1 Glossary GLOSSARY AfDB African Development Bank AFROSAI African Organization of Supreme Audit Institutions AMTO Assisted Medical Treatment Orders ARDCZ Association of Rural District Councils of Zimbabwe BCC Budget Call Circular BEAM Basic Education Assistance Module BOP Balance of Payments BSP Budget Strategy Paper BSP-Z Better Schools Programme-Zimbabwe CAPEX Capital Expenditure CCT Conditional Cash Transfer CIT Corporate Income Tax CNFA Cultivating New Frontiers in Agriculture COFOG Classification of Functions of Government CPI Consumer Price Index CSC Civil Service Commission DEA Data Envelopment Analysis DFID Department for International Development (United Kingdom) DPO Development Policy Operation DRRR Disaster Response and Risk Reduction DSA Debt Sustainability Analysis DSS Department of Social Services ECD Early Childhood Development EDF Education Development Fund EMIS Education Management Information System EMTP Education Medium Term Plan ETF Education Transition Fund FAO United Nations Food and Agriculture Organization FY Fiscal Year GDP Gross Domestic Product GEC Girls Education Challenge GFMIS Government Financial Management Information System GFSM Government Finance Statistics Manual i Glossary GNU Government of National Unity GoZ Government of Zimbabwe GPE Global Partnership for Education HIV/AIDS Human Immunodeficiency Virus Infection and Acquired Immune Deficiency Syndrome HQ Headquarters HSCT Harmonized Social Cash Transfer System IBRD International Bank for Reconstruction and Development ICT Information and Communications Technologies IDA International Development Association IDBZ Infrastructure Development Bank of Zimbabwe IFIs International Financial Institutions IFMIS Integrated Financial Management Information System IFRS International Financial Reporting Standards IMF International Monetary Fund INTOSAI International Organization of Supreme Audit Institutions I-PRSP Interim Poverty Reduction Strategy Paper IPSAS International Public Sector Accounting Standards IT Information Technology KPA Key Performance Area LA Local Authority LAPF Local Authorities Pension Fund LEAP Livelihood Empowerment Against Poverty program LLECE Latin American Laboratory for Assessment of the Quality of Education MAMID Ministry of Agriculture, Mechanisation and Irrigation Development MASAF Malawi Social Action Fund MDA Ministries, Departments, and Agencies MIS Management Information System MLGPWNH Ministry of Local Government, Public Works, and National Housing MoAMID Ministry of Agriculture, Mechanisation and Irrigation Development MoFED Ministry of Finance and Economic Development MoHCC Ministry of Health and Child Care MoPSE Ministry of Primary and Secondary Education MPSLSW Ministry of Public Service, Labor and Social Welfare MSMECD Ministry of Small and Medium Enterprises and Cooperative Development MTEF Mid-Term Expenditure Framework MTFF Mid-Term Fiscal Framework MWAGCD Ministry of Women Affairs, Gender and Community Development MYIEE Ministry of Youth, Indigenization and Economic Empowerment NGO Non-Governmental Association ii Glossary NPL Non Performing Loan NSSA National Social Security Authority OAG Office of the Auditor General OECD Organisation for Economic Co-operation and Development OPC Office of the President and Cabinet OSISA Open Society Initiative for Southern Africa OVC Orphans and Vulnerable Children PAC Public Accounts Committee PASEC Program on the Analysis of Education Systems PAYG Pay-As-You-Go PCW Public Community Works PER Public Expenditure Review PFM Public Financial Management PFMA Public Finance Management Act PFMEP Public Financial Management Enhancement Project PICES Poverty, Income, Consumption, Expenditure Survey PIM Public Investment Management PIRLS Progress in International Reading Literacy Study PISA Programme for International Student Assessment PIT Personal Income Tax PLAP Performance Lag Address Programme PSNP Productive Safety Net Programme PPPs Public-private Partnerships PSC Public Service Commission PSIP Public Sector Investment Programme PSPF Public Service Pension Fund RBB Results-Based Budgeting RBM Results-Based Management RBZ Reserve Bank of Zimbabwe SACMEQ Southern and Eastern Africa Consortium for Monitoring Educational Quality SACU Southern African Customs Union SADC Southern African Development Community SAP Systems Application Products SDC School Development Committee SDG Sustainable Development Goal SDR Special Drawing Rights SEDCO Small Enterprises Development Corporation   SERA - USAID Strategic Economic Research and Analysis Program (USAID) SERA State Enterprises Reform Agency SFA Stochastic Frontier Analysis iii Glossary SMEDCO Small and Medium Enterprises Development Corporation SMP Staff Monitored Program SEPs State-Owned Enterprises SSN Social Safety Nets STAP Seasonal Targeted Assistance Program TA Technical Assistance TCPL Total Consumption Poverty Line TDIS Teacher Development Information System TIMSS Trends in International Mathematics and Science Study TMS Teacher Minimum Standards UCAZ Urban Councils Association of Zimbabwe UCT Unconditional Cash Transfer UIS Institute for Statistics (UNESCO) UNDP United Nations Development Programme UNESCO United Nations Educational, Scientific and Cultural Organization UNICEF United Nations International Children's Emergency Fund USAID United States Agency for International Development USD United States Dollar VAT Value-Added Tax WFP United Nations World Food Program ZAMCO Zimbabwe Asset Management Company ZIA Zimbabwe Investment Authority ZIMASSET Zimbabwe Agenda for Sustainable Socio-Economic Transformation ZIMRA Zimbabwe Revenue Authority ZIMSTAT Zimbabwe National Statistics Agency ZIMVAC Zimbabwe Vulnerability Assessment Committee ZINARA Zimbabwe National Roads Administration ZINWA Zimbabwe National Water Authority iv Table of Contents TA B L E O F C O N T E N T S Glossary ........................................................................................................................................ i Table of Contents ....................................................................................................................... v Acknowledgements .................................................................................................................... vii Executive Summary .................................................................................................................... viii 1. Introduction ............................................................................................................................. 1 2. Sector Governance .................................................................................................................. 2 A. Institutional Arrangements .......................................................................................................... 2 National Level ............................................................................................................................... 2 Provincial and District Level ........................................................................................................ 3 School level .................................................................................................................................... 4 B. School Financing Mechanisms ..................................................................................................... 5 National Budget ............................................................................................................................ 5 School Fees .................................................................................................................................... 8 Donor Financing ........................................................................................................................... 8 C. Accountability ................................................................................................................................ 10 Public Financial Management Systems ...................................................................................... 10 Human Resource Management .................................................................................................. 11 Results-Based Management Techniques .................................................................................. 11 School-Based Management Systems ......................................................................................... 13 3. Efficiency and Effective Service Delivery ............................................................................. 14 A. Adequacy and Sustainability ....................................................................................................... 14 B. Allocative Efficiency ....................................................................................................................... 14 C. Technical Efficiency ....................................................................................................................... 19 4. Equity ........................................................................................................................................ 26 A. Access and Infrastructure Matters ............................................................................................. 26 B. Resource Allocations Across Schools and School Types .......................................................... 28 C. School Fees and Levies ................................................................................................................ 31 Annex 1 ......................................................................................................................................... 33 Annex 2 ........................................................................................................................................ 36 Figures Figure 1: Organizational Structure of the MoPSE ................................................................................... 2 Figure 2: Provincial Education Office Organizational Structure ...................................................... 3 Figure 3: District Education Office Organizational Structure ............................................................ 3 Figure 4: Structure of the System of Primary and Secondary Education ............................................. 4 Figure 5: Typology of Primary (P) and Secondary (S) Schools .......................................................... 5 Figure 6: Estimated Financing of Primary and Secondary Education by Source, 2014 .................. 5 Figure 7: MoPSE Budget and Expenditure, 2014 (US$ Millions) ......................................................... 6 Figure 8: MoPSE Wage Bill Expenditure, 2011-2015 (US$ Millions) .................................................. 6 Figure 9: MPSLSW Funding of BEAM, 2010-2014 (US$ Millions) ......................................................... 8 Figure 10: Annual Average Spending on Selected Non-Salary Recurrent Costs, 2012-14 ............ 9 Figure 11: Overall Trends in MoPSE Expenditure, 2009-2014 .......................................................... 15 Figure 12: Budget Execution of MoPSE Non-Employment Expense Categories, 2014 .................. 16 Figure 13: MoPSE Spending per Student in Primary and Secondary Education, 2009-2014 ....... 17 v Table of Contents Figure 14: Student-Teacher Ratios in Zimbabwe (2014) and OECD Countries (2013) ................... 17 Figure 15: Enrollment and Examination Pass Rates, 2009-2014 ...................................................... 19 Figure 16: Selected Service Delivery Indicators by Level of Education and Province, 2014 .......... 21 Figure 17: Grade 7 Pass Rates in Relation to Selected School Resources, 2013 ............................. 24 Figure 18: Enrollment Rates and Student to Classroom Ratios, 2014 ................................................. 27 Figure 19: School Spending per Student by Education Level and Funding Source, 2013 .............. 29 Figure 20: School Spending per Student by Education Level and Category, 2013 ......................... 30 Figure 21: Number of Computers per 100 Students and School Fees, 2014 ................................. 32 Tables Table 1: Capital Expenditure as a Percent of Total Education Expenditure ........................................ 15 Table 2: Secondary to Primary Ratio in Public Expenditure per Student ............................................. 16 Table 3: Selected Service Delivery Indicators by Level of Education and Province, 2014 ............. 23 Table 4: Enrollment Rates by Level of Education and Sex, 2014 ...................................................... 26 Table 5: School Income per Student by Level of Education and School Category, 2013 ............... 29 Table 6: School Expenditure per Student by Level of Education and School Category, 2013 ...... 31 Table 7: Top Categories of School Levies and Fees by Level of Education, 2013 .......................... 32 Table 8: MoPSE Budget and Expenditure, Economic Classification, 2009-15 Education, 2013 .... 33 Table 9: MoPSE Budget and Expenditure, Administrative Classification, 2009-15 ........................ 34 Table 10: MoPSE Education Budget and Expenditure, Approximated Program Classification, 2009-15 ..................................................................................................................................................... 35 Table 11: Funding of Major Activities Under the EDF, 2012-2014 ................................................... 36 Table 12: Major Contributors to the EDF, 2012-2014 ............................................................................ 36 Table 13: GPE Funding, 2014/15 .............................................................................................................. 37 Table 14: Unit Costs from the EDF, 2012/13-2014/15 ............................................................................ 37 Boxes Box 1: Program/Results-Based Budgeting in the MoPSE ....................................................................... 12 Box 2: Using Technology to Improve Learning in Large Class Environments: Case of South Korea ... 18 Box 3: Measuring Efficiency at School Level in Primary and Secondary Education ...................... 20 Box 4: International Programs of Student Learning Assessment ................................................... 25 vi Acknolwedgements ACKNOWLEDGEMENTS This volume is the fourth in a series of Public Expenditure Reviews (PERs) prepared jointly by the Government of Zimbabwe and the World Bank. Each report benefited from the support of the Honorable Patrick Chinamasa, Minister for Finance and Economic Development and Mr. W. Manungo, Permanent Secretary of this Ministry. Each report further benefited from the guidance of Camille Nuamah, World Bank Country Manager for Zimbabwe; Mark Roland Thomas, Manager, Macroeconomics & Fiscal Management Global Practice and Guang Zhe Chen, World Bank Country Director for Zimbabwe. The joint team was led by Mr. Z.R. Churu, Principal Director, National Budgets at the Ministry of Finance and Economic Development and at the World Bank by co-team leaders, Johannes (Han) Herderschee and Leif Jensen, both Senior Economists, Macroeconomics & Fiscal Management Global Practice supported by Marko Kwaramba, Economist in the same Global Practice. At the Ministry of Finance and Economic Development the team comprised of Fidelis Ngorora, Director, Public Sector Investment Programme; Samuel Phiri, Principal Economist; Marcos Nyaruwanga, Chief Economist and Brian Goredema, Chief Economist. The volumes were edited by Sean Lothrop, Oscar Parlback and Dean Thompson. Cybil Maradza (Design Consultant) prepared the design and typesetting. Nyasha Munditi (Consultant) completed copy editing. Photos presented in the report were taken by Arne Hoel. Farai Sekeramayi-Noble organized the workshops and other communication events and managed all contracts involved. This volume four of the Public Expenditure Review (PER) series was prepared jointly by Zimbabwe’s Ministry of Primary and Secondary Education (MOPSE) and the World Bank. At the Ministry of MOPSE the team benefited from the support of the Honorable L. Dokora, Minister of Primary and Secondary Education and Permanent Secretary Dr. S.J. Utete-Masango. The MOPSE team was led by Mr. Peter Muzawazi (Director of Policy and Planning, Research and Development, MOPSE) with invaluable contributions received from Hon. Dr. L.D.K. Dokora (Minister, MOPSE) and Dr. S.J. Utete- Masango (Permanent Secretary, MOPSE). School-level data that formed the basis of the analysis was made available by the MOPSE Education Management Information System (EMIS) team led by Mr. Langton Tambandini. The World Bank team consisted of Igor Kheyfets, Senior Economist, Education Global Practice; Janine Mans, Consultant, USAID Zimbabwe Strategic Economic Research and Analysis (SERA) Program; Thanh Thi Mai, Senior Education Specialist, Education Global Practice; and Tapfuma Ronald Jongwe, Consultant, Education Global Practice with input from Johannes (Han) Herderschee; Leif Jensen; Camille Nuamah, all mentioned above; and Sajitha Bashir, Practice Manager, Education Global Practice. Peer review comments from Fadila Caillaud (Senior Economist GED01) are gratefully acknowledged. The report was finalized in October 2016 and macroeconomic indicators for 2016 were updated on March 6, 2017. vii Primary and Secondary Education EXECUTIVE SUMMARY Control/Challenges • Zimbabwe spends significant resources on education, both public and private. The current economic growth and fiscal challenges leave little room for further increase of this spending envelop in the short and medium term. • The need to provide equal opportunities to quality education for all Zimbabwean children is paramount, which calls for more effective use of the available resources. • This PER analysis points to several imbalances in the education spending. Zimbabwe’s secondary- to-primary-student public spending ratio of 1.66:1 far exceeds the averages for OECD countries (1.15:1) and other developing countries with similar demographic characteristics (ranging from 0.83- 1.14:1). Of particular note is the very high share of public spending on salaries, crowding out public resources for infrastructure development and quality improvement. Total families’ contribution is equally large but the capacity to contribute varies with poorer communities not capable to pay fully as their income is limited. This resulted in a wide variation of actual per student spending at the school level which depends almost exclusively on parents’ contribution and posed the risk of students dropping out and not completing schools as expected. Choices/Options • One way to improve the effectiveness of resource utilization is to budget and monitor spending using unit costs. For the wage bill, the key variable is the Pupil-Teacher Ratio. Zimbabwe’s average PTR for junior education is not high, and even low by international standards in secondary education. PTRs can however go as high as 84 in junior and 61 in secondary while they can also go as low as 5 in junior and 3 in secondary, pointing to the existence of very crowded classes as well as extremely small classes. For the capital spending, the key variable is the Pupil-Classroom Ratio. Again there is a significant variation in this ratio across levels of education and across provinces. • The unit cost approach will guide the discussion on whether (i) they are reasonable; and (ii) rooms for reduction (if deemed unreasonably high) or increase (if deemed unreasonably low) can be identified. This will help frame the discussion on who need more resources and why, enabling the savings to be used for equalization or need-based targeting investment to take place. • It will also strengthen the Program-Based Budgeting efforts that are being implemented in the education sector in Zimbabwe. The specification of the outcomes and in particular outputs to be delivered at the end of the budget cycle should be developed in light of the unit costs. Spending proposal should be justified in terms of how many more students will be served. Unit cost budgeting will also enable a more structured and transparent framework to make trade-off decisions (between levels of education, salary vs non-salary/capital or public/private share) especially when the system faces hard budget constraints. Coordination • This approach calls for making good use of essential education and finance data. Zimbabwe has rich databases (EMIS, Examination, TDIS, Payroll and IFMIS) but they need to be linked, analyzed and used in the budgeting, implementation and monitoring processes. • Evidence generated by the data will need to be shared and used by the key decision making bodies (MoPSE, Ministry of Higher and Tertiary education, Public Service Commission, Ministry of Finance and Economic Development and Provinces) in a coordinated manner. By working with a broader set of partners to make the most of the available data, Zimbabwe could develop innovative solutions to tackle the problems in education even in a fiscally challenging situation. viii Primary and Secondary Education 1 INTRODUCTION As Zimbabwe recovered from the protracted economic crisis from 2000 to 2008, the Government of Zimbabwe (GoZ) made major efforts to restore service delivery in education. Public funding for primary and secondary education rebounded from two percent of GDP in 2009 to about 5.4 percent of GDP in 2013. Owing to strong investment in education by the GoZ, households, and donors, pass rates by 2013 had risen to more than 50 percent for seventh grade exams,² and textbook/pupil ratios had reached 1:1 for core subjects.³ School attendance recovered. Primary net attendance rose from 90 percent in 2009 to 94 percent in 2014. Secondary net attendance increased from 45 percent to 58 percent over the same period. ECD programs expanded, with about 22 percent of three- to five-year- old children attending ECD programs in 2014 - up from 18 percent in 2009.⁴ Expanding access to high-quality education is a major policy priority for Zimbabwe. The Zimbabwe Agenda for Sustainable Socio-Economic Transformation (ZimAsset), the GoZ’s plan to achieve sustainable development and social equity, provides a strong mandate for investment in education. It calls on the government to “continue to improve the quality of education from ECD to vocational and tertiary levels to enhance literacy levels and skills development.”⁵ The goal of the Education Medium Term Plan (EMTP) 2011-2015, endorsed by the Cabinet in May 2012, is to “revitalize the provision of relevant, quality, inclusive and holistic education, sport, arts and culture for all Zimbabweans in line with the MDG targets by 2015,” and set ambitious targets across seven strategic priorities, including: (i) restore the professional status of teachers; (ii) revitalize learning quality and relevance; (iii) restore and improve conditions of learning; (iv) ensure quality assurance and staff development; (v) reinvigorate school and system governance and management; (vi) focus resources on those with greatest needs; and (vii) revitalize sports, arts, and culture. When economic growth slowed again in 2012, the MoPSE and donors involved in education conducted a strategic review and agreed on a revised operational plan accounting for the smaller than anticipated fiscal space for education. The development of the next five-year sector plan for 2016-2020 is underway. Initial public consultations involved 600,000 participants at more than 8,000 school centers. Though the need for fiscal funding for education is universally accepted, communities voiced desires to contribute to the education of their children. A school financing policy is to be developed under this plan, recognizing the importance of both public and private education funding. The following volume analyzes public, household, and donor expenditures in primary and secondary education during economic recovery from 2009 to 2015, focusing on the effectiveness, efficiency, and equity of spending. The analysis covers all expenditures financed by the Ministry of Primary and Secondary Education (MoPSE),⁶ and expenditures of the Basic Education Assistance Module (BEAM) program of the Ministry of Public Service, Labor and Social Welfare’s (MPSLSW). It also covers donor and household financing. ² Up from 39.7 percent in 2009. ³ An increase from 2008, when the textbook/pupil ratio was about 1:8 for primary students and 1:16 for secondary students, as per DFID monitoring reports and the OSISA. ⁴ According to the 2014 MICS Report and the 2009 MICS Report, respectively. ⁵ Page 39, Zimbabwe Agenda for Sustainable Socio-Economic Transformation (ZimAsset). October 2013-December 2018. Government of Zimbabwe. ⁶ The MoPSE was previously known as the Ministry of Education, Sport, Arts and Culture, but its name was changed in 2013. 1 Primary and Secondary Education 2 SECTOR GOVERNANCE A . I N S T I T U T I O N A L A R RA N G E M E N T S National Level The MoPSE is the key national-level institution responsible for primary and secondary education. Its mission is to “promote and facilitate the equitable provision of quality, inclusive and relevant Infant, Junior and Secondary Education.”⁷ The MoPSE sets policy; monitors performance; maintains quality assurance; supports capacity building and operations at the provincial and district levels; and manages resources and personnel. Figure 1 illustrates the MoPSE’s core organizational structure. Figure 1: Organizational Structure of the MoPSE Minister STATUTORY BODIES Deputy Minister National Library and Documentation Services Zimbabwe Schools’ Examination Council Permanent Secretary Infant Education & Learner Support Services Finance & Administration Junior Education Internal Audit Secondary & Non-Formal Education Human Resources & Discipline Education Coordination & Development Legal Services Source: MoPSE Establishment Table, as of December 31, 2013 (latest available). Other national-level ministries are involved in education through specialized programs - most focusing on students’ health and well-being, or promoting equity in education spending. As mentioned MPSLSW administers the BEAM program, which provides financing directly to primary and secondary schools to cover fees for economically disadvantaged children, especially orphans. The Ministry of Youth, Indigenization and Economic Empowerment supports technical and vocational education programs in the country. Finally, the Ministry of Higher and Tertiary Education, Science and Technology Development oversees the upper levels of Zimbabwe's education system. ⁷ Zimbabwe MoPSE Vision, Mission and Core Values. 2 Primary and Secondary Education Provincial and District Level At the sub-national level, the MoPSE has offices in each of Zimbabwe’s ten administrative provinces, and each of its 72 districts. The Provincial Office is largely responsible for: (1) implementing policies determined at the national level; (2) overseeing certain HR management and recruitment functions for the province; (3) assisting with planning and infrastructure; (4) managing the process of registration of new schools; and (5) supporting district level offices and schools as needed. District offices, on the other hand, are continuously engaged with schools in districts, conducting financial oversight and education quality assurance, and helping schools manage other issues as they arise. The structures of the provincial and district education offices are shown in Figures 2 and 3. Figure 2: Provincial Education Office Organizational Structure Provincial Education Director Deputy Provincial Director Planning & Infrastructure Deputy Provincial Director - Infant & Junior Secondary & Non-Formal Education Officers Education Officers Education Inspection Education Inspection Learning Finance & Internal Human Other Support Services Administration Audit Resources Education Accounting Internal Audit Recruitment Sports & Psychologist Recreation Records Keeping Evaluation Speech Therapist Arts & Culture Other admin Management Special Needs Source: MoPSE Establishment Table, as of December 31, 2013 (latest available). Figure 3: District Education Office Organizational Structure District Education Director Education Finance & Human Other Inspection Administration Resources Infant & Junior Accounting Recruitment Literacy Coord. Secondary Records Keeping Evaluation Arts & Culture (from province) Management Sports & Recreation Source: MoPSE Establishment Table, as of December 31, 2013 (latest available). 3 Primary and Secondary Education School Level For the formal primary and secondary systems, Zimbabwe follows a modified 7-4-2 structure,⁸ with an additional two years of early childhood education. Following a renewed effort in 2005 to reinforce ECD, Zimbabwe added two additional years to primary education, and in 2014, split Infant Education and Junior Education.⁹ Infant Education includes two years of pre-school (ECD A and ECD B), and Grades 1 and 2. Junior Education includes Grades 3 through 7. Lower secondary is commonly referred to as Ordinary Level (O-Level), and includes Form 1-4. Upper secondary, referred to as Advanced Level (A-Level), includes two years known as “Lower 6” and “Upper 6.” Students participate in terminal exams at the end of Grade 7, at the end of Form 4 (O-level), and at the end of Upper 6 (A-level).¹⁰ This progression is summarized in Figure 4. Figure 4: Structure of the System of Primary and Secondary Education Lower Upper Infant Junior Secondary Secondary ECD A Grade 3 ECD B Grade 4 Form 1 Grade 1 Grade 5 Grade 7 Form 2 O-level Lower 6 A-level Grade 2 Grade 6 Exam Form 3 Exam Upper 6 Exam Grade 7 Form 4 Zimbabwe categorizes schools at the primary and secondary levels based on their organizational structure, location, and status of registration with the government. Students may attend either government schools or non-government schools (religious, trust schools, company-owned, community, or others). Schools are further classified according to their location, which determines funding received from the national level. The designations P1 and S1 refer to primary and secondary schools in low- density urban areas, while P2 and S2 refer to schools in high-density urban areas. P3 and S3 refer to schools in rural areas. These categories were created to differentiate between the resources available to different schools. In addition, these categories are used to define the levels of fees that each type might charge its pupils.¹¹ Finally, a school meeting all statutory requirements for government registration will be classified as a “registered school”. Schools that have not yet met these requirements, but are on their way to achieving them, are known as “satellite schools” and operate under the guidance of a registered “mother” school. Figure 5 summarizes these overlapping categories of schools. ⁸ As of 2013, the UNESCO Institute for Statistics (UIS) classified Zimbabwe as having a 7-2-4 school structure, which is unaligned with the nationally-defined school system structure of 7-4-2. ⁹ Provisions are set out in the Secretary’s Circular No. 14 of 2004 (ECEC); Director’s Circular No. 12 of 2005 (ECD); and SI 106 of 2005 (ECD centers); 2014 was the first year when Infant Education represented a separate sub-vote in the budget for the MoPSE. ¹⁰ Progression from primary to O-Levels is unimpeded by results of 7th grade exams. However, some schools may have selection criteria based on exam scores. Progression from O-Levels to A-Levels is contingent on passing exams, and schools admit students based on performance on exams. ¹¹ This issue is discussed in more detail later in this report. 4 Primary and Secondary Education Figure 5: Typology of Primary (P) and Secondary (S) Schools Registered Government Satellite Registered Non-Government Satellite P1/S1 P2/S2 P3/S3 Source: Authors’ presentation. In addition, Zimbabwe has a number of niche schools, such those serving special needs students. B. SCHOOL FINANCING Figure 6: Estimated Financing of MECHANISMS Primary and Secondary Education by Source, 2014 Primary and secondary education in Zimbabwe is financed through several channels, including the national budget, school fees and levies, donor financing, and other locally generated funds. In 2014, public and private sources each accounted for just under US$800 million, with donor funds adding approximately US$50 million more to the sector (see Figure 6). National Budget The MoPSE's budget should normally fund a mixture of salaries; non-salary recurrent Sources: MoPSE financial reports, EMIS 2014 data, spending; transfers; to provinces; schools and donor reports. and districts, and capital spending. In practice, however, most budgeted and obligated funding is dedicated to salary expenses, leaving little to finance other needs. Despite increases in overall government spending on primary and secondary education, financing of non-salary recurrent expenses remains low and continues to fall. Spending by the MoPSE increased from US$176 million in 2009 to an estimated US$787 million in 2014. However, non- employment recurrent expenditures fell from about $9 million in 2009 to about US$5.9 million in 2014. On a per pupil basis, this represented a drop from US$2.77 per pupil in 2009 to US$1.46 in 2014. Of MoPSE's total budget of US$877 million, only US$38 million (4 percent) was allocated to non-personnel costs in 2014 - and only US$8 million (1 percent) was spent (see Figure 7). In 2014, about 64 percent of MoPSE's budget was allocated to primary education (including ECD), and 34 percent to secondary education. In 2014, MoPSE budgeted three percent to education administration, coordination, and development - though spent only one percent. 5 Primary and Secondary Education Figure 7: MoPSE Budget and Expenditure, 2014 (US$ Millions) A. By economic classification B. By administrative classification Source: MoPSE financial reports. Within the employment costs classification of MoPSE’s budget, substantial growth has occurred in the category of employment allowances, and to a lesser extent base salaries, in primary and secondary education. The MoPSE’s expenditures on the wage bill grew by 83 percent from US$ 487 million in 2011 to US$ 893 million in 2015. Primary education in 2015 accounted for two-thirds of the wage bill, but the growth in wage bill expenditure was split equally between primary and secondary education, with wage bills in both sub-sectors nearly doubling over five years. Meanwhile, expenditure on administration grew slightly. In terms of MoPSE’s expenditure categories within primary and secondary education, employment allowances grew by 120 percent and base salaries 60 percent. Housing and transportation allowances were the main factors in employment allowance costs. Total expenditure on housing allowances grew by more than 150 percent from US$ 71 million in 2011 to US$ 180 million in 2015, while spending on transport allowances rose by more than 110 percent from US$ 71 to US$ 150 million. The sustainability of public education spending depends in large part on the MoPSE’s ability to slow the growth of spending on these allowance categories - and on the wage bill as a whole. Figure 8: MoPSE Wage Bill Expenditure, 2011-2015 (US$ Millions) A. Total Primary and Secondary Education B. Administration $295 m (34%) 6 Primary and Secondary Education C. Primary Education D. Secondary Education Source: The Ministry of Finance and Economic Development. Most spending for school-level recurrent expenditures is based on locally sourced revenues. However, nongovernment schools receive some support from per capita school grants, and government schools receive tuition grants. The central government transfers such grants to schools to purchase school books and learning supplies. Under normal circumstances these transfers would support a fixed amount per pupil. Yet education provinces, having received much less grant revenue than anticipated in the tight fiscal environment, have rationed funds to support the neediest schools. Current low levels of public spending on non-salary support are inadequate to meet the basic costs of running a school, such as paying for teaching and learning supplies, and school utilities. This low level also cannot sufficiently maintain the operations of the MoPSE’s national, provincial, or district offices, which often lack funding for basic needs, such as fuel to conduct audits or quality assurance functions. Similarly, spending on capital expenses is far behind in meeting the needs of the sector. Overall capital spending averaged about US$3.4 million per year between 2009 and 2014—representing only 0.6 percent of overall spending. This spending fell far short of budgeted amounts, which averaged US$8.9 million per year, or about 1.6 percent of the overall budget. Building grants, which are transfers from the central government budget for school capital improvement projects, are schools’ main source of funding for undertaking capital improvements. A final source of funding from the national budget is the BEAM program. BEAM is a transfer program established in 2001 to improve access to quality education for orphans and vulnerable children (OVC). The GoZ initiated funding for BEAM, but donors began contributing by late 2008. This program, which is managed by the Department of Social Services (DSS) of the MPSLSW, transfers funds directly to primary and secondary schools to cover fees and levies of eligible students, and examination fees in secondary schools. From 2010 to 2014, the GoZ budgeted an annual average of US$11.8 million to BEAM, though the yearly budgeted amount varied somewhat—from US$15 million in 2010, 2012, and 2014, to only US$1.5 million in 2013. Actual spending on BEAM also varied, ranging from US$13.5 million in 2010 to only US$1 million in 2013. In 2012 and 2014, $15 million was budgeted, but just US$5 million and US$7 million were spent in these years respectively, as shown in Figure 9. 7 Primary and Secondary Education School Fees Figure 9: MPSLSW Funding of BEAM, Due to fiscal constraints, financing of non- 2010-2014 (US$ Millions) salary educational expenses remains low and falling, which has left many schools without sufficient resources to meet their basic needs. Consequently, the primary and secondary education systems have developed a complex system of school fees and levies,¹² which helps to finance learning materials, pay utilities, finance capital projects, fund non-teaching staff, and in some cases supplement teacher incomes. The national level mandates school fees to cover basic educational costs,¹³ and fees differ according to the type of school (e.g. P1, P2, etc.). School levies, however, are determined by schools to cover other costs, as agreed annually by the School Development Committee (SDC) and approved at the provincial level. These include contributions to fund capital projects, school utility costs, or teaching and learning materials not covered by Source: MPSLSW. the school fees.¹⁴ According to the 2014 EMIS figures, the median school fee paid by households is US$40 per year per primary pupil, and US$95 per year per secondary pupil. This amount varies widely, however, based on the type and location of the school. The reliance on private fees has grown in recent years, putting additional pressure on households unable to afford these payments. In the second half of 2015, for example, the cost of pre-primary and primary education—as measured by the consumer price index (CPI) —grew by more than 11 percent, representing a sharp increase in out-of-pocket costs for many Zimbabwean households. The topic of private fees is discussed in more detail in the section on equity in Zimbabwe’s education system. Donor Financing Domestic sources of financing for education in Zimbabwe, especially financing for recurrent spending, are supplemented by funding from international partners, which play a modest but strategic role in filling the gap in education. Donors contribute approximately US$50 million to Zimbabwe’s education sector each year. From 2009, the largest source of support has been the Education Transition Fund (ETF) and the Education Development Fund (EDF). The ETF is a multi-donor funding mechanism set up in 2010. The ETF’s first phase focused on supporting the education sector to regain its footing after Zimbabwe’s period of instability. The second phase of the ETF reinforced gains by making strategic investments, including helping the GoZ to reduce the textbook-pupil ratios of core subjects¹⁵ from 1:10 to 1:1.¹⁶ As Zimbabwe moved from stabilizing to strengthening the education system, the ETF 1 and 2 were superseded by the EDF in 2012. Also a multi-donor funding mechanism, the EDF provided about US$115 million in funding from 2012 to 2015. Managed by UNICEF, the EDF receives funding from Germany, the United Kingdom, the European Union, Sweden, Finland, Norway, ¹² Some schools may differentiate between types of levies – such as School Development Committee (SDC) levies, building levies, sports levies, and boarding levies or fees, if applicable. Many schools will combine fees and levies in their financial records, and report only the total amounts charged to parents. For simplicity in this report, we designate only between fees, which are a government mandated, and levies. ¹³ They are meant to cover only teaching and learning supplies, though it is unclear if this is closely monitored. ¹⁴ Prior to 1999, fees paid at the school level were collected in a treasury account and redistributed based on government priorities. However, a decision was made to institute a “School Services Fund”, which allows schools to retain fees collected at the source, and use them to finance non-salary costs at that school. ¹⁵ The four core subjects include English, mathematics, environmental science, and one vernacular language. ¹⁶ http://www.educationandtransition.org/wp-content/uploads/2007/04/Zimbabwe_EEPCT_2010_Report.pdf. 8 Primary and Secondary Education and the Open Society Initiative for Southern Africa (OSISA). Table 11 in Annex 2 provides details on major activities funded by the EDF. The Global Partnership for Education (GPE)—a multilateral organization helping developing countries to improve their education systems—began supporting Zimbabwe in 2013, through a grant of US$ 23.6 million for programming between 2013 to 2015. The lead coordinating agency for GPE in Zimbabwe is the United Kingdom’s Department for International Development (DFID). The lead managing agency is the United Nations Children’s Fund (UNICEF). Of the about $50 million provided by the EDF and the GPE in 2014, roughly half went directly to help schools implement their development plans.¹⁷ Table 13 in Annex 2 provides details on the major activities funded by the GPE. DFID also supports activities in Zimbabwe related to education. This includes £12 million dedicated to education under DFID’s £24 million contribution to the Child Protection Fund (CPF), a multi-donor trust fund, which supports phase II of the government’s National Action Plan for Orphans and Vulnerable Children. In addition, DFID supports two centrally funded projects under the Girls Education Challenge (GEC), providing £24 million to two NGOs to support additional bursaries for girls’ education, and a transformative program to empower girls within their schools and communities. In addition, donors have contributed to the BEAM program, which subsidizes education for orphans and vulnerable children. Donors began contributing to BEAM in late 2008 through the CPF. The first round of funding, however, had negligible impact due to hyperinflation. During the next round of funding from 2009 to 2011, BEAM disbursed US$60.2 million directly to primary and secondary schools to cover fees, levies, and examination fees in secondary schools—45 percent of which was provided by donors. ¹⁸ Since 2012, donors have phased out support for BEAM, though several donors continued to support the CPF through end-2015. Given the GoZ’s funding shortfalls for non-salary recurrent spending, donors have provided the most resources for such costs (see Figure 10). From 2012 through 2014, EDF provided more than five times the spending of MoPSE on school monitoring and supervision, and 7.5 times the spending of MoPSE on educational materials and supplies. Notes: Figure 10: Annual Average Spending on ¹ School monitoring, supervision, and support: Selected Non-Salary Recurrent Costs, For the EDF, includes amounts spent on the 2012-14 School monitoring, supervision, and support sub- program; For the MoPSE, includes amounts spent on vehicle maintenance, oil and fuel, and quality assurance programs for each sub-vote. ² Education materials and supplies: For the EDF, includes amounts spent on the purchase and delivery of teaching and learning materials, and provision of text books sub-programs; For the MoPSE, includes all amounts spent on education materials and supplies line items, as well as line items for programs, including the Better schools program, Schools on the shop floor, School library book fund, Education revitalization, Zim-science kit, Syllabus development and printing, E-learning in secondary schools, and Rural pre-schools. Note: Data presented in Figure 10 does not include expenditure covered by school fees and levies. As noted above, school fees and levies contribute significantly to these costs ¹⁷ http://www.dailynews.co.zw/articles/2015/02/27/increase-education-budget-unicef. ¹⁸ Smith, Harvey, Patrick Chiroro and Paul Musker “2012 Zimbabwe: Evaluation of the Basic Education Assistance Module Programme” UNICEF, accessed online 4/16/2015 http://www.unicef.org/evaldatabase/files/BEAM_Evaluation_Final_Report.pdf. 9 Primary and Secondary Education Current spending allocations are insufficient to meet school funding requirements: many P3 and S3 schools have only basic facilities and supplies, and households bear most costs. As donor support diminishes, the MoPSE will need to consider options for increasing budgetary spending to sustain school operations. For example, the MoPSE will soon need to start replacing ETF2/EDF-funded textbooks, which were provided in 2012 but have about a five-year life span. Assuming that the MoPSE replaces a quarter of the old textbooks each year from 2017 to 2020, the MoPSE will need to spend US$2.27 million per year at the primary level, and US$1.39 million per year at the secondary level.¹⁹ Table 14 in Annex 2 provides details on unit costs of supplies and services delivered under the ETF2/EDF. C . A C C O U N TA B I L I T Y Over the past several years, Zimbabwe has put in place new systems and approaches to enhance accountability in using resources at all levels of the education sector. Such approaches include strengthening national public financial management (PFM) and human resource management systems, improving performance management, and reinforcing mechanisms for school-based management. Public Financial Management Systems At the national level, the Public Financial Management System (PFMS) allows the treasury to release funds only if a corresponding budget allocation is authorized and funds are earmarked to cover that expenditure. This system prevents overspending on approved line items,²⁰ except for public utility costs, which are billed post-service based on actual use. Overall, this policy promotes accountability and predictability in the whole PFM system. The MoPSE has strengthened its internal auditing department, though weaknesses remain due to financing gaps. The MoPSE has four chief internal auditors at its national offices—up from one in 2014. At the local level, the Ministry has about 70 auditors, and about 20 vacancies in provinces due to a lack of qualified candidates. Auditors are not always able to fulfil their responsibilities in the field as they lack funding for vehicle repairs and fuel for supervision missions. As a result, auditors tend to more closely supervise schools in urban areas, while smaller, rural schools go several years between audits. The audit department is mounting an ambitious effort to audit all 8,451 institutions (of which about 8,300 are schools) under their purview in a four-year period, though funding for this effort is not guaranteed. Provincial and district offices benefit from the PFMS and electronic payroll, but few goods and services are procured at this level due to low funding for non-employment costs. Provinces and districts have another source of funding for non-salary costs: the Better Schools Programme-Zimbabwe (BSP-Z). Under the BSP-Z, schools transfer between $0.10 and $1 per child of collected school fees to the district level, which provides a fraction to the provincial level. Provincial and district staff are meant to use these resources to finance supervision and teacher trainings benefitting schools. Yet there is less transparency in using these funds, which do not run through the PFMS, as do funds transferred from the central treasury. One recent innovation to improve transparency is UNICEF’s fuel smart-card system, which provides cards to provincial and district offices permitting the easy monitoring and supervision of fuel costs. The MoPSE may consider adopting this system to monitor fuel costs financed through other government programs—such as the BSP-Z. ¹⁹ These calculations include replacement of 4 core textbooks at the primary level (at a cost of $0.86 per textbook) and 4 core textbooks at the secondary level (at a cost of $1.32 per textbook). The projected number of students requiring textbooks is based on estimated enrollments of primary and secondary student, which was calculated by taking an average of progression/ repeat rates by grade, as well as growth rates for the entry of students into ECD A, ECD B, and Grade 1 (as not all students who enter Grade 1 will have necessarily have completed both ECD A and ECD B). ²⁰ Unless a manual order is executed to allow for a cost overrun. 10 Primary and Secondary Education Human Resource Management The MoPSE manages more than 120,000 employees—most of whom are teachers throughout the country. The high number of dispersed teachers makes it challenging to maintain a qualified workforce, ensure teachers and staff report for duty, manage discipline issues, and identify and address cases of absenteeism. Salary payments are submitted to several levels of inspection. Both the Public Service Commission (PSC) and the treasury must concur on salary payments after the head of each department confirms that employees on payroll were on-duty during the period in question. Payroll is executed through an electronic payroll system. Hiring at the provincial and district levels is overseen by the PSC, but the process was recently revised. In mid-2013, the PSC assumed full jurisdiction over hiring decisions of the MoPSE, which proved problematic as the PSC did not always hire teachers for posts for which they were qualified. In January 2015, an adjustment was made to the hiring decision process: the PSC remains on the hiring committee and must provide approval, but MoPSE staff at the provincial and district levels have roles in ensuring that the most qualified candidates are selected. The MoPSE has implemented mechanisms to better manage teacher qualifications. To better catalogue and track its teaching staff, the MoPSE began implementing in 2013 the electronic TDIS database, which includes a census of all teaching staff, their credentials and specialties, and matches between qualifications and current assignments. This database also helps to minimize the possibility of ghost workers. Under its 2011-2015 Medium Term Plan and Operations Strategy, with support from DFID and UNICEF, the MoPSE started a Teacher Development Strategy that established Teacher Professional Standards (TPS). This strategy supports a Continuous Professional Development Program to improve the skills of teachers, implementation of a Performance Lag Address Programme (PLAP), and upgrading skills of ECD paraprofessionals and teachers without O-level math, science, and language. Results-Based Management Techniques The GoZ began implementing Results-Based Management (RBM) in 2006, with a view to reinforcing public sector performance and service delivery through better strategic planning, more efficient use of resources, and improved monitoring and reporting on performance information. In Zimbabwe, RBM involves four main initiatives: (1) personnel performance system, (2) monitoring and evaluation, (3) management information system, and (4) Results-Based Budgeting (RBB). These reforms have been driven by the Office of the President and Cabinet (OPC), which created the Department of Reforms within the OPC to move the process forward. The Ministry of Finance and Economic Development and the PSC have also supported these reforms. A few of these initiatives made strides in recent years. A results-based personnel evaluation system was deployed throughout the public sector. Public sector employees are now annually evaluated against specific results related to their job description. The government is beginning to roll out RBB with the MoPSE and two other pilot ministries that adopted this approach in 2015. More information on RBB is included in Box 1 below. 11 Primary and Secondary Education Box 1: Program/Results-Based Budgeting in the MoPSE In late 2014 and early 2015, the MoPSE began transitioning to Program/Results-Based Budgeting. This form of budgeting re-groups planned and actual expenditures on a set of outputs these expenditures aim to provide, and links spending to output, outcome and impact indicators. Program/Results-Based Budgeting has advantages over traditional forms of line item budgeting, including, among others: (1) increased focus on service provision and sector strategy priorities over administrative divisions; (2) improved planning and operations because real costs of service delivery are clearer; (3) increased information to promote the efficiency and accountability of results; and (4) more informed tradeoffs for budget negotiations.²¹ To begin this process, the MoPSE defined its programs and sub-programs, established key indicators, set baselines and targets for indicators, and translated the current year’s budget from line item format into program budget format. The programs and sub-programs are defined as follows: Program 1: Management & Support Services Program 2: Education Research & Development • Sub-Program 1: H.R. Management & • Sub-Program 1: Curriculum Development Development • Sub-Program 2: Policy Research & Planning • Sub-Program 2: Financial Management • Sub-Program 3: Internal Audit • Sub-Program 4: Legal Services • Sub-Program 5: Administration • Sub-Program 6: Information Technology • Sub-Program 7: Ministerial Affairs and P.S. Program 3: Infant Education Program 4: Junior Education • Sub-Program 1: Teaching & Learning • Sub-Program 1: Teaching & Learning • Sub-Program 2: Quality Assurance • Sub-Program 2: Quality Assurance • Sub-Program 3: Non-Formal Education Program 5: Secondary Education Program 6: Learner Support Services • Sub-Program 1: Teaching & Learning • Sub-Program 1: Learner Welfare Services • Sub-Program 2: Quality Assurance • Sub-Program 2: Special Needs Education • Sub-Program 3: Non-Formal Education • Sub-Program 3: Psychological Services The resulting “program budget” provides a much clearer picture of the costs of service delivery for the ministry. This is because several major costs for service provision cross administrative units in the budget structure, thus obscuring the picture of how the ministry spends its budget. ²¹ http://www.imf.org/external/pubs/ft/wp/2003/wp03169.pdf. 12 Primary and Secondary Education For example, all of the ministry’s spending on school construction and maintenance is executed through the Administrative and General sub-vote, despite representing costs for infrastructure in primary (infant and junior) and secondary education. Another example is budget allocations to infant education. Based on the administrative structure, in the 2015 budget the MoPSE allocated only about US$ 1.8 million to infant education. This amount is much below the actual allocation, however, as the budget for teachers in infant education and grades 1 and 2 was provided in the junior education subvote. The program structure corrects this issue; in the 2015 program budget, infant education received an estimated allocation of US$ 110 million. Annex 1 provides the trends in administrative and program-based budgeting and expenditure. RBB is one of the core elements of the GoZ’s efforts to implement RBM. Efforts to implement RBB have been ongoing since around 2011, though progress accelerated in late 2014 and early 2015. In 2011 and 2013, staff at the Ministry of Finance and select line ministries were trained on RBB concepts. These trainings did not result in major adjustments in budgeting practices until 2014 and early 2015, when authorities signaled to donors that implementing RBB was a top priority. School-Based Management Systems At the school level, the School Head plays a major role in managing school finances. The School Head is involved in annual budgeting and strategic planning exercises, and in administering the budget on a day-to-day basis. If sufficient resources are available, some schools may also have a bursar or school accountant who supports financial management. That said, School Heads or Teachers-in-Charge are teachers rather than managers by training, and often take on school financial management responsibilities without strong backgrounds in financial management. In 2013 and 2014, as many as 90 percent of all School Heads underwent training in financial management at public service training centers with the assistance of UNICEF. A full financial management manual was developed and disseminated, along with reporting templates. Accountability of funds was reported to have improved significantly. This progress could be confirmed if more regular audits are possible. Zimbabwe has developed a system of SDCs as a control mechanism for school finance and service delivery. SDCs have a role in reviewing the annual school development plan and the annual budget required to implement the plan, before the plan and budget go to the full Parents’ Assembly for authorization. The annual budget is accompanied by the estimated school fees and levies required to mobilize annual funding requirements. At the end of the year, the SDCs review the final accounts for the year. Furthermore, SDCs designate a smaller finance committee to help monitor financial management issues more regularly and make decisions on the use of funds. The finance committee should be composed of the School Head, the Deputy Head, a chairperson and a vice chairperson. In addition, each school should have a procurement committee that oversees purchases, evaluates quotations, and justifies purchases based on competitive procurement processes.²² Given the role of SDCs in managing schools, the ETF I and the MoPSE trained two to four SDC members from each school in 2010 to monitor the delivery of learning materials provided by the ETF.²³ The Finance and Administrative Manual provides guidelines for processes and requirements for SDCs, finance committees, and procurement committees. More training or materials for SDC leaders may be warranted, since turnover is quite high, as parents may only be able to commit for one or two years, or leave after their children graduate from school. ²² For more see, GoZ, “Financial Management Training Manual for the Non-Finance Education Managers,” MoPSE, Finance Administration, October 2013. ²³ This activity covered 5519 out of 5644 primary schools in Zimbabwe. For more details, please see “School Development Committee Capacity Reinforcement Project in Zimbabwe April 2011.” SNV Netherlands Development Organisation, April 2011. 13 Primary and Secondary Education EFFICIENCY AND EFFECTIVE 3 SERVICE DELIVERY A . A D E Q U A C Y A N D S U S TA I N A B I L I T Y Zimbabwe’s public expenditure on primary and secondary education has grown considerably since the nadir of the economic crisis in 2008. Overall, the voted budget allocated to the MoPSE increased fivefold—from $177 million in 2009 to $876 million in 2014, then to $890 million in 2015. In actual budget expenditures, spending by the ministry increased from $176 million in 2009 to an estimated $787 million in 2014. However, government spending does not yet fully finance the needs of general education. In fact, private sources are estimated to have contributed $779 million to primary and secondary schools in 2013. Donors contribute an additional $40-50 million to the sector each year, with a large portion going directly to schools. Compared to neighboring countries, Zimbabwe fares well in public spending allocated to primary and secondary education. For example, primary and secondary education (including ECD) in the 2015 MoPSE budget made up about 20 percent of the total 2015 national budget. In comparison, nearby Zambia allocated only 16 percent of its 2015 budget to primary and secondary education. The difference is more pronounced measured as a share of GPD. Though Zimbabwe allocated 6.2 percent of GDP to primary and secondary education in 2015—three times what it spent in 2009—Zambia budgeted only 3.9 percent in 2015 (a modest increase from 2009). Despite the vast improvements since 2008, the adequacy and sustainability of public spending remain a concern for the education sector. The large share of spending from private and, to a lesser extent, donor sources point to inadequate public financing to meet basic educational needs. Financing of non-salary recurrent and capital costs, as discussed below, largely falls within the realm of private contributions. The sustainability of relying on off-budget sources to support schools’ basic operations is a concern for this fast growing sector. B. ALLOCATIVE EFFICIENCY One way to measure the efficiency of public expenditure is to assess whether resources are optimally allocated across different expenditure categories to achieve desired objectives. In other words, are funds being spent in the right areas to ensure a quality education, or would reallocating resources make the sector more efficient? In allocative efficiency, a key concern in Zimbabwe is the high share of the MoPSE budget devoted to salaries. Employment costs dominate spending by the ministry, accounting for virtually the entire increase in MoPSE spending between 2009 and 2014 (see Figure 11). Employment costs rose steadily from $167 million to $779 million over this period. Meanwhile, non-salary recurrent and capital costs decreased from $10 million to $8 million—despite a growing student population. Zimbabwe’s 99 percent of public spending devoted to employment costs—with only 1 percent spent on all other inputs—is among the most lopsided ratios in the world. 14 Primary and Secondary Education Figure 11: Overall Trends in MoPSE Expenditure, 2009-2014 A. Including employment costs B. Excluding employment costs Source: MoPSE financial reports. Non-salary spending is insufficient for two reasons; the first is that the allocation approved in the annual budget is not adequate to fund school investment and operating costs. In the 2014 budget, MoPSE was allocated US$877 million, but only US$38 million—about 4 percent—was allocated to non-salary spending. In the 2015 budget, MoPSE’s allocation to non-salary spending fell to US$17 million—or less than 2 percent. In 2014, this non-salary allocation equaled US$9 per student, far less than necessary to ensure proper learning conditions. In fact, of the total resources for capital expenditures in primary schools, most comes from private sources. In secondary schools, capital financing is sourced more equally between public and private sources. Compared to other countries, Zimbabwe’s school system generally underfunds capital investment (see Table 1), and the low share of publicly financed investment is particularly troubling. Table 1: Capital Expenditure as a Percent of Total Education Expenditure PRIMARY SECONDARY Brazil (2011) 8.6 7.5 Chile n/a n/a Indonesia (2013) 14.0 7.9 Mexico (2012) 2.5 2.6 Turkey (2012) 4.6 6.5 OECD average (2012) 7.1 6.8 Zimbabwe (2013) 5.1 3.4 Of which from public sources: 0.3 1.7 Of which from private sources: 4.8 1.7 Sources: Calculations using data from the EMIS and MoPSE financial reports; OECD (2015), “Education at a Glance,” Table B6.1. 15 Primary and Secondary Education The second reason for inadequate spending on non-salary items is low budget execution. In 2014, the MoPSE reported a 90 percent execution rate of its overall budget. However, no non-salary line item surpassed 61 percent. MoPSE disbursed only US$0.4 million of its budgeted US$0.7 million for maintenance costs. Budget execution for goods and services, together with current and capital transfers, reached only 29 percent. Meanwhile, acquisitions of fixed capital assets and the implementation of special budget programs—such as quality assurance, e-learning, and school feeding—received less than 1 percent of allocated amounts (see Figure 12). In comparison, budget execution for employment costs reached 93 percent. As a result, less than US$2 was spent per child from the 2014 MoPSE budget to finance operating and capital costs in Zimbabwe’s schools. Delaying the release of funds for non-salary spending has deeply negative consequences. Maintenance and operations must be financed according to a regular schedule, and capital investment projects are especially dependent on predictable funding. The failure to release budgeted funds undermines the ability of local education authorities to design and execute performance plans, and undermines the credibility of national authorities. Figure 12: Budget Execution of MoPSE Non-Employment Expense Categories, 2014 Sources: MoPSE financial reports. The allocative efficiency of MoPSE’s spending can also be reviewed by looking at the budget split between primary and secondary Table 2: Secondary to education.²⁴ Primary schools enroll 76 percent of Primary Ratio in Public Expenditure Zimbabwe’s four million students but receive 64 percent of the MoPSE’s total spending. Secondary per Student schools account for 24 percent of enrollment, and 35 percent of MoPSE spending.²⁵ Though Brazil (2012) 0.98 delivering the secondary school curriculum is more costly, as it requires more teaching hours Chile (2013) 0.87 and learning materials per student, the secondary Indonesia (2013) 0.83 to primary ratio in public spending per student is out of line with international comparators. In Mexico (2012) 1.14 Zimbabwe, secondary education receives 1.66 Turkey (2012) 1.13 OECD average (2012) 1.15 ²⁴ For the purpose of this analysis, ECD is included with primary Zimbabwe (2014) 1.66 education. A broader discussion of this topic can be found in the equity section below. ²⁵ The rest, 1% of MoPSE expenditure in 2014, was spent on Sources: Calculations using data from the EMIS and MoPSE administration and coordination. financial reports; OECD (2015), “Education at a Glance,” Table B1.1a. 16 Primary and Secondary Education times as much funding per student as primary education (in 2014, $301 and $181, respectively, with the latter including ECD).²⁶ This ratio has been generally stable in recent years, though it increased slightly from 2009 to 2012. In parallel, the total amount spent per student more than quadrupled between 2009 and 2014 (see Figure 12). In comparison, the secondary to primary ratio of spending per student among OECD countries is approximately 1.15. Brazil, Chile, and Indonesia, which support education systems with young and growing student populations, are more generous to primary education (see Table 2). Figure 13: MoPSE Spending per Student in Primary and Secondary Education, 2009-2014 Sources: MoPSE financial reports and EMIS reports. Figure 14: Student-Teacher Ratios in Zimbabwe (2014) and OECD Countries (2013) Sources: EMIS 2014 preliminary report and OECD (2015), “Education at a Glance,” Table D2.2. ²⁶ All but $2 of this per student amount is used to finance salaries. 17 Primary and Secondary Education As 99 percent of MoPSE spending goes to personnel costs, the difference in per-student allocations between the primary and secondary levels can be attributed to differences in student-teacher ratios. According to the 2014 EMIS, the average student-teacher ratio was 35 in ECD classes, 36 in primary schools, and 23 in secondary schools.²⁷ Though ECD and primary school classes are large by necessity—there are not enough classrooms and qualified teachers—secondary school classes and student-teacher ratios are smaller because more resources are available. In fact, Zimbabwe’s student-teacher ratio in secondary schools is lower than that of Chile, a high-income member of the OECD (see Figure 14). Since only half of Zimbabwe’s students progress to secondary education, the relatively generous resource allocations to secondary schools (exhibited in small class sizes) may be unaffordable in a country working to ensure adequate ECD coverage.²⁸ Box 2: Using Technology to Improve Learning in Large Class Environments: Case of South Korea South Korea is widely viewed to be a pioneer in using Information and Communication Technologies (ICTs) in the classroom. South Korea’s first master plan for ICT in education dates to 1996, and educators have since increasingly integrated ICTs into education at all levels. South Korea’s innovations and dedication to quality teaching have paid off – Korea routinely scores at the top of Programme for International Student Assessment (PISA) rankings for math, science and reading—despite having higher pupil-teacher ratios than other countries in the OECD. A recent trend in the South Korean approach is “SMART Education”—that is, education which is “Self-directed”, keeps students “Motivated”, “Adapts” to the learners’ needs, includes a rich set of “Resources” in the system, and embeds “Technology” throughout the learning process. ICT education in South Korea includes a mix of training on the use of ICTs and ways to use ICTs to teach other subjects, such as language arts, mathematics, and science. The ICT skills curriculum covers computing skills (hardware and software), the role of information in society and ethical issues related to technology, and other IT skills, such as data analytics and web development. Training on ICT skills is widespread at the elementary (primary) level: 77 percent of primary students receive such training. This falls to 22 percent in middle school and 35 percent in high school. This model demonstrates the importance of introducing ICTs to students at an early age to maximize impact. ICTs are used to teach more traditional subjects, such as mathematics, science and language arts, through interactive, digital textbooks, which in many cases supplement physical learning materials. These textbooks allow students to engage with the subject matter in a self-paced manner through interactive learning modules, games, tests, and activities. Teachers have access to learning outcome information to better understand the challenges facing individual students, and customize their teaching methods to individual students. South Korea’s success is due to several factors: alignment of education policy; investments in ICTs at the school level; e-learning embedded in the curricula; trainings of teachers on ICTs in the classroom, and private sector partnerships to develop education content innovations through digital textbooks and e-classroom modules. The master plan for ICT in education was a key step for South Korea to begin mainstreaming ICT in its education system. Building on that experience, the World Bank is working with the MoPSE to develop a draft policy framework for ICT in Education. The policy framework is intended to provide cost-effective, innovative and sustainable options for using ICT to connect schools, districts, and provinces with one another and with the Head Office. The framework will also support decision-making, policy analysis, management, and administration of Zimbabwe’s education system. Sources: Jeong, Euisuk, “Digital Textbook Project in Korea: An Adaptive Model of Digital Textbook in Korea.” Presentation at the Global Symposium on ICT in Education 2014, Gyeongju, Republic of Korea, November 2014 ²⁷ The ECD ratio showed substantial variation across different provinces, ranging from 23 in Bulawayo to 42 in Masvingo. Meanwhile, the primary and secondary ratios were relatively uniform throughout the country, ranging between 33 and 39, and 20 and 24, respectively. ²⁸ Larger class sizes do not necessarily lead to poor education outcomes. An interesting example of a country achieving high quality education with larger than average class sizes is South Korea (Republic of Korea), which has done so partly through the use of technology in the classroom (see Box 2). 18 Primary and Secondary Education Lim, Sungbin, “2014 White Paper on ICT in Education Korea.” Korea Education and Research Information Service, December 2014. Sohn, Byeong-Gil, “ICT in primary and secondary education policy and prospects in Korea.” Presentation at the IADB Conference The Role of ICT4E Policy in Education Transformation, Montevideo, Uruguay, April 2011. In sum, the per-student allocations from public sources are skewed toward employment costs over recurrent and capital expenses, and toward secondary education over primary education and ECD. These imbalances present challenges in allocative efficiency, and heavily influence the equity of public resource allocation. Equity-related aspects are explored in the equity section below. C. TECHNICAL EFFICIENCY In addition to examining resource allocation, efficiency can be measured by assessing how well resources are being utilized to achieve desired objectives after allocations are made.²⁹ This approach, known as technical efficiency, is the study of expenditures in relation to results, expressed through various service delivery indicators. These indicators can be input-oriented, such as student-teacher ratio, student- qualified teacher ratio, student-classroom ratio, computers per 100 students, average class size, and average school size. Indicators can also be output-oriented, such as enrollment rates, completion rates, student learning outcomes, and so forth. Several indicators are discussed in the equity section below. This section focuses on assessing efficiency by analyzing unit costs, student-teacher ratios, and student learning outcomes. Despite the nearly fivefold increase in the primary and secondary education budget from 2009 to 2014, key output measures in education barely changed. According to preliminary 2014 EMIS data, net enrollment rates have fallen slightly since 2009 from 97.7 percent to 92.2 percent, while enrollment in secondary education rose from 38.6 percent to 49.6 percent. The increase in secondary education enrollment was most evident at the A-level (Form 5-6), where net enrollment grew from 4.8 percent in 2009 to 7.2 percent. Meanwhile O-level (Form 1-4) enrollment rates experienced a slight decline. The pass rates of the three main examinations changed little since 2009. The Grade 7 exam, taken by pupils at the end of primary education, improved the most: pass rates increased from 39.7 percent to 50.2 percent from 2009 to 2013, though this is still below the pre-crisis rate of 70 percent in 2007. Despite Zimbabwe’s expanding education budget since 2009, pass rates for the O-level and A-level exams stayed largely unchanged at around 20 percent and 85 percent, respectively—though rates fluctuated moderately from year to year. Figure 15: Enrollment and Examination Pass Rates, 2009-2014 A. Enrollment rates B. Pass rates Source: EMIS 2014 preliminary report. ²⁹ Measuring the efficiency of frontline service delivery units in the public sector is a growing field utilizing various methods to assess the relative efficiency of achieving the desired outcomes by schools, clinics, public administration offices, etc., with a given level of inputs (see Box 3). 19 Primary and Secondary Education As reflected in Figure 16 and Table 3, the unit costs of providing education vary across Zimbabwe’s provinces. The school-level expenditure per student—excluding centrally paid staff salaries—ranges from $35 to $319 in primary schools, and from $267 to $841 in secondary schools. The provinces with the highest costs are not remote, rural areas where costs are expected to be greater owing to higher transport costs and lower class sizes in less dense areas. Instead, the two metropolitan provinces— Bulawayo and Harare—are where spending per student is the highest. This is because urban households tend to have higher living costs and a greater ability to pay school fees, which makes available more resources to urban schools.³⁰ Student-teacher ratios are relatively uniform across the country. The average number of students per teacher employed in primary schools (including ECD) ranges from 32 to 39, while secondary schools average from 20 to 24 across provinces. The large difference in students per teacher between primary and secondary schools is arguably suboptimal. By international standards, Zimbabwe’s provinces range between Chile (25) and the United Kingdom (18) in students per teacher in secondary education,³¹ which may be unaffordable given Zimbabwe’s emergence from a deep economic crisis. Box 3: Measuring Efficiency at School Level in Primary and Secondary Education An education system aims to develop students’ cognitive and non-cognitive skills, consistent with a country’s education policy and priorities. To achieve desired outcomes, which can be measured through student learning assessments or other methods, schools use a mix of available inputs. These inputs, which contribute to the learning process to different degrees, include: teachers with varying levels of training; non-teaching staff; financing—both public and private; learning materials and technology; educational facilities; and so forth. Additional non-school factors influence how inputs are converted into learning outcomes. These factors include the socioeconomic background of students (e.g., parental education and income levels) and characteristics of schools and communities, such as urban-rural location. Non-school factors are outside the control of the school system, but can impact student performance. Technical efficiency is generally measured by comparing the level of output achieved with a given level of input, or by comparing the level of input needed to achieve a given level of output. An education “production function” can be constructed to determine the technical efficiency of each production unit (e.g., a school or a district), which will calculate an “efficiency score” for each unit. Various techniques can derive such scores—some simple and other complex. At its most straightforward, an efficiency analysis can compare schools achieving high marks with low levels of inputs (e.g., low levels of per-student spending and/or below-average socioeconomic status of students). Such schools are deemed relatively technically efficient, while schools with high levels of inputs but low scores are relatively inefficient. Sophisticated methods to calculate efficiency scores include econometric techniques, such as parametric and non-parametric estimation. Two common methods from literature in education and health are Data Envelopment Analysis (DEA) and Stochastic Frontier Analysis (SFA). Both approaches calculate an efficiency score for each school or district by constructing an “efficiency frontier” against which all schools can be mapped. The distance to this frontier is considered that school’s efficiency score. In DEA, for example, units lying on the frontier are considered “perfectly efficient”, and receive the maximum score of 1. This analysis requires reliable measures of student learning outcomes, and detailed school-level data on education inputs (e.g., spending by school on various expenditures, student- teacher ratios) and school context (students’ socioeconomic background). It also requires a high familiarity with advanced econometric techniques, and an ability to work with large data sets using modern statistical software. Increasingly, ministries of education and ministries of finance ³⁰ A thorough review of cost drivers in the urban provinces is underway to investigate whether these preliminary findings are a result of errors in the underlying data or arise from legitimate spending differences among Zimbabwe’s schools. ³¹ OECD (2015), “Education at a Glance.” Table D2.2. OECD calculations done using full-time equivalent teachers. 20 Primary and Secondary Education in different countries are beginning to invest in data and analytical capacities to enable staff to execute such analyses, with a view to identifying how to efficiently deliver high-quality education in resource-scarce environments. For more information see: Coelli, Timothy, D.S. Prasada Rao, Christopher O’Donnel and George Bettese, An Introduction to Efficiency and Productivity Analysis, Second Edition. New York: Springer Science and Business Media Inc., 2005. Cornali, Federica, “Effectiveness and Efficiency of Educational Measures: Evaluation Practices, Indicators and Rhetoric.” Sociology Mind, Vol 2, No. 3, July 2012. Santin, Daniel and Gabriela Sicilia. “Measuring the efficiency of public schools in Uruguay: main drivers and policy implications.” Latin American Economic Review, 2015. Smith, Peter and Andrew Street, “Analysis of Secondary School Efficiency: Final Report.” Department for Education and Skills, United Kingdom, 2006. Yet a sizeable share of teachers lack proper qualifications. The ratio of students per qualified teacher is substantially higher—between 36 and 68 in primary schools (including ECD), and between 26 and 38 in secondary schools. Nationwide, the share of unqualified teachers in primary school is 11 percent, reaching as high as 39 percent in Matabeleland North province. In secondary education, 26 percent of teachers lack proper qualifications. The share of unqualified ECD instructors is a whopping 67 percent nationally, but reaches 90 percent in Matabeleland North and South provinces. Teacher training colleges are working to fill the gap with qualified ECD teachers, but the current output of about 2,000 new teachers per year struggles to keep up with demand.³² Resolving this challenge will be essential for Zimbabwe to attain the Sustainable Development Goal (SDG) #4, which aims to ensure that all girls and boys have access to quality ECD, care, and preprimary education by 2030. In the short term, officials may need to explore alternative methods of providing ECD in areas with high teacher shortages, such as technology-based ECD models, which can support interactive audio instruction. Figure 16: Selected Service Delivery Indicators by Level of Education and Province, 2014 A. Primary Schools 21 Primary and Secondary Education B. Primary Schools Notes: ¹ Expenditure data collected in 2014 corresponds to the 2013 academic year and includes school-level expenditure from all financing sources (public, private, and donor). ² Number of teachers as reported to the TDIS in 2014. ³ Data for primary schools include ECD due to an inability to obtain separate expenditure data for ECD grades. Source: Calculations using EMIS 2014 and TDIS data. ³² MoPSE estimates that 16,568 ECD teachers are currently needed in the system, which now employs 3,960 qualified and 8,164 unqualified ECD teachers. 22 Table 3: Selected Service Delivery Indicators by Level of Education and Province, 2014 Of which: Service Delivery Indicators NUMBER NUMBER TOTAL NUMBER QUALIFIED UNQUALIFIED EXPENDITURE STUDENTS STUDENTS OF OF EXPENDITURE¹ OF TEACHERS TEACHERS PER PER PER SCHOOLS STUDENTS ($) TEACHERS² STUDENT ($) TEACHER QUALIFIED TEACHER Primary Schools³ 5,863 3,086,516 271,431,813 86,253 70,052 16,201 88 36 44 Bulawayo 130 115,738 21,246,786 3,634 3,236 398 184 32 36 Harare 225 275,831 88,059,984 7,342 6,745 597 319 38 41 Manicaland 865 496,274 35,286,672 13,896 11,897 1,999 71 36 42 Mashonaland Central 488 289,072 10,792,146 7,503 5,271 2,232 37 39 55 Mashonaland East 688 334,607 25,004,527 9,350 7,733 1,617 75 36 43 Mashonaland West 719 359,117 26,185,494 9,574 7,644 1,930 73 38 47 Masvingo 866 419,934 25,413,742 12,111 10,771 1,340 61 35 39 Matabeleland North 580 212,777 7,467,995 6,021 3,127 2,894 35 35 68 Matabeleland South 507 178,345 9,580,328 5,028 3,742 1,286 54 35 48 Midlands 795 404,821 22,394,138 11,794 9,886 1,908 55 34 41 Secondary Schools 2,424 979,644 427,018,764 43,361 32,171 11,190 436 23 30 Bulawayo 52 52,226 43,900,729 2,475 1,927 548 841 21 27 Harare 94 96,165 77,502,707 4,238 3,390 848 806 23 28 Manicaland 391 152,696 49,900,186 7,085 5,903 1,182 327 22 26 Mashonaland Central 209 76,678 20,466,497 3,166 2,000 1,166 267 24 38 Mashonaland East 331 121,296 46,114,519 5,164 3,609 1,555 380 23 34 Mashonaland West 348 113,819 32,297,458 4,839 3,370 1,469 284 24 34 Masvingo 338 130,445 47,559,414 5,837 4,289 1,548 365 22 30 Matabeleland North 174 53,790 43,205,171 2,648 1,815 833 803 20 30 Matabeleland South 157 54,582 23,040,522 2,251 1,713 538 422 24 32 Midlands 330 127,947 43,031,559 5,658 4,155 1,503 336 23 31 Notes: ¹ Expenditure data collected in 2014 corresponds to the 2013 academic year and includes school-level expenditure from all financing sources (public, private, and donor). ² Number of teachers as reported to the TDIS in 2014. 23 ³ Primary schools include ECD. Primary and Secondary Education Source: Calculations using EMIS 2014 and TDIS data Primary and Secondary Education Partly due to shortages in school financing and qualified teachers, learning outcomes vary substantially across the country. In primary schools, an average of 49 percent of internal candidates passed Grade 7 (G7) exams in 2013. However, the pass rate ranged from 32 percent (Matabeleland North) to 80 percent (Bulawayo). In secondary schools, 23 percent of O-level internal candidates passed the 2013 exam—ranging from 17 percent in Matabeleland North to 27 percent in Harare. Of those who reached A-levels, the vast majority (83 percent) passed in 2013—ranging from 70 percent in Matabeleland North to 88 percent in Mashonaland East. Not surprisingly, a strong correlation is observed between provinces performing well on exams and those where school resource endowments were more plentiful.³³ This is particularly striking at the primary school level. Provinces with higher per-student spending tend to have higher G7 pass rates (see Figure 17). Provinces with fewer qualified teachers—and higher ratios of students per qualified teacher—tend to have lower G7 pass rates. As consistently shown in international research literature, spending levels are generally not a strong predictor of student learning outcomes, but a minimum level of financial resources is a necessary (but not sufficient) condition for learning. Zimbabwe’s schools and provinces (such as Matabeleland North) that fail to meet this condition are likely to continue providing a low quality education to students. Figure 17: Grade 7 Pass Rates in Relation to Selected School Resources, 2013 A. School Expenditure per Student (US$) B. Students per Qualified Teacher Note: ◊ = national average. Source: Calculations using EMIS 2014 and TDIS data Though supporting a robust system of national examinations, Zimbabwe lacks an effective mechanism for system-wide quality assessment. Various international, regional, and national mechanisms for measuring education quality are being used throughout the world. Zimbabwe is a member of the Southern and Eastern Africa Consortium for Monitoring Educational Quality (SACMEQ), but Zimbabwe’s last set of publicly available SACMEQ assessment results is from 2007. To better understand the relationship between student learning outcomes and the allocation of resources— including issues relating to equity and efficiency—Zimbabwe might consider joining an international program providing regular measurements of the quality of education. One such program is PISA for Development, which is being implemented by the OECD (see Box 4). ³³ While student learning outcomes are affected by a wide range of factors, resource availability is likely to play a role in determining learning outcomes in Zimbabwe. 24 Primary and Secondary Education Box 4: International Programs of Student Learning Assessment The last 20 years has seen tremendous growth in the use of international systems for student learning assessment. Such assessments aim to provide information on education system performance, typically relative to an agreed set of standards or learning goals, to inform education policy and practice. International assessments can be similar to ones used at the national or subnational levels, but they introduce an element of comparability and standardization across countries, facilitating easy benchmarking of results internationally. Popular international systems for student assessment include TIMSS (Trends in International Mathematics and Science Study), PIRLS (Progress in International Reading Literacy Study), and PISA (Programme for International Student Assessment). Regional systems also exist, such as SACMEQ (Southern and Eastern Africa Consortium for Monitoring Educational Quality), PASEC (Programme d’Analyse des Systèmes Educatifs / Program on the Analysis of Education Systems), and LLECE (Latin American Laboratory for Assessment of the Quality of Education). Though systems differ in their goals and the methods for measuring achievement, the top scoring education systems on international assessments are typically considered among the most successful and most emulated in the world. Countries and cities with top scores on PISA 2012, for example, included Shanghai, Hong Kong, Singapore, Japan, South Korea, and Finland—all of which have a reputation for delivering high quality education to their students. The OECD, which runs the PISA initiative, recently developed a program of assessments designed to measure the skills and knowledge of 15 year olds in developing countries. PISA for Development, as it is called, will conduct its first assessments from 2015 and 2018. Seven countries from Africa, Asia, and Latin America have signed up so far, including Cambodia, Ecuador, Guatemala, Honduras, Paraguay, Senegal, and Zambia. In addition to providing a single reference for rigorously gauging progress on educational quality and equity, PISA for Development emphasizes institutional capacity building to strengthen countries’ national assessment systems. By participating in international programs, such as PISA for Development, Zimbabwe can enhance its use of data in managing the education system by adding robust measures of student learning outcomes and skill acquisition to the pool of assessment information already available for policymakers. Sources: https://www.oecd.org/pisa/aboutpisa/pisa-for-development-participating-countries.htm and http://saber.worldbank.org/index.cfm?indx=8&pd=5&sub=0. 25 Primary and Secondary Education 4 EQUITY A . A C C E S S A N D I N F RA S T R U C T U R E M A T T E R S Access to education is one of the basic concepts of educational equity. Access is commonly measured using enrollment rates. In such terms, Zimbabwe has achieved widespread and nearly universal access in primary education: 92 percent of boys and girls aged 6-12 are enrolled in primary school (see Table 4). Except for Harare, no province has a net enrollment rate³⁴ below 90 percent for boys and girls. However, the story is different in ECD and secondary education. The net enrollment rate for ECD was only 25 percent in 2014, with Harare the lowest among provinces at 9 percent. Zimbabwe has commendable gender parity in ECD enrollment, and four provinces (Manicaland, Masvingo, Matabeleland North and South) have reached 30 percent net ECD enrollment. However, much work remains to be done to ensure adequate ECD coverage throughout Zimbabwe. Table 4: Enrollment Rates by Level of Education and Sex, 2014 MALE (%) FEMALE (%) TOTAL (%) GPI GER NER GER NER GER NER GER NER ECD (ages 4-5) 39.5 19.0 39.3 19.3 39.4 19.2 0.99 1.02 Primary (ages 6-12) 109.4 91.9 106.4 92.5 107.9 92.2 0.97 1.01 Lower Secondary / Forms 1-4 (ages 13-16) 73.2 50.8 73.7 56.7 73.4 53.7 1.01 1.12 Upper Secondary / Forms 5-6 (ages 17-18) 12.9 7.6 9.9 6.8 11.4 7.2 0.77 0.90 Notes: ¹ GPI = Gender Parity Index = female enrollment rate / male enrollment rate. ² GER = Gross Enrollment Rate (among children of all ages). NER = Net Enrollment Rate (among children of grade-appropriate ages). Source: The EMIS 2014 preliminary report. Secondary education faces equity challenges, especially in the transition from lower to upper secondary school. Based on net enrollment, 54 percent of 13-16 year olds attend lower secondary (Forms 1-4), but only seven percent of 17-18 year olds enroll in upper secondary. Girls outnumber boys in lower secondary, though the gap is narrowed when considering gross enrollment rates. This implies that a larger share of boys enrolls in Forms 1-4 at a later age. In upper secondary, net enrollment rates are low for both boys (7.6 percent) and girls (6.8 percent). Gross enrollment rates for girls and boys ³⁴ Net enrollment rates are calculated as the number of children enrolled in each level of education from the appropriate age group (e.g., ages 3-5 for ECD) divided by the total number of children in appropriate age group. Gross enrollment rates are calculated as the total number of children enrolled, regardless of age, divided by the total number of children in the appropriate age group. 26 Primary and Secondary Education diverge in upper secondary – with boys attaining a gross enrollment rate of 13 percent and girls a gross enrollment rate of 10 percent. This suggests scope to encourage girls to continue studies in line with the SDG to eliminate gender disparities and ensure equal access to all levels of education. Both boys and girls enroll in A-level classes at a later than expected age. Net enrollment is starkly different across provinces in lower secondary, ranging from 43 percent in Harare to 65 percent in Mashonaland East, and in upper secondary, ranging from 3 percent in Matabeleland North to 10 percent in Bulawayo. In many parts of Zimbabwe, access to education is impaired by inadequate infrastructure. In primary and secondary education, provinces with higher pupil per classroom ratios tend to have lower enrollment rates (see Figure 18). Harare has the lowest enrollment rates in the country. Harare also has the most crowded classrooms—56 students per primary classroom, and 52 per secondary classroom, substantially exceeding the national averages of 45 and 44, respectively. However, higher enrollment rates in ECD are associated with higher pupil per classroom ratios, suggesting schools may be reluctant to turn children away, instead opting to pack them into severely overcrowded classrooms. The quality of ECD likely suffers as a result. In some provinces, pupil per classroom ratios in ECD exceed 100, such as 107 in Mashonaland Central and 124 in Matabeleland North. Moreover, education authorities would likely find it difficult to provide a quality ECD education even at the national average of 73 children per classroom, suggesting a need to build or identify additional classrooms. Figure 18: Enrollment Rates and Student to Classroom Ratios, 2014 A. ECD B. Primary C. Secondary Note: ◊ = national average. Source: The EMIS 2014 preliminary report. Expanding access to education in these areas requires scaling up public resources and capital investment to fill the infrastructure gap. The most pressing need is in ECD, where reducing the pupil per classroom ratio to 40 students will require at least 4,812 new ECD classrooms—almost double the current stock—assuming no increase in ECD enrollment. Reducing the ratios to 40 students per classroom in primary and secondary education will require constructing 7,911 new primary classrooms and 2,056 new secondary classrooms. Based on MoPSE estimates, these figures translate into a shortage of 1,252 primary schools and 804 secondary schools. The ECD program has been fully incorporated into the primary school curriculum under the Infant Module. The targeted 1,252 primary schools will be mandated to have full-fledged Infant Module Centres. Such ECD investments are anticipated to be included in the new Education Sector Plan for 2016-2020. One concern about the boom of new community schools, which resulted from the 2000 land reform, is that many were not established in line with the normative framework. The rules of the Directors Policy Circular No. 73 of 1991, which guides the establishment and planning of new schools, were not fully applied in many recently set-up satellite schools. The MoPSE is making efforts to ensure that all schools established henceforth comply with guidelines, and is 27 Primary and Secondary Education considering amending the Circular to accommodate the Infant School Module (ECD). The MoPSE may give priority to public over private schools when establishing new institutions, with a view to reducing user fees in selected areas.³⁵ At present, infrastructure conditions in many schools do not provide an adequate learning environment. Half of Zimbabwe’s schools report having no electricity, and one-third have insufficient or inconsistent water supply. Among primary schools, 52 percent lack electricity, though this share is even higher among P3 schools (58 percent), and primary schools in certain provinces, such as in Matabeleland North (67 percent). Sufficient and consistent water supply is available in only 65 percent of primary and 64 percent of secondary schools in the country. B. RESOURCE ALLOCATIONS ACROSS SCHOOLS AND SCHOOL TYPES The resources available to different primary and secondary schools varies substantially across Zimbabwe. According to 2014 EMIS data, schools reported a total of $814 million in income during the 2013 academic year. Of this, $779 million (96 percent) came from private sources, $32 million (4 percent) from public sources,³⁶ and $3 million (0.4 percent) from external sources. This amounts to $134 in school revenues per student enrolled in primary education, and $458 per student in secondary education (see Figure 19). Private financing comes from a multitude of levies and fees. Invariably, schools serving children from more disadvantaged socioeconomic backgrounds have less capacity to collect private funds, leading to highly unequal resources available to different categories of schools. Though public resources are generally allocated to categories of schools that have lower capacity to generate own-source revenue (such as P3 and S3 schools), the distribution of private funds raised by schools is highly regressive. Public financing is transferred to schools through four main channels: per capita grants, salary grants, building grants, and BEAM payments. All of these are distributed in a generally progressive manner. For example, a larger share of public transfers benefit P3 and S3 schools, whose students tend to come from the most disadvantaged backgrounds, who make up 79 percent and 69 percent of primary and secondary school students, respectively. External donor funds reach schools in a generally regressive manner: P1 and S1 schools receive four times as much funds per student as other schools. But the most regressive flows are from private sources. In 2013, P1 schools received $608 per student in 2013—twice the amount received by P2 schools, and a staggering nine times more than P3 schools. Similarly, S1 schools—receiving $1,827 per student—collected four times as much in fees and levies as S2 schools, and eight times more than S3 schools (see Table 5). ³⁵ The proliferation of community schools in rural areas has likely exacerbated the need for infrastructure investment in satellite facilities, which do not meet the standards to become registered schools. This, in turn, has led to higher fees to be charged to households in these areas. ³⁶ These totals exclude staff salaries, which are paid directly by the government and make up the vast majority of public spending on education. 28 Primary and Secondary Education Figure 19: School Spending per Student by Education Level and Funding Source, 2013 A. Primary B. Secondary Total = $134 per student Total = $458 per student Source: Calculations using EMIS 2014 data. Table 5: School Income per Student by Level of Education and School Category, 2013 Of which ($): TOTAL TOTAL INCOME PER PRIVATE PUBLIC EXTERNAL ENROLLMENT INCOME STUDENT (US$) (US$) Primary P1 140,562 86,143,013 613 608 3 1.⁹⁴ P2 509,328 138,322,746 272 268 3 0.⁴⁷ P3 2,386,888 181,685,080 76 70 5 0.⁵⁹ Secondary S1 82,367 152,686,184 1,854 1,827 22 4.⁹⁰ S2 196,654 95,268,389 484 474 9 1.²⁴ S3 611,129 159,883,535 262 239 21 1.²⁹ Source: Calculations using EMIS 2014 data. Large differences in spending per student on categories ranging from textbooks to teacher training profoundly alter the opportunities available to students in different schools. Primary schools spent an average of US$89 per student in 2013—just under a third of which went to salaries³⁷ 29 Primary and Secondary Education and benefits (US$28) and building construction and repairs (US$17). Secondary schools spent five times as much—US$480 per student—of which compensation made up a slightly smaller share (US$127). The rest was spread more or less evenly across sports; textbooks and learning equipment; buildings and repairs; boarding, and other costs (see Figure 20). Figure 20: School Spending per Student by Education Level and Category, 2013 A. Primary B. Secondary Total expenditure = $89 per student Total expenditure = $480 per student Source: Calculations using EMIS 2014 data. However, large differences in equity were observed in level of education. Primary schools in more affluent areas (P1) spent on average US$634 per child, four times more than P2 schools and 14 times more than P3 schools (see Table 6). P3 schools spent only US$45 per student in the 2013 school year—one-third of which (US$14) funded building construction and repairs, with little left for textbooks and learning materials (US$5) and school management and teacher training (US$4). The differences are nearly as stark in secondary education. Though S1 schools spent US$2,049 per student, S2 and S3 schools spent $638 and US$217, respectively. S1 schools thus spent three times as much as S2 schools, and nine times as much as S3 schools. In short, though P3 and S3 schools enroll three-quarters of all students in Zimbabwe, they have next to no resources that can be used on teacher training and learning materials—two crucial inputs in the educational process. ³⁷ This does not include the salaries of staff officially employed and paid by the government. 30 Primary and Secondary Education Table 6: School Expenditure per Student by Level of Education and School Category, 2013 Of which: SCHOOL MANAGEMENT & TRAINING EXPENDITURE PER STUDENT (US$) TEXTBOOKS, EQUIPMENT, EXAMS SALARIES & FRINGE BENEFIT TOTAL EXPENDITURE (US$) SPORT, ARTS, CULTURE UTILITIES, TRANSPORT, TOTAL ENROLLMENT BUILDING REPAIRS COMMUNICATION BOARDING OTHER Primary P1 140,562 89,110,942 634 316 28 58 51 12 23 65 80 P2 509,328 73,822,220 145 38 7 15 36 9 2 15 23 P3 2,386,888 108,459,511 45 9 4 2 5 5 1 14 5 Secondary S1 82,367 168,775,448 2,049 721 81 109 139 269 278 145 307 S2 196,654 125,441,547 638 145 29 25 81 169 60 53 76 S3 611,129 132,772,786 217 41 13 13 43 13 26 33 34 Source: Calculations using EMIS 2014 data C. SCHOOL FEES AND LEVIES As demonstrated above, private financing accounts for the overwhelming majority of resources available at the school level. According to 2014 EMIS data, virtually all schools charge some combination of levies and fees. The median amount charged by primary schools was $40 per student in 2013. The median for secondary schools was $95. These amounts vary widely across school categories—from $36 in P3 schools to $253 in P1 schools, and from $80 in S3 schools to $643 in the wealthier S1 schools. Fees and levies finance everything from capital improvements and classroom construction to staff salaries and learning supplies. In 2013, the single largest category of private funds for primary schools was the building levy, which accounted for one-quarter ($101 million) of all funds raised by these schools (see Table 7). The Centre/SDA levy ($96 million) and the tuition fees ($94 million) were not far behind. For secondary schools, tuition fees brought in the most revenue ($112 million), followed by boarding fees ($75 million) and the Centre/SDA levy ($71 million). 31 Primary and Secondary Education Table 7: Top Categories of School Levies and Fees by Level of Education, 2013 PRIMARY SCHOOLS SECONDARY SCHOOLS 1. Building Levy $101 million 1. Tuition Fees $112 million 2. Centre/SDA Levy $96 million 2. Boarding Fees $75 million 3. Tuition Fees $94 million 3. Centre/SDA Levy $71 million 4. General Purpose Fund $33 million 4. General Purpose Fund $25 million 5. School Activities (including 5. Building Levy $19 million Sports) Levy $21 million Total Income of Total Income of Primary Schools $406 million Secondary Schools $408 million Of which, Levies and Fees: $389 million Of which, Levies and Fees: $390 million Source: Calculations using EMIS 2014 data The level of fees and taxes collected at each school varied substantially according to parents’ ability to pay. By law, the level of fees is set by the central government, agreed between the school administration and the respective SDC, and approved by the District Education Office. In reality, the payment demanded from families each term (combining all levies and fees) is often determined by the administration’s assessment of what families can afford, along with school needs. Cases of non- payment are widespread, but a recent government decision has barred schools from denying education to children whose families cannot pay. Conversations with several heads of schools confirmed that this decision has resulted in lower levels of payment compliance. Schools’ heavy reliance on highly variable fees has created a system in which parents’ ability to pay determines the availability of infrastructure and learning materials in a particular school, perpetuating inequalities. The number of computers in a school, for example, is closely related to the total amount of taxes and fees charged by the school. This relationship is particularly strong in secondary schools where fee levels explain nearly one-third of the variation in the total number of computers (see Figure 21). Also, investments in teacher training and school infrastructure reflect the level of private financing received by each school. Figure 21: Number of Computers per 100 Students and School Fees, 2014 A. Primary B. Secondary Note: A small number of outliers (10 schools) with fees exceeding $10,000 are omitted from the analysis. Source: Calculations using EMIS 2014 data. 32 Primary and Secondary Education At present, no equalization mechanism exists to ensure an adequate and equitable distribution of resources across Zimbabwe’s schools.³⁸ Studies have shown that equalizing education financing can contribute to improvements in access to education and learning outcomes for disadvantaged populations³⁹ The introduction of an equalization mechanism to allocate financial resources more equitably across schools and school types should, therefore, be considered in Zimbabwe—potentially within the framework of the existing BSP-Z program. ANNEX 1: MoPSE BUDGET FINANCING OF EDUCATION Table 8: MoPSE Budget and Expenditure, Economic Classification, 2009-15 2009 2010 2011 2012 2013 2014 2015 % Average Change 2009 - 2009 - 2014 Budget (US$) 2014 Employment Costs 137,570,966 290,430,541 494,378,000 685,002,000 709,857,000 838,340,000 873,210,000 509% 525,929,751 Goods and Services 10,542,668 5,284,325 10,126,000 5,211,000 7,033,000 9,356,000 4,606,000 -11% 7,925,499 Maintenance 1,286,336 431,546 1,767,000 761,000 751,000 721,000 546,000 -44% 952,980 Current transfers 18,542,575 12,124,628 7,170,000 12,800,000 11,250,000 9,430,000 2,063,000 -49% 11,886,201 Programs 2,843,181 300,080 1,480,000 1,429,000 3,346,000 4,112,000 1,712,000 45% 2,251,710 Acquisition of fixed capital assets 4,145,120 1,389,604 3,743,000 4,500,000 7,900,000 7,710,000 5,535,000 86% 4,897,954 Capital transfers 2,399,880 3,192,999. - 1,500,000 10,000,000 7,000,000 2,465,000 192% 4,015,480 TOTAL 177,330,726 313,153,724 518,664,000 711,203,000 750,137,000 876,669,000 890,137,000 394% 557,859,575 Expenditure (US$) Employment Costs 166,718,167 286,917,000 486,886,459 662,409,609 733,044,336 779,488,299 - 368% 519,243,978 Goods and Services 1,462,262 5,151,338 5,851,981 2,912,613 3,404,192 2,722,201 - 86% 3,584,098 Maintenance 154,671 420,750 915,833 589,038 465,423 437,998 - 183% 497,286 Current transfers 7,387,802 12,212,818 6,179,818 1,295,999 4,110,000 2,737,900 - -63% 5,654,056 Programs 69,774 300,080 474,688 109,215 69,996 22,293 - -68% 174,341 Acquisition of fixed 154,236 1,401,682 2,975,585 3,270,633 7,047,349 57,510 - -63% 2,484,499 capital assets Capital transfers 438,561 3,292,999 - - - 2,000,000 - 356% 955,260 TOTAL 176,385,473 309,696,668 503,284,364 670,587,107 748,141,296 787,466,201 - 346% 532,593,518 ³⁸ The MoPSE reports that an equalization grant was used in the past to provide additional funding to underfunded schools, but was discontinued during the 1990s. ³⁹ See, for example, Gordon and Vegas (2004), “Education Finance Equalization, Spending, Teacher Quality and Student Outcomes: The Case of Brazil’s FUNDEF.” Available at: http://www.rti.org/files/conferences/intl-educ-finance-05/background/equity/gordon_ vegas_06_01_04.pdf. 33 Primary and Secondary Education Table 9: MoPSE Budget and Expenditure, Administrative Classification, 2009-15 2009 2010 2011 2012 2013 2014 2015 % Average Change 2009 - 2009 - 2014 Budget (US$) 2014 Administration 20,421,798 6,996,796 9,850,000 10,082,000 9,910,000 13,770,000 13,045,000 -33% 11,838,432 Education Coordination and Development 935,664 2,641,205 6,383,000 6,435,000 8,428,000 8,652,000 5,887,000 825% 5,579,145 Secondary Education 46,473,153 102,057,201 163,726,000 235,937,000 247,190,000 295,120,000 297,830,000 535% 181,750,559 Primary Education 109,500,111 201,458,521 338,705,000 458,749,000 484,609,000 559,127,000 573,375,000 411% 358,691,439 Of which: Junior Education 109,500,111 201,458,521 338,705,000 458,749,000 484,609,000 553,692,000 571,604,000 406% 357,785,605 Infant Education - - - - - 5,435,000 1,771,000 n/a n/a TOTAL 177,330,726 313,153,724 518,664,000 711,203,000 750,137,000 876,669,000 890,137,000 394% 557,859,575 Expenditure (US$) Administration 10,967,966 7,030,354 9,385,186 9,893,597 8,643,650 9,462,902 - -14% 9,230,609 Education Coordination and Development 195,695 2,740,677 3,810,728 3,044,674 8,127,563 1,565,977 - 700% 3,247,552 Secondary Education 54,954,952 102,187,980 168,928,069 226,856,698 256,359,765 273,014,882 - 397% 180,383,724 Primary Education 110,266,860 197,737,657 321,160,380 430,792,138 475,010,318 503,422,440 - 357% 339,731,632 Of which: Junior Education 110,266,860 197,737,657 321,160,380 430,792,138 475,010,318 503,022,440 - 356% 339,664,966 Infant Education - - - - - 400,000 - n/a n/a TOTAL 176,385,473 309,696,668 503,284,364 670,587,107 748,141,296 787,466,201 - 346% 532,593,518 34 Primary and Secondary Education Table 10: MoPSE Education Budget and Expenditure, Approximated Program Classification, 2009-15 ⁴⁰ 2009 2010 2011 2012 2013 2014 % Change Average 2009 - 2009 - Budget (US$) 2014 2014 Program 1: Management and Support Services 16,619,004 5,175,580 8,730,714 8,408,302 8,431,520 11,987,350 -28% 9,892,079 Program 2: Education Research and Development 769,173 2,584,332 3,071,000 2,935,000 1,424,400 2,651,040 245% 2,239,157 Program 3: Infant Education 19,200,649 36,261,426 63,281,196 85,634,566 89,808,796 108,828,532 467% 67,169,194 Program 4: Junior Education 92,872,296 166,623,106 278,042,467 376,214,652 400,350,558 454,896,301 390% 294,833,230 Program 5: Secondary Education 47,058,011 102,050,334 163,975,658 236,387,642 248,618,517 296,538,022 530% 182,438,031 Program 6: Learner Support Services 811,593 558,946 1,562,965 1,532,837 1,503,209 1,767,755 118% 1,289,551 TOTAL 177,330,726 313,253,724 518,664,000 711,203,000 750,137,000 876,669,000 394% 557,876,242 Expenditure (US$) Program 1: Management and Support Services 10,736,879 5,097,278 8,551,683 8,070,010 7,999,245 8,697,000 -19% 8,192,016 Program 2: Education Research and Development 195,503 2,683,803 893,456 973,316 1,105,193 1,565,977 701% 1,236,208 Program 3: Infant Education 20,607,684 35,575,152 60,439,874 81,649,121 90,587,951 94,168,645 357% 63,838,071 Program 4: Junior Education 89,690,666 163,595,510 263,404,850 351,602,921 389,804,917 409,323,795 356% 277,903,776 Program 5: Secondary Education 54,952,679 102,185,864 169,222,136 227,505,442 257,830,152 272,944,882 397% 180,773,526 Program 6: Learner Support Services 202,063 559,060 772,366 745,752 813,838 765,902 279% 643,164 TOTAL 176,385,473 309,696,668 503,284,364 670,587,107 748,141,296 787,466,201 346% 532,593,518 ⁴⁰ This program structure was agreed to by the MoPSE in February 2015. Note that these are estimates only, based on assumptions developed for the 2015 budget to map expenditures from individual line items to programs. 35 Primary and Secondary Education ANNEX 2: DONOR FINANCING OF EDUCATION Table 11: Funding of Major Activities Under the EDF, 2012-2014 2012 2013 2014 2015 Disbursed Disbursed Estimated Planned Overall (i) School and Systems Governance, including school grants 3,790,280 8,487,105 25,934,678 27,469,000 65,681,063 Sector Wide Planning 1,066,796 188,794 915,648 335,000 2,506,238 School Improvement Grants 1,166,652 6,562,788 23,267,393 26,379,000 57,375,833 School monitoring, supervision & support 1,556,831 1,735,524 1,751,637 755,000 5,798,992 (ii) Quality of teaching and learning 13,535,204 11,652,900 4,351,961 6,711,000 36,251,065 Teacher Quality 1,115,320 2,097,942 1,710,927 3,605,000 8,529,189 Learning Outcomes Purchase and delivery of teaching & 380,979 268,256 876,910 1,415,000 2,941,145 learning materials 2,311,597 7,543,182 947,287 650,000 11,452,066 ZELA learning assessment 1,997,398 646,906 816,837 1,041,000 4,502,142 Provision of text books 7,729,910⁴¹ 1,096,614 - - 8,826,523 (iii) Out-of-school young people 120,663 2,150,072 4,005,193 1,587,000 7,862,929 Policy sector analysis - - 30,822 157,000 187,822 Return to Mainstream Education - 1,389,560 2,531,526 1,430,000 5,351,086 Out of school technical education 120,663 760,513 1,442,845 2,560,000 4,884,020 Administrative Costs 1,091,903 1,848,887 1,899,975 1,800,000 6,640,765 TOTAL 18,538,050 24,138,964 36,191,807 37,567,000 116,435,821 Table 12: Major Contributors to the EDF, 2012-2014 2012 2013 2014 Overall (Current US$ million) Received Received Received Received DFID 14,196,469 11,048,988 10,619,220 35,868,464 EC 5,672,229 15,645,153 945,517 21,560,554 Germany 13,337,579 7,395,767 19,131,538 38,431,277 Finland 3,688,732 3,032,192 92,808 6,883,996 Norway 6,689 - - 6,689 Sweden 500,000 - 500,000 OSISA 691,686 215,320 1,869,159 2,776,165 Pooled (DFID, AusAID, Finland, New Zealand) 2,377,627 - - 2,377,627 TOTAL 39,971,011 37,837,420 32,658,242 108,404,771 ⁴¹ These textbooks were purchased in 2011. 36 Primary and Secondary Education Table 13: GPE Funding, 2014/15 Budget Received Cumulative Spending (Current US$) to Date to May 2015 Component 1 1.1 Early Reading Initiative 3,211,960 1,745,153 1.2 Performance Lag Address Programme 3,354,449 1,168,140 Component 2 2.1 Teacher Professional Standards 1,854,583 1,328,971 2.2 Teacher Development Information System 2,691,294 1,287,742 Component 3 3. Education Sector Plan 2016-2020 434,900 105,950 Management and M & E costs 923,167 716,267 Harare Overheads 315,702 176,483 TOTAL 12,786,055 6,528,705 Table 14: Unit Costs from the EDF, 2012/13-2014/15 Unit Cost Definition Baseline 2013/14 2013/14 2014/15 Text book unit cost (Primary Av cost per school textbook $4 in ETF 1 and School Text Book-core) procured including delivery $2 in EDF $0.86 $0.86 $0.86 Text book unit cost (Primary Av cost per school textbook School Text minor Languages) procured including delivery for minor Languages -Tonga and Sesotho No baseline set $11.11 Text book unit cost (Secondary Av cost per school textbook School Text Book) procured including delivery $2 $1.32 $1.32 $1.32 Text book unit cost (Secondary Av cost per school textbook School Text book- minor procured including delivery for minor Languages) Languages -Tonga and Sesotho No baseline set $3.82 Teacher Training cost Average cost per teacher for the delivery of training No baseline set $406 $708 $660 Teacher Training cost-Science Kit Av cost per teacher trained on Science kit No baseline set $230.82 $230.82 Unit cost of provision of materials Ave cost per science kit procured – Science Kits unit cost and delivered. $4,000 - $3,532.62 $3,532.62 Cost of delivering per capita grant Total expenditure per pupil per pupil per year (including overheads) under the school grants program, divided by number of children $10 $115.95 $102.70 $9.64 Cost of conducting a school visit No baseline set $22.00 Source: UNICEF 37 Zimbabwe Education Public Expenditure Review Primary and Secondary Education 52