67319 AL ASSISTANCE PROGRAM AND PUBLIC EXPENDITURE REVIEW 5 SOCIA THE WORLD BANK OFFICE JAKARTA Indonesia Stock Exchange Building Tower II/12th Floor Jl. Jend. Sudirman Kav. 52-53 Jakarta 12910 Tel: (6221) 5299-3000 Fax: (6221) 5299-3111 Website: www.worldbank.org/id THE WORLD BANK 1818 H Street N.W. Washington, D.C. 20433, U.S.A. Tel: (202) 458-1876 Fax: (202) 522-1557/1560 Website: www.worldbank.org Printed in February 2012. Designed by Hasbi Akhir (hasbi@aisukenet.com) Cover photograph provided by Ryca C. Rawung. Photographs on pages 5 and 9 provided by Rythia Afkar. Copyright protection and all other rights reserved. The Social Assistance Program and Public Expenditure Review policy notes 1 through 8 together comprise Volume 2 of Protecting Poor and Vulnerable Households in Indonesia report. Both the report and the policy notes are products of the World Bank. The findings, interpretations, and conclusions expressed herein do not necessarily reflect the views of the Board of Executive Directors of the World Bank or the Governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, 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 or acceptance of such boundaries. For any questions regarding this report, please contact Vivi Alatas (valatas@worldbank.org) or Jon Jellema (jjellema@worldbank.org). BANTUAN SISWA MISKIN CASH TRANSFERS FOR POOR STUDENTS SOCIAL ASSISTANCE PROGRAM AND PUBLIC EXPENDITURE REVIEW 5 Bantuan Siswa Miskin | Cash Transfers for Poor Students Table of Content TList of Abbreviations, Acronyms and Indonesian Terms 3 Executive Summary 4 1. Background 8 2. Objective, Program Size, and Benefit Adequacy 12 3. Targeting 18 4. Potential Impacts 21 5. Cost Effectiveness 25 6. Implementation 27 7. Public Financial Management and Sustainability 28 8. Summary and Recommendations 31 References 33 List of Figures Figure 1: Education Attainment by consumption quintile, 2000 and 2010 10 Figure 2: Education Expenditures by school level, 2009 11 Figure 3: BSM Expenditure and Coverage, 2008-2010 12 Figure 4: Inflation-adjusted BSM Benefit Levels, 2008-2010 14 Figure 5: BSM Flow of Funds in Kemdikbud 17 Figure 6: BSM Flow of Funds in Kemenag 17 Figure 7: BSM Coverage and Incidence by expenditure decile, 2009 20 Figure 8a: Education Expenditure versus BSM Benefits 21 Figure 8b: Education Expenditure (excl. transport) versus BSM Benefits 21 Figure 9: Education Cost Profile versus Education Benefits 22 Figure 10: Net Enrollment Rates by school level, 2000-2009 24 Figure 11: Budget Execution Ratio at Kemdikbud, 2008-2010 28 Figure 12: Budget Execution Ratio at Kemenag, 2008-2010 28 Figure 13: Budget Disbursement Rate at Kemdikbud, 2008-2010 29 Figure 14: Budget Disbursement Rate at Kemenag, 2008-2010 29 List of Tables Table 1: BSM Benefit Levels and Number of Beneficiaries by school level and type, 2008-2010 13 Table 2: BSM at a Glance 13 Table 3: BSM Expenditure Summary, 2008-2010 15 Table 4: BSM Expenditure by Ministry, Directorate General, and Directorate, 2008-2009 16 Table 5: Characteristics within Indonesian Populations, 2009 18 Table 6: Characteristics within BSM and non-BSM Populations, 2009 19 Table 7: Spending Efficiency Indicators, 2008-2010 26 Table 8: BSM Financial Sustainability under RPJM 30 2 List of Abbreviations, Acronyms and Indonesian Terms AFC Asian Financial Crisis Bidik Misi Beasiswa Pendidikan bagi Calon Mahasiswa Berprestasi (Scholarship program for prospective students with good academic achievement and coming from underprivileged families) BKM Bantuan Khusus Murid (Scholarships program in compensation for fuel subsidy decrease) BLT Bantuan Langsung Tunai (Unconditional cash transfer) BOS Bantuan Operational Sekolah (School Operational Grants) BPS Badan Pusat Statistik (Central Statistics Agency - Statistics Indonesia) BSM Bantuan Siswa Miskin (Cash transfer for poor students) CPI Consumer Price Index DG Directorate General Dinas Agama Local religious affairs agency Dinas Pendidikan Kemdikbud regional - level office GDP Gross Domestic Product Jamkesmas Jaminan Kesehatan Masyarakat (Health insurance scheme for the poor, formerly Askeskin) Kemdikbud Kementerian Pendidikan dan Kebudayaan (Ministry of Education and Culture, MOEC) Kemenag Kementerian Agama (Ministry of Religious Affairs, MORA) Kemenkeu Kementerian Keuangan (Ministry of Finance, MOF) KPPN Kantor Pelayanan Perbendaharaan Negara (State treasury service office) LHS Left hand side (of graph) M&E Monitoring and Evaluation MoU Memorandum of Understanding NER Net Enrollment Rates PFM Public Finance Management PKH Program Keluarga Harapan (Conditional cash transfer) Podes Potensi Desa (Survey of village potential) PPLS Pendataan Program Perlindungan Sosial (Data collection for targeting social protection programs) PT Pos Perseroan Terbatas Pos Indonesia (National post office system) Raskin Beras Miskin (Program for sale of subsidized rice to the poor) Rp Indonesian Rupiah RPJM Rencana Pembangunan Jangka Menengah (Medium-Term Development Plan, MTDP) Satker Satuan Kerja (Implementing Unit) SD Sekolah Dasar (Elementary school) SMA Sekolah Menengah Atas (Senior secondary school, now SMU) SMERU SMERU Research Institute SMP Sekolah Menengah Pertama (Junior secondary school) SMU Sekolah Menengah Umum (Senior secondary school, formerly SMA) SSN Social Safety Net Susenas Survei Sosio-Ekonomi Nasional (National Socio-Economic Survey) US$ United States Dollars 3 Executive Summary Primary enrollment rates in Indonesia are near 100 percent for all students, but students from poor and vulnerable households have a difficult time completing higher levels of education. Poor households most often have heads of households with primary school education or lower while rates of illiteracy among households below the poverty line are approximately double that of nonpoor households. For the most recent cohort (with completed elementary and secondary education) less than 50 percent of those from poor households who enrolled in first grade made it to junior secondary and less than 20 percent made it to senior secondary (compared to 90 and over 75 percent, respectively, of those from the richest 20 percent of households who enrolled in first grade). Poor households also state that if higher education has to be rationed in their household because it is too expensive, it is more often male children who will be sent and female children kept back. Real costs of education, especially at the senior secondary level, are on the rise and real education expenditures are rising fastest for poorer households. Between 2003 and 2009, the real (inflation-adjusted) cost of attending primary school was roughly constant, junior secondary cost fell slightly, and senior secondary cost rose, leaving the total cost of a 12-year education higher in real terms. Between 2006 and 2009, real expenditures on education rose 120 to 160 percent for the poorest fifth of Indonesian households while for the richest fifth real expenditures decreased slightly. The Bantuan Siswa Miskin (BSM) program, which provides cash transfers for school attendance, could provide the right solution for poor households facing rising costs and increased risk of low education. The BSM program provides transfers from central education agencies directly to students or schools in which students sit. Both Kementerian Pendidikan dan Kebudayaan (Ministry of Education and Culture, Kemdikbud) and Kementerian Agama (Ministry of Religious Affairs, Kemenag) have BSM transfers providing cash payments once enrollment, attendance and other criteria 4 have been verified. The amount of the transfers provided rises with the level of education, from Rp 360,000 for primary school to approximately Rp 1.2 million (per year) for a university student. BSM may have even larger society-wide benefits when it promotes an educated, curious, and productive workforce with social skills and characteristics - like healthy behaviors and aversion to crime - that are valuable public goods. BSM spending has risen rapidly and it now ranks as the third-largest household-based transfer (by central government expenditure) behind Raskin and Jamkesmas, the GOI’s in-kind food transfer and health fee waiver (respectively). BSM spending has increased in line with the rapid expansion in target beneficiaries. In 2010, Rp 3.6 trillion (around US$ 397 million) was spent on the program, equivalent to 4 percent of central government education expenditures. BSM accounts for 10 percent of all central government resources devoted to household-based social assistance (SA). Around half of BSM spending goes to primary and junior secondary school students, with the remaining going to senior high school students and university students at similar portion. The BSM program is actually 10 independently-run initiatives that together cover all levels of education (including vocational education) at secular and religious public schools. Unlike other household-based transfers, the BSM initiatives have neither a central coordinating unit nor a unified budget. Within each institution, separate units independently manage and execute initiatives for students from each level of schooling and for vocational education. The Kemenag-run BSM initiatives for university scholars are further fragmented by religious affiliation (of partner universities). In total, there are 10 BSM initiatives, each with its own manual, fund flow structure, and implementing procedures. There is little coordination between initiatives, even those located in the same institution. 5 Bantuan Siswa Miskin | Cash Transfers for Poor Students The BSM program is national in scope but reaches very few students overall and does a poor job of identifying poor students. In 2009 program coverage (through the senior secondary level) was still small at 2.3 percent of all 6 to 18 year olds in Indonesia. Overall coverage of enrolled children is 3 percent. However, students from the poorest 40 percent of households account for approximately half of all BSM scholarships (and half of all rupiah distributed through the BSM program) while the middle-class and richer households in the top 60 percent capture an equal 50 percent of all BSM scholarships. In other words, a rupiah from the BSM program is equally likely to end up the hands of a poor student as in the hands of a nonpoor student. BSM does not target those who are unfamiliar with the school system and its administrators. BSM initiatives typically identify potential scholarship recipients by soliciting nominations from schools and school committees. Students nominated must have already achieved consistent attendance and demonstrated “good behavior” confirmed by the principal. Recently enrolled students or prospective new entrants have very little chance of being selected; likewise, those who have not made themselves known to the principal are unlikely to be selected. Households cannot nominate their own children and there is currently no formal appeals process. BSM does not effectively address difficult and costly transition periods – between elementary and junior secondary, and again between junior and senior secondary – when the overwhelming majority of dropouts occur. Verification procedures plus slow rates of disbursement together mean that recipients typically receive a BSM transfer for their first schooling year only after their second schooling year has already started. The BSM program does not distribute scholarships for the final grades of any schooling level (the sixth year for elementary and the last semester of the third year for junior and senior secondary) while BSM is not integrated across primary, junior secondary, and senior secondary education providers or across the subdirectorates in Kemdikbud and Kemenag in charge of those levels of schooling. The result is that BSM transfers are not available for poor students as they make transitions to junior and senior secondary school and BSM scholarships for the first year in the post-transition, higher-level school are delayed until the following year. As education expenditures rise especially steeply between education levels and are always greater at the beginning of the registration period (at any school level), that BSM cannot facilitate expenditures during these period is doubly unfortunate. Support operations like monitoring and evaluation are partly delegated to schools, which are also the main agency in charge of delivering BSM funds to beneficiaries. This has led to weakness in non-benefit program operations. Within the 10 different BSM initiatives, very few budgeted funds are spent on socialization, monitoring and evaluation, and complaints, appeals, or grievances. Socialization is typically provided to the schools and school committees who are responsible for nominating the beneficiaries, but not directly to the beneficiaries or their parents, guardians, or caretakers. Though monitoring and evaluation instruments do exist, they do not provide quality information to higher levels of program management and they do not enter into the program reform and evaluation cycle. 6 7 1. Background While increased access to education has long been a policy objective in Indonesia, the poor remain behind the non-poor in education levels and outcomes. Increased access to education has long been an explicit goal of education policy in Indonesia… From shortly after independence, the Government of Indonesia (GOI) has been devoting resources to initiatives for expanding enrollment in primary, secondary, and tertiary enrollment for all citizens. In 1970, the gross primary enrollment rate was under 70%; by 1995, universal primary enrollment was the norm; by 2009, even gross secondary enrollment rates were topping 80 percent and a 9-year education was compulsory. Enrollment in tertiary education has risen from approximately 2000 students in 1945 to nearly 4 million in 2008. …and more recently, nationwide scholarship instruments and fee waivers have been introduced to increase education achievement among the poor and disadvantaged… During the Asian Financial Crisis (AFC) at the end of the 1990s, the GOI introduced a School Grants initiative that included funding for means-tested cash transfers (conditional upon school attendance) intended to cover school fees and related costs for households affected by the crisis.1 In the early part of the 2000s, another scholarship scheme for children from poor households was part of a 1 The School Grants program did not provide scholarships per se; the use of the conditional cash transfers was unmonitored, though it was intended that a recipient household would use those funds to keep their children in school. 8 package of spending reorientation away from fuel subsidies. In 2003, Education Law #20 explicitly obligated the Government to provide scholarships in order to ensure an equal-opportunity 9-year compulsory basic education for all citizens. In 2005, a new school grants progam – Bantuan Operasi Sekolah, or BOS – again had a mechanism for providing cash support directly to targeted households for non-tuition costs.2 Later in 2008 and 2009, the Bantuan Siswa Miskin (BSM) program was introduced to help poor students facing still-elevated education costs. …but poor households remain far behind rich households in education achievement. While gross primary school enrollments have been comfortably over 90 percent for at least the last decade, these good starts have not led to improvements in educational achievement beyond the primary level for poor households. Students from poorer households drop out in large numbers during the transition from primary school to junior secondary school and again during the transition from junior to secondary school – see Figure 1 below. In 2010, nearly 100 percent of 26 to 28 year olds had attained at least three years of basic education, while about 86 percent of the poorest and 98 percent of the richest 26 to 28 year olds attained at least six years of basic education. However, attainment rates of at least a 7th-grade education drop to 44 percent for poor households (and to 90 percent for rich households).3 Achievement gaps at the senior secondary level are larger: 2010 attainment rates of at least 10th-grade education are about 50 percentage points higher in the richest than in the poorest households.4 2 It has historically been an option for the BOS program to target own funds to poor households. However, it has not been mandatory that BOS funds are spent as transfers directly to households. 3 This gap in attainment rates between rich and poor students actually widened slightly from 2000 to 2010 even while more households from all quintiles were participating in all levels of education. In other words, most of the increase in education attainment rates is coming from the nonpoor deciles. 4 Gaps (between students from poor and rich households) in senior secondary attainment have also been rising: in 2000 for the same group (26 to 28 year olds), the gap in attainment rates of at least a 10th-grade education was only about 45 percentage points. The richest 40 percent of 26 to 28 year olds have increased their rates of attainment of at least 12 years of education by approximately 15 percentage points from 2000 to 2010 while similar rates in the poorest 20 percent of 26 to 28 year olds have only risen by 2 percentage points. 9 Bantuan Siswa Miskin | Cash Transfers for Poor Students Figure 1: Education 2010 Attainment by 100 consumption quintile, 90 2000 and 2010 80 Percent of 26-28 year olds 70 60 50 40 30 20 10 0 2000 100 90 80 Percent of 26-28 year olds 70 60 50 40 30 20 10 0 Sources and Notes: Susenas 2000, 2010 and World Bank staff calculations. A major difficulty for poor households is financial: secondary education costs, including opportunity costs, are significant. Figure 2 below shows that the costs of sending one child to enroll in secondary education is costly for all households: out of pocket costs (including transportation) range from Rp 1.8 million to Rp 2.6 million, with poorer households paying slightly less and richer households slightly more. As a percent of overall expenditure, secondary education can be prohibitively expensive for the poorest households – regular costs to send one child to senior secondary school consume approximately 20 percent of a poor household’s overall expenditure. For all households, registration fees and other school fees like tuition, as well as school committee fees, make up the bulk of the costs of the out of pocket costs of education. For poor households, not only do school charges represent a higher share of overall expenditure, but the opportunity costs of education – incurred as foregone income when a child attends school instead of working – will be relatively larger also, making secondary education doubly expensive for poor households. Government fee-reduction efforts notwithstanding, the total costs of education have risen in real terms between 2003 and 2009, and real education expenditures are rising much faster for poor than for nonpoor households. 10 Figure 2: Education average SMU Expenditures cost is 100% 3,000,000 of 2009 by school poverty line level, 2009 average SMP 2,500,000 cost is 76% of 2009 poverty line Rp, nominal 2,000,000 average SD cost is 47% of 2009 1,500,000 poverty line 1,000,000 500,000 - Source: Susenas 2009 BSM was designed in recognition of the lingering gaps in education indicators and the high costs of schooling. The BSM program delivers cash transfers to students who have demonstrated enrollment, attendance, and good behavior in school. Like the PKH conditional cash transfer (see “Social Assistance Progam and Public Expenditure Review 5: PKH” in this collection) BSM potentially addresses both the demand for education (by providing funds conditional upon verified attendance) and the cost of access. As it is targeted to students from poor households, it could be a powerful tool for reducing the education gap between poor and non-poor households. In addition, it could address a potential gender gap in secondary schooling specifically among poorer households, who say that when education has to be rationed because of high costs, it is often a boy who gets secondary education while his sister stays home or goes to work.5 This note assesses the operation and implementation of the BSM program to determine how well poor households are served. It provides quantitative analysis of the coverage, incidence, and average benefit levels of BSM to determine both the progressivity of the program’s targeting and the adequacy of benefit levels. Qualitative information on program delivery and program operations will also shed light on areas for reform. A review of the program’s design and operating guidelines will offer enhancements to the program’s effectiveness and relevance. An evidence-based appraisal of the BSM program can provide inputs to the GoI as it continues to try to achieve higher levels of education for all citizens as well as its Pro-Poor development initiatives and the Millenium Development challenges. 5 SMERU (2008) 11 Bantuan Siswa Miskin | Cash Transfers for Poor Students 2. Objective, Program Size, and Benefit Adequacy Bantuan Siswa Miskin is a centrally-funded cash transfer to current public school students from poor households intended for use on education fees and other non-fee costs of attending school. Introduced in 2008, BSM is a collection of tax-financed cash transfers to public-school students from poor households. BSM programs exist at all public schools across all levels of education and provide currently enrolled students from poor households, selected by school administrators, with an annual cash transfer in one lump-sum installment. The funds are intended for use on education fees and other non-fee costs of attending school, such as transportation to school and uniforms. By helping poor students meet these expenses BSM aims to provide opportunities for pupils from poor households to obtain higher-level education and to reduce dropouts. BSM programs are tax-financed by the central government and do not require any contributions or explicit cost-sharing on the part of beneficiaries, local governments, or schools. Figure 3: BSM 4,200 7.0 Target Beneficiaries (millions of students) Expenditure and 3,600 6.0 Coverage, 2008- Rp billions (nominal) 3,000 5.0 2010 2,400 4.0 1,800 3.0 1,200 2.0 600 1.0 0 0.0 2008 2009 2010 Expenditure (LHS) Target no. beneficiaries (RHS) Sources and Notes: Kemenkeu (MoF), Kemdikbud and Kemenag data. BSM programs assisted over 4.5 million poor students in 2009 across all levels of schooling. BSM benefits rise with education level. In its inaugural year (2008), BSM targeted around 3 million students across all levels of schooling; in 2010 official coverage was at nearly 6 million students (Table 1 and Figure 3). Benefit levels vary by level of schooling, ranging from Rp 360,000 (around US$ 36) for primary school students to Rp 1,200,000 (around US$ 120) for university students. The per-student benefit levels are similar to those provided by Indonesia’s pilot conditional cash transfer (PKH, Program Keluarga Harapan) and to unit costs used in the large-scale school operational assistance program (BOS, Bantuan Operasional Sekolah). 12 Table 1: BSM Number of Beneficiaries Benefits Amounts Benefit Levels (Rp) 2008 2009 2010 and Number of Beneficiaries by Kemdikbud 2,136,473 2,988,628 4,123,204 school level and Primary School 898,400 1,796,800 2,277,039 360,000 type, 2008-2010 Junior High School 499,105 523,667 591,129 531,000 Senior High School 732,620 577,791 613,967 780,000 University 6,348 90,370 641,069 1,200,000 Kemenag 886,411 1,571,873 1,820,360 Primary School 358,492 645,556 714,642 360,000 Junior High School 274,027 544,861 645,033 720,000 Senior High School 204,922 316,282 382,903 780,000 University 48,970 65,175 77,781 1,200,000 Total 3,022,884 4,560,501 5,943,564 Source: Program manual and World Bank staff estimates. Number of beneficiaries estimated from the ratio of social assistance budget to benefit amounts. The cash transfer for each school year is conditional upon completion of the academic year and is transferred to students in the subsequent school year (for this reason, students in the final year of school at each level are not eligible for the scholarship). Funds are channeled to students either directly through PT Pos bank accounts or delivered by school personnel. Table 2: Official name Bantuan Siswa Miskin (BSM) BSM at a Program type Cash transfer for poor students Glance Program Type and inaugural year (start/usage year) Permanent, tax-financed, 2008 Coverage National (100% provinces, 100% districts) Number of beneficiaries (2010) 5.9 million Official value of benefit Rp 360,000-1,200,000 depending on level of schooling Public expenditure (2010) Rp 3,607 billion (US$ 397 million) Administrative cost per recipient (2010) Rp 15,608 (US$ 1.56) Percent of poor 6-18 years old covered (year) 3.2% (2009) Key policy and executing agencies Kemdikbud, Kemenag Kemenag, Kemdikbud (verification of beneficiary lists), Key implementation agencies (role) Education service providers (targeting and eligibility, fund distribution) Support operations partners (role) PT Pos/appointed bank (fund distribution) Local Government participation Targeting, verification, socialization, monitoring and evaluation Sources and Notes: Kemenkeu, Susenas 2009, program manuals and World Bank staff calculations. 13 Bantuan Siswa Miskin | Cash Transfers for Poor Students BSM benefit levels are not adjusted for general or school-cost-specific inflation, resulting in noticeable erosion in value over time. Like most other cash transfer programs in Indonesia, benefit levels for the BSM cash transfers are not adjusted annually for inflation and have remained unchanged since the 2008 launch. The real value of BSM benefits has declined by 13 to 14 percent between 2008 and 2010 if adjusted using poverty basket inflation (see Figure 4 below). This is on top of substantial increases in education costs from primary to junior secondary to senior secondary in any given year; BSM benefits do increase across schooling levels, but at a much slower rate than the increase in the costs of schooling (see Section 4 below). In addition, there is evidence that expenditures on education for poor households have risen in real terms by 20 to 50 percent between 2006 and 2009, so BSM benefits are becoming less adequate in both real terms and in terms of education purchasing power. If left unaddressed, the failure to adjust for rising costs of living and the rising real cost of education could undermine BSM’s objectives. Figure 4: 1,500,000 Inflation-adjusted Rp, constant 2009 prices BSM Benefit 1,200,000 Levels, 2008-2010 900,000 600,000 300,000 - Sources and Notes: Kemenkeu, BPS, and World Bank staff calculations. BSM nominal benefits have been deflated using the “poverty basket” price index. BSM consumes less than 5 percent of the central government’s budget for education but is the third largest household-targeted social assistance program in Indonesia. In 2010, Rp 3.6 trillion (around US$ 397 million) was spent on the program, which represented just 4 percent of central government education expenditures and a similar proportion of the combined budgets of Kemdikbud and Kemenag. However, as a social assistance scheme it is a relatively large program: BSM expenditures are third only to Raskin and Jamkesmas and accounted for 13.7 percent of all central government resources devoted to such programs.6 BSM in 2010 was roughly triple the size (by expenditure) of the conditional cash transfer, PKH, and it intended to reach over 7 times the number of beneficiaries. In its first year, senior secondary BSM accounted for the largest share of spending (Tables 3 and 4). However, in order to support a new government regulation (No.48/2008) mandating nine years of compulsory basic education, most of the expansion in 2009 was accounted for by increased resourcing for scholarships at the primary and junior secondary school levels. 6 Excluding expenditures on the GOI’s emergency and temporary unconditional cash transfer, BLT. BLT is not a continuous program and has been distributed twice, in 2005 to 2006 and again from 2008 to 2009. 14 Table 3: BSM 2008 2009 2010 Expenditure Total BSM (constant 2009 prices, Rp bn) 1,238 2,562 3,607 Summary, 2008- 2010 - Primary school level (SD) 274 875 1,077 - Junior secondary school level (SMP) 318 786 890 - Senior secondary school level (SMA) 600 718 778 - University level 46 183 863 Analytical series: Total BSM (Constant 2009 prices, Rp bn) 1,343 5,856 3,339 Total BSM (US$, Rp million) 127 247 397 Share of central education expenditures (%) 2.2 3.0 4.0 Share of combined Kemdikbud and Kemenag 2.1 3.0 expenditures (%) 4.1 Share of central government SA spending (%) 3.9 10.0 13.7 Memo items: Target number of beneficiaries (Million) 3.0 4.6 5.9 Average benefit level per beneficiary (Rp) 597,708 555,338 606,912 Sources and Notes: Kemdikbud and Kemenag records from Directorates General, Kemenkeu BPS, and World Bank staff calculations. Nominal expenditures have been deflated using the GDP deflator. irectorate General. Nominal spending is deflated using the GDP deflator. BSM consists of 10 independent initiatives with responsibilities and budgets delineated by type and level of education. BSM consists of 10 largely independent scholarship programs with neither a central coordinating unit nor a unified budget. BSM students attending secular public schools are managed by Kemdikbud, while BSM students attending religious public schools are managed by Kemenag. Kemdikbud initially accounted for nearly all BSM spending, but greater resources were allocated to Kemenag in 2009 (Table 4). Within both Kemdikbud and Kemenag, BSM budgets and administration are further fragmented by education level. For example, separate directorates within the Directorate General (DG) of Basic and Middle Education Management administers each BSM program for primary, junior secondary and senior secondary secular school students while university scholarships are administered by the DG for Higher Education. Within Kemenag, the program is fragmented by religion. Primary to senior high scholarships are administered by the Secretariat General with a majority of resources allocated to Muslim schools, while university scholarships are each administered by the DG representing each of the major religions (Muslim, Christian, Catholic, Hindu and Buddhist). This fragmentation makes it difficult to trace and connect the programs to each other, complicating analysis of efficiency and effectiveness. It also means eligible students may not keep BSM benefits as they progress from one level of schooling to another. For example, a primary school BSM beneficiary moving into junior high school must be renominated by the higher level nominating committee; there is no guarantee that higher-level committees coordinate with lower-level committees. This arrangement could undermine the programs’ objective to provide better access to education; see Section 6 below. 15 Bantuan Siswa Miskin | Cash Transfers for Poor Students Table 4: BSM 2008 2009 Expenditure by Kemdikbud (Rp billion, nominal) Ministry, Directorate DG of Basic and Middle Education Management 539 1,050 General, and Directorate, 2008- Directorate of Kindergarten and Primary School Development 254 654 2009 Directorate of Junior High School Development 286 396 DG of Middle Level of Education 569 478 Directorate of Senior High School Development (General and 569 478 Vocational) DG of Higher Education 8 105 Sub-Total Kemdikbud 1,116 1,633 Kemenag (Rp billion, nominal) Secretariat General 122 921 DG of Islamic Education 1 5 DG of Christian community guidance 0 1 DG of Catholic community guidance 0 1 DG of Hindu community guidance 0 1 DG of Buddha community guidance 0 0 Sub-Kemenag 122 929 Total 1,238 2,562 Sources and Notes: World Bank staff calculations based on Directorate General. Funds are channeled to students via three different methods, complicating monitoring (Figures 5 and 6). The first approach, employed by the Directorate of Kindergarten and Primary School Development within Kemdikbud (and also commonly used by cash transfer programs implemented by the Ministry of Social Affairs, Kemensos), is direct central execution with disbursement of funds through PT Pos: funds are released by the Treasury Office in Kemenkeu directly to PT Pos following the signing of a Memorandum of Understanding (MoU); PT Pos then distributes the funds directly to targeted students at the nearest PT Pos office within 40 days of receiving the funds. The second approach, employed by the directorates for junior and senior high school at Kemdikbud, is execution by local governments: funds are first distributed to provincial governments via the deconcentration fund mechanism, then on to district governments, and finally to schools who disburse the funds to students directly. The third approach, employed mainly by Kemenag7, is direct execution by schools: BSM funds are disbursed by local Treasury Offices directly to religious institutions (schools and universities), who in turn disburse the benefits directly to students. Having three different fund flow mechanisms complicates monitoring and evaluation by increasing the difficulty of tracing the flow of funds and monitoring the multiple actors and agencies involved across the 10 initiatives. 7 Kemenag’s approach partially reflects the fact that religious affairs are the sole prerogative of the central government, so local governments are not involved and there are no regional Religious Affairs Offices. 16 Figure 5: BSM Flow of Funds in Kemdikbud Kemdikbud Directorate General Directorate General of Primary and of Higher Education Middle Education Directorate Directorate Directorate of of Senior of Junior Directorates Primary School of Higher High School High School Development Development Development Education Province/ District Treasury Office Treasury Office Government (KPPN) (KPPN) University Local Treasury PT Pos Indonesia (Higher Education Office (KPPN) institution) Targeted Schools Paying PT Pos beneficiaries (Students) Targeted Targeted beneficiaries beneficiaries (Students) (Students) Figure 6: BSM Kemenag Flow of Funds in Kemenag Directorate General Directorate General of Directorate Directorate General of Secretariat Directorate of Hindu of Muslim Education Christian Community General of Catholic Buddhist Community General Community Community Province/Districts Province/Districts Religious Affairs Religious Affairs Office Office Local Treasury Local Treasury Office (KPPN) Office (KPPN) University Religious (Higher Education Schools institution) Targeted Targeted beneficiaries beneficiaries (Students) (Students) Sources and Notes: based on program manuals from Directorates General in Kemdikbud (Figure 5) and Kemenag (Figure 6). 17 Bantuan Siswa Miskin | Cash Transfers for Poor Students 3. Targeting BSM is intended for public and private school students from poor households, but many non-poor students end up with BSM transfers. BSM scholarships reach very few students overall and poor students are weakly prioritized (Table 5, Table 6 and Figure 6). BSM transfers are present in every province and district in Indonesia, but they cover only 2.3 percent of all children aged 6 to 18. For basic primary (SD), junior secondary (SMP) and senior secondary levels, coverage of students from the poorest 20 percent of households is 4.0, 3.4, and 1.7 while coverage of students from the richest 20 percent of households is 1.7, 1.9, and 1.2 percent, meaning scholarship coverage becomes less progressive at higher levels of schooling. Because benefit levels are higher for higher levels of schooling (see Table 1 above), this means the richest 60 percent of households have a slightly larger share of the total rupiah value of BSM transfers than do the poorest 40 percent. Total scholarships (rather than rupiah amounts) are split approximately equally between students from the poorest 40 percent and students from the richest 60 percent of households, meaning poor and vulnerable students and nonpoor students have about an equal change of receiving a BSM scholarship. Sections 4 and 6 below provide more details on the BSM selection process, which is done primarily by schools and sometimes in consultation with school committees, and which may lead to the relative weak prioritization of students from poor households.8 Table 5 demonstrates that households with BSM students are on average better off than average poor households. For example, access to safe water and sanitation as well as illiteracy and education are generally better (while agricultural employment is lower) for an average BSM-recipient household than for an average poor household. On the other hand, households with a BSM recipient are more often larger than an average poor household. This contrasts with the BLT emergency cash transfer program (see” Social Assistance Program and Public Expenditure Reveiew 2: BLT” in this collection) which on more frequently found households exhibiting the non-expenditure correlates of poverty. Table 5: % of all % of poor % of BSM Characteristics Indonesians who: population who: recipients who: within Do not have access to bottled, tap, or Indonesian 19 30 24 well water Populations, Do not have access to private 2009 sanitation 35 58 50 Live in villages without a primary 1 2 4 school Live in villages without a junior 36 42 38 secondary school Live in rural areas 52 63 62 Live with more than 5 household 27 43 59 members Have less than primary education 19 32 28 Are illiterate 9 15 9 Work in agriculture sector 38 59 49 Sources and Notes: Susenas 2009, Podes 2008, and World Bank staff calculations. In the BSM column and “Work in agriculture sector” row, the cell entry refers to the share of working heads of households with BSM students. “Work in …” refers to shares of working individuals, not all Indonesians. 8 Also see “Targeting Poor and Vulnerable Households in Indonesia” World Bank 2012a for a comparison of targeting procedures and outcomes in BSM and other Cluster 1 initiatives. 18 Table 6 demonstrates that poor households not receiving BSM (incorrectly excluded) are slightly better educated than poor households with a BSM recipient. Non-poor households receiving BSM (incorrectly included) have similar education levels to the excluded poor, but are less frequently rural, involved in informal jobs, or working in the agricultural sector. This implies that households in both rural and urban areas may be selected to receive scholarships based on characteristics other than income poverty alone. Table 6: Poor households Non-poor households Characteristics BSM non-BSM BSM non-BSM within BSM and non-BSM Household or Household head is/has: (% of households) Populations, Primary school or less 60 56 55 52 2009 Agricultural Sector 60 60 39 35 Formal Employee 45 46 61 65 Urban 31 29 48 50 (average among households) Household members 5.8 5.8 4.8 4.9 Child dependency ratio 1.1 1.1 0.8 0.8 Source: Susenas 2009 and World Bank calculations. Figure 6 demonstrates that while poorer households do have shares of total benefits averaging above 10 percent (per expenditure decile), shares among the three richest deciles average over 5 percent. BSM progressivity declines across schooling levels as well: by senior secondary level BSM coverage in the 9th expenditure decile (or the 2nd-richest 10 percent of households in Indonesia) is roughly equal to the shares of either the poorest or 2nd-poorest 10 percent of households. This is partially due to the design of the BSM program, which combines ten-independently run initiatives (see Section 2 above) with an absence of outreach to school-leavers, school committee beneficiary selection, (see Section 6 below), and a verification and payment system which both create disbursement delays which together mean that transitioning students cannot be served (see Sections 5, 6 and 7 below). Transitioning students from poor households drop out at very high rates; as a consequence by the senior secondary level there are very few enrolled students from eligible poor households left in the education system. 19 Bantuan Siswa Miskin | Cash Transfers for Poor Students Figure 7: BSM Coverage and Incidence by expenditure decile, 2009 Source: Susenas 2009 & World Bank staff calculations BSM is one of the few social assistance programs with a relatively larger presence in rural areas. While BSM coverage levels are not significantly different in rural and urban areas, rural areas have a share of total BSM benefits that is 25 percent greater than urban areas’ share. In the Eastern Indonesia provinces of Nusa Tenggara and Papua, the combined share of total BSM scholarships is greater than those provinces’ combined share of the total nationwide 6 to 18 year old population, meaning students from those provinces are more than proportionally represented in the BSM programs. BSM’s overall coverage is higher in areas with a relatively elevated number of SMU students from poor households, but this is likely due to BSM regional quota calculations, which are based on BPS lists of poor households, and does not necessarily lead to better prioritization of poor school-age children in such areas. 9 9 Susenas 2009 20 4. Potential Impacts BSM design features do not address the difficulties in achieving education that poor households face. Scholarship amounts are 30 percent of the costs of education… Figure 2 above shows education expenditures as high as 100 percent of the poverty line while Figure 1 above shows that poor households do not translate high initial enrollments into long careers in the education system. Figure 8a shows that official BSM amounts are only about one- third of total education expenditures at any level and also that as education costs rise from primary to junior secondary to senior secondary, scholarship amounts rise at a much slower rate. In contrast, the earliest scholarship schemes in Indonesia (circa the Asian Financial Crisis) provided cash transfers equal to 68, 56, and 37 percent of the overall out-of- pocket costs of education at the primary, junior secondary, and senior secondary levels respectively.10 Figure 8a: Education 3,000,000 Expenditure versus BSM Benefits 2,500,000 2,000,000 1,500,000 1,000,000 500,000 0 Source: Susenas 2009 & World Bank staff calculations Figure 8b: Education Expenditure (excl. 3,000,000 transport) versus BSM Benefits 2,500,000 2,000,000 1,500,000 1,000,000 500,000 0 Source: Susenas 2009 & World Bank staff calculations 10 Chen et al, 2009 (Table 2 pg 12 and Figure 5, pg 27.) 21 Bantuan Siswa Miskin | Cash Transfers for Poor Students …and are not delivered in amounts that account for increased entry costs. Figure 8 below shows that the costs of transition (from a lower school level to a higher school level) are significantly higher than the costs of continuation within the same level. Registration fees and other one-time induction fees are typically larger the first time a student enters the next-highest education level. Figure 8 also shows that households incur larger costs at the beginning of any school year (i.e., no matter the school level). BSM scholarships, which are actually delivered in one large tranche sometime after enrollment in the next year of schooling, do not take account of the one-time increase in costs at the beginning of a schooling year and the beginning of a schooling level. Finally, Figure 9 demonstrates that even an evenly-divided and quarterly-disbursed BSM scholarship would not be an appropriate solution; instead, benefits should be “front loaded” toward the beginning of the school year and total benefits should increase more sharply between school levels. Figure 9: 1,200,000 Education Cost Profile 1,000,000 versus Education 800,000 Benefits transition 600,000 transition 400,000 200,000 Sources and Notes: Susenas 2009 & World Bank staff calculations. BSM scholarship disbursements are usually delivered as one lump-sum payment, not in installments. In the figure above, a BSM disbursement is imagined to be divided evenly into four quarterly payments to better illustrate the difference between BSM amounts and household schooling expenditures at different points throughout the school year and across the regular transition path from primary to junior secondary to senior secondary. For additional detail on the timing and size mismatches in BSM, please see Section 6 below. The value of BSM amounts is eroding for three additional reasons.11 First, BSM amounts are fixed in nominal terms from year to year while Indonesian yearly price inflation has averaged 8.8 to 12.5 percent (as measured by the headline CPI and GDP deflators, respectively) from 2003 to 2009; nominal education costs have risen at yearly rates of over 27, 18, and 13 percent for primary, junior secondary, and senior secondary education (respectively) during the same period; and real education expenditures are rising much faster for poor than for nonpoor households.12 Secondly, the opportunity costs of education (measured as foregone wages and household production) rise as children acquire more productive physical capital (in larger and more agile bodies) and human capital (in more advanced “soft skills” like social understanding and reasoning abilities)13, so opportunity costs are rising through education levels. Finally, the BSM program is not necessarily delivering the entire stated benefit amount. For example, in 2009, 83 percent of students who reported receiving BSM scholarships reported receiving only 39 to 50 percent of stated benefit levels14, meaning de facto BSM transfers may be as low as 10 percent of average education costs in those schooling levels. Scholarship distribution is ultimately done by school-based committees. As for other social assistance schemes, a province-level quota of beneficiaries is determined by the implementing agency, in this case Kemdikbud and Kemenag. 11 BSM transfers do not depend on distance from school or other variable costs. 12 Average year-on-year inflation. CPI and the GDP deflator are taken World Development Indicators. Inflation in education costs are measured from household reports using the Susenas socioeconomic survey in 2003 and 2009. 13 Rising opportunity costs of education are indirectly confirmed by impact analysis of the BLT unconditional cash transfer (see “Social Assistance Program and Public Expenditure Review 2: BLT”, including references, in this collection). Overall, households receiving the poor-household-targeted BLT transfers had children who exited formal labor and participated in school at higher rates than similar households who did not receive the transfer. However, in urban areas, the BLT effect was erased (and even reversed): households with the program saw higher rates of child labor and lower rates of school participation. In Indonesia, the wage returns to an extra year of education are lower in urban areas than rural areas – see Indonesia Jobs Report (World Bank 2010) – meaning total opportunity costs of an extra year of education are higher for urban residents. 14 Susenas 2009 22 Province-based offices then allocate province-level quotas to districts. Finally, district offices allocate their district-level quotas to individual schools. Schools, school committees, and school principals are then responsible for allocating their quota among beneficiaries. Province and district-level quotas are designed to come from the national poor household registry maintained by BPS. However, district shares in the scholarship program do not necessarily match district shares of poor households. The nominating group (school committees, principals and others) is able to distribute the school’s allocated quota among a greater number of beneficiaries; add to, modify, or select certain of the criteria used by BPS to identify poor households to request greater allocations; and ignore the BPS-produced poor household registry because, for example, it excludes too many poor households with eligible children. In other words, program guidelines for targeting procedures are often modified and the resulting distribution of benefits does not closely track the distribution that would have occurred based on the original quotas alone. Selection by school committees and other design features emphasize currently enrolled students. Eligibility criteria for potential scholarship recipients includes verified attendance at school, a letter from the school principal verifying the student comes from a poor household, a nomination from the school principal, and evidence of good behavior as determined by the school principal. There are no guidelines and no explicit funding for outreach activities; no advance socialization to communities that precedes beneficiary selection; and no on-demand application system or appeals process that households may use to apply for scholarships directly. As a consequence, BSM selects students who are already attending school and who are already known by the school committee and most importantly, the principal. In other words, BSM distributes scholarships to students who are already exposed to the local school system and does not reach out to students with low levels of exposure. The BSM cash transfer is conditional on verified attendance in the previous calendar year, so funds delivery can lag enrollment by one year or more and scholarships are not applicable in transition years. Eligibility criteria state that primary-level BSM recipients must be verified to be a student in grade 1 to grade 5 attending at least 75 percent of the time; for secondary-level BSM recipients, the criteria are similar. BSM transfers actually distributed in any year are based on the previous year’s eligibility, so transfers received by a student sitting in 2nd grade are for verified attendance in 1st grade. This means BSM transfers cannot be used to pay the fees that are incurred when a student first enters the education system. Likewise, a 6th-grade student is not eligible any longer for a primary-level BSM transfer and 7th-grade student’s eligibility for the junior-secondary BSM program depends on verified 75 percent attendance in junior secondary. Therefore the last possible payments under the primary-level BSM program are during the 6th-grade year (for 5th grade attendance) and a student entering junior secondary cannot receive BSM transfers until he or she enters 8th grade. In practice, this creates gaps at the beginning and at the end of a schooling level, or precisely when poor households face the greatest difficulty in ensuring school-age children continue their studies. BSM transfers are not always routed directly to beneficiaries. Once lists of school-appointed beneficiaries have been compiled and approved by Kemdikbud, BSM transfers then proceed through the post office system down to the local level. The district post office is responsible for notifying regional education offices when they have received funds available for withdrawal; the regional offices are then responsible for notifying schools; and schools are responsible for notifying students. Upon notification funds can be withdrawn by students directly (provided they have a student ID card and a letter from their school verifying beneficiary status) or they can be withdrawn on behalf of the student (provided the school representative can present an authorization letter signed by the student). According to the manual, if funds are withdrawn by the school, the school has at most 12 days to make the final distribution to the student. Neither the frequency of school withdrawal of scholarship funds nor final distribution to students after school withdrawal is monitored by the scholarship program. Beneficiaries have no way to monitor their BSM transfers if they are delayed by any of the agencies involved. The available evidence indicates that the BSM program is not having large effects on enrollments. Figure 10 details net enrollment rates (NER) at the primary and secondary level from 2000 onward. It has been reported elsewhere that the AFC-era scholarship programs did allow households to maintain primary school enrolment15, and this is confirmed 15 However, two to three years after the crisis itself, the AFC-era scholarship program’s own internal monitoring unit found that 90 percent of parents with primary-school aged children claimed that their children would stay in school without scholarships. In addition, while primary school enrolments and attendance were protected, the AFC-era scholarship program did not increase the likelihood of transition from primary to secondary level. 23 Bantuan Siswa Miskin | Cash Transfers for Poor Students by the primary NER time series which has been roughly constant (above 90 percent) from 2000 to 2009. Junior and senior secondary NER improvement peaked in 2003 and has been roughly constant since even though the BOS, BKM, and later BSM schemes were introduced during this period. Figure 10: Net 100% Enrollment Rates by Percent of official age-group enrolled school level, 2000- (at corresponding school level) 2009 80% 60% SD SMP 40% SMU 20% 0% Source: Susenas, various years. 24 5. Cost Effectiveness Assessing the efficiency of spending on BSM programs as a whole is very challenging due to high levels of fragmentation and varied implementation. High-level budget data classifies virtually all BSM spending as social assistance expenditure, but these classifications exclude information on salary expenditure. Budget and management fragmentation and decentralized implementation (by schools) makes it difficult to track the multitude of administrative expenses, particularly when execution takes place at the level of the education provider. Third, each BSM initiative has different beneficiary numbers, benefit levels, coverage areas, flow of funds and implementation systems, making it difficult to quantify and compare administrative costs of each intervention or aggregate them consistently into total BSM spending. Efficiency indicators presented here should be interpreted with caution for these reasons. The intervention for non-religious primary school students exhibits extremely low administrative costs, possibly signaling underdeveloped administrative systems. Spending on the intervention for primary school students administered by Kemdikbud – which is the largest of the interventions, accounting for 30 percent of total program spending – was examined in detail. The analysis indicates an extremely low administrative overhead ratio of just 1 percent of total program expenditure during 2008 to 2010 (Table 7). Moreover, the rapid increase in the number of beneficiaries over three years of the program has resulted in a further decline in administrative costs per beneficiary to 52 US cents in 2010. These costs appear extremely low compared to other cash transfer interventions in Indonesia and by international standards.16 One explanation for the low administrative budget could be that schools perform most of the costly safeguarding activities like socialization and targeting. Additional support could also be provided by local government, for example according to Government Regulation # 47/2008 which specifies local government contribution to education initiatives. No matter what local governments have explicitly or implicitly been tasked with, low costs at the central level could signal an underdeveloped administrative and oversight architecture. For example, it appears that very little is spent on training or monitoring and evaluation activities, both of which could undermine the effectiveness of the program. 16 For example, Indonesia’s temporary but large scale unconditional cash transfer (BLT) is estimated to have had average administrative costs of US$ 5 per beneficiary (5 percent overhead) in 2008-09, while the pilot conditional cash transfer program (PKH) is estimated to have had average costs of US$ 21 per beneficiary (14 percent overhead) in 2009; see “Social Assistance Program and Public Expenditure Review 2: BLT” and“Social Assistance Program and Public Expenditure Review 6: PKH” in this collection. An international survey of 16 cash and near cash programs found that the administrative costs of well executed interventions cluster in range of 8 to 15 percent of total costs, with a mean of around 8 percent (Grosh et al, 2008). 25 Bantuan Siswa Miskin | Cash Transfers for Poor Students Table 7: Spending BSM at Kemdikbud, primary school level 2008 2009 2010 Efficiency Indicators, Unit Cost (Total spending/no. beneficiaries, Rp) 283,841 363,954 365,225 2008-2010 Administrative costs per beneficiary (Non-benefits/ 5,131 4,406 5,225 No. beneficiaries, Rp) in US$** 0.51 0.44 0.52 Administrative overhead ratio (Non-benefits/Total 2% 1% 1% spending) Cost of delivering benefits ratio (Non-benefits/ 2% 1% 1% Benefits) Memo items: No. of beneficiaries 898,400 1,796,800 2,247,222 Number of civil servants n.a. n.a. n.a. Value of annual scholarship for the poor (Rp) 360,000 360,000 360,000 Total spending (Rp bn)* 255 654 821 o/w benefits 250 647 809 o/w Non-benefits 5 8 12 o/w Civil servant salaries*** 1 1 1 o/w Admin/other 0 1 1 o/w Targeting 1 2 3 o/w Follow up 0 0 0 o/w Socialization 1 4 5 o/w Training 0 0 0 o/w Evaluation (M&E) 1 1 1 Source and Notes: World Bank staff calculations based on Directorate General of Treasury, Kemenkeu, Data. *While original budget data allocates personnel expenses to the DG as a whole, expenses for each individual Activity/ Directorate have been divided evenly. **Rp/US$ rate assumed to be 10,000. *** Salary costs are estimated using the total budget for salaries and honorarium under the same Satker (implementing agency) divided by total activities within the Satker. 26 6. Implementation Support operations like socialization and monitoring and evaluation are not designed to reliably indicate areas for program reform and improvement. Guidelines for socialization activities are not fully developed in BSM operations manuals. For example, even general guidelines for BSM socialization content, frequency and timing are not covered in program manuals. The structure of socialization activities is covered – provincial offices socialize to district offices while district offices then are responsible for socialization to sub-district offices and schools – but there is no monitoring of the activities actually carried out by various levels. Program manuals do not mention any socialization activities for selected beneficiaries, potential beneficiaries, poor households, communities, education advocates and other stakeholders. All together, the scant socialization that is carried out is not producing beneficiaries or communities knowledgeable enough to monitor the program from the bottom up or participate in either safeguarding or accessing the program. Monitoring and evaluation activities are regularly undertaken… For example, in 2009, the Kemdikbud BSM program completed its scheduled monitoring and evaluation activities in 130 districts in 33 provinces. In following years, a different set of districts was selected for such activities; the rotating schedule allows Kemdikbud to regularly visit all of Indonesia’s 498 districts. The purpose of the these activities, according to program manuals, is to ensure the program is effective for households and for implementing partners and that it achieves stated goals and meets targets. The directorate hosting the particular scholarship program, together with the province and district offices of Kemdikbud (Dinas Pendidikan), collects information and directly observes program areas such as the selection process, the resulting scholarship allocation, fund distribution realization, and the withdrawal process. Typically observation takes place at the school level and above and participants include the school principal, beneficiary students, the district-level education official responsible for handling the scholarship program, and post office officials. …but produce ineffective information that is not incorporated into a continuous program improvement cycle. For example, in 2009, reports confirmed that district education offices conducted a verification of chosen beneficiaries. They also stated that a majority of verified students met eligibility criteria as defined in the manual and that the majority of principals conducted the selection process according to the manual. Not included in the report was a discussion of whether the selection procedure as designed produced a pro-poor distribution of benefits or whether scholarship benefits reached the most vulnerable households. Similarly, in 2007 reports confirmed that 100 percent of post offices knew their BSM responsibilities, but the reports make no mention of how many were late in distributing payments nor what the cause of tardy benefit distribution was. See Protecting Poor and Vulnerable Households in Indonesia for more detail on monitoring and evaluation efforts in Indonesian ministries and agencies including Kemdikbud. Household or individual application as well as complaint and grievance reporting arrangements are not provided except indirectly through the school-based selection committees themselves. These processes are not mentioned in program manuals. The BSM program delegates selection to school-based committees and does not explicitly provide any formal way to challenge these selections; given program logic, therefore, the lack of an application or grievance system makes sense. However, it inadvertently provides those with access to or knowledge of the selection committee and its activities an advantage in suggesting recipients for BSM scholarships. Again, it is precisely these households who likely already have experience accessing and navigating education-provider systems. 27 Bantuan Siswa Miskin | Cash Transfers for Poor Students 7. Public Financial Management and Sustainability After weak budget execution in the first year a sharp improvement was seen in 2009. The first year of the scholarships program witnessed weak budget execution rates (Figures 11 and 12). The Kemenag BSM programs struggled in 2008, executing less than 20 percent of their total BSM budget. By 2009 BSM execution had improved to 100 percent (or over) in both agencies. This could be the result of different program implementation mechanisms. As noted above, because Kemenag has no local offices, it is required to monitor BSM implementation in each district without assistance from local government, which could significantly slow operation Figure 11: Budget Execution Ratio at Figure 12: Budget Execution Ratio at Kemenag, Kemdikbud, 2008-2010 2008-2010 120 120 100 100 80 80 60 60 40 40 20 20 0 0 2008 2009 2010 Scholarship at Primary Scholarship at Senior Total Program Scholarship at Junior Scholarship at University Sources and Notes: Kemenkeu DG of Treasury database and World Bank staff calculations. The fiscal year and school year do not coincide; this disrupts implementation and results in budget disbursement that is skewed towards the second half of the year, especially at Kemenag. BSM disbursement typically begins in June of each year and continues slowly over the following six months (Figures 13 and 14). This pattern primarily reflects the timing of the school year, which begins in June, while the fiscal year commences in January. In 2010 Kemdikbud improved disbursement, achieving nearly 40 percent disbursement by June with a further 37 percent of the budget disbursed during the final quarter of the year. Kemenag disbursement in 2010 was even later than in 2008: over 56 percent of total disbursed funds were released during the final quarter of the year. The bottom-up, school-initiated identification, selection, and approval of beneficiaries (which also starts in June) is time-consuming and leads to delays. This partly explains the skewed distribution profile, but cannot explain the notable differences between Kemdikbud and Kemenag performance as they rely on the same targeting procedures. Future budget implications of the BSM program are limited (Table 8). The RPJM 2010-14 outlines ambitious expansion plans for BSM with the target number of planned to roughly double from its 2009 level to almost 8 million students by 2011. This would be enough scholarships to potentially reach full coverage of school-aged children from poor households. After 2014, RPJM numbers for BSM gradually decline, presumably in line with an expected decline in poverty rates. If average unit costs remain unchanged then annual spending on BSM would rise from Rp 2.56 trillion in 2009 to around 4.1 trillion by 2014, keeping BSM shares of total central government spending unchanged (around 0.4- 0.5 percent). 28 Figure 13: Budget Disbursement Rate at Figure 14: Budget Disbursement Rate at Kemdikbud, 2008-2010 Kemenag, 2008-2010 Percent Percent 100 100 80 80 60 60 40 40 20 20 0 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Month - 2008 Month - 2009 Month - 2010 Cum - 2008 Cum - 2009 Cum - 2010 Sources and Notes: Kemenkeu DG of Treasury database and World Bank staff calculations Addressing inadequate benefit levels, non-indexation of benefits, poor targeting, undercoverage, and weak implementation in order to enhance the program’s impacts and effectiveness will require more resources. If (1) benefit levels are increased to bring them into line with real education expenses, are adjusted annually for inflation, and are revised to include both a transition bonus and a transportation allowance; (2) administrative resources for safeguarding are increased to around 8 percent of total costs; and (3) coverage levels are increased to 100 percent of poor students and 50 percent on near-poor students; then the current BSM budget would roughly quadruple to Rp 8 trillion 2015. Under this reform and expansion scenario, the share of 1st-grade enrollees from poor households expected to reach 12th grade (and beyond) would increase by over 50 percent, and BSM would consume approximately 9 percent of the combined Kemdibkbud and Kemenag budgets (from 3 percent currently). This scenario assumes that BSM scholarships positively affect transition rates for poor students, so more students are kept in the education system and continue receiving BSM scholarships each year, including transition years. 29 Bantuan Siswa Miskin | Cash Transfers for Poor Students Table 8: BSM 2009 2010 2011 2012 2013 2014 Financial Sustainability Baseline (RPJM) under RPJM Total budget allocation 2,562 3,607 4,132 4,151 4,150 4,130 (Nominal, Rp billion) Share of central government 0.4 0.5 0.5 0.5 0.4 0.4 spending (%) Average unit cost per 561,759 606,912 527,615 538,851 549,667 560,090 beneficiary (Rp) Target number of 4,560,501 5,943,564 7,830,735 7,703,654 7,550,092 7,374,249 beneficiaries - Primary school 2,442,356 2,991,681 4,556,220 4,280,780 4,010,200 3,743,210 - Junior secondary school 1,068,528 1,236,162 1,935,100 1,886,020 1,815,840 1,735,700 - Senior secondary school 894,073 996,870 1,212,374 1,409,813 1,595,011 1,765,298 - University 155,545 718,850 127,041 127,041 129,041 130,041 Share of potential target 52 70 96 98 99 101 population covered (%) Sources and Notes: RPJM 2010-14 and World Bank staff estimates and projections. 30 8. Summary and Recommendations BSM could be a valuable program for poor households and for Indonesia generally if it were better targeted, better socialized, and benefit packages were revised to correspond to the risks that poor students face. Poor households are not translating primary school enrolments into frequent success in higher levels of education. Education costs, especially for senior secondary, are rising in real terms. Poor households in particular are facing the biggest increases in real education expenditures (at all schooling levels). By providing benefits adequate for meeting the real costs of education precisely when those higher costs arrive, the BSM program could serve an important risk-mitigation function and inducement to higher education for poor households. Unfortunately the BSM program does not have any of these characteristics and its effectiveness is consequently lowered. Targeting and allocation results could be rapidly improved through use of the most recent registry of poor households, the PPLS11. The school committee targeting and nomination system is not effective in identifying students currently out of school; those likely to leave school; or the entire universe of potential students living in poor households. Targeting should proceed based on the upcoming 2011 list of poor households (PPLS11) and the unified targeting system when it becomes available. Households not on the PPLS list but who believe they have a qualified student should be able to appeal to the school directly (see also the recommendations for an improved socialization strategy below). BSM design should be revised so that the program is able to provide reliable relief for students and households during the riskiest periods of an educational career. Enrollment rates in basic education are high for everyone, but students from poor households drop out at alarmingly elevated rates between the last grade of basic primary and the first grade of junior secondary. For the poor who make it to junior secondary, there is a similarly alarming rate of drop-out between junior and senior secondary levels. BSM design characteristics (conditionality, slow annual verification and disbursement, and a lack of delegated responsibility for continuing students) exacerbate these risks: BSM scholarships are essentially unavailable for students finishing up primary school and entering junior secondary school and those completing junior secondary school and entering senior secondary school. BSM benefits should be recalculated to be commensurate with the total costs of education. The cash transfers provided by the BSM program are a small fraction of the total costs of education; that fraction has been eroding as the costs of education have risen. In addition to tuition and registration fees (which in real terms have been rising while BSM scholarship amounts have not been adjusted) all households pay for uniforms, books and materials, and fees for exams, school committees, extracurricular activities, and transportation. Transportation alone can double the total cost of attendance. All told, a poor household pays about 20 percent of total household expenditures to send one child to senior secondary school and BSM cash transfers cover less than half of that amount. BSM benefits should be calibrated to reflect the higher costs of schooling at higher levels. The total costs of a year of schooling rise by roughly 100 percent between the last year of primary and junior secondary levels and by over 100 percent again between junior and senior secondary levels. BSM benefits are not currently calibrated to with these considerable increases. BSM cash transfers should be delivered when needed. Current BSM delivery schedules see students receiving cash transfers for, say, 4th grade attendance after they have paid fees and enrolled in 5th grade (in other words a full year or more after school begins). To effectively assist households with the costs of education, BSM transfer delivery should be timed so that students receive funds in time to pay for registering and enrolling. Additional portions of the total cash transfers should be delivered, for example, soon after first-semester attendance has been verified, second-semester attendance has been verified, and when full-year attendance and passing grades have been verified. 31 Bantuan Siswa Miskin | Cash Transfers for Poor Students BSM cash transfers could include a “graduation bonus”. In addition to calculating appropriate BSM cash transfer levels and determining an appropriate delivery schedule, the BSM program should consider providing a “graduation bonus” for those students transitioning between elementary and junior secondary and for those transitioning between junior and senior secondary. This graduation bonus provides an incentive for continuing education during a period when many drop out and also helps smooth the costly transition to a new higher level of fees. BSM needs intra-agency rationalization. Organizational reform should have as its goal a better service standard for students and internal reforms should focus on revising BSM organization so that scholarships can follow the life-cycle of a student during her educational career. Currently BSM organizes around divisions (by schooling level and by type of religious instruction) within the implementing agencies, which produces a design failure for transitioning students who are not served by any BSM option. Consolidating the programs into one scholarship program should produce economies of scale in the delivery of the support operations. Socialization to potential beneficiaries needs to be carried out thoughtfully and frequently. Partly as a result of current procedures (selection of BSM recipients by school officials and lack of an application procedure), very few school- age children from poor households know about the BSM program nor if they are technically eligible. Additionally, the uncertainty regarding the continuity of the BSM scholarship across schooling levels makes it less likely that BSM provides a positive change in expectations sufficient to draw students into the system permanently. Examples from Indonesia and internationally convincingly demonstrate that scholarships are only effective when disadvantaged households are provided enough knowledge to use them properly. BSM reforms should therefore include a thoughtfully developed and effectively delivered socialization and outreach campaign to all potential beneficiaries and their communities. 32 References Chen, D., Agustina, C.D., Ragatz, A., and Setiawan, I. 2009. “Scholarship Programs in Indonesia: Past, Present, and Future.” Jakarta, Indonesia: World Bank Grosh, M., del Ninno, C., Tesliuc, E., & Ouerghi, A. 2008. For Protection and Promotion: The Design and Implementation of Effective Safety Nets. Washington, DC: World Bank. Smeru. 2006. “A Rapid Appraisal of the Implementation of the 2005 Direct Cash Transfer Program in Indonesia: A Case Study in Five Kabupaten/Kota.” Research Report. Jakarta, Indonesia: The Smeru Institute. Smeru. 2008. “Qualitative Baseline Study for PNPM-Generasi and PKH: The availability and Use of the Maternal and Child Health Services and Basic Education Services in the Provinces of West Java and East Nusa Tenggara.” Research Report. Jakarta, Indonesia: The Smeru Institute. Smeru. 2009. “A Rapid Appraisal of the Implementation of the 2008 Direct Cash Transfer Program and Beneficiary Assessment of the 2005 Direct Cash Transfer Program in Indonesia. Research Report.” Jakarta, Indonesia: The Smeru Institute. World Bank. 2006. Making the New Indonesia Work for the Poor. Jakarta, Indonesia: World Bank. World Bank. 2010. Indonesia Jobs Report: Towards Better Jobs and Security for All. Jakarta, Indonesia: World Bank. World Bank. 2012a. Targeting Poor and Vulnerable Households in Indonesia. Jakarta, Indonesia: World Bank. World Bank. 2012b. Protecting Poor and Vulnerable Households in Indonesia. Jakarta, Indonesia: World Bank. World Bank. 2012c. “Social Assistance Program and Public Expenditure Review 1: Public Expenditure Review Summary.” Jakarta, Indonesia: World Bank. World Bank. 2012d. “Social Assistance Program and Public Expenditure Review 2: BLT.” Jakarta, Indonesia: World Bank. World Bank. 2012e. “Social Assistance Program and Public Expenditure Review 3: Raskin.” Jakarta, Indonesia: World Bank. World Bank. 2012f. “Social Assistance Program and Public Expenditure Review 4: Jamkesmas.” Jakarta, Indonesia: World Bank. World Bank. 2012g. “Social Assistance Program and Public Expenditure Review 5: BSM.” Jakarta, Indonesia: World Bank World Bank. 2012h. “Social Assistance Program and Public Expenditure Review 6: PKH.” Jakarta, Indonesia: World Bank. World Bank. 2012i. “Social Assistance Program and Public Expenditure Review 7: JSLU, JSPACA, and PKSA.” Jakarta, Indonesia: World Bank. World Bank. 2012j. “Social Assistance Program and Public Expenditure Review 8: History of Social Assistance in Indonesia.” Jakarta, Indonesia: World Bank. 33 Bantuan Siswa Miskin | Cash Transfers for Poor Students 34 Indonesia Social Assistance Program and Public Expenditure Reviews are policy notes summarizing current issues in the Government of Indonesia’s major household-targeted social assistance programs. Design, implementation, coverage, impacts, budgetary demands, cost effectiveness, stakeholder perceptions, and program history will be examined through qualitative and quantitative analysis for each program: conditional and unconditional cash transfers, subsidized rice distribution, a health fee waiver, cash scholarships, and cash transfers for the neglected elderly, disabled, and at-risk youth. Evidence-based appraisal can assist policymakers in evaluating programs and deciding whether and how they are achieving Indonesia’s social protection goals. Reviews 1 through 8 together comprise a companion volume to the “Protecting Poor and Vulnerable Households in Indonesia” report. “Protecting Poor and Vulnerable Households in Indonesia” provides a public expenditure review of the wider social assistance sector, summarizes results from Reviews 1 through 8, explores gaps in both population coverage and risk coverage of current social assistance programming, and offers recommendations and suggested reforms for achieving an integrated and coordinated household-based social assistance system. THE WORLD BANK, JAKARTA OFFICE Indonesia Stock Exchange Building Tower II/12th Floor Jl. Jend. Sudirman Kav. 52-53 Jakarta 12910 Tel: (6221) 5299-3000 Fax: (6221) 5299-3111 Website: www.worldbank.org/id