gender inequalities are observed across provinces. School buildings constructed through the external budget were on average 50 percent more expensive compared with those funded through the core budget. However, the results should be treated carefully. A financial sustainability analysis shows that expenditures will need to increase 180 percent in constant values if full participation in primary school is to be achieved (meeting the Millennium Development Goals (MDGs)) by 2020. 1 This working paper was produced by Mary Venner (DFID Consultant) for Afghanistan Public Expenditure Review based on information available as of March 31, 2010. 2  This Public Expenditure Review (PER) on the whole supports MoE’s current reform direction. Policy recommendations are basically in line with the ongoing reform process. The key expenditure framework recommendations are: - Effective implementation of the Budget Planning and Expenditure Tracking Database (BPET). - Better planning and execution of operating expenditures including the timeliness and value of quarterly allotments to PEDs. - No further elaboration of MoE program budgeting and simplified program budgeting for MoHE. The key expenditure and financing recommendations are: - Agreements are needed on multi-year operating budget funding strategies between MoF and MoE/MoHE, including an adequate non-salary budget for the Operation and Maintenance (O&M) of school buildings. - Progressive moves towards the allocation of funding based on the number of students to address regional inequalities, and direct grants to schools for non-salary costs. - Positive preference to women in teacher training, investment in girls’ schools and the construction of suitable boundary walls around the schools to address gender inequalities.  Regarding the effectiveness of aid, the NESP has functioned as a platform for coordination between MoE and donors. In order to further improve coordination and aid effectiveness, a step-by-step approach based on the existing reform process should be taken to encourage future Sector-Wide Approaches (SWAps). 3 A. Introduction and Background Objective 1. Strong demand for education results in financing pressures. There was limited investment in education in Afghanistan over the past few decades. By 2001, slightly more than one million students were enrolled and the adult literacy rate was 28 percent. However, the low participation in education was not due to a lack of interest. Since 2001, the number of children enrolled in primary and secondary schools increased rapidly, reaching six million in 2008. Moreover, the people of Afghanistan have expressed their wish that the government acknowledge their constitutional right to a free education to the tertiary level.2 This places significant pressure on government finances and engenders high donor dependence. It is essential the government make best use of its resources by directing funding to the highest priorities and maximizing the efficiency of education expenditures. 2. Objectives. The government aims to have the best education system possible and has set ambitious objectives. For example, MoE is aiming for developed country-standard teacher qualifications and classroom sizes. However there are major challenges to reaching these goals in the short term. This Annex assesses the extent to which the government spends public resources on education in an appropriate and effective manner. More specifically, this Annex: - Outlines the institutional framework and current status of the education system in Afghanistan. - Reviews the progress of the 2005 PFM Review policy recommendations. - Reviews government strategies and policies. - Assesses the current expenditure framework. - Analyses the fiscal situation and assesses future financing requirements. - Recommends policies to improve the use of public resources. Education System and Outcomes 3. Responsibilities are divided between several ministries (Figure 1). MoE operates primary and secondary schools and government-funded Islamic schools. MoE also provides teacher training, community literacy programs and operates a number of vocational education institutions. MoHE is responsible for tertiary education through universities including teacher’s colleges. The Ministry of Labor, Social Affairs and Martyrs and Disabled (MLSAMD) is responsible for vocational skills training through the National Skills Development Program (NSDP). About 200 kindergartens are also operated by MLSAMD, mainly in Kabul. In addition, the National Solidarity Program (NSP) has provided funds directly to communities for school construction. NGOs and donors also play a role through community-based schools and development projects. The private education sector is also developing. MoE has so far licensed about 160 private schools in Kabul and the provinces, and there are an increasing number of private vocational colleges and universities. 2 Article 43 of the Constitution 4 Figure 1 Education System in Afghanistan Age cohort 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 Number of Students Ministry in Grade 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 (2007) Charge Community Based Education (CBE) 158,482 MoE Primary (compulsory) 4,669,110 MoE Lower Secondary (compulsory) 735,697 MoE Higher Secondary 271,144 MoE Islamic Education 91,362 MoE Teacher Training (primary) 14,294 MoE Technical and Vocational Institutions 10,366 MoE Tertiary, including teacher training 52,000 MoHE Source: Education Management Information System (EMIS) and MoHE 4. Enrollment has significantly improved. The number of students in primary and secondary schools increased from 1.1 million in 2001 to 6.1 million in 2008. Concurrently, the net primary education enrollment rates reached 61 percent in 2007.3 The enrollment of girls has increased from almost zero in 2001 to one-third of all enrollments. There are currently 20 tertiary institutions including teacher education colleges where the number of students has increased from 22,500 in 2002 to 52,000 in 2008. In contrast to quantity indicators, limited information is available on quality indicators. According to examination results in 2007, failure rates vary from grade to grade4, but about 21 percent of primary and secondary students did not qualify to sit exams due to lack of attendance. Development since the 2005 PFM Review 5. Progress against 2005 PFM Review recommendations (Table 1). Substantial progress has been made in the education sector. The first NESP for the period 2006/07 through 2010/11 was formulated in 2007, and then further updated for the period between 2010/11 through 2014/15 in January 2010. The role of PEDs in administration and budget management has been significantly enhanced. Also, efforts have been made to improve the quality of education through better teacher qualifications and training, for example. In terms of public financing, MoE’s improved donor coordination system has to some extent contributed to aligning the external budget with the NESP. 6. Progress has also been made in areas not covered by PFM Review recommendations. Restructuring of the MoE has strengthened its planning and budgeting functions and integrated the preparation of the development and operating budgets that in the past were prepared by different units. MoE was one of the first ministries to pilot program budgeting and has now successfully developed three annual budgets (2007/08 through 2009/10) on a program basis. There has been significant expansion in tertiary education funding and student numbers since 2005. Donors have been working with tertiary institutions and faculty to improve the quality of education, but limited progress has been made reforming higher education policy and institutional arrangements, or modernizing financial planning and management. 3 Seventy-four percent for boys and 46 percent for girls. The net enrollment measures the number of students in the relevant age group relative to the total number of children in that age group in the population. 4 On average, failure rates were 5.3 percent between grade one and grade 12. However, the rates reached 11 percent at grade four. 5 Table 1 Summary of Progress against 2005 PFM Review Recommendations Recommendations Current Status Gender disparity NESP identifies gender disparity as important. Some Address supply side constraints (e.g. more schools in initiatives are being implemented but specific strategies geographical proximity for girls) need to be clarified. Regional disparity NESP provides limited information on this issue. Develop more targeted financial programs and a simple Formula-based funding has not been considered. formula-based funding mechanism Quality of education Efforts to improve quality have focused on improving Define quality standard, ascertain baseline and monitor teacher qualifications and training. The non-salary progress; increase non-salary recurrent expenditures operating budget is still below requirements. Secondary education A new curriculum finalized and textbooks being Curriculum and textbook reform; teacher training; prepared. Considerable investment in teacher training. school construction; strategies and policies on private The number of school buildings has increased. Private secondary education school policy established. Higher education Donors are supporting partnerships with foreign Overhaul academic programs and course structures. universities and developing faculty skills. Although Increase overall level and efficiency of financing. funding levels have increased efficiency is still an issue. Develop quality assurance and accreditation mechanisms. Highly centralized education administration The role of PEDs has been significantly enhanced. De-concentrate administrative functions to PEDs and/or Administration in Higher Education remains heavily schools centralized. Limited/unpredictable public financing Financing for core budget still unpredictable. Donor Promote SWAps, the harmonization of donor funding, coordination in MoE is improving. SWAps are under and improvements in financial management and consideration. Legislation for private education is in procurement. Develop regulatory framework for place. User fees for Higher Education are not politically private financing. Charge user fees for services such as possible dormitories. Key Obstacles 7. Quantity and quality of education data. Analysis of the education sector is hampered by the lack of reliable data including population statistics, limiting for example, the calculation of net and gross enrollment rates. Key measures of equity and efficiency, and the information needed to project future budget requirements are not readily available. Significant improvements in information regarding students, schools and teachers have been made but the data remains flawed. Budget reporting has also improved but the data needs to be treated with caution. Appendix 1 discusses problems with the available education data in greater detail. 8. Security. Anti-government forces have targeted schools, teachers and students, girls’ schools in particular. In 2008, 650 schools were closed because of concerns about security and 140 teachers were killed or wounded.5 It is difficult to recruit staff to work in these areas and existing staff are understandably reluctant to be transferred there. This has a negative effect on the delivery of education services and the ability of children to participate, and also compounds gender and regional disparities. The security situation also poses problems for education planning and management. The 2007 Education Management Information System (EMIS) survey was unable to collect data from roughly 200 schools and was not able to verify data on a further 400. In addition, it has not been possible to register all teachers or audit payrolls in insecure areas. 5 (www.quqnoos.com, 1 February 2009) 6 B. Education Sector Strategies 9. Outcome indicators need to be developed. The government has prepared a number of strategic documents including the MDGs, the 2006 Afghanistan Compact, the 2008 Afghanistan National Development Strategy (ANDS) (Table 2) and the 2007 NESP. MoHE completed the National Higher Education Strategic Plan for 2010-2014 with support from United Nations Educational, Scientific and Cultural Organization (UNESCO) in late 2009. These strategic documents tend to focus on inputs such as enrollments, teachers and curriculum, rather than outcomes such as literacy, academic performance and skill attainment. Despite the data limitations it is reasonable to assume for the time being that increased and improved inputs will produce the desired outcomes. A stronger focus on measuring education outcomes will need to be developed progressively as the capacity of the system to collect relevant data improves. Table 2 ANDS Benchmarks in Education and Current Status Benchmarks Current Status  The net enrollment rate for boys and girls in primary grades  In 2007, total net enrollment was Primary and will be at least 75 percent and 60 percent respectively. estimated at 74 percent for boys Secondary Education  A new curriculum will be operational in all secondary schools. and 46 percent for girls.  The number of female teachers will increase by 50 percent.  In 2007 there were just over  Seventy percent of teachers will have passed a competency 40,000 female teachers and test. 109,000 male teachers.  University enrollment will reach 100,000, with at least 35  There are currently 52,000 Education percent female student participation. students at tertiary institutions. Higher  The curriculum in Afghanistan’s public universities will be Eighteen percent are female. revised to meet the development needs of the country and private sector growth.  150,000 men and women will be trained in marketable skills development through public and private means by the end of 2010  Non-formal education, apprenticeship programs and vocational Skills -- education and training will provide unemployed older youths and adult workers with skills relevant to the evolving needs of employers and the economy 10. Shifting focus to longer-term education development needed. There are differences of opinion on whether the education system should be regarded as still in an emergency phase or whether the focus should shift to longer-term development. High demand, lack of school buildings, large numbers of unqualified teachers and serious security challenges suggest elements of the emergency situation remain. However, more than seven years after the end of the Taliban regime, the government is shifting its focus away from the numbers of students and classrooms, to what happens in those classrooms and whether children benefit from school attendance. Strategic Planning in MoE 11. The 2007 NESP is a key strategic document focusing on the activities of both MoE and donors. Annual operational plans for implementing the NESP are developed for each program, starting 12 months before the fiscal year. These form the basis for estimating annual budget requirements and targets. The operational plans are revised to account for actual resources and available staff following approval of the annual budget. The donors have been very supportive of the NESP initiative in part because it provides an opportunity to align their activities with the government’s plans. 12. Lack of Prioritization. The NESP’s financial estimates relate mainly to development investments. The significant effort and resources required simply to operate existing schools 7 and manage teachers are not included. A particular problem is the lack of prioritization between different activities and targets. In fact, funding has not been forthcoming in either the core or external development budget for a large number of NESP activities. 13. Monitoring and Evaluation. The planning department is responsible for monitoring the strategic plan through monthly reports from program managers. Quarterly progress reporting from the provinces was initiated in 2008 but not all PEDs have complied. MoE is also working with the Ministry of Economy (MoEc) to monitor the ANDS targets. The planning department also intends to implement performance evaluations of development activities, including external budget projects, in the future. 14. The NESP was revised in early 2010. Led by the planning and evaluation department, MoE finalized the revised NESP covering 2010/11-2014/15 in January 2010. The new NESP was formulated through participation process by stakeholders and by the strong ownership of MoE. More attention is paid for monitoring and evaluation, and medium-term costing was updated. Strategic Planning of MoHE 15. MoHE strategic plan for 2010-14 was finalized in late 2009 with the support from the World Bank, USAID and UNESCO. The strategic plan set out a vision and goals for higher education. The strategic plan consists of two main programs such as: Program 1: To educate and train skilled graduates to meet the socio-economic development needs of Afghanistan, enhance teaching, research and learning and encourage service to the community and the nation. Program 2: To lead and manage a coordinated system of higher education comprising universities, institutions, and community college dedicated to providing high quality tertiary education. C. Expenditure Framework 16. PFM in MoE/MoHE is a key aspect of effective management and the improvement of education delivery. MoE has made significant improvements in PFM, integrating the development and operating budgets, and decentralizing budget and administration authority to PEDs. MoHE, on the other hand, continues to operate a highly centralized budget management system. Individual tertiary institutions have little input into budget preparation and limited discretion over the use of resources. Development projects are managed separately from the operating budget and are largely driven by donors. 17. MoE has been restructured in recent years to place financial responsibility for Figure 2 Organization of planning, budget, both the operating and development finance and audit functions in MoE budgets in a single budget department Minister of Education (under Finance and Accounting) and to Internal Audit strengthen the strategic focus of the planning department (Figure 2). Deputy Minister Planning and Administration and Finance Evaluation Directorate Procurement for both development and Human Resource operating budget expenditures has also Management Strategic Planning been brought together in a reorganized Finance and Accounting procurement department. However, (including Budget and Payroll) Monitoring while the development budget is Procurement EMIS implemented mainly by international Source: MoE and national Technical Assistance (TA), 8 the operating budget is implemented by civil servants using manual procedures. This is largely due to the different and more complex allotment processes and reporting requirements of the development budget. MoE has extended the scope of participation of provincial directors in the budget process. However, the capacity for budget preparation is still developing and there is limited devolution of responsibility for its implementation. Budget Formulation 18. The budget formulation process has been significantly improved. MoE budgets were prepared on a program and provincial basis between 2007/08 and 2009/10. This is a significant change from the previous practice where the budget department prepared an operating budget and the planning department separately prepared a development budget. The initial estimates for the 2009/10 operating and development budgets were prepared by each program director with assistance from the Program Budgeting Unit that coordinated the process. MoE faced tight deadlines and a lack of adequate advance notice from MoF, including revisions to the budget ceiling that resulted in the proposed budget being lowered. This precluded discussion about budget adjustments with either PEDs or program managers. The NESP has provided MoE with the basis for multiyear expenditure planning, but further work is needed including more realistic scheduling of development projects and better linkages with the MTFF. 19. Operating budget formulation. Wage and salary estimates are prepared from the bottom up on the basis of individual salary entitlements and allowances. By contrast, non-salary budget requirements are based on previous years’ expenditures, MoF ‘norms’, and requests from provinces. The wages and salaries budget is not based on detailed projections for deployment of new staff. MoF significantly reduced MoE’s initial non-salary budget needs estimates during budget negotiations, and MoE considers the 2009/10 budget insufficient to cover all staff salary entitlements. There are no reliable forward estimates of the future costs of existing activities (e.g. school construction and maintenance, P&G reforms). 20. Core development budget formulation. The core development budget is almost entirely financed by donors underlining the importance of effective coordination. MoE estimates its core development budget requirements on the basis of activities and cost estimates in the NESP, however the actual approved budget has been well below these estimates. For example, the actual core development budget in 2009/10 (US$106 million) is less than one-quarter of the MoE estimate (US$385 million). However, MoE has not been able to expend the allocated core development expenditures in previous years: the execution rate in 2008/09 was just 38 percent. 21. Program budgeting. Program budgeting in MoE is seen as one of the best cases. Factors contributing to the success are: (i) MoE had just completed the NESP and identified and costed its programs; (ii) program budgeting was incorporated in the significant reform process that was already underway within MoE, and (iii) the use of international and local consultants with relevant experience. However, inadequate performance monitoring data and the lack of financial resources for achieving ambitious program objectives limits the ability of program budgeting to improve service delivery.  Benefits. In some respects program budgeting has produced benefits, although many of these derive from the strategic planning process rather than from program budgeting itself. There are clearly advantages in having an agreed set of policy objectives, strategies, defined activities and progress targets instead of the previous undefined and uncoordinated activities. The program structure provides a consistent and transparent basis for both financial and non-financial reporting, including EMIS statistics. Extending 9 program reporting to the provincial level has enabled more detailed information, although there are questions about the accuracy of the program classification of expenditures.  Difficulties and challenges. Both MoE and MoHE are critical of pilot program budgeting, arguing that it is too complex, has been implemented too quickly with inadequate consideration of capacity or preparation. Pilot program budgeting has generated a considerable amount of additional work and created problems in budget execution. Submission documents are complex and generally can only be prepared by international advisers and national consultants. The program budget submission for the 2008/09 budget for example, comprised more than 1,000 pages of detailed information about activities and sub-activities. Although requirements were simplified significantly, a submission of more than 90 pages was still required for the 2009/10 program budget.  Impacts on PEDs. Operating budget allotments by program at the provincial level started in 2007/08. This increased the amount of work required in managing the budget in PEDs and Mustofiats (Provincial office of MoF). Neither MoE nor MoF were adequately prepared for this change. Most of the provincial budget execution work is done manually and there were consequentially delays in budget expenditure6. Program allotments also reduced the flexibility previously available to PEDs in the use of budget allocations. There were cases of overspending on some allotments and of funds not being available to cover salaries. It also generated a need for a larger number of amendments to allotments and increased the number of payment transactions to be processed. For example, instead of making one payment transaction for salaries each month it is now necessary to process separate payments for each program. 22. Provincial budgeting. Seventy-eight percent of the operating budget is allocated to the provincial level7. There is therefore a strong case for PEDs to be closely involved in budget formulation and execution. In 2007/08, MoE prepared a manual on the budget and finance responsibilities of Provincial Directors and Finance Directors and provided training based on the manual (Box 1). In the past, PEDs had little idea of the amounts they could expect to spend. The new provincial budgeting process in MOE provides PEDs with an estimate of their total annual allotment amount and requires them to develop a plan for spending it, including the timing and methods of procurement. Box 1 Financial Responsibilities of Provincial Education Directors  Expenditure and procurement planning.  Recording and monitoring allotments.  Timely execution of the budget.  Effective use of the budget.  Recording expenditure commitments and ensuring they do not exceed the allotment.  Compliance with the procurement law.  Correct processing of payments.  Timely acquittal of advances.  Safe custody of public property and assets.  Monthly reconciliation of accounts with MoF and reporting on expenditures to the center.  Coordination and monitoring of development activities with donors. Source: MoE 6 The number of forms that needed to be prepared by the MoE Budget Department to issue allotments to provinces, for example, increased from 185 to around 1,480. 7 Moreover, the actual benefit of much of the expenditure implemented by the centre may go to the provinces. Annex 2 discusses this issue. 10 23. External Budget. Unlike the core budget there is no requirement for the external budget to be approved by parliament. However, to ensure smooth implementation of external budget projects, MoE expects donors to obtain prior approvals from the minister or relevant deputy minister. MoE uses the NESP to align the external and core budgets. The Education Partnership Working Group in 2008 conducted a mapping of donor activities against the NESP programs for donor coordination. MoE maintains various records of external donor activities in the planning, budget and construction departments that do not necessarily align with MoF’s donor assistance database. Each dataset has its own purpose, but there would be benefits from improved information sharing between MoE and MoF. 24. The Budget Planning and Expenditure Tracking Database (BPET) is expected to be ready in 2009. BPET will assist MoE with the internal management of financial information including: (i) program and provincial budget planning and allotment management, (ii) consolidated program finance and performance reports, (iii) centralized data on procurement contracts and donor grants, and (iv) reporting on the status of advances. BPET will facilitate information sharing within MoE departments and potentially replace the large number of separate spreadsheets used to manage the development budget. In time, it will also allow the paper- based procedures for implementing the operating budget to be rationalized. However, there are several challenges. For example, information entered into the system will need robust quality control procedures and all allotment and expenditure data needs to be reconciled with the AFMIS. Budget Execution 25. MoE continues to face challenges in effective budget execution despite the improvements of budget formulation. Operating budget execution is slow and erratic and the execution rates of core development budgets are very low. Program/provincial budgeting is of little benefit if the budget cannot be effectively executed. 26. Operating budget. There are concerns about the timeliness of the distribution of allotments and their execution especially in provinces. MoF does not start the allotment process until the budget is approved (usually just before the start of the fiscal year) creating a delay of several weeks or more before PEDs and Mustofiats are authorized to make commitments and expenditures. This means that teachers and other staff do not receive salaries in the first month and sometimes not until three months into the year. Regarding goods and services, PEDs are unable to initiate procurement procedures before the quarterly allotments are issued. Budget Execution Guidance from MoF requires that all procurements be completed within a fiscal year creating a rush to spend the budget in the final two months, which is highly likely to produce inefficient and inappropriate expenditures. 27. Teacher salaries. Wages and salaries (including those of non-teachers) account for more than 90 percent of MoE operating expenditures. Monthly payrolls are compiled based on attendance records submitted by each school. PEDs compile the records and submit a payment request to Mustofiats. Salaries are paid in cash, in front of witnesses, and the employee signs to confirm receipt. However, staff in remote locations may delegate one person to receive salaries for all employees, or may only come to collect their pay every few months. There is scope for this system to be abused in various ways, such as adding ‘ghost’ staff on the payroll and skimming ‘commissions’ from cash payments8. MoE has taken several steps to tighten administration including registering and confirming the identity of all teaching staff and conducting an internal audit of payroll processes. In 2009, MoE plans to improve on- 8 MoE identified about 5,000 possible ghost teachers during its teacher registration process in 2008/09 (out of almost 200,000 staff), and removed them from the payrolls. 11 time salary payments and reduce the scope for corruption by paying salaries through banks9, thus eliminating unofficial “deductions”. 28. Development budget. MoE’s core development budget execution rates are lower than the overall disbursement rates10. This reflects the slow procurement process, delays in payments and over-optimistic budget estimates by both donors and the government. Although there are some positive developments in the procurement department, it remains a major cause of the delays in core budget execution (Box 2). 29. Predictability of funding in provinces. In 2008/09, there were many adjustments to the MoE budget that were passed on as increases, decreases or transfers between codes in provincial and program allotments at the secondary level. This means that PEDs have little certainty about the availability of resources for the quarter, resulting in delays in procurement and expenditures. The quarterly allotment system authorizes provinces to enter commitments up to the level of the allotment, but this may not be sufficient to cover large purchases, nor do allotments take seasonality into account. 30. Mustofiats. MoE is concerned about the efficiency and speed of expenditure processing by Mustofiats. It is suggested that some Mustofiats limit the number of transactions processed at a time, work short daily hours and are inoperative for a large part of the month. Sixteen of the 34 Mustofiats are connected to the AFMIS system and are able to provide automated, real time transaction processing and reporting11. The remainder relies on manual ledgers and provides ex-post reports to the centre that run between one and two months behind. Extension of the AFMIS to all Mustofiats and in due course to PEDs would improve the timeliness and accuracy of budget execution and enable MoE to monitor provincial budget execution more closely. Box 2 Procurement of construction projects The reorganisation of the procurement department and procedural changes have significantly improved the procurement of construction services. The number of procurement contracts signed in 2008/09 exceeded the total number processed in the preceding three years. Nevertheless, the department has identified the following causes for delays in construction projects. Construction season. The best time to start construction is March, which is also the beginning of the fiscal year. The minimum procurement timeframe for construction projects is between two months and three months so planning for procurement of building work should commence before the start of the fiscal year, to ensure that the project can be completed before winter. However, procurements from discretionary funds cannot be initiated until the budget is approved. Inadequate understanding of procurement process. The lack of knowledge about procurement processes and rules remains a concern and some program managers do not appreciate the time required for procurement. This leads to rushed procurements and a greater risk of mistakes. Use of donor procurement systems adds additional burdens to the existing procedure. For example, World Bank-funded projects are required to use World Bank procurement regulations rather than the national law and there are multiple points at which the Bank’s approval is required. In 2008/09, 96 projects required prior approval of the procurement decision by the World Bank. Special Procurement Committee (SPC). Projects above a certain threshold value must be referred to SPC for approval. This adds considerable time to the finalization of some contracts. Recent amendments to the Procurement Law have increased the threshold at which a contract needs to be reviewed by SPC and set deadlines for the Committee to respond. This is expected to accelerate the procurement process. 9 As of April 2009, 122,917 out of 160,918 teachers have bank accounts. 10 In 2008/09, the disbursement rates of MoE were 37 percent, lower than the overall rates of 43 percent. Please see Annex 2 on the explanation of core development budget. 11 As of June 2009. 12 Audit and Reporting 31. Internal Audit. The new Internal Audit unit, comprising an auditor in each province and nine in the central office, was established in 2008. The unit, which reports directly to the Minister, has the authority to conduct audits of financial management, procurement and administration of both operating and development budgets, and in due course plans to undertake program policy, management and performance audits. However the capacity for the latter will take time to develop. Thus far, the unit has carried out financial compliance audits and is starting procurement audits. The first priority has been an ongoing audit of payroll processing. The audits, together with the teacher registration process (which was also audited by the unit) appear to have considerably improved compliance and reduced the number of payroll ‘mistakes’. The auditors encountered resistance and significant difficulties in carrying out their work in some provinces and have been unable to conduct audits in insecure areas. 32. Reporting. There are significant delays in reporting on operating budget execution in the provinces, through both MoF and MoE’s own budget monitoring processes, due to the system of monthly manual reporting where the AFMIS is not in use. As the execution of the development budget is more centralized, within-year monitoring reports are more up to date and there is less need to collect data from the provincial offices. Provincial development expenditures are made through a system of cash advances that are recorded in the AFMIS as disbursements before the funds have actually been spent. There is as yet no system for reporting on the use of budget resources by individual schools or institutions. Ministry of Higher Education 33. Limited reforms in budget formulation and execution. Responsibility for operating and development budget formulation and execution remains in separate departments. Administration of the tertiary education sector is highly centralized with individual institutions having limited autonomy and no independent budget authority. Each university is treated as a branch of MoHE and receives quarterly budget allotments determined by the central budget office. There is limited consultation with institutions on the budget and all procurement is carried out in the centre. 34. Core development budget formulation. The core development budget consists primarily of construction projects, but there is no active consultation process with universities. Rather the central planning department waits for universities or donors to take the initiative to identify projects. In the absence of a strategic plan, MoHE generally accepts any project that is proposed by donors and conducts limited monitoring of external budget activities. Although there is some liaison between the government and the major donors, this is mainly initiated by the donors themselves. 35. Program budgeting. MoHE has not yet completed pilot program budgeting for 2008/09. MoHE rejected the program structure developed by international consultants (based on universities) and there was insufficient time before the finalization of the 2009/10 budget to develop a new structure. MoHE is still developing a program budget but considers the MoF’s current process too bureaucratic, centrally imposed and unsustainable. 36. Audits. MoHE has an internal audit unit and is subject to external audit visits from the Control Audit Office (COA) every six months. The quality of these audit exercises is not known. As a relatively small ministry, with around 6,000 staff in 20 institutions, MoHE avoids some of the administrative challenges of larger ministries in monitoring staff numbers and budget expenditures. It does not believe there is a significant problem with unauthorized employment or fraudulent salary payments. 13 D. Level and Distribution of Expenditures and Financing Overall Education Sector Financing 37. International comparison. Afghanistan spends more on education as a proportion of GDP than South Asian countries like India, Pakistan and Bangladesh (Table 3). However, the country’s higher education expenditures as a share of GDP should be examined with caution. Afghanistan’s GDP is low, a large amount of donor funding is included in total education sector spending, and there is a backlog of infrastructure projects currently being addressed in the development budget component12. On the other hand, the ratio of education expenditures to total government expenditures is lower than the South Asia average. Table 3 Education Sector- International Comparison Expenditure as % of Total Government Country Expenditure as % of GDP Expenditure (WDI) Afghanistan (2007/08) 4.6 12.7 Iran 5.5 18.6 India 3.2 10.7 Pakistan 2.9 12.2 Bangladesh 2.5 14.2 Source: UNESCO UIS and WDI Most recently available data. 38. The external budget accounts for half of Figure 3 Composition of Education Sector total expenditures. Between 2003/04 Funding (2003/04-2008/09) and 2008/09, total education expenditures 13 were US$2.1 billion, or External Operating Expenditures eight percent of total national Budget 41% 49% expenditures. During this period the external budget accounted for 41 percent. Core operating expenditure was 49 percent of the total, mainly due to MoE operating expenditures associated with Core Development teacher salaries (Figure 3). The figures do Expenditures not take into account formal or informal 10% out-of-pocket expenditures on public Source: MoF education. There is no data on these costs but out-of-pocket expenses include the cost of school clothing and stationery, informal payments to teachers and possibly, contributions to their non-salary expenditures like transportation, rent and food. 12 Please see Annex 2 on this discussion on the overall expenditures. 13 Including MoE, MoHE, Ministry of Information and Culture, Science Academy and National Olympic Committee However, the figure does not include amounts that the Ministry of Labor has spent on vocational training or spending by donors and NGOs on vocational training activities, which are classified to a different ANDS sector. Spending on school construction under NSP is also not included. 14 Figure 4 Financing Sources of Development Expenditures in the 2009/10 budget A. Core Budget B. External Budget Denmark Other, 4% 11% France, 4% Norway, 4% Germany, 6% Afghanistan 16% Canada, 7% USAID, 45% ARTF 55% Sweden, 11% World Bank 18% India, 19% Source: MoF 39. Sources of core and external development financing. In the 2009/10 budget, total education development expenditures are US$250 million14. The core development budget is US$137 million and the external budget is US$116 million. More than 90 percent of the total development budget (including both core and external) is externally financed by donors, while the government provides the remaining 10 percent. The ARTF is expected to finance more than half the core development budget, followed by the World Bank (Figure 4A). USAID accounts for almost half of the external development budget, followed by India and Sweden (Figure 4B). Expenditures 40. Education expenditures have Figure 5 Education Sector Expenditures continuously increased in absolute terms, (US$ million, share in GDP) although the share of GDP has been US$million %GDP 400 6.0% gradually falling (Figure 5). Total 350 education expenditures (including both %GDP (RHS) 5.5% 300 core and external expenditures) increased 5.0% 250 from US$250 million in 2003/04 to Core Budget 200 4.5% US$513 million in 2008/09. While External Budget 150 external budget expenditures have been 4.0% 100 stable, core budget expenditures have 50 3.5% constantly increased 15 . Total 0 3.0% expenditures as a share of GDP have fallen 2003/04 2004/05 2005/06 2006/07 2007/08 2008/09 from 5.7 percent in 2003/04 to 4.4 Source: MoE, World Bank percent in 2008/09. In the education sector, MoE spends more than three-quarters of total education expenditures followed by MoHE (about 10 percent). 41. Effective management of teachers’ salaries is an important factor in delivering education services because they account for 90 percent of MoE core operating expenditures. Regarding the current teacher salary system, the existing grade structure for teachers emphasizes seniority rather than competence and responsibility, and is based on the existing civil service salary scale that is very compressed (highest US$65 and lowest US$25 a month) (Table 4). Until recently, the average salary including allowances and bonuses of MoE staff (including 14 Including MoE and MoHE 15 External budget expenditures in 2008/09 may not represent the real picture mainly due to data collection and reporting problems by donors. 15 administration staff) was US$75 per month. Monthly salaries were increased by Af 1,300 (US$26) in May 2008, in response to the protests by teachers over the rapid increase in food prices. Although the increased costs (about US$40 million) were eventually financed through the ARTF, the payments began without clearly identified funding source. 42. Teacher salaries Table 4: Salary scales for teachers and other civil servants after P&G reforms. (US$ per month) The highest and Priority Reform Civil Civil Service Teacher Pay lowest teacher Service and Pay and and Grading salaries in the new Restructuring pay scale Grading proposal scale are US$172 (PRR) pay scale High point 65 239 650 172 and US$118 per Low point 25 82 100 118 month, respectively Source: MoF, World Bank (Table 4).16 Pay and Grading (P&G) reform aims to emphasize teacher competence and make performance assessments the key criteria for merit-based hiring and promotion. P&G reform also intends to replace the current multiplicity of allowances and extra payments with a single salary scale. This is expected to simplify salary processing and reduce the scope for fraudulent payments. MoE hopes the proposed new salary scale, which will significantly increase teacher pay, will attract more qualified teachers and university graduates to the profession. 43. MTFF estimates increased teacher salaries will cost an additional US$85 million per year from 2012. A further US$51 million has been included in the estimate for an extra 37,500 teachers to be recruited by 2012 at the new salary levels (Table 5). However, MoE believes the MTFF estimate is far below actual requirements. A process of re-grading existing positions and assigning staff to new salary levels is being developed, however it is not yet clear how positions will be distributed between the higher and lower salary grades. This will determine the actual cost of the new salary rates. It is also unclear how many of the existing allowances will be removed from the pay system. Table 5 Estimated P&G Costs for Teachers US$ million 2009/10 2010/11 2011/12 2012/13 2013/14 Current salary budget estimate for MoE 248 255 262 268 275 Estimated cost of teacher P&G and new teachers 14 52 101 136 136 Percentage increase in salary costs 5.6 20.5 38.7 50.8 49.5 Source: MoF Equity 44. Regional equity. EMIS data identifies unequal distribution of schools and teachers across provinces, which is reflected in significant variations in participation rates. For example, primary net enrollment rates range from about 30 percent in some areas to almost 100 percent in others. Differences in population densities and terrain, infrastructure, as well as political factors, security issues and differences in the demand for education affect the distribution of education resources. 16 Annex 2 explains more on P&G. 16  Operating expenditures. Operating Figure 6 Relation between Number of Students expenditures are almost in line with and Operating Expenditures in 2007/08 the number of students in different 700 R² = 0.8214 areas (Figure 6). This relatively 600 Badakshan equal distribution is mainly due to 500 wages and salaries of teachers. The 400 national student-to-teacher ratio Panjsher Herat 300 was 40:1 in 2007/08. Although there were more than three times 200 the number of students in Helmand 100 classrooms (70) than in Nuristan - Zabul (21), MoE reports that there are 0 100,000 200,000 300,000 400,000 500,000 600,000 between 30 and 50 students per Source: MoE, World Bank teacher in 25 of the country’s 34 Figure 7 Relation between Number of Students provinces. and Development Expenditures in 2007/08  Core development expenditures17. 200 By contrast, there is little 180 Faryab 160 relationship between core 140 development budget expenditures 120 Nangahar and the number of students (Figure 100 2 7). Faryab and Nangahar provinces 80 R = 0.0219 spent a relatively high core 60 development budget, while Kapisa 40 20 province did not spend a core Kapisa - development budget. 0 100,000 200,000 300,000 400,000 500,000 600,000  Total development expenditures Source: MoE, World Bank (including both core development Figure 8 Relation between New Schools and and external expenditures). Number of Students in 2007/08 External budget expenditures may 300 have offset inequitable core R = 0.6987 2 250 development expenditures. As provincial external budget 200 Kandahar Kabul city allocations are not available, 150 school construction data is used as a proxy for total development 100 18 Helmand budget expenditures . Although 50 Nimroz the location of most school Zabul 0 building projects (especially 0 100,000 200,000 300,000 400,000 500,000 600,000 700,000 800,000 external budget) is determined by Source: MoE, World Bank donors, a strong correlation between the number of new school buildings and students suggests that construction projects to some extent target areas with overcrowding (Figure 8). While some insecure provinces (e.g. Helmand and Zabul) are under the trend line, school construction in 17 Development budget expensed by the central ministry is excluded. About two thirds of the development budget is executed by provinces. 18 The Construction Department database records information on all school building projects constructed by MoE or handed over by external donors since 2006, including schools constructed by PRTs, NGOs and communities. The database contains over 4,000 records but the Department does not have complete information for all projects. 17 Kandahar (also a very insecure province) is well above the trend line, possibly partly due to Canada’s preference for the province19.  Policy measures. Although the NESP acknowledges the need to address regional inequalities, there is no specific plan for addressing the issue. To date, the distribution of resources has mainly responded to demands from PEDs but a more targeted strategy to improve supply in underserved areas should be developed. The 2005 PFM Review suggested a formula approach to address the issue of regional inequity as a more transparent mechanism for allocating resources, including teaching staff. This will require an improvement in the quality of data such as enrollment numbers and school costs. As a formula approach will result in significant resource reallocations, the approach should be implemented gradually, focus on increased funding to less well-resourced provinces within an overall increase in education resources, and take absorption capacity into account. Moreover, a devolved formula funding approach will require developments in governance and accountability at the local level. 45. Gender equality. Girls’ participation has always been low and is reflected in the lower literacy rate for women. While female participation has improved in recent years it varies considerably across the country, ranging from almost 90 percent in Nimroz to less than 10 percent in Helmand and Uruzgan for primary grades. 20 Participation rates for girls are considerably lower in secondary schools. The 2005 PFM Review analyzed factors affecting attendance by girls and found that supply issues such as the distance to the school, adequacy of the building, the availability of a separate school for girls and the presence of a female teacher were more significant than demand-side constraints such as the family’s attitude to education.  Number of female teachers across Figure 9 Female students per female teachers provinces. Twenty-seven percent 350 of current teaching staff are 300 women but they are concentrated 250 in urban areas. Around 70 percent of teachers in Kabul city schools 200 are women but there are many 150 areas of the country where there 100 are no female teachers 21 . The 50 ratio of female teachers to female - ity H an Pa is an ag d an an n K ar G z r t h N zni K n B sa La tika st kh students therefore varies ira ho du an K ga ta Sa ara gh ho h i C h hl m al ha ap is da H G un k hs ilm an B ad gh K F ul ur ak m ab B ad significantly around the country, K B ranging from 23 students per Source: MoE, World Bank teacher in Kabul City to more than 300 in Kunar and Khost (Figure 9).  Policy measures. The low female participation rate in some parts of the country is recognized as an important issue in the NESP and ANDS. The NESP lists several proposed initiatives to promote girls education. They include demand side initiatives such as a communication strategy on girls’ education, the training School Advisory Councils on gender issues, and organizing special events to campaign for girls’ education. The NESP aims to increase the number of female teachers by 50 percent by 2010. Based on EMIS 19 Canada’s PRT is responsible for Kandahar Province. 20 Net enrollment rates based on EMIS survey and CSO population data. 21 According to MoE, about 90 percent of districts do not have female teachers. 18 data on teacher numbers, this would require an additional 20,000 female teachers to be recruited between 2008 and 201022. To meet this target, two-thirds of the 10,000 new teachers recruited each year will have to be women. In 2007, 36 percent of Teacher Training College (TTC) students were girls. 46. Equity issues in MoHE. The small size of the tertiary sector suggests that access to higher education is very inequitable. Given the resource constraints, this will be difficult to address in the short term. The constitutional requirement for free tertiary education is intended to ensure equity of access, but in fact free tuition and accommodation for all students is equivalent to a system of non-means tested scholarships that is likely to favor those who are already well off. There are few additional benefits or incentives for disadvantaged students to attempt university studies. The number of women in tertiary education is very low. The need to travel and live away from home is one of the factors that may discourage women, as well as students from rural areas, from pursuing tertiary-level studies. There is a tension however between the goal of equity for women and rural students, and the policy of improving quality by creating larger, more professional, central universities. To improve gender and regional equity in tertiary education, the first step should be to expand access to secondary education that remains concentrated in urban areas. Efficiency 47. Effectiveness of expenditure is adversely affected by low staff capacity and efficiency. Most teachers have no formal teacher training and some have not finished primary school. In- service teacher training programs and competency tests for existing teachers aim to address this in the coming period. The quality of pre-service training for teachers is poor and resources used for teacher training at the tertiary level are wasted because many graduates do not become teachers. Qualifications officially required for teachers are unrealistic given the current state of the education system and the urgent need for more teachers. Less resource- intensive training options with lower entrance requirements are needed in the short term to fill the gap. Average class sizes and teacher/student ratios reported by MoE are low by developing country standards. Larger class sizes would allow more efficient use of teaching resources and classrooms. The capacity of education administration staff is generally low, as it is in other ministries, and the proportion of unskilled ‘ajeer’ support staff on the payroll (16 percent of all staff) appears high. 48. Government and donors have different views on school construction efficiencies. MoE has raised concerns about donor-financed school construction projects through the external budget, arguing they are unnecessarily expensive, and occasionally of low quality. MoE points to several cases where it demolished school buildings erected by NGOs because of poor workmanship. Donors on the other hand point to excessive delays in the implementation of projects funded through core development budget and examples of misadministration in the use of core budget funds. Donors argue that higher project costs in the external budget are due to higher standards of design and construction. 49. School construction costs by core budget are roughly 50 percent lower. An analysis based on construction department data shows that there is a significant gap in average costs between 22 The EQIP II project includes funding for measures to increase the number of female teachers including a household scholarship and financial incentive scheme to encourage girls to enter a two year TTC program and become teachers on completion. The program aims to recruit 5,000 girls into teacher training over three years. 19 core development and external budget expenditures (Table 6).23 Average construction costs per classroom for the core development budget were US$8,407, while those funded by USAID (including ones implemented by other agencies) were US$12,614, for example. Table 6 Average School Construction Costs in 2006-2008 Total cost Number of Average per Average per Donor (US$ million) schools school (US$) classroom (US$) Core Budget Development Budget 13.9 73 190,568 8,407 EQIP 15.9 298 53,211 8,912 External Budget PRT 14.1 95 148,048 11,603 USAID 14.7 146 100,385 12,614 Source: MoE, World Bank 50. Further studies and cooperation between the government and donors are required. However, school construction contracts arranged through MoE for Education Quality Improvement Program (EQIP) projects for example, include only the actual cost of building works. MoE provides administrative overheads such as design, supervision and procurement through the operating budget or separate donor TA funds. USAID, on the other hand, argues that it has higher design standards for its construction projects and includes additional services in its contracts, including independent quality inspections. A systematic analysis of the unit costs of key activities such as school construction would improve aid effectiveness in the education sector. Sustainability and Estimates of Future Education Expenditures 51. Significant financial resources are required to meet MDGs in 202024. The MDGs aim to achieve full participation in primary education by 2020. Based on available population forecasts, this implies that by 2020 the total number of students will increase by 5.9 million, requiring an additional 147,000 teachers and 84,000 new classrooms (based on current policies on class sizes and teacher/student ratios). However, neither MoE nor MoF have conducted estimates of future education expenditures associated with full participation. A rough analysis suggests that the annual MoE budget will need to increase by around US$553 million from US$313 million in 2008/09 (preliminary actual) (Table 7). 52. Wages and salaries are the main source of the increase25. The need for additional teacher salaries comes from two sources: the increase in student numbers and the implementation of P&G. An adequate number of administrative staff is also essential for effective education management. The other major pressures on the budget will be from infrastructure and building maintenance costs. According to the current plan it appears the existing school construction backlog will be addressed in the next few years. However, further increases in student numbers means another 84,000 classrooms will be required. The education budget has, to date, allocated insufficient resources for operating and maintenance costs. In the future, both existing and new school buildings will incur significant operation and maintenance. 23 Data covers 2006 and onwards. Data on the cost of construction and the number of classrooms are not available for all schools. The database identifies the donor and location but not the type of building so it may include teachers colleges, Islamic schools and dormitories as well as primary and secondary schools. 24 These financial estimates assume that other obstacles to full participation in education can be overcome, such as the supply of teachers, security issues, cultural attitudes and economic disincentives. The analysis envisages a total increase in teacher numbers of around 13,300 per year, which is higher than MTFF estimates of 12,500 additional teachers per year over the next three years. 25 Appendix 2 explains assumptions and calculations of each item. 20 Table 7 Summary of Additional Financing Requirement for MoE in 2020 (constant value, US$ million) Increased Full participation in participation in Total secondary primary school school Operating Expenditures 485 Wages and salaries 231 69 301 Additional teacher salaries 135 64 199 Additional support staff salaries 9 4 13 Impacts of P&G 1/ 87 1 88 Non-salary - - 184 Additional non-salary costs 13 6 19 O&M of existing infrastructure investments - - 53 O&M of new classrooms 76 36 112 Development Expenditures for new classrooms 2/ 46 22 68 Total increase in expenditures - - 553 Increase in students (millions) 4.0 1.9 5.9 Increase in teachers (thousands) 99 47 147 Increase in classrooms (thousands) 57 27 84 1/ including P&G for existing teachers; 2/ annual average costs between 2010 and 2020 Source: MoE, World Bank 53. Financing the education sector. Increased financing will require: (i) a significant increase in domestic revenues; (ii) prioritization of the education sector, and (iii) assistance from the international community. A projection of future financing requirements, based on realistic participation targets integrated into the MTFF is the first step to fully financing the education sector. E. Aid Effectiveness and Donor Coordination 54. Donor Activities. While increasing access to basic education has been the government’s immediate priority, donors have also emphasized the need to pay attention to the quality of education. Striking a balance between the two objectives is challenging. Initial donor assistance focused on emergency school rehabilitation, textbook supply and encouraging children back to school. Development activities more recently have focused on improving quality through training teachers, improving the curriculum and textbooks and direct support to schools. A significant amount of donor funding is also directed to the construction of schools and other education facilities, and to adult literacy programs. The activities of major donors are found in Appendix 3. 55. Donor coordination. Since 2007, the NESP has been an important tool for improving coordination between donors and MoE26. An Education Development Forum organized by MoE to discuss the NESP in February 2008 led to the establishment of an Education Partnership Working Group, chaired by the Canadian International Development Agency (CIDA). MoE is in the process of establishing an ‘Education Development Board’ that will provide technical advice to MoE through ad hoc working groups on specific issues27. However, responsibility for donor coordination within MoE is somewhat unclear. The planning 26 Before 2007 there were numerous formal and informal donor coordination forums dealing with specific issues within the sector. However, there was no effective overall donor coordination forum. 27 The Board will be jointly chaired by MoE and CIDA and will have a secretariat, funded by CIDA, and resources to commission work by consultants. Membership represents major donors, NGOs and MoE. 21 department has a major role through the NESP and maintains a database of external budget projects. There is also a donor coordination unit in the Minister’s office. 56. Possibility of SWAps in MoE. The 2005 PFM Review recommended the possibility of SWAps for MoE. While SWAps can be a strong approach to improve aid effectiveness, immature implementation would waste scarce resources. In addition, each sector has different degrees of readiness for SWAps. In the case of MoE (Table 8), the 2007 NESP set a clear nationally- owned sector policy and strategy, and other progress has been made (e.g. government-led donor coordination). However, there are some areas to be improved further. For example, monitoring and PFM (especially procurement) are not strong enough for SWAps. Although MoE is not fully ready to shift to SWAps, a step-by-step approach should be taken through the ongoing reform process. Program budgeting is expected to improve linkages in the budget cycle (budget formulation, execution and monitoring) and to improve monitoring and evaluation. Achieving SWAps requirements at a program level will also help MoE to shift to SWAps in the coming period. Table 8 Summary Assessments of MoE SWAps Readiness Technical Pre-Condition Status of MoE Policy and Strategy Developing new strategy Institutional Framework Should be improved Sector Budget and Public Financial Management PMF capacity being developed Sector Performance and Monitoring System M&E being developed Sector and Donor Coordination Education development board has strengthened Donor Support Options and Harmonization Willingness among donros Source: Author based on Peter Jensen (2009) Sector-Wide Approach: Assessment of the Suitability and Readiness of SWAp Development Assistance for Afghanistan- Phase II 57. Donor assistance and sustainability. In addition to their strong leadership, donors have also contributed to the development of the education sector through funding and technical assistance. There is thus a risk that unforeseen changes in the policies of donors towards Afghanistan, or towards the education sector, would undermine the recent pace of development. The ministry has around 1,200 national consultants funded by donors (e.g. EQIP employs around 700 local TA). Salaries for donor-funded local TA are much higher than civil servants’ salaries. Participation of civil servants in most reform activities is limited. This poses several risks. National consultants are employed on a short-term basis and there is a loss of expertise when they leave. Most importantly however, national TAs are expensive and their employment is entirely dependent on donor funding. MoE and donors recognize the danger of over-reliance on TA and there are several exercises underway to map and evaluate TA positions in MoE. F. Conclusion and Policy Recommendation28 58. This PER on the whole endorses the current direction of reform and the main recommendation is that MoE maintains the current momentum. Less progress has been seen in the tertiary education sector, however, MoHE has finalized its strategic plan in late 2009. In spite of recent progress, the sector continues to face many difficulties. There are a number of further actions that could be taken, not only by the relevant sector ministries but also by MoF, the government and donors to bolster the development of education in Afghanistan. Outcomes and performance monitoring 28 Most of the suggested actions have been categorised as short term, meaning they should be addressed during the next year. A number of recommendations have been noted as longer-term issues, for consideration over the next two to three years. 22  Progressively implement education quality and outcomes assessment by developing simple arrangements for measuring basic student achievements and identifying a limited number of key measures of performance against NESP and ANDS goals, while taking care to minimize costs and administrative burdens.  To provide a more realistic picture of participation rates, change the practice of keeping children on the enrollment register for an extended period after they stop attending, or make adjustments to provincial enrollment data to reflect the number of ‘permanently absent’ students.  In the longer term, develop processes for reporting on resources used by each school, but only if these processes do not add significantly to the workload or reduce the efficiency of budget execution. PFM and Expenditure Framework  The government should prepare a realistic multiyear operating budget funding strategy for the education sector so MoE and MoHE can plan for the expansion of staff numbers and the allocation of resources between provinces and solicit adequate infrastructure maintenance budgets. Donors, MoE and MoF should also develop a multiyear core development budget-funding plan for the sector.  Improve the collection and reporting of performance indicator data before implementing further enhancements to program budgeting in MoE. Implementation in MoHE should follow a simplified approach focused at the program level. MoF should adopt budget formulation and execution procedures that support program budgeting and budget integration (for example, in budget hearings and negotiations).  Where appropriate, budget and administration responsibilities should be decentralized to PEDs. This must be supported by continuing capacity development for PED staff in budget planning, execution and reporting, adequate funding for administration facilities and equipment, and continued development of internal audit functions at the provincial level.  MoE should prioritize the effective implementation of BPET to ensure that it is fully utilized and consistent with the AFMIS. MoF to continue to roll out AFMIS to all Mustofiats and in due course to PEDs to improve the timeliness and accuracy of budget execution and enable MoE to monitor provincial budget execution more closely.  MoF and MoE should pay more attention to planning and execution of the operating budget including: (i) agreeing to forward estimates of operating budget requirements based on approved staffing levels; (ii) improving the timeliness and predictability of quarterly allotments to PEDs, and (iii) implementing processes in PEDs and the central office to ensure that AFMIS data and MoE records are consistent and that any discrepancies are adequately resolved. In the longer term, MoF and MoE should provide PEDs access to the AFMIS to accelerate expenditure processing, and extend use of BPET to PEDs for expenditure processing.  MoE should continue to improve teacher payroll administration including an internal audit of payroll processes and payments through bank accounts, and carefully monitor implementation of P&G for teachers to ensure that its initial objectives are achieved and that the cost to the budget is controlled.  MoE should take further steps to improve development budget execution rates while MoF, MoE and donors ensure that the multiyear development budget is based on realistic assessments of what can be implemented in each year. 23  MoE and donors shall ensure continuation of the Internal Audit Unit with suitably qualified, independent audit staff.  In the longer term, conduct a survey to assess the level of private out-of-pocket expenditures on education. On the basis of these results, develop strategies, if necessary, to reduce costs to families and improve equity of access to education. Equity  MoE should formulate a comprehensive strategy for promoting gender equity in education including by: (i) giving positive preference to women in teacher training and in employment as teachers; (ii) accelerating training of teachers, particularly female teachers; (iii) ensuring that separate girls schools, and in particular boundary walls around the school, are constructed in areas where this is a factor affecting girls attendance, and (iv) providing adequate sanitation and other facilities at schools. In the longer term, consider possible incentives to encourage women, as well as rural students and the less well off, to enter university.  MoE to develop a strategy to target operating and development funding in areas of low school participation, explore options for equalizing operating and development funding levels and staff allocations between provinces and between schools. In the longer term: (i) move progressively towards allocating financial resources based on numbers of schools and students and policy objectives rather than numbers of teachers, and (ii) provide direct grants to schools for non-salary expenditures to ensure equitable distribution and adequate resourcing. Teacher training  Conduct a tracking survey of primary and secondary teaching graduates to determine how many are employed as teachers.  Develop strategies to rapidly increase the number of trained teachers at primary and secondary levels including: - Accelerated training programs with less demanding entrance requirements that focus on teaching skills rather than academic qualifications. - Improved arrangements for credit transfer and recognition of prior experience to make it easier for trained primary teachers and people with other qualifications to become secondary teachers.  Develop strategies for selecting students who are more likely to become teachers when filling places in tertiary teacher training courses. Aid effectiveness and donor coordination  To the extent possible donors should provide long term and predictable funding to enable better education sector planning and implementation. Donors should improve the alignment of external budget reporting with MoE’s program structure.  Efforts should be made to direct donor resources through government systems and procurement processes to the extent possible, supported by capacity development and monitoring assistance. A step-by-step approach towards SWAps is recommended.  MoE and donors will ensure value for money in the use of national TAs while at the same time training civil servants, but avoid the premature replacement of national TAs and consultants with civil servants at lower salaries. It is also important to harmonize TA salaries in order to reduce the cost of TAs and improve sustainability. 24 Ministry of Higher Education  Implement strategic planning and financial management and administration reforms in MoHE including the integration of operating and development planning and decentralization of administration functions to universities.  If means tested charges for university food and accommodation are not feasible in the short term, MoHE should review policies on assigning students to institutions to encourage them to study close to home to minimize the need for dormitory costs. 25 Appendix 1 Data Limitations in the Education Sector Students and enrollment  In 2007 MoE carried out a comprehensive Education Management Information Survey (EMIS) of all schools and institutions. Provincial and district officials conducted a smaller follow-up survey in 2008 but the results are not yet available.  The EMIS report29 attempted to calculate net enrollment ratios in each province based on population data provided by the Central Statistics Organization (CSO). However, the underlying population estimates for participation rate calculations may be inaccurate. There has been no complete census in Afghanistan for several decades. MoE is developing its own estimates of the expected number of school-age students, to supplement CSO data.  On the other hand, the student enrollment data may also be incorrect. Schools generally keep students on the school roll for three years even when they have stopped attending classes. Between 5 and 25 percent of enrolled students in each province were reported as ‘permanently absent’ in the 2007 survey. The calculation of enrollment rates, teacher-student ratios and per capita funding amounts is difficult in these circumstances. Until reliable provincial population data are available, and enrollments reflect actual participation in education, the available statistical data are mainly useful for tracking changes over time.  Most general education schools cater to both primary and secondary grades for both boys and girls making it difficult to produce disaggregated data on the level of education or gender other than through the EMIS survey. Analysis of budget inputs by level of education or gender is therefore not feasible. Teacher numbers  MoE has made considerable progress in monitoring the number of teaching staff. The lack of central personnel records linked to the payroll and irregular and unreliable salary payment procedures in the past meant there was no clear picture of the number of teachers working in the sector. This led to suggestions of excessive recruitment, inflated staff numbers and ‘ghost’ workers.  Agreement has been reached with MoF on the maximum number of teachers (Tashkeel) that will be funded by the budget, and MoE has taken steps to ensure that this is in fact the number employed and working. Several rounds of teacher registration have been carried out including verification of documents, photographing each person and issuing ID cards. The education sector experiences a high turnover rate – 5 percent per month according to some reports – and an additional 12,500 new positions have been created each year so the number of MoE employees is constantly changing. Responsibility for recruiting staff and maintaining personnel records now rests with PEDs, reducing the time required to fill positions and improving the accuracy of records. The efficiencies from devolution of responsibility to the provincial level need to be balanced, however, against the scope for fraud in salary payments. Robust independent auditing of teacher numbers and payroll, recently initiated by the Internal Audit department, needs to be maintained. Financial data  There have also been improvements in the reliability of the data used by MoE to manage the budget. The monthly reports the provincial offices provide to the central ministry on spending against their budget allotments are generally delayed (or not submitted at all). While they often 29 Ministry of Education, 1386 (2007) Schools Survey: Summary Report. 2008 26 conflict with MoF records, the extension of the AFMIS to some Mustofiats has improved the timeliness of expenditure reporting, making the AFMIS a more reliable source of information. MoE is able to access up-to-date expenditure data through AFMIS terminals in the ministry. However, MoE need to pay more attention to ensuring that MoF records in the AFMIS are reconciled with MoE data.  Information on donor-provided, off-budget resources (i.e. the external budget) is particularly difficult to obtain. Not all donors provide adequate information and there is usually a timing lag. The MoE budget department keeps its own record of external funding for education projects. These various sources of information differ in timing, level of detail and basis of classification.  Finally, there appears to be no information available on the magnitude of out-of-pocket education expenses parents face. While school tuition is officially free, parents may need to pay for uniforms and stationary and there are anecdotal reports of unofficial enrollment charges and fees for textbooks and examinations. 27 Appendix 2: Sustainability and Estimates of Future MoE Expenditures The estimates of additional financial resources required in 2020 shown in Table 5.7 are based on the following assumptions. It should be emphasized that these financial estimates assume that full participation in primary grades is achievable by 2020 and do not take into account non-financial obstacles such as a deterioration in the security situation, cultural attitudes, economic factors that may discourage education participation, and the special needs of nomads and disabled children. Actual budget allocations in 2020 will need to be based on actual enrollment numbers.  Additional primary and secondary teachers (US$199 million). The total number of primary aged children will be 8.8 million in 202030. If all primary aged children enroll there will be 4 million more students than the estimated enrollment in 200931. At the current average ratio of 40 students per teacher this implies a need for around 99,000 additional primary teachers by 2020. MTFF estimates each teacher recruited on the new P&G salary will be paid roughly US$1,360 per year, at a total cost of $134.9 million annually. Assuming dropout rates between grades seven and 12 remain constant, the number of secondary students will increase by 1.9 million, requiring the recruitment of an additional 47,000 secondary teachers to maintain the 40:1 student/teacher ratio. The estimated annual salary cost of these new employees at the P&G rate is US$64.4 million.  Additional support staff (US$13 million). Non-teaching staff currently account for 21 percent of all MoE employees: 5 percent are administrative officials and 16 percent are unskilled ‘ajeer’ staff. An adequate number of administrative staff is essential for effective management of the education system and to enable teachers to focus on teaching. Assuming that the current ratio of non-teaching staff is maintained, there will be a need for an additional 5,000 administration staff and 16,000 ajeer staff for primary education, and 2,000 administrative staff and 8,000 ajeer for secondary education. There is no data on the average salary rates of non-teaching staff so it is assumed that administrative staffs receive an average salary of US$67 per month (based on 2007 averages before P&G reforms) and ajeers receive the lowest civil service salary of US$25.  Impact of P&G on existing staff (US$88 million). The MTFF estimates that the annual cost of implementing P&G for existing teachers by 2020 will be around $85.6 million at current prices. The impact of P&G increases on MoE civil servants, including the new staff needed for increased participation, is not clear. It is likely to be lower than the increases in teachers’ salaries given the large number of low paid ajeer staff. Calculations presume a 20 percent salary increase for new staff.  Development expenditures on new classrooms (US$68 million). Out of 9,062 schools, almost 5,000 did not have buildings in 200732. Since 2006, 1,732 school buildings have been built, 1,933 are under construction and 479 are in the planning stage.33 This suggests that around 4,145 schools will have been constructed between 2006 and the end of 2009, so the backlog of existing school construction requirements identified in EMIS may be addressed over the next few years, possibly within currently committed funding. However, the number of students, and the need for new classrooms grows every year. If most schools operate in two shifts with a 30 MoE estimates 31 This estimate includes children who enroll in Islamic schools for their primary education and the cost estimates are assumed to encompass the cost of expansion of grades 1 to 6 in the Islamic education system. There were 63,000 primary Islamic education students in 2007. 32 EMIS data 33 There is no link between the information on schools constructed or under construction in the database and EMIS data from 2007 on the number of schools required. 28 class size of 35, an additional 57,000 primary grade classrooms will be required by 2020 and a further 27,000 for secondary grades. The total construction costs for primary classrooms will be US$504 million between 2010 and 2020 (roughly US$46 million annually), and US$240 million, or US$21.9 million annually, for secondary grades.  Maintenance costs of new and existing school buildings34. The construction department database does not provide data on the value of all schools built since 2006. On the basis of the available data, it is estimated that about US$350 million has been invested in school construction in recent years. This means that roughly US$52 million per year will be required to maintain existing school buildings at a depreciation rate of 15 percent. This is significantly higher than the non-salary operating expenditures foreseen in the 2009/2010 budget. In addition, between 2010 and 2020, US$68 million per year will be needed to build new primary and secondary classrooms. The cumulative investments translate into annual maintenance costs of US$112 million in 2020 to maintain the value of this investment. 34 In Afghanistan, the construction of a school building does not necessarily create a need for additional staff costs. In most cases, the school is operating without a building and the staff are already employed. Most of the additional costs relate to maintenance. 29 Appendix 3 Activities by Major Donors in the Education Sector General Education  The World Bank’s major contribution has been through the Education Quality Improvement Program (EQIP) that commenced in 2004. EQIP focuses its activities at the school and community level by providing School Management Committees with grants for quality improvements including building works and educational materials. Since 2006, EQIP has established 8,593 School Management Committees, provided US$7.6 million for school improvement grants and funded construction of 828 schools. The program also funds teacher education activities. Implementation of EQIP is one of MoE’s key activities. The program employs around 400 national consultants in the central ministry and PEDs. The second phase of EQIP began in 2008 and is jointly funded by the World Bank and other donors either through ARTF or through external budget contributions. The total cost of EQIP II is estimated at US$186 million over four years.  The US’s major basic education programs focus on teacher training for both primary and secondary teachers. The Building Education Support Systems for Teachers (BESST) program works with MoE to improve primary teacher quality, and the Higher Education Project works with MoHE to improve the performance of its 16 teacher education faculties. USAID has also contributed towards printing new textbooks to support the revised curriculum and provides funds for the establishment of community-based schools in areas where there are no government schools.  Denmark’s contribution has been primarily in textbook production and distribution and in providing technical assistance for administrative reform in MoE. Denmark is also building schools in Helmand province.  Japan is providing funding for school construction in Kabul city and in other provinces including areas bordering Pakistan.  A number of donors are investing in adult literacy programs. Japan is funding a UNESCO program and the US funds UN Habitat. Together, these two programs will provide literacy programs to all provinces.  Other significant donor activities have included UNESCO’s support for MoE Strategic Planning (funded by Norway). Similar assistance is now planned for MoHE. Tertiary Education  In 2005 the World Bank launched a four-year ‘Strengthening Higher Education Program’ that focuses on six major universities (in Balkh, Herat, Kandahar, Nangahar, and two in Kabul) and provides funding for university partnership programs with foreign universities as well as block grants.  In addition to providing support to 16 education faculties in provincial universities, the US funds professional development assistance programs for Afghan university staff, including opportunities to study overseas.  The US has also established the American University of Afghanistan, a private university focusing on business management, IT and public administration. The university currently has around 300 students. Many of these are in the foundation year course that provides a bridge between Afghanistan’s secondary education system and university undergraduate classes.  Many external donor contributions to the tertiary education sector are in the form of scholarship programs to study abroad. These include 500 scholarships to India (US$1.8 million per year), Australian Development Scholarships (US$0.6 million in 2009) and a French scholarship program (US$2.2 million in 2009). 30 The original had problem with text extraction. pdftotext Unable to extract text.