47514 Investing in Indonesia's Education at the District Level An Analysis of Regional Public Expenditure and Financial Management THE WORLD BANK OFFICE JAKARTA Indonesia Stock Exchange Building, Tower II/12-13th Fl. Jl. Jend. Sudirman Kav. 52-53 Jakarta 12910 Tel: (6221) 5299-3000 Fax: (6221) 5299-3111 THE WORLD BANK The World Bank 1818 H Street N.W. Washington, D.C. 20433 USA Tel: (202) 458-1876 Fax: (202) 522-1557/1560 Email : feedback@worldbank.org Website : www.worldbank.org Printed in February 2009 Investing in Indonesia's Education at the District Level: An Analysis of Regional Public Expenditure and Financial Management is a product of staff of the World Bank. The findings, interpretation and conclusion expressed herein do not necessarily reflect the views of the Board of Executive Directors of the World Bank or the government they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denomination and other information shown on any map in this work do not imply any judgment on the part of the World Bank concerning the legal status of any territory or the endorsement of acceptance of such boundaries. For any questions regarding this report, please contact Mae Chu Chang (mchang@worldbank.org) and Cut Dian Rahmi Agustina (cagustina@ worldbank.org) Investing in Indonesia's Education at the District Level An Analysis of Regional Public Expenditure and Financial Management Glossary of Terms APBD Regional Government Budget Pelaksanaan Anggaran Satuan (Anggaran Pendapatan dan Kerja Perangkat Daerah) Belanja Daerah) DPRD Provincial/District House of APBN State Budget (Anggaran Representatives (regional Pendapatan dan Belanja parliament) (Dewan Negara) Perwakilan Rakyat Daerah) BAKD Local Government Financial DSSD Decentralized Social Services Administration Guidance Delivery (Bina Administrai Keunangan GDP Gross Domestic Product Daerah) GDS Governance and Bappeda Regional Development Decentralization Survey Planning Agency (Badan GER Gross Enrollment Rate Perencanaan Pembangunan GoI Government of Indonesia Daerah) GRDP Gross Regional Domestic Bappenas National Development Product Planning Agency (Badan HDI Human Development Index Perencanaan Pembangunan Inpres Presidential Grants (Instruksi Nasional) Presiden) Bawasda Regional Monitoring Agency Kabupaten District (regency) (Badan Pengawasan Daerah) Kecamatan Subdistrict BEC Basic Education Capacity Kelurahan Urban village BKD Regional Civil Service Board Kepmen Ministerial Decree (Keputusan (Badan Kepegawaian Daerah) Menteri) BKKM Scholarship for Student Keppres Presidential Decision from Poor Family (BeaSISWA (Keputusan Presiden) Keluarga Kurang Mampu) Km Kilometer BKM Special Assistance for Students Kota City (urban district) (Bantuan Khusus Murid) KUA General Budget Policy BOMM Operational Assistance for (Kebijakan Umum Anggaran) Quality Management (Bantuan LAKIP Government Agency Operasional Manajemen Performance Accountability Mutu) Report (Laporan Akuntabilitas BOS School Operational Assistance Kinerja Instansi Pemerintah) (Bantuan Operasional Sekolah) LG Local Government BPS Central Bureau of Statistics LPMP Educational Quality Guarantee (Badan Pusat Statistik) Agency (Lembaga Penjamin Bupati District Head Mutu Pendidikan) CCT Conditional Cash Transfer MBE Managing Basic Education DAK Special Allocation Fund (Dana MenPan State Ministry of State Alokasi Khusus) Administration Reform DASK Work Unit Budget Document (Kementerian Negara (Dokumen Anggaran Satuan Pendayagunaan Aparatur Kerja) Negara) DAU General Allocation Fund (Dana MoF Ministry of Finance Alokasi Umum) MoHA Ministry of Home Affairs Desa Village MoRA Ministry of Religious Affairs Dinas Local Technical Agency Office MoNE Ministry of National Education DPA-SKPD Work Unit Budget Execution MPR People's Consultative Document (Dokumen Assembly (Majilis ii Investing in Indonesia's Education at the District Level Permusyawaratan Rakyat) RKA-SKPD Education Annual Work and MSS Minimum Service Standard Budget Plan (Rencana Kerja Musrenbang Annual Development dan Anggaran Satuan Kerja Coordination Meeting Perangkat Daerah) (Musyawarah Perencanaan RKPD District Government Annual Pembangunan) Work Plan (Rencana Kerja NER Net Enrollment Rate Pemerintah Daerag) NGO Non-Governmental RPJMD Local Government Medium- Organization Term Development Plan NTB West Nusa Tenggara (Nusa (Rencana Jangka Menengan Tenggara Barat) Pemerintah Daerah) NTT East Nusa Tenggara (Nusa SD Primary School (Sekolah Dasar) Tenggara Timur) SDN State Owned Primary School O&M Operations and Maintenance (Sekolah Dasar Negeri) PAD Own-Source Revenue SIKD Regional Finance Information (Pendapatan Asli Daerah) System (Sistem Informasi PEACH Public Expenditure Analysis Keuangan Daerah) and Capacity Harmonization SISWA Education System PER Public Expenditure Review Imrpovement through Sector- Perpu Regulation in Lieu of Law Wide Appraches (Peraturan Pemerintah SKPD Regional Government's Penggati Undang-Undang) Working Unit (Satuan Kerja Perda Regional Regulation (Peraturan Pemerintah Daerah) Daerah) SMK Vocational High School PFM Public Financial Management (Sekolah Menengah Kejuruan) PKU Program Management Unit SMP Junior High School (Sekolah Podes BPS Village Potential Survey Menengah Pertama) (Potensi Desa) STR Student Teacher Ratio PPA Tentative Budget Ceilings and Sub-national Budget Consolidated Budget Priorities (Pagu dan Prioritas consisting Provincial and Anggaran) District Budgets. PREM Poverty Reduction and Susenas BPS National Socio-Economic Economic Management Survey (Survei Sosial Ekonomi RAPBS Annual Budget of Primary Nasional) School (Rencana Anggaran SWA Sector-Wide Approach dan Pendapatan Belanja Unicef United Nations Children's Fund Sekolah) Unesco United Nations Educational, Regional Budget Consolidated Budget Scientific and Cultural consisting Central Organization Government Budget WB World Bank (Deconcentrated), Provincial WDI World Development Indicators Budget, and District Budget Renja-SKPD Work Unit Annual Work Plan (Rencana Kerja Satuan Perangkat Daerah) Renstra Educational Development Strategy (Rencana Strategis) RGDP Regional Gross Domestic Product RIPS School Master Plan (REncana Induk Pengembangan Sekolah) An Analysis of Regional Public Expenditure and Financial Management iii Foreword Since the implementation of decentralization in 2001, district governments have been getting increased responsibilities to provide education services to the citizen. Basic and secondary education management authority has been fully transferred from the central to the provincial and district government. District education expenditure has grown rapidly both in terms of level and as a share of national education expenditure. The amount of district education expenditure has increased from Rp. 26 trillion in 2001 to 52 trillion in 2006 and they constitute 50 percent of the total national education public expenditures in 2006. Moreover, the education sector at the district level are also getting more priorities with the enactment of Law No. 20/2003 on the National Education System which requires central and sub-national governments to allocate a minimum of 20 percent of their budgets to the education sector. In order for this significant allocation of education spending to translate into improved education outcomes, it is important to understand the spending pattern of the district governments, so that a proper assessment on the effectiveness and efficiency of the district government spending can be carried out. This review of district education expenditures is one of the outputs of a larger set of analytical and design activities that will prepare the ground for System Improvement through Sector Wide Approaches (SISWA) Program in basic education. We hope that this report can benefit the SISWA team, governments, and other stakeholders through its analysis on the current trends in education investments and outputs and can serve as a model for the broader program of local education expenditure analysis which hopefully will be carried out by the district government themselves to help them do better planning and budgeting and to inform future spending decisions. Mae Chu Chang Wolfgang Fengler Human Development Coordinator Senior Economist The World Bank Indonesia The World Bank Indonesia iv Investing in Indonesia's Education at the District Level Acknowledgments This report was prepared by a core team led by Cut Dian Agustina (EASPR), together with Elif Yavuz (EASPR) and Ahmad Zaki Fahmi (EASPR). Excellent research and data analysis assistances was provided by Sukmawah Yuningsih and Adrianus Hendrawan (EASPR). Important substantive comments were also received from Andrew B. Ragatz (EASHD), Vincente Paqueo (EASHD) and Eduardo Velez Bustillo (Sector Manager, EASHD). Valuable inputs were obtained from the Government of Indonesia, especially Prof. Dr. Suyanto and his team from the Directorate General of Primary and Secondary Education (Ministry of Education). We are also grateful for the assistances provided during the field visit from the Bappeda and Provincial Education Offices in North Sumatra, Central Java, North Sulawesi, and East Nusa Tenggara and District Education Offices in Asahan, Kota Binjai, Kota Magelang, Wonosobo, Kota Menado, Minahasa, Belu and Timor Tengah Selatan. A larger group within the World Bank contributed valuable inputs to this report, for which the core team expresses their appreciation: Ratna Kesuma, Prima Setiawan, Rosfita Roesli, Dandan Chen (EASHD), Meltem Aran, Sukarno Wirokartono, Bastian Zaini, Ahya Ihsan, Adrianus Hendrawan, Bambang Suharnoko (EASPR), and Ajay Tandon (HDNHE). Special thanks also go to Peter Milne for editing and Arsianti for assisting with formatting and production. Overall guidance was provided by Wolfgang Fengler (Senior Economist) and Mae Chu Chang (Human Development Coordinator). An Analysis of Regional Public Expenditure and Financial Management v Table of Contents Glossary of Terms ii Foreword iv Acknowledgments v Table of Contents vi Summary 9 Chapter 1. Introduction 11 1.1. The Role of Districts in Investing in Education 11 1.2. Data Collection, District Selection, and Capacity Building for District Expenditure Analysis 12 1.3. Outline of the Review 12 Chapter 2. The Legal Framework and Flow of Funds for the Education Sector 15 2.1 The Legal Framework 15 2.2. The Flow of Funds for the Education Sector 18 Chapter 3. Education Expenditures at the Local Level 21 3.1. The Big Picture: Public Education Spending in Indonesia 21 3.2. Education Spending and the Importance of the Districts as a Spending Unit 22 3.3. District Expenditure Patterns and Trends 24 3.4. Economic and Programs Composition 26 3.5. Household Out-of-Pocket Spending 31 3.6. Spending at the School Level 32 Chapter 4. District Education System and Outputs 35 4.1. Education Infrastructure and Facilities 35 4.2. Human Resources in Education: Teachers and Students 37 4.3. Education Expenditure, Outputs, and Equity 39 4.4. Efficiency at the District Level: Identifying Best Practice 45 Chapter 5. District Planning, Budgeting, and Monitoring for the Education Sector 47 5.1. Planning and Budgeting 47 5.2. Monitoring and Evaluation 50 5.3. Financial Management Assessment in Papua 51 Chapter 6. Key Findings and Policy Options Regarding the SWA 53 Annexes 57 References 61 vi Investing in Indonesia's Education at the District Level Tables Table 1.1. District covered by this study 13 Table 2.1. Education spending as percentage of local government spending, 2006 16 Table 2.2. Summary of intergovernmental financing arrangement based on PP No. 38/2007 17 Table 3.1. National public expenditure on education (central + province + district), 2001-08 22 Table 3.2. Nominal education expenditures by level of government, 2001­06 22 Table 3.3. Share of development and routine expenditures by level of government, 2001-04 23 Table 3.4. Routine expenditure distribution by level of sub-national government, 2002-06 24 Table 3.5. Education expenditure as a share of total expenditure in visited districts, 2001-06 25 Table 3.6. Composition of routine and development education expenditure, 2001-05 27 Table 3.7. Budget planned vs. realized in the visited districts, 2005 28 Table 3.8. Public and apparatus spending on education 30 Table 3.9. BOS fund allocation based on school level and type in the visited districts, 2005 33 Table 4.1. Average service areas for schools in the visited districts, 2006 36 Table 4.2. Schools per 1,000 school-aged children in the visited districts, 2006 36 Table 4.3. Student teacher ratios (STR) in the visited districts, 2006 37 Table 4.4. Some indicators of school availability 43 Table 4.5. District expenditures per poverty quintile, 2005 43 Table 5.1. School monitoring mechanism 51 Table 1. Education spending by central government (decon), and local governments, and household out of pocket, 2005 57 Table 2. Local government education expenditure, 2001-06 58 Table 3. Routine and development expenditure on education, 2004-06 58 Table 4. Economic classification of education routine spending, 2006 59 Table 5. Public and apparatus spending on education 2005 and 2006 59 Table 6. NER, literacy rates, and mean years of schooling, 2005 60 Table 7. Number of students, classroom condition, and number of schools, 2006 60 Figures Figure 2.1. Flow of funds of education expenditures 19 Figure 3.1. Trends in national education expenditures, 2001­08 22 Figure 3.2. Education spending by economic classification, level of government and routine district spending composition 23 Figure 3.3. Source of district education expenditures, 2005 24 Figure 3.4. District education expenditures by source of fund in visited districts, 2005 24 Figure 3.5. Education expenditure and percentage of total expenditure, 2001-06 25 Figure 3.6. Education expenditures in the visited districts, 2001-06 25 An Analysis of Regional Public Expenditure and Financial Management vii Figure 3.7. Per capita education expenditure in the 10 visited districts, 2006 26 Figure 3.8. Economic composition of routine expenditures in visited districts, 2006 27 Figure 3.9. Program composition of the visited districts, 2006 29 Figure 3.10. The breakdown of public and apparatus spending in 10 visited districts, 2006 31 Figure 3.11. Household out-of-pocket expenditure in the visited districts, 2004-06 32 Figure 4.1. Primary schools per 1,000 primary school-aged children by sub-districts in Kab, Belu and Kab. Wonosobo, 2006 36 Figure 4.2. STR for junior and senior secondary levels by sub-district in Kab. Belu, 2006 37 Figure 4.3. Over- and under-supply of elementary teachers by district and sub-district, 2006 38 Figure 4.4. Teacher qualifications in the visited districts, 2006 39 Figure 4.5. Primary school net enrollment rates of the visited districts, 2002-06 40 Figure 4.6. Junior secondary net enrollment rates of the visited districts, 2002-06 40 Figure 4.7. Senior secondary net enrollment rates of the visited districts, 2002-05 40 Figure 4.8. Primary school drop-out rates of the visited districts, 2002-06 41 Figure 4.9. Mean years of schooling in the visited districts, 2001-05 41 Figure 4.10. Literacy rates in the visited districts, 2001-06 41 Figure 4.11. Junior secondary school net enrollment rates by income quintile, 2006 42 Figure 4.12. Spending per capita, net enrollment rates at junior and senior secondary levels 42 Figure 4.13. Best practice frontier of education sector performance at the district level 45 Figure 4.14. Distribution of output index by input quintile 45 Figure 5.1 The normative framework for annual budgeting preparation process 49 Figure 5.2. Overall PFM scores for all assessed sub-national governments 52 Figure 5.3. Average scores for each strategic area 52 Figure 1. Total district revenue and education spending, 2005 57 Boxes Box 2.1. The legal background of Indonesia's "20 percent rule" 16 Box 3.1. What drives the education expenditure pattern? 28 Box 4.1. Minimum service standards in the education sector 44 Box 5.1. Lost in translation: How medium-term plans are translated intothe education sector plan 48 Box 5.2. School financial management and formula funding in Kota Magelang 50 viii Investing in Indonesia's Education at the District Level Summary Highlights in education development the primary and junior secondary levels are more · In2006,56percentofeducationexpenditure significant in kabupaten (district) than kota (city) was spent at the sub-national level. areas. This situation suggests a higher transition District governments are the main spenders, rate from primary to secondary schools in urban accounting for 51 percent of total spending, than rural districts. while provincial governments account for just · There are variations in access to schools over 5 percent. These shares of total education and in teacher supply, but almost all of spending demonstrate the trend in education the districts visited score higher than the service delivery, with district government shares national average. Access to schools for being relatively high compared with the central districts nationally and in most of the visited government. districts is good at all school levels, although · Districteducationexpenditurehasincreased greater variation is found among sub-districts. since decentralization. However, the budget In terms of student teacher ratios (STRs), based shareofdistricteducationspendinghasbeen on WDI most of visited districts have below decreasing. This decreasing trend, particularly average primary STRs for low- and middle- since 2005, may have been influenced by BOS income countries, indicating an adequate supply (school operational assistance) transfers from of teachers. the central government. · Aggregate spending at the sub-national Key issues level in 2006, as well as in most of the · Although districts spend the majority of districts visited, achieved the "20 percent the total education budget, this spending rule", if salaries are included in the is mostly on non-discretionary routine estimation. The Constitutional Court ruled in expenditure. At the sub-national level more 2007 that the "20 percent rule" would include than 90 percent of routine expenditure is all education spending (including salaries) at all allocated to personnel spending. levels of government. · Secondary level enrollment is still · Net enrollment rates (NERs) at the primary problematic for many districts and across school level in most of the visited districts income groups. The bridge between primary are close to universal, except in some remote and secondary education is important if the districts in Papua. Districts with high NERs at transition rate is to be increased. For people the primary level also tend to have high NERs at in lower income quintiles, access to education the junior secondary level. Differences in NERs at is still a problem and here transportation and An Analysis of Regional Public Expenditure and Financial Management 9 Summary education fees continue to play a major role. In or disincentives in order to improve the budget certain districts significant challenges remain in composition. One of the mechanisms to achieve increasing literacy rates. this is by using the budget composition of local · The condition of education infrastructure in governments as an indicator for deconcentrated many districts in Indonesia is still poor. This spending targeting. poor condition is all the more worrying because · Local governments need to re-evaluate only less than 1 percent of routine spending is teacher allocations and employment, allocated towards maintenance and operational particularly with regard to uneven teacher spending. distribution. Local governments should also · There is still an unequal distribution of link their teacher allocation planning with schools and teachers, particularly among the incentives program as required by Law sub-districts. A problem of over- and under- No. 14/2005 on Teachers and Lecturers. Re- supply of teachers, as well as insufficient evaluation of staffing needs to be supported numbers of teachers who meet the minimum by a clear staffing framework since teachers qualification standards, is observed in most are members of the national civil service. The districts. In primary schools, most teachers have government needs to provide not only for staff only graduated from senior high schools, despite transfers between schools within a district, the legal requirement that teachers have at least but also staff transfers between districts and a bachelor degree. between provinces. · An incoherent planning and budgeting · Invest in teaching quality through teacher process is still observed in many districts certification together with adequate both within the sector and between local mechanisms and performance control. Poor governments and Education Dinas. This is education quality is an issue for both rural and probably caused not only by the low capacity of urban areas. Certification of teachers can help local governments, nor is it entirely explained by to enhance learning outcomes if adequate the lack of a regulatory framework, but largely mechanisms and performance controls, such due the absence of any planning culture and the as teacher attendance and teaching quality lack of performance accountability. measurements, are implemented. · A nationwide system of performance measurement is required to ensure that Recommendations local governments are accountable in · District governments need to prioritize developing education outcomes, as well secondary education. This could be achieved as synchronizing budgeting and planning by increasing budget allocations for programs in their regions. Such a system would act as that provide scholarships and encourage drop- a balance between the full autonomy that is outs to return to school. Improvement in the provided to districts in spending their funds and literacy rate will require not only an expansion of the outcomes they achieve with those resources. the school system to increase NERs but also an An appropriate performance accountability effort to reach the non-literate population who system can be instituted by using the minimum are now beyond school age. Local governments service standards (MSS) on a preliminary basis. should include specific programs in their education budget priorities that attempt to combat illiteracy and provide a second chance to those who have been missed out on the opportunity to enroll in formal education. · Districts need to improve their capacity to assess the efficiency of their spending and identify key issues within the education sector. Technical assistance might be needed to this end. In addition, the central government can also monitor the composition of local government expenditures and provide incentives 10 Investing in Indonesia's Education at the District Level Chapter 1. Introduction 1.1. The Role of Districts in this information to inform future spending decisions. With respect to the issue of governance, and in light Investing in Education of increased expenditure allocations for education to the sub-national level, the GoI hopes to strengthen the The Government of Indonesia (GoI) has made capacity of districts such that they use their resources investing in the education sector a major priority well. An important priority here is the strengthening and, in recent years, has been allocating an of good planning, budgeting, procurement, financial increasing percentage of its budget towards management and accountability practices. education. National public spending on education rose from 2.8 percent in 2001 to 3.1 percent in The forthcoming Education System Improvement 2006 relative to GDP. The fuel subsidy reductions throughSectorWideApproaches(SISWA)Program have allowed the GoI to reallocate public resources is a large budgetary support program that aims to education spending, for example through the to tackle challenges related to governance, school operational assistance (BOS) and conditional management and resource allocations. The cash transfers (CCT). Remarkably, education has now SISWA program's main development objective is to displaced interest payments on debt as the largest increase basic education enrollment, completion and item in the GoI budget. Public spending on education quality by providing support for the implementation is estimated to increase further to 3.0 percent in 2008 of the educational development strategies (Renstra) according to budget data. not just of the Ministry of National Education (MoNE) and the Ministry of Religious Affairs (MoRA), but also Since 2001, district governments have been of sub-national governments. In order to ensure responsiblefortheprovisionofeducationservices commitment and ownership of the SISWA program, and, as a consequence, now spend the majority a large component will provide grants to district of the total education budget. In 2006, about 56 education offices (education Dinas) based on their percent of the GoI education budget was executed performance on basic education outcome indicators. at the sub-national level. District governments are the main spenders, accounting for about 51 percent The review of district education expenditure of total spending, with provinces accounting for the and financial management is an output from other 5 percent. the Basic Education Capacity Program, which is aimed at preparing for the launch of the In order for public spending to translate into Government's larger Siswa program under the improved performance outcomes, it is crucial that new Renstra 2010-14 for basic education. Hence, district governments have sufficient financial this paper is one of the outputs of a larger set of management capacity. In addition, it is imperative analytical and design activities that will prepare the for districts to be able to evaluate the effectiveness ground for a sector wide approach (SWA) program in and efficiency of their spending and, subsequently, use basic education. An Analysis of Regional Public Expenditure and Financial Management 11 Chapter 1 Introduction 1.2. Data Collection, District In addition, the project team also visited the provincial education offices to ensure that any comprehensive Selection, and Capacity Building district data, compiled or summarized at the province for District Expenditure Analysis level, would be taken into account when performing the analysis for the various districts. Furthermore, cross-checking at the provincial level allowed for the This activity is primarily focused on district verification not only of completeness and transparency educationexpenditureandfinancialmanagement of the district data, but also for an assessment of analysis, and was implemented jointly by the the data transfer mechanisms in the various regions World Bank Jakarta Education Team and the and provide an opportunity to document potential Poverty Reduction and Economic Management shortcomings in information systems within the (PREM) Team. The methodology applied in this regional education system. analysis has been developed by the PREM team and is known as the PEACH approach, which stands for The data were collected by a team of research Public Economic Analysis and Capacity Harmonization. analysts that visited a large number of The PEACH has been used in a number of provinces stakeholders ranging from district education Gorontalo, Aceh, Nias and Papua and is a offices (Dinas) to schools and local government programmaticapproachtoundertakingcomprehensive planning units (Bappeda). Most districts were visited expenditure analysis together with local stakeholders for a period of three days and the collected data were at the sub-national level. In this analysis the approach supplemented by additional data collection at the was used to analyze specifically the expenditures for corresponding provincial Dinas office. Generally, the the education sector at the district level. provincial Dinas office staff facilitated the meetings with the district teams. Regarding analysis at the Since it requires significant effort to carry out school level, the team complemented its field research rigorous expenditure reviews at the district with findings coming from the GDS2 survey for those level as already demonstrated by the various districts (out of the 10 selected) that were included in PEACH initiatives completed so far only 10 the survey, namely: Kab. Asahan, Kota Manado, Kab. districts were covered in this initial effort. This Belu, Kab. Timtengsel, and Kab. Jayawijaya. district sectoral public expenditure review (PER), however, serves as a model for the broader program 1.3. Outline of the Review of local PERs expected to be carried out in relation to the national SISWA program. This PER comprises six chapters. Following this In terms of the selection of districts for this introductory chapter, the second chapter covers the expenditure analysis exercise, the activity team legal framework for financing the education sector, as aimed to align the selection process with the well as the flow of funds spent on education services. district targeting approach used for the SISWA. The third chapter is the core of the paper and Since the SISWA aims to provide budgetary support in analyzes trends and patterns in public expenditures. an equitable manner to districts with a wide range of The economic and programmatic composition of capacity across Indonesia, the team decided to target expenditures at the district level are also discussed here. districts with a relatively good reputation in terms of Furthermore, household expenditures on education, budgeting, planning and expenditure management, as well as spending at the school level, particularly as well as districts that are known to be `poor' and through the new BOS program, are analyzed. The lacking in capacity in the human development fourth chapter focuses on the educational system and sector.1 its outcomes at the district level. System inputs, such as education infrastructure and human resources, To ensure regional equity, the team decided to are examined and (distributional) outputs in terms focus on five provinces, located in five different of enrollment rates, literacy rates, etc, are discussed. regions/islands of Indonesia and, within those The fifth chapter reviews the planning and budgeting provinces, target rural as well as urban districts. mechanisms found at the district level. Finally, the sixth chapter of this report contains policy options and the main messages of particular importance to 1 Here `poor' are understood as both the fiscally poor, as well as the SISWA program. the economically poor. 12 Investing in Indonesia's Education at the District Level Chapter 1 Introduction Table 1.1. District covered by this study Characteristics of Regions Fiscal Resources GRDP Poverty Urban/Rural Province North Sumatra Kab Asahan Low Medium Medium Rural Kota Binjai Medium High Low Urban Province Central Java Kab Wonosobo Low Low High Rural Kota Magelang High High Medium Urban Province North Sulawesi Kab Minahasa High Medium Low Rural Kota Manado Medium High Low Urban Province NTT Kab Timur Tengah Selatan Medium Low High Rural Kab Belu Medium Low Medium Rural Province Papua Kab Jayawijaya Low Low High Rural Kab Jayapura High Medium High Rural Source: World Bank staff calculations based on BPS/MoF data (Susenas, SIKD, GRDP). Detailed Description of Main Deliverables/Outputs from the Activity A number of outputs were delivered under this activity, of which this district expenditure review is one. The deliverables are summarized below. District Database for Selected Districts: The district database contains the budget data compiled in the field visits, which are mostly local expenditure and budget data, as well as qualitative information regarding the planning and budgeting processes in the various institutions. Further, data concerning outputs for main education performance indicators have been included and extracted mostly from existing data sources, such as Susenas, Podes and SIKD. In order to allow for trend analysis and to enlarge the likelihood of making viable cross-district comparisons, both expenditures and allocations for as many recent years as available were collected (2001-06/07). The format of the database is in Microsoft Excel, although some parts are complemented with STATA datasets. District Education PER The district education PER provides inputs into the SISWA program and thereby aims to improve project design and single out district needs in terms of public financial management and expenditure analysis support. The PER is not only focused on the description of trends and patterns in district expenditures, but also reviews the legal framework with regard to education spending and the planning and budgeting process. District Level Education Expenditure Analysis Toolkit One sub-component of this PER is a toolkit for districts on how to undertake expenditure analysis in the education sector. The toolkit is based on the experience and methodology coming out of the World Bank's PEACH work and has been specifically adapted here for the education sector. This output is aimed at the strengthening district capacity and its dissemination is planned under the second component of the SISWA and the Basic Education Project. An Analysis of Regional Public Expenditure and Financial Management 13 Chapter 1 Introduction 14 Investing in Indonesia's Education at the District Level Chapter 2. The Legal Framework and Flow of Funds for the Education Sector In this chapter the legal framework and the flow of on salaries from this benchmark.2 The central funds for the education sector are discussed. The government is responsible for allocating a share of system described here is largely applicable at the districts' budgets towards the salaries of teachers who national level, as the majority of the regulations are are appointed by the government. Funds for teacher effective across the entire country, although regional salaries are transferred by the central government regulations were in place in some of the districts visited. to district government budgets through general The flow of funds structure also allows for national allocation fund (DAU) transfers. The allocation of comparisons and hence can feed into the knowledge funds to education units (education service group that and know-how needed for the preparation of the provides formal, non-formal and informal education SISWA district performance grant component. in each level and type of education) by central and sub-national governments are also transferred in the form of grants. 2.1 The Legal Framework Aggregate spending at the sub-national level in 2006, as well as in the 10 districts visited, was The responsibilities of sub-national governments still far from meeting the 20 percent mandate if with regard to budget allocations for the salaries are excluded in the estimation. None of education sector have been defined by a variety the 10 districts has been able to meet the mandate. of laws and regulations. Among these laws and In fact, all 10 districts have education expenditures regulations are most importantly: Law No. 20/2003 (excluding salary expenditures) that are less than 10 on the National Education System, Law No. 14/2005 percent of their local budgets. Kab. Timtengsel and on Teachers and Lecturers, and the latest Government Kab. Wonosobo have the highest shares, with 8.9 and Regulation No. 38/2007 on intergovernmental 6.6 percent of total expenditure (excluding salaries), relations and sub-national roles and functions. The respectively, allocated to the education sector. The contents of these legal documents are discussed picture is, however, changed significantly when below. salaries are included in the estimation. The aggregated districts at the national level and most of the visited 2.1.1. Education Financing Responsibilities at districts have reached or surpassed the 20 percent the Central and Sub-national Levels mandate, except for the ones in Papua. The overall share of education spending to total expenditure LawNo.20/2003ontheNationalEducationSystem requirescentralandsub-nationalgovernmentsto allocate a minimum of 20 percent of their budgets 2 In 2007, the Constitutional Court specified that teacher salaries would be included in the 20 percent budget allocation to to the education sector, excluding spending education. This paper, however, only includes brief discussion on this issue as the revision of Law 20/2003 is still unavailable. An Analysis of Regional Public Expenditure and Financial Management 15 Chapter 2 The Legal Framework and Flow of Funds for the Education Sector including teachers' salaries at the national level Compared with other visited districts, Kab. accounts for 28.3 percent, and only 6.3 percent if Jayawijaya and Kab. Jayapura in Papua have the salaries are excluded. lowest shares of overall education spending. One of the reasons for the lower spending of education in Box 2.1: The legal background of Indonesia's these districts is the relatively low education spending "20 percent rule" per capita student, as well as fewer schools and teachers. · 1945: Indonesia Constitution stipulates in Article 31: (1) "Every citizen has the right to education" The spending structure within the education and; (2) "The government shall establish and sector largely explains the difficulty in meeting conduct a national educational system which shall be regulated by law." the target stipulated by Law No. 20/2003, rather · 2002: Nearly 60 years later, this Constitution article than funding constraints. The significance of the was amended to specify: "The state should allocate salary component in sub-national expenditures was a minimum of 20 percent from the APBN budget striking in the districts visited, leaving only small to education expenditures, to respond to national shares available for other spending items. A similar education needs." pattern is observed when analyzing aggregate sub- · 2003: Later, Law No. 20/2003 on the National national expenditures for the education sector, as on Education System (Part 4, Art. 49) again redefines average 96 percent of district routine expenditures the 2002 benchmark. The 2003 law narrows the go towards salaries or incentives. Lack of resources is range of spending items that count toward the 20 not the reason behind the low non-salary education percent target by excluding salaries. As stated: spending, because districts have received massive "Education funds, excluding salary of educators increases in transfers (DAU), particularly in 2006. The and service education expenditure, are allocated at increase in transfers has significantly boosted district a minimum 20 percent of the APBN and a minimum revenues, thus topping up the funds still further fails of the APBD." to address the main structural problem. · 2007: The Constitutional Court Decree No. 24/ PUU-V/2007 specified that teacher salaries will be included in the 20 percent budget allocation to education. · 2008: The Constitutional Court Decree No. 13/PUU- VI/2008 provides an ultimatum for the government to comply with the 20 percent share. Source: Investing in Indonesia's Education, World Bank (2007a) and Constitutional Court decree. Table 2.1. Education spending as percentage of local government spending, 2006 Education spending share Education spending share Districts (incl. salary) (excl. salary) Kab. Asahan ­ North Sumatra 39.9 4.6 Kota Binjai ­ North Sumatra 26.3 3.7 Kab. Wonosobo ­ Central Java 37.0 6.6 Kota Magelang ­ Central Java 30.6 2.2 Kab. Minahasa ­ North Sulawesi 35.1 5.8 Kota Manado ­ North Sulawesi 30.2 2.4 Kab. Timur Tengah Selatan ­ NTT 35.7 8.9 Kab. Belu ­ NTT 32.2 6.4 Kab. Jayawijaya ­ Papua 14.0 3.8 Kab. Jayapura ­ Papua 17.8 5.0 Average of 10 districts 29.0 5.1 Aggregate districts (national) 28.3 6.3 Source: World Bank staff calculations based on SIKD/MoF and districts' APBD data. 16 Investing in Indonesia's Education at the District Level Chapter 2 The Legal Framework and Flow of Funds for the Education Sector 2.1.2 Intergovernmental Financing support for secondary and vocational education, and Arrangements special-needs education. Provinces can also provide additional support or subsidies for early childhood education, primary and non-formal education, as well The intergovernmental financing arrangements as higher education. Finally, district governments are for the education sector have been partially mainly responsible for providing resources on early defined in Government Regulation (PP) No. childhood education, and primary and non-formal 38/2007. This new government regulation specifies education. Districts do not have any additional the assignment of roles and responsibilities of each support responsibilities. For a summary of these government level for all the decentralized sectors. arrangements, see Table 2.2. The regulation has addressed some issues on intergovernmental roles and responsibilities which, In terms of additional support or subsidies according to many decentralization stakeholders, had from the central government to the districts, previously not been clearly defined. However, further these are channeled mostly through so called clarifications on issues such as sectoral financing deconcentrated funds, while supplements from are still needed, particularly for those sectors that the provinces mostly stem from their own-source were decentralized to a considerable extent, such as revenues. Among programs that are subsidized by education and health. The assignments for central, the central and provincial governments are the school provincial and district governments within the rehabilitation program, teacher training, subsidies for education sector can be divided into five sub-sections poor families (BKKM and BKM), youth reintegration related to: (i) policy; (ii) planning and financing; (iii) programs (reintegrating school dropouts), teacher curricula; (iv) infrastructure and facilities; and (v) incentives, etc. In addition to deconcentrated funds education personnel. that are channeled through provinces, the central government has also provided direct support to The planning and financing sub-section of primary and junior high schools through school the new regulation classifies the financing operational assistance (BOS), although primary responsibilities of each level of government by schools are supposed to be a district government levels of education and program. The central responsibility. As for school rehabilitation programs, government is responsible for the provision of particularly for primary schools, these are often overall guidance on financial support for each level jointly performed by central, provincial and districts of education and program. Furthermore, the central governments under a specific MoU agreement with governmentisresponsiblefortheprovisionofresources an agreed contribution from each party. Such co- for the higher education level and cross subsidies financing efforts occur in most districts and hence for early childhood education, primary, secondary, the central and provincial governments still play and non-formal education, as well as for special an important role in rehabilitation efforts, thereby education services. As for provincial government, challenging traditional decentralized responsibilities. the primary responsibilities include providing financial Table 2.2. Summary of intergovernmental financing arrangement based on PP No. 38/2007 Central Province District Primary financing - Higher education - Secondary education - Early childhood responsibility - Vocational education education - Special education service - Primary education - Non-formal education Additional support / - Early childhood education - Early childhood subsidy - Primary education education - Secondary education - Primary education - Non-formal education - Non-formal education - Specific education - Higher education - Special education service Source: Government Regulation No. 38/2007. An Analysis of Regional Public Expenditure and Financial Management 17 Chapter 2 The Legal Framework and Flow of Funds for the Education Sector 2.1.3 Teacher Incentives The central government's direct transfer of funds to the school level is implemented on a per student basis. The types of transfer include the Teacher incentives are regulated by Law No. recent BOS program, targeted to elementary and junior 14/2005 on Teachers and Lecturers, which covers secondary schools, and the operational assistance for professional incentives, functional incentives, quality management (BOMM, or Bantuan Operasional special incentives, and additional benefits. Manajemen Mutu) program to support vocational and Professional incentives are provided to those teachers senior secondary schools. At present, these programs who are certified through a competency examination constitute the majority of funds at the school level conducted by a nationally authorized institution. and are hence of high importance. Functional incentives are given to civil servant teachers, as well as to non-civil servant teachers, at both public Regardingthedeconcentratedfunds,theprojects and private schools. Special incentives are available to cover a variety of activities and vary in their teachers who teach in specific areas, such as remote disbursement mechanisms. Deconcentrated funds areas, districts in border areas, disaster areas or other are often used for projects such as school or classroom emergency situations. In addition to these incentives, reconstruction and school quality improvements. teachers can also receive supplementary benefits in Depending on the particular design of a project, the the form of insurance, awards, scholarships, etc. funds may be channeled directly into school accounts, or through provincial Dinas accounts. The special Central and sub-national governments are all characteristics of deconcentrated projects relate to legally responsible for providing resources from the role of the province in the flow of funds, as the their budgets to support teacher incentives. The provincial Dinas are responsible and accountable professional and special incentives are to constitute for the supervision and management of those the same amount as the current base salaries. The projects within their jurisdictions. In this case, the functional incentive for civil servant teachers depends implementing units (school and/or district Dinas) are on their rank in the civil service. Functional incentives required to submit progress and completion reports for non-civil servant teachers, however, are subsidized; to the provincial Dinas. the level depends on the financial capacity of the government. This incentive program is relatively new Another financing mechanism is through and implemented only in fiscal year 2007. APBN projects, which are funded, as well as executed, directly by the center. In these types of projects, the central government directly provides 2.2. The Flow of Funds for the goods and/or services to schools. An example is Education Sector the national multimedia project, whereby schools receive multimedia equipment (TVs, computers, etc) distributed directly by central government agencies. Funding for the education sector at the district level comes through multiple channels, most In regions where representative offices exist, notably the central and sub-national (province central funds are sometimes still channeled and district) governments (Figure 2.1). Central to these offices from the center. An example government funds are channeled through several of this type of office is the Educational Quality mechanisms that include: direct transfers to schools; Guarantee Agency (LPMP, or Lembaga Penjamin deconcentrated funds/projects; APBN projects; and Mutu Pendidikan). Among other tasks, the LPMPs are central representative office projects. responsible for providing training to teachers in the provinces where such agencies are based. 18 Investing in Indonesia's Education at the District Level Chapter 2 The Legal Framework and Flow of Funds for the Education Sector Figure 2.1. Flow of funds of education expenditures LEVEL BUDGET NATIONAL MINISTRY OF MINISTRY OF EDUCATION FINANCE APBN Tugas Central DAU /SDA DAK Dekon BOS Pembantuan Functions Provincial PROVINCIAL Budget : Provincial Central Own DAU /SDA Education Office Government Source Agencies in the Revenue PAD Regions District Budget : DISTRICT DAU/ SDA District Education Own PAD Office Source DAK Revenue Schools Fund originated from MoF Fund originated from Provincial Budget Fund originated from MoNE Fund originated from District Budget sectoral budget Source: Interviews with district and provincial education Dinas officials. In addition to the central government, provincial Resources allocated to the education sector by governments also provide funds to finance provincial governments are not always under the activities executed at the district/school level. management of the provincial education Dinas. Provinces often pay the salaries of contract teachers, In the case of the school rehabilitation program and provide incentives to teachers in remote areas, and teacher incentives in remote areas in North Sumatra finance school rehabilitation activities. Recently, the and NTT, the funds are classified as social assistance MoNE has signed a number of MoUs with several expenditure and the transfers go directly into the provincial and district governments regarding the district government account, bypassing the provincial co-financing of school rehabilitation. The financing Dinas. This adds to the complexity of sub-national share for each level of government concerned in this expenditure analysis, particularly with regard to fund arrangement is different for each province depending flows coming from the provincial level. on the fiscal capacity of the district, with some districts requiring more assistance than others. In the At the district level, the processes of allocating five provinces visited, the contributions of provincial operational funds from district governments to governments varied from 15 percent in North Sulawesi schools are not uniform across districts. In most to about 30 percent in Central Java. districts, the education Dinas is responsible for the An Analysis of Regional Public Expenditure and Financial Management 19 Chapter 2 The Legal Framework and Flow of Funds for the Education Sector planning, budgeting, and execution of expenditures related to coordination activities and education management. However, there are a number of flows that are allocated directly to schools for some districts, such as Kota Manado, and provide resources based on a per student formula to both elementary, secondary and high school levels. In other districts, such as Kab. Minahasa, schools receive fixed allocations on a quarterly basis. Nevertheless, district Dinas share a common process in terms of coordinating the budgeting process for primary schools. All primary schools were required to submit their annual budgets (RAPBS) to Dinas in all districts examined. For junior secondary (SMP) and senior secondary (SMA/SMK) schools, however, the budgeting process varies. In some districts, such as Kota Binjai, schools have their own budget documents and there is relatively little interference from the Dinas before they submit their annual budget proposals to the district planning and budgeting committee. In other districts, such as Kab. Minahasa and Kab. Belu, the Dinas still coordinate and supervise the planning and budgeting of every individual school, including district junior and senior secondary schools. Moreover, in all districts visited the education Dinas were in charge of the planning, budgeting, and budget execution processes for capital, and operations and maintenance funds. 20 Investing in Indonesia's Education at the District Level Chapter 3. Education Expenditures at the Local Level In this chapter, mainly education expenditures at the decrease during the crisis and a dip in 2004. The local level are discussed. However, in order to set the decrease in spending in 2004 was jointly caused by stage for this discussion, the chapter also provides an low budget execution and a crowding-out effect in analysis of the role of the districts within the national most social sectors due to increasing fuel subsidies. public expenditure system for education (Section 3.1 Education expenditures peaked in 2003, when they and 3.2). In addition to this `big picture' overview, accounted for about 14 percent of overall national private (i.e. household) expenditures and expenditures expenditures (Table 3.1). Education expenditures at the school level are analyzed, as these form the as a share of GDP in 2005 relative to 2004 slightly most significant additions towards local education declined, at about 2.6 percent, as did the ratio of spending. overall national expenditures to GDP. In the absolute figures, education expenditures rebounded by 8 percent in 2005 and showed an even higher increase 3.1. The Big Picture: Public of about 24 percent in 2006. Budget figures for 2007 indicate another 9 percent increase, with the share of Education Spending in Indonesia national education spending at 15.4 percent of total national expenditures. Since the mid-1990s, Indonesia has seen an upward trend in government expenditure on education.3 The two exceptions were a temporary 3 In this chapter education spending for central government is defined following the sectoral budget classification. From Sector 11: Education, National Culture, Belief in God Almighty, Youth and Sports Sector, the sub-sectors 11.1 Education and 11.2 Official and Informal Education are included in the analysis, which together account for 98 percent of the sector total. An Analysis of Regional Public Expenditure and Financial Management 21 Chapter 3 Education Expenditures at the Local Level Table 3.1. National public expenditure on education (central + province + district), 2001-08 Rp trillion Education 2001 2002 2003 2004 2005 2006 2007* 2008** Nominal National Education Expenditures 40.1 49.8 63.2 61.4 73.2 102.5 118.5 135.7 National Education Expenditures (2006 63.4 70.5 83.7 76.7 82.8 102.5 111.3 116.4 prices) Growth Real National Education - 11.1 18.7 (8.4) 8.0 23.9 8.6 4.5 Expenditures Education Exp. (% Total National Exp.) 11.0 14.2 14.7 13.0 13.9 15.3 15.4 13.6 National Education Exp. (% GDP) 2.8 2.7 3.1 2.7 2.6 3.1 3.0 3.0 Total Nominal National Expenditures 363.7 350.6 429.9 472.8 528.1 671.9 768.5 1,000.1 Source: World Bank staff calculations based on MoF and SIKD data. Note: * = estimated sub-national budget, ** = APBN-P and estimated sub-national budget. Figure 3.1. Trends in national education 3.2. Education Spending and the expenditures, 2001­08 Importance of the Districts as a 160,000 18 Spending Unit 140,000 16 14 120,000 12 100,000 In 2006, the majority of education expenditures, 10 about 56 percent, was spent at the sub-national 80,000 rupiah 8 Percent level. District governments are the main spenders, 60,000 Billion 6 accounting for 51 percent of total spending, while 40,000 4 provincial governments account for only just over 5 20,000 2 percent. These shares of total education expenditures 0 0 demonstrate the trend in education service delivery, 2001 2002 2003 2004 2005 2006 2007 2008* with the shares of the district governments being Total national education spending nominal relatively high compared with the central government. Total national education spending (2006=100) % of education to total expenditure In 2005 and 2006 there was an increase in central Source: World Bank staff calculations based on data from MoF. expenditures. This can be partly explained by Note: Figures for 2007 and 2008 are partially based on budget additional new spending on the GoI's BOS program, estimates, as opposed to expenditures which started in 2005 and originates at the central level (Table 3.2). Table 3.2. Nominal education expenditures by level of government, 2001­06 Rp trillion 2001 % 2002 % 2003 % 2004 % 2005 % 2006 % Central 12.1 30 13.5 27 21.4 34 18.1 29 29.3 40 45.3 44 Development 8.2 60 9.2 62 16.0 69 12.5 63 - - - Routine 3.9 40 4.2 38 5.4 31 5.6 37 - - - Provincial 1.9 5 4.0 8 3.9 6 3.8 6 3.9 5 5.4 5 Development 1.4 70 2.6 66 3.1 80 3.0 79 3.0 78 4.3 79 Routine 0.6 30 1.4 34 0.8 20 0.8 21 0.9 22 1.1 21 District 26.2 65 32.6 65 38.3 60 39.8 64 40.0 55 51.8 51 Development 3.0 11 4.6 14 5.3 14 4.6 12 5.6 14 9.7 19 Routine 23.2 89 28.0 86 33.0 86 35.2 88 34.5 86 42.1 81 Total National Expenditures 42.3 100 51.3 100 64.8 100 61.8 100 73.2 100 102.5 100 Source: World Bank staff calculations based on data from MoF. 22 Investing in Indonesia's Education at the District Level Chapter 3 Education Expenditures at the Local Level Although the districts spend the majority of the Furthermore,themajorityofroutineexpenditures total education budget, these expenditures are at the sub-national level is allocated to personnel mostly non-discretionary routine expenditures. spending, followed by goods expenditures (Table Hence, while decentralization formally devolved the 3.4). Hence, although sub-national governments responsibilities for education down from the central account for a significant share of expenditures in the level to the sub-national level, the majority of the education sector, they have limited fiscal space for development budget is still spent by the central development expenditures. In 2004, development government. Since 2001, the central government expenditures comprised about 30 percent of national has been consistently covering more than 55 percent consolidated expenditures on education, whereas in of total development spending (and typically more 2003 they accounted for a little more, at about 38 than 60 percent), with districts only covering about percent. The slight decrease in education spending in a quarter (Table 3.3). Thus, local governments have 2004 resulted predominantly from a decline in central surprisingly little discretion in managing funds and development spending. shaping the key education sector decisions. Table 3.3. Share of development and routine expenditures by level of government, 2001-04 2001 2002 2003 2004 Total development expenditures (Rp trillion) 12.6 16.4 24.4 20.1 Central development as % of total 65 56 66 62 Province development as % of total 11 16 13 15 District development as % of total 24 28 22 23 Total routine expenditures (Rp trillion) 27.7 33.6 39.2 41.6 Central routine as % of total 14 13 14 13 Province routine % of total 2 4 2 2 District routine as % of total 84 83 84 85 Source: World Bank staff calculations based on data from MoF. Figure 3.2. Education spending by economic classification, level of government and routine district spending composition 2% Province 13% Central Government 4% Salaries Development Routine 85% District 96% 33% 67% Source: World Bank staff calculations based on data from MoNE. An Analysis of Regional Public Expenditure and Financial Management 23 Chapter 3 Education Expenditures at the Local Level Table 3.4. Routine expenditure distribution by Figure 3.3. Source of district education level of sub-national government, 2002-06 expenditures, 2005 Percent BOS, 6% Composition of 2002 2003 2004 2005 2006 routine expenditure Personnel expenditure 94 94 95 95 96 Goods expenditure 4 4 3 4 3 HH out of pocket expenditure, O&M expenditure 0 1 1 1 0.4 LGs spending, 31% 53% Travel expenditure 0 0 0 0 0.5 Miscellaneous 2 2 0 0 0 expenditure Other expenditure 0 0 0 0 0 Sectoral central spending, 10% Total routine 100 100 100 100 100 expenditure Source: World Bank staff calculations based on SIKD/MoF, districts' Source: World Bank staff calculations based on data from MoF. APBD data, and Susenas/BPS. In accordance with these national trends, district governments in the 10 visited districts are the 3.3. District Expenditure Patterns largest contributors to their respective district and Trends education budgets. District government shares of expenditures amount to about 90 percent of total education expenditures in the three districts of Kab. In Indonesia, total education spending at the Jayawijaya, Kab. Jayapura, and Kab. Timtengsel. district level reached Rp 74.4 trillion in 2005. Resources from central government in the form of This was a combination of district government deconcentrated funds appear to play only a limited expenditures, deconcentrated funds spent in the role in the education sector in most of the districts, regions, and private out-of-pocket expenditures on except for Kota Manado, where deconcentrated funds education. Total education expenditure has increased account for around 41 percent of total education by 5.9 percent since 2004. District governments spent expenditures. At the same time, household out- the largest share of total education expenditure in of-pocket payments are significant and on average 2005, accounting for 53 percent of total expenditure constitute about 20 percent of the total district (about Rp 41.8 trillion). Out-of-pocket expenditures expenditures in the 10 districts examined. for education are the second-largest component of expenditures, at Rp 24.3 trillion or 31 percent of Figure 3.4. District education expenditures by total expenditure. As a largely decentralized sector, source of fund in visited districts, 2005 the central government only spent a relatively small share on education, accounting for 16 percent of 100% total education expenditure. Deconcentrated funds 80% for education declined from Rp 12.2 trillion in 2004 to 60% Rp 8.2 trillion in 2005. 40% 20% 0% Kab. AsahaKn ota Binjai ab. Wonosoobo K K ab. Jayapuraational N ta Magelanbg Ka . Minahasaa Manado Kot Kab. Timtengsel Kab. Belu Kab JayawKijaya Deconcentration fund LGs HH out-of-pocket Source: World Bank staff calculations based on SIKD/MoF and districts' APBD data. 24 Investing in Indonesia's Education at the District Level Chapter 3 Education Expenditures at the Local Level Aggregate district government spending on expenditures during the same period. On the other education has increased in real terms since the hand, Kab. Minahasa experienced a significant decline implementation of decentralization in 2001. in its spending, from Rp 177.2 billion in 2001 to Rp86.3 Education spending increased from Rp 29.3 trillion in billion in 2006 due to the division of Kab. Minahasa 2001 to Rp 35.1 trillion in 2004 and further to Rp into four new districts. Kab. Jayawijaya experienced a 36.8 trillion in 2006. As a share of total expenditure, decline in 2004 and 2005 as spending realization was spending on education shows an opposite decreasing still very low by the end of the calendar year. For the trend from 37 percent in 2001 to 33 percent in 2004 other districts, the trend has been relatively stable with and 28 percent in 2006. The decline in 2005 and 2006 an average annual growth rate of about 3.5 percent. might be due to the increase of central government support through the BOS program that started in July Figure 3.6. Education expenditures in the visited 2005. districts, 2001-06 250,000 Figure 3.5. Education expenditure and percentage of total expenditure, 2001-06 200,000 40,000 38 150,000 35,000 36 34 30,000 million 100,000 32 Rp 25,000 30 20,000 50,000 Billion 28 Rp 15,000 Percent 26 10,000 0 24 a 5,000 22 Belu Asahan Binjai Menado 0 20 Magelang Minahasa Kab. Jayapur Kab. Kota Wonosobo Timtengsel Jayawijaya Kota 2001 2002 2003 2004 2005 2006 Kab. Kab. Kota Kab. Kab. Kab. Total education expenditure % of total expenditure 2001 2002 2003 2004 2005 2006 Source: World Bank staff calculations based on SIKD/MoF and Source: World Bank staff calculations based on SIKD/MoF and districts' APBD data (at constant 2002 prices). districts' APBD data (at constant 2002 prices). Most visited districts have increased their In terms of education shares of total expenditure, spending on education since the implementation the trend varies among the 10 districts. In most of decentralization, except for Kab. Minahasa. cases, at least one district from each province has Kota Manado has seen a significant increase in its experienced a declining share in its budget allocation education expenditures, followed by Kab. Wonosobo. for education. The sharpest decline was experienced Education expenditures increased from Rp 15.0 billion by Kab. Minahasa, which also suffered the largest in 2001 to Rp 100.5 billion in 2006 for Kota Manado, decline in its total expenditure figure. The share of whereas Kab. Wonosobo more than doubled its Table 3.5. Education expenditure as a share of total expenditure in visited districts, 2001-06 Percent Districts 2001 2002 2003 2004 2005 2006 Kab. Asahan 50.0 45.3 45.5 45.5 44.9 39.9 Kota Binjai 21.7 29.5 31.2 32.8 31.0 26.3 Kab. Wonosobo 32.8 40.9 35.0 40.9 40.3 37.0 Kota Magelang 35.9 29.9 35.4 34.1 36.0 30.6 Kab. Minahasa 52.6 61.0 56.0 49.6 38.3 35.1 Kota Manado 8.2 36.6 10.2 42.1 37.6 30.2 Kab. Timtengsel 41.4 35.7 34.0 22.8 38.3 35.7 Kab. Belu 35.8 30.7 34.8 35.0 35.8 32.2 Kab. Jayawijaya 22.0 18.7 17.6 23.6 20.5 17.8 Kab. Jayapura - 6.2 7.2 21.7 22.7 14.0 Source: World Bank staff calculations based on SIKD/MoF and districts' APBD data. An Analysis of Regional Public Expenditure and Financial Management 25 Chapter 3 Education Expenditures at the Local Level education to total expenditure declined from 52.6 3.4. Economic and Programs percent in 2001 to 35.1 percent in 2006. Conversely, Kota Manado, which is situated in the same province Composition as Kab. Minahasa, saw a significant increase in its allocation to the education sector, from 8.2 percent The vast majority of district education in 2001 to 42.1 percent in 2004, and subsequently to expenditures constitutes routine spending, 30.2 percent in 2006. Kab. Jayapura also experienced leaving limited space for development spending an increase in its allocation to education, while Kab. at the kabupaten or kota level.4 On average, from Jayawijaya saw the opposite trend. Kab. Belu in NTT, 2001 to 2006, routine expenditure for all districts in however, showed a stable allocation pattern in terms Indonesia accounted for 87 percent of total education of its expenditures to education, within a narrow expenditures. Yet, while constituting of a relatively range of 31 to 36 percent between 2001 and 2006. small share to total expenditure, development expenditure experienced a significant increase from The majority of districts visited have per capita Rp 3.0 trillion in 2001 to Rp 9.7 trillion in 2007. A education expenditures that are higher than similar trend was observed in the 10 visited districts. the national average. Kab. Jayapura in Papua More than 90 percent of education expenditures were has the highest per capita spending on education at allocated to routine spending in Kab. Asahan in 2001- almost Rp 800,000 per person. Kab. Asahan in North 05. Nevertheless, relative increases for development Sumatra and Kab. Wonosobo in Central Java have the spending were found in Kab. Wonosobo, Kota lowest per capita spending level at slightly above Rp Magelang, and Kab. Belu, while Kota Manado 200,000. Kab. Asahan has a relatively low poverty and Kab. Jayawijaya saw declining allocations to rate, at 12 percent, while in contrast Kab. Wonosobo development expenditure. Kab. Timtengsel showed a has a poverty rate exceeding 30 percent. Although relatively constant allocation pattern, with an average further research is required, this quick snapshot shows of 13.3 percent allocated to development in 2001- that education spending per capita is not necessarily 05. higher in the districts with lower poverty rate. Figure 3.7. Per capita education expenditure in the 10 visited districts, 2006 Kab. Jayapura Kota Magelang Kab. Minahasa Kab. Jayawijaya Kota Menado Kab. Timtengsel Kota Binjai Kab. Belu National Kab. Wonosobo Kab. Asahan 0 100 200 300 400 500 600 700 800 900 (Rp'000) Source: World Bank staff calculations based on BPS, SIKD/MoF and districts APBD data. 4 World Bank 2007(b) indicates that while decentralization formally devolved the responsibilities for education to the central level, the majority of development budget in education is still spent by central government. 26 Investing in Indonesia's Education at the District Level Chapter 3 Education Expenditures at the Local Level Table 3.6. Composition of routine and development education expenditure, 2001-05 Percent 2001 2002 2003 2004 2005 2006 District Rtn Dev Rtn Dev Rtn Dev Rtn Dev Rtn Dev Rtn Dev Kab. Asahan 93.0 7.0 92.6 7.4 92.6 7.4 93.1 6.9 96.5 3.5 89.5 10.5 Kota Binjai 89.0 11.0 91.0 9.0 84.1 15.9 91.1 8.9 90.3 9.7 89.2 10.8 Kab. Wonosobo 96.7 3.3 92.5 7.5 88.5 11.5 84.2 15.8 90.7 9.3 87.8 12.2 Kota Magelang 97.2 2.8 92.7 7.3 71.2 28.8 87.7 12.3 88.2 11.8 94.2 5.8 Kab. Minahasa 98.2 1.8 99.9 0.02 98.9 1.1 96.8 3.2 88.6 11.4 84.8 15.2 Kota Manado 83.2 16.8 98.8 1.2 91.8 8.2 96.4 3.6 96.0 4.0 92.7 7.3 Kab. Timtengsel 87.1 12.9 90.5 9.5 89.5 10.5 84.4 15.6 82.2 17.8 77.0 23.0 Kab. Belu 95.4 4.6 94.5 5.5 86.1 13.9 87.2 12.8 87.8 12.2 84.2 15.8 Kab. Jayawijaya 68.5 31.5 76.9 23.1 86.9 13.1 97.9 2.1 90.7 9.3 75.2 24.8 Kab. Jayapura - - 51.3 48.7 45.8 54.2 92.1 7.9 80.1 19.9 74.1 25.9 All districts Indonesia 89.3 10.7 86.9 13.1 87.1 12.9 88.9 11.1 86.1 13.9 81.3 18.7 Source: World Bank staff calculations based on SIKD/MoF and districts' APBD data. A worrying observation relates to the levels of Education spending on personnel absorbs the operations and maintenance expenditure, which bulk of routine functions at the aggregate level are less than 1 percent of routine education and in the visited districts. In 2006, personnel spending. The low allocations for this function spending accounted for more than 90 percent of are far from sufficient to bring any improvement in routine spending for districts in Indonesia. For most of education infrastructure and facilities. It appears the visited districts the number was similar, with more that most districts still rely on central and provincial than 95 percent of routine resources for the education assistance in terms of maintenance, or classify such sector going towards personnel. For Kab. Minahasa, expenditures under other categories. Kab. Asahan, and Kota Manado, personnel spending reached almost 100 percent of routine spending. Spending on goods and services accounts for about 4 percent and less of routine spending for districts overall and all visited districts, respectively. Figure 3.8. Economic composition of routine expenditures in visited districts, 2006 100% 80% 60% 40% 20% 0% r Kab. Asahan Kota Binjai . Wonosobo a Magelang r Tenan gah National Kab Kot Kab. MinahasaKota Menado Kab. Belu Kab. TimuSelat Kab. Biak NumfoKab. JayapuraKab. Jayawijaya Personnel Goods Maintenance Travel Others Source: World Bank staff calculations based on SIKD/MoF and districts' APBD data. An Analysis of Regional Public Expenditure and Financial Management 27 Chapter 3 Education Expenditures at the Local Level Budget absorption rates are generally high for approval of the APBD. Until fiscal year 2007, many recurrent expenditure, while absorption for district governments failed to approve their APBDs development expenditure is less optimal. In three until after January 1. This may delay the disbursement out of the 10 districts studied, the absorption rate of capital expenditure and cause projects to lag behind does not reach 90 percent. One of the factors that schedule. may hinder full absorption of the budget is the late Table 3.7. Budget planned vs. realized in the visited districts, 2005 Rp million Routine Expenditure Development Expenditure Districts Plan Realized % Realized Plan Realized % Realized Kab. Asahan 169,597.7 169,389.8 99.9 6,358.8 6,165.2 97.0 Kota Binjai 56,260.3 543,32.0 96.6 5,819.1 5,808.6 99.8 Kab. Wonosobo 131,356.5 122,133.4 93.0 12,637.4 12,580.8 99.6 Kota Magelang 59,860.4 54,670.5 91.3 8,760.9 7,348.7 83.9 Kab. Minahasa 89,064.5 88,447.5 99.3 11,425.3 11,425.0 100.0 Kota Manado 108,452.8 108,338.4 99.9 5,672.0 4,526.3 79.8 Kab. Timtengsel 84,365.9 86,648.1 102.7 19,641.3 18,720.5 95.3 Kab. Belu 75,116.1 74,685.0 99.4 15,544.6 10,344.0 66.5 Kab. Jayawijaya 50,683.8. 46,436.1 91.6 5,119.5 4,745.0 92.7 Kab. Jayapura 50,586.8 50,133.2 99.1 12,490.0 12,490.0 100.0 Source: World Bank staff calculations based on SIKD/MoF and districts' APBD data. Box 3.1. What drives the education expenditure pattern? The consistent pattern across districts on high salary shares are partly driven by three factors namely: (i) the expansion on the number of teachers, (ii) inherited teachers from centralized era, and (iii) lack of flexibility in teacher's mobility. The expansion on the teacher numbers is somewhat based on the subjective perception that there is still an undersupply of teachers in district schools. Based on the teacher employment and deployment survey conducted by the World Bank in 2005, 74 of 276 primary schools that claim to have an undersupply actually show an oversupply according to the current teacher entitlement formula. This general perception might explain as well the growing numbers of teachers in districts in Indonesia as well as in the visited districts, where in average these districts have 5 percent of growth rate in their number of teachers between 2004 and 2006. The large number of civil servant teachers inherited from centralized era contributed to the problem of excess supply of teachers. Prior to decentralization era, teachers were allocated to schools align with the expansion of school Inpres program. However, with the falling of birth rate, the needs of primary school have declined and some districts have even merged their schools, creating an excess supply of teachers. The World Bank Education Sector Review (2005) indicates that prior to decentralization, Indonesia have already had over 2 million teachers for the entire school system in Indonesia. Of this figure, 93 and 62 percent that were employed in primary and junior secondary schools respectively are civil servants. The transfer of responsibility in the decentralization era has caused districts to inherit the large wage bill of those civil servant teachers employed in centralized era. The fact that civil servant teachers are employed under the national civil service regulation provides district with less flexibility for teacher employment and deployment. Proper sizing of teachers were constrained by the facts that there is still ambiguity on who is responsible in terms teacher employment and deployment across government in the country. However, the new Government Regulation No. 38/2007 on intergovernmental relations and sub-national roles and functions is expected to clarify and address the uncertainty in the responsibility of teacher deployment and employment. 28 Investing in Indonesia's Education at the District Level Chapter 3 Education Expenditures at the Local Level Analyzing expenditures at the program level Budget allocations for staff capacity building and shows that districts' priorities vary, but in 2006 educational development program, institutional most of the districts targeted the Education management program, and planning, Infrastructure and Facilities Development implementing, monitoring and evaluation Program, as well as the Pre-school and Basic program remain low. The allocations for the teacher Education Program. Five out of ten districts and education personnel capacity building program allocated 50 percent or more of their non-salary range between 1 and 14 percent of the non-salary budgets to the Education Infrastructure and Facilities budget. Kab. Minahasa hardly allocates any resources Development Program. Kab. Wonosobo has the to this program, while Kota Binjai allocates close to smallest allocation to this particular program, but 14 percent. There is a tendency for kota to allocate that is mostly because it classifies its allocations on larger shares to these programs than the kabupaten, infrastructure and facilities under basic education and except in the case of Kab. Belu. The institutional secondary education programs. After infrastructure and organizational management program receives and facilities, the majority of districts' second-largest even lower allocations, and seems far from being on allocation at the program level is to pre- and basic the list of priorities. It seems that allocations to this education programs that focus on capacity building program are less than 1.0 percent of budgets in all and quality enhancement. The visited districts in of the visited districts. Similar results are observed North Sumatra, Central Java, and South Sulawesi when analyzing the planning, implementation, allocated more than one-fifth of their budget plans monitoring and evaluation program, which includes to these programs, which include activities such as activities such as data collection and coordination, science competitions, mathematics advancement work plan preparation, socialization of the education programs and exercise programs, as well as support strategic plan, etc. Most of the districts allocate less for the preparation of final school examinations. than 1 percent, except for Kota Magelang and Kab. Timtengsel, which allocate 5 and 6 percent of their non-salary planned budgets, respectively. Figure 3.9. Program composition of the visited districts, 2006 100% Planning, Implementation, Monitoring & Education Development Evaluation Program 80% Credit Score, Administration Clarification, Mutation, Pension, & Death Assessment Program Youth and Sport Development Program 60% Art and Culture Development Program Scholarship Program Institutional and Organization 40% Management Program SNP Acreditation Quality Enlargement & Enhancement Program 20% Teacher and Education Staff Competency Development Program Education Infrastructure & Facilities Development Program 0% Access to Non-Formal Education Enhancement program Binjai Belu Secondary Education Development Asahan Manado Program Jayapura Pre-School and Basic Education Kota Wonosobo Magelang Minahasa Timtengsel Kab. Jayawijaya Kab. Development program Kota Kab. Kota Kab. Kab. Kab. Kab Source: World Bank staff calculations based on districts DASK and LAKIP data 2006. An Analysis of Regional Public Expenditure and Financial Management 29 Chapter 3 Education Expenditures at the Local Level Underthe`public'and`apparatus'budgetformat, The general administration component received most visited districts spend a large share of their the largest allocation in both apparatus and education expenditures on the public category.5 "public" category. General administration on The introduction of Ministerial Decree No. 29/2002 average accounts of more than 55 percent of has changed the district government budget format spending in apparatus and "public" category. On from `routine' and `development' classifications to average, operations and maintenance only accounts `apparatus' and `public' expenditure categories. The for 6 percent of apparatus spending and 25 percent Format No. 29/2002 of expenditure affects mainly the of "public" expenditure category. However, Kab. structure of expenditure, while the revenue structure Asahan and Kota Binjai in North Sumatra, as well has remained largely the same. Under this format the as Kab. Jayapura in Papua, spent the majority of focus of spending is on beneficiaries rather than on their "public" expenditure category on operations programs and projects. A large allocation of spending and maintenance and capital. Spending on capital on the public category, however, has created a investment accounts for the smallest proportion, somewhat biased classification, since most districts averaging less than 1 percent for apparatus and about categorize their spending based on a subjective 15 percent of the "public" expenditure category. perception that education belongs to public service. Overall, although education expenditure allocated a The inconsistent classification of the education sector large share of the spending to the public category, the under the Format No. 29/2002 among districts is proportion goes mainly to general administration and partly due to unclear guidance and vague definitions leaves only a small proportion for capital. of what belongs to public and apparatus spending. Table 3.8. Public and apparatus spending on education Percent 2005 2006 District Apparatus Public Fin. Assist Total Apparatus Public Fin. Assist Total Kab. Asahan 96.58 3.39 0.03 100.0 89.73 10.27 - 100.0 Kota Binjai 90.42 9.58 - 100.0 89.85 10.15 - 100.0 Kab. Wonosobo - 97.61 2.39 100.0 - 97.48 2.52 100.0 Kota Magelang 11.32 88.68 - 100.0 13.70 86.30 - 100.0 Kab. Minahasa 2.05 97.95 - 100.0 3.09 96.91 - 100.0 Kota Manado - 100.00 - 100.0 - 100.00 - 100.0 Kab. Timtengsel 14.62 85.38 - 100.0 17.02 82.98 - 100.0 Kab. Belu 0.61 99.39 - 100.0 0.77 99.23 - 100.0 Kab. Jayapura 85.49 14.51 - 100.0 74.15 25.85 - 100.0 Kab. Jayawijaya 89.90 10.10 - 100.0 75.21 24.79 - 100.0 Source: World Bank staff calculations based on SIKD/MoF and districts' APBD data. 5 For the purpose of time series analysis, the classification based on routine and development expenditure is adopted in this review and not on public and apparatus classification. 30 Investing in Indonesia's Education at the District Level Chapter 3 Education Expenditures at the Local Level Figure 3.10. The breakdown of public and apparatus spending in 10 visited districts, 2006 Apparatus Public 100% 100% 90% 80% 80% 70% 60% 60% 50% 40% 40% 30% 20% 20% 0% 10% Kab. Kota Kota Kab. Kab. Kab. Belu Kab. 0% Asahan Binjai Magelang Minahasa Timtengsel Jayawijaya Kab. Kota Kab. Kota Kab. Kota Kab. Kab. Belu Kab. Asahan Binjai Wonosobo Magelang Minahasa Menado Timtengsel Jayapura General admin Operational & Maintenance Capital General admin Operational & Maintenance Capital Source: World Bank staff calculations based on SIKD/MoF and districts' APBD data. 3.5. Household Out-of-Pocket Most of the 10 visited districts, household out- of-pocket spending declined in 2006 compared Spending with previous years. Two districts in North Sulawesi and one district in North Sumatra experienced a At the national level, districts' out-of-pocket significant decline in their out-of-pocket spending spending contributes significantly to total between 2004 and 2006. Household spending in Kab. education expenditures and has increased by 32 Minahasa dropped in real terms from Rp 51.7 billion percent between 2004 and 2005. According to the in 2004 to Rp 11.8 billion in 2006, while spending by Governance and Decentralization Survey (GDS) data, households in Kota Manado declined by 64 percent the majority of household spending on junior and from Rp 42.9 billion in 2004 to only Rp 45.5 billion in senior secondary schools is directed towards books, 2006. Out-of-pocket spending declined from Rp 22.7 writing materials and transportation. Expenditures billion in 2004 to Rp 7.8 billion in 2006 for Kota Binjai. on enrollment and monthly schools fees are relatively Kab. Asahan and Kab. Belu show relatively stable small, probably due to the impact of the BOS program, expenditure patterns at the household level, although which provides support from the central government there was a slight decline in 2006 compared with the directly to schools. According to Aran (2007), of previous year. Kab. Jayapura and Kab. Jayawijaya the households that had at least one child enrolled are the only districts that experienced increases in in a primary school that received BOS, 72.6 percent their household spending between 2005 and 2006. reported that their monthly school fees were reduced The significant decline that was experienced by the in the 2004-05 academic year, while at the junior three districts mentioned above is probably the result secondary level 82.6 percent of households that had of the scholarship programs provided by district children enrolled in schools that received the BOS governments, which is an addition to the BOS support program reported fee reductions. from the central government. An Analysis of Regional Public Expenditure and Financial Management 31 Chapter 3 Education Expenditures at the Local Level Figure 3.11. Household out-of-pocket expenditure in the visited districts, 2004-06 60,000 2004 2005 2006 50,000 40,000 30,000 million Rp 20,000 10,000 0 Kab. Kota Binjai Kab. Kota Kab. Kota Kab. Kab. Belu Kab. Kab. Asahan Wonosobo Magelang Minahasa Menado Timtengsel Jayawijaya Jayapura Source: World Bank staff calculations based on Susenas/BPS data. the 1,251 primary and junior secondary schools did 3.6. Spending at the School Level not receive BOS (or did not know if they had received BOS). In addition, according to the Susenas panel School budget consists of funding from multiple data, 74.2 percent of households with children in sources, with the largest share coming from primary school and 64.6 percent of households with the recently launched BOS program. The BOS children at junior secondary school thought that at program started in July 2005 and funds channeled least one of their children's schools had received BOS.6 through the program are directly transferred from There was some regional variation in the perceived central government to each school's bank account. coverage of BOS: for instance, while in Java and Bali The funds are intended to lower or eliminate school 77.8 percent of households with children in primary fees, while maintaining the quality of education. school (and 67.3 percent of households with children According to the BOS guidelines, the operational in junior secondary school) reported that their child's costs can include the cost of registration of new school received BOS, in Papua only 37.8 percent of students, text books and reference books, stationery, households with children in primary school (and 35.5 examinations, teacher development and training, percent of those in junior secondary school) reported school maintenance, transport costs for poor students, that their child's school received BOS funding. remuneration of temporary teacher, and electricity, water and telephone. The allocation of BOS covered almost 40 million students in 2007, accounting for Rp 11.8 trillion. Before the implementation of the BOS, the The program distributed 73 percent of the funds to largest share of school budgets came from the the primary school level and the remaining 27 percent district government and household/community to the junior secondary level. The size of the grant contributions. In 2006, the BOS constituted more to a school is determined based on the number of than 50 percent of school budgets on average. In students. The per-pupil BOS allocation has increased SDN 1 Kalijajar and SDN 1 Candimulyo -- the visited from Rp 235,000 to Rp 276,000 per student per schools located in Kab. Wonosobo in Central Java -- year for primary schools, and from Rp 324,500 to the BOS program contributed up to 82 and 91 percent Rp 376,000 per student per year for junior secondary of the total school budget in 2006, respectively. schools since 2007. By covering school operational costs as well as lowering and often eliminating school All public and private schools in Indonesia fees, the BOS program is expected to encourage poor are eligible to apply for BOS and a recent children to attend school and thereby assist in meeting quantitative assessment based on GDS 2 data found that the BOS program coverage is very 6 There may be a downward bias in the coverage from wide. In the GDS 2 sample, only one school out of household data if program socialization to households was sub-optimal. 32 Investing in Indonesia's Education at the District Level Chapter 3 Education Expenditures at the Local Level the government's mandate of completing the nine However, 67.5 percent of schools that reported having years of basic education. an `adequate' number of teaching staff still chose to spend part of BOS on teacher honoraria. In terms of the use of BOS funds, it was found thatschoolsusedmostofthefundingfor`quality- Greater socialization of the program is still enhancing' rather than `access-increasing' requested by respondents in the GDS 2 survey. investments. GDS 2 data report the responses from Aside from the large share of principals who reported school principals and school committees in the use of that they had received sufficient information, 20 BOS funding at the school level. According to these percent of school principals nonetheless reported that data, at the primary school level 95.4 percent of the socialization of the BOS program was inadequate. school principals reported that BOS funding was used From the parents' side, 79 percent of respondents for textbooks and stationery, 74.5 percent reported had children in primary and junior secondary schools it was used for maintenance of school facilities, 71.1 that received BOS funding. Of these parents, only percent reported it was used for teacher honoraria 73 percent reported having received socialization on and salaries, and only about 46.5 percent said it school fees and costs. This reflects both the need and was used to assist poor students. The percentage of the demand of more program socialization to schools junior secondary schools that reported assisting poor and parents. students using the BOS grant was higher, with 56.7 percent of school principals reporting that the funds The allocation of BOS by level of education for the were used for providing aid to poor students. visited districts was 63 percent to primary school level and 37 percent to the junior secondary Theneedsdeterminedattheschoollevelreflected school level in 2005.7 The BOS funds in the visited to a degree the use of BOS funds. However, the districts were allocated to more than 3,500 schools relationship between the needs of schools and and 663,000 students in 2005, of which 13 schools the use of BOS funds was at times unclear. For are special schools for disability students with 491 instance, at schools where the number and quality pupils in total. As for the distribution of BOS based of teachers is `adequate' as reported by the school on school category, on average around 67 percent of committee and school principal, one would expect BOS funds were allocated to public schools, while the to see less spending from BOS on teacher salaries. remainder went towards private schools. Table 3.9. BOS fund allocation based on school level and type in the visited districts, 2005 Rp million Districts Primary level Junior secondary Public school Private school Kab. Asahan 16,528 5,793 19,427 2,891 Kota Binjai 2,071 2,310 3,080 1,301 Kab. Wonosobo 9,943 8,800 16,496 2,247 Kab. Minahasa 3,750 1,940 3,289 2,401 Kota Manado 5,370 2,940 4,231 4,080 Kab. Belu 7,447 2,257 6,316 3,388 Kab. Timtengsel 5,640 1,693 3,562 3,771 Kab. Jayawijaya 3,135 1,292 2,876 1,552 Kab. Jayapura 1,097 1,774 2,186 685 Source: World Bank staff calculations based on MoNE data (http://bos.dit-plp.go.id/aboutBOS.php). 7 The number covers public and private schools at primary and junior secondary level, but excludes religious schools that are in similar level with primary and junior secondary school. An Analysis of Regional Public Expenditure and Financial Management 33 Chapter 3 Education Expenditures at the Local Level Most visited districts received BOS in 2005. Kab. Asahan received the largest allocation, at more than Rp 22 billion in 2005, followed by Kab. Wonosobo at Rp 18.7 billion. Meanwhile, the two districts in Papua received the lowest allocations, corresponding to the number of students in these districts. Based on GDS 2 survey, some schools in districts in Bengkulu, Gorontalo, North Maluku, NTT, and North Sumatra had to pay a special fee to retrieve the funds, while some schools in districts in Papua reported that they had to pay a special `tax' of 11.5 percent to retrieve the BOS grant. In the five visited districts that are also covered by the GDS 2, the allocation of BOS funds is reported largely for books. On average for the five districts, 95 percent of school principals perceived that the funds had an impact on book availability, 71 percent reported that it had improved school maintenance, and 47 percent believed that BOS had enhanced the education access of poor students. Some schools in these five districts have also been able to provide support for the financing of transportation costs of poorer students, particularly schools in Kab. Timtengsel in NTT. Although most schools did not return payments they had previously received from parents or the community after the implementation of BOS, in some cases the money was returned either in part or completely, even when the schools received smaller amount of BOS than expected. 34 Investing in Indonesia's Education at the District Level Chapter 4. District Education System and Outputs 4.1. Education Infrastructure and In most districts, the main text book is not always available on a one book-one student basis. Data Facilities from the GDS 2 survey in the respective districts show that on average at the primary level only 50 percent In general, the condition of education of students have one Indonesian language book and infrastructure in many districts in Indonesia is one math book, while at the junior secondary level still poor. Poor education infrastructure is particularly the percentage is even lower at 41 percent. As for apparent at the primary level. In 2006 about 25 the visited districts covered in the GDS 2 survey, percent of classrooms across Indonesia were classified only Kab. Asahan and Kab. Timtengsel provide one by the MoNE as badly damaged. Another 30 percent main book for students. Kab. Asahan provides one was lightly damaged. The condition of schools at the Indonesian language book and one math book per junior and secondary school levels is better, but even primary student, while Kab. Timtengsel provides one so an estimated of 19 and 10 percent of classrooms, math book per student at both primary and junior respectively, were classified as damaged. secondary levels. Poor classroom condition was reflected in the 10 There are variations in access to schools in the visited districts. Kab. Belu and Kab. Timtengsel in visited districts, but all score better than the NTT have the highest percentage of primary school national average. Access in most districts is good classrooms that are heavily damaged. About 40 at all school levels, although greater variation is found percent of classrooms in these two districts were among sub-districts. The service area ranges from badly damaged and only 35 percent of classrooms 0.2km2 to 40km2 for primary levels, 1km2 to 212km2 are considered acceptable. In contrast, Kota Binjai for junior secondary levels, and 1km2 to 732 km2 in North Sumatra and Kota Magelang in Central Java for senior secondary levels. As expected, smaller have the highest percentage of acceptable classrooms service areas are found in the cities (Kota Binjai, Kota condition, with more than 75 percent of classrooms Magelang, and Kota Manado) since these districts are being in a good condition. For junior and secondary smaller in geographical size and have more schools schools, most districts visited in eastern Indonesia, than kabupaten. Districts in Papua have the highest excluding Kota Manado, have classroom conditions schoolserviceareascomparedwithotherdistricts. Kab. worse than the visited districts in Central Java and Jayapura has particularly large service areas compared North Sumatra. The rehabilitation of schools needs with the national average, with primary, junior and to be made a priority at the district level, particularly senior secondary schools covering areas as large since poor classroom conditions affect the quality and 148km2, 547km2, and 1,030km2 for the three levels, outcomes of investments in the education sector. respectively. The national average stands at 23km2 for An Analysis of Regional Public Expenditure and Financial Management 35 Chapter 4 District Education System and Outputs primary schools, 115km2 for junior secondary schools, scores below the national average of seven. The and 292km2 for senior secondary schools. distribution of schools, however, is more unequal among many sub-districts. For example, in Kab. Belu, Table 4.1. Average service areas for schools in the sub-district of Malaka Barat has a ratio that is the visited districts, 2006 six times lower than that of the sub-district of Sasita Mean, while the sub-district of Selomerto in Kab. Junior Senior District Primary secondary secondary Wonosobo has a ratio of four times higher than sub- district Mojotengah within the same district. The lower Kab. Asahan 7 39 51 number of schools in some districts and sub-districts, Kota Binjai 1 2 4 however, does not necessarily reflect insufficiency in Kab. Wonosobo 2 14 58 school services, as some schools may have a limited Kota Magelang 0.2 1 1 number of students. In this case, merging schools Kab. Minahasa 3 13 43 and students might improve efficiency. Kota Manado 1 2 3 Kab. Belu 8 53 163 Table 4.2. Schools per 1,000 school-aged Kab. Timtengsel 9 62 232 children in the visited districts, 2006 Kab. Jayawijaya 40 212 732 Junior Senior District Primary Kab. Jayapura 148 547 1,030 Secondary Secondary National 23 115 292 Kab. Asahan 5 2 1 Average Kota Binjai 5 3 2 Source: World Bank staff calculations based on various District Kab. Wonosobo 5 1 1 Education Profile and Susenas/BPS data. Kota Magelang 6 3 2 The ratio of primary schools per 1,000 primary Kab. Minahasa 12 6 2 school-aged children shows that the number of Kota Manado 6 4 2 schools in some of the visited districts is lower Kab. Belu 6 2 1 than the national average. On average, the visited Kab. Timtengsel 7 2 1 districts have seven primary schools per 1,000 primary Kab. Jayawijaya 4 3 1 school-aged children. The highest number of schools Kab. Jayapura 9 6 3 is found in Kab. Minahasa, while the lowest number National 7 3 2 is found in Kab. Jayawijaya, with 12 and 4 primary Average schools per 1,000 primary school-aged children, respectively. Again, with only four primary schools per Source: World Bank staff calculations based on various District Education Profile and BPS data. 1,000 primary school-aged children, Kab. Jayawijaya Figure 4.1. Primary schools per 1,000 primary school-aged children by sub-districts in Kab, Belu and Kab. Wonosobo, 2006 Kab. Bel Kab. Wonosobo Sasita mean selomerto sukoharjo Raihat kepil Malaka tengah kalibawang Lamaknen kaliwiro Kakuluk mesak kalikajar Rinhat wadaslintang kejajar Kobalima wonosobo Tasifeto barat watumalang Tasifeto timur leksono Malaka timur kertek Atambua sapuran garung Malaka barat mojotengah 0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14 16 Source: World Bank staff calculations based on various District Education Profile and BPS data. 36 Investing in Indonesia's Education at the District Level Chapter 4 District Education System and Outputs 4.2. Human Resources in Table 4.3. Student teacher ratios (STR) in the visited districts, 2006 Education: Teachers and Students Junior Senior District Primary secondary secondary The number of the students per classroom is Kab. Asahan 21 15 15 typically much lower than the national standard Kota Binjai 22 13 12 in most districts. At the primary level, the number of Kab. Wonosobo 19 17 13 students per classroom ranges from 16 to 32 students Kota Magelang 17 15 12 per classroom. This is lower than the ceiling of 30 to Kab. Minahasa 16 13 18 40 students per classroom set in the national standard. Kota Manado 19 12 17 The numbers are also lower in the junior and senior Kab. Belu 17 23 42 secondary levels, with the exception of Kab. Belu and Kab. Timtengsel 19 15 15 Kab. Jayawijaya. Kab. Belu has 44 and 47 students per Kab. Jayawijaya 24 21 16 classroom for the junior and senior secondary levels, Kab. Jayapura 15 12 12 respectively, while Kab. Jayawijaya has 48 students National 19 13 13 per classroom at the junior secondary level. average Source: World Bank staff calculations based on various District Student-teacher ratios (STRs) at the junior and Education Profile/Dinas Education and MoNE. senior secondary levels in many visited districts are above the national average. On average, the Despite adequate numbers of teachers in most sample districts have STRs of 19 at the primary, 16 at districts, there are highly unequal allocations of the junior secondary, and 17 at the senior secondary teachers across sub-districts. In Kab. Belu, the STR levels, while the national averages are 19, 13, and 13, varies from 9 to 40 at the junior secondary level and respectively. The highest STR at the primary level is from 17 to 54 in the senior secondary level. As for found in Kab. Jayawijaya in Papua. As for the junior the primary level, the STR ranges from 15 to 27 in and secondary levels, the highest STRs are found in Kota Manado and from 12 to 32 in Kab. Timtengsel. Kab. Belu, with 23 at the junior secondary level and 42 Looking at geographic location, sub-districts within at the senior secondary level. The World Development cities have relatively smaller variations in STR than in Indicators (WDI) 2006 show that low- and middle- rural districts. As for the visited districts taken overall, income countries on average have STRs of 31 at the the two districts in NTT have larger variations among primary level, and 24 for junior secondary level. Based their sub-districts than other districts in teacher on WDI, most of visited districts have below average distribution. Of particular concern is the senior STRs for low- and middle-income countries, indicating secondary level in the sub-districts in Kab. Wonosobo an adequate supply of teachers at the primary level. and Kab. Belu, where some students have no teacher. This could be because the school is located in another sub-districts due to the remoteness of certain sub- districts. Figure 4.2. STR for junior and senior secondary levels by sub-district in Kab. Belu, 2006 Sub-dist rict in Kab.Timtengse l Sub-dist rict in Kab.Tim tengse l Sub-dist rict in Kab.Tim tengse l (prima ry level) (junior secondary level) (senior secondary level) Nunkolo Lasiolat Kualin Mollo Utara Malaka Timur Kota Toianas Malaka Tengah Koto'olin Malaka Tengah Lamakanen Amanuban Timur Tasifeto Barat Polen Raihat Boking Tasifeto Barat Malaka Barat Amanuban Tengah Weliman Amanuban Barat Lamakanen Raimanuk Batuputih 0 5 10 15 20 25 30 35 0 5 10 15 20 25 30 35 40 45 0 10 20 30 40 50 60 Source: World Bank staff calculations based on various District Education Profile/Dinas Education. An Analysis of Regional Public Expenditure and Financial Management 37 Chapter 4 District Education System and Outputs Problems in teacher distribution can partly An over-supply of teachers will particularly affect be explained by the preference of teachers the education budget in terms of the composition for employment in urban areas, giving rise to of part-time and full-time employed teachers. shortages of teachers in rural areas. Adding The share of part-time teachers reaches 11 percent in more teachers will not solve the problem of teacher Kab. Minahasa and 28 percent in Kab. Wonosobo.8 distribution, while re-assessing the teacher/staff A high share of part-time teachers leads to higher placement might provide a better solution. The personnel costs given that part-time teachers are more required teacher supply at the primary level is expensive than full-time teachers on an hourly basis.9 calculated based on the number of classrooms. The national education PER (World Bank, 2007a) As mentioned in the minimum service standards, suggests that, in order to improve cost effectiveness, every primary school should have a minimum of six increasing certification of these teachers should be classroom teachers, one sports teacher, one religion encouraged with the aim of employing them on a teacher, and one school principal. On this basis, in full-time basis. six out of 10 of the visited districts there is an over- supply of teachers in primary schools, although an Teacher qualifications at the primary level often under-supply of teachers still exists within some sub- fail to fulfill Ministry Regulation No. 19/2005, districts. In addition, STRs at the primary level in most which requires teachers to have a bachelor districts are half the targeted STR of 40:1 indicated by degree. Most teachers at the primary level have only the minimum service standards. If more teachers are graduated from senior secondary school (SMA) or are added, this will need to be carefully planned by looking diploma II graduates. More than 75 percent of teachers at geographic location and taking into account other in the two districts in NTT and around 60 percent of circumstances that can affect student numbers over teachers in the two districts in North Sulawesi are high time. A mechanism to re-allocate teachers from over- school graduates. The district average shows that supplied areas to rural or under-supplied districts and only 12 percent of primary teachers have bachelor sub-districts should be encouraged, together with degrees. The share of teachers with bachelor degrees a mechanism to enable schools to regroup within a is higher at the junior and senior secondary levels. At specific geographic area. the junior secondary level, bachelor degree teachers Figure 4.3. Over- and under-supply of elementary teachers by district and sub-district, 2006 Districts % Sub -districts 120 1000 100 80 teachers 500 of 60 40 0 20 0 -500 -20 Over/unde-supply -1000 -40 -60 -1500 -80 -100 Belu Binjai Kab Kab Kota Kab Belu Kota Binjai Kab Kota Minahasa Manado Numfor Asahan Timtengsel Jayawijaya Wonosobo Kota Magelang Kab Timtengsel Wonosobo Manado Asahan Magelang Kab Kota Biak Kab % of sub-districts with under-supply teachers Kab Kab Kab Kota Kab % of sub-districts with over-supply teachers Source: World Bank staff calculations based on various District Education Profile/Dinas Education. 8 Part-time teacher is defined as teacher with teaching hour less than 24 hours per week based on minimum teaching hour requirement in Law No. 14/2005 on Teachers and Lecturers. Figure is estimated using Sakernas/BPS data. 9 See more on hourly based teachers' salaries in World Bank, 2007a. 38 Investing in Indonesia's Education at the District Level Chapter 4 District Education System and Outputs accounted for more than 70 percent in Kota Binjai and The GDS survey indicates that teacher Kab. Wonosobo. However, 33 percent of teachers absenteeism in sample districts is relatively low. at the junior secondary level in Kab. Timtengsel are Data in four visited districts also covered in the GDS only high school graduates. The share of bachelor survey show that the share of absent teachers to degree graduates is significantly higher at the senior total teachers in school is less than 10 percent. The secondary level, with an average 82 percent of lowest level is found in Kota Manado, with hardly teachers holding bachelor degrees. Districts with any absent teachers, while a relatively high level is the highest share of teachers having completed found in Kab. Jayawijaya, at 32 percent of primary graduate studies are found in Kota Magelang, Central teachers absent and 29 percent of junior secondary Java. Overall, no districts meet the standard service teachers absent during school time. Reasons for minimum in terms of teacher qualifications. Based on absence were given as being engaged in tasks outside Regulation No. 19/2005, 90 percent of their teachers school, sickness, on leave, or attending other private are required to have bachelor degrees, while the three business. The aggregate share of all sample districts in education levels in all visited districts are still below the GDS shows 6 and 3 percent of teachers absent at this requirement. Kab. Wonosobo was found to be the primary and junior secondary levels, respectively. the closest in meeting the requirement at the senior The survey on teacher deployment and employment, secondary level, where 89.4 percent of teachers have however, shows that teacher absenteeism rate is 19 at least a bachelor degree. percent at the national level. Figure 4.4. Teacher qualifications in the visited districts, 2006 4.3. Education Expenditure, Primary 100% Outputs, and Equity 80% 60% 40% 4.3.1. Education Outputs 20% 0% Asahan Kota Wonosobo Magelang Minahasa Manado Kab. Timtengsel Kab. Kota Binjai Kab. Kota Kab. Belu Kab. The primary school net enrollment rates (NER) in most of the visited districts are close to universal, with the exception of Kab. Jayawijaya. Kab. <=SLTA D-I D-II D-III S1 S2 Jayawijaya lags behind other districts with an average Junior enrollment rate of 72 percent in 2005. Enrollment 100% rates shows a relatively stable trend towards universal 80% primary school enrollment in most districts. Kab. 60% Timtengsel and Kab. Jayawijaya experienced the most 40% 20% significant improvements, while Kab. Jayawijaya still 0% Asahan Kota Wonosobo Magelang Minahasa Manado Kab. Timtengsel lags significantly. The current enrollment rate in Kab. Kab. Binjai Kab. Kota Kab. Kota Belu Kab. Jayawijaya means that it is one of only seven districts out of 451 districts in Indonesia with NERs of less <=SLTA D-I D-II D-III S1 S2 than 80 percent, all located in Papua. This shows that Senior while most districts in Indonesia are close to achieving universal enrollment at the primary level, many districts 100% in Papua still lag behind. 80% 60% 40% 20% 0% Asahan Kota Wonosobo Magelang Minahasa Manado Kab. Timtengsel Kab. Kota Binjai Kab. Kota Kab. Belu Kab. <=SLTA D-I D-II D-III S1 S2 Source: World Bank staff calculations based on various District Education Profile/Dinas Education. An Analysis of Regional Public Expenditure and Financial Management 39 Chapter 4 District Education System and Outputs Figure 4.5. Primary school net enrollment rates Only two out of 10 visited districts have relatively of the visited districts, 2002-06 high NERs at the senior secondary level. The expansion of student enrollment at the high school 100 level remains challenging for many districts. The cities 90 80 (Kota Magelang, Kota Binjai and Kota Manado) have 70 relatively higher NERs at the senior secondary level 60 than the rural districts. Nonetheless, enrollment rates 50 in the cities are still less than 75 percent of the senior 40 secondary school-aged population. In the more rural 30 districts, including Kab. Wonosobo, Central Java, 20 10 enrollment rates barely reach 40 percent of the senior 0 secondary school-aged population. Asahan Kota Wonosobo Kota Minahasa Kota Belu TTS Jayapura Jayawijaya Binjai Magelang Manado 2002 2004 2006 Figure 4.7. Senior secondary net enrollment Source: World Bank staff calculations based on Susenas/BPS data rates of the visited districts, 2002-05 100 In general, NERs at the junior secondary level are 90 lower than those at the primary level. Districts 80 with high NERs at the primary level also tend to 70 60 have high NERs at the junior secondary level, with 50 the exception of Wonosobo. Differences in NERs 40 at the primary and junior secondary levels are more 30 20 significant in kabupaten than city areas. The three 10 visited cities have less than 10 percent difference 0 Asahan Kota Binjai Wonosobo Magelang Minahasa Kota Kota Belu TTS Jayapura Jayawijaya Manado between the primary and junior secondary levels. This 2002 2004 2005 situation suggests a higher transition rate from primary to secondary schools in urban districts than in rural Source: World Bank staff calculations based on Susenas/BPS data. districts. Kab. Wonosobo in Central Java presents an interesting case: despite having a high primary NER, Districts with low enrollment rates do not its junior and senior secondary NERs are close to those necessarily have high drop-out rates. Drop-out found in Kab. Jayawijaya, Papua. This highlights the rates, defined as the percentage of children in each policy challenge of achieving nine years of mandatory school-age group who attend school but do not schooling in lagging districts. finish, are relatively low in all districts, even those districts where NERs are relatively low. For example, Figure 4.6. Junior secondary net enrollment Kab. Wonosobo, where the high school enrollment rates of the visited districts, 2002-06 rate has been low for several years, has a zero percent high school drop-out rate. Kab. Jayawijaya is similar, 100 with the lowest primary school NER but a drop-out 90 rate that is lower than four other districts. In the 80 case of these two districts, the low enrollment rates 70 60 are probably due to the high number of school-aged 50 children who have never attended school. Providing 40 greater access to school should be a priority in these districts. In the case of districts with higher enrollment 30 20 rates, keeping children in school needs to be prioritized 10 to avoid increasing drop-outs. Programs to provide 0 Asahan Kota WonosoboKota Minahasa Kota Belu TTS Jayapura Jayawijaya scholarship and attract youth back to school needs to Binjai Magelang Manado be strengthened and reflected in budget allocation. 2002 2004 2006 Source: World Bank staff calculations based on Susenas/BPS data. 40 Investing in Indonesia's Education at the District Level Chapter 4 District Education System and Outputs Figure 4.8. Primary school drop-out rates of the Kab. Jayawijaya, whereas in Kab. Minahasa and Kota visited districts, 2002-06 Manado the literacy rate almost reaches 100 percent. Unsurprisingly, districts that have recently seen an 5.5 5 increase in enrollment rates, such as Jayawijaya, face (%) 4.5 4 a challenge to improve literacy rates for their entire Rate 3.5 populations. Any improvement in the literacy rate Out 3 2.5 in these districts will require not only an expansion Drop 2 1.5 of the school system to increase NERs but also an 1 0.5 effort to reach the non-literate population who are 0 now beyond school age. The local government should Binjai Belu TTS Asahan include in their education budget priorities specific Kota Wonosobo Magelang Manado Minahasa Jayapura Jayawijaya programs that attempt to combat illiteracy and provide Kota Kota a second chance to those who have been missed the 2002 2004 2005 opportunity to enroll in formal education. Source: World Bank staff calculations based on Susenas/BPS data. Figure 4.10. Literacy rates in the visited districts, The mean number of years of schooling for 2001-06 the entire population has stagnated in some districts. These districts, such as Kab. Wonosobo, % 100.0 Kab. Timtengsel and Kab. Belu, are also districts with relatively low NERs above the primary school level. 80.0 A steady improvement in mean years of schooling 60.0 occurred in Kab. Asahan, Kab. Minahasa and Kab. 40.0 Jayawijaya, albeit from a low base. The trend in mean 20.0 years of schooling is in line with the NER trend at the 0.0 junior secondary level in these districts. Meanwhile, in Binjai Belu Kota Magelang and Kota Manado the mean number Asahan Kota Kota Timur Tengah Kota Wonosobo Magelang Minahasa Manado Jayapura Jayawijaya of years of schooling for the entire population has surpassed the number of years required to finish high 2004 2005 2006 school. Source: World Bank staff calculations based on Susenas/BPS data. Figure 4.9. Mean years of schooling in the visited districts, 2001-05 4.3.2. Expenditure and Equity in Education 12 10 Enrollment at the primary level is relatively 8 equal across income quintiles, although there are concerns that this is not the case at the higher 6 MYS levels. At the primary level, NERs for the two highest 4 income quintiles are close to 100 percent, while there 2 are slight variations among other income groups. 0 Kab. Kota Binjai Kab. Kota Kab. Kota Magelang, Kota Binjai and Kab. Minahasa have Kota Kab. Kab. Belu Jayapura Jayawijaya Asahan Wonosobo Magelang Minahasa Manado Timtengsel universal NERs at the primary level, even for the lowest income quintile. The variation among income groups 2001 2002 2003 2004 2005 increases at higher education levels. Junior secondary Source: World Bank staff calculations based on Susenas/BPS data. NERs for high income quintiles reach close to, and even exceed, 90 percent, while for the lowest income Despite the increase in enrollment rates, some group the average is 70 percent. The rate declines districts still face the challenge of improving to an average of 76 and 38 percent, respectively, the literacy rate for the whole population. The for lowest and highest income groups at the senior literacy rate is as low as 38 percent of the population in secondary level. An Analysis of Regional Public Expenditure and Financial Management 41 Chapter 4 District Education System and Outputs Lower enrollment in the lowest income quintile Although there is no significant correlation is one of the major causes of overall low NERs between education spending per capita in some districts. The enrollment rate in the and enrollment rates, the two variables are highest income quintile tends to be high even in nonetheless positively correlated, A relatively those districts with low NERs. In Kab. Jayawijaya and higher correlation can be seen between spending per Kab. Wonosobo, the low overall NERs at secondary capita in 2004 and NERs at the senior secondary level education levels are mostly due to the low enrollment in 2005, rather than at the junior secondary level. rates in the lower income quintiles. Increasing per capita spending on education may help to increase the net enrollment rate particularly at the Figure 4.11. Junior secondary school net secondary level. enrollment rates by income quintile, 2006 The relatively limited number of schools is 100.0 another characteristic of lower enrollment 90.0 districts. Around 60 percent of all villages in Kab. 80.0 70.0 Jayawijaya do not have a single primary school in the 60.0 village and the average distance from these villages 50.0 to the nearest school is 7km. At the secondary level 40.0 30.0 the average distance to the nearest school is 23km in 20.0 Kab. Jayawijaya, compared with 1km and 2km in Kota 10.0 Binjai and Kota Magelang, respectively. The distance 0.0 Kab. Kab. Kab. Kab. Kab. Kab. Timor Kab. Kota Kota Manado increases at higher school levels, reaching 32km in Asahan Belu Jayapura Jayawijaya Minahasa Tengah WonosoboMagelang Selatan Kab. Jayawijaya. Nevertheless, abundant numbers of Q1 Q2 Q3 Q4 Q5 schools do not necessarily lead to higher enrollments, Source: World Bank staff calculations based on Susenas/BPS data. as in the case of Kab. Wonosobo, where the distance to school at the junior and senior secondary levels is relatively shorter, but the NER is still low. Figure 4.12. Spending per capita, net enrollment rates at junior and senior secondary levels Senior Secondary Junior Secondary 900,000 900,000 04 800,000 800,000 04 700,000 700,000 capita 600,000 capita 600,000 per 500,000 per 500,000 exp 400,000 exp 400,000 300,000 300,000 Education 200,000 200,000 Education 100,000 100,000 0 0 0 20 40 60 80 100 20 30 40 50 60 70 80 90 100 Senior Secondary NER (%) Junior Secondary NER (%) Source: World Bank staff calculations based on SIKD/MoF and Susenas/BPS data. 42 Investing in Indonesia's Education at the District Level Chapter 4 District Education System and Outputs Table 4.4. Some indicators of school availability Village with no % of villages with no Avg distance to Avg distance to Districts elementary school elementary school school (km) for SMP school (km) for SMA Kab. Asahan 12 4.4 4 6 Kota Binjai 0 0 1 2 Kab. Wonosobo 0 0 3 6 Kota Magelang 1 7.1 2 1 Kab. Minasaha 11 5.7 3 19 Kota Manado 3 3.4 2 16 Kab. Belu 12 5.8 7 5 Kab. Timtengsel 3 1.3 10 4 Kab. Jayawijaya 221 60.4 23 36 Kab. Jayapura 38 28.8 19 32 Source: World Bank staff calculations based Podes/BPS data. Lower education shares (non-personnel The limited number of schools within some education spending) in rich districts is partly districts might be due to the limited budgets of explained by high private household expenditure district governments for the construction of new on education. Kab. Asahan and Kota Manado, two schools at the secondary level. Thus, although most rich districts, have private household expenditure district governments are allocating higher proportions on education of Rp 67 billion and Rp 59 billion, of their non-salary budgets for infrastructure and respectively, with shares of education excluding facilities, in reality they still depend to a large degree salaries of only around 2 percent of total expenditure. on central and provincial support to fund new school Meanwhile, poor districts, such as Kab. Jayawijaya projects. and Kab. Timtengsel, have only Rp 4 billion and Rp 7 billion in out-of-pocket education expenditure, and The spending pattern at districts suggests that education spending excluding salaries of 6 and 7 poor districts have lower per capita spending, percent of total expenditure, respectively. In addition, as well as lower per student spending.10 Lower richer districts also allocate a larger proportion of their per student spending is partly explained by the lower budgets to other major sectors, such as health and educational attainment and enrollment of students infrastructure, thus shifting their allocations away in the poor districts compared with rich districts. In from education. terms of share of expenditure allocated to education, there is only a slight difference between the poor and rich districts both for shares including and excluding personnel. Table 4.5. District expenditures per poverty quintile, 2005 District Per capita district Education expenditure Non-personnel Private HH out-of- quintile expenditure (Rp) per school student (Rp) education % tot exp pocket (Rp million) Poorest 858,296 1,097,255 6.99 26,067 2 807,164 941,367 5.02 43,121 3 595,605 1,048,785 5.38 62,138 4 896,367 1,402,839 5.01 66,642 Richest 912,590 1,574,280 4.68 114,132 Source: World Bank staff calculations based on SIKD/MoF and Susenas/BPS data. 10 High per capita spending in the Kab. Jayapura in Papua is more an exception than a regular pattern for poor district nationally. An Analysis of Regional Public Expenditure and Financial Management 43 Chapter 4 District Education System and Outputs Box 4.1. Minimum service standards in the education sector The manual of minimum service standards (MSS) for education was released by the MoNE in 2001, based on the Education Ministerial Decree No. 53/2001. The MSS provide technical specifications for regions on how to conduct education activities from kindergarten up to the secondary level. In general, the manual is divided into eight main components, which include the school curriculum, students, education personnel and teachers, facilities and infrastructure, organization, financing, school management, and community participation. Under every main component there is a set of indicators related to the minimum service standards. The responsibility for achieving these indicators is divided between central, provincial, district governments, and the schools. For example, on the `personnel and teachers' component, the indicator of availability of qualified teachers is under the authority of provinces, districts and schools. At the moment, some provinces, such as Central Java and Yogyakarta, have prepared their own MSS based on the manual provided by the MoNE. Four years after the MoNE released the MSS manual the government issued a new regulation, PP No. 19/2005 on the National Education Standards. The regulation follows Law No. 20/2003 on the National Education System and covers broader aspects of education standards. In 2007, another government regulation, PP No. 38/2007, was issued on the assignment of responsibilities across inter-governmental levels, including the education sector. Both government regulations emphasize the need to update the previous MSS manual. Updating the manual is particularly important to align the various regulations as well as to avoid contradictions. The alignment of contents is needed, for example, in the categorization of the main components for minimal standards, while avoiding contradictions is important in the financing responsibility. PP No. 38/2007 indicates that the financing of senior high schools is under the provincial responsibility, while the MSS manual indicates that senior high school financing is the responsibility of all government levels (central, provincial and district). Although there is no solid theory or practice that can be used as a basis for preparing the MSS, the government or the MoNE needs to take into account the following while updating and revising the MSS: the measurability of indicators and the accessibility of indicators. The issue of measurability of indicators particularly applied to some indicators such as work mechanisms, school performance, and administration discipline. These indicators require further elaboration for measuring the minimal assessment. Finally, accessibility of indicators relates mainly to the availability and the level of difficulty regarding data collection for the indicators. The most important issue to consider, however, is the usability of MSS. Currently, there is no formal reporting channel with regard to the fulfillment of MSS requirements of provinces, districts and schools. Government Regulation PP No. 19/2005 indicates that central and sub-national governments need to include MSS in their work programs. However, there is no strong enforcement nor are there penalties for those that do not comply or follow the MSS. A World Bank report (2005) found that the motivation for adopting MSS was often only to meet the bureaucratic reporting needs rather than accountability to citizens. Strong enforcement together with clear accountability mechanisms are needed in order to put the MSS into practice and to improve education service delivery through adherence to standards. 44 Investing in Indonesia's Education at the District Level Chapter 4 District Education System and Outputs 4.4. Efficiency at the District Level: of the boxplot in the graph below) is considerable for districts in the second and third quintiles of input Identifying Best Practice index. The top performer in each quintile generally corresponds to the districts at the best practice Results based on the so-called "best practice frontier. These districts, particularly the one in the frontier" analysis show wide variations in lower quintile of input index warrant further study of education sector performance across districts. their high performance despite resource constraints. This analysis attempts to develop a best practice frontier for those districts that maximize their outputs Figure 4.14. Distribution of output index by given a certain level of inputs, and hence it can also be input quintile interpreted as the level of efficiency of the education Quintiles of inputs index system in a district. The plotted line in Figure 4.13 represents the maximum level of output achieved for 4 any given level of input and is called the production frontier. Kab. Halmahera Tengah, Kota Tomohon, and index 2 Kota Kendari are among the best achieving districts in terms of using only low to medium levels of inputs, output Average of 0 while Kota Padang and Kota Yogyakarta score the highest given their medium to high level of inputs. -2 The 10 visited regions are shown as red dots in Figure Distribution 4.13. Kab. Biak Numfor, Kab. Minahasa, Kota Binjai, Kota Manado, and Kota Magelang perform relatively -4 well given their input levels. 1 2 3 4 5 Source: World Bank staff calculations based on Susenas/BPS and Figure 4.13. Best practice frontier of education districts' budget data. sector performance at the district level The "best practice frontier" was developed using 4 Kota KendariKota Padang Kota Yogyakarta the framework applied by Tandon (2005) and 3 Kota Tomohon Binjai Magelang the WHO (2005) for the health sector. "Outputs" 2 Halmahera Manado Tengah Minahasa and "inputs" in this regard are represented by the 1 respective output and input indexes. The index has Index 0 Asahan been constructed through factor analysis techniques, Output-1 where several variables with different underlying -2 dimensions are combined into a single composite -3 BeluWonosobo variable. The variables that constitute the input -4 Timtengsel index are: (i) realized education expenditure; (ii) -5 -6 -4 -2 0 2 4 6 8 10 12 average service area of schools (primary, junior high, Input Index and secondary, in km2); (iii) number of teachers per level of school; and (iv) the number of schools per Source: World Bank staff calculations based on Susenas/BPS and 1,000 students at each age group. The use of these districts' budget data. inputs is assumed to contribute to the production of Some districts seem to have problems several education outputs that are captured in the transforming higher levels of inputs into higher output index. These are: (i) net enrollment rates at levels of output achievement. Figure 4.14 shows the different school levels; (ii) adult literacy rates; and that the average output index for districts in the (iii) means years of schooling. The data used for the lowest quintile of the input index is higher than the current exercise are for 2005. average for districts in the second quintile, while the average output index for districts who are in the While it should be approached with caution, the fourth quintile of input index are almost equal to best practice frontier can be useful for several average for district in the third quintile. Variation in purposes, among others to identify districts output achievement (as demonstrated by the length that are outperforming others despite resource An Analysis of Regional Public Expenditure and Financial Management 45 Chapter 4 District Education System and Outputs constraints, or likewise identify districts that are lagging behind despite the high levels of inputs. In the analysis here, the second case seems to be more prominent. In general, higher levels of inputs correlate with higher levels of achievement. However, some districts seem to be less efficient than others. Kab. Asahan is less efficient than Kota Manado, given that the two districts are using almost similar levels of inputs, while Kab. Asahan obtains significantly fewer outputs. Kab. Wonosobo is also relatively inefficient compared with Kab. Timtengsel, because the district uses substantially higher levels of inputs but obtains similar levels of outputs as Kab. Timtengsel. Kab. Halmahera Tengah and Kota Sawahlunto are both interesting outliers, because at their minimal levels of inputs they achieve output levels higher than many other districts, including those with more substantial levels of inputs. Due to the limitation of this study, unfortunately the reasons for difference in performance of these districts have not been thoroughly examined in this study. Further case studies need to be conducted to understand the reasons behind why some districts achieved higher level of performance than other districts. 46 Investing in Indonesia's Education at the District Level Chapter 5. District Planning, Budgeting, and Monitoring for the Education Sector 5.1. Planning and Budgeting and the annual budget plans from schools (RAPBS). The sub-division for human resources is responsible for preparing the annual teacher needs assessment. The The regulatory framework for the planning sub-division for planning and program development and budgeting process in the education sector then compiles and refines these documents, and mandates concurrence between the process at produces a draft of the RENJA SKPD (Working Unit the district government level and the sectoral Annual Work Plan). The annual work plan is then level. As illustrated in Figure 5.1, the district education discussed during the annual development planning sector medium-term plan (Renstra Dinas) should be coordination meeting (Musrenbang) at the local level. derived from the local government medium-term This meeting should be attended by representatives development plan (RPJMD). The education annual from all local government working units and also working plan (Renja-SKPD) undergoes a two-way community representatives. The outcomes of the synchronization process with the district government annual meeting are an input for the final draft of the annual working plan (RKPD). The education annual district government annual work plan. The tentative working and budget plan (RKA-SKPD) is then budget ceiling and priorities (PPA) and a draft of the developed based on the budget ceiling and priorities general budget policy (KUA) are then discussed with (PPA) and is used to formulate district government the local parliament and jointly signed off. Based on annual budget (APBD). The final education budget the budget ceiling and priorities, the education Dinas document, the budget implementation document prepares the final budget and work plan (RKA SKPD) (DPA-SKPD) can only be formulated after the APBD for the coming year. The finance unit of the district is approved by both the district head/mayor and the government then compiles all RKA SKPD into a budget district/city parliament (DPRD). draft (RAPBD) document. The district parliament and government subsequently discuss this draft and, after In practice the annual planning and budgeting being finalized, the draft becomes the final budget routines in the education Dinas start when each (APBD), which has to be approved by the provincial sub-division prepares its annual plan. The plan government. The education Dinas then concludes by should be normally prepared using the input from the preparing a budget execution document (DPA SKPD). annual school survey, medium-term plan documents, An Analysis of Regional Public Expenditure and Financial Management 47 Chapter 5 District Planning, Budgeting, and Monitoring for the Education Sector Box 5.1. Lost in translation: How medium-term plans are translated intothe education sector plan Not long after the new mayor was elected in 2005, Kota `Pasir Putih' began to prepare a new medium-term plan for the period 2005-10. The plan calls for a bold new vision of "Pasir Putih as World Class Tourism City by 2010, towards a safe, competitive, prosperous, just, and dignified community". To achieve this vision, four strategic objectives were identified: (i) creating effective and efficient government and public service system; (ii) existence of tourism-based city spatial plan; (iii) developing infrastructure that meets international standards; and (iv) creating an enjoyable urban environment. To achieve these four priorities, the city government will follow 22 strategies and 12 sub-strategies. Education is part of a sub-strategy under strategy number four, which relates to the creation of effective public services. The strategy for the education sector reads as follows: "to provide broad access to education through subsidies to the poor population". Elsewhere in the plan under the heading of "Strategic Breakthrough Matrix", the city government formulated a matrix delineating goals, strategies, policies and programs for each Dinas. The plan sets out clear and quantifiable targets for the education Dinas, namely: (i) improve net enrollment rates for all levels of schools from 63.1 to 80.70; (ii) increase students' average grade for all school levels from 5.74 to 7.29; (iii) increase the absorption rate of secondary school graduates to industries; and (iv) implement school management systems based on national education standards. The matrix also specifies the different programs that complement these objectives. To achieve objective 1, the following programs are mentioned: BOS for SD, SMP and SMA; Retrieval and Community Schooling Program (KEJAR) for those who cannot attend formal schools; building new schools and classrooms; provision of school furniture; and scholarship programs. However, these relatively clear and quantifiable targets stated in the medium term plan seem to get lost in the education sector plan. While many of the programs mentioned in the city's plan are incorporated in the education sector plan, there seems to be a disconnect between these two plans. First, the matrix of objectives of the education Dinas does not follow the structure of the city's plan. Moreover, the first few pages of the education sector plan are more about programs to improve office administration. Second, the education sector plan does not contain specific quantifiable targets on enrollment or on average grades. Hence, instead of becoming a more detailed elaboration of the city's medium-term plan, the education sector plan seems to have pursued a different path. Unfortunately, the case of Kota `Pasir Putih' is not uncommon. The medium-term plan of Kabupaten `Rancak' (2004-2008) starts with a good analysis of problems in the education sector. It identifies improving enrollment figures and the quality of education as the two main priorities of the district government, followed by an analysis of the proximate causes of low enrollment and quality. The plan mentions the following problems: lack of access of the poorest group and distance to schools, uneven distribution of teachers between urban and rural areas. The plan also contains a matrix of programs that need to be implemented for the four-year period along with an indicative budget. The listed programs seem to be addressing the identified problems and, among others, entail training for teachers, scholarships for poor students, and the construction of new schools. Again, the relatively reasonable medium-term plan of the district does not seem to be followed by a clear medium-term plan from the education sector. The education sector plan does not have any quantitative targets and the program matrix is not as specific as was the one in the local government's medium-term plan. Even worse, the budget (APBD) that describes the district's actual activities does not seem to be aligned with the medium-term plan. For example, despite the fact that the medium-term plan emphasizes the uneven distribution of teachers between rural and urban areas as a priority area, for the three years of medium-term plan implementation (2004 to 2007) there appears to be no budget allocated for teacher incentives to work in remote areas. 48 Investing in Indonesia's Education at the District Level Chapter 5 District Planning, Budgeting, and Monitoring for the Education Sector Bothmedium-termandannualplanningprocesses In most districts visited, the Dinas coordinates seem to be mostly prepared by Dinas officials the budgeting processes for schools at the with little apparent consultation with other elementary level, while the budgeting processes stakeholders. In the visited districts, the medium- of schools vary between regions at the secondary term plan documents (RPJM/Renstra) for education level. In most districts, junior and senior secondary Dinas are usually publicly available. Interviews during schools are treated as independent working units the field visits further indicated that the preparation (SKPD) that manage their own budgetary processes of these documents is led and coordinated by the and individually submit their budgets and work planning and program development sub-division of plans (RKA-SKPD). Exceptions are Kab. Timtengsel, the education Dinas in consultation with other sub- Kab. Jayapura, and Kota Magelang, where the Dinas divisions in the Dinas.11 The main reference document still coordinates school budgeting processes at the for this exercise is the regional medium-term plan secondary level. (RPJMD/Renstrada). Education statistics, which are collected through annual school surveys, provide The current planning and budgeting practices another input for the planning process. By using in kabupaten leave little room for stakeholders these statistics, the education planner estimates the to participate. In current practices, the available need for additional teachers, classrooms, and other channels for the school and grassroots stakeholders school needs. Other documents that are used include to inform the planning and budgeting process are RIPS (Rencana Induk Pengembangan Sekolah, or the mostly through the annual school survey and, to some School's Master Plan), which have to be submitted by extent, the annual budgeting cycle via the annual schools to the Dinas. development coordination meeting (Musrenbang), where the work unit plan is discussed together with the Figure 5.1 The normative framework for annual results from bottom-up planning meetings. However, budgeting preparation process this forum is rarely used effectively for channeling the aspirations of sector specific stakeholders, such as LG/ TAPD SKPD school committees or teachers' associations. District level planning and budgeting often do take into account funding from central and RPJMD Renstra-SKPD provincial governments. Deconcentrated projects should use inputs from local Dinas in selecting the location in which to implement their programs. Then, the Dinas at the district level usually know in advance RKPD Renja-SKPD which programs or projects are likely to take place in their localities and can prioritize their spending to avoid overlap with other programs. District participation, however, does not include the selection of those KUA/ PPA RKA-SKPD programs that are most suitable for implementation in their districts, or the monitoring and evaluation of the programs. APBD DPA-SKPD Source: UU No. 25/2004 and PP No. 58/ 2005. 11 A typical structure of education Dinas consists of sections based on school level under its responsibility (SD, SMP, SMA/SMK), and several sections based on its functional responsibilities (human resources, planning and program development, and administration/finance). An Analysis of Regional Public Expenditure and Financial Management 49 Chapter 5 District Planning, Budgeting, and Monitoring for the Education Sector Box 5.2. School financial management and formula funding in Kota Magelang Kota Magelang is a small city with a population of only 129,854 people. The city's development theme is to become a `service city' for other districts in its vicinity. There is substantial interest from the surrounding districts in the education services provided by Kota Magelang, and this is one of the reasons behind the current development theme. In order to accommodate the increasing demand for education, as well as to attract more demand, improvements in education quality is thus perceived as necessary. The education Dinas in Kota Magelang has conducted several education programs with the assistance of donors, such as the Decentralized Social Service Delivery (DSSD) program and the Managing Basic Education (MBE) program. The DSSD is focused on improving the transparency of school budgets, and aims to produce a unified budget reporting format, while the MBE program is focused more on financial management that feeds into the formulating of funding formulas. The unified budget reporting standards coming from the project have now been made as a permanent procedure for schools in Kota Magelang. Schools have to provide a unified accountability report, based on their funding resources, to the Dinas. The resources mostly come from the government, the school's own businesses (i.e. cooperatives), and community contributions. Each school report is collected by the Dinas to be subsequently included into a unified report. Based on this report, the education Dinas, district governments, donors, as well as community, know the level of allocation and expenditure by levels of education. To encourage the submission of unified budget reporting by schools, the education Dinas provides certain school assistance subsidies only when they submit their reports according to the budget standards. The school subsidies are particularly intended to meet their objective in improving the education quality. The subsidy to school is allocated based on a specific funding formula developed under District Basic Education program. The formula takes into account a set of indicators such as number of students, number of poor students, students score, national examination result, teacher cost, etc. As a result of budget unified reporting, as well as funding formula program, which takes into account other factors (not just based on student numbers), the Dinas will be able to identify the actual school needs and assists in a more equitable funds distribution. 5.2. Monitoring and Evaluation Apart from regular monitoring activities, districts also perform random monitoring on activities financed by the APBD. Monitoring Annual school surveys and school visits by tends to be unsystematic and performed randomly superintendents are the two mechanisms for without a robust sampling methodology. Apart from Dinas to collect information on schools. These DAK financed activities, which have relatively clear constitute the backbone of the school performance guidelines on who and what should be monitored, monitoring system at the district level. Table monitoring of APBD-financed activities is usually 5.1 summarizes the main features of these two limited to the regular reporting from schools on the mechanisms. Kota Binjai is the only visited district use of funds. Reporting on activities financed by where the district Dinas has attempted to design its deconcentrated expenditures is done directly from own monitoring instrument, which it plans to roll schools to the program management unit (PMU) at out in 2007. The information-collection instrument the provincial level. There is no obligation for schools used in Binjai is similar to the MoNE-designed annual to copy the report to the Dinas at the district level. school survey. However, in Binjai, the education Dinas In almost all districts visited, there was no attempt attempts to obtain more detail on some questions, to consolidate the deconcentrated projects into one such as the number of students per classroom, single report. completeness of administrative documents, and the condition of each type of facilities. 50 Investing in Indonesia's Education at the District Level Chapter 5 District Planning, Budgeting, and Monitoring for the Education Sector Table 5.1. School monitoring mechanism Mechanism Instrument Information Collected Main use of Information Annual School MoNE-designed ·Student condition: enrollment, ·Input into annual planning Survey questionnaire, uniform across drop out, no. of graduating, no. of and budgeting process districts in Indonesia. continuing to higher level. ·Estimate of annual teacher ·Teacher condition: by level of needs education, by taught subject. ·Annual education statistics · Classroom condition. (profil pendidikan) · Amount and sources of fund received. School Visit Visit by school superintendent ·Spot check on quality of learning, ·On the spot problem (pengawas) ·Current issues faced by schools. solving ·Input into budgeting proposal Source: Various interviews during the field visits. Systematic evaluations on the efficacy of APBD- 5.3. Financial Management financed activities do not appear to exist. No district visited has ever conducted an efficacy study Assessment in Papua on its APBD financed programs. Hence, decisions regarding the inclusion of activities in the annual The capacity of district governments to manage budget and the medium-term plan are not based on local public finance was assessed through the systematic evaluations. Public Financial Management (PFM) Survey. The surveywasappliedtosub-nationalgovernmentsoverall Dinas are obliged to submit a performance and not specifically to education Dinas. Although not accountability report (Lakip), but there is no directly linked with the education Dinas, the results clear framework on how to follow up on this do reflect to some extent the performance in PFM of report. Based on Presidential Decree No. 7/1999, local officials, including Dinas officials. any government department, including the Dinas both at the province and district levels, is obliged The PFM assessment of district governments to prepare an annual performance accountability and the provincial government in Papua was report (Lakip), to be submitted to the State Ministry conducted through the revised PFM assessment of State Administrative Reform (Menpan). Almost framework. The original version was developed by all districts visited prepared this report annually, with DG BAKD (Bina Administrasi Keuangan Daerah/ Local the exception of Biak Numfor. Generally, the report Government Financial Administration Guidance) of outlines all activities under the management of the the MoHA and the World Bank in 2005 and has been Dinas, including the amount of money spent from all applied to around 50 sub-national governments. The sources (including the APBN), the number of outputs PFM framework is divided into nine strategic areas: achieved, and a comparison with the targets planned. (i) local regulatory framework; (ii) planning and As with budget performance indicators, there are cases budgeting; (iii) cash management; (iv) procurement; where inputs, such as the amount of money disbursed, (v) accounting and reporting; (vi) internal audit; (vii) are being reported as outputs of activity performed. debt and investment; (viii) asset management; and (ix) Once the reports are submitted, there seems to be no external audit and monitoring. clear framework on how they should be used or how they can feed into future policy planning. An Analysis of Regional Public Expenditure and Financial Management 51 Chapter 5 District Planning, Budgeting, and Monitoring for the Education Sector Figure 5.2. Overall PFM scores for all assessed Figure 5.3. Average scores for each strategic sub-national governments area 100.00% Overall LOCAL REGULATORY FRAMEWORK 90.00% W/o procurement 100.00% EXTERNAL AUDIT AND 80.00% MONITORING 80.00% PLANNING AND BUDGETING 60.00% 70.00% 40.00% 60.00% ASSET MANAGEMENT 20.00% CASH MANAGEMENT 0.00% 50.00% 40.00% DEBT AND PUBLIC INVESTMENT PROCUREMENT 30.00% 20.00% INTERNAL AUDIT ACCOUNTING AND REPORTING 10.00% Source: PFM Survey, 2007. 0.00% uke ura or Kab.Paniai Kab.Nabire Mimika. BintangJayawijaya eraKab Kab. M Papua Jayapura Kab.Peg Kab. Kab. Jayap. Low capacity in financial units and working Kab Biak-NumfProv. Kota units of sub-national governments lead to poor Source: PFM Survey, 2007. accountingandreporting. Thefinancialunitsinmost districts struggled to cope with the various regulations The capacity of district government and the on local financial management enacted during the provincial governments in Papua to manage local past five years that affected the budget structures public finance ranges from poor to acceptable. of sub-national governments. Decentralization of As shown in Figure 5.2, sub-national governments some financial management functions to district located around the capital (Prov. Papua, Kota Jayapura government working units (SKPD) and the additional and Kab. Jayapura) score significantly better than those tasks specified in Ministry Decree No. 13/ 2006 to in the highlands or landlocked regions (Kab. Paniai, provide cash management and financial reporting to Kab. Jayawijaya, and Kab. Peg. Bintang). On average, each SKPD further complicated the situation. While planning and budgeting, accounting and reporting, the objective of the decree is to reduce the burden, it and external audit and monitoring are the weakest does not envisage that these additional tasks require areas, while procurement and internal audits are the increased capacity not readily available to district strongest. Figure 5.3 shows that other than those two governments. As a result, staff in the financial units strongest areas, no area has an average score of more now need to guide, monitor and advise financial than 50 percent. It should be noted, however, that officers in the SKPD on how to perform their tasks. the procurement area cannot be assessed effectively due to the ad hoc nature of procurement committees and an inability to capture the "behind the scenes" activities. The low score in planning and budgeting was mainly caused by incomplete and weak planning documents, as well as cluttered budgeting processes. Law No. 25/2004 specified that districts should have a medium-term development plan (RPJMD) and an annual district working plan (RKPD). In addition, each working unit (SKPD) of the district needs to have a strategic plan (Renstra) and an annual SKPD working plan (Renja). While the RPJMD and RKPD are available, only a few SKPDs within the region surveyed that have either a Renstra or a Renja. There are inconsistencies among the various planning documents. For example, the Renstra and, to a lesser extent, the RKPD are often not aligned with the RPJMD. 52 Investing in Indonesia's Education at the District Level Chapter 6. Key Findings and Policy Options Regarding the SWA 1) The Legal Framework and · Also, in regions where such offices exist, central fundsarestillchanneledthroughtherepresentative Flow of Funds for the Education offices of the central government. Sector · Complicating the flow of funds is the fact that resources allocated to education by provincial governments are not always under the management of the education Dinas, because Key Findings: they are classified as "social assistance" and hence flow directly into district governments' accounts. Legal Framework · Law No. 38/2007 on functional assignments still lacks clarity on the roles and responsibilities of Key Policy Options: different levels of government. The obligatory and discretionary functions of each level of · Identifyingunclearareasinfunctionalassignments government remain vague. For example, in by developing a road map that provides a clear financing responsibility, whereas the district is framework on the responsibilities of central, responsible for providing education funds at provincial, and district governments, including the primary level, the central and provincial the appropriate contribution size of each level of governments can still contribute a significant government in education financing of districts. share for education financing at the district level. · The road map should also enable district · Legally, both central and district governments governments to have greater participation in are obliged to contribute towards (co-finance) the planning and execution of deconcentrated incentives for teachers in special area. However, spending in their districts. The aim is to provide in practice budget allocations or contributions them with adequate financing and managerial from each government level towards this item autonomy to execute their functional assignments remain unclear. These incentives are particularly (based on the principle that money follows important in attracting teachers to remote areas. function). Flow of Funds · Consistency in terms of the channeling · The central government still spends a significant mechanism of education spending from province share of total education expenditures -- to district. In addition, the correct categorization particularly rehabilitation and development of education spending, which is supposed to be spending -- potentially limiting the policy-making recorded under the education sector, is necessary ability of the districts. in order to monitor education fund flows to the An Analysis of Regional Public Expenditure and Financial Management 53 Chapter 6 Key Findings and Policy Options Regarding the SWA districts. Numerous fungibility issues may arise makes up the majority of funding at the due the issue of `misclassification'. school level. o The BOS program has had more impact 2) District Education Expenditures on quality enhancement than on access increasing investment. BOS funds are perceived as having more impact on Key Findings: the availability of books, stationery and school maintenance, as well as honoraria · Overall Spending for temporary teachers, than on helping Since decentralization, as expected most districts poor students. have increased their spending on education. · Out-of-Pocket Spending Since 2005, however, the share of spending on o Out-of-pocket spending remains education appears to be decreasing as revenues significant but has declined slightly. rise, possibly as a result of the BOS transfers The majority of household spending (hence through a substitution effect). is directed towards books, stationery · Economic Classification and transportation. Expenditures on Most district education spending is routine and enrollment and monthly school fees of this the largest part goes towards salaries. are relatively small, probably due to the In general, there is limited space for spending impact of the BOS program. on development. In most districts, there are negligible allocations from education spending Key Policy Options: towards maintenance, although many districts have classrooms in poor condition. · The central government needs to monitor the · Functional/Programmatic Analysis composition of local government expenditures o There is a clear emphasis on increasing and provide incentives/disincentives in order to access to facilities/infrastructure, as improve the budget composition. One of the opposed to investments more directly mechanisms to achieve this is by using the budget targeted at quality enhancements, composition of local governments as an indicator particularly at the secondary level of for deconcentrated spending targeting. education. · District governments should include an analysis o Districts appear to have only limited of unit costs in district planning and budgeting funds for management/operations. to identify school needs in order to cover their o Limited funds for improving teacher operational costs. An analysis of unit costs will quality almost certainly impair teacher then assist districts in planning and allocating capacity. their budgets appropriately by channeling funds This appears to be the case in some to schools that have difficulty in narrowing their of the analyzed districts and may budget gaps (the gap between available funds equally be an issue at the national and the operational costs of the school). level. · Strengthening the monitoring and evaluation of However, caution is required in the utilization of BOS funds is important if the interpretingthedata,asshortcomings program is to achieve its objective of widening are observed at the district level in access to education, particularly for the poor. The the classification of programmatic/ tendency for BOS funds to be utilized more for functional budgeting for improving teacher honoraria than for poor students could teacher quality. serve to undermine the primary objective of the Given this issue of inconsistent BOS program if not addressed. classification, determining expenditure by education level has proved problematic. · BOS Funds o The sustainability of the program is an important consideration given that BOS 54 Investing in Indonesia's Education at the District Level Chapter 6 Key Findings and Policy Options Regarding the SWA 3) Education Systems and Outputs outperformed others with higher input levels. These successful districts should be studied further. Key Findings: Key Policy Options: Infrastructure / Material Challenges · Infrastructure quality in most districts is poor, · Anationwidesystemofperformancemeasurement especially at the primary level. This may be the is required to ensure that local governments are result of limited funding for operations and accountable in developing education outcomes maintenance at the district level. · Initial observations also point towards shortages in their regions. Such a system would act as a balance between the full autonomy that is of essential learning materials in some districts, provided to districts in spending their funds and predominantly books. the outcomes they achieve with those resources. Number of Schools · The team observed an unequal distribution · Districts needs the capacity to assess the efficiency of their spending and identify key issues within of schools, especially in certain sub-districts. the education sector. This will provide important However, with the exception of some regions inputs for their planning and budgeting process. of eastern Indonesia, the number of schools To this end, technical assistance will help districts overall appears sufficient. In order to provide a to improve their capacity in planning and comprehensive overview of the situation analyses budgeting. should also include transportation modes, as · Localgovernmentsneedtoplanteacherallocation these are important in determining the size of and employment effectively, particularly with school service areas. regard to the issue of unequal distribution of Teacher Distribution · Similar unequal teacher distributional issues were teachers. Local governments should also link their teacher allocation planning with the incentives observed, particularly at the sub-district level. program as required by Law No. 14/2005 on Teacher Qualification · Most teacher qualifications are still not up to Teachers and Lecturers. the standards/level set in Law No. 14/2005 on Teachers and Lecturers. It is important to clarify which level of government will be responsible for 4) Planning, Budgeting, and teacher certification costs. Monitoring for Education at the Enrollment · Despite almost universal coverage at the primary District Level level, significant challenges remain in certain districts, particularly in remote areas, such as Papua. Key Findings: · Secondary level enrollment is still problematic and it is important to increase the transition rate Planning and Budgeting between primary and secondary education. · The regulatory framework for planning and · For people in the lower income quintiles access to budgeting in the education sector mandates education is still a problem, with transportation concurrence between the process at the local and education fees continuing to act as barriers government and the sectoral levels. to access. · Both the medium-term and annual planning · Literacy rates are still low in certain districts processes seem to be prepared by the Dinas with and there may be a need for non-formal adult limited consultation with other stakeholders. education in some areas. · District planning and budgeting processes do try Best Practice (Districts Performance) to take into account funding from central and · Districts with a higher level of inputs do provincial governments. not necessarily see higher levels of output · Most districts share a common process in that achievement. the Dinas coordinate the budgeting process for · Some districts with lower input levels have primary schools. However, for junior and senior An Analysis of Regional Public Expenditure and Financial Management 55 Chapter 6 Key Findings and Policy Options Regarding the SWA secondary schools the process is different since most districts treat them as independent working units that manage their own budgetary processes and individually submit their budgets and work plans. Monitoring · Annual school surveys and school visits by superintendents are the two mechanisms through which the Dinas collect information and monitor schools: o Apart from regular monitoring activities, districts also perform random monitoring on activities financed by the APBD. o Systematic evaluations of the efficacy of APBD-financed activities do not appear to exist. o The Dinas are obliged to submit performance accountability reports, although there is no clear framework on how to use these reports for subsequent policy improvement. Key Policy Options: · The absence of a coherent planning and budgeting process is not only caused by the lack of capacity of local governments or the absence of a regulatory framework linking planning and budgeting. The dearth of a planning culture and the lack of performance accountability are also likely contributors to the problem. Hence, in addition to improving capacity in planning and budgeting, changing the bureaucratic culture and instigating an appropriate performance accountability system are also likely to prove necessary. · Any capacity-building activities at any level should work in line with the existing system and budget cycle. Capacity-building activities need to be carefully planned in order to enhance the current system and avoid conflicting with and duplicating it. 56 Investing in Indonesia's Education at the District Level Annexes Figure 1. Total district revenue and education spending, 2005 600,000 500,000 million) (Rp 400,000 2005 300,000 Spending 200,000 Education 100,000 0 0 500,000 1,000,000 1,500,000 2,000,000 Revenue 2005 (Rp million) Source: World Bank staff calculations based on SIKD/MoF Table 1. Education spending by central government (decon), and local governments, and household out of pocket, 2005 Percent Districts Deconcentration LGs HH out-of-pocket Total Kab. Asahan 0.00 72.37 27.63 100 Kota Binjai 0.17 60.59 39.25 100 Kab. Wonosobo 0.33 78.76 20.91 100 Kota Magelang 0.53 73.78 25.68 100 Kab. Minahasa 2.03 82.58 15.39 100 Kota Menado 41.82 38.22 19.96 100 Kab. Timtengsel 0.62 93.45 5.92 100 Kab. Belu 0.57 81.83 17.60 100 Kab. Jayawijaya 0.06 92.46 7.48 100 Kab. Jayapura 1.03 94.99 3.98 100 Source: World Bank staff calculations based on SIKD/MF and districts' APBD data. An Analysis of Regional Public Expenditure and Financial Management 57 Annexes Table 2. Local government education expenditure, 2001-06 Rp million Districts 2001 2002 2003 2004 2005 2006 Kab. Asahan 148,449 141,451 159,201 159,857 140,340 151,315 Kota Binjai 28,538 45,766 54,088 53,024 48,077 51,255 Kab. Wonosobo 64,977 95,861 110,022 126,837 107,692 113,363 Kota Magelang 42,338 42,272 59,321 52,865 50,595 52,406 Kab. Minahasa 177,156 199,877 223,540 131,045 9,839 86,274 Kota Menado 15,013 73,634 24,463 102,091 90,225 100,524 Kab. Timtengsel 90,332 76,522 79,424 54,970 84,233 93,819 Kab. Belu 60,723 57,298 69,531 70,916 67,973 75,077 Kab. Jayawijaya 79,064 75,040 73,266 50,237 40,915 77,456 Kab. Jayapura - 21,183 28,784 51,718 50,062 53,949 Note: Kota Manado data 2006 are based on planned budget. Source: World Bank staff calculations based on SIKD/MoF and districts APBD data (at constant 2002 prices). Table 3. Routine and development expenditure on education, 2004-06 Rp million 2004 2005 2006 Districts Routine Development Routine Development Routine Development Kab. Asahan 148,864 10,993 135,412 4,928 135,440 15,875 Kota Binjai 48,293 4,731 43,433 4,643 45,719 5,536 Kab. Wonosobo 106,853 19,985 97,634 10,057 99,516 13,848 Kota Magelang 46,375 6,489 43,704 6,890 49,352 3,054 Kab. Minahasa 126,863 4,182 70,706 9,133 73,138 13,135 Kota Menado 98,463 3,628 86,607 3,618 93,138 7,386 Kab. Timtengsel 46,375 8,594 69,267 14,965 72,209 21,610 Kab. Belu 61,845 9,071 59,704 8,269 63,238 11,839 Kab. Jayawijaya 49,161 1,076 37,121 3,793 41,172 13,574 Kab. Jayapura 47,629 4,089 40,077 9,985 40,003 13,946 Note: Kota Manado data 2006 are based on planned budget. Source: World Bank staff calculations based on SIKD/MoF and districts APBD data (at constant 2002 prices). 58 Investing in Indonesia's Education at the District Level Annexes Table 4. Economic classification of education routine spending, 2006 Rp million Districts Personnel Goods Maintenance Travel Others Financial Total Assistance Kab. Asahan 189,665 1,673 14 274 - - 191,626 Kota Binjai 62,375 1,726 539 45 - - 64,685 Kab. Wonosobo 131,879 4,097 548 226 - 4,049 140,799 Kota Magelang 68,896 649 150 129 - - 69,825 Kab. Minahasa 101,781 1,515 132 51 - - 103,479 Kota Menado 131,059 522 154 40 - - 131,775 Kab. Timtengsel 99,548 1,510 398 708 - - 102,164 Kab. Belu 85,047 2,885 370 1,169 - - 89,472 Ksb. Jayawijaya 55,868 1,747 525 112 - - 58,252 Kab. Jayapura 55,587 743 195 73 - - 56,598 Note: Kota Manado data 2006 are based on planned budget. Source: World Bank staff calculations based on SIKD/MoF and districts APBD data (at current prices). Table 5. Public and apparatus spending on education, 2005 and 2006 Rp million 2005 2006 Districts Apparatus Public Financial Total Apparatus Public Financial Total Assist. Assist. Kab. Asahan 135,536 4,758 46 140,340 135,774 15,540 - 151,315 Kota Binjai 43,469 4,607 - 48,077 46,050 5,205 - 51,255 Kab. Wonosobo - 105,119 2,573 107,692 - 110,501 2,862 113,363 Kota Magelang 5,729 44,866 - 50,595 7,182 45,223 - 52,406 Kab. Minahasa 1,639 78,200 - 79,839 2,665 83,609 - 86,274 Kota Menado - 90,225 - 90,225 - 100,524 - 100,524 Kab. Timtengsel 12,311 71,921 - 84,233 15,968 77,851 - 93,819 Kab. Belu 417 67,556 - 67,973 581 74,496 - 75,077 Kab. Jayawijaya 36,551 4,104 - 40,655 41,172 13,574 - 54,746 Kab. Jayapura 42,795 7,266 - 50,062 40,003 13,946 - 53,949 Note: Kota Manado data 2006 are based on planned budget. Source: World Bank staff calculations based on districts APBD data (at constant 2002 prices). An Analysis of Regional Public Expenditure and Financial Management 59 Annexes Table 6. NER, literacy rates, and mean years of schooling, 2005 NER (%) Literacy rate Mean years of Districts Primary Junior Senior (%) schooling Kab. Asahan 97.25 93.05 58.69 95.89 7.29 Kota Binjai 97.98 90.99 73.37 98.33 8.79 Kab. Wonosobo 98.21 66.26 31.69 87.53 6.02 Kota Magelang 99.54 93.94 72.39 95.99 9.25 Kab. Minahasa 96.13 87.87 64.26 99.19 8.35 Kota Menado 98.86 93.15 68.54 99.40 9.61 Kab. Timtengsel 95.00 66.07 33.41 79.18 6.39 Kab. Belu 93.34 70.74 44.04 79.60 6.16 Kab. Jayawijaya 96.68 93.25 69.56 96.39 7.75 Kab. Jayapura 78.96 61.89 32.97 38.34 6.18 Source: World Bank staff calculations based on SUSENAS/BPS data 2005. Table 7. Number of students, classroom condition, and number of schools, 2006 No of schools per 1,000 Students per classroom % badly damaged school-aged population (per Districts (%) classrooms level) Primary Junior Senior Primary Junior Senior Primary Junior Senior Kab. Asahan 29 38 41 29.5 4.1 3.9 5 2 1 Kota Binjai 32 37 27 4.8 2.6 3.5 5 3 2 Kab. Wonosobo 25 36 36 31.0 1.6 1.6 5 1 1 Kota Magelang 27 37 34 2.4 1.2 1.7 6 3 2 Kab. Minahasa 16 28 32 18.0 14.3 10.6 12 6 2 Kota Menado 26 34 31 21.9 3.3 1.2 6 4 2 Kab. Timtengsel 26 44 47 43.2 4.5 2.5 6 2 1 Kab. Belu 18 35 36 38.8 13.2 7.1 7 2 1 Kab. Jayawijaya 30 48 40 - 32.5 17.3 4 3 1 Kab. Jayapura 22 35 36 - 19.1 21.4 9 6 3 Source: World Bank staff calculations based on Districts Education Profile 2006. 60 Investing in Indonesia's Education at the District Level References Aran, Meltem (2007). Note on "Pro-Poor Targeting and the Effectiveness of Indonesia's Fuel Subsidy Reallocation Programs". Jakarta. Indonesia Ghozali, Abbas. "Analisis Sejarah Kebijakan, Penyelenggaraan, dan Kondisi Pendidikan Dasar serta Implikasinya pada Pendidikan Dasar Gratis". Makalah individual untuk studi Pendidikan Gratis yang diselenggarakan oleh BAPPENAS. Jakarta. Indonesia. Pemerintah Daerah Kabupaten Belu (2004). Rencana Strategis Kabupaten Belu 2004-2008 Pemerintah Daerah Kabupaten Belu (2004). Rencana Strategis Dinas Pendidikan Kabupaten Belu Periode 2004- 2008 Pemerintah Daerah Kabupaten Minahasa (2003). Rencana Stratejik Dinas Pendidikan dan Kebudayaan Kabupaten Minahasa Tahun 2003-2007 Pemerintah Daerah Kabupaten Timor Tengah Selatan (2004). Rencana Strategis Kabupaten Timor Tengah Selatan 2004-2008 Pemerintah Daerah Kabupaten Timor Tengah Selatan (2005). Rencana Strategis Dinas Pendidikan dan Kebudayaan Kabupaten Timor Tengah Selatan Tahun 2005-2009 Pemerintah Daerah Kota Binjai. Peraturan Walikota Binjai nomor 050-6525 tahun 2005 tentang Rencana Pembangunan Jangka Menengah Daerah Kota Binjai 2006-2010. Pemerintah Daerah Kota Magelang (2005). Peraturan Daerah nomor 9 tahun 2005 tentang Rencana Pembangunan Jangka Menengah Kota Magelang Tahun 2005-2010. PemerintahDaerahKotaManado(2005).PeraturanDaerahnomor04tahun2005tentangRencanaPembangunan Jangka Menengah Daerah Kota Manado 2005-2010. Pemerintah Daerah Kota Manado (2005). Matriks Program Lima Tahunan (RPJMD dan Renstra SKPD). Dinas Pendidikan Kota Manado Pemerintah Indonesia (2007). Peraturan Pemerintah Republik Indonesia nomor 38 tahun 2007 tentang Pembagian Urusan Pemerintahan antara Pemerintah, Pemerintahan Daerah Provinsi, dan Pemerintahan Daerah Kabupaten. An Analysis of Regional Public Expenditure and Financial Management 61 References Pemerintah Indonesia (2005). Peraturan Pemerintah Republik Indonesia nomor 19 tahun 2005 tentang Standar Nasional Pendidikan. Pemerintah Indonesia (2005). Undang-Undang Republik Indonesia nomor 14 tahun 2005 tentang Guru dan Dosen. Pemerintah Indonesia (2003). Undang-Undang Republik Indonesia nomor 20 tahun 2003 tentang Sistem Pendidikan Nasional. Pemerintah Indonesia (2002). Keputusan Menteri Dalam Negeri nomor 29 tahun 2002 tentang Pedoman Pengurusan, Pertanggungjawaban dan Pengawasan Keuangan dan Belanja Daerah, Pelaksanaan Tata Usaha Keuangan Daerah dan Penyusunan Perhitungan Anggaran Pendapatan dan Belanja Daerah. Pemerintah Indonesia (2004). Undang-Undang No. 32 tahun 2004 tentang Pemerintahan Daerah. Pemerintah Indonesia (2004). Undang-Undang No. 33 tahun 2004 tentang Perimbangan Keuangan antara Pemerintah Pusat dan Pemerintahan Daerah. Tandon, Ajay (2005), Measuring Eficiency of Macro Systems: An Application to Millennium Development Goal Attainment, Asian Development Review, Vol 22, no. 2, pp. 108-125 WHO (2005), " Sub National Health System Performance Assessment in Indonesia". processed World Health Organization, Geneva World Bank (2005). "Education in Indonesia: Managing the Transition to Decentralization", volume 1 ­ volume 3. Jakarta, Indonesia. WorldBank(2007a)."InvestinginIndonesia'sEducation:Allocation,Equity,andEfficiencyofPublicExpenditures". Jakarta, Indonesia. World Bank (2007b). Indonesia Public Expenditure Review. "Spending on Development: Making the Most of Indonesia's New Opportunities". Jakarta. Indonesia World Bank. "Teacher Employment and Deployment in Indonesia: Opportunities for Equity, Efficiency, and Quality Improvement." Jakarta. Indonesia. 62 Investing in Indonesia's Education at the District Level