The Promise of Education in Indonesia OVERVIEW Consultation Edition The Promise of Education in Indonesia OVERVIEW Consultation Edition This volume is a product of the staff of the International Bank for Reconstruction and Development/The World Bank. The findings, interpretations, and conclusions expressed in this paper do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guar- antee the accuracy of the data included in this work. Copyright © 2019 The World Bank The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission maybe a violation of applicable law. The International Bank for Reconstruction and Development/ The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of this work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978-750-8400, fax 978-750-4470, http://www.copyright.com/. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Pub- lisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail pubrights@ worldbank. org. Contents Abbreviations  iii Introduction  1 Education sector diagnostic   7 GOAL 1: Boost learning  12 RECOMMENDATION 1: Ensure that students reach at least minimum learning and development standards at each level of the system   14 GOAL 2: Provide learning for all   16 RECOMMENDATION 2: Act to improve learning outcomes of the lowest performers   18 GOAL 3: Start early  20 RECOMMENDATION 3: Make quality early childhood education accessible to all   23 GOAL 4: Serve everyone  24 RECOMMENDATION 4: Ensure that all students, including the disabled, succeed   27 GOAL 5: Improve teaching  28 RECOMMENDATION 5: Improve teacher recruitment, training, and professional development; experiment with incentives to increase accountability   30 GOAL 6: Increase learning for employment   32 RECOMMENDATION 6: Expand access and improve the quality of TVET and tertiary education   35 GOAL 7: Manage for performance   36 RECOMMENDATION 7: Strengthen accountability mechanisms   38 GOAL 8: Align institutions for learning   40 RECOMMENDATION 8: Support existing institutions to improve service delivery  44 Conclusion  45 Notes  46 References  47 Acknowledgments  49 Figures Figure 1 Number of teachers and students in MoEC and MoRA institutions  3 Figure 2 Most provincial national exam scores miss the national passing grade  8 Figure 3 NES scores are not useful for diagnosing the causes of low learning  9 Figure 4 Indonesia compares poorly with neighbors such as Vietnam  13 Figure 5 The quality of preschools varies widely across settings, and average quality is low  22 Tables Table 1 Summary of key recommendations  4 Table 2 Accreditation status of higher education institutions  34 Boxes Box 1 The potential of technology in education  1 Box 2 Learning poverty and learning inequality  8 Box 3 Holding all actors to account  43 The Promise of Education in Indonesia • iii Abbreviations 3T Border, remote, underdeveloped (Terdepan, Terluar, Tertinggal) AKSI Indonesian Student Competency Assessment (Asesmen Kompetensi Siswa Indonesia) APBD District-level funds ASEAN Association of Southeast Asian Nations BAN-PAUD National Accreditation Board for Early Childhood Education (Badan Akreditasi Nasional Pendidikan Anak Usia Dini) BAN-PT National Accreditation Board for Higher Education (Badan Akreditasi Nasional Perguruan Tinggi) BAPPENAS Ministry of National Development Planning BLK Working and Training Center (Balai Latihan Kerja) BOP-PAUD School Operational Assistance Grant from subnational government (Bantuan Operasional Pendidikan) BOS School Operational Assistance (Biaya Operasional Sekolah) DAKs Earmarked transfers Dapodik Data Pokok Pendidikan DAU General allocation fund DINAS Province-level education offices ECED Early childhood education and development IQF Indonesian Qualification Framework IT Information technology KEMENPAN RB Ministry of State Apparatus Utilization and Bureaucratic Reform (Kementerian Pendayagunaan Aparatur Negara dan Reformasi Birokrasi) LKP Course and Training Institutions (Lembaga Kursus Dan Pelatihan) LPDP Institute of Education Fund Management (Lembaga Pengelola Dana Pendidikan) LPMP Education Quality Assurance (Lembaga Penjaminan Mutu Pendidikan) LPTKs Teacher Training Institutes (Lembaga Pendidikan Tenaga Keguruan) MoEC Ministry of Education and Culture MoF Ministry of Finance MoHA Ministry of Home Affairs MoM Ministry of Manpower MoRA Ministry of Religious Affairs MoSA Ministry of Social Affairs MoV Ministry of Villages MSS Minimum Service Standards NES National Education Standards OECD Organisation for Economic Co-operation and Development PAUD Early Childhood Education (Pendidikan Anak Usia Dini) PAUD DAK Special Allocation Fund (Pendidikan Anak Usia Dini Dana Alokasi Khusus) PERDA Regional Regulation (Peraturan Daerah) iv • The Promise of Education in Indonesia PIP Program Indonesia Pintar PISA Programme for International Student Assessment PNS Civil servant (Pegawai Negeri Sipil) PPP Purchasing power parity RISKESDAS National Health Survey RPJMN Medium-Term National Development Plan SMA Senior secondary schools (Sekolah Menengah Atas) SMK Vocational high schools (Sekolah Menengah Kejuruan) SMP Junior secondary schools (Sekolah Menengah Pertama) STEM Science, technology, engineering, and mathematics SAKERNAS National Labor Force Survey (Survei Angkatan Kerja Nasional) Susenas National Socioeconomic Survey (Survei Sosial Ekonomi Nasional) SUPAS Inter-census Population Survey (Survei Penduduk Antar Sensus) TIMSS Trends in International Mathematics and Science Study TVET Technical and vocational education and training UN National Exam (Ujian Nasional) USBN Locally designed and administered test The Promise of Education in Indonesia • v Introduction Indonesia needs an education system Indonesia is large and growing rapidly to match its development goals With 268 million people (2018), Indonesia comprises In a speech following his 2019 reelection, Indone- more than 17,000 islands, spread over 5,000 kilo- sian President Joko Widodo “Jokowi” declared his meters and spanning three time zones, strategically aim to develop an adaptive, productive, innovative, located between the Pacific and Indian oceans. A and competitive Indonesia that will make the country diverse country, Indonesia is the world’s most pop- one of the strongest in the world. He highlighted that ulous Muslim-majority nation. Resource-endowed the key to this more prosperous future is developing and rapidly urbanizing, Indonesia is the world’s 10th human resources (State Address, August 2019). To largest economy and, if current growth rates are sus- prosper, Indonesia needs an education and training tained, it is expected to become the 4th largest econ- system that can enhance the well-being of its citizens, omy by 2050.1 Its population has a median age of improve its human capital, and achieve its economic 28.8 years and is expected to exceed 318 million by and development goals. But the current education 2045.2 Indonesia spent approximately 20 percent of system delivers insufficient student learning. For the national budget on education each year over the example, while science scores on international tests past decade.3 have been increasing, learning levels are still 19 points below those predicted by Indonesia’s income (World Strengthening human capital is crucial Bank 2018d). To achieve the President’s vision, a com- for Indonesia’s future success prehensive change in the education and training sys- tem is needed to deliver on its promise and support Major policy reforms in previous decades have dra- the country’s full participation in the fourth industrial matically improved access to education, including revolution and harness the benefits of Indonesia’s raising the average years of education for individu- demographic dividend. als 20–25 years old from 6.95 years in 1987 to 10.94 BOX 1 The potential of technology in education Indonesia started instituting computer-based testing in Moving forward, Indonesian classrooms and teacher- 9th and 12th grade national exams in 2014. Globally, this training programs can pilot and gradually introduce proven change often marks a turning point in the integration of technological aids to the teaching and learning process to technology into classrooms (Omidiyar 2019). However, the enhance and support teachers and administrators, not to use of education technology (EdTech) is still only in the very replace them. MoEC and MoRA can work with the private early stages in Indonesia (Google–Temasek 2018). sector and other partners to evaluate EdTech products EdTech start-up firms point to low levels of tech skills and identify low-cost, high-impact products that can be and a lack of incentives to adopt new approaches, leading used increase equity in student learning outcomes at scale. to low levels of EdTech uptake among teachers and school A popular approach to EdTech integration is large-scale leaders. This makes integration of technology difficult tablet distribution, which runs the risk of being both costly (Bahrdwaj and Yarrow forthcoming). Parents, teachers, and ineffective (e.g., American Institutes for Research and school leaders point to the fact that the benefits of 2015). Rather than rushing, it is essential to accompany Indonesian EdTech products remain unproven. This skep- hardware with high-quality curriculum-aligned software ticism is prudent, since some EdTech products have no and teacher and administrator training as well as internet impact or even reduce student learning (J-PAL 2019). connectivity if EdTech is to improve teaching and learning equitably at scale. The Promise of Education in Indonesia • 1 years in 2018 (Susenas 1987, 2018) in a context of system at 16 years old with 10.94 years of education rapid population growth. Since 2002, further efforts (Susenas 2018). But many of those who complete sec- have dramatically raised spending and expanded ondary education do not have the skills needed in the enrollment in a large and complex education system labor market and end up in low paying occupations (figure 1). (World Bank calculations based on Sakernas). Low Advances in computing and technology as a part of skills reflect poor basic education and poor alignment Industry 4.0 are expected to dramatically change the between education institutions’ curricula and labor way society works and interacts. The growing econ- market needs. More than 55 percent of students do omy needs increased human capacity in basic skills, not achieve minimum mastery in literacy and math, as well those for technology-enhanced occupations to and, as they engage in TVET and higher education, expand the number of good paying jobs (World Bank the taught curriculum tends to be misaligned with 2019b). These changes make improving human capi- today’s market needs or those expected for Industry tal essential to enable Indonesia to achieve its ambi- 4.0 (World Bank 2018a). tions and reach its full potential. Student learning results remain low, Indonesia ranks 87th on the World and inequality in learning outcomes Bank’s Human Capital Index is increasing While Indonesia has made significant progress in Despite the large increase in spending and resources, recent years, it is still hamstrung by a human cap- student learning results remain low, and inequal- ital deficit. Indonesia ranks 87th of 157 countries on ity in learning outcomes is increasing. The learning the World Bank 2018 Human Capital Index, which gap between the bottom and top 50 percent of stu- assesses countries’ future productivity based on their dents by household income increased from one year education and health outcomes. Indonesia’s score on of learning in 2003 to two years of learning in 2015 the 2018 Human Capital Index was 0.53.4 This means (World Bank 2018a). This overview of the forthcoming that, on average, Indonesian workers of the next gen- Indonesia Education Flagship report examines ways eration will be only 53 percent as productive as they to strengthen education reforms to boost the learning could be under the benchmark of 14 years of learning outcomes of all Indonesian students. It focuses on how and full health. the education system can deliver on the promise of Improving Indonesia’s human capital is a complex human capital for Indonesia. More specifically, it looks and long-term agenda, which must be at the core of at the changes the central government can make to the government’s growth strategy. It requires upgrad- improve its approach either directly for areas under ing the education system at all levels, from early its control, or indirectly by guiding and supporting childhood education through tertiary education and subnational levels, including provinces, districts, and lifelong learning opportunities. schools. The overview builds on two previous World Bank studies: the 2018 World Development Report, which examined education around the world, and The time to deliver human capital is now Growing Smarter, which looked at education in East Each year 4.2 million Indonesians leave the education Asia and the Pacific. system (Susenas 2018). The average student exits the 2 • The Promise of Education in Indonesia FIGURE 1 Number of teachers and students in MoEC and MoRA institutions STUDENTS INSTITUTIONS 437,811 937,170 2,409,692 765 7,399 758,178 15,926 1,437 3,176,155 492,830 23,265 1,288,075 1,686 22,153,241 6,763 3,495,570 6,732 131,974 2,585,169 15,733 7,540,555 23,227 3,333,265 16,270 TEACHERS MoEC institutions public primary MoEC institutions private primary 24,251 94,648 MoEC institutions public junior secondary 199,130 MoEC institutions private junior secondary 46,065 MoEC institutions public senior secondary 214,372 MoEC institutions public junior secondary 47,497 757,173 MoRA institutions public primary MoRA institutions private primary MoRA institutions public junior secondary 371,700 MoRA institutions private junior secondary MoRA institutions public senior secondary MoRA institutions public junior secondary 707,540 239,287 ECED institutions, ECED private: 204,495 ECED students, 324,580 ECED institutions, public: 204,495 310,834 public: 4,154 ECED teachers, NPNS teachers: 611,103 ECED teachers, MoEC public primary PNS teachers: MoEC private primary 45,107 MoEC public junior secondary MoEC private junior secondary MoEC public senior secondary MoEC public junior secondary MoRA public primary ECED students, MoRA private primary private: 5,964,507 MoRA public junior secondary , MoRA private junior secondary MoRA public senior secondary MoRA public junior secondary The Promise of Education in Indonesia • 3 TABLE 1 Summary of key recommendations GOAL RECOMMENDATION WHO OPTIONS FOR HOW Boost learning Ensure that students MoEC MoEC could reduce and revise the NES indicators to focus on •  reach at least MoRA measurable and observable aspects of the education process minimum learning Local level that are more closely linked to learning. and development supervisors Subnational stakeholders can develop budgets and learning •  standards at each level (pengawas) improvement plans to improve student learning. of the system Principal and MoHA can require these plans and assess goal achievement; •  teacher working MoEC can provide technical support. groups MoEC can revise the national curriculum to focus on •  competencies rather than facts. Provide learning Act to improve MoEC MoEC and MoRA can implement national assessments in •  for all learning outcomes of MoRA primary to identify learning inequities. the lowest performers Provinces Through in-service teacher training provided at subnational •  Districts levels, all actors can ensure that teachers know how to use Schools assessment results to support students. Teachers Provinces and districts can use student learning data to identify •  Supervisors the lowest 40 percent of schools and students. Principal and MoEC and MoRA can provide special capacity support to •  teacher working consistently low-performing schools and districts. groups Schools and teachers can preferentially support lowest •  LPMP performing students. Start early Make quality early MoEC (DG ECED Government, led by MoEC, can issue a policy statement making •  childhood education and Community two years of preprimary education compulsory; share roadmap accessible to all Education) to achieve this by 2030. MoRA Increase public funding to ECED and seek alternative, •  BAPPENAS innovative approaches to funding. MoHA Use PAUD DAK to increase the supply of quality early childhood •  MoV education. Provinces Use a socialization campaign to stimulate registration of PAUD •  Districts services and higher enrollment. Villages Improve collaboration among PAUD stakeholders and improve •  data collection on PAUD services, teachers, and learners. Serve everyone Ensure that all MoEC Provide support to students at high risk of exclusion and identify •  students, including the MoRA them early. disabled, succeed MoSA Remove barriers to continuing schooling by adapting learning •  Provinces environments. Districts Train teachers to identify and work with disabled students; •  Schools refine the curriculum to be more inclusive. Use BOS, BOP-PAUD to reduce cost of schooling, PIP for •  subsidies to disadvantaged families to enroll and keep children in school. Improve teaching Improve teacher MoEC Strengthen recruitment processes for all teacher types to ensure •  recruitment, training MoRA only the highest qualified candidates work with children. and professional KEMENPAN RB MoEC and MoRA can establish/enforce procedures around •  development; Province induction, probation, and teacher assessment. experiment with District Provinces and districts can require supervisors to monitor and •  incentives to increase Schools supervise student assessment by teachers; these are used to accountability Principals inform teaching and learning. Supervisors Strengthen working groups to support their efforts to increase •  LPTKs quality and decrease disparities among schools. MoEC and MoRA can include result of formative and summative •  student assessments in teacher appraisal. Districts and provinces can experiment with ways to increase •  accountability through incentives. 4 • The Promise of Education in Indonesia TABLE 1 continued GOAL RECOMMENDATION WHO OPTIONS FOR HOW Increase learning Expand access and MoM Establish a Skills Development Council with strong private •  for employment improve quality of MoEC sector participation. TVET and tertiary MoRA Develop labor market information system to guide policymakers •  education MoF and job seekers on their TVET decisions. BAN-PT MoM can lead the development of competency frameworks that •  reflect private sector needs. TVET institutions can increase their capacity to deliver •  graduates with these competencies. MoF can finance accreditation agencies to assure independence •  and capacity to undertake accreditation. MoEC and MoRA can consolidate small, low-quality private •  universities, improve the quality of tertiary institutions; and increase the independence and financing of the tertiary accreditation board. Manage for Strengthen MoHA MoHA and MoEC can develop a simple education quality •  performance accountability MoEC index drawing on improved MSS, NES, and student learning mechanisms (through MoRA measures. better data tracking Parents (school MoEC and MoRA can require districts to evaluate student •  and verification) committees) learning at primary level, support them on strategies to improve Teachers learning. Schools Districts can communicate results to parents and teachers, •  Districts support schools and teachers to remediate gaps. Provinces Schools can use results to improve teacher practices, mobilize •  community support, and provide additional services for students. MoEC can support schools and provinces to improve data •  reporting; MoHA can mandate independent verification of data, with financial sanctions for misreporting. Align institutions Support existing MoEC Support school improvement and enhance student outcomes •  for learning institutions to improve MoRA by building the capacity of existing actors (such as working service delivery MoHA groups, school committees). Districts Incentivize and hold accountable districts through performance- •  Provinces based budgeting and capacity building and support. Teachers Improve the soon-to-be-implemented performance-based •  Principal and BOS program (BOS Kinerja) by using objective indicators in the teacher working scoring mechanism. groups School committees LPMP LPTKs Table 1 abbreviations: BAN-PAUD = Badan Akreditasi Nasional Pendidikan Anak Usia Dini, BAN-PT = Badan Akreditasi Nasional Perguruan Tinggi, BAPPENAS = Ministry of National Development Planning, BOP-PAUD = Bantuan Operasional Pendidikan, or School Operational Assistance Grant from subnational government, BOS = Biaya Operasional Sekolah, School Operational Assistance, ECED = Early childhood education and development, LPMP = Lembaga Penjaminan Mutu Pendidikan, or Education Quality Assurance, LPTK = Lembaga Pendidikan Tenaga Keguruan, or Teacher Training Institutes, MoEC = Ministry of Education and Culture, MoF = Ministry of Finance, MoHA = Ministry of Home Affairs, MoM = Ministry of Manpower, MoRA = Ministry of Religious Affairs, MoSA = Ministry of Social Affairs, MoV = Ministry of Villages, MSS = Minimum Service Standards, NES = National Education Standards, PAUD-DAK = Pendidikan Anak Usia Dini Dana Alokasi Khusus, PIP = Program Indonesia Pintar, TVET = Technical and Vocational Education and Training. The Promise of Education in Indonesia • 5 Education sector diagnostic Indonesia has achieved much … … but needs to focus more on learning Education is a central part of the government’s Indonesia has made the right moves but needs development agenda to work differently to achieve learning Education is central to the Indonesian government’s Despite important progress in prior years, most stu- development agenda. Since the early 2000s, Indo- dents do not meet the national learning targets Indo- nesia has implemented a broad range of education nesia has set for itself. Measures of learning show reforms, including decentralizing much of the educa- challenges in primary grades (40 percent of 2nd tion system,5 improving the achievement of teacher graders cannot recognize two-digit numbers and 50 qualifications, and increasing education spending, up percent of 4th graders cannot arrange a series of four- 200 percent in real terms from 2002 to 2018 (World digit numbers by value), and learning remains low as Bank calculation). These reforms have expanded students move across grades (World Bank data 2011). access to education, particularly among disadvan- Learning is low both in absolute terms, below national taged children. The additional resources for the sector targets, and in relative terms when compared with mandated by the constitutional amendment of 2002 neighboring countries (World Bank 2018a). Despite successfully financed the expansion of education ser- recent growth in learning as measured by the PISA, it vices and increased the number of teachers for new will take 50 years for Indonesia to reach the average schools and classrooms, as well as for kindergartens OECD score (World Bank 2018a). To reach its human and other early childhood programs (World Bank capital potential, Indonesia must now work differently. 2018a). Learning poverty and learning inequality are Enrollments are up by more than 10 million both high Since 2000, the total enrollment of students has Learning inequality is high between regions, between increased by more than 10 million (25 percent), mostly schools and within schools (box 2). Some provinces in in secondary education. The average rate of increase Indonesia, especially those in the central region, per- between 2000 and 2015 was 0.26 years of education form well on the national exam, while others, often in per chronological year—more than doubling the rate the east and far west, perform poorly (figure 2). The in the 50 years prior. Between 2002 and 2017, enroll- difference between the average of the three top per- ments of youth ages 16–18 increased from 50 percent forming provinces and the three lowest performing to 71 percent. 6 provinces on the 12th grade exam for SMA (senior secondary schools) is 21 points on a 100-point scale. Indonesia has made gains in math and science Only 4 of the 34 provinces had an average 12th grade while educating more children score above the minimum passing score of 55. The Of countries participating in the Programme for results are even lower for the 9th grade exam (SMP, International Student Assessment (PISA), Indonesia junior secondary schools), and for technical and voca- recorded the highest gain in mathematics between tional schools (SMK, 12th grade exam). Districts with 2003 and 2015, a positive outcome that coincided higher incomes, large urban centers, and greater with a rapid expansion of enrollment, including stu- implementation capacity tend to do better than lower dents from low socioeconomic conditions. Improv- income, more rural districts with lower implementa- ing learning in math while expanding enrollment tion capacity (World Bank 2013). was a major achievement, and scores in science also improved. Students consistently fail to meet the country’s own learning standards There is little evidence of effective support or pressure to improve teaching and learning in classrooms based on these poor results. Instead, the movement has been away from testing and accountability for deliver- ing student learning. The national exam at the end of The Promise of Education in Indonesia • 7 primary was transformed into a locally designed and administered test (USBN), and the UN (Ujian Nasi- BOX 2 Learning poverty and learning inequality onal) or national exam for 9th and 12th grades has Learning poverty is the share of children in a country who are not been a graduation requirement since 2015 (Per- either not enrolled in school or not proficient in reading at age mendikbud 58/2015). While some decentralization of 10. Indonesia’s learning poverty rate is estimated at 35 percent, assessment is in line with the broader political pro- though the data used for this calculation is from 2011 since cess of decentralization, Indonesia’s students need a there is now no nationally representative test of student learn- national assessment in at least one grade in primary ing outcomes at the primary level (World Bank 2019c). school to make sure that students who are not learn- ing foundational skills in reading and math are priori- Learning inequality is the difference in achievement between tized for support. In response to the poor results, the the poorest quintile and the richest quintile. In Indonesia, this education system at all levels should focus on improv- difference is large. PISA data show that the gaps are large and ing learning. growing. The next round of PISA data, to be released shortly, should provide more data on this important indicator of equity. The complexity of a decentralized system has created challenges Indonesia’s education system is both complex and very large. It is the 4th largest education system in the Most provincial national exam scores miss the FIGURE 2  world with 3.9 percent of the world’s student popu- national passing grade lation. Aligned with the overall decentralization pro- Average senior secondary Ujian Nasional score by province, 2019 cess, Indonesia has since 1999 decentralized much of the education system so that it now involves multiple Maluku 41.2 Passing grade actors at the central, provincial, district, and school Papua 41.4 levels (Law 23/2014 on Regional Autonomy and Maluku Utara 41.8 Sulawesi Barat 42.9 Law 33/2004 on Fiscal Balance). The formal system Aceh 43.0 collectively employs 3.3 million teachers educating Nusa Tenggara Timur 43.4 53.1 million children in 1st through 12th grades under Sulawesi Selatan 43.9 the Ministries of Education and Culture (MoEC) and Sulawesi Utara 44.3 of Religious Affairs (MoRA). An additional 231,446 Sulawesi Tengah 44.5 early childhood education centers support the early Nusa Tenggara Barat 44.9 learning of 7.4 million children (DAPODIK 2019, MoEC Sulawesi Tenggara 45.2 2019). And 4,670 higher education institutions provide Papua Barat 45.2 services to 8 million students. The nonformal voca- Gorontalo 45.3 tional training system comprises more than 40,000 Jambi 46.2 institutions under the supervision of MoEC and Minis- Sumatera Utara 46.6 try of Manpower, as well as some line ministries. Kalimantan Barat 48.1 Kalimantan Tengah 48.4 Kalimantan Utara 49.3 The decentralization process has underlined Lampung 49.5 weak checks and balances in education delivery Bengkulu 49.9 Decentralization is well suited to a large system such Sumatera Selatan 50.1 as Indonesia’s, but smaller districts tend to have low Riau 50.2 capacity to manage their education services, with National 50.4 negative impacts on spending efficiency and student Sumatera Barat 51.7 learning (Al-Samarrai 2013, World Bank 2018a). The Jawa Barat 52.0 decentralization process also revealed weak systems Kalimantan Selatan 52.1 of checks and balances in education service delivery Kepulauan Bangka Belitung 52.3 between central and subnational levels, and among Banten 52.4 central levels (Al-Samarrai 2013; World Bank 2017). Bali 52.6 Kalimantan Timur 53.3 Jawa Timur 53.8 Coordinating multiple actors at multiple levels Kepulauan Riau 57.0 is difficult Jawa Tengah 58.1 Two key ministries—MoEC and MoRA—oversee for- D I Yogyakarta 64.6 mal education.7 But other ministries and institutions D K I Jakarta 65.5 are also involved, such as the Ministry of Home Affairs 0 10 20 30 40 50 60 70 (MoHA), the Ministry of National Development Plan- Source: World Bank, based on Ministry of Education and Culture data 2019. https://hasilun .puspendik.kemdikbud.go.id. 8 • The Promise of Education in Indonesia ning (BAPPENAS) as well as KEMENPAN RB, the spend a considerable amount of time filling out and Ministry of Villages, and the Coordinating Ministry compiling all this information each year. While both of Human Development and Culture, among others. the MSS and NES signals correlate with each other, Decentralization laws shifted the management of neither closely correlates with student learning, one schools under MoEC to more than 34 provinces and of the central functions of an education system (figure 500 districts administering some 340,000 schools 3, NES and Student Achievement). This means that and other learning institutions across Indonesia’s the information requested by the central government more than 17,000 islands. Some 42,800 schools are from schools and subnational governments, which is classified as “3T” (Terdepan, Terluar, Tertinggal, or used for decisionmaking, is not necessarily linked to border, remote, underdeveloped). The districts’ highly improvements in system performance. Since the data varied institutional capacities and socioeconomic and are not checked by direct observation, some are likely geographic conditions affect their ability to deliver inaccurate as well. Further, no data are publicly avail- education services effectively and efficiently (World able on MSS achievement at the time of publication, Bank 2017). Coordinating so many actors at different suggesting a lack of interest or follow-up on this basic levels is not an easy task. of-service delivery metric. quality-­ Central government data requests from Spending is still relatively low and can be better provinces and districts don’t correlate with targeted student learning Although Indonesia officially allocates 20 percent of The central government focuses on multiple sets of its national budget to education, it is spending less, indicators, sending mixed signals to provinces and as a percentage of GDP, than comparable countries. districts. One set of signals comes from the Minimum For example, Indonesia’s expenditure on education as Service Standards (MSS) of MoHA, which are few in a percentage of GDP was 3.3 percent in 2014, falling number but regularly revised. All of the standards to 3.0 percent in 2018, compared with spending by should be achieved, and while some subnational Malaysia at 4.7 percent in 2017 and Vietnam at 4.4 governments exceed them, many others fail to meet percent in 2016 (UNESCO). Furthermore, Indonesia them year after year, with little to no accountability is among the countries with the lowest expenditure (World Bank 2018a). A separate, very large set of sig- in PPP terms among countries participating in PISA nals comes from the National Education Standards (World Bank 2018d). (NES), a subjective set of 595 questions for school principals, with no external verification of the reported Most public financing of education comes from information.8 The NES survey includes a total of 2,055 the center and is managed at the subnational level questions for principals, teachers, supervisors, stu- To fund the education sector in a decentralized context, dents, and school committees, obliging schools to the central government supports local g ­ overnments FIGURE 3 NES scores are not useful for diagnosing the causes of low learning Junior secondary Ujian Nasional score and NES score, 2017 70 75.9 73.6 73.7 74.3 73.8 72.6 75.5 75.0 80 72.1 71.3 73.2 72.0 71.2 69.3 70.7 71.2 68.9 71.5 68.9 68.4 68.5 60 64.6 66.5 65.0 66.6 70 60.7 59.4 62.5 Ujian Nasional score (0–100) 54.5 54.2 56.0 55.1 60 50 49.6 46.3 50 Percent 40 40 30 30 20 20 10 10 62.1 58.9 58.9 57.7 57.2 57.0 56.5 55.6 55.1 54.8 54.7 54.7 54.1 53.9 53.2 53.0 52.8 52.1 52.1 51.8 51.4 51.4 51.0 50.9 50.5 50.1 49.1 48.7 48.5 48.5 48.0 47.4 47.2 46.9 0 0 DI Yogyakarta Papua Barat Jawa Barat Sumatera Utara DKI Jakarta Sulawesi Utara Maluku Sulawesi Tenggara Kalimantan Tengah Jawa Timur Sulawesi Selatan Maluku Utara Jawa Tengah Riau Kepulauan Riau Bali Kalimantan Selatan Gorontalo Sulawesi Tengah Nusa Tenggara Timur Kalimantan Timur Sumatera Barat Papua Lampung Nusa Tenggara Barat Kep. Bangka Belitung Kalimantan Utara Sumatera Selatan Jambi Kalimantan Barat Aceh Banten Sulawesi Barat Bengkulu Ujia n Na sional score (left axis) Avera ge NES achievement (right a xis) Source: World Bank analysis drawing from https://puspendik.kemdikbud.go.id/hasil-un/ and http://pmp.dikdasmen.kemdikbud.go.id/. The Promise of Education in Indonesia • 9 through fiscal transfers. Most of the public financing It is difficult for provinces and districts to know to the education sector comes from the central gov- how to focus their discretionary spending ernment, but nearly two-thirds of education spend- The bulk of the Indonesian education budget com- ing is managed by subnational governments, namely prises fiscal transfers to schools and to district and provinces and districts.9 The national government provincial offices. BOS grants, one of the largest allocated Rp 492 trillion for education in 2019. Of of these transfers, are intended to inject funds into this amount, Rp 52 trillion was for education services schools to keep children enrolled and give schools under MoRA, while MoEC was allocated Rp  36 tril- some flexibility in managing their own funds. Sup- lion. The majority, Rp 308 trillion, was transferred to porting this flexibility and the decentralization effort local governments (Presidential Regulation 107/2017, in general, the government has moved both to anchor Presidential Regulation 129/2018), and the remainder the principles of school-based management in the allocated to tertiary education and other education national education system and to provide a frame- spending. These large local government transfers work of national standards. However, these National include allocations to: Education Standards and the Minimum Service Stan- dards are neither verified nor closely correlated with • The general allocation fund (DAU), mostly for student learning outcomes. The wide variation in sub- recurrent expenditures of the local governments national capacity makes improving learning outcomes including teacher salaries. DAU makes up 34 per- even more difficult (World Bank 2013). cent (Rp 168.8 trillion) of the total allocated at the national level for general education in 2019.10 Educating to reap the demographic dividend • Earmarked transfers (DAKs), including: will pay off – Teacher professional and special allowances, The new administration is taking action to reverse the which make up 12 percent (Rp 56.8 trillion). country’s human capital shortfall. It is implement- ing an ambitious program of investing in people to – The Bantuan Operational Sekolah (BOS) per improve health, nutrition, and education outcomes, student school grant, which is 10 percent (Rp all key for developing human capital and a more pro- 51.2 trillion), while the transfer for preschool ductive labor force. (BOP PAUD) is 1 percent (Rp 4.4 trillion). With 50 percent of Indonesians under the age of – A special allocation fund for education con- 30, the population is very young.11 A demographic struction (DAK-Fisik), which is 3 percent (Rp dividend—from having more workers in relation to 16.8 trillion). dependents—is already materializing, and appropri- ate policies can ensure that the country benefits from Schools, districts, and provinces control most it. On one side, a large number of young people are of the inputs that determine learning entering the labor market with the potential to boost MoEC’s authority, according to the Education Law overall productivity and economic growth. On the of 2003, is focused on hiring civil servant teachers, other side, the number of school-age children is start- establishing curricula and competency standards, ing a gradual decline, which will eventually free up and administering student learning assessments. This resources to improve education quality (Supas 2015).12 means that basic inputs for student learning—such The dividend is expected to peak between 2020 and as the availability and quality of textbooks and other 2030, when the share of the working age population teaching and learning materials, as well as in-ser- and the potential for increased output per capita will vice teacher training and monitoring and supporting be at their highest. This opportunity is rapidly slipping teachers, principals and schools—fall largely under away as this “golden generation” leaves the education the authority of districts and provinces. To improve system, though an expanded range of lifelong learn- student learning, subnational spending and initiatives ing opportunities could sustain the dividend longer. need to be aligned with regulations and support for If this opportunity is missed, Indonesia will likely not learning at the center. The current lack of alignment reap the predicted benefits of Industry 4.0 and risk between student achievement and the MSS and NES being saddled with a less productive workforce for the monitoring systems needs to change in order for stu- next generation. dent learning to increase at scale. 10 • The Promise of Education in Indonesia GOAL 1 Boost learning 12 • The Promise of Education in Indonesia Why is this important? 100 in 2018; the minimum passing score is 55 (MoEC 2019). This means that students, on average, fail the Human capital is critical for future success summative secondary exam. There may be no direct Strengthening human capital is crucial for Indone- consequences, since the students are still eligible to sia’s future success so that it can provide the skills graduate and eventually enter higher education. But to fully participate in Industry 4.0 and, through a teaching and learning need to improve for Indonesia higher-skilled population, harness the benefits of its to meet its own standards and to realize their ambi- demographic dividend. tions of improved human capital and foster economic growth. Education can equip students with skills to lead productive lives Indonesia participates in all major international Education is a basic human right and should equip tests students with the skills they need to lead healthy, pro- ductive, and meaningful lives. Education can improve As part of an enduring commitment, Indonesia has an individual’s economic opportunities, promote participated in all major international tests since 1990, health and general well-being, and expand the ability including the OECD’s Programme for International to make choices. For societies, education can expand Student Assessment (PISA) and Trends in International economic opportunities, promote social mobil- Mathematics and Science Study (TIMSS). Even with an ity and equity, and make institutions function more improvement in PISA results from 2012 to 2015, Indo- effectively (World Bank 2018b). All of these benefits nesia is still ranked below most neighboring countries depend not simply on years of schooling, but also on (figure 4). The TIMSS 4th grade assessment in 2015 student learning. Education technology (EdTech) has showed similar results, with Indonesia ranking among the potential to support student learning in the class- the lower-­achieving countries (Beatty et al. 2018). room as recent evaluations from other middle-income country contexts demonstrate (Muralidharan et al. Money helps, but how it is spent is important 2017, see text box above). Increased learning is often associated with increased resources in the minds of policymakers, but more resources do not automatically translate into improved How is Indonesia doing? learning (World Bank 2018b). Limited accountabil- Increased spending is not delivering more ity and capacity constrain improvements in learning learning in Indonesia. Capacity constraints limit the poten- Despite the increased spending on education, many tial impacts of district support to better education, Indonesian students are not learning enough and the school-based management, and community involve- country is not catching up with its neighbors. The aver- ment for MoEC and MoRA schools, especially in age score across all subjects and school types for the low-income and rural areas. School leaders are often national end-of-secondary exam was 49.5 points of not focused on learning, may not know how to spend FIGURE 4 Indonesia compares poorly with neighbors such as Vietnam (Share of the population by level of achievement, percentage points) INDONESIA VIETNAM OECD 6 6 0.1 6 1.1 5 0.1 5 2.5 5 7.2 4 1.9 4 15.8 4 20.5 3 11.7 3 35.2 3 27.9 2 30.9 2 32.5 2 23.2 1 55.4 1 13.9 1 20.1 Note: Students with achievement below 2 in the PISA achievement scale are considered functionally illiterate. Source: World Bank 2018a. The Promise of Education in Indonesia • 13 ­ iscretionary funds to improve learning, and may not d Jenderal Aplikasi 2018). This overlapping (Direktorat ­ be held accountable for persistently poor learning responsibility of multiple authorities leads to a lack of results (World Bank 2015). Signals from the center clarity and high transaction costs for the implemen- about what is important are reflected in the MSS and tation of EdTech approaches. NES, which are not aligned with student learning. EdTech firms are concentrated on test Education technology holds promise, but has preparation for students at the junior- not yet delivered for Indonesia secondary and senior-secondary levels MoEC, MoRA, and KomInfo have governance In the private sector, EdTech firms tend to target sec- responsibility to provide oversight to the EdTech sec- tors of the education market where there is money to tor. Responsibility for the EdTech agenda at MoEC be made, and so are concentrated on test-preparation resides with the Center for Information and Commu- for students at the junior-secondary and senior-sec- nication Technology for Education (Pusat Teknologi ondary levels. This means that priority areas such as Informasi dan Komunikasi untuk Pendidikan or reading in early grades or adaptive learning for chil- Pustekkom).13 Pustekkom’s areas of responsibility dren with disabilities receive less attention. The focus overlap closely in some areas with the product and of most private EdTech firms is Java, and Jakarta service offerings of private EdTech firms. MoRA also specifically, leaving more remote areas underserved develops online learning tools for teachers and stu- (Bahrdwaj and Yarrow forthcoming). IT connectivity dents as well as school administration technologies, needs to be addressed to ensure equity of access, and and it governs the use of ICT in schools under its IT literacy should not be assumed because it is a real authority. MoEC (formerly MoRTHE) governs online barrier in remote locations. Effective use of EdTech tertiary education, while KomInfo governs electronic products at scale to equitably support student learn- transactions and, with the assistance of partners, ing in Indonesia is a major challenge facing the Indo- multiple internet safety and digital literacy programs nesian education system today. RECOMMENDATION 1 Ensure that students reach at least minimum learning and development standards at each level of the system • Focus on quality of learning and provide more support to low achievers to improve the country’s overall performance. • Guide and support learning, with more emphasis on helping teachers improve, on measuring outcomes, and on stressing 21st century skills. • Students must achieve at least a minimum standard of learning and development at every level of education. What can be changed or improved? • EdTech initiatives to equitably increase student learning can be supported in MoEC and MoRA Focus more on student learning and outcomes schools through partnerships with the private sec- For Indonesia to reach its education goals, it needs tor. These private sector options can complement to shift from relying primarily on additional resources existing public sector online learning resources to focus more directly and explicitly on improving stu- and be tested to identify successful and cost- dent learning and outcomes at all levels of the system. effective approaches that can be used at scale. The results of both national and international exams There is a need for a clear vision for the role of indicate that action is needed urgently. curriculum-aligned EdTech use in classrooms by • The foundations for later learning must be pro- teachers, which could focus on rural and remote vided in early childhood development programs, areas where highly–skilled instruction is in short built upon in later grades to ensure mastery of lit- supply. Developing EdTech with a focus on lower-­ eracy and numeracy. income and rural and remote areas will increase 14 • The Promise of Education in Indonesia equity and mitigate the risk that technology may • Local supervisors (pengawas) and MoEC’s primarily benefit urban schools with high-speed province-­ level education quality assurance (Lem- internet connectivity. baga Penjaminan Mutu Pendidikan—LPMP) can work together systematically with the province-­ • In addition to the central content required of any level education offices, the DINAS, and principal/ education system, 21st century skills for Industry teacher working groups to plan budget allocations 4.0, in such areas as communication, collabora- and activities to improve student learning in each tion, and critical thinking can be expanded in a school. There is no one-size-fits-all solution, but by revised curriculum and then taught early on and working together, local teams will be better able to reinforced throughout the lifelong learning pro- find local solutions to improve learning. MoEC can cess. Given its expected impact on the economy make technical support available to regions that and lives of Indonesians, climate change could be are struggling, and MoHA can require detailed an important topic to feature across disciplines. learning improvement plans with results-based • At later stages, particularly for vocational educa- budget allocations from each level of subnational tion, partnerships with the private sector can ensure government down to the school. that the skills taught to students are responding to • MoEC can revise the national curriculum to focus market needs. A revised governance structure is more on skills and competencies needed in the needed to promote private sector participation in labor market. Curricular reform is notoriously the TVET system. lengthy and expensive, but the process should be • Access to lifelong learning opportunities can be started soon. Overall coherence of the curriculum, increased by improving the quality of the supply of as well as sequencing between grades and forward these opportunities and incentivizing the demand planning for textbook supply, teacher training, and so that low- and middle-skilled employees can get assessment mechanisms are essential to achieve training for continuing employment. Systematic better results than the 2013 curriculum reform experimentation and evaluation of education tech- process. nology (EdTech) can help achieve rapid reskilling • MoEC and MoRA can support the evaluation of and upskilling at scale. different approaches for integrating EdTech into schools at the province and district level, and use What are the options to implement the results to determine what programs are most this change? effective for improving learning at the least cost. MoEC and MoRA can articulate a vision for the Learning should be guided and supported equitable use of curriculum-aligned student learn- Learning starts with the interaction between students ing support by teachers and encourage provinces and teachers in schools but must be guided and sup- and districts to provide the necessary school infra- ported by districts, provinces, and the central gov- structure, teacher training, and safety and security ernment. No magic pill will improve student learning before rolling out major technology initiatives. throughout the system, but many things can be done to support student learning: • EdTech also holds promise for learning for working adults to reskill and upskill. MoEC and MoRA can • MoEC can reduce and revise the NES indicators to work with districts, provinces, and the private sec- focus more on measurable and observable aspects tor to evaluate different models and support inte- of the education process that are more closely gration of those that are shown to be effective at linked to learning. This will send a system-wide sig- increasing learning and cost-efficient. nal that learning is important. And it can provide information to schools about where they should Putting the focus on quality (general student learning focus their attention to improve outcomes. and school performance) and equity (support to low performing students and schools) is the best way to improve the country’s performance. The Promise of Education in Indonesia • 15 GOAL 2 Provide learning for all Why is this important? Significant gaps in achievement are related to family wealth Links are strong between education, health, and Inequity based on differences in household income longer productive lives and wealth is a serious challenge. Household surveys Learning is an equalizer, promoting more equal indi- indicate that considerations related to the cost of vidual lifetime outcomes, shared prosperity, and pov- education account for more than half of cases where erty reduction. Internationally, each additional year of parents do not send their child to primary school, or schooling raises an individual’s earnings by approxi- where children drop out of school (Susenas MBSP mately 8–10 percent (World Bank 2018b, following 2015). Furthermore, significant gaps in achievement Patrinos and Montenegro 2015). Learning includes are related to family wealth. The PISA results for cognitive skills, which equip workers with knowledge 2009 showed that more than 40 percent of girls from that makes them more productive and allows them richer families achieved a score for mathematics at or to take advantage of new technologies and adapt to above level 2, while fewer than 10 percent of girls from changing work. Beyond productivity and economic poorer families did. growth, the links are strong between education, bet- ter health outcomes, and longer lives (World Bank Two obstacles impede making better use of 2018b). If the outcomes of learning are not more equi- assessment results tably achieved within a population, the benefits will Student assessment is a key step in the process of accrue to some over others and leave the excluded raising awareness of the importance and challenges even further behind. around learning. Indonesia faces two major obstacles to make better use of assessment results: How is Indonesia doing? • Student learning levels in core subjects such as More Indonesians are in school but learning math, science, and reading and writing Indone- too little sian often are not seen as important by many Reforms over the past two decades have brought stakeholders—parents, schools (including school many Indonesians from disadvantaged socioeco- committees), the community, and districts—when nomic conditions into schools, but their learning in many cases they are problematic and deserve levels remain low and inequality remains a serious immediate attention (see figure 2). problem. As wealthier Indonesians rapidly improve • There is a cultural aversion to identifying and label- their learning outcomes, the poor advance more ing low performers, especially children. As in many slowly, so the inequality in years of education is now other countries, Indonesians tend to prefer to cele- being replicated with inequality in learning outcomes. brate success and to avoid conflict and visible fail- For example, between 2003 and 2015, PISA scores ure (Mulder 2005). of students from households in the bottom 50 per- cent of the income distribution remained stable, while Boys and girls have roughly equal access to student scores in the top 50 percent of income rose. education, though girls have better outcomes The widening difference can be expressed in school Indonesia improved its Gender Parity Index (GPI) for years; the gap of about one school year in 2003 wid- school participation for children ages 7–12 from 0.89 ened to about two school years in 2015 (World Bank in 1971 to 1.00 in 2018.14 The current national GPIs for 2018a). Actions to improve reforms should therefore school participation rates for ages 13–15 and 16–18 prioritize interventions to support low-performing are also impressive, at 1.02, demonstrating that students, schools, and districts, particularly in poor females are enrolling and staying in secondary edu- and rural areas. cation at slightly higher rates than boys (Yarrow et al. forthcoming). One constant is poverty: the poorer a The system focuses on inputs more than district or family, the more likely it is to have low enroll- outcomes and more on high achievers ments and learning for both boys and girls. Indonesia’s education system tends to focus on equity of inputs not outcomes, and on the high achievers. National averages mask considerable local For example, district offices tend to work with better variation performing schools, and teachers often focus on bet- These national averages mask variations at the district ter students, leaving the weaker provinces, districts, level, including cases of significant male and female schools, and students behind with little realistic chance disadvantage. For example, the variations include the for improvement (Shaeffer and Arlianti 2019). On the significant difference of the 9th grade national exam financial side, poorer districts tend to benefit in finan- scores in Bantul Regency, Yogyakarta Province, where cial formulas to distribute resources, but there is no girls outperform boys by 6.6 percentage points on a similar program to support development of capacity. The Promise of Education in Indonesia • 17 100-point scale, and the high enrollment gap in South Post-school aspirations differ for young men Buton Regency, South Sulawesi Province, where the and women percentage of boys ages 16–18 enrolled is twice as Despite largely similar outcomes, post-school aspi- high as the percentage of girls enrolled (Yarrow et al. rations are markedly different for young men and forthcoming based on 2018 test and enrollment data). women. For example, there are marked differences by This local variation means that the most effective gender in the share of young Indonesians who want to approaches to achieving gender parity in education enter STEM-related careers (favoring males) or more are likely to be driven by localized data analysis and service-oriented careers (favoring females) (World locally driven policies and actions. This district- and Bank estimates based on PISA 2015). province-level approach to addressing gender dis- School-based violence is an issue affecting both parities can be complemented at the national level by boys and girls, and more needs be done to make documenting positive examples of addressing gender schools safe spaces for learning. More than 20 per- imbalances successfully and by providing support for cent of Indonesian students age 13–17 report being capacity and momentum building. bullied in the last 30 days (WHO 2015). Bad enough in itself, violence also reduces educational attainment and learning. Safe schools benefit everyone, so teach- ers need respectful work environments and confiden- tial channels for reporting inappropriate behavior. RECOMMENDATION 2 Act to improve learning outcomes of the lowest performers • Make help for low-performing students, schools, and districts a priority. • Use high-quality student assessments to diagnose issues and inform instruction. • Harness learning data to identify lowest-performing schools and provide extra assistance to them. What can be changed or improved? to low-performing districts, schools, and students. Otherwise, low levels of human capital are likely to persist. A culture of classroom assessment can identify gaps in student development and learning Continue to improve student assessments Measurement makes learning visible, but without follow-up The national primary exam was abolished in 2015 (Permendik- action and adjustment, assessment is worth little for improving bud 58/2015). Despite this negative development, MoEC has learning, teaching, and schools. At the most basic level, a cul- improved its system of assessment through the introduction of ture of classroom assessment can be fostered to identify gaps AKSI (Asesmen Kompetensi Siswa Indonesia or Indonesian Stu- in student development and learning and to help resolve them. dent Competency Assessment), a sample-based assessment, Assessment can highlight where support is most needed, but and increased exam integrity for the remaining 9th and 12th support needs to be provided to be fully utilized. Indonesia’s grade national exams, creating an opportunity for more detailed education system can focus more on how to improve out- and useful analyses of achievement at district and school levels. comes for low-performing students and schools rather than Indonesia should continue to improve its student assessment on increasing the performance of existing high-performers— system and, most importantly, needs to act on evidence from that is, not only raising the average achievement of schools the assessments to make schools work for all learners. and students but also reducing the gaps between the best and the worst. Provide schools with information on student achievement by grade and by question Make help for low-performing students, schools, and AKSI is closely linked to PISA and to a lesser extent TIMSS in its districts a priority question design and is part of an effort to improve Indonesia’s To overcome student assessment obstacles, Indonesian lead- poor showing on these assessments. The AKSI initiative covers ers at all levels can foster a process that recognizes under- multiple grades and includes the capacity to provide schools performance in learning and destigmatizes targeted assistance 18 • The Promise of Education in Indonesia with information on student achievement by grade, by subject, and struggling students. Teachers are already required and by question. This initiative could be expanded. Schools and (on paper) to include tutoring as a part of their 40 weekly subnational governments could use results to improve teacher academic hours (Permendikbud 15/2018); this time can be practices, mobilize community support, and provide additional focused on addressing identified learning gaps in individual services for students who are behind the curricular learning students. goals. • Consistently low-performing schools could receive special coordinated support from provincial and district offices, Use high-quality student assessments to diagnose supervisors, principal and teacher workings groups, other issues and inform instruction more successful schools. and the LPMP. • Use AKSI-for-schools, a standardized, formative, school- • MoEC and MoRA can continue improving the integrity of the based assessment in second or third grade, to help teach- 9th grade and 12th grade exam by expanding computer-­ ers, schools, and communities identify early weaknesses in based testing and online assessment and linking them to learning. broader EdTech integration initiatives. • Implement a national assessment in 4th or 5th grade to • Central ministries: provide information to the district and central government – MoEC and MoRA can require a school-based assess- about student learning outcomes in primary school and act ment and a national assessment in two different primary on that information to support districts and schools that are grades to help identify and then address learning ineq- not achieving the desired outcomes. Avoid an end-of-pri- uities. mary exam in 6th grade, which is less likely to be used to identify lower performers but instead deter students from – Results could be reported to all stakeholders. The exam progressing, and more likely to be politically challenging. design can be simple and short, to capture essential skills and competencies for what primary students should • Link the early and late primary grade assessments to in-ser- learn. vice teacher-training support to make sure teachers know – MoEC and MORA can help districts design strategies to how to use this information to target support to all students support working groups, schools, and teachers in order and especially to low-performing students. to remediate poor student learning and teaching prac- • Use the student learning data produced by these and tices based on the results of the national primary student existing assessments to identify the lowest 40 percent of assessment. schools and students at primary, junior, and senior second- • District education offices should: ary levels. Reward and encourage high-performing schools to support and work closely with low-performing schools – Organize a formative assessment of 3rd or 4th grade through teacher and principal working groups and zones. students at the beginning of the school year; AKSI for Reward and encourage high-performing districts to support schools is an example of a promising approach. low-performing districts. – Share student and classroom results with parents and • Continue improving the integrity of the 9th grade and 12th teachers within three months of the formative, school- grade exams. based assessment, along with a plan to improve the results. – Through the structures of zones and working groups, sup- What are the options to implement this port teachers and schools to remediate student learning change? gaps through student tutorials, in-service teacher train- Indonesian leaders at all levels of government can foster ing, mentoring, and other approaches. recognition of under-performance in learning • Schools can: To overcome these obstacles, Indonesian leaders at all levels of government can help foster recognition of under-performance – Use the national assessment results to improve teacher in learning and destigmatize targeted assistance to low-per- practices, mobilize community support, and provide forming districts, schools, and students. additional services for students who are behind in mas- tering the curricular learning goals. • Districts and provinces can send a strong signal that all chil- dren can learn, and that school leaders and teachers are – Use the school-based formative assessment results to accountable for ensuring that this happens and for using identify in what grades, subjects, and subject content student learning data to identify and support weak classes students perform less well and adjust the syllabus and teaching methods as required. The Promise of Education in Indonesia • 19 GOAL 3 Start early Why is this important? and qualified teachers for preschool learning, as well as the lack of attention for children with special needs The engineers, artists, and political leaders of and disabilities, holds back education outcomes. tomorrow need kindergarten today The acquisition of knowledge, values, behaviors, and Spending on early childhood education is low skills is a lifelong process that starts at birth and con- relative to other education investments tinuous throughout our lives. Successful education Indonesia has expanded participation in early child- systems promote these outcomes starting at prep- hood care and education. But investment in early rimary, gradually building on them with increasing childhood education is well below what is needed. complexity as the child develops and progresses There is also a major divide between rich and poor in to higher levels of education. Research shows that the access to and quality of early childhood education neural circuits (series of synapses) form sequentially provision. While Government Regulation 2/2018 on and cumulatively so that the robustness of progres- MSS requires local governments to support one year of sively more complex brain structures depends on the preprimary education, early childhood education is not robustness of foundational neural networks (World a compulsory part of the national education program Bank 2018b). Abundant evidence suggests that and thus often receives insufficient public funding at investment in early childhood development (focused subnational levels. on children 0–8 years of age) is more cost-effective and is particularly important to build socio-emo- Almost 12 million children ages 3–6 are not tional skills in high demand in today’s workplaces. Put enrolled in preschool education another way, efforts to transform the skill sets of stu- In mid-2019, MoEC took the positive step of revising dents with interventions targeting the senior second- downward the reported early childhood education ary or university level are unlikely to succeed if these and development (ECED) enrollment numbers (MoEC students have not engaged with rigorous learning 2019). Aligning more closely with Susenas and other materials and processes throughout their education, data sources, these lower numbers indicate that 11.7 starting with high-quality early childhood education. million children ages 3–6 are not enrolled in preschool education (MoEC 2019 and Susenas). In addition, the A good start in life makes all the difference quality of ECED services is modest, based on a sam- International research suggests that investments in ple from 10 districts (figure 5) (Brinkman et al. 2017). early childhood education generate high payoffs cog- nitively, economically, and socially (Gertler et al. 2014, The regulatory and budgetary environment is Heckman 2000). Early childhood education is most complex effective when linked with health services (including Indonesia’s early childhood education system goes by nutrition), diagnostics to detect any learning delays the name PAUD. The term PAUD is commonly used and disabilities, and interventions to mitigate them to refer to both ECED as a sector and as a facility with emotional support and family involvement— that delivers childcare and early learning services to rather than simply providing spaces and teachers. children ages 0–6 (Pendidikan—education and Anak Money spent on preschool programs generates a Usia Dini—early childhood; in a broader definition higher return on investment than the same spending used across sectors, the P stands for Pengembangan on schooling (Heckman and Masterov 2007). or development).17 At the national level, there are fiscal transfers to PAUD service providers for registered stu- dents. This BOP-PAUD in 2019 is worth IDR 600,000 How is Indonesia doing? per child per year (MoEC Juknis BOP Guidelines 2019). Students must be ready to learn At the village level, support for ECED services is usually One of the drivers behind Indonesia’s current human from the community (more than 95 percent of these capital shortfall is children entering school unprepared services are coded as “private” in DAPODIK and are to learn. In Indonesia, 55 percent of children 5–6 years community-­based  https://referensi.data.kemdikbud. old are enrolled in preprimary, but only 22 percent of go.id/index21.php). Additional funds can come from children 3–4 years old (Susenas 2018). This means district-level BOS, occasional discretionary funds from overall that only 38.5 percent of children ages 3–6 are the province or district, and in some cases village funds enrolled in preprimary (the official age for entry into (dana desa) regulated by the Ministry of Villages. All primary is 7 years in Indonesia). As in other countries, this creates a complex regulatory and budgetary poverty, malnutrition, remoteness, the lack of facilities environment. The Promise of Education in Indonesia • 21 FIGURE 5 The quality of preschools varies widely across settings, and average quality is low Project playgroup (N=236) 3.1 81% of centers in Indonesia meet minimum ECED Kindergarten (N=221) 2.9 standards, but only 43% meet a slightly higher threshold based Non-project playgroup (N=70) 2.7 on ECERS-R. Islamic kindergarten (N=50) 2.5 0 1 2 3 4 5 6 7 Source: Brinkman et al. 2017. Stunting and malnutrition limit human capital An ambitious campaign to reduce childhood development stunting is under way… Despite some progress, Indonesia’s rates of child- The government is implementing an ambitious pro- hood stunting and malnutrition seriously limit its gram to reduce stunting, which is expected to make a human capital development (Rokx, Subandoro, and major contribution to improving the country’s human Gallagher 2018). According to the National Health capital ranking. The US$14.6 billion National Strategy Survey (RISKESDAS), in 2018 about 30 percent of to Accelerate Stunting Prevention, launched in August Indonesia’s children under 5 years old (almost 9 mil- 2017 with World Bank support will benefit 48 million lion children) were stunted, down from 37 percent in pregnant mothers and children under 2 in the next 2013, although the rate is higher in some regions such four years. as Aceh and Sulawesi. Stunting is the impaired growth and development that children experience from poor … helping children and families with a package nutrition, repeated infection, and inadequate psycho- of key services social stimulation. To combat stunting, pregnant mothers and children under the age of 2 need to simultaneously access key Stunting is largely irreversible services. These services include basic immunization, Children are defined as stunted if their height-for- breastfeeding, dietary diversity, clean drinking water age is more than two standard deviations below the and sanitation, early childhood stimulation and edu- World Health Organization’s Child Growth Standards cation, food insecurity measures, and a birth certifi- median. Stunting occurs in the first 1,000 days of life, cate to make sure children can access future benefits. can be traced back to the period of pregnancy and the Indonesia is now focusing on providing these key serv- health and nutrition status of the expectant mother. ices as a package. The investment is already improv- It is largely irreversible. Long-lasting consequences ing Indonesian families’ access to key quality services, include diminished mental ability and learning capac- from health and nutrition to education and sanitation, ity, poor school performance, reduced earnings, and and is part of a large national movement to reduce increased risks of nutrition-related chronic diseases Indonesia’s high stunting rates. such as diabetes, hypertension, and obesity. 22 • The Promise of Education in Indonesia RECOMMENDATION 3 Make quality early childhood education accessible to all • Make quality early childhood education compulsory and accessible to all. • Strengthen the coverage and quality of ECED by ensuring sufficient funding; publish roadmap to achieve universal PAUD enrolment by 2030. • Incentivize PAUD expansion through grants for additional centers and encourage better collaboration among stakeholders. What can be changed or improved? What are the options to implement this change? • MoEC, MoHa, and MoRA can propose a law making two years of preprimary education compulsory • The MoEC Minister (with the support of the Ministry of Finance, MoHA, Ministry of Villages, and MoRA) can • Use district-level funds (APBD) to expand the number make two years of preprimary education compulsory for and improve the quality of PAUD services using a staged all children. Develop a roadmap for phased implementa- approach, prioritizing children by age and socioeconomic tion, including financing and technical support. MoHA can background for one year of preschool, and then work on advocate for districts to pass district legislation (PERDA: additional years for younger children. Praturan Daerah) to finance and implement PAUD services • Use the new RPJMN to gradually push districts to achieve using APBD. Access can be to daycare centers, play groups, 100 percent enrollment in one year of kindergarten, iden- kindergartens, and a range of other services for children tify support for districts to achieve this goal including hir- under the PAUD umbrella. Parenting programs should be ing qualified teachers, and hold to account those that fail expanded with clear links to the stunting agenda. to reach targets in subsequent years. This can be a phased • BAPPENAS (with support from MoEC, MoHA, and MoRA) requirement, since more than 20,000 villages in Indonesia can plan implementation of this commitment to two years lack a PAUD center (MoHA 2019). of preprimary with a staged approach—prioritizing children • PAUD expansion could be incentivized by an output-based by age group and socioeconomic background—starting grant or new PAUD-DAK-Fisik for districts to build new with one year preprimary that meets the minimum service PAUD centers meeting a small number of key criteria and standards and provides funding for poor families and rural registering existing PAUD service providers to ensure that areas. Conduct budget analyses on the financing gap to those providers meeting the minimum requirements benefit provide the supply side (infrastructure and operational from BOP-PAUD. costs, teacher salaries, professional development, and so on). Issue an implementation regulation that defines roles • Develop a “socialization” campaign to stimulate both the and responsibilities of various stakeholders in ECED. registration of all PAUD services and the higher enrollment of children in PAUD services. An unknown number of PAUD • Ensure that all sectoral stakeholders in PAUD communi- service providers are not registered with MOEC, so their cate this roadmap to relevant line directorates, and to dis- children are not entered in DAPODIK and do not receive trict and village governments to secure their commitment BOP-PAUD. Of course, many children are not enrolled at all. through local policies and budgets (working across MoEC, MoRA, MoHA, MoV, and BAPPENAS). • The government can strengthen the coverage and quality of ECED by giving it sufficient funding within the current 20 • Organize national and local campaigns and identify national percent education envelope and improving the governance and local champions (such as the Bunda PAUD) to raise framework by ensuring that minimum quality standards are awareness of the benefits of ECED and increase demand for met. The recent mandating of minimum service standards ECED services. for ECED (Peraturan Pemerinta 2/2018) with the technical • Both increase public funding to ECED (from central min- guidelines (Permendikbud 32/2018) is an important step istries, district budgets, and village funds) and seek alter- forward. But if the standards are not enforced to ensure native, innovative approaches to such funding (such as minimum levels of quality, children will develop and learn public-private partnerships, the private sector). less than they should, and human capital will not reach its full potential. The Promise of Education in Indonesia • 23 GOAL 4 Serve everyone Why is this important? Providing accessible facilities is proving difficult Providing accessible facilities (such as ramps and wide Education for all requires inclusive education doors) to children with physical impairments, is prov- systems and schools ing difficult for many schools and their personnel to Many factors exclude children from attending school, manage. So is assisting children with cognitive delays and many more exclude them from learning in the caused by conditions such as dyslexia, and handling classroom. Yet education for all—and ensuring that children with more complex cognitive and socio-emo- all children learn what they are meant to learn— tional challenges such as autism and hyperactivity. The remains essential in both fulfilling children’s rights result is that a large percentage of children with delays and ensuring their participation in the development of and disabilities never enter school or never advance their community and their nation. This requires inclu- to higher levels of education (Male and Wodon 2017, sive education systems and schools, where inclusive UNESCO Institute of Statistics 2017, 2018). education is defined as “a transformative process that ensures full participation and access to quality learn- Many of these children can be included in ing opportunities for all children, young people and “regular” schools adults, respecting and valuing diversity, and eliminat- Having access to, and opportunities for success in, ing all forms of discrimination in and through educa- education is important because it is every child’s tion” (UNESCO Cali Statement 2019). right—and evidence shows that many of these chil- dren can be included in “regular” schools and in Exclusion has many underlying factors learning and others, with more complex needs, in Gender is a major factor of exclusion, though less so in inclusive schools or schools for special needs. Such Indonesia, but many other factors are also important: inclusiveness is important because of the potential • Remoteness—Children living in rural and remote contribution such children can make to their own— areas that may have few educational facilities and and to national—development and because their par- many barriers to reach the ones that exist. ticipation in education can demonstrate the diversity of experience, which can contribute to creating toler- • Poverty—Children of families who cannot afford ant and just societies (Vargas-Baron 2019, Olusanya, to send their children to school and/or need their Krishnamurthy, and Wertlieb 2018). children to supplement family income. • Language—Children whose mother tongue is dif- How is Indonesia doing? ferent from the language of the school and there- fore have difficulty in gaining literacy in the national Overcoming exclusion from schooling and language. learning Remoteness. In Indonesia, more than 53,000 MoEC “Inclusive education” is often limited to and MoRA primary schools have fewer than 100 stu- disabilities and special needs dents, of which 37,441 schools have fewer than 60 stu- “Inclusive education” is often limited, as in Indonesia, dents (Dapodik and EMIS 2018). This is largely due to to its original focus on disabilities and special needs, the presence of small communities in rural and remote but the more general definition is broader, cover- regions where the population does not permit larger ing many kinds of barriers to education. One of the enrollments. The schools are often of low quality, with most intractable barriers to a child’s participation in inadequate facilities and relatively untrained teach- schooling and in learning is the existence of delays ers (or trained teachers not motivated to remain in and disabilities, whether sensory, intellectual, men- such schools for long). The solution favored by many tal, or physical (or more than one). The first barrier is ministries of education (and finance) is to close such often the shame felt by the family, resulting in the child schools and merge them to create a larger, suppos- being hidden or kept out of public spaces. Early diag- edly more economically viable entity. Often a solution, nosis of the disability, through early childhood inter- this sometimes comes at the sacrifice of the one insti- vention programs for example, might lead to early tution—the school—that serves as the social hub of mitigation, but the specialists required to make such the community. diagnoses and manage such efforts at mitigation are rare, especially in rural and remote areas. And early Multigrade teaching is the model of choice in childhood development personnel and even primary many advanced nations school teachers seldom have the skills and knowledge Another possible solution—multigrade teaching required to identify disabilities, even simple ones such where one teacher teaches more than one grade—is as slightly impaired sight and hearing (which could be found in many places in the world and is the model helped by simply moving such children to the front of of choice in many advanced nations (Little 1995). the classroom). But it is rarely seen in Indonesia, despite successful The Promise of Education in Indonesia • 25 development agency pilot projects and training pro- Children with special needs tend to drop out grams (e.g., UNICEF, USAID PRIORITAS, and INO- Children with special needs who enroll in school tend VASI). One reason for this has been the reluctance to drop out as they move up the system and encounter of MoEC to accept anything less than the presumed challenges such as examination protocols that are not international norm of “one class, one teacher.” A sec- appropriate to their disability, physically inaccessible ond is the oversupply of teachers (mostly contracted) facilities, and even fewer teachers trained in special even in remote schools and even where there are not needs (World Bank forthcoming). The enrollment rates enough classrooms. And a third is that multigrade for children with at least one type of physical difficulty teaching requires a modified curriculum and addi- at primary level is 91 percent. This falls to 49 at junior tional teacher training which development agencies and 27 percent at senior secondary levels. can provide but LPTK generally do not. So, while the need for multigrade teaching for better quality National policy stipulates that children with remains, this need generally is not met in Indonesia. special needs be integrated into the education system Indonesia has been ensuring enrollment of Indonesian national policy stipulates that children with children from poor families special needs be integrated into the education system Poverty. Indonesia has been successful in ensuring through dedicated special needs schools and through enrollment of children of poor families, through such inclusive schools, which include children with and programs as BOS (which provides per capita funds without disabilities. While a number of important reg- for school and therefore, in theory, negates the need ulations support inclusive education, including the Per- for school fees) and PIP (Program Indonesia Pintar), mendiknas 70/2009, implementation has been lacking which provides funds for education to families classi- in many provinces and districts. Teachers often do not fied as poor, as well as PKH (Program Keluarga Hara- have the training to fully integrate children with physi- pan), which provides grants to poor families. cal, psychological, and learning disabilities, and social stigmas are often high, leading parents to conceal chil- Indonesian students of primary age with a dren, particularly those with sensory disabilities. disability are more than 10 times more likely to be out of school Despite the presence of some 700 languages, Disability. In Indonesia, having a disability greatly most are not officially used in the education increases a child’s likelihood of being out of school. system A disabled person is defined in Indonesia as one Language. Despite the presence of some 700 lan- who has long-term physical, mental, intellectual, or guages in Indonesia (petabahasa.kemendikbud.go. sensory impairments and, when interacting with the id), most are not officially used in the education system. environment and society, encounters difficulty in There is increasing evidence from around the world participating fully and effectively (Undang-Undang that gaining initial literacy in one’s mother tongue 8/2016). A 2016 study found that Indonesian students leads to higher achievement not only in the various of primary age with a disability are more than 10 times subjects in the curriculum but also, eventually, in the more likely to be out of school than children without national language (World Bank 2018). Some mother disabilities (UNICEF 2016). tongue programs are found in large language groups such as Balinese, Javanese, Sundanese, and the lin- The prevalence of children ages 7–18 years guistically rich provinces of Papua, but others are usu- with at least one type of physical difficulty is ally relegated to “local curriculum content” rather than 0.26 percent as the language of instruction and initial literacy. The most recent national figure for the prevalence of children ages 7–18 years with at least one type of Enforcing existing regulations and providing physical difficulty (visual/auditory/motor-sensory) is teachers with training would help 0.26 percent (Susenas 2018). Including other types Discussions with provincial and district education of functional impairment—such as behavioral and officers as part of a World Bank study (Yarrow, Afkar, learning challenges, inability to understand commu- Sudarti, and Cooper forthcoming) reveal a lack of clar- nication, and self-care—boosts the total prevalence ity about implementation, since children with special rate to 0.46 percent, though this is lower than global needs come under the auspices of the province, yet the average estimates for “severe disability” (WHO 2011). province is not responsible for primary schools, leaving According to Susenas, the proportion of enrolled stu- an ambiguous area of responsibility for primary-age dents with at least one physical impairment is 0.24 children with disabilities. Enforcing existing regulations, percent at primary, 0.16 percent at junior secondary, refining the curriculum for children with disabilities, and and 0.12 percent at senior secondary levels. providing teachers with training on appropriate strat- egies to teach students with disabilities would help to improve access to and the quality of inclusive education. 26 • The Promise of Education in Indonesia RECOMMENDATION 4 Ensure that all students, including the disabled, succeed • Ensure that all students, including the disabled and those in remote areas, succeed. • Identify children with disabilities as soon as possible so that early help can be provided; train teachers to work with children who have disabilities. • Ensure that small rural and remote schools can provide quality education. What can be changed or improved? What are the options to implement this change? • Improve instructional practices and support so that all students—from urban/rural locations, boys/ • Remoteness. MoEC can help ensure that multi- girls, with disabilities and without—can succeed. grade teaching is included in pre-service education programs and provided to all candidate teachers, • Ensure that small rural and remote schools can that the national curriculum is adapted for it, and provide quality education despite their disadvan- that adequate facilities and materials are provided tages by ensuring that teachers assigned to these to facilitate it. schools understand and can practice multigrade teaching. • Poverty. MoEC and MoF can continue to expand the provision of BOS and BOP-PAUD to reduce • Promote the mother tongue as the language of the cost of schooling and of PIP to provide subsi- instruction in early childhood development pro- dies to disadvantaged families to help ensure their grams and the early grades of primary school children enroll and remain in both nonformal and where the majority of children in a class speak the formal schools. same non-Bahasa Indonesia language at home, leading to a smooth transition to mastery of • Language. Where appropriate, MoEC can encour- Bahasa Indonesia in later years of primary school. age use of the mother tongue in PAUD programs and the early grades of primary school and provide • Ensure that at the district-level children with dis- adequate materials and teacher training to imple- abilities are identified as early as possible, pro- ment such mother tongue programs and ensure a vided with early childhood interventions where successful transition to mastery in Bahasa Indone- possible, and eventually enrolled in preschool and sia. then primary school, and appropriately served. • Disability. MoEC, MoHA, and the Ministry of Social • Refine the curriculum for children with disabilities Welfare can work together to ensure the early iden- and provide teachers with training on appropriate tification of children with delays and disabilities by strategies to teach students with disabilities, which working with schools and village/neighborhood would help to improve the access to and the qual- authorities. Provide early childhood interventions ity of inclusive education if combined with accessi- and financial and technical support to PAUD pro- ble infrastructure and equipment. grams and primary schools to enroll these children and ensure that they are included in learning to the extent possible. Invest and promote school designs facilitating access for children with disabilities. The Promise of Education in Indonesia • 27 GOAL 5 Improve teaching 28 • The Promise of Education in Indonesia Why is this important? How is Indonesia doing? Good teaching is central to learning Only two-fifths of the 421 teacher training insti- Good teachers are central to student learning. To tutions are accredited improve their quality, Indonesia needs to assist them Indonesia’s large number of teacher training institu- more effectively, both before they enter the classroom tions (421) produce more than three times the number and throughout their careers. Without consistently of teacher candidates required by the public service better teaching, Indonesian students will not achieve system. This very large number of teacher candidates, the foundations for later learning or gain the skills for approximately 300,000 in 2017, include many who the 21st century workplace in a competitive and glo- are underqualified, linked to the fact that 58 percent balized economy. of the teacher training institutions are not accred- ited.15 There is a need to reorient from the quantity of More than 3.3 million teachers work in teacher graduates to the quality of teacher graduates. Indonesian classrooms Worryingly, very few of high performers on PISA want More than 3.3 million teachers work in Indonesian to become teachers (PISA 2015). classrooms every day, along with 294,000 profes- sors and lecturers at the tertiary level and 656,000 Classroom teachers have different hiring in ECED (Statistik Pendidikan Indonesia 2017/2018, processes EMIS 2018, Statistik Pendidikan Tinggi 2018). For Classroom teachers fall into different categories, with students to learn, teaching has to be effective, since different hiring processes and different levels of qual- well-trained and motivated teachers are the most fun- ification, pay, and benefits. For MoRA, civil servant damental ingredient for learning after the students teachers and principals make up 19.2 percent of the themselves. workforce, and non-civil servants 80.8 percent (Simpa- tika 2018). For MoEC, civil servants are 48 percent of the teacher workforce, while teachers hired by commu- nity foundations (yayasan) are 16 percent, and honor- ary teachers hired by the districts are 28 percent, while part-time teachers (often hired directly by schools) make up the remaining 7 percent (MoEC 2018). The Promise of Education in Indonesia • 29 RECOMMENDATION 5 Improve teacher recruitment, training, and professional development; experiment with incentives to increase accountability • The caliber of teaching needs further improvement, including through better hiring and continued professional development. • Indonesia should ensure that it has enough highly qualified teachers in the right locations, particularly in rural, remote, and low-performing schools. • It should also experiment with ways to increase teacher accountability through incentives. What can be changed or improved? servants (PNS) or as contract and honorarium teachers, is cen- Indonesia can afford to hire only the most qualified tral to improving student learning (de  Ree 2016). In addition, candidates teacher pay within school can be made more equitable, made Indonesia can insist on hiring only the most qualified candidates dependent on performance and less on the hiring mechanism. to become teachers. It can educate and pay them well, deploy Honorarium teachers’ salaries are way below minimum wage, them efficiently and equitably across the country, while provid- and less than a fifth of certified civil servant teachers’ income ing incentives and support for continuous improvement. There (World Bank forthcoming). needs to be continual development of teachers’ skills through more effective professional development, including through 2. Improve preservice teacher preparation lower-cost online options if proved effective. Given the need The high civil-servant teacher salary and certification payment is to reach more than 4 million teachers, new strategies should encouraging people to enter teacher education institutions, and be tested and scaled. And to keep the best teachers in the the high demand for this education has encouraged the open- classroom, robust teacher evaluation systems should be imple- ing of additional private (and usually low-quality) teacher educa- mented and linked to incentives based on performance. Pre- tion institutions. Preservice teacher education can be improved paring teachers better requires targeted reforms, coordinated with an updated curriculum, blended approaches to offline, efforts, and clear and consistent implementation of regulations online, and distance teaching and learning, the appointment across independent training and decentralized administrative of lecturers with experience in the education level for which systems—a major challenge. they are training new teachers, and more in-school and better Additional policy actions are needed to improve the pool of supervised teaching practice, beginning from the first year of candidates entering the career, enhance the quality of preser- the candidates’ education. This can be linked to more robust vice training at teacher colleges (LPTKs) and in-service training, engagement by the accreditation body of teacher training insti- and improve the teacher’s incentives to deliver higher quality tutes, as well as publication of the rate of acceptance of gradu- education. ates of individual institutions to civil service teaching positions. Five ways to improve the caliber of teaching Hire highly qualified teachers while resisting political 3.  Indonesia can ensure that it has the right number of highly pressures qualified teachers in the right locations, particularly in low-per- Numbers of non-civil servant teachers have grown in recent forming, remote, and rural schools, and that teachers are per- years, with uneven quality control. With a large pipeline of forming at their best. With 55 percent of civil servant teachers retirements, up to one million new civil-servant teachers are retiring over 10 years starting in 2018 (about 960,000 individu- expected to be hired in the next 10 years (World Bank 2018c). als) (World Bank 2018c), there are major opportunities and risks While some existing highly qualified contract teachers can be to reshaping the teacher workforce for the next generation. hired into newly opened civil servant positions, no candidates Here are five ways to accomplish this: who lack qualifications should be selected despite political and other non-professional reasons for doing so. 1. Attract the best candidates and pay them more No one should be teaching who is not qualified—whether equitably  it is schools hiring honor teachers, districts and provinces hir- Many teachers lack the basic subject knowledge to effectively ing contract teachers, or PNS teachers hired centrally. All new support student learning (Ragatz 2015, Al-Samarrai et al. 2013). teachers, regardless of hiring mechanism, should have some Hiring only highly qualified teacher candidates with strong minimum subject knowledge and meet standards for the pro- knowledge of the subjects they will teach, whether as civil fession. 30 • The Promise of Education in Indonesia 4. Ensure continuous professional development with province and districts. Subject knowledge of the sub- Teacher competencies can be  continually improved through ject(s) to be taught can be one of the main requirements for high-quality teacher professional development linked to teachers hired into new and vacant PNS posts. career progression and promotion. This can begin by serious • To build on current reforms, MoEC and MoRA can improve processes of induction and probation and continue through professional development by enforcing procedures around systematic and regular assessment processes. Special focus induction, probation, and teacher assessment, ensuring should be on the design and use of student learning assess- greater coordination at the local level among LPTKs, dis- ments to improve teaching and student learning. trict governments and other actors working with teachers Professional development is often sporadic due to the vari- and strengthening teacher, principal, and supervisor work- ation in funding of activities by districts, minimal evidence of ing groups and coaching to support their efforts to increase impact, and the fact that teachers in remote schools tend to the quality of teaching and decrease disparities in learning have fewer opportunities.  This is partially addressed by the achievement among schools. recent  Zonasi reform (Permendikbud 51/2018), but more work and attention are needed to effectively support teachers, super- • MoHA, MoEC, and MoRA can work together to further test visors, and school principals. The teacher and principal work- and then adapt existing teacher incentive programs more ing groups can be strengthened by increasing their resources, broadly. blending on-the-job training and in-the-job mentoring, and expanding their responsibilities. KIAT Guru improves teacher presence, service There is a high level of interest in using online learning to performance, and student learning outcomes in remote improve teacher practices and student learning. Excellent evi- primary schools dence exists on the efficacy of some EdTech interventions, but KIAT Guru (Kinerja dan Akuntabilitas Guru/Teacher Perfor- no rigorous evidence exists for online learning in Indonesia. mance and Accountability) is a pilot program that aims to Before committing resources to specific online learning pro- improve teacher presence, service performance, and student grams, MoEC and MoRA can work with online teacher train- learning outcomes in remote primary schools. Absenteeism in ing providers to evaluate products in the public and private remote schools (19 percent) is twice the national rate (9 percent) domains before selecting specific options to test. (ACDP 2014), with negative consequences on student pres- ence, retention, and learning outcomes (ACDP 2014, UNICEF  xperiment with ways to increase accountability 5. E 2012, Usman et al. 2004). A key feature of KIAT Guru is that through incentives it empowers communities, including parents, to hold teachers Certified civil servant teachers who are absent two days out of accountable and ties the payment of the Teacher Special Allow- five receive the same payment as those who work all five days ance (Tunjangan Khusus Guru) to teacher service performance. and come early and stay late to help struggling students. So, It works by having community members develop a joint agree- few teachers have incentives based on performance. Indo- ment with teachers to improve the learning environment in nesia has piloted the use of incentives (the KIAT Guru pilot in school and at home, verify teacher presence using an Android- rural areas as well as the Daerah Khusus Ibukota Jakarta pro- based application, and evaluate service delivery performance gram), and these can be adapted and tested more widely to through a scorecard. try to improve both equity and performance. Some existing teacher allowances can be made conditional, tied to objective KIAT Guru results were statistically and significantly and observable indicators such as attendance and professional better than control schools development to improve teaching competence. One year after the pilot was launched in 2016, student learning was assessed. Language learning outcomes improved from 37.5 percent to 50 percent, and math outcomes from 37.4 percent What are the options to implement this to 48.8 percent. Teacher presence in school improved from 78 change? percent to 83 percent, and classrooms with teachers increased • MoEC and MoRA can help attract the best teacher can- from 81 percent to 87 percent. KIAT Guru results were statisti- didates by increasing the status of accredited LPTKs and cally and significantly better than control schools (at 0.19 and actively advertising among senior secondary graduates that 0.17 standard deviations in mathematics and language respec- teaching is a worthwhile and profitable career. tively) (Gaduh et al. forthcoming). The pilot is a collaboration of MoEC, TNP2K, five rural district governments, and the World • To improve teacher preparation, MoEC and MoRA can Bank. Starting in 2019, the Government of Indonesia is expand- jointly raise standards for enrollment in LPTKs while exert- ing KIAT Guru and adapting the mechanism to urban secondary ing more quality control over LPTK curriculum, lecturers, schools. KIAT Guru provides evidence-based policy for the gov- and methods of teaching practice. ernment to introduce effective conditions for the US$6  billion • MOEC, MoRA, and MoHA can set minimum standards for of annual spending on teacher allowances, including Teacher hiring of teachers across contract types by working closely Certification Allowance (Tunjangan Profesi Guru). The Promise of Education in Indonesia • 31 GOAL 6 Increase learning for employment Why is this important? Universities and TVET institutions have limited guidance on what to teach their students Low quality of the labor force translates into Indonesia lacks a useful labor market information sys- low labor productivity and lower overall com- tem to identify occupations in high demand and allow petitiveness universities and TVET institutions to focus on these An educated and skilled workforce is a prerequisite occupations. Even though Indonesia has defined for economic growth and shared prosperity. Indone- Indonesia Qualification Framework (IQF), progress sia has favorable conditions for economic growth, but has been limited in defining the occupations and the slow progress in human capital development so far different tasks that occupations require (competency has limited the country in achieving its potential. Low frameworks), and much of this preparation has been quality of the labor force translates into low labor pro- with limited private sector participation. As a result, ductivity and lower overall competitiveness. Indone- universities and TVET institutions have limited guid- sia’s labor productivity is one-fourth Malaysia’s, and ance on what to teach their students. Added to this, the estimated contribution of education to long-term accreditation and certification are limited, allowing economic growth is 1.8 percentage points per year high heterogeneity in quality and relevance among lower than Vietnam’s (World Bank 2018a). For exam- service providers. ple, in the tourism sector, Indonesia is a leader in its attractiveness (4 of 46) but among the last in the qual- Indonesian universities rank poorly at the global ity of its general human resources (45 of 46) (WTTC level 2015). University and vocational education and train- Indonesian universities rank poorly at the global level. ing can focus on needs of the workplace so that grad- Only three Indonesian universities rank in the top 500 uates are prepared for good jobs. globally in the QS World University Ranking 2020.16 Low results are related to the lack of qualified faculty Companies find it hard to source professional members, among other issues; the RPJMN targets and managerial staff increasing the number of lecturers with PhDs from 16 Indonesia needs practical solutions to improve its percent (2018) to 20 percent by 2024 (RPJMN 2020– human capital quality and compete in a globalized 24, BAPPENAS). Given the difficulty of upgrading the economy, with an educated workforce required for faculty in the short run, funding alone will not make a success in Industry 4.0. To engage more successfully significant impact on student results. in the global economy, Indonesia needs the right set of highly qualified technicians, professionals, and The Indonesian government has launched skilled workers to perform tasks the economy needs. various initiatives to improve quality The basic education system should generate the fun- The Indonesian government has launched various ini- damental knowledge and skills that can then comple- tiatives to improve the quality and competitiveness mented by the TVET and higher education systems. of universities and higher education institutions. Pro- As enrollments in all education levels have increased, grams and incentives include introducing lecture-fo- entrants to the labor market are now better edu- cused scholarships to improve the quality of faculty cated. But as the demand for skills is growing with the members, providing infrastructure funding, and offer- economy, employers have difficulty finding the right ing online courses. skills for their needs. There is ample evidence that companies are finding it increasingly hard to source Most underperforming universities are private professional and managerial staff (Di Gropello 2011, and unaccredited Enterprise Survey 2015). Until 2018 there were more universities unaccredited in Indonesia than accredited, and some of them have How is Indonesia doing? been unaccredited long enough to have graduated classes of students. Most of the underperforming uni- Graduates do not match labor market needs versities are private and unaccredited (BAN-PT 2018). University and TVET systems have expanded rapidly in Many of them are too small to benefit from economies Indonesia in recent years, with 123 public universities of scale and to offer substandard educational ser- and 3,195 private universities in 2018 (MoRTHE Per- vices, which in turn produce poor quality graduates. formance Report 2017 and MoRTHE Annual Report This is one of the drivers behind the high unemploy- 2018). Even so, the private sector persistently notes ment among Indonesian graduates.  MoEC has the that graduates do not match labor market needs. This authority to reduce the number of underperforming in part reflects the lack of a proper skills development universities through consolidation, but it can move system with strong participation of the private sector, more decisively to resolve this issue. which could help set the priorities for investments and anticipate and validate labor market needs. The Promise of Education in Indonesia • 33 TABLE 2 Accreditation status of higher education institutions majority of those familiar with the frameworks have limited capacity to implement the framework’s study Ranking 2016 2017 2018 2019 plans. Accredited A 48 65 85 95 • Many institutions lack adequate infrastructure and Accredited B 336 531 725 881 trained teachers. There is wide variation in quality Accredited C 733 954 1,164 1,267 among training providers, and most do not have the Unaccredited 2,158 1,726 1,319 1,059 appropriate infrastructure and trained teachers to Source: BAN-PT Report (2018) and BAN-PT Executive Council presentation (2019). implement competency frameworks. • Engagement with the private sector is limited, with The overall tertiary sector needs to expand some positive exceptions. For example, SMKs in the At the same time, the overall tertiary sector needs to maritime sector can send all their graduates abroad expand. The recent opening of the sector to foreign uni- to work on the field, and SMKs in the tourism sector versities is an important step, but is highly constrained, have direct links to international cruise lines (World limiting them to partnerships with a small number of Bank 2017). existing universities or to specific geographic areas near • The different accreditation systems that monitor the capital. While the partnership approach may bene- TVET institutions need to be strengthened (and in fit a small number of existing Indonesian institutions, an theory assure the quality of these programs). Their approach that allows more foreign satellite campuses to reach is limited in the number and type of programs open more quickly and in more regions of the country they can assess. And certification of graduates fol- would better address rising demand and could offer stu- lowing the competency frameworks is still limited dents high-quality options. due to the limited number of certification providers and the lack of competency frameworks. Private sec- Promoting access to high-quality international tor demand for certification remains low. opportunities • The current governance structure has failed to pro- Nonetheless, the government is allocating substantial duce clear signals on future needs to guide the func- resources to promote Indonesian students’ access to tioning of the overall system high-quality international opportunities. Through the Lembaga Pengelola Dana Pendidikan (LPDP) program, MoEC is implementing the SMK revitalization it allocated 31 trillion IDR (2.2 billion USD) in accumu- program lated funds since 2010 and financed 18,446 domestic There have been several major regulations to improve and international scholarships for students through the the TVET system in Indonesia in recent years. One of end of 2017 (LPDP Annual Report 2017). In 2018, 55 tril- the most important is Presidential Regulation 9/2016 lion IDR were put into the LPDP fund, an amount similar on the revitalization of SMKs. Even though it focuses to MoEC’s total budget. on SMKs, it set clear mandates to finalize the develop- ment of competency frameworks and to promote better The TVET system skills and reskills technicians private sector participation and coordination among and low-level professionals training institutions. Under this regulation, MoEC is The current TVET system comprises formal and nonfor- implementing the SMK revitalization program, which mal institutions. The formal institutions include Sekolah aims to improve the quality of teachers and infrastruc- Menengah Kejuruan (Vocational High Schools [SMK]) ture, strengthen links with the private sector. There are and the Polytechnics. The nonformal institutions include similar programs to revitalize polytechnics and BLKs. Balai Latihan Kerja (Working and Training Center [BLK]) and the Lembaga Kursus dan Pelatihan Training Insti- A Pre-employment Card will provide 2 million tution [LKP]). It is estimated that more than 40,000 Indonesians with resources to engage in TVET institutions provide vocational and education training. training These institutions are under MoEC, the Ministry of Man- Given the urgency that government leadership is putting power, and the Ministry of Religious Affairs. Some of on skills for Industry 4.0, it is expected that financing for these institutions are created and managed by the line these types of programs will increase. The government ministries, such as tourism, industry, and transportation. has announced a Pre-employment Card that, starting in 2020, will provide 2 million Indonesians with resources The different TVET institutions have common issues to engage in TVET training. The government has also While each type of TVET institution has unique challenges, just launched a tax deduction program to promote there are some common issues for the sector overall: training by private companies (Peraturan Pemerintah • Large numbers of TVET institutions have limited Nomor 45 Tahun 2019, Peraturan Menteri Keuangan knowledge of the competency frameworks, and the Nomor 128 Tahun 2019). 34 • The Promise of Education in Indonesia RECOMMENDATION 6 Expand access and improve the quality of TVET and tertiary education • Tertiary education is crucial for shaping the workforce that will enable Indonesia to be competitive. • TVET and tertiary education sector should expand to meet rising demand, including by easing entry for foreign institutions. • Balance expansion with robust accountability mechanisms. What can be changed or improved? What are the options to implement this change? Establish a governance structure to guide the overall skills development system with strong participation of • Government of Indonesia can establish a Skills the private sector. This structure can set priorities in Development Council with strong participation terms of labor market needs and future expectations. of the private sector to oversee the overall skills development. The council would comprise the • Establish a reliable, timely, and easily accessi- coordinating ministries such as Bappenas and ble labor market information system to identify the ministries traditionally in charge of the skill labor market needs for use by training institutions, provision, including MoEC, Ministry of Manpower employers, students, and job seekers. The system (MoM), and MoRA—as well as line ministries with could build on the existing job-matching platform training facilities such as the Ministry of Industry, (Ayokitakerja) to help guide existing workers toward the Ministry of Transportation, and the Ministry of growing or higher paying sectors based on skills Tourism. required and training possibilities by occupation. • MoM can develop a Labor Market Information Sys- • Establish a dynamic mechanism for private sector tem to monitor the evolution of the labor demand participation in developing competency frame- and supply, but also to provide information to job works for all occupation levels of the IQF. Com- seekers on occupations. petency frameworks can benefit from frameworks already defined in the ASEAN context. • MoM can collaborate with line ministries to final- ize the development of competency frameworks, • Ensure that universities and TVET institutions have and to supervise their use in the skills development the right infrastructure and teachers to deliver the institutions. competency frameworks. Mechanisms to share resources among institutions should be explored • MoEC can expand the Revitalization of SMK pro- to maximize the use of existing infrastructure. gram subject to evaluation of current results. MoEC can continue consolidating the supply of • Have established and reliable certification and SMK, merging ones with limited capacity with accreditation systems recognize skills. Improve the SMKs with higher capacity. Continue and evaluate protocol and instruments used in the accredita- programs to Revitalize Polytechnics (by MoEC) and tion process, and have an external party audit the BLKs (by MoM). MoEC can strengthen the capacity results. of the Technology Transfer Office. • Have the TVET and tertiary education sector • MoF can establish direct financing to accreditation expand to meet rising demand, including easing agencies of universities and TVET institutions to entry for foreign institutions. assure independence and capacity to undertake • Balance expansion of tertiary and TVET with robust accreditation. accountability mechanisms, and measure account- • MoEC can increase the internationalization of the ability through accreditation and performance-­ higher education system by allowing greater free- based funding. dom for foreign higher education institutions to • Ensure that existing scholarship mechanisms, par- provide services to Indonesian students across the ticularly the LPDP, deliver on their potential. country. The Promise of Education in Indonesia • 35 GOAL 7 Manage for performance 36 • The Promise of Education in Indonesia Why is this important? Limited accountability and capacity are con- than 17,000 islands. MoEC directly manages only 9 straining possible improvements in learning percent of the budget for the system it oversees (based Capacity constraints limit the potential impacts of on 2018 financial data; World Bank 2018a). The dis- school-based management and community involve- tricts and provinces have highly varied socioeconomic ment for MoEC and MoRA schools, especially in and geographic conditions and institutional capacities low-income and rural areas. Research by the World which affect their ability to deliver education services Bank and others indicates that school leaders often effectively and efficiently (World Bank 2017). are not focused on learning, may not know how to spend discretionary funds to improve learning, and The governance structure requires districts to are often not held accountable for persistently poor fulfill a set of minimum service standards learning results (World Bank 2018a). The governance structure requires districts to fulfill a set of minimum service standards. While some sub- national governments exceed these minimum stan- How is Indonesia doing? dards, many fail to meet them year after year, with MoEC directly manages only 9 percent of the little consequence (World Bank 2018a). Reporting budget for the system it oversees measures in the current system include reporting on Decentralization laws shifted the management of BOS spending, publishing exam results and tracking schools under MoEC to 34 provinces and more than NES scores, but the resulting information (positive or 500 districts administering some 340,000 schools negative) is rarely used actively to allocate resources and other learning institutions across Indonesia’s more and technical support. The Promise of Education in Indonesia • 37 RECOMMENDATION 7 Strengthen accountability mechanisms • Keep better track of education trends by improving MoEC and MoRA databases. • Hold stakeholders and decisionmakers accountable for improving education quality by establishing an Education Quality Index. • Data from the index can be used to direct assistance to lagging districts and schools. What can be changed or improved? What are the options to implement this change? Guide stakeholders and decisionmakers and hold them accountable for improving education quality by For the Education Quality Index establishing an Education Quality Index for districts • MoHA can work with MoEC and MoF to develop and provinces on student learning, education expen- the technical guidelines for the subnational spend- ditures, and system performance. Information from ing classification regulation (Government Regu- the Quality Index can be made public at the presiden- lation 12/2019) and plan to support subnational tial level, and the information should flow to provinces, governments to implement the detailed education districts, working groups, schools, and classrooms so expenditure reporting guidelines to help answer that principals and teachers know in what areas their key questions about education spending, including students need help—and this help can then be pro- student unit costs by level of education, and spend- vided. ing on teacher training by teacher type (PNS vs. contract teachers) and level (primary vs. secondary). • Integrate and improve MoEC and MoRA data- bases to provide accurate and up-to-date data • MoHA, MoEC, and MoRA can work together to (and trends revealed by the data) to decisionmak- develop a simple quality index, drawing from ers across ministries and levels of government. improved and simplified versions of MSS and NES as well as measures of student learning. MoHA can • Ensure that the databases clearly identify inequi- use the index to identify districts and provinces not ties and disparities in the system—such as between meeting the minimum performance targets and in provinces and districts, urban/rural/remote schools need of more support. MoEC can provide assis- and large/small schools, and high-performing and tance to improve learning and school functioning low-performing schools—so that action can be to these identified districts and provinces, since taken to reduce any existing disparities. Require MoHA is empowered to instruct subnational gov- district and province education officials to pass a ernments what to do, while MoEC is qualified to customized online training course on the analysis suggest to them what they should do to improve of their own education data with a focus on identi- their performance. MoRA can perform both func- fying disparities. tions within their system. • Design and establish verification mechanisms for • The Office of the President could announce the data on school and PAUD registration, infrastruc- results of each education quality index ranking ture, staffing, and student access (enrollment) and each year to publicize the results in the national learning outcomes for MoEC and MoRA. political discussion, signaling the importance of student learning and system performance by prais- ing those that improved their results and calling on lagging regions to improve. • Databases: – MoHA can mandate independent data verifica- tion as well as financial sanctions for misreporting. – MoEC can support districts and provinces to improve data collection and reporting both up the system to the ministry and down the system to the schools. 38 • The Promise of Education in Indonesia GOAL 8 Align institutions for learning Why is this important? the following fundamental steps to align institutions around the goal of improving learning: Getting governance right is essential to overall education system performance • Focus public spending on improving the equity of Governance is the set of rules and procedures that student outcomes not only by raising the average establishes formal relationships between institutions achievement level of all students but also by nar- and different levels of authority. These relationships rowing the gap between the best-performing and are influenced by incentives, sanctions, and informal worst-performing students and schools. relationships and by how much the formal rules are • Select and support teachers throughout their careers enforced. while rewarding increased effort and improved results. As is aligning education systems with learning The World Development Report, Learning to Realize • Ensure that children are ready to learn when they Education’s Promise, highlighted the importance of start school by expanding and improving the qual- aligning education systems with learning (World Bank ity of early childhood education services. 2018b). “Alignment” here refers to ensuring symmetry • Use high-quality student assessments starting in and synchronization of the objectives of actors at all primary school to identify gaps and disparities in levels of the system. For example, are school directors learning outcomes and use the results to inform focused on supporting their teachers to improve the instruction and school improvement planning. learning of all of their students, or is their main goal increasing the school’s examination results by focusing • Improve TVET and tertiary education to produce a on the best students? Do districts and provinces hire more competent and adaptable workforce. the best qualified technical staff to support teachers and schools, or do some of them make appointments Aligning institutions is challenging with so to underqualified individuals as rewards for politi- many areas and layers of government involved cal support or for other personal reasons, as Rosser in education and Fahmi (2016) and others have documented? Are To improve education results, overall governance actors at the central level focused on assessing learn- of the education sector can be improved by taking ing achievement for all students, or are they focused action to expand the capacity of subnational educa- on other goals that may detract from this? For exam- tion governance, better align regulations and close ple, are they working on complex and time-intensive regulatory gaps. Learning is a complex process that measurement systems that don’t closely correlate with requires the alignment and coordination of all institu- learning (see figure 3)? tions involved—a process especially challenging as so many areas and layers of government are involved in Many goals are not connected to student learning education in Indonesia. When all major actors at each level in the system focus on learning in addition to other goals, the system can Upgrading subnational capacity and improving be said to be aligned with learning, since actors at accountability are the highest priority challenges every level are pointing toward (or lined up with) the Upgrading subnational capacity and improving same goal. Multiple goals are normal for all levels of accountability are the highest priority challenges a complex education system, but when some or many in ensuring alignment. Indonesia should now put a of these additional goals are not connected to student greater focus of learning outcomes by enhancing the learning or are even barriers to student learning, then tracking, monitoring, and evaluation of results and the system is not aligned with learning. then acting on these results to increase the quality of, and reduce disparities in, the education system. How is Indonesia doing? School principals are key to a decentralized Indonesia can build on its recent reforms to service delivery system improve learning Improving school management and accountability is Recent reforms to introduce performance-based BOS expected to improve student learning. While principals (BOS Kinerja) and a standard classification of district are essential to school-based management in Indone- spending (Government Regulation 12/2019) go in the sia, many do not have adequate training or knowledge right direction and indicate an appetite for results- of school management and leadership or of the cur- based reform. Effectiveness in increasing accountabil- riculum content teachers are meant to be teaching. ity for results will depend on the appropriateness and They are thus unable to lead their teachers or motivate relevance of the reform’s goals and the mechanisms their students in ways that will achieve better student to achieve them. The new presidential administra- outcomes (Bandur 2012; Vernez et al. 2016; Pradhan tion can work with subnational governments to take et al. 2014). Too often, principals are selected based The Promise of Education in Indonesia • 41 on years of teaching or on political and other per- actors accountable when this does not occur). sonal considerations rather than their ability to lead their school (Rosser 2018). Although policy requires Identify and support the lowest-performing that new principals be selected from a list of quali- students, schools, and regions fied candidates, this is not universally implemented Provinces, districts, and working groups can focus in Indonesia’s decentralized system. The appointment resources and expertise on identifying and helping of unqualified and underqualified school leaders is the lowest-performing schools and early childhood an impediment to student learning. In high-perform- education services and on increasing their capacity to ing systems, principals are not only managers of the implement—and be held accountable for—success- institution they head but also instructional leaders who ful school-based management. While MoRA’s system take a positive role in improving the quality of teaching is centralized, it too can shift to identifying and sup- and learning at their schools (World Bank 2018b). porting its lowest-performing students, schools, and regions, while holding them accountable for making School committees should become more active progress toward specific goals (box 3). in overseeing school activities Schools nominally have parent committees that can Spending is not always efficient, but the depth play an important role in school management, but of the problem is not clear many lack the authority, capacity, and resources—and In theory, subnational governments are required to the support of their school’s staff—to make much of a allocate at least 20 percent of their budget to edu- difference. The capacities of school committees need cation, but there is evidence that funds intended for to be improved for them to effectively participate in education are sometimes diverted to other local pri- activities such as results-based budgeting, operational orities or spent ineffectively (Suryadarma 2012, Al-Sa- planning and monitoring, and accountability. Without marrai and Cedran-Infantes 2013). There is a pressing such support, it is likely that higher-income commu- need to develop the human resource capacity of dis- nities, which tend to have a population with greater trict education offices and schools, especially in more capacities and a greater awareness of the importance disadvantaged regions. of education, will benefit more from increasing decen- tralization, leading to more effective school-based Basic information is lacking on the resources management and stronger school committees. Low- local governments allocate to education er-income communities—with a less educated popu- Districts use nonstandard budget classification sys- lace, perhaps less interest in education, and often less tems, making it difficult or even impossible to com- time to get involved in school affairs—will benefit less. pare spending, and some districts do not classify significant amounts of spending by program or func- The committees can be genuinely involved in tion. A recent government regulation (Government school decisionmaking Regulation 12/2019) lays the foundation for a standard School principals can ensure that the committees spending classification system for education and other meet and are genuinely involved in school decision- sectors across subnational governments. The impact making rather than being rubber stamps for a prin- of this reform will depend both on the usefulness of cipal’s decisions (as they sometimes are). MoEC and the detailed classifications adopted and on the robust- district education offices should launch a communi- ness of implementation arrangements to ensure that cation campaign directed at principals, teachers, and districts use them. communities on the importance of school committees and how they can contribute to school management; Implementation is key training can also be provided to school committee Overall, Indonesia’s education reforms have addressed members to help them fulfill their role. many of the right issues, but challenges in implemen- tation have led to uneven results. Going forward, Accountability for results should be top-down improving learning in Indonesia is about how poli- and bottom-up cies and interventions are implemented in any given Real accountability for results is a prerequisite for effec- context. Large improvements in Indonesia’s human tive management and autonomy in a decentralized sys- capital depend on shifting how the education system tem. This accountability should be both top-down and operates—specifically in aligning and strengthening bottom-up, with local leaders and government offices the capacities, effectiveness, and accountability of being accountable to higher levels as well as to citizens teachers and principals in schools and of all the other and civic organizations (World Bank 2017). To improve local, regional, and national stakeholders. outcomes for MoEC and MoRA schools as well as the wide range of early childhood education services, there Monitor the effectiveness and efficiency of is a need to align budgets, spending, and activities district education spending toward improvements in learning and to hold system Recent efforts by the central government to more 42 • The Promise of Education in Indonesia BOX 3 Holding all actors to account The senior secondary enrollment target for 2014 mum Service Standards,” especially in remote still has not been met in 2019, and is not planned areas for education, and implementation in 90 to be met by 2024 (box figure 1). No direct sanc- percent of districts, but without specific targets tions have followed this repeatedly missed target, for individual standards (RPJMN 2015–2019). The and this lack of any repercussions is common 2020–2024 RPJMN, reflecting increasing decen- throughout the system (World Bank 2018a). tralization, calls for 100 percent achievement of Other essential targets from the 2010 Medi- the newly revised minimum service standards and um-Term National Development Plan (RPJMN) for increased targeting of district-level budgets to have not been met and have disappeared com- achieve the standards across all sectors (RPJMN pletely from the objectives. For example, the 2010 2020–24). While the planning goalposts are con- RPJMN targeted improvements in minimum ser- tinually moved and, in some cases, removed com- vice standards, focusing on 17 different standards pletely, no one is held to account for successes (RPJMN 2010–14). The next planning document or failures in education system improvement or referred to “expediting the fulfillment of Mini- student learning BOX FIGURE 1 Continually missing targets for senior secondary enrollment rates 100 91.6 90 82.84 85 84.02 79.2 80 70 64.28 60 50 40 30 20 10 0 2010 2014 2019 2014 2019 2024* Baseline Target Source: RPJMN 2010–2014; RPJMN 2015-19, expected targets for 2020–2024; RPJMN (Rencana Pembangunan Jangka Menengah Nasional). https://www.bappenas.go.id/files/rpjmn/Narasi%20RPJMN%20IV%202020-2024_Revisi%2028% 20Juni%202019.pdf. efficiently use BOS grants (BOS Kinerja) and teacher teacher training programs, support to schools, and allowances (Kiat Guru pilot) can be matched by com- student learning. But other districts and provinces, plementary actions by districts. The effectiveness and even with the funds available, continue to struggle to efficiency of district education spending needs to be meet the most basic minimum levels of quality year monitored, especially in smaller districts where mana- after year. gerial capacity tends to be lower. Schools have bounded autonomy on spending Link financial transfers more explicitly to quality Schools in Indonesia have significant autonomy under Funding allocations to schools and other educational Education Law 2005. They manage resources and have institutions can be more explicitly linked to increasing autonomy, with some constraints, in how they spend access and equity and assuring quality in the system, BOS and other resources, including those provided by such as achieving accreditation or meeting national the district. Some schools have the capacity to spend standards (and narrowing the gap between the best- these resources well and could benefit from increased and worst-performing schools in this process). More autonomy, while others lack the capacity or incentive to than US$20 billion will flow from the central govern- do so. A key challenge is to decide how much autonomy ment to districts and provinces for education in 2019, to give to schools in a decentralized system in which according to World Bank calculations. Some subna- school and district capacity and motivation vary widely. tional governments invest this financing in effective The Promise of Education in Indonesia • 43 Stronger quality assurance mechanisms are Performance-based management is urgently urgently needed needed Indonesia has set up centrally appointed quality Performance-based management—where districts, assurance bodies for education in each province provinces, and the central government allocate resourc- (LPMPs), but so far there is no evidence of their hav- es and time to support low-performing schools and re- ing an impact on learning outcomes. The LPMPs gions to improve—is an urgently needed adaptation, have strong potential to contribute to the education and stronger quality assurance can assist in this process. sector at the district and school level. But in many cases, the relationship between the LPMPs and local Ensure that basic inputs are available education offices and schools is underdeveloped. International evidence shows that each child having In addition, LPMP actions are restricted to policy their own good quality and grade-appropriate text- interventions related to the Directorate General of book is a key ingredient for learning. Diverse reading Basic and Secondary Education as the LPMP are materials are essential for children to learn to read. functionally linked to that Directorate General. Their Though 33 percent of Indonesian principals indicated constrained scope of action is a missed opportunity, a shortage of textbooks on PISA 2015, there is no reli- for example on teacher and school leadership qual- able data on the extent to which children in Indonesia ity issues. Nor is there a well-established, competent have access to these vital inputs. Urgent attention is body to assist districts in improving the quality and needed to address this challenge, both on the data quantity of early childhood education, another seri- side (gathering information that matters for learning) ous capacity gap. and acting on it (managing for performance and hold- ing key actors accountable). RECOMMENDATION 8 Support existing institutions to improve service delivery • Indonesia can build on its reforms to improve learning quality. • It can work through the institutions now in place to enhance accountability and promote results-based change. • And it can link financial transfers more explicitly to quality. What can be changed or improved? What are the options to implement this change? • Support school improvement and enhance stu- dent outcomes using the building blocks already • MoHA, MoEC, and MoRA can work together to in place—principal and teacher working groups, reform the current system, which is not set up in school committees, education quality assurance a way that districts are being incentivized or held institutes (LPMPs) and training colleges (LPTK), and accountable for producing good student learning the DINAS and their supervisors. All these building outcomes. District offices (and schools) are not blocks need further capacity development, and the held accountable by parents, and their poor results resulting aligned architecture of support can be do not have an impact on budgets. One option directly involved in improving teacher performance. to address this is moving to performance-based budgeting (for stronger performers) and providing • Work district by district to make staff more capable capacity building and support (for weaker perform- and accountable for the work they do, including ers). This approach would be similar to that in the clarifying the role of every DINAS unit in enhanc- performance-based BOS program: if schools per- ing learning outcomes and requiring DINAS staff form better, they will get more funds as a reward to to remain in their positions for a minimum period encourage them to perform even better. of time following capacity-­ strengthening activities. • MoEC can improve the newly implementing • Incentivize and hold accountable districts through performance-­ based BOS program (BOS Kinerja) performance-based budgeting and capacity build- by simplifying the scoring mechanism to fewer and ing and support. more objective observable criteria. • Encourage all education stakeholders to partici- pate in education service delivery. 44 • The Promise of Education in Indonesia Conclusion A new round of reforms can create the path to success Indonesia has huge potential to upgrade its human capital for the 21st century. To get education right, it can raise awareness of the importance of education and ensure that high-quality teacher candidates are recruited, trained well, supported and held accountable for learning. It can make early childhood education accessible and compulsory. It can concentrate on improving the lowest performers and thereby reduce learning inequities in the system. And it can focus on using data to upgrade quality. Earlier reforms have put the building blocks in place for achieving the overarching goal of improving learning, and a new round of reforms can create the path to success. The Promise of Education in Indonesia • 45 Notes 1. Based on a PPP measured GDP. 2. https://indonesia.unfpa.org/sites/default/files/pub-pdf/Proyeksi%20Penduduk%202015-2045_.pdf 3. Average of national education spending as a percent of national spending from 2008 to 2018 was 18.5 per- cent (World Bank COFIS Database). 4. Human Capital Index: https://www.worldbank.org/en/publication/human-capital 5. MoRA constitutes 15 percent of the pretertiary education system and is not decentralized 6. https://data.worldbank.org/indicator/SE.SEC.NENR?locations=ID&view=chart 7. The higher education function of the Ministry of Research, Technology and Higher Education (MoRTHE) was transferred to the Ministry of Education and Culture in October 2019. 8. The NES Junior Secondary questionnaire contains 595 questions for school principals, 563 for supervisors, 547 questions for each teacher, and 162 questions for students, 188 questions for school committee (Kemen- dikbud 2018). 9. This proportion includes the general allocation fund (DAU), with an estimated amount going to education. The amount going to education from this unconditional grant is estimated by MoF and does not represent actual reported expenditures by subnational governments, as actual reported expenditure amounts by sub- national governments in education are not tracked. 10. A specific amount of DAU is allocated by MoF for education, however, the amount of DAU actually spent on education by subnational governments is estimated (allocation not execution). 11. https://www.indexmundi.com/indonesia/demographics_profile.html 12. Based on projections of the 2015 population survey (Supas), the population of primary age children has already begun to decline. However, the rate of decline is slow; the number of Indonesian children age 5–9 in 2018 is estimated at 22,043,000, while the number of children age 5–9 in 2025 is estimated at 21,906,000. 13. The Pustekkom budget in 2018 is IDR 197,753,183,000 as per Ministry of Finance’s budget document (RKKL DIPA Kemdikbud, 2018) for MoEC. This allocation covers multiple areas, such as enhancing school connectiv- ity, developing digital learning resources, improving educators’ ICT skills, and increasing ICT use for teaching and learning. 14. World Bank calculations using Population Census 1971 and Susenas 2017. 15. http://lldikti12.ristekdikti.go.id/2016/12/15/mayoritas-lptk-belum-terakreditasi.html 16 https://www.topuniversities.com/university-rankings/world-university-rankings/2020 46 • The Promise of Education in Indonesia References ACDP Indonesia: Education Sector Analytical and Capacity Di Gropello, E. 2013. “Role of the Education and Training Sector in Development Partnership. 2017. “Teacher Retirement Pattern Addressing Skill Mismatch in Indonesia.” In D. Suryadarma and 2017–2027: An Opportunity to Improve Teacher Manage- G. Jones (Eds.), Education in Indonesia, pp. 236-266. ISEAS– ment,” Working Paper, June 2017. Yusof Ishak Institute, Singapore. ACDP Indonesia: Education Sector Analytical and Capacity Devel- Di Gropello, E., A. Kruse, and P. Tandon. 2011. Skills for the Labor opment Partnership. 2014. Study on Teacher Absenteeism in Market in Indonesia: Trends in Demand, Gaps, and Supply. Indonesia. ACDP, Jakarta. World Bank, Washington, DC. Al-Samarrai, S. 2013. Local Governance and Education Perfor- Direktorat Jenderal Aplikasi. 2018. Literasi Digital Pilar Masyarakat mance: A Survey of the Quality of Local Education Governance Informasi Indonesia. https://aptika.kominfo.go.id/2018/11/ in 50 Indonesian Districts. World Bank, Jakarta Indonesia. literasi-digital/ Al-Samarrai, S. and P. Cerdan-Infantes. 2013. “Where Did All the Enterprise Survey. 2015. Indonesia. https://microdata.worldbank. Money Go? Financing Basic Education in Indonesia.” Educa- org/index.php/catalog/2665 tion in Indonesia, 109-138. Gaduh, A., M. Pradhan, J. Priebe, and D. Susanti. Forthcoming. Al-Samarrai, S., M. C. Chang, A. B. Ragatz, S. Shaeffer, R. Steven- Camera, Score, Action? Incentivizing Teachers in Remote son and J. J. De Ree. 2013. Teacher Reform in Indonesia: The Areas. Role of Politics and Evidence in Policy Making: Main Report Gertler, P. J., J. J. Heckman, R. Pinto, A. Zanolini, C. Vermeersch, (English). Directions in Development; Human Development, S. Walker, S. M. Chang, and S. Grantham-McGregor. 2014. World Bank, Washington DC. “Labor Market Returns to an Early Childhood Stimulation Inter- American Institutes for Research (AIR). 2015. Evaluation of vention in Jamaica.” Science 344:6187, 998–1001. LAUSD’s Instructional Technology Initiative Year 2 Report. Google–Temasek. 2018. https://www.thinkwithgoogle.com/intl/ Washington, DC. https://www.air.org/sites/default/files/down- en-apac/tools-resources/research-studies/e-conomy-sea loads/report/LAUSD-Instructional-Technology-Initiative-Exec- -2018-southeast-asias-internet-economy-hits-inflection-point/ Sum-August%202015.pdf. Heckman, J. J. and D. V. Masterov. 2007. “The Productivity Argu- Australia Indonesia Partnership for Economic Governance ment for Investing in Young Children.” Review of Agricultural (AIPEG). 2017. Women’s Economic Participation in Indonesia: Economics 29:3, 446–493. A Study of Gender Inequality in Employment, Entrepreneur- Heckman, J. J. 2000. “Policies to Foster Human Capital.” ship, and Key Enablers for Change. AIPEG, the Australian Research in Economics 54:1, 3–56. Department of Foreign Affairs and Trade (DFAT), and Monash J-PAL Evidence Review. 2019. “Will Technology Transform Edu- University’s Centre for Development Economics and Sustain- cation for the Better?” Abdul Latif Jameel Poverty Action Lab, ability (CDES), Melbourne, Australia. Cambridge, MA. Bahrdwaj, R. and N. Yarrow. Forthcoming. “EdTech in Indonesia: Little, A., 1995. Multigrade Teaching: A Review of Practice and Are We There Yet?” Research, Serial No. 12. Overseas Development Administra- Bandur, A. 2012. “School-based Management Developments: tion, London. Challenges and Impacts.” Journal of Educational Administra- Male, C. and Q. T. Wodon. 2017. Disability Gaps in Educational tion, September. Attainment and Literacy. World Bank, Washington, DC. Beatty, A., E. Berkhout, L. Bima, T. Coen, M. Pradhan, and D. Ministry of Education and Culture. 2018. Peraturan Menteri Pendi- Suryadarma. 2018. “Indonesia Got Schooled: 15 Years of Rising dikan dan Kebudayaan 15. Enrolment and Flat Learning Profiles,” RISE Working Paper Ministry of Education and Culture. 2018. Statistik Pendidikan Indo- 18/026, Research on Improving Systems of Education, Oxford, nesia 2017/2018. United Kingdom. Ministry of Education and Culture. 2019. Dapodik. Brinkman, S. A., A. Hasan, H. Jung, A. Kinnell, N. Nakajima, and Ministry of Education and Culture. 2019. National Exam Data, M. Pradhan. 2017. “The Role of Preschool Quality in Promoting PUSPENDIK. Child Development: Evidence from Rural Indonesia.” European Ministry of Finance. 2019. Peraturan Pemerintah Nomor 45 Tahun Early Childhood Education Research 25:4, 483-505. 2019. Cerdan-Infantes, P., Y. Makarova, S. Al-Samarrai, D. Chen, P. Ministry of Health. 2018. RISKESDAS, presentation “Hasil Utama Cerdan-Infantes, Y. Makarova, C. Al-Samarrai, D. Chen. 2013. RISKESDAS 2018.” Spending More or Spending Better: Improving Education Ministry of Home Affairs. 2019. http://bangda.kemendagri.go.id/ Financing in Indonesia. Policy brief. World Bank, Washington berita/baca_kontent/1399/pemerintah_dorong_program_ DC. satu_paud_satu_desa. De Ree, J. J. 2016. How Much Teachers Know and How Much It Ministry of Religious Affairs. 2018. EMIS. Matters in Class: Analyzing Three Rounds of Subject-specific Ministry of Research, Technology, and Higher Education. 2018. Test Score Data of Indonesian Students and Teachers. Policy Statistik Pendidikan Tinggi 2018. Research Working Paper WPS 7556. World Bank, Washington, Mizunoya, S., S. Mitra, and I. Yamasaki. 2016. “Towards Inclusive DC. Education: The impact of Disability on School Attendance in Denboba, A. D., R. K. Sayre, Q. T. Wodon, L. K. Elder, L. B. Raw- Developing Countries,” Innocenti Working Papers 2016-03, lings, and J. Lombardi. 2014. Stepping Up Early Childhood UNICEF Office of Research, Florence, Italy. Development: Investing in Young Children for High Returns. Mulder, N. 2005. Inside Indonesian Society: Cultural Change in World Bank, Washington, DC. Java, Penerbit Kanisius, Bangkok, Thailand. The Promise of Education in Indonesia • 47 Muralidharan, K., A. Singh, and A. J. Ganimian. 2017. “Disrupt- WHO. 2015. Global School-based Student Health Survey 2015. ing Education? Experimental Evidence on Technology-Aided https://www.who.int/ncds/surveillance/gshs/2015_Indonesia_ Instruction in India.” American Economic Review 109:4, 1426-60. GSHS_Fact_Sheet.pdf Olusanya, B., V. Krishnamurthy, and D. Wertlieb. 2018. Re: Global World Bank. 2015. Reviewing 10 Years of Indonesia’s School Initiatives for Early Childhood Development Should Be Disabil- Grants Program. World Bank, Washington, DC. https:// ity Inclusive. Pediatrics. 141. e20174055. 10.1542/peds.2017- www.indonesia-investments.com/upload/documents/ 4055. World-Bank-Bantuan-Operasional-Sekolah-Indonesia- Omidyar Network. 2019. Scaling Access and Impact: Realizing the School-Grants-Program-Indonesia-Investments.pdf Power of EdTech. www.omidyar.com/ ———. 2017. Indonesia Economic Quarterly: Decentralization that Organization for Economic Cooperation and Development OECD. delivers. December. World Bank, Jakarta, Indonesia. 2016. PISA 2015 Results: Excellence and Equity in Education. ———. 2018a. Indonesia Economic Quarterly: Learning More, Vol. 1. OECD, Paris, France. Growing Faster. June. World Bank, Jakarta, Indonesia. Pradhan, M., D. Suryadarma, A. Beatty, M. Wong, A. Gaduh, A. ———. 2018b. World Development Report 2018: Learning to Alisjahbana, and R. Artha, R. 2014. “Improving Educational Realize Education’s Promise. World Bank, Washington, DC. Quality Through Enhancing Community Participation: Results ———. 2018c. Mapping Indonesia’s Civil Service. World Bank, from a Randomized Field Experiment in Indonesia.” American Washington, DC. http://documents.worldbank.org/curated/ Economic Journal: Applied Economics 6:2, 105-126. en/643861542638957994/pdf/126376-Mapping-Indonesia PricewaterhouseCoopers. 2017. The World in 2050: How -Civil-Service-14977.pdf Will the Global Economic Order Change? February. ———. 2018d. Growing Smarter: Learning and Equitable Devel- PricewaterhouseCoopers, London, United Kingdom. opment in East Asia and Pacific. World Bank, Washington, DC. Psacharopoulos, G. and H.A. Patrinos. 2018. Returns to Invest- ———. 2019a. Indonesia Economic Quarterly: Oceans of Oppor- ment in Education: A Decennial Review of the Global Litera- tunity. June. World Bank, Jakarta ture. Policy Research Working Paper, WPS 8402. World Bank, ———. 2019b. World Development Report: The Changing Nature Washington, DC. of Work. World Bank, Washington, DC. Ragatz, A. B. 2015. A Video Study of Teaching Practices in TIMSS ———. 2019c. Ending Learning Poverty: What Will It Take? World Eighth Grade Mathematics Classrooms: Understanding What Bank, Washington, DC. Teaching Practices are Used, Why They are Used and How ———. Forthcoming. Primary Education in Remote Indonesia: They Relate to Student Learning. World Bank, Washington DC. Survey Results from West Kalimantan and East Nusa Tenggara. Rokx, C., A. Subandoro, and P. Gallagher. 2018. Aiming High: World Travel and Tourism Council. 2015. Benchmark Report— Indonesia’s Ambition to Reduce Stunting. World Bank, Wash- Indonesia. ington, DC. Yarrow, N., R. Afkar, S. Sudarti, and R. Cooper. Forthcoming. Rosser, Andrew; Fahmi, Mohamad. 2016. The Political Economy Gender and Inclusion in Indonesian Education. of Teacher Management in Decentralized Indonesia. Policy Research Working Paper;No. 7913. World Bank, Washington, Government laws and regulations DC. Law of the Republic of Indonesia 22/1999 Concerning Regional Rosser, Andrew. 2018. Beyond Access: Making Indonesia’s Educa- Administrations. tion System Work. Lowy Institute, Sydney Australia Presidential Regulation 107/2017. Shaeffer, S. and R. Arlianti. 2019. Education Innovations in East Presidential Regulation 129/2018. Java: A Review and Analysis of INOVASI’s Stocktake Study, Law 23/2014 on Regional Autonomy. INOVASI, Jakarta, Indonesia. Law 33/2004 on Fiscal Balance. Simpatika. 2018. Permendikbud 58/2015. Statistics Indonesia. 2015. Badan Pusat Statistik (BPS), the Central Permendikbud 15/2018. Bureau of Statistics, Jakarta, Indonesia. Permendikbud 32/2018. Suryadarma, D. 2012. “How Corruption Diminishes the Effective- Permendikbud 51/2018. ness of Public Spending on Education in Indonesia.” Bulletin of Undan-Undang 8/2016. Indonesian Economic Studies 48:1, 85-100. Government Regulation 12/2019. Susenas. Various years, Badan Pusat Statistik. UNESCO. 2018. Education and Disability: Analysis of Data from 49 Countries. Information Paper 49, March. UNESCO Institute Websites of Statistics, Montreal. http://apkapm.data.kemdikbud.go.id/index.php/cberanda/ UNESCO. 2019. Cali Statement. https://unesdoc.unesco.org/ apkapmsekolahmadrasah?kode_wilayah=000000& ark:/48223/pf0000370910 tahun=2018 UNICEF. 2012. We Like Being Taught: A Study on Teacher Absen- https://data.worldbank.org/indicator/SE.SEC.NENR?locations= teeism in Papua and West Papua, Technical Report. ID&view=chart UNICEF. 2014. UNICEF Blog, Remote Schools: Inspiring Papua’s http://bangda.kemendagri.go.id/berita/baca_kontent/1399/ Children to Go the Extra Mile for Education. Monday, 14 April pemerintah_dorong_program_satu_paud_satu_desa 2014. https://data.worldbank.org/indicator/SE.SEC.NENR?locations= Usman, S., Akhmadi, and Suryadarma, D. 2004. When Teachers ID&view=chart are Absent: Where Do They Go and What is the Impact on https://www.bappenas.go.id/files/rpjmn/Narasi%20RPJMN%20 Students? SMERU Research Institute, Jakarta. IV%202020-2024_Revisi%2028%20Juni%202019.pdf Vargas-Barón. E., J. Small, D. Wertlieb, H. Hix-Small, R. Gómez https://www.bps.go.id/statictable/2010/03/19/1525/indikator Botero, K. Diehl, P. Vergara, and P. Lynch. 2019. Global Survey -pendidikan-1994-2018.html of Inclusive Early Childhood Development and Early Childhood https://www.indexmundi.com/indonesia/demographics_ Intervention Programs. RISE Institute, Washington, DC. profile.html Vernez, G., R. Karam, and J. H. Marshall. 2016. Transforming https://www.topuniversities.com/university-rankings/world Indonesia’s Centralized Education System to School-Based -university-rankings/2020 Management. RAND Corporation, Santa Monica, CA. https://www.worldbank.org/en/publication/human-capital WHO. 2011. World Report on Disability for children aged 0–14, petabahasa.kemendikbud.go.id following Global Burden of Disease 2004. 48 • The Promise of Education in Indonesia Acknowledgments This report was prepared by a team led by Noah Yarrow, com- prising Rythia Afkar, Deepali Gupta, Susiana Iskandar, Ratna Kesuma, Javier Luque, Sylvia Njotomihardjo, Andy Ragatz, Rosfita Roesli, Sheldon Schaeffer, Ruwiyati Purwana, Michael Tjahjadi, and Wisnu Harto Adi Wijoyo. The team is grateful to the peer reviewers Kathleen Whimp, Dewi Susanti, Fadila Cail- laud, and Elizabeth Ninan, as well as to Camilla Holmemo for inputs and guidance. The report was prepared under the over- all guidance of Rodrigo Chaves (Country Director, Indonesia and Timor Leste) and Toby Linden (Practice Manager, Educa- tion, East Asia and Pacific). A team at Communications Development, Inc., led by Bruce Ross-Larson and including Joe Caponio, Jeremy Clift, Debra Naylor, Peter Redvers-Lee, and Elaine Wilson, edited, designed, and laid out the report. Funding from the Australian Government, Department of Foreign Affairs and Trade supported the report’s research and production. The Promise of Education in Indonesia • 49