© 2023 The World Bank 1818 H Street NW, Washington DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved This work is a product of the staff of The World Bank. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Attribution—Please cite the work as follows: “World Bank. 2023. Indonesia Poverty Assessment – Pathways Towards Economic Security. © World Bank.” All queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. CONTENTS Abbreviations ........................................................................................................................................................................................................................................................................... i Foreword....................................................................................................................................................................................................................................................................................... iii Pendahuluan ............................................................................................................................................................................................................................................................................. iv Acknowledgement ............................................................................................................................................................................................................................................................... v Executive Summary............................................................................................................................................................................................................................................................... ix Ringkasan Eksekutif ............................................................................................................................................................................................................................................................... xix 1. Context ................................................................................................................................................................................................................................................................................. 1 Looking back ..................................................................................................................................................................................................................................................................... 1 Looking ahead ................................................................................................................................................................................................................................................................. 3 This report ........................................................................................................................................................................................................................................................................... 4 2. Poverty and Inequality Trends.............................................................................................................................................................................................................................. 7 National ................................................................................................................................................................................................................................................................................ 9 Sub-national........................................................................................................................................................................................................................................................................ 14 Opportunities over the lifecycle.............................................................................................................................................................................................................................. 17 3. Poverty Reduction.......................................................................................................................................................................................................................................................... 29 Poverty reduction............................................................................................................................................................................................................................................................. 29 Inequality .............................................................................................................................................................................................................................................................................. 47 4. Shocks..................................................................................................................................................................................................................................................................................... 55 Health and employment............................................................................................................................................................................................................................................. 56 Climate change and natural disasters................................................................................................................................................................................................................. 58 Prices ....................................................................................................................................................................................................................................................................................... 63 COVID-19 .............................................................................................................................................................................................................................................................................. 64 5. Policy Recommendations......................................................................................................................................................................................................................................... 73 Creating better opportunities.................................................................................................................................................................................................................................. 74 Protecting against poverty......................................................................................................................................................................................................................................... 75 Financing pro-poor investments............................................................................................................................................................................................................................ 76 Improving future policies............................................................................................................................................................................................................................................ 77 References .................................................................................................................................................................................................................................................................................. 79 Annex ............................................................................................................................................................................................................................................................................................ 85 List of Figures Figure 2.1: GDP growth (LHS) and GDP -per-capita (RHS) from 1990 to 2021............................................................................................................................ 7 Figure 2.2: Terms-of-trade in US$ billions and growth of gross fixed capital formation from 2007 to 2019............................................................ 7 Figure 2.3: GDP composition from 1997 to 2021........................................................................................................................................................................................... 7 Figure 2.4: Annualized employment growth and change in productivity, bubble size reflects value-add of sec............................................... 7 Figure 2.5: Terms-of-trade in US$ billions and growth of gross fixed capital formation from 2007 to 2019............................................................ 8 Figure 2.6: Poverty headcount rates using $1.90, $3.20, and $5.50 per-day 2011 PPP as well as national poverty line (NPL)...................... 10 Figure 2.7: Relative and absolute change in poverty at US$ 3.20 2011 PPP from 2009 to 2018/2019.......................................................................... 10 Figure 2.8: Poverty rates for Indonesia and its economic peers............................................................................................................................................................ 11 Figure 2.9: Log GDP per capita (PPP) vs poverty rate for peers.............................................................................................................................................................. 11 Figure 2.10: Share of population classified as structurally poor, economically insecure, and economically secure............................................... 12 Figure 2.11: Consumption growth among the bottom 40 and across the population, for Indonesia and peer countries................................ 13 Figure 2.12: Gini coefficient for Indonesia and its economic peers....................................................................................................................................................... 13 Figure 2.13: Gini coefficient from 2002 to 2022................................................................................................................................................................................................. 13 Figure 2.14: Unfair inequality in consumption and labor income, across years............................................................................................................................. 13 Figure 2.15: Poverty rates based on official poverty lines, by regions.................................................................................................................................................. 16 Figure 2.16: Poverty rates for Nusa Tenggara (NT), Maluku Papua (MP) and other regions.................................................................................................... 16 Figure 2.17: Share of the poor and the general population in rural and urban areas................................................................................................................ 17 Figure 2.18: Share of the (extreme) poor, by region........................................................................................................................................................................................ 17 Figure 2.19: Educational attainment among young adults (19-25 yo) in Indonesia................................................................................................................... 17 Figure 2.20: Educational attainment among Indonesian household heads.................................................................................................................................... 17 Figure 2.21: Access to infrastructure services in 2015.................................................................................................................................................................................... 18 Figure 2.22: Access to infrastructure services in 2022.................................................................................................................................................................................... 18 Figure 2.23: Infant mortality for Indonesia and its economic peers...................................................................................................................................................... 19 Figure 2.24: Stunting levels for Indonesia and its economic peers........................................................................................................................................................ 19 Figure 2.25: Learning outcomes in Indonesia and its economic peers............................................................................................................................................... 19 Figure 2.26: Maternal mortality in Indonesia and its economic peers................................................................................................................................................. 19 Figure 2.27: Childhood mortality rates, by wealth quintile......................................................................................................................................................................... 21 Figure 2.28: Vaccination rates and dietary diversity among children, by wealth quintile........................................................................................................ 21 Figure 2.29: Total fertility rates among women, by wealth quintile....................................................................................................................................................... 21 Figure 2.30: Educational attainment among the top 20 (top) and bottom 20 percent (bottom), household heads............................................ 21 Figure 2.31: Sub-national human capital index relative to GPD per capita...................................................................................................................................... 22 Figure 2.32: Childhood mortality rates in 2012 and 2020, by region.................................................................................................................................................... 22 Figure 2.33: Labor force participation, by education and for women.................................................................................................................................................. 24 Figure 2.34: Real labor income, by education from 2001 to 2021.......................................................................................................................................................... 24 Figure 2.35: Share of household heads by sector and type of employment, by poverty status, for 2003 and 2021.............................................. 25 Figure 2.36: Formalization by education, from 2001 to 2021..................................................................................................................................................................... 25 Figure 2.37: Gender gap in years of education completed among adults and 15-19-year-olds......................................................................................... 26 Figure 2.38: Attitudes around women’s work, among women and men.......................................................................................................................................... 26 Figure 2.39: Poor and non-poor households in 2019, by demographic classification............................................................................................................... 27 Figure 3.1: Annualized contributions of growth and redistribution to poverty reduction.................................................................................................. 30 Figure 3.2: Elasticity (left) and semi-elasticity (right) of poverty to per-capita growth........................................................................................................... 30 Figure 3.3: Elasticity of poverty to per capita growth.................................................................................................................................................................................. 30 Figure 3.4: Ravallion-Huppi decomposition of poverty reduction, by urban/rural................................................................................................................... 30 Figure 3.5: Poverty vs. urbanization rate by year............................................................................................................................................................................................. 31 Figure 3.6: Contributions of endowment and return to poverty reduction, by urban/rural............................................................................................... 31 Figure 3.7: Gender and age-cohort poverty rates for 2014...................................................................................................................................................................... 32 Figure 3.8: GDP growth (LHS) and GDP -per-capita (RHS) from 1990 to 2021............................................................................................................................. 32 Figure 3.9: Annualized changes in poverty rate and population share from 2014 to 2019 for demographic and economic groups of households............................................................................................................................................................................................................................. 33 Figure 3.10: Female effect on poverty status, with and without controlling for dependency ratio, for 2019............................................................. 33 Figure 3.11: Annualized real wages by sector...................................................................................................................................................................................................... 33 Figure 3.12: Wage premium for gender, formalization, location, education, and sector.......................................................................................................... 34 Figure 3.13: Contributions of endowment and return to wages, by gender, formalization, and location.................................................................... 34 Figure 3.14: Annualized employment growth (2014 to 2019) by median sector wage and employment share (bubble size) in 2019 35 Figure 3.15: Average volume and annual growth of exports of goods and services................................................................................................................. 36 Figure 3.16: Export to GDP ratio vs. change in GDP per capita................................................................................................................................................................. 36 Figure 3.17: Manufacturing export competitiveness...................................................................................................................................................................................... 36 Figure 3.18: Export sophistication............................................................................................................................................................................................................................... 36 Figure 3.19: Growth in labor productivity.............................................................................................................................................................................................................. 36 Figure 3.20: Average annual growth (2010 to 2019) of unit labor cost per output and labor cost per hour.............................................................. 36 Figure 3.21: Sectoral contributions to poverty reduction............................................................................................................................................................................ 37 Figure 3.22: Sectoral composition of household heads from 2006 to 2019, by poverty status and urban/rural...................................................... 37 Figure 3.23: Share of working-age population in high-skilled jobs vs. log GDP per capita (Indonesia highlighted).............................................. 39 Figure 3.24: Changes in endowment and returns to education for real wages by income deciles, 2014 versus 2019........................................ 39 Figure 3.25: Consumption share for rent payments, by urban/rural and decile in 2019.......................................................................................................... 40 Figure 3.26: Hours in lost congestion by city per year................................................................................................................................................................................... 40 Figure 3.27: Annual inflation by product category........................................................................................................................................................................................... 42 Figure 3.28: Retail price for rice in US$ per kg, across countries.............................................................................................................................................................. 42 Figure 3.29: Government expenditure relative to GDP per capita......................................................................................................................................................... 44 Figure 3.30: Government revenues relative to GDP per capit................................................................................................................................................................... 44 Figure 3.31: Absolute (left) and relative (right) incidence of indirect tax by income decile................................................................................................... 44 Figure 3.32: Total consumption of exempted goods and services by consumption decile (2019)................................................................................... 45 Figure 3.33: Poverty impact, cost and efficiency of modeled direct transfers and subsidies................................................................................................ 45 Figure 3.34: Efficiency of individual transfer and subsidy programs..................................................................................................................................................... 45 Figure 3.35: Government education expenditure relative to GDP per capita................................................................................................................................. 47 Figure 3.36: Government health expenditure relative to GDP per capita.......................................................................................................................................... 47 Figure 3.37: Consumption growth across periods, annualized................................................................................................................................................................. 48 Figure 3.38: Theil inequality decomposition between urban/rural and provinces...................................................................................................................... 48 Figure 3.39: Relative real wages by education, compared to tertiary education.......................................................................................................................... 49 Figure 3.40: Share of sector of employment of household heads.......................................................................................................................................................... 49 Figure 3.41: Share of household consumption by category and decile, for 2019........................................................................................................................ 50 Figure 3.42: Impact of fiscal policy on Gini index, across countries....................................................................................................................................................... 51 Figure 3.43: Concentration of direct transfers and subsidies, across consumption deciles, in 2017................................................................................. 52 Figure 3.44: Share of benefits (direct transfers and subsidies) relative to consumption, by consumption deciles................................................. 52 Figure 3.45: GIn-kind education benefit as share of market income.................................................................................................................................................... 52 Figure 3.46: In-kind health benefit as share of market income................................................................................................................................................................ 52 Figure 4.1: Ratio of idiosyncratic to covariate susceptibility to fall into poverty, for 2011 and 2019, by urban and rural................................. 55 Figure 4.2: Account ownership among adults 15+....................................................................................................................................................................................... 55 Figure 4.3: Barriers to open accounts, 2021....................................................................................................................................................................................................... 55 Figure 4.4: Access to emergency funds and their sources, 2021.......................................................................................................................................................... 55 Figure 4.5: Share of working age adults reporting sickness in bottom 40 and top 60........................................................................................................... 57 Figure 4.6: Facilities visited for outpatient and inpatient care, for bottom 40 and top 60.................................................................................................... 57 Figure 4.7: Share of household members with insurance coverage for bottom 40 and top 60....................................................................................... 57 Figure 4.8: Type of health insurance for bottom 40 and top 60............................................................................................................................................................ 57 Figure 4.9: Number of natural disasters................................................................................................................................................................................................................ 58 Figure 4.10: Estimated agricultural yields in 2030............................................................................................................................................................................................. 58 Figure 4.11: Share of households adopting adverse strategies to cope with crises.................................................................................................................... 61 Figure 4.12: Contributions to consumption growth for low-ambition scenario, by income decile.................................................................................. 62 Figure 4.13: Contributions to consumption growth for medium-ambition scenario, by income decile....................................................................... 62 Figure 4.14: Year-on-year inflation for 2022, by item group........................................................................................................................................................................ 63 Figure 4.15: Loss of purchasing power due to shock in prices, as in Scenario A........................................................................................................................... 63 Figure 4.16: Unmitigated poverty impact (in pp) of price shocks ......................................................................................................................................................... 64 Figure 4.17: Change of consumption after compensating for price shocks, by decile, for Scenario A .......................................................................... 64 Figure 4.18: Google mobility and Oxford stringency index as well as number of new COVID-19 cases....................................................................... 64 Figure 4.19: Number of added/lost workers since previous year, by sector and informal/formal...................................................................................... 66 Figure 4.20: Number of added/lost workers since previous year, by type of employment.................................................................................................... 66 Figure 4.21: Number of added/lost workers since previous year, by gender and informal/formal................................................................................... 66 Figure 4.22: Likelihood of employment in August 2020, after controlling for individual characteristics....................................................................... 66 Figure 4.23: Share of social protection beneficiaries in March 2021..................................................................................................................................................... 67 Figure 4.24: Share of program beneficiaries assessing benefits as adequate.................................................................................................................................. 67 Figure 4.25: Consumption growth Incidence curve for 2020 to 2021, by urban/rural.............................................................................................................. 69 Figure 4.26: Ravallion-Huppi decomposition for 2020 to 2021................................................................................................................................................................ 69 Figure 4.27: Coping strategies of households in the bottom 40 and top 20................................................................................................................................... 69 Figure 4.28: Share of students attending face-to-face learning............................................................................................................................................................... 70 Figure 4.29: Share of students using mobile learning apps or online schooling in November 2020............................................................................... 70 Figure 5.1: GDP growth (LHS) and GDP -per-capita (RHS) from 1990 to 2021............................................................................................................................. 73 List of Boxes Box 2.1: Poverty for Indonesia is assessed based on revised 2011 PPP estimates .............................................................................................................. 10 Box 2.2: Definition of economic insecurity................................................................................................................................................................................................... 12 Box 2.3: The advantage of using absolute poverty lines to compare poverty across Indonesia’s provinces...................................................... 15 Box 2.4: Access to opportunities in Indonesia’s lagging regions.................................................................................................................................................... 23 Box 3.1: Analytical framework for poverty reduction in Indonesia................................................................................................................................................ 31 Box 3.2: Commitment to Equity (CEQ) framework.................................................................................................................................................................................. 43 Box 3.3: Top incomes and measurement issues........................................................................................................................................................................................ 48 Box 3.4: Education and health services in the context of decentralization.............................................................................................................................. 53 Box 4.1: Voices, formal and informal community mechanisms to mitigate shock impacts........................................................................................... 59 Box 4.2: Voices, impact of a health shock followed by a mudslide................................................................................................................................................ 60 Box 4.3: Modeling parameters for distributional climate impact simulations........................................................................................................................ 62 Box 4.4: Voices, residents in Kediri on COVID-19 effects....................................................................................................................................................................... 65 Box 4.5: Voices, receiving COVID social assistance................................................................................................................................................................................... 68 Box 4.6: Voices, challenges of online learning............................................................................................................................................................................................ 71 ABBREVIATIONS AFC Asian Financial Crisis ANC Ante-natal Care Asabri Asuransi Angkatan Bersenjata Republik Indonesia; Indonesian Armed Forces Insurance ASEAN Association of Southeast Asian Nations B40 Bottom 40 Percent of Households Bansos Bantuan Sosial; Social Assistance BPJS Badan Penyelenggara Jaminan Sosial; Social Security Administration Agency BPS Badan Pusat Statistik; Central Bureau of Statistics BRI Bank Rakyat Indonesia; Indonesian People's Bank CEQ Commitment to Equity CIT Corporate Income Tax COP26 United Nations Climate Change Conference CPI Consumer Price Index D Number of Dependent in the Household DTKS Data Terpadu Kesejahteraan Sosial; Integrated Social Welfare Database E Number of Earners in the Household ECD Early Childhood Development EM-DAT Emergency Events Database EP Extreme Poor EU European Union FAO Food and Agriculture Organization GDP Gross Domestic Product GHG Greenhouse Gas GIEWS Global Information and Early Warning System on Food and Agriculture GVCs Global Value Chains GoI Government of Indonesia HIC High Income Country HiFy High Frequency Household Phone Survey IFLS Indonesian Family Life Survey IFPRI International Food Policy Research Institute ILO International Labor Organization IPL International Poverty Line Jamkesda Jaminan Kesehatan Daerah; Regional Health Insurance Jamkesmas Jaminan Kesehatan Masyarakat; Public Health Insurance JHT Jaminan Hari Tua; Pension Program JKK Jaminan Kecelakaan Kerja; Work-accident Insurance Program JKM Jaminan Kematian; Death Insurance Program JKN Jaminan Kesehatan Nasional; National Health Insurance JKP Jaminan Kehilangan Pekerjaan; Unemployment Insurance KOTAKU Kota Tanpa Kumuh; City Without Slums Program LIC Low Income Country Pathways Towards Economic Security Indonesia Poverty Assessment i LMIC Lower Middle-Income Country M40 Middle 40 Percent of Households Mo Months MoH Kementerian Kesehatan Republik Indonesia; Ministry of Health MoSAUBR Menteri Pendayagunaan Aparatur Negara dan Reformasi Birokrasi Republik Indonesia; Ministry of State Apparatus Utilization and Bureaucratic Reform MP Maluku-Papua ND-GAIN Notre Dame Global Adaptation Initiative NDC Nationally Determined Contribution NEET Neither in School nor Employed NGO Nongovernment Organization NP Non-poor NPL National Poverty Line NT Nusa-Tenggara NTM Non-tariff Measures OECD Organisation for Economic Co-operation and Development OLS Ordinary Least Squares PA Poverty Assessment PAUD Pendidikan Anak Usia Dini; Early Childhood Education PBI Penerima Bantuan Iuran; Recipient Contribution Assistance PEKKA Pemberdayaan Perempuan Kepala Keluarga; Empowerment of Female Heads of Family PIP Program Indonesia Pintar; Smart Indonesia Program PISA Programme for International Student Assessment PIT Personal Income Tax PKH Program Keluarga Harapan; Family Hope Program PMT Proxy-means Test PP Percentage Point PPP Purchasing Power Parity Puskesmas Pusat Kesehatan Masyarakat; Community Health Center R Rounds of HiFy Survey R&D Research and Development Raskin Beras Miskin; Rice for the Poor Program Sakernas Survei Angkatan Kerja Nasional; Labor Survey Sembako Sembilan Bahan Pokok; Affordable Basic Food Program SPS Sanitary and Phytosanitary Measures Susenas Survei Sosial Ekonomi Nasional; National Socio-Economic Survey T20 Top 20 Percent of Households T60 Top 60 Percent of Households TFR Total Fertility Rate UMIC Upper Middle-Income Country VA Value-Add VAT Value-Added Tax WelTrAC Welfare Tracking in the Aftermath of Crisis ii Pathways Towards Economic Security Indonesia Poverty Assessment FOREWORD I ndonesia arguably met its goal to eliminate extreme poverty when it reached 1.5 percent in 2022. Sustained economic growth combined with social protection has made this progress possible. Indonesia can now set its sights higher to improve the lives of the still one-third of Indonesians who remain economically insecure. As Indonesia aims to become a high-income country by 2045, our analysis in this report uncovers opportunities as well as some important roadblocks to further progress. Even though economic growth has contributed to poverty reduction, nearly all sectors in rural, agricultural areas and in cities suffer from low productivity, while human capital development lags peer countries and half of Indonesian women remain excluded from the labor force. Indonesians need better work opportunities offering higher income in higher productivity sectors. Despite a fast- growing digital sector, only one in ten Indonesian workers has a high-skilled job, and not enough workers have skills to take advantage of these opportunities when they arise. These are areas in which policies can make a difference. Our analysis offers some perspectives that lead to a few recommendations. One set of recommendations focus on creating better opportunities. Integration into global value chains would increase Indonesia’s productivity and help take advantage of its growing digital economy. Urban areas need investments to allow them to become the engines of productivity growth we have seen in other countries, while enhancing agricultural productivity would provide better livelihoods for farmers. More affordable, quality childcare would in turn help open opportunities for women. Another set of recommendations focus on protecting people from staying and falling into poverty. Indonesia is prone to shocks, especially from weather-related incidents. Between 1990 and 2021, Indonesia experienced more than 300 natural disasters affecting more than 11 million people, with climate-related disasters accounting for around 70 percent of total disasters in this period. As usual, the poor and economically insecure carry a disproportionate burden when a disaster strikes. Like many countries Indonesia needs to look at scaling up social protection, including both social assistance and insurance, as well as increasing financial inclusion. It is also vital to enhance the resilience of infrastructure to shocks. These measures would require resources, but Indonesian policymakers have options to increase financing for these “pro-poor investments”. As our analysis of taxation and subsidy policies points out, Indonesia has opportunities to increase tax revenues while reducing spending on less effective and often regressive energy and agricultural subsidies. In this context, improving the efficiency and quality of sub-national governments’ administration and spending, especially on education and healthcare, is also key to increase the quality of public services. Our hope is that this Poverty Assessment will help inform and broaden public dialogue on opportunities and challenges as well as on possible solutions to creating better opportunities and protecting against poverty. Satu Kahkonen Country Director World Bank Indonesia Pathways Towards Economic Security Indonesia Poverty Assessment iii PENDAHULUAN IIndonesia boleh dibilang telah mencapai tujuannya untuk memberantas kemiskinan ekstrim ketika kemiskinan tersebut mencapai 1,5 persen pada tahun 2022. Pertumbuhan ekonomi yang berkelanjutan digabung dengan perlindungan sosial telah memungkinkan kemajuan ini. Indonesia sekarang dapat menetapkan sasaran yang lebih tinggi untuk meningkatkan kehidupan sepertiga penduduk Indonesia yang secara ekonomi masih tidak aman. Karena Indonesia bertujuan untuk menjadi negara berpenghasilan tinggi pada tahun 2045, analisis kami dalam laporan ini mengungkap beberapa peluang serta hambatan penting untuk kemajuan lebih lanjut. Meskipun pertumbuhan ekonomi telah memberi kontribusi terhadap pengentasan kemiskinan, hampir semua sektor, di pedesaan, daerah pertanian dan di perkotaan, memiliki produktivitas yang rendah, sementara pembangunan sumber daya manusia tertinggal dari negara-negara yang setara dan setengah dari perempuan Indonesia tetap tersisih dari angkatan kerja. Masyarakat Indonesia membutuhkan kesempatan kerja yang lebih baik yang menawarkan penghasilan lebih tinggi di sektor-sektor dengan produktivitas yang lebih tinggi. Meskipun sektor digital berkembang pesat, hanya satu dari sepuluh pekerja Indonesia yang memiliki pekerjaan dengan keterampilan tinggi, dan tidak cukup banyak pekerja yang memiliki keterampilan yang tepat untuk memanfaatkan peluang ini di saat peluang tersebut muncul. Ini adalah wilayah di mana kebijakan dapat membuat perbedaan. Analisis kami menawarkan beberapa perspektif yang mengarah pada beberapa rekomendasi. Serangkaian rekomendasi berfokus pada menciptakan peluang yang lebih baik. Integrasi ke dalam rantai nilai global dapat meningkatkan produktivitas Indonesia dan membantu memanfaatkan pertumbuhan ekonomi digitalnya. Daerah perkotaan membutuhkan investasi agar daerah-daerah tersebut dapat menjadi mesin pertumbuhan produktivitas yang telah kita lihat di negara-negara lain, sementara meningkatkan produktivitas pertanian dapat memberikan penghidupan yang lebih baik bagi para petani. Fasilitas penitipan anak yang lebih terjangkau dan berkualitas dapat membantu membuka peluang bagi perempuan. Serangkaian rekomendasi lainnya berfokus pada melindungi masyarakat dari keterpurukan dalam kemiskinan. Indonesia rentan terhadap guncangan ekonomi, terutama dari kejadian-kejadian terkait cuaca. Antara tahun 1990 dan 2021, Indonesia mengalami lebih dari 300 bencana alam yang menimpa lebih dari 11 juta orang, dengan bencana terkait iklim mencapai sekitar 70 persen dari total bencana pada periode ini. Seperti biasa, masyarakat miskin dan tidak aman secara ekonomi memikul beban yang tidak proporsional ketika terjadi bencana. Seperti banyak negara lainnya, Indonesia perlu meningkatkan perlindungan sosial, termasuk bantuan dan jaminan sosial, serta meningkatkan inklusi keuangan. Penting juga untuk meningkatkan ketangguhan infrastruktur terhadap guncangan. Langkah-langkah tersebut membutuhkan sumber daya, tetapi para pembuat kebijakan Indonesia memiliki pilihan untuk meningkatkan pembiayaan bagi “investasi yang berpihak pada masyarakat miskin” tersebut. Seperti yang ditunjukkan oleh analisis kami mengenai kebijakan perpajakan dan subsidi, Indonesia memiliki peluang untuk meningkatkan penerimaan pajak sambil mengurangi pengeluaran untuk subsidi energi dan pertanian yang kurang efektif dan seringkali bersifat regresif. Dalam konteks ini, peningkatan efisiensi dan kualitas administrasi dan belanja pemerintah daerah, terutama untuk pendidikan dan kesehatan, juga menjadi kunci untuk meningkatkan kualitas pelayanan publik. Harapan kami adalah Kajian Kemiskinan ini akan membantu memberi informasi dan memperluas dialog publik tentang peluang dan tantangan serta kemungkinan solusi untuk menciptakan peluang yang lebih baik dan melindungi dari kemiskinan. Satu Kahkonen Country Director World Bank Indonesia iv Pathways Towards Economic Security Indonesia Poverty Assessment ACKNOWLEDGEMENT The Poverty Assessment was co-led by Rabia Ali (Senior Economist, EEAPV) and Utz Pape (Senior Economist, EEAPV). The core team consisted of Samuel Nursamsu (Economist, EEAPV), Anissa Rahmawati (Consultant, EEAPV), Imam Setiawan (Economist, EEAPV) and Putu Sanjiwacika Wibisana (Consultant, EEAPV) as well as assistance from Christal Ng (Consultant, EEAPV). The core team was supported by Dyah Nugraheni (Program Assistance, EACIF). Background papers and analyses were prepared by Rabia Ali (Senior Economist, EEAPV), Ade Febriady (Consultant, EEAPV), Daniel Halim (Economist, HGNDR), Sean Hambali (Consultant, SEAS2), Utz Pape (Senior Economist, EEAPV), Timothy (Irfan) Kortschak (Consultant, EEAPV), Ririn Purnamasari (Senior Economist, EEAPV), Virgi Sari (Economist, EEAPV), Imam Setiawan (Economist, EEAPV), Bambang Suharnoko (Koko) Sjahrir (Senior Economist, EEAPV), and Iqbal Wibisono (Consultant, EEAPV). The team would like to thank the following colleagues for comments and inputs: Rythia Afkar (Economist, HHCDR), Salman Alibhai (Senior Financial Sector Specialist, EEAF2), Vikas Choudhary (Senior Agriculture Specialist, SEAAG), Zelalem Debebe (Senior Economist – Health, HEAHN), Anastasiya Denisova (Senior Economist, HEASP), Hannah Duncan (Consultant, SEAS2), Sara Giannozzi (Senior Social Protection Specialist, HEASP), Gracia Hadiwidjaja (Social Protection Specialist, HEASP), Indira Hapsari (Senior Economist, EEAM2), David Kaczan (Senior Economist, SEAE1), Csilla Lakatos (Senior Economist, EEAMS2), Wael Mansour (Senior Economist, EEAM2), Dino Merotto (Lead Economist, HSPJB), Somil Nagpal (Senior Health Specialist, HEAHN), Daniel Nieto (Senior Public Sector Management Specialist, EEAG1), Shinsaku Nomura (Senior Economist, HEAED), Anthony Obeyesekere (Economist, EEAM2), Rong Qian (Senior Economist, EEAM2), Francesco Strobbe (Lead Financial Sector Specialist, EEAF2), Ekki Syamsulhakim (Senior Social Protection Specialist, HEASP), Emcet Tas (Senior Social Development Specialist, SEAS2), Alika Tuwo (Agriculture Economist, SEAAG) and Asha Williams (Senior Social Protection Specialist, HEASP). The team also received important advice from the peer reviewers Nadia Belghith (Senior Economist, EEAPV), Ana Maria Boudet (Senior Social Scientist, EPVGE), Carlos Castelan (Lead Economist, ELCPV), P. Facundo Cuevas (Senior Economist, EAEPV), Ambar Narayan (Lead Economist, EAWPV) and Sailesh Tiwari (Lead Economist, EEAPV). Guidance was provided by Rinku Murgai (Practice Manager, EEAPV), Matthew Wai-Poi (Lead Economist, EEAPV), Habib Rab (Lead Economist, EEAM2), Achim Schmillen (Practice Leader, HEADR), Bolormaa Amgaabazar (Operations Manager, EACIF), Aaditya Mattoo (Chief Economist, EAPCE), Hassan Zaman (Regional Director, EEADR) and Satu Kahkonen (Country Director, EACIF). Aldo Morri (Consultant, HSPJB) edited the report thoroughly and Robert Waiharo (Temporary, EAEM1) helped with the final layout while Florence Micheltorena prepared infographics. The team held consultations and received important comments from BAPPENAS, BKF, BPS and TNP2K as well as SMERU. Financial support for this work was provided by the Government of Australia’s Department of Foreign Affairs and Trade through the Australia-World Bank Indonesia Partnership (ABIP). Pathways Towards Economic Security Indonesia Poverty Assessment v vi Pathways Towards Economic Security Indonesia Poverty Assessment Pathways Towards Economic Security Indonesia Poverty Assessment vii Indonesia has made impressive gains in reducing poverty, with previously lagging regions catching up, and the Government’s goal to eliminate extreme poverty by 2024 practically met. Photo: © Josh Estey/World Bank viii Pathways Towards Economic Security Indonesia Poverty Assessment EXECUTIVE SUMMARY Overview Indonesia can build on its impressive track-record of poverty reduction to tackle more ambitious poverty reduction targets. Indonesia has made impressive gains in reducing poverty, with previously lagging regions catching up, and the Government’s goal to eliminate extreme poverty by 2024 practically met. As an aspiring upper middle-income country, however, Indonesia may want to widen its focus beyond extreme poverty by moving from the US$ 1.90 2011 PPP poverty line to higher lines for middle-income countries. The focus should also include economically insecure households susceptible to falling back into poverty. Is Indonesia’s current effort ready for this challenge? Human capital outcomes are disappointing and worrying geographic disparities remain. Low productivity still prevents households from becoming economically secure. Shocks, including from climate change, continue to threaten reversal in poverty gains. We identify several major pathways to tackle these challenges in a comprehensive and sustainable manner (Figure ES1). (i) Create better opportunities in higher productivity and low-carbon work to help households become economically secure. (ii) Improve social protection and financial inclusion to mitigate harm from future shocks. (iii) Develop a more effective fiscal system for more pro-poor investments contributing to human capital through better public service delivery. (iv) Close data and knowledge gaps to improve future policies to support this agenda. FIGURE ES1: Four pathways with policy priorities (green) towards economic security can tackle key challenges (orange) faced by Indonesia COVID-19 Global uncertainties Climate Change Towards economic security Creating better opportunities Protecting against poverty Financing pro-poor investments Increase agricultural productivity Better and more agile social assistance Re-examine the use of VAT exceptions Make urban areas engines of growth Social insurance covering all workers Increase taxes on alcohol, tobacco, sugar and carbon Enable high-productivity and low-carbon sectors Increase nancial inclusion Remove energy and agricultural subsidies Improve a ordability and quality of childcare Infrastructure investments to create resilience Improve sub-national administrative capacity Improving future policies Strengthen o cial statistics Enable data use Close analytical gaps Pathways Towards Economic Security Indonesia Poverty Assessment ix Executive Summary Progress and challenges while inequality slowly declined. Poverty, measured at Trends the lower middle-income country line of US$ 3.20 2011 Having eradicated nearly all extreme poverty, PPP per day, also declined steeply from 61 percent in 2002 Indonesia can now turn to broadening its definition of to 16 percent in 2022. Increased domestic consumption poverty commensurate with its middle-income status. drove poverty reduction in the last decade, contributing Extreme poverty, defined by living on less than US$ 1.90 to job growth in a tight labor market and increased real 2011 PPP per day, dropped from 19 percent in 2002 to wages. The largely inclusive nature of growth (Figure 1.5 percent in 2022 (Figure ES2), practically meeting the ES3) reversed the previous trend of rising inequality Government’s objective to eradicate extreme poverty when economic growth mostly benefitted wealthier ahead of its expected schedule of 2024. A small amount households (Figure ES4). of extreme, frictional poverty is likely to remain for some time. With extreme poverty almost eliminated, poverty Poverty reduction was broad-based, allowing most reduction strategies must widen their focus to also lagging regions to catch up, except for rural areas in include poor – but not extremely poor – households. two provinces. Poverty converged from 46 percent in Lower middle-income countries use higher poverty lines urban areas and 73 percent in rural areas in 2002 to 16 set to US$ 3.20 2011 PPP per day. percent in both urban and rural areas in 2022. Today, over half of the poor (56 percent) reside in urban areas. Similar Even with a broader definition of poverty, Indonesia but slower convergence occurred between regions. has made remarkable poverty reduction progress The two main lagging regions, Nusa Tenggara (NT) and FIGURE ES2: Poverty dropped starkly from 2002 to 2022 when FIGURE ES3: Annualized consumption growth (by consumption measured with the absolute international poverty lines percentile) became more pro-poor from 2011 onwards 80 6 5 60 4 Percent Percent 40 3 2 20 1 0 0 1 6 11 16 21 26 31 36 41 46 51 56 61 66 71 76 81 86 91 96 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022 $3.2 $3.2 Urban $3.2 Rural 2002-2010 2011-2014 $1.9 O cial 2014-2019 2019-2021 FIGURE ES4: Inequality increased substantially from 2002 until FIGURE ES5: Poverty rates across regions are converging, when 2010 before stagnating and dropping from 2014 until 2019 using absolute poverty estimates and slightly increasing due to COVID-19 in 2021 60 0.4 50 40 Gini Coe cient Percent 30 0.35 20 10 0 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022 0.3 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 $1.9 Nusa Tenggara $1.9 Maluku-Papua $1.9 Other O cial Nusa Tenggara Old Methodology New Methodology O cial Maluku-Papua O cial Other x Pathways Towards Economic Security Indonesia Poverty Assessment Executive Summary FIGURE ES6: Indonesia’s fiscal policies only have limited impact on inequality, especially compared to middle income countries 0 Change in Gini Index (points) -5 -10 -15 -20 Spain Uruguay Panama United States Croatia Mauritius Romania South Africa Argentina Brazil Mexico Namibia Georgia Venezuela Costa Rica Botswana Dominican Republic Colombia China Ecuador Peru Iran Turkey Belarus Malaysia Jordan Russia Albania Guatemala Paraguay Indonesia (2012) Indonesia (2017) Indonesia (2019) eSwatini Lesotho Zambia Tunisia Kenya Ukraine Honduras El Salvador Mongolia Bolivia India Nicaragua Moldova Egypt Tanzania Ghana Sri Lanka Comoros Ivory Coast Uganda Burkina Faso Togo Mali Ethiopia Niger Gambia Tajikistan Guinea -25 Low income High income Upper middle income Lower middle income Cash taxes and transfers In-kind spending on H+E Net scal impact Maluku-Papua (MP), saw their poverty rates decline by reduction, but with limited benefits for reduced poverty 50 percentage points from around 80 percent in 2002 to and inequality. Fourth, social assistance is more effective below 30 percent in 2022, compared to a drop of about in reducing poverty (Figure ES8) but still insufficient 40 percentage points in the remaining regions (Figure to provide needed coverage and benefits despite its ES5). However, rural areas in Nusa Tenggara and Maluku scale-up. Challenges in updating the targeting database continue to lag. increase inclusion and exclusion errors. Fifth, tight fiscal space led to under-investment in education and health, Fiscal policies contributed to reduce poverty and and—exacerbated by low administrative capacity of inequality, but low government revenue collection sub-national governments—hindered improvement in and costly subsidies reduce fiscal space and limit human capital outcomes and geographic disparities. more pro-poor investments. Inequality decreased by around 3 points of the Gini coefficient through fiscal Low-productivity challenge policies (Figure ES6). This is substantially less than the More than one-third of Indonesians remain at a such a range of 5 to 15 points for most middle and high-income low level of economic insecurity that a shock can push countries for several reasons. First, fiscal revenues are low them into poverty. In 2019, 40 percent of Indonesians relative to GDP compared to peers (Figure ES7). Second, were economically insecure. Most of these households agricultural subsidies are high and distort the market, are non-poor but can fall into poverty when exposed without obvious benefits for the poor. Third, costly to a shock. Economically insecure households can be energy subsidies have re-emerged after a temporary forced to adopt adverse coping strategies, diminishing FIGURE ES7: Indonesia’s government revenues relative to GDP FIGURE ES8: Transfers are significantly more effective in per capita remains low, limiting space for investments reducing poverty than energy subsidies 60 10.0 50 8.0 40 6.0 30 4.0 Malaysia 2.0 20 Thailand Indonesia 0.0 10 2012 2019 2012 2019 2012 2019 Philippines 0 Poverty impact Cost (% GDP) E ciency 3 3.5 4 4.5 5 5.5 (in pp) Log GDP per capita Transfers Subsidies Total Pathways Towards Economic Security Indonesia Poverty Assessment xi Executive Summary their physical and human capital assets, which, in turn, This absence of a productivity-increasing structural reduces short and long-term productivity. They may transformation undermines Indonesia’s potential, not also anticipate shocks and adopt conservative or risk- only in sustainably reducing poverty and economic averse production and investment strategies, reducing insecurity, but also in economic growth. their productivity even in the absence of a shock. Thus, regardless of whether adopted after or before a shock, Low urban migration limits productivity gains adverse coping strategies reduce long-term productivity, as fewer workers can take advantage of positive which in turn lowers chances of securely escaping poverty. agglomeration forces. Urban areas in Indonesia gained more productivity from agglomeration forces than from Agriculture and low value-add (low-VA) services more productive workers moving to urban areas. The remained the most important drivers of poverty official Indonesian urbanization trend is largely due to reduction, even though the work is often not very change of classification as rural areas increased density productive or sufficient to escape poverty. Agricultural to become more urban, rather than rural households incomes drove rural poverty reduction. However, many moving to urban areas. Nevertheless, urbanization is, farmers remained poor as they were constrained to and will remain, an important force. Even though urban low-productivity subsistence and rice production. A areas offered most higher-productivity work, such as distortionary set of incentives for agricultural producers in manufacturing and high-VA services, the number and high prices for staples due to import restrictions of such opportunities was insufficient. In addition, contribute to slow diversification to higher value cash- urban areas suffered from high cost of living (due to crops, for which the soil might be more suited in some housing costs), traffic congestions undermining urban areas. Low-VA services played a key role in poverty connectedness, and high air pollution. Thus, urban areas reduction particularly in urban areas, with the share of were not able to attract more workers, hence limiting workers rising in this sector. However, this work is often further agglomeration gains. This also limited their spill- informal and low-productivity, with many workers over effects into nearby rural areas, providing fewer remaining poor. opportunities for diversification from agriculture. High-skilled jobs remain scarce in Indonesia, Many women remained excluded from the labor limiting pathways towards economic security. Some force, constrained by cultural norms and home more productive opportunities were available—in care responsibilities, thus limiting livelihoods manufacturing and high-VA services, for example. opportunities for households. While above 80 percent However, not enough workers had the right skills to of men (although on a slowly decreasing trend) are in take advantage of these opportunities. At the same the labor force, only about 50 percent of women are time, the number of such high-skilled jobs – often found either employed or looking for work. Cultural norms in manufacturing – remained well below expected played an important role, translating into labor market levels relative to Indonesia’s development status. In discrimination. Women earned less than men, driven by a fact, premature deindustrialization reduced the output specific “female effect”. They also had care responsibilities share of manufacturing from 48 percent in 2002 to 41 for dependent household members, limiting their percent in 2019 while the service sector expanded participation in the labor force. This explains a persistent from 36 to 46 percent. While service-led development small gender poverty gap, especially for women around is possible, the increasingly inward-looking economy child-bearing age. While caring for household members missed out on productivity increases from global value is work, it is often a less remunerative activity than chain integration and export competition. Productivity participating in the labor market. This limits livelihoods for of services dropped from an average of 4.0 percent from households, and can make the difference between being 2000 to 2013 to 1.7 percent from 2014 to 2019 as growth poor, economically insecure, or economically secure. of low-VA outpaced high-VA service jobs (Figure ES10). xii Pathways Towards Economic Security Indonesia Poverty Assessment Executive Summary FIGURE ES9: Indonesia’s human capital index is lower FIGURE ES10: Labor productivity growth is dropping than peers, with some areas lagging far behind especially in industry and services 7.0 Gross xed capital formation (growth; LHS) 1 6.0 .8 5.0 4.0 .6 3.0 .4 2.0 1.0 .2 6 8 10 12 0.0 Real GDP per capita, 2011 log PPP Agriculture Industry Services Other countries Kalimantan Nusa Tenggara Sulawesi -1.0 Java-Bali Maluku Papua Sumatera 2002 - 2009 2010 - 2013 2014 - 2019 Human capital outcomes in Indonesia are slowly be 54 percent as productive as if receiving full education improving but remain below peer countries, especially and health. This is not only relatively low compared to in the Indonesian provinces of Maluku-Papua and Indonesia’s peers (Figure ES9), but also exhibits strong Nusa Tenggara, undermining productive potential of geographic disparities. Nusa Tenggara and Maluku-Papua, the population and exacerbating inequality. Access have worse outcomes, comparable to countries with to basic education is nearly universal since 2015, except significantly lower GDP per capita, a continued cause of for Nusa Tenggara and Maluku-Papua, where primary inequality in the medium and long-term. school enrollment rates stagnated at around 80 percent. At the secondary level, enrollment rates between poor Shock challenges and non-poor converged but plateaued at a relatively Shocks, such as COVID-19, can threaten poverty low level of around 80 percent. Learning quality remains reduction progress. The COVID-19 pandemic pushed a concern, as the expected 12.4 years of schooling Indonesia’s economy into a recession before rebounding translates to only 7.8 learning-adjusted years. Indonesia’s in 2021. This provided a stark example of a severe shock maternal mortality rate, and other key health indicators, affecting employment and health. It altered poverty fluctuated and remained significantly higher compared reduction significantly, affecting better-off, but not to peers. Accordingly, Indonesia’s human capital index the richest, households most, especially in urban areas improved only slightly from 0.5 in 2010 to 0.54 in 2020; (Figure ES11). The Government rapidly scaled-up social this means that a child born in Indonesia today would only assistance, reaching more beneficiaries and increasing FIGURE ES11: COVID-19 affected consumption growth from 2020 FIGURE ES12: Share of social protection beneficiaries in March to 2021 (shown by consumption percentile) in urban areas much 2021, who received any benefit since the onset of the pandemic more strongly than in rural areas 5 50 4 40 3 30 20 2 Percent 10 1 0 0 Bottom 40% Middle 40% Top 20% Bottom 40% Middle 40% Top 20% Bottom 40% Middle 40% Top 20% 1 7 13 19 25 31 37 43 49 55 61 67 73 79 85 91 97 -1 -2 Urban Rural Sembako Card PKH Bansos Tunai Pathways Towards Economic Security Indonesia Poverty Assessment xiii Executive Summary FIGURE ES13: Share of program beneficiaries assessing FIGURE ES14: Increase in prices in the first half of 2022 significantly benefits as adequate deteriorated purchasing power, but less for richer households 6 Pre-pandemic 4 Percent 2 0 Last three months 1 2 3 4 5 6 7 8 9 10 Income decile Food 0 20 40 60 80 100 Household energy price (Fuel and electricity) Percent Transportation Not at all Partially Mostly Completely All three categories the level of benefits. However, not all households in need last decades, its greenhouse gas (GHG) emissions rose received benefits, nor were benefits always adequate. sharply, making Indonesia the seventh largest emitter in Even among the bottom 40, less than 40 percent received the world. Indonesia is the world’s largest coal exporter, benefits from the expanded social assistance programs with coal exports representing 2 percent of GDP, or 13 (Figure ES12). Less than half of program beneficiaries self- percent of total goods exports. In 2021, the Government assessed the benefits of current programs as adequate at committed to substantial reduction of GHG emissions the time of the COVID-19 pandemic (Figure ES13). Also, and reaching net-zero emission by 2060. Phasing out coal the social insurance program did not manage to protect will particularly affect coal-producing communities, with all workers. In particular, informal workers often neither coal mining workers representing 0.2 percent of total had no access to sick leave nor were not eligible for formal employees in 2018. However, a larger number government unemployment insurance. of coal workers are informally employed. With coal mining concentrated in specific areas and communities, Climate change is expected to increase the frequency phasing-out coal will directly decrease employment in and severity of natural shocks, which can trap poor mining but also indirectly through firms depending on households into poverty and push economically coal mining and their workers. insecure households back into it. Between 1990-2021, Indonesia experienced more than 300 natural disasters, Combined with global uncertainties, such as Russia’s including 200 floods, affecting more than 11 million invasion of Ukraine, these risks threaten poverty people. Climate-related disasters already accounted reduction progress in Indonesia if households are not for around 70 percent of the total number of disasters protected. The war in Europe has triggered high volatility from 1990 to 2021. Although climate change affects in prices, especially for food and fuel. The purchasing the whole population, the poor and economically power of households in Indonesia deteriorated (Figure insecure are likely to carry a disproportionate burden. ES14), mostly because of the increase in food prices and They rely more often on agriculture, which is particularly the large food share in consumption. At the same time, negatively affected, and they often live in areas prone to the Government kept fuel prices constant by implicitly risk without resources to protect assets and less savings increasing fuel subsidies, adding to fiscal strains given to recover. the larger budgetary needs. With insufficient access to social protection and financial services, poor and Climate change mitigation will particularly harm economically insecure households are less able to workers in carbon-intensive industries if they are not weather shocks and might have to resort to adverse protected. Alongside Indonesia’s development over the coping strategies. xiv Pathways Towards Economic Security Indonesia Poverty Assessment Executive Summary Pathways towards economic security Increasing agricultural productivity can boost On the path to high-income, Indonesia’s poverty agricultural incomes. Enhanced agricultural productivity reduction policies need to be broadened through using climate-smart approaches can provide better a multi-pronged approach: creating better livelihoods for farmers and allow them to escape opportunities, protecting households against poverty, poverty, which is particularly relevant for households in and focusing fiscal resources on pro-poor investments, remote areas. For the two-thirds of poor, rural agricultural while promoting better information and evidence for households, their work is insufficient to escape poverty decision making. Given Indonesia’s development and given low productivity. Improving agricultural extension ambition, a broader definition of poverty, for example services and market access can boost agricultural around the US$ 3.20 2011 PPP poverty line, would be productivity. Removing agricultural subsidies focused more adequate. Better opportunities are needed in on food production can encourage farming of cash rural areas, through higher agricultural productivity, as crops, often better suited for some soil conditions. well as urban areas, by making cities engines of growth. Current subsidies are expensive and show little benefits. Higher productivity growth in low-carbon sectors Removing food import barriers can also help, as they can boost incomes and reduce poverty, while taking keep food prices high without helping poor farmers— advantage of digital opportunities. However, shocks are since most are net consumers of food—while diverting inevitable and will become more frequent with climate resources from higher value crops. change, but resilience can be fostered to minimize their harm. With about one-half of the non-poor population Investing in urban infrastructure can unlock the susceptible to falling back into poverty, better resilience potential of cities to act as engines of growth and and protection are needed. These measures will require amplify rural spill-over effects. Urban areas need public investments in a fiscally tight space. Policies need investments to become engines of productivity growth. to ensure cost-effective design while raising revenues Nurturing a more meaningful structural transformation and lift constraints to improve human capital equitably can create more opportunities for workers in urban areas. across the country. Finally, policy makers need to close Investments in urban infrastructure can help lower the remaining data and knowledge gaps to inform more cost of living in urban places. Together, these make cities effective policies. more attractive places to live. More workers moving to urban areas increase agglomeration forces, helping to Creating better opportunities unlocking productivity gains. This also contributes to job creation in nearby rural areas, creating opportunities Policies can support the private sector to create better, outside agriculture. higher-productivity jobs, in the context of climate change, the ongoing redesign of global value chains Better opportunities are needed in low-carbon sectors (GVCs), and digitalization. To continue reducing broad poverty and help households to reach economic with high productivity growth to boost incomes and security, better opportunities are needed. Enhanced reduce poverty. Competitiveness policies, including less agricultural productivity can provide better livelihoods restrictive trade and foreign direct investment policies for farmers and allow them to escape poverty. Urban as well as more effective anti-competitive policies, can areas need investments to allow them to become foster job growth, while eco industrial parks and circular engines of productivity growth. Better opportunities economy solutions can lower the carbon footprint of in low-carbon sectors with high-productivity growth can boost incomes. Integration into global value chains high-productivity sectors. Integration into GVCs attracts provides opportunities for Indonesia to increase its foreign direct investment for exports and can increase productivity through competitiveness. Digitalization productivity, especially in low-carbon sectors. The current similarly provides opportunities, and Indonesia can take global remapping of GVCs provides opportunities for advantage of its growing digital economy. Finally, more Indonesia to bolster its integration, but this would affordable and high-quality childcare can create jobs and require reversing increasingly restrictive trade policies provide opportunities for women to join the labor force. Pathways Towards Economic Security Indonesia Poverty Assessment xv Executive Summary and opening the economy for external opportunities. for example, through regular updating of the targeting Similarly, digitalization can provide opportunities but database and calibrating eligibility criteria to reflect new requires digital skills, connectivity, and a supportive poverty definitions. Third, adequacy of benefits can be policy environment. At the same time, workers need to improved. Such an improved social assistance system will be equipped with the right skill mix to prepare for new better mitigate negative shock effects on households, jobs; for example, policies must increase the level and who will therefore reduce the use of destructive coping quality of secondary and especially tertiary education strategies and be better able to make longer-term and invest in technical and vocational trainings (TVET). investments in higher-productivity activities. Offering affordable childcare can create jobs, foster Expanding coverage of social insurance to all workers female labor force participation, and improve can increase protection and productivity. In addition productivity. With affordable childcare, women can shift to social assistance, social insurance can help mitigate from unpaid to higher-productivity work, improving the impact of harmful shocks. Unemployment and labor market skills and firm productivity. Childcare helps health shocks are the most important household-level close the gender wage gap, which is still substantial shocks, and unemployment and health insurance can in Indonesia. Childcare creates jobs, and fosters early provide protection. However, unemployment insurance childhood learning, with long-term benefits for an is now only available to salaried, usually formal, workers economy’s productivity. in Indonesia. In addition, health shocks often have implications for labor incomes, due to lower productivity Protecting against poverty or unavailability to work because of sickness or care needs. Only formal workers have protection for these A combination of social assistance, social insurance, events currently. Thus, poorer households, who have less financial inclusion, and resilient infrastructure investments can help keep households out of poverty. secure work, benefit the least from protection, not only Better opportunities are essential to sustainably lift making them susceptible to falling into poverty, but also households out of poverty and economic insecurity. limiting progress on inequality. However, social protection measures need to complement job creation to help poor households and to Including the poor in the digital financial system can protect others from falling into poverty. Social assistance play a critical role in creating shock resilience and can be better targeted and be more comprehensive. A more agile social assistance system and expanded reducing poverty. Many Indonesian households remain coverage of social insurance, including informal workers, unbanked; even though financial inclusion has improved, are needed to improve household resilience against half of all adults in the bottom 40 still did not have a falling into poverty. Improved financial inclusion can help bank account in 2021. The lack of an account reduces households smooth income shocks without resorting the ability to save, which can smooth consumption to adverse coping strategies. Investments in resilient during a shock and replace lost assets. It also excludes infrastructure and climate-smart agricultural production households from receiving digital payments—for are also important to limit the impact of shocks. example, from government delivering social assistance Scaling-up social assistance includes improving quickly and efficiently in response to a shock. Including targeting and providing more adequate benefits. more households in digital financial services can foster COVID-19 provided lessons on how to improve resilience against shocks as a complement to social Indonesia’s social assistance system. First, coverage of assistance and insurance. Establishing a well-functioning the targeting database can be expanded beyond the and fully interoperable payment system together with bottom 40 percent to include all households, to support digital IDs and open banking policies can expand financial swift and flexible expansion of targeting in the case of services and make them more attractive for households, shocks. Second, targeting accuracy can be improved— ultimately contributing to increased resilience. xvi Pathways Towards Economic Security Indonesia Poverty Assessment Executive Summary Investing in resilient infrastructure and climate- household consumption, they are also consumed by smart investments can reduce the harmful effects of richer households and usually in greater amounts. One- natural disasters. Shocks from disasters put poverty third of potential VAT revenues (0.7 percent of GDP) in reduction progress at risk. Although poor households Indonesia are lost through the current exemptions are not necessarily more exposed to natural disasters, structure, enough to have funded the entire expanded they are less resilient and, thus, suffer the most from social assistance budget in 2019. Tobacco, alcohol, shocks. For example, in areas affected by the September and sugar-sweetened beverages have adverse health 2018 earthquake in central Sulawesi, over one in five effects, with large cost implications for public health. households from the bottom 40 percent were still in Increasing tax on these goods will reduce their temporary housing seven months later, compared to 13 consumption, saving costs for the public health system percent of the top 20 percent. Climate change will also while generating government revenue. Finally, a carbon reduce expected agricultural yields due to changes in tax can increase revenue while making investments in precipitation, temperature, and extreme weather events. high-carbon sectors less attractive. This will help increase Thus, investments in resilient infrastructure and climate- Indonesia’s competitiveness—for example, with respect smart agricultural production are important to limit to exports to countries that levy import tariffs on high- shock devastation in the first place. carbon products, like the EU’s carbon border adjustment mechanism. These reforms can hurt poor households, Financing pro-poor investments potentially reducing their income, but social assistance programs can compensate households. This would cost Increasing tax revenues and removing wasteful subsidies only a fraction of revenues gained but have a much can create fiscal space to make pro-poor investments, while increased sub-national administrative capacity larger effect on reducing inequality. can improve public services. Investments in education, health, and social protection will require more financial Removing energy and agricultural subsidies can raise resources than currently available. Tax revenues can be further fiscal resources. Energy subsidies are costly and increased through a reduction of value-added tax (VAT) ineffective in reducing poverty and inequality. While exemptions as well as excise taxes on tobacco, alcohol, an ambitious reform in 2015 started to reduce energy and sugar-sweetened beverages, which will create beneficial health effects. A carbon tax can raise revenue subsidies, social assistance was not scaled-up fast and incentivize a shift to a low-carbon economy, enough with sufficient compensation. This might have while reducing air pollution. Removing distortionary contributed to a political economy gravitating back to subsidies—especially for energy and agriculture—can subsidies, which returned from costing 0.7 percent of also create additional fiscal resources. A well-functioning GDP in 2016 to 1.7 percent of GDP in 2019. However, social assistance system can mitigate the adverse effects they reduced poverty only by 2.4 percentage points, as on the poor from these measures, at a fraction of the cost much as a core set of social assistance programs that of current policies. The additional fiscal resources from these measures could be redirected to finance pro-poor cost only 0.4 percent of GDP. Social assistance is not only investments to create better jobs and alleviate poverty. more efficient to reduce poverty but it is also strongly In addition, improving the administrative capacity of progressive in lowering inequality. Most fuel subsidies, sub-national governments would increase spending on the other hand, are not well targeted and can even be quality, especially in education and health, to improve regressive, while contributing to higher GHG emissions. human capital and attenuate geographic disparities. The Government also spends 2 to 3 percent of GDP on Removing VAT exemptions and increasing taxes on agriculture, mostly on subsidies for agricultural products. alcohol, tobacco, sugar, and carbon can generate However, subsidies are not well targeted to poor farmers, additional government revenue. A practical way to are largely ineffective, distort the agricultural market, quickly increase VAT revenue is to eliminate exemptions and undermine agricultural productivity. Revisiting and preferred rates for various goods and services. While agricultural expenditures to enhance competitiveness these items often represent a greater share of poorer and productivity can lead to large fiscal savings. Pathways Towards Economic Security Indonesia Poverty Assessment xvii Executive Summary Increasing sub-national administrative capacity overall outcomes and make them more equitable, while can improve quality of spending, service delivery, helping overcome stark geographic disparities in non- and human capital, while attenuating geographic monetary poverty. disparities. Indonesia started to decentralize about two decades ago. Sub-national governments (SNGs) became Improving future policies responsible for about 40 percent of total government Strengthening official statistics to enable data use and expenditures for service delivery in education and close analytical gaps can help inform and improve health. However, the quality of subnational spending policy design. Closing some important gaps can improve is limited in both allocative and technical efficiency. official statistics. For example, Indonesia needs to create Allocative efficiency suffers from misalignment of SNG an absolute poverty line and create an appropriate rural resources, under-serving areas with higher poverty rates, consumer price index (CPI). Use of Indonesia’s impressive thus exacerbating geographic disparities and worsening data collection can be increased by providing more inequality. Technical efficiency is undermined by open access to data. New challenges—such as the role growing SNG budgets without improvement of service of structural transformation and informality, and their delivery outcomes. Improving administrative capacity, implications for poverty—will need new policies based with a focus on the lowest-capacity SNGs, can improve on new and better data and evidence. xviii Pathways Towards Economic Security Indonesia Poverty Assessment RINGKASAN EKSEKUTIF Ikhtisar Indonesia memiliki rekam jejak pengentasan kemiskinan yang mengesankan, dan dapat membangun diatas keberhasilan tersebut untuk mengatasi target pengentasan kemiskinan yang lebih ambisius. Indonesia telah mencapai hasil yang mengesankan dalam mengurangi kemiskinan, dengan daerah-daerah yang sebelumnya tertinggal berhasil mengejar ketertinggalan nya dan tujuan Pemerintah untuk mengentaskan kemiskinan ekstrim pada tahun 2024 secara praktis telah terpenuhi. Namun demikian, sebagai calon negara berpenghasilan menengah ke atas, Indonesia mungkin ingin memperluas fokusnya di luar kemiskinan ekstrem dengan beralih dari garis kemiskinan US$ 1,90 2011 PPP (paritas daya beli) ke garis yang lebih tinggi untuk negara berpenghasilan menengah. Fokusnya juga harus mencakup rumah tangga yang secara ekonomi tidak aman, yang rentan jatuh kembali ke dalam kemiskinan. Apakah Indonesia siap menghadapi tantangan ini? Hasil sumber daya manusia masih belum memuaskan dan masih ada perbedaan geografis yang mengkhawatirkan. Produktivitas yang rendah masih menghalangi rumah tangga untuk menjadi aman secara ekonomi. Guncangan, termasuk dari perubahan iklim, terus mengancam upaya pengentasan kemiskinan. Kami mengidentifikasi beberapa jalur utama untuk mengatasi tantangan ini secara komprehensif dan berkelanjutan (Gambar ES1). (i) Menciptakan peluang yang lebih baik dalam produktivitas yang lebih tinggi dan pekerjaan rendah karbon untuk membantu rumah tangga menjadi aman secara ekonomi. (ii) Meningkatkan perlindungan sosial dan inklusi keuangan untuk memitigasi kerugian akibat adanya guncangan di masa mendatang. (iii) Mengembangkan sistem fiskal yang lebih efektif untuk investasi yang lebih berpihak pada masyarakat miskin yang berkontribusi pada sumber daya manusia melalui pemberian layanan publik yang lebih baik. (iv) Menutup kesenjangan data dan pengetahuan untuk memperbaiki kebijakan di masa mendatang untuk mendukung agenda ini. GAMBAR ES1: Empat jalur dengan prioritas kebijakan (hijau) menuju ketangguhan ekonomi dapat mengatasi tantangan utama (oranye) yang dihadapi Indonesia COVID-19 Ketidakpastian Global Perubahan iklim Menuju ketahanan ekonomi Menuju ketahanan ekonomi Melindungi dari kemiskinan Membiayai investasi yang berpihak pada masyarakat miskin Menuju ketahanan ekonomi Bantuan sosial yang lebih baik dan lebih responsif Meninjau kembali penggunaan pengecualian PPN Menjadikan perkotaan sebagai mesin pertumbuhan Jaminan sosial yang mencakup semua pekerja Menaikkan pajak alkohol, tembakau, gula, dan karbon Mendukung sektor dengan produktivitas tinggi dan rendah karbon Meningkatkan inklusi keuangan Hapus subsidi energi dan pertanian Meningkatkan keterjangkauan dan kualitas Investasi infrastruktur untuk menciptakan ketangguhan Meningkatkan kapasitas administrasi daerah fasilitas penitipan dan perawatan anak Memperbaiki kebijakan di masa depan Memperkuat statistik resmi Mendukung penggunaan data Menutup celah analitis Pathways Towards Economic Security Indonesia Poverty Assessment xix Ringkasan Eksekutif Kemajuan dan tantangan Bahkan dengan definisi kemiskinan yang lebih Tren luas, Indonesia telah mencapai kemajuan yang luar Setelah memberantas hampir semua kemiskinan biasa dalam pengentasan kemiskinan, sementara ekstrim, Indonesia kini dapat beralih untuk ketimpangan perlahan menurun. Kemiskinan, yang memperluas definisi kemiskinannya sesuai dengan diukur pada lini negara berpenghasilan menengah status pendapatan menengahnya. Kemiskinan ekstrim ke bawah sebesar US$ 3,20 2011 PPP (paritas daya yangdidefinisikan sebagai hidup dengan kurang dari US$ beli) per hari juga menurun tajam dari 61 persen 1,90 2011 PPP (paritas daya beli) per hari, turun dari 19 pada tahun 2002 menjadi 16 persen pada tahun 2022. persen pada tahun 2002 menjadi 1,5 persen pada tahun Peningkatan konsumsi dalam negeri mendorong 2022 (Gambar ES2), secara praktis memenuhi tujuan pengentasan kemiskinan dalam dekade terakhir, Pemerintah untuk memberantas kemiskinan ekstrim berkontribusi terhadap pertumbuhan lapangan kerja lebih cepat dari jadwal yang diharapkan pada tahun di pasar tenaga kerja yang ketat dan peningkatan 2024. Sejumlah kecil kemiskinan friksional yang ekstrim upah riil. Sifat pertumbuhan yang inklusif ini (Gambar kemungkinan besar akan tetap ada untuk beberapa ES3) membalikkan tren peningkatan ketimpangan waktu. Dengan kemiskinan ekstrim yang hampir hilang, sebelumnya di saat pertumbuhan ekonomi sebagian strategi pengentasan kemiskinan harus memperluas besar menguntungkan rumah tangga yang lebih kaya fokus nya agar mencakup juga rumah tangga miskin (Gambar ES4). – tetapi tidak sangat miskin. Negara berpenghasilan menengah ke bawah menggunakan garis kemiskinan Pengentasan kemiskinan berbasis luas, yang yang lebih tinggi yang ditetapkan sebesar US$ 3,20 PPP memungkinkan sebagian besar daerah tertinggal (paritas daya beli) 2011 per hari. untuk mengejar ketinggalan, kecuali daerah pedesaan GAMBAR ES2: Kemiskinan turun drastis dari tahun 2002 hingga 2022 GAMBAR ES3: Pertumbuhan konsumsi tahunan (berdasarkan persentil jika diukur dengan garis kemiskinan internasional absolut konsumsi) menjadi lebih berpihak pada masyarakat miskin sejak 2011 dan seterusnya 80 6 5 60 4 Percent Percent 40 3 2 20 1 0 0 1 6 11 16 21 26 31 36 41 46 51 56 61 66 71 76 81 86 91 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022 96 $3.2 $3.2 Urban $3.2 Rural 2002-2010 2011-2014 $1.9 O cial 2014-2019 2019-2021 GAMBAR ES4: Ketimpangan meningkat secara substansial dari tahun GAMBAR ES5: Tingkat kemiskinan di seluruh wilayah mengalami 2002 hingga 2010 sebelum stagnan dan menurun dari tahun 2014 konvergensi, saat menggunakan estimasi kemiskinan absolut hingga 2019 dan sedikit meningkat akibat COVID-19 pada tahun 2021 60 0.4 50 40 Gini Coe cient Percent 30 0.35 20 10 0 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022 0.3 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 $1.9 Nusa Tenggara $1.9 Maluku-Papua $1.9 Other O cial Nusa Tenggara Old Methodology New Methodology O cial Maluku-Papua O cial Other xx Pathways Towards Economic Security Indonesia Poverty Assessment Ringkasan Eksekutif GAMBAR ES6: Kebijakan fiskal Indonesia hanya berdampak terbatas pada ketimpangan, terutama jika dibandingkan dengan negara- negara berpenghasilan menengah 0 Change in Gini Index (points) -5 -10 -15 -20 Spain Uruguay Panama United States Croatia Mauritius Romania South Africa Argentina Brazil Mexico Namibia Georgia Venezuela Costa Rica Botswana Dominican Republic Colombia China Ecuador Peru Iran Turkey Belarus Malaysia Jordan Russia Albania Guatemala Paraguay Indonesia (2012) Indonesia (2017) Indonesia (2019) eSwatini Lesotho Zambia Tunisia Kenya Ukraine Honduras El Salvador Mongolia Bolivia India Nicaragua Moldova Egypt Tanzania Ghana Sri Lanka Comoros Ivory Coast Uganda Burkina Faso Togo Mali Ethiopia Niger Gambia Tajikistan Guinea -25 Low income High income Upper middle income Lower middle income Cash taxes and transfers In-kind spending on H+E Net scal impact di dua provinsi. Kemiskinan menurun dari 46 persen di banyak investasi yang berpihak pada masyarakat daerah perkotaan dan 73 persen di daerah pedesaan miskin. Ketimpangan menurun sekitar 3 poin koefisien pada tahun 2002 menjadi 16 persen baik di daerah Gini melalui kebijakan fiskal (Grafik ES6). Capaian ini perkotaan maupun pedesaan pada tahun 2022. Saat dibawah capaian sebagian besar negara berpenghasilan ini, lebih dari separuh penduduk miskin (56 persen) menengah dan tinggi yang berkisar antara 5 hingga tinggal di daerah perkotaan. Konvergensi serupa tetapi 15 Gini poin karena beberapa alasan. Pertama, lebih lambat terjadi antar daerah. Dua daerah tertinggal pendapatan fiskal relatif terhadap PDB termasuk rendah utama, Nusa Tenggara (NT) dan Maluku-Papua (MP), dibandingkan dengan negara-negara setara (Gambar mengalami penurunan tingkat kemiskinan sebesar ES7). Kedua, subsidi pertanian tinggi dan mendistorsi 50 poin persentase dari sekitar 80 persen pada tahun pasar, tanpa manfaat nyata bagi kaum miskin. Ketiga, 2002 menjadi di bawah 30 persen pada tahun 2022, subsidi energi yang mahal muncul kembali setelah dibandingkan dengan penurunan sekitar 40 poin adanya pengurangan sementara, tetapi dengan manfaat persentase di wilayah yang tersisa (Gambar ES5). Namun yang terbatas untuk mengurangi kemiskinan dan demikian, daerah pedesaan di Nusa Tenggara dan ketimpangan. Keempat, bantuan sosial lebih efektif Maluku masih tetap tertinggal. dalam mengurangi kemiskinan (Gambar ES8) tetapi masih belum cukup untuk memberikan cakupan dan Kebijakan fiskal berkontribusi dalam mengurangi manfaat yang dibutuhkan meskipun telah ditingkatkan. kemiskinan dan ketimpangan, tetapi penerimaan Tantangan dalam memperbarui Data Terpadu pemerintah yang rendah dan subsidi yang berbiaya Kesejahteraan Sosial (DTKS), yang merupakan data tinggi mengurangi ruang fiskal dan membatasi lebih rujukan untuk penetapan target, dapat meningkatkan GAMBAR ES7: Pendapatan pemerintah Indonesia relatif terhadap PDB GAMBAR ES8: Dana transfer jauh lebih efektif dalam mengurangi per kapita tetap rendah, sehingga membatasi ruang untuk investasi kemiskinan daripada subsidi energi 60 10.0 50 8.0 40 6.0 30 4.0 Malaysia 2.0 20 Thailand Indonesia 0.0 10 2012 2019 2012 2019 2012 2019 Philippines 0 Poverty impact Cost (% GDP) E ciency 3 3.5 4 4.5 5 5.5 (in pp) Log GDP per capita Transfers Subsidies Total Pathways Towards Economic Security Indonesia Poverty Assessment xxi Ringkasan Eksekutif kesalahan inklusi dan eksklusi. Kelima, ruang fiskal yang di sektor ini. Namun demikian, pekerjaan ini seringkali ketat menyebabkan kurangnya investasi di bidang bersifat informal dan produktivitasnya rendah, dengan pendidikan dan kesehatan, dan – diperparah oleh banyak pekerja yang tetap miskin. rendahnya kapasitas administrasi pemerintah daerah – menghambat peningkatan hasil sumber daya manusia Pekerjaan berketerampilan tinggi tetap langka di dan kesenjangan geografis. Indonesia, membatasi jalan menuju keamanan ekonomi. Beberapa peluang yang lebih produktif Tantangan produktivitas yang rendah sebenarnya tersedia – seperti di bidang manufaktur Lebih dari sepertiga penduduk Indonesia rentan jatuh dan jasa dengan NT tinggi . Namun demikian, tidak miskin jika terkena guncangan Pada tahun 2019, cukup banyak pekerja yang memiliki keterampilan 40 persen penduduk Indonesia tidak aman secara yang tepat untuk memanfaatkan peluang ini. Pada saat ekonomi. Sebagian besar rumah tangga ini tidak miskin yang sama, jumlah pekerjaan berketerampilan tinggi tetapi dapat jatuh miskin jika terkena guncangan. Rumah tersebut – yang sering didapati di bidang manufaktur tangga yang tidak aman secara ekonomi dapat dipaksa – tetap jauh di bawah tingkat yang diharapkan jika untuk mengadopsi strategi yang merugikan mereka dibandingkan dengan status pembangunan Indonesia. untuk mengatasi guncangan seperti mengurangi aset Faktanya, deindustrialisasi prematur mengurangi pangsa fisik dan sumber daya manusia mereka, yang pada keluaran manufaktur dari 48 persen pada tahun 2002 gilirannya mengurangi produktivitas jangka pendek menjadi 41 persen pada tahun 2019 sementara sektor dan jangka panjang. Mereka juga dapat mengantisipasi jasa bertumbuh dari 36 menjadi 46 persen. Meskipun guncangan dan mengadopsi strategi produksi dan pembangunan berbasis jasa dimungkinkan, ekonomi investasi yang konservatif atau menghindari risiko, yang semakin berorientasi ke dalam kehilangan potensi mengurangi produktivitas mereka bahkan tanpa adanya peningkatan produktivitas dari integrasi rantai nilai guncangan. Jadi, terlepas dari apakah diadopsi setelah global dan persaingan ekspor. Produktivitas sektor jasa atau sebelum guncangan, strategi penanggulangan turun dari rata-rata 4,0 persen dari tahun 2000 hingga yang merugikan mengurangi produktivitas jangka 2013 menjadi 1,7 persen dari tahun 2014 hingga panjang, yang pada gilirannya menurunkan peluang 2019 karena pertumbuhan pekerjaan jasa dengan NT untuk keluar dari kemiskinan dengan aman. rendah melampaui pekerjaan jasa dengan NT tinggi (Gambar ES10). Tidak adanya transformasi struktural Sektor pertanian dan jasa dengan nilai tambah yang meningkatkan produktivitas melemahkan potensi rendah (NT rendah) tetap menjadi pendorong utama Indonesia, tidak hanya dalam menurunkan kemiskinan pengentasan kemiskinan, meskipun pekerjaan dan kerawanan ekonomi secara berkelanjutan, tetapi tersebut seringkali tidak terlalu produktif atau juga dalam pertumbuhan ekonomi. tidak cukup untuk mendukung upaya keluar dari kemiskinan. Pendapatan pertanian mendorong Migrasi perkotaan yang rendah membatasi pengentasan kemiskinan di pedesaan. Namun demikian, peningkatan produktivitas karena lebih sedikit banyak petani tetap miskin karena mereka terkendala pekerja yang dapat memanfaatkan kekuatan pada produktivitas rendah dalam strategi pemenuhan aglomerasi positif. Peningkatan produktivitas di daerah kebutuhan hidup dan produksi beras. Serangkaian perkotaan lebih banyak dipicu oleh kekuatan aglomerasi insentif yang menyimpang bagi produsen pertanian dibandingkan karena pekerja yang lebih produktif dan harga bahan pokok yang tinggi karena pembatasan pindah ke daerah perkotaan. Tren urbanisasi resmi impor berkontribusi pada lambatnya diversifikasi ke Indonesia sebagian besar disebabkan oleh perubahan tanaman komersial bernilai lebih tinggi, di mana tanah klasifikasi karena daerah pedesaan meningkatkan di beberapa daerah mungkin lebih cocok. Sektor kepadatannya menjadi lebih ke perkotaan, daripada jasa dengan NT rendah memainkan peran penting rumah tangga pedesaan pindah ke daerah perkotaan. dalam pengentasan kemiskinan, khususnya di daerah Namun demikian, urbanisasi akan tetap menjadi kekuatan perkotaan, dengan jumlah pekerja yang meningkat yang penting. Meskipun daerah perkotaan menawarkan xxii Pathways Towards Economic Security Indonesia Poverty Assessment Ringkasan Eksekutif GAMBAR ES9: Indeks modal manusia Indonesia lebih rendah dari GAMBAR ES10: Pertumbuhan produktivitas tenaga kerja menurun negara-negara lain, dengan beberapa daerah tertinggal jauh terutama di sektor industri dan jasa 7.0 Gross xed capital formation (growth; LHS) 1 6.0 .8 5.0 4.0 .6 3.0 .4 2.0 1.0 .2 6 8 10 12 0.0 Real GDP per capita, 2011 log PPP Agriculture Industry Services Other countries Kalimantan Nusa Tenggara Sulawesi -1.0 Java-Bali Maluku Papua Sumatera 2002 - 2009 2010 - 2013 2014 - 2019 sebagian besar pekerjaan dengan produktivitas lebih laki-laki (walaupun dengan tren yang secara perlahan tinggi, seperti di sektor manufaktur dan jasa dengan menurun) berada dalam angkatan kerja, hanya sekitar 50 NT tinggi, jumlah peluang seperti itu tidak mencukupi. persen perempuan yang bekerja atau sedang mencari Selain itu, daerah perkotaan memiliki biaya hidup yang pekerjaan. Norma budaya memainkan peran penting, tinggi (karena biaya perumahan), kemacetan lalu lintas yang diterjemahkan senagai diskriminasi pasar tenaga yang mengganggu keterhubungan perkotaan, dan kerja. Perempuan berpenghasilan lebih rendah dari pria, pencemaran udara yang tinggi. Dengan demikian, daerah didorong oleh “efek perempuan” tertentu. Mereka juga perkotaan tidak dapat menarik lebih banyak pekerja, memiliki tanggung jawab merawat anggota keluarga, sehingga membatasi keuntungan aglomerasi lebih membatasi peran serta mereka dalam angkatan kerja. lanjut. Hal ini juga membatasi efek limpahan mereka ke Hal ini menjelaskan adanya kesenjangan kemiskinan daerah pedesaan terdekat, memberikan peluang yang gender yang kecil, terutama bagi perempuan di sekitar lebih kecil untuk diversifikasi dari pertanian. usia subur. Sementara merawat anggota rumah tangga adalah pekerjaan, seringkali merupakan kegiatan Banyak perempuan belum menjadi bagian dari yang kurang memberikan hasil banyak dibandingkan angkatan kerja, dibatasi oleh norma budaya dan dengan berperan serta dalam pasar tenaga kerja. Hal ini tanggung jawab perawatan keluarga di rumah, membatasi mata pencaharian rumah tangga, dan dapat sehingga membatasi peluang mata pencaharian membuat perbedaan antara menjadi miskin, tidak aman untuk rumah tangga. Sementara lebih dari 80 persen secara ekonomi, atau aman secara ekonomi. GAMBAR ES11: COVID-19 memengaruhi pertumbuhan konsumsi dari GAMBAR ES12: Persentase penerima manfaat perlindungan sosial tahun 2020 hingga 2021 (ditunjukkan oleh persentil konsumsi) di pada Maret 2021, yang menerima manfaat apa pun sejak awal perkotaan jauh lebih kuat dibandingkan dengan di pedesaan pandemi 5 50 4 40 3 30 20 2 Percent 10 1 0 0 Bottom 40% Middle 40% Top 20% Bottom 40% Middle 40% Top 20% Bottom 40% Middle 40% Top 20% 1 7 13 19 25 31 37 43 49 55 61 67 73 79 85 91 97 -1 -2 Urban Rural Sembako Card PKH Bansos Tunai Pathways Towards Economic Security Indonesia Poverty Assessment xxiii Ringkasan Eksekutif GAMBAR ES13: Pangsa penerima manfaat program yang menilai GAMBAR ES14: Kenaikan harga pada paruh pertama tahun 2022 manfaat sudah memadai menurunkan daya beli secara signifikan, tetapi lebih rendah untuk rumah tangga yang lebih kaya 6 Pre-pandemic 4 Percent 2 0 Last three months 1 2 3 4 5 6 7 8 9 10 Income decile Food 0 20 40 60 80 100 Household energy price (Fuel and electricity) Percent Transportation Not at all Partially Mostly Completely All three categories Capaian sumber daya manusia di Indonesia perlahan- Tantangan dari guncangan lahan membaik tetapi tetap di bawah negara-negara Guncangan, seperti COVID-19, dapat mengancam setara, terutama di provinsi Maluku-Papua dan Nusa kemajuan pengentasan kemiskinan. Pandemi Tenggara di Indonesia, melemahkan potensi produktif COVID-19 mendorong perekonomian Indonesia ke penduduk dan memperburuk ketimpangan. Akses dalam resesi sebelum pulih kembali pada tahun 2021. ke pendidikan dasar hampir merata sejak tahun 2015, Hal ini memberikan contoh nyata tentang guncangan kecuali Nusa Tenggara dan Maluku-Papua, di mana hebat yang berdampak pada pekerjaan dan kesehatan. angka partisipasi sekolah dasar stagnan di sekitar 80 Hal ini mengubah pengentasan kemiskinan secara persen. Di tingkat menengah, angka partisipasi antara signifikan, karena berdampak pada sebagian besar penduduk miskin dan tidak miskin menyatu tetapi rumah tangga yang relatif lebih kaya, tetapi bukan yang tetap pada tingkat yang relatif rendah, sekitar 80 persen. terkaya, terutama di daerah perkotaan (Gambar ES11). Kualitas pembelajaran tetap menjadi perhatian, seperti Pemerintah dengan cepat meningkatkan bantuan yang diharapkan 12,4 tahun sekolah menjadi hanya 7,8 sosial, menjangkau lebih banyak penerima manfaat dan tahun yang disesuaikan dengan pembelajaran. Angka meningkatkan tingkat manfaatnya. Namun demikian, kematian ibu di Indonesia, dan indikator kesehatan tidak semua rumah tangga yang membutuhkan utama lainnya, berfluktuasi dan tetap jauh lebih tinggi menerima manfaat, juga tidak selalu menerima manfaat dibandingkan dengan negara-negara lain. Oleh karena yang memadai Kurang dari 40 persen dari masyarakat itu, indeks modal manusia Indonesia hanya sedikit 40 persen termiskin menerima manfaat dari perluasan meningkat, dari 0,5 pada tahun 2010 menjadi 0,54 program bantuan sosial (Gambar ES12). Kurang dari pada tahun 2020; artinya, anak yang lahir di Indonesia separuh penerima manfaat program menilai sendiri saat ini hanya 54 persen yang produktif jika mendapat manfaat program saat ini sebagai memadai pada saat pendidikan dan kesehatan penuh. Ini tidak hanya relatif pandemi COVID-19 (Gambar ES13). Demikian pula, rendah dibandingkan negara-negara setara lainnya program jaminan sosial tidak berhasil melindungi semua (Gambar ES9), tetapi juga menunjukkan disparitas pekerja. Secara khusus, pekerja informal seringkali geografis yang kuat. Nusa Tenggara dan Maluku-Papua, tidak memiliki akses ke cuti sakit atau tidak memenuhi memiliki hasil yang lebih buruk, sebanding dengan syarat untuk mendapatkan jaminan pengangguran negara-negara dengan PDB per kapita yang jauh lebih pemerintah. rendah, menjadi penyebab berlanjutnya ketimpangan dalam jangka menengah dan panjang. Perubahan iklim diperkirakan akan meningkatkan frekuensi dan tingkat keparahan guncangan alam, yang dapat menjebak rumah tangga miskin ke dalam kemiskinan dan mendorong rumah tangga yang tidak xxiv Pathways Towards Economic Security Indonesia Poverty Assessment Ringkasan Eksekutif aman secara ekonomi kembali ke dalam kemiskinan. pangan yang besar. Pada saat yang sama, Pemerintah Antara tahun 1990-2021, Indonesia mengalami lebih dari mempertahankan harga bahan bakar secara konstan 300 bencana alam, termasuk 200 banjir, yang berdampak dengan secara implisit meningkatkan subsidi bahan pada lebih dari 11 juta orang. Bencana terkait iklim telah bakar, yang semakin menambah beban fiskal mengingat mencapai sekitar 70 persen dari total jumlah bencana adanya kebutuhan anggaran yang lebih besar. Dengan dari tahun 1990 hingga 2021. Meskipun perubahan iklim akses yang tidak memadai ke perlindungan sosial dan memengaruhi seluruh penduduk, masyarakat miskin dan layanan keuangan, rumah tangga miskin dan tidak tidak aman secara ekonomi kemungkinan besar akan aman secara ekonomi kurang mampu menghadapi menanggung beban yang tidak proporsional. Mereka guncangan dan mungkin harus menggunakan strategi lebih sering mengandalkan pertanian, yang terkena penanggulangan yang merugikan. dampak negatif, dan mereka sering tinggal di daerah yang rawan risiko tanpa sumber daya untuk melindungi Jalan menuju keamanan ekonomi aset dan lebih sedikit tabungan untuk pulih kembali. Di jalur menuju pendapatan tinggi, kebijakan pengentasan kemiskinan Indonesia perlu diperluas Mitigasi perubahan iklim secara khusus akan melalui pendekatan multi-cabang: menciptakan merugikan pekerja di sektor industri padat karbon peluang yang lebih baik, melindungi rumah tangga dari jika tidak dilindungi. Seiring dengan pembangunan kemiskinan, dan memfokuskan sumber daya fiskal pada Indonesia selama beberapa dekade terakhir, emisi investasi yang berpihak pada masyarakat miskin, sambil gas rumah kaca (GRK) meningkat tajam, menjadikan mendorong pemanfaatan informasi dan bukti yang Indonesia sebagai penghasil emisi terbesar ketujuh di lebih baik untuk pengambilan keputusan. Mengingat dunia. Indonesia adalah pengekspor batu bara terbesar pembangunan dan ambisi Indonesia, definisi kemiskinan di dunia, dengan ekspor batu bara mewakili 2 persen dari yang lebih luas, misalnya di sekitar garis kemiskinan US$ PDB, atau 13 persen dari total ekspor barang. Pada tahun 3,20 2011 PPP , akan lebih memadai. Diperlukan peluang 2021, Pemerintah berkomitmen untuk mengurangi emisi yang lebih baik di daerah pedesaan, melalui produktivitas GRK secara substansial dan mencapai emisi net-zero pada pertanian yang lebih tinggi, serta daerah perkotaan, tahun 2060. Penghapusan batubara secara bertahap dengan menjadikan kota sebagai mesin pertumbuhan. akan berdampak terutama pada masyarakat penghasil Pertumbuhan produktivitas yang lebih tinggi di sektor batubara, dengan jumlah pekerja tambang batubara rendah karbon dapat meningkatkan pendapatan dan sebesar 0,2 persen dari total pekerja formal pada tahun mengurangi kemiskinan, sekaligus memanfaatkan 2018. Namun demikian, jumlah yang lebih besar dari peluang digital. Namun demikian, guncangan tidak pekerja batubara dipekerjakan secara informal. Dengan dapat dihindari dan akan menjadi lebih sering dengan pertambangan batu bara terkonsentrasi di wilayah dan adanya perubahan iklim, tetapi ketangguhan dapat komunitas tertentu, penghapusan batu bara secara dipupuk untuk meminimalkan kerugiannya. Dengan bertahap akan secara langsung mengurangi lapangan sekitar setengah dari penduduk tidak miskin yang kerja di pertambangan tetapi juga secara tidak langsung rentan untuk jatuh kembali ke dalam kemiskinan, melalui perusahaan-perusahaan yang bergantung pada diperlukan ketangguhan dan perlindungan yang lebih pertambangan batu bara dan para pekerjanya. baik. Langkah-langkah ini akan membutuhkan investasi publik dalam ruang fiskal yang ketat. Kebijakan perlu Dikombinasikan dengan ketidakpastian global, memastikan desain hemat biaya sambil meningkatkan seperti invasi Rusia ke Ukraina, risiko ini mengancam pendapatan dan menghilangkan kendala untuk kemajuan pengentasan kemiskinan di Indonesia meningkatkan sumber daya manusia secara merata di jika rumah tangga tidak terlindungi. Perang di Eropa seluruh negeri. Yang terakhir, para pembuat kebijakan telah memicu volatilitas harga yang tinggi, terutama perlu menutup kesenjangan data dan pengetahuan untuk makanan dan bahan bakar. Daya beli rumah yang masih ada untuk dapat memberi informasi bagi tangga di Indonesia memburuk (Grafik ES14), terutama pengambilan kebijakan yang lebih efektif. karena kenaikan harga pangan dan porsi konsumsi Pathways Towards Economic Security Indonesia Poverty Assessment xxv Ringkasan Eksekutif Menciptakan peluang yang lebih baik Berinvestasi dalam infrastruktur perkotaan dapat membuka potensi kota untuk menjadi mesin Kebijakan dapat mendukung sektor swasta untuk menciptakan pekerjaan yang lebih baik dengan pertumbuhan dan memperkuat dampak limpahan produktivitas lebih tinggi, dalam konteks perubahan pedesaan. Kawasan perkotaan membutuhkan investasi iklim, desain ulang rantai nilai global (global value untuk menjadi mesin pertumbuhan produktivitas. chains, GVC) yang sedang berlangsung, dan digitalisasi. Memelihara transformasi struktural yang lebih bermakna Untuk dapat terus mengurangi kemiskinan yang luas dapat menciptakan lebih banyak peluang bagi para dan membantu rumah tangga mencapai keamanan pekerja di daerah perkotaan. Investasi dalam infrastruktur ekonomi, diperlukan peluang yang lebih baik. Peningkatan produktivitas pertanian dapat memberikan perkotaan dapat membantu menurunkan biaya hidup mata pencaharian yang lebih baik bagi para petani dan di perkotaan. Secara bersama-sama, hal tersebut memungkinkan mereka untuk keluar dari kemiskinan. membuat kota menjadi tempat yang lebih menarik Daerah perkotaan membutuhkan investasi untuk untuk ditinggali. Lebih banyak pekerja yang pindah ke memungkinkan mereka menjadi mesin pertumbuhan daerah perkotaan meningkatkan kekuatan aglomerasi, produktivitas. Peluang yang lebih baik di sektor rendah membantu meningkatkan produktivitas. Hal ini juga karbon dengan pertumbuhan produktivitas tinggi dapat berkontribusi pada penciptaan lapangan kerja di daerah meningkatkan pendapatan. Integrasi ke dalam rantai nilai global memberikan peluang bagi Indonesia untuk pedesaan terdekat, menciptakan peluang di luar sektor meningkatkan produktivitasnya melalui daya saing. pertanian. Digitalisasi juga memberikan peluang, dan Indonesia dapat memanfaatkan pertumbuhan ekonomi digitalnya. Diperlukan adanya peluang yang lebih baik di sektor Yang terakhir, fasilitas penitipan anak yang lebih rendah karbon dengan pertumbuhan produktivitas terjangkau dan berkualitas tinggi dapat menciptakan tinggi untuk meningkatkan pendapatan dan lapangan kerja dan memberikan peluang bagi perempuan untuk bergabung dengan angkatan kerja. mengurangi kemiskinan. Kebijakan daya saing, termasuk kebijakan perdagangan dan investasi asing Peningkatan produktivitas pertanian dapat langsung yang tidak terlalu ketat serta kebijakan anti meningkatkan pendapatan pertanian. Peningkatan persaingan yang lebih efektif, dapat mendorong produktivitas pertanian dengan menggunakan pertumbuhan lapangan kerja, sementara kawasan pendekatan cerdas iklim dapat memberikan industri ramah lingkungan dan solusi ekonomi sirkular penghidupan yang lebih baik bagi petani dan dapat menurunkan jejak karbon dari sektor produktivitas memungkinkan mereka untuk keluar dari kemiskinan, tinggi. Integrasi ke dalam rantai nilai global (global yang sangat relevan untuk rumah tangga di daerah value chains, GVC) menarik investasi asing langsung terpencil. Bagi dua pertiga rumah tangga pertanian untuk ekspor dan dapat meningkatkan produktivitas, pedesaan yang miskin, pekerjaan mereka tidak khususnya di sektor rendah karbon. Pemetaan kembali mencukupi untuk dapat keluar dari kemiskinan GVC secara global saat ini memberikan peluang bagi mengingat produktivitas yang rendah. Meningkatkan Indonesia untuk memperkuat integrasinya, tetapi hal ini layanan penyuluhan pertanian dan akses pasar dapat memerlukan adanya perubahan kebijakan perdagangan meningkatkan produktivitas pertanian. Menghapus yang semakin ketat dan membuka ekonomi untuk subsidi pertanian yang berfokus pada produksi pangan peluang eksternal. Demikian pula, digitalisasi dapat dapat mendorong pertanian tanaman komersial, memberikan peluang tetapi membutuhkan keterampilan seringkali lebih cocok untuk beberapa kondisi lahan. digital, konektivitas, dan lingkungan kebijakan yang Subsidi saat ini mahal dan hanya menunjukkan sedikit mendukung. Pada saat yang sama, para pekerja perlu manfaat. Menghilangkan hambatan impor pangan juga dibekali dengan bauran keterampilan yang tepat untuk dapat membantu, karena hambatan tersebut membuat mempersiapkan diri bagi jenis pekerjaan baru; misalnya, harga pangan tetap tinggi tanpa membantu petani kebijakan harus meningkatkan tingkat dan kualitas miskin – karena sebagian besar adalah konsumen bersih pendidikan menengah dan khususnya pendidikan tinggi pangan – dan mengalihkan sumber daya dari tanaman dan berinvestasi dalam pendidikan dan pelatihan vokasi bernilai lebih tinggi. teknis (Technical and Vocational Trainings, TVET). xxvi Pathways Towards Economic Security Indonesia Poverty Assessment Ringkasan Eksekutif Menawarkan fasilitas penitipan anak yang terjangkau cepat dan fleksibel jika terjadi guncangan. Kedua, dapat menciptakan lapangan kerja, mendorong peran akurasi penetapan target dapat ditingkatkan – misalnya, serta angkatan kerja perempuan, dan meningkatkan melalui pemutakhiran basis data penetapan target produktivitas. Dengan fasilitas penitipan anak yang secara rutin dan penyesuaian kriteria kelayakan untuk terjangkau, perempuan dapat beralih dari pekerjaan mencerminkan definisi kemiskinan yang baru. Ketiga, tidak berbayar ke pekerjaan dengan produktivitas lebih kecukupan manfaat dapat ditingkatkan. Sistem bantuan tinggi, meningkatkan keterampilan pasar tenaga kerja sosial yang lebih baik seperti itu akan mengurangi dan produktivitas perusahaan. Fasilitas penitipan anak dampak kejutan negatif pada rumah tangga dengan membantu menutup kesenjangan upah gender, yang lebih baik, dan dengan demikian akan mengurangi masih cukup besar di Indonesia. Fasilitas penitipan penggunaan strategi penanggulangan yang merusak anak menciptakan lapangan kerja, dan memupuk dan lebih mampu melakukan investasi jangka panjang pembelajaran anak usia dini, dengan manfaat jangka dalam kegiatan produktivitas yang lebih tinggi. panjang untuk produktivitas ekonomi. Memperluas cakupan jaminan sosial ke seluruh Menciptakan peluang yang lebih baik pekerja dapat meningkatkan perlindungan dan produktivitas. Selain bantuan sosial, jaminan sosial Kombinasi bantuan sosial, jaminan sosial, inklusi keuangan, dan investasi infrastruktur yang tangguh dapat membantu mengurangi dampak guncangan dapat membantu rumah tangga keluar dari kemiskinan. yang merugikan. Guncangan pengangguran dan Peluang yang lebih baik sangat penting untuk kesehatan merupakan guncangan tingkat rumah tangga mengangkat rumah tangga keluar dari kemiskinan yang paling penting, dan jaminan pengangguran serta dan kerawanan ekonomi secara berkelanjutan. Namun jaminan kesehatan dapat memberikan perlindungan. demikian, langkah-langkah perlindungan sosial Namun demikan, di Indonesia jaminan pengangguran perlu melengkapi penciptaan lapangan kerja untuk saat ini hanya tersedia bagi pekerja bergaji, biasanya membantu rumah tangga miskin dan melindungi masyarakat lainnys agar tidak jatuh ke dalam kemiskinan. pekerja formal. Selain itu, guncangan kesehatan sering Bantuan sosial dapat lebih tepat sasaran dan lebih berimplikasi pada pendapatan tenaga kerja, karena komprehensif. Diperlukan adanya sistem bantuan produktivitas yang lebih rendah atau tidak tersedianya sosial yang lebih responsif dan perluasan cakupan pekerjaan karena sakit atau membutuhkan perawatan jaminan sosial, termasuk bagi para pekerja informal, kesehatan. Saat ini, hanya pekerja formal yang memiliki untuk meningkatkan ketangguhan rumah tangga agar perlindungan untuk kejadian-kejadian tersebut. Dengan tidak jatuh ke dalam kemiskinan. Inklusi keuangan yang lebih baik dapat membantu rumah tangga mengatasi demikian, rumah tangga yang lebih miskin, yang memiliki guncangan pendapatan tanpa menggunakan strategi pekerjaan yang kurang terjamin, adalah yang paling penanggulangan yang merugikan. Investasi dalam sedikit mendapat manfaat dari perlindungan, tidak hanya infrastruktur yang tangguh dan produksi pertanian yang membuat mereka rentan jatuh ke dalam kemiskinan, cerdas iklim juga penting untuk membatasi dampak tetapi juga membatasi kemajuan ketidaksetaraan. guncangan. Mengikut-sertakan masyarakat miskin dalam Peningkatan bantuan sosial mencakup peningkatan sistem keuangan digital dapat memainkan peran kualitas penetapan target dan pemberian manfaat penting dalam menciptakan ketangguhan terhadap yang lebih memadai. COVID-19 memberikan pelajaran guncangan dan mengurangi kemiskinan. Banyak mengenai cara meningkatkan sistem bantuan sosial rumah tangga Indonesia tetap tidak memiliki rekening Indonesia. Yang pertama, cakupan basis data penetapan bank; meskipun inklusi keuangan telah meningkat, target dapat diperluas melampaui 40 persen masyarakat setengah dari semua orang dewasa di masyarakat di desil terbawah untuk mencakup semua rumah tangga, 40 terbawah masih belum memiliki rekening bank pada untuk mendukung perluasan penetapan target yang tahun 2021. Dengan tidak memiliki rekening mengurangi Pathways Towards Economic Security Indonesia Poverty Assessment xxvii Ringkasan Eksekutif kemampuan menabung, yang dapat memperlancar Pembiayaan investasi yang berpihak pada konsumsi saat terjadi guncangan dan mengganti aset masyarakat miskin yang hilang. Hal ini juga menyebabkan rumah tangga Meningkatkan penerimaan pajak dan menghilangkan tidak dapat menerima pembayaran digital – misalnya, subsidi yang tidak efisien dapat menciptakan ruang dari pemerintah yang memberikan bantuan sosial fiskal untuk melakukan investasi yang berpihak pada dengan cepat dan efisien sebagai tanggapan terhadap masyarakat miskin, sementara peningkatan kapasitas guncangan. Melibatkan lebih banyak rumah tangga administrasi daerah dapat meningkatkan layanan dalam layanan keuangan digital dapat menumbuhkan publik. Investasi dalam pendidikan, kesehatan, dan perlindungan sosial akan membutuhkan lebih banyak ketangguhan terhadap guncangan sebagai pelengkap sumber daya keuangan daripada yang tersedia saat bantuan dan jaminan sosial. Membangun sistem ini. Penerimaan pajak dapat ditingkatkan melalui pembayaran yang berfungsi dengan baik dan pengurangan pembebasan pajak pertambahan sepenuhnya dapat dioperasikan bersama dengan ID nilai (PPN) serta cukai atas tembakau, alkohol, dan digital dan kebijakan perbankan yang terbuka dapat minuman berpemanis, yang akan menciptakan dampak memperluas layanan keuangan dan menjadikannya kesehatan yang menguntungkan. Pajak karbon dapat meningkatkan penerimaan dan mendorong peralihan lebih menarik bagi rumah tangga, yang pada akhirnya ke ekonomi rendah karbon, sekaligus mengurangi berkontribusi pada peningkatan ketangguhan. pencemaran udara. Menghapus subsidi yang terdistorsi – khususnya untuk energi dan pertanian – juga dapat Berinvestasi dalam infrastruktur yang tangguh dan menciptakan sumber daya fiskal tambahan. Sistem investasi cerdas iklim dapat mengurangi dampak bantuan sosial yang berfungsi dengan baik dapat merugikan dari bencana alam. Guncangan akibat memitigasi dampak negatif bagi masyarakat miskin dari bencana membahayakan kemajuan pengentasan langkah-langkah tersebut, dengan sebagian kecil dari biaya kebijakan saat ini. Sumber daya fiskal tambahan kemiskinan. Meskipun rumah tangga miskin belum dari langkah-langkah tersebut dapat diarahkan untuk tentu lebih rentan terhadap bencana alam, mereka membiayai investasi yang berpihak pada masyarakat kurang tangguh sehingga paling menderita akibat miskin untuk menciptakan pekerjaan yang lebih baik guncangan. Misalnya, di daerah yang terkena dampak dan mengentaskan kemiskinan. Selain itu, peningkatan gempa bumi pada bulan September 2018 di Sulawesi kapasitas administratif pemerintah daerah akan Tengah, satu dari lima rumah tangga dari masyarakat di meningkatkan kualitas belanja, terutama di bidang pendidikan dan kesehatan, untuk meningkatkan sumber desil 40 terbawah masih menempati rumah sementara daya manusia dan mengurangi kesenjangan geografis. tujuh bulan kemudian, dibandingkan dengan 13 persen dari masyarakat di desil 20 persen teratas. Perubahan Menghapus pembebasan PPN dan menaikkan pajak iklim juga akan mengurangi hasil pertanian yang atas alkohol, tembakau, gula, dan karbon dapat diharapkan karena perubahan curah hujan, suhu, dan menghasilkan tambahan penerimaan pemerintah. Cara peristiwa cuaca ekstrem. Oleh karena itu, investasi dalam praktis untuk meningkatkan penerimaan PPN dengan infrastruktur yang tangguh dan produksi pertanian yang cepat adalah dengan menghilangkan pengecualian cerdas iklim penting untuk membatasi kehancuran dan tarif pilihan atas pajak untuk berbagai barang dan akibat guncangan sejak awal. jasa. Sementara barang-barang tersebut seringkali merupakan pangsa yang lebih besar dari konsumsi rumah tangga yang lebih miskin, barang-barang tersebut juga dikonsumsi oleh rumah tangga yang lebih kaya dan biasanya dalam jumlah yang lebih banyak. Sepertiga dari potensi penerimaan PPN (0,7 persen dari PDB) di Indonesia hilang melalui struktur pembebasan PPN saat ini, cukup untuk mendanai seluruh anggaran bantuan sosial yang diperluas pada tahun 2019. Tembakau, alkohol, dan minuman berpemanis memiliki dampak xxviii Pathways Towards Economic Security Indonesia Poverty Assessment Ringkasan Eksekutif kesehatan yang negatif, dengan implikasi biaya yang melemahkan produktivitas pertanian. Meninjau kembali besar bagi kesehatan masyarakat. Menaikkan pajak atas belanja pertanian untuk meningkatkan daya saing dan barang-barang tersebut akan mengurangi konsumsinya, produktivitas dapat menghasilkan penghematan fiskal menghemat biaya untuk sistem kesehatan publik yang besar. sekaligus menghasilkan penerimaan pemerintah. Yang terakhir, pajak karbon dapat meningkatkan penerimaan Meningkatkan kapasitas administrasi daerah dapat dan membuat investasi di sektor karbon tinggi menjadi meningkatkan kualitas belanja, pemberian layanan, kurang menarik. Hal ini akan membantu meningkatkan dan sumber daya manusia, sekaligus mengurangi daya saing Indonesia – misalnya, terkait dengan ekspor kesenjangan geografis. Indonesia mulai melakukan ke negara-negara yang mengenakan tarif impor untuk desentralisasi sekitar dua dekade lalu. Pemerintah daerah produk-produk berkandungan karbon tinggi, seperti (Pemda) bertanggung jawab atas sekitar 40 persen dari mekanisme penyesuaian batas karbon UE. Reformasi total belanja pemerintah untuk penyediaan layanan di ini dapat merugikan rumah tangga miskin, berpotensi bidang pendidikan dan kesehatan. Namun demikian, mengurangi pendapatan mereka, tetapi program kualitas belanja daerah terbatas, baik dalam efisiensi bantuan sosial dapat memberi kompensasi kepada alokatif maupun teknis. Efisiensi alokatif mengalami rumah tangga. Ini hanya akan menelan biaya sebagian ketidaksejajaran sumber daya Pemda, daerah yang kecil dari penerimaan yang diperoleh tetapi memiliki kurang terlayani dengan tingkat kemiskinan yang dampak yang jauh lebih besar dalam mengurangi lebih tinggi, sehingga memperparah kesenjangan ketimpangan. geografis dan memperburuk ketimpangan. Efisiensi teknis diperlemah oleh peningkatan anggaran Pemda Menghapus subsidi energi dan pertanian dapat tanpa adanya peningkatan hasil penyampaian layanan. meningkatkan sumber daya fiskal lebih lanjut. Subsidi Meningkatkan kapasitas administratif, dengan fokus energi mahal dan tidak efektif dalam mengurangi pada Pemda yang berkapasitas paling rendah, dapat kemiskinan dan ketimpangan. Sementara reformasi meningkatkan hasil keseluruhan dan menjadikannya yang ambisius pada tahun 2015 mulai mengurangi lebih adil, sambil membantu mengatasi kesenjangan subsidi energi, bantuan sosial tidak ditingkatkan dengan geografis yang mencolok dalam kemiskinan non- cukup cepat dengan kompensasi yang memadai. Hal ini moneter. mungkin telah berkontribusi pada ekonomi politik yang kembali ke subsidi, yang kembali dari biaya 0,7 persen dari Memperbaiki kebijakan di masa depan PDB pada tahun 2016 menjadi 1,7 persen dari PDB pada Memperkuat statistik resmi untuk memungkinkan tahun 2019. Namun demikian, subsidi tersebut hanya penggunaan data dan menutup kesenjangan mengurangi kemiskinan sebesar 2,4 poin persentase, analitis dapat membantu memberi informasi bagi sebanyak seperangkat program bantuan sosial inti yang pengambilan kebijakan dan meningkatkan desain biayanya hanya 0,4 persen dari PDB. Bantuan sosial kebijakan. Menutup beberapa celah penting dapat tidak hanya lebih efisien untuk mengurangi kemiskinan meningkatkan statistik resmi. Misalnya, Indonesia perlu tetapi juga sangat progresif dalam menurunkan menciptakan garis kemiskinan absolut dan menciptakan ketimpangan. Di sisi lain, sebagian besar subsidi BBM indeks harga konsumen (IHK) pedesaan yang sesuai. tidak tepat sasaran dan bahkan dapat bersifat regresif, Penggunaan pengumpulan data Indonesia yang tetapi berkontribusi terhadap emisi GRK yang lebih mengesankan dapat ditingkatkan dengan menyediakan tinggi. Pemerintah juga membelanjakan 2 hingga 3 akses data yang lebih terbuka. Tantangan baru – seperti persen dari PDB untuk pertanian, sebagian besar untuk peran transformasi struktural dan informalitas, serta subsidi produk pertanian. Namun demikian, subsidi implikasinya terhadap kemiskinan – akan membutuhkan tersebut tidak tepat sasaran bagi petani miskin, sebagian kebijakan baru berdasarkan data dan bukti yang baru besar tidak efektif, mendistorsi pasar pertanian, dan dan lebih baik. Pathways Towards Economic Security Indonesia Poverty Assessment xxix CHAPTER 1 CONTEXT Photo: © Ed Wray/World Bank xxx Pathways Towards Economic Security Indonesia Poverty Assessment 1. CONTEXT Looking Back range of sectors fundamentally altering the structure of Indonesia’s extensive literature on poverty reduction the economy, livelihoods, and poverty. highlights successes, including progress in poverty reduction, as well as recurring challenges, such as low productivity Since 2006, inequality in Indonesia continued to in both urban and rural economies, non-monetary rise into the 2010s. The main factors contributing deprivations, and inequality, including a “digital divide”. to inequality were inequality of opportunity, a low- productivity work trap for less educated workers, Indonesia has sustained impressive economic growth and poverty reduction. Over the last half century, Indonesia has experienced rapid and sustained concentration of financial assets, and greater vulnerability to shocks for the poor compared to the non-poor.4 Inequality of opportunity to access education meant economic growth averaging 5.3 percent, despite the that children in poorer households did not develop skills massive shock of the 1997 Asian Financial Crisis (AFC). It needed to obtain a well-paying job later in life. They progressed from a low-income to a lower middle-income would join the ranks of less educated workers trapped country before reaching upper middle-income status in in informal, low-productivity jobs without the right skills 2019.1 Extreme poverty, measured at US$ 1.90 PPP 2011, needed in a modern economy. These jobs might not fell from an estimated 69 percent of the population in even pay enough to escape poverty, while increasing 1984, the first point at which the measure becomes wage inequality. In addition, financial resources were relatively consistent, to 2 percent by 2021.2 Using the becoming more concentrated in the hands of a few lower middle-income country poverty line, or US$ 3.20 wealthy households.5 This exacerbated income gaps in PPP 2011, poverty dropped from 91 percent in 1984 the current generation and would increase human and to 18 percent in 2021. The President of Indonesia, Joko financial resource inequality in the future. Finally, shocks Widodo, announced the same year the country’s goal to more negatively affected poor households as they often eradicate extreme poverty by 2024. did not have the ability to even cope even with small shocks. For them, a shock quickly eroded their ability to The World Bank’s last major poverty analysis for earn, save, and invest in health and education. Indonesia in 2006 expanded the focus beyond monetary poverty to include economic insecurity, Urbanization has accelerated, partly driving non-monetary poverty dimensions, and regional Indonesia’s economic rise, but has not benefitted disparities.3 By 2006, Indonesia had recovered from all of Indonesia’s urban residents equally.6 Indonesia the Asian Financial Crisis and started to benefit from was—and still is—urbanizing rapidly. However, poverty the global commodity boom with high growth. Even levels were higher than countries with similar levels of though inequality started to rise, poverty dropped urbanization.7 Even though cities provided better jobs rapidly, reaching 61 percent, at US$ 3.20 PPP 2011. and better access to services, the gaps between the poor Three poverty features were identified at that time as and non-poor remained stark. In particular, households salient for Indonesia: (i) a high degree of vulnerability to residing at the fringes of urban centers were not able to poverty for a many people, (ii) the serious nature of non- fully reap the benefits of urban living. They were exposed income poverty, and (iii) regional disparities. Since then, to higher living costs and long commute times, often in Indonesia has transformed with changes at the global, the absence of reliable public transportation. regional, national, and sub-national levels and across a 1 The country fell back to lower middle-income status in 2020 with the impact of COVID-19. 2 The national poverty measurement methodology has not stayed constant 4 World Bank 2016; Asian Development Bank 2020b; Hill 2021. over this time, but this will not change the overall trend of a very large decrease in poverty. See Hill 2021 for detailed discussion of poverty reduction 5 Credit Suisse 2019. back to the 1960s. 6 Roberts, Gil Sander, and Tiwari 2019. 3 World Bank 2006. 7 Asian Development Bank 2022. Pathways Towards Economic Security Indonesia Poverty Assessment 1 Context Rural development continued to depend on However, a lack of middle-class jobs hampered agriculture, but diversification into non-agricultural productivity and put at risk livelihoods for a growing incomes has started to play a larger role.8 Over the class of economically insecure households.15 Even last two decades, most rural households remained before the COVID-19 pandemic, Indonesia struggled engaged in agriculture. While agriculture can provide a to create enough jobs providing sufficient income pathway out of poverty, many rural farmers remained to belong to the middle class. Instead, many workers poor, struggling with low productivity, poor market remained trapped in low-VA services with limited access, and exposure to shocks without access to productivity. As Indonesia’s population begins to age and resilience measures.9 Not surprisingly, rural households lose its “demographic dividend” over the next decade, were, hence, diversifying their income, but often in low enabling the private sector to create more middle-class value-add (low-VA) services. These were able to provide jobs becomes even more important to sustain economic an escape from poverty, but hardly offered a pathway growth, as well as to continue reducing poverty. towards the middle class. Digitalization can create new opportunities, but if Significant progress has also been made on several not well managed can heighten the risk of a “digital dimensions of gender equality. These included divide”.16 Indonesia has made big strides in improving achieving gender parity in gross school enrollment rates10 connectivity. Access to the internet jumped from and expanding basic health services, although these 13 percent in 2011 to 51 percent in 2019. However, improvements had not yet translated into increased households in rural and remote areas continued to have women’s economic participation. Indonesia’s female significantly lower access, while poorer households labor force participation rates have remained low and struggled to afford good quality access. Nevertheless, were stubbornly stagnant over the last two decades.11 the benefits of digitalization in Indonesia cannot be overstated. Using eCommerce, consumers benefitted More generally, upward mobility has been a success from more choice, lower prices, and more comfort, story in Indonesia, and remains critical to reach its although the high cost of logistics, limited connectivity, high-income country ambitions.12 Of everyone who and low trust in digital payments constrained the was poor or vulnerable to fall into poverty in 1993, benefits from eCommerce. For workers, digitalization over half had become either middle class or realistically offered new opportunities in the gig economy or as aspiring to it twenty years later. By 2016, one in five digital entrepreneurs, but also transformed existing Indonesians had entered the middle class, compared to opportunities by increasing productivity. But taking less than 7 percent at the beginning of the millennium.13 advantage of these opportunities requires digital skills, Continued expansion of the middle class in the years and poorer households and remote areas often did not and decades to come underpin Indonesia’s ambitions have the connectivity and skills to take advantage of to become a high-income country, driving economic eCommerce and digital opportunities. Digitalization also growth, widening, and deepening the tax base, and improved delivery of public services, but – as COVID-19 expanding the constituency for better governance.14 has shown – fragmentation of data and systems as well as a lack of coordination diminished the ability to fully exploit these benefits. 8 World Bank 2020b. 9 Asian Development Bank 2019. 10 Parity in gross enrollment rates at the primary school level was achieved in the mid-1990s, and at the secondary level by mid-2000s. (UNESCO Institute for Statistics (UIS). UIS.Stat Bulk Data Download Service. Accessed October 24, 2022. apiportal.uis.unesco.org/bdds). 11 World Bank 2020a; Cameron, Suarez, and Rowell 2018. 12 World Bank 2018b. 13 Pratomo, Syafitri, and Anindya 2020. 15 Wihardja and Cunningham 2021. 14 World Bank 2019a; see report for class definitions. 16 World Bank 2021a. 2 Pathways Towards Economic Security Indonesia Poverty Assessment Context Looking ahead Many of the poor live in high-risk areas and will need to Indonesia is an aspiring upper middle-income country become more resilient, which will require infrastructure facing new challenges such as climate change, COVID-19, as well as social investments.24 Thus, the readiness of and global uncertainties in its inclusive growth path. Indonesia’s social protection and disaster response frameworks will be tested in the medium term while The path towards high-income country status will be questions will also be raised about the distributional challenging. The transition to high-income status often impacts of policy measures aimed at both adaptation involves a change in growth strategies,17 a challenge that and mitigation. has been successfully navigated only by few countries. Some of them in East Asia, such as South Korea, pursued Indonesia will need to recover from COVID-19’s a strategy of labor-intensive, export-led manufacturing disruption of economic and social progress. The while taking advantage of a youthful population and global coronavirus pandemic (COVID-19) dramatically demographic dividend.18 However, these successes slowed economic activity around the world, causing occurred in a context without the headwinds that governments to implement lockdown measures, Indonesia and the region are now facing with an aging individuals to reduce both their mobility and economic population19 and shifting regional and global trade activity, and disrupting firms’ production processes. patterns.20 Sustaining inclusive economic growth in the These broader economic shifts affected both firms’ region requires fostering greater economic mobility demand for labor and workers’ ability and willingness to while enhancing economic security across the income work, with potentially structural implications for post- distribution.21 COVID-19 economies. Indonesia was no exception. The Government deployed various fiscal and other policy Climate change affects the economy and livelihoods responses to cushion the impact of the crisis while through domestic and international mitigation recognizing the importance of including Indonesia’s measures as well as adaptation.22 International most vulnerable in the recovery, setting a target of mitigation measures, such as the EU’s carbon border eliminating extreme poverty by 2024.25 Despite falling adjustment mechanism, are likely to reduce demand into recession, the recovery is already underway— for carbon-intensive goods, affecting countries like although the implications of the pandemic are not yet Indonesia which export such goods. In addition, fully understood. domestic policies to reduce carbon emissions such as, for example, the Nationally Determined Contributions (NDCs), will affect relative prices and the structure of the economy, creating winners and losers. Strategies need to be put in place to identify and compensate the losers. Although nationally the number of mining workers is relatively low, in Kalimantan almost 1 million workers, representing 8 percent of the population of the region, work in mining, emphasizing the need for contextual solutions. Adaptation will also continue to play a large role, given the increasing frequency and intensity of natural disasters and other environmental shocks.23 17 Bulman, Eden, and Nguyen 2016. 18 World Bank 1993; Asian Development Bank 2007. 19 World Bank 2015a; Widianto and Isdijoso 2020. 20 World Bank 2022i. 21 World Bank 2018a. 22 World Bank 2022h. 24 Kementerian PPN (Bappenas) 2021. 23 Asian Development Bank 2022. 25 World Bank 2020e; World Bank 2020g. Pathways Towards Economic Security Indonesia Poverty Assessment 3 Context Russia’s invasion of Ukraine further creates rather than the extreme poor, given the extreme poor uncertainties for the future. Prices for most now represent a small and declining 2 percent of the commodities have risen significantly in 2022 after the population in 2021. We analyze spatial differences with start of Russia’s invasion of Ukraine, and are expected greater consideration of prices, rather than focusing to remain high in the medium-term.26 In addition to only on official poverty lines, which shift every year, exacerbating food insecurity globally, they can increase thus limit comparability across provinces. This section inflation, magnifying financial vulnerabilities. Even also presents trends beyond poverty, expanding the though Indonesia exports commodities and benefits focus to economically insecure households, as well as from improved terms-of-trade, it imports food and non-monetary dimensions of poverty by deploying a domestic prices are increasing. Thus, the shock affects “lifecycle opportunity” approach. the economy and livelihoods, and careful mitigation measures are needed at a time when both household Second, this Poverty Assessment explores drivers and government finances and coping mechanisms are of poverty reduction and inequality to understand already severely strained by the pandemic. challenges on the path towards higher income. Examining the structural drivers of poverty and This report inequality reduction, the analysis zooms into the This Poverty Assessment for Indonesia recommends period from 2014 to 2019. We discuss COVID-19 effects policies to promote inclusive growth and shared prosperity starting in 2020 in-depth in the following chapter. This in the context of COVID-19, climate change, and global allows extracting medium-term structural drivers of uncertainties based on existing and new analysis of trends poverty reduction without confounding the analysis and drivers of poverty and inequality. with COVID-19 issues. We look at the drivers of poverty reduction through a simple framework, including Facing these challenges, Indonesia will need to derive demographics, employment and education, prices, and a careful way forward to meet its economic and taxes, and public spending policies. Taxes and public social objectives, while broadening its poverty focus. spending are particularly relevant in the current context The 2022 Indonesian Poverty Assessment considers of a tighter fiscal position. The analysis of drivers of household welfare in Indonesia as the country and inequality looks back to previous periods and explains the world emerge haltingly from the COVID-19 crisis, inequality trends over time. deal with climate change adaptation and mitigation, and are exposed to global uncertainty. It employs a Third, this Poverty Assessment analyzes and discusses broadened definition of poverty given the gains in shocks in the context of climate change, COVID-19, reducing extreme poverty. It asks what can be done to and global uncertainties. Idiosyncratic and covariate eradicate extreme poverty, lift the remaining poor and shocks can destroy households’ livelihoods. Idiosyncratic economically insecure into economic security, and drive shocks affect households but leave their communities greater inclusivity. First, this report reviews trends and developments over the last two decades. The first chapter updates poverty and inequality trends based on the US$ 3.20 PPP (2011) poverty line the World Bank uses to define poverty.27 In addition, we show trends for extreme poverty, defined using the US$ 1.90 PPP (2011) poverty line. However, the focus of the report is on the poor World Bank 2022a; Bank Indonesia 2022. 26 This report deploys the 2011 PPP poverty lines for alignment with the 27 Government’s goal (Box on p. 8). 4 Pathways Towards Economic Security Indonesia Poverty Assessment Context largely unaffected. In contrast, co-variate shocks, such growth towards economic security. Three as natural disasters, affect whole communities at once. complementary pathways to reach economic security In addition, climate change mitigation measures— emerge, supported by crucial data, and we identify that is, coal and carbon taxes—will affect specific type knowledge gaps in each area to improve future of workers, but also affect households overall due to policies: overall economic effects. Price shocks recently triggered (i) Create better opportunities to increase productivity by Russia’s invasion of Ukraine also affect livelihoods. and its implications for economic growth, poverty Finally, this section will discuss COVID-19 in more detail and inequality reduction. given its large and recent negative effects on livelihoods, (ii) Better protect against poverty by safeguarding including a discussion of drivers of poverty reduction in poverty reduction progress and building resilience those years. against shocks. Fourth, based on the analysis, we offer policy (iii) Finance pro-poor public investments in the context recommendations to foster inclusive and sustainable of limited fiscal space. Pathways Towards Economic Security Indonesia Poverty Assessment 5 CHAPTER 2 POVERTY AND INEQUALITY TRENDS Photo: © World Bank 6 Pathways Towards Economic Security Indonesia Poverty Assessment 2. POVERTY AND INEQUALITY TRENDS After experiencing high growth rates during the commodity rubber, and crude oil. The improved terms-of-trade boom, Indonesia’s growth moderated, but with domestic translated into increased investments (Figure 2.2), consumption picking up, generated higher labor income creating jobs and spurring domestic consumption, than capital gains. which in turn stimulated the service sector. However, the agricultural sector declined only slowly together with A fter a strong recovery from the 1997-98 Asian Financial Crisis (AFC), Indonesia enjoyed solid growth fueled by a commodity boom until 2013. an early shift from the manufacturing to the services sector.28 Following a 13 percent contraction in 1998, Indonesia’s From 2014 to 2019, the tailwinds of the commodity output growth recovered strongly (Figure 2.1) driven boom receded, revealing the natural resource by a commodity boom since 2003. The commodity dependence of Indonesia’s capital investments. The boom, thanks to rapid growth in China, India, and other ending of the commodity boom exposed Indonesia to emerging economies, led to high demand for many the typical aftermath resource-dependent countries commodities and increases in their prices. Indonesia experience, with growth dropping to an average of 5 benefitted from increased international demand for percent per year. With declining terms-of-trade, the twin commodities it exported, such as coal, crude palm oil, surplus in current and fiscal accounts turned into twin FIGURE 2.1: GDP growth (LHS) and GDP -per-capita (RHS) FIGURE 2.2: Terms-of-trade in US$ billions and growth of gross from 1990 to 2021 fixed capital formation from 2007 to 2019 15 15,000 10 140 10 10,000 8 130 5 5,000 120 6 110 0 0 4 100 -5 -5,000 2 90 -10 -10,000 0 80 -15 -15,000 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 Terms of trade (RHS) GDP growth (LHS) GDP per capita, PPP 2017 (RHS) Gross xed capital formation (growth; LHS) FIGURE 2.3: GDP composition from 1997 to 2021 FIGURE 2.4: Annualized employment growth and change in productivity, bubble size reflects value-add of sec 100 6 80 4 Services Employment growth, percent Industry Industry Services 60 2 40 0 Agriculture Agriculture 20 -2 0 -4 0 2 4 6 1997 1999 2001 2003 2005 2007 2009 2011 2013 2015 2017 2019 2021 Change in productivity, percent Agriculture Industry Services 2000-2013 2014-2019 28 World Bank 2015c. Pathways Towards Economic Security Indonesia Poverty Assessment 7 Poverty and Inequality Trends deficits exerting pressures on both the monetary and workers from agriculture to low-productivity services.30 fiscal situation. Despite some success in diversifying the While service-led development is possible,31 growth in economy in previous years, gross fixed capital investment productivity of services in Indonesia dropped from an declined with terms-of-trade (Figure 2.2) exposing its average of 4.0 percent from 2000 to 2013 to 1.7 percent resource dependence. from 2014 to 2019 as growth of low-VA outpaced high- VA service jobs. Despite slow-down in the economy, Indonesian households benefitted from strong employment As a result, Indonesia’s structural transformation, growth, even though this came in service sector and although lagging other countries, was still sufficient with limited productivity gains. The output share to briefly allow the economy to reach upper-middle of manufacturing dropped from 48 percent in 2002 income status. Indonesia’s agricultural share of GDP to 41 percent in 2019. In the same period, the service remains relatively high, while its manufacturing share sector expanded from 36 to 46 percent (Figure 2.3) is prematurely dropping (Figure 2.5). Non-tradable with annualized growth in employment in services services are increasing, but tradable services—often increasing from 3.0 between 2000 and 2013 to 4.1 of higher productivity—stagnating at a low level. percent between 2014 and 2019 (Figure 2.4). Hence, the Nevertheless, Indonesia graduated to an upper middle- economy continued the deindustrialization that began income country in 2019 and became the seventh largest at the turn of the century. This phenomenon—often economy in the world ranked by GDP, and the only G20 called premature deindustrialization29—is reflected in nation in Southeast Asia. minuscule productivity increases gained from a shift of FIGURE 2.5: Terms-of-trade in US$ billions and growth of gross fixed capital formation from 2007 to 2019 Agiculture Manufacturing 25 80 20 % of ttl. value added 60 % of WAP 15 40 10 2019 20 1991 5 0 0 4 6 8 10 12 4 6 8 10 12 Log of GDP per capita (2017 international $) Log of GDP per capita (2017 international $) IDN IDN Non−tradable services (including construction) Tradable services 60 60 50 50 % of real value added % of real value added 40 40 1991 2018 30 30 20 20 10 10 0 0 4 6 8 10 12 4 6 8 10 12 Log of GDP per capita (2017 international $) Log of GDP per capita (2017 international $) IDN IDN Source: Authors’ compilation based on World Bank 2023 30 World Bank 2020c. 29 Rodrik 2015. 31 Nayyar, Hallward-Driemeier, and Davies 2021. 8 Pathways Towards Economic Security Indonesia Poverty Assessment Poverty and Inequality Trends The COVID-19 pandemic, which reached Indonesia Indonesia achieved impressive reduction in extreme in March 2020, made Indonesia’s achievement to poverty, having virtually achieved the goal of upper middle-income status short-lived. The COVID-19 eradicating extreme poverty. Indonesia’s extreme pandemic severely disrupted the economy, first due to poverty rate dropped from 18.8 percent in 2002 to 2.7 far-reaching mobility restrictions, and later by the spread percent in 2019 (Figure 2.6), using the US$ 1.90 2011 of the virus itself. This took a significant toll on economic PPP per day. Amidst these promising developments, the activity and triggered Indonesia’s first recession since Government of Indonesia (GOI) committed in 2020 to the AFC. The economy shrank by 2.1 percent in 2020, eradicating extreme poverty by 2024. Indeed, extreme necessitating a downgrading to lower middle-income poverty continued to drop further to 1.5 percent in status.32 Its contraction in real GDP, however, was smaller 2022, basically eradicating extreme it. A small amount of than in most ASEAN countries, in part due to more frictional poverty is likely to remain, with further progress limited mobility restrictions and a large government being difficult to monitor given measurement error and fiscal package put in place to support households, firms, statistical inaccuracies. and healthcare.33 The contraction was also much smaller than that experienced during the AFC (Figure 2.1). Poverty, more adequately defined at US$ 3.20 2011 Growth rebounded to 3.7 percent in 2021, supported PPP, similarly dropped steeply, but remained higher by public expenditures, exports, and improved terms- than for peer countries. The share of the poor defined of-trade.34 as living below the poverty line for lower-middle income countries at US$ 3.20 2011 PPP dropped from National 61 percent in 2002 to 20 percent in 2019 and further Indonesia made impressive gains in poverty reduction with to 15.7 percent in 2022. While the pace of poverty more inclusive growth and declines in inequality since reduction is comparable to peers (Figure 2.9), the poverty 2014, but more focus on people beyond the extreme poor rate is slightly above other countries with a similar PPP is warranted. GDP per capita (Figure 11) and still higher than for its regional peers (Figure 2.10). Going forward, a broader To maintain consistency with the Government’s definition of poverty, e.g., around the lower-middle definition of extreme poverty, this report continues to income poverty line, is more appropriate in a country like use 2011 PPP poverty lines instead of the new 2017 Indonesia, which is on the verge of becoming a higher- PPP estimates. The World Bank adopted the 2017 PPP middle income country. exchange rates for global poverty monitoring, revising slightly the estimates for poverty also for Indonesia (Box However, progress almost stagnated using official 2.1). The revision is driven by declines in purchasing poverty estimates due to complications in calculating power against the US$ and increases in the real value comparable official poverty lines between years. Using of the global poverty lines. However, the revision is official poverty lines, poverty declined from 18 percent in not an actual increase in poverty due to economics, 2002 to 11 percent in 2014 before reaching 9.4 percent in but a change in measurement and its definition. It is 2019, followed by a slight uptick to 10.1 percent in 2021, recommended to use the 2017 PPP estimates for future then dropping back to 9.5 percent in 2022. However, analysis and poverty tracking as they are based on the official poverty lines are updated annually at the improved data collection and methodology and, hence, province-level based on prices and diet for an ever- better capture relative price differences across countries. wealthier reference group. This makes the poverty lines However, given the Government’s current objective to hard to compare as they shift each year at the province eradicate extreme poverty measures at the 2011 PPP, this level (Box 2.3). Therefore, this Poverty Assessment defines report will continue to use the 2011 PPP estimates. extreme poverty at the international poverty line (IPL) of US$ (1.90) 3.20 2011 PPP35. 32 OECD 2021. 33 International Monetary Fund 2022. USD 1.90 (2011 PPP) is equivalent to IDR 10282.4; USD 3.20 (2011 PPP) is 35 34 World Bank 2022i. equivalent to IDR 17317.7. Pathways Towards Economic Security Indonesia Poverty Assessment 9 Poverty and Inequality Trends Box 2.1: Poverty for Indonesia is assessed based on revised 2011 PPP estimates Indonesia’s poverty rate under the 2017 PPP estimates is higher than under 2011 PPP estimates. The World Bank adopted the 2017 PPP exchange rates for global poverty monitoring in fall 2022. The new global poverty lines of US$2.15, US$3.65, and US$6.85 reflect the typical national poverty lines of low income, lower middle-income, and upper middle- income countries in 2017 prices. Given improvements in data collection and methodology for the 2017 PPP estimates, the new poverty lines represent better measures of poverty. For Indonesia, a higher poverty rate will be reported when expressed in 2017 PPP, particularly at the lower middle and upper middle-income lines (Figure Box 2.1.1). The different poverty rates do not represent a change in poverty in Indonesia. The change in the poverty rate when using 2017 PPP is consistent over time (Figure Box 2.1.2), emphasizing that this is not an actual change in poverty. The change can be explained by two factors: (i) declines in purchasing power against the US$, with Indonesia having become more expensive in relative terms; and (ii) increases in the real value of the global poverty lines, as many upper middle- income countries raised the standards by which they determine people to be poor since the previous update.36 FIGURE BOX 2.1.1: Indonesia’s poverty rate at 2011 and FIGURE BOX 2.1.2: Extreme poverty headcount 2017 PPP estimates for 2019 rates for Indonesia 61.9 100 52.2 80 60 24.7 40 19.9 20 2.7 4.4 0 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022 International poverty Lower middle income Upper middle income line ($1.9 vs $2.15) class poverty lines class poverty lines ($3.2 vs $3.65) ($5.5 vs $6.85) IPL 2011 PPP IPL 2017 PPP LMIC 2011 PPP LMIC 2017 PPP Revised 2011 PPP 2017 PPP Note: IPL: International Poverty Line; LMIC: Lower Middle Income Class Poverty Line. Source: World Bank Dataviz For consistency with the Government’s goal to eradicate extreme poverty, we apply the 2011 PPP poverty lines consistently. The Government has set a target of eliminating extreme poverty by 2024, measured by the 2011 PPP poverty lines. Hence, this Poverty Assessment will continue to define (extreme) poverty using the 2011 PPP poverty lines, to ensure consistency and alignment with Government’s objectives. FIGURE 2.6: Poverty headcount rates using $1.90, $3.20, and FIGURE 2.7: Relative and absolute change in poverty at US$ 3.20 $5.50 per-day 2011 PPP as well as national poverty line (NPL) 2011 PPP from 2009 to 2018/2019 100 0 80 -5 60 -10 Percent Percent -15 40 -20 20 -25 0 -30 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022 Relative Absolute NPL $1.9 $3.2 $5.5 Indonesia Vietnam Philippines China Source: Authors’ calculations using SUSENAS (2002-2022) Source: Authors’ compilation using World Development Indicators 10 Pathways Towards Economic Security Indonesia Poverty Assessment Poverty and Inequality Trends FIGURE 2.8: Poverty rates for Indonesia and its economic peers FIGURE 2.9: Log GDP per capita (PPP) vs poverty rate for peers 20 9.9 Thailand 15 9.7 Log GDP per capita (PPP) China Percent 10 9.5 9.3 Indonesia 5 Vietnam 9.1 0 8.9 Philippines Philippines Indonesia Vietnam China Malaysia Thailand Philippines Indonesia Vietnam China Malaysia Thailand 8.7 Lao PDR 8.5 0 10 20 30 40 $1.90 $3.20 US$ 3.20 PPP poverty rate (2018) Source: Author’s compilation using World Development Indicators, using latest available Source: Authors’ compilation using World Development Indicators. estimates: Indonesia 2022, Thailand 2020, China 2019, Philippines and Vietnam 2018, Malaysia 2015. Despite progress, one in six households are investments at the household level. Reliance on adverse structurally poor; that is, they have insufficient assets strategies when coping with income shocks —such to escape poverty. The level of structural poverty as the sale of productive assets— can further reduce dropped from 40 percent in 2011 to 16 percent in 2019 productivity.37 Even before shocks, economically (Figure 2.10), slightly outpacing the reduction in poverty. insecure households may anticipate them and adopt Structurally poor households have consumption levels conservative or risk-averse production and investment below US$ 3.20 2011 PPP, and are likely to remain poor strategies that lower consumption and/or investment.38 in the future (Box 2.2). This generally occurs when a Thus, regardless of whether poor household adopt household’s physical assets and/or human capital adverse coping strategy after or before shocks, they endowments are too low to allow the household to reduce long-term productivity, which in turn lowers their generate adequate income and sustain a consumption chances of securely escaping poverty. level above the poverty line.36 Since 2014, consumption growth has been stronger In addition, a large share of the population remains for Indonesia’s bottom 40 percent of households, economically insecure as they still are susceptible to a reversal from commodity-boom years. Until becoming poor in the future. In 2019, 40 percent of 2014, consumption growth—especially during the Indonesians were economically insecure (Figure 2.10). commodity boom—was biased against the poorest Most of these households are non-poor, but can fall 40 percent of households (bottom 40). From 2002 to into poverty when being exposed to a shock (Box 2.2). 2005, the consumption growth rate for the bottom 40 The share of economically insecure households has was only 80 percent of the annualized consumption hardly changed since 2011, but a substantial share of growth for all households, and only 65 percent of that structurally poor became economically insecure, while from 2006 to 2013. This trend reversed from 2014 to a similar share of economically insecure households 2019 when consumption of the bottom 40 grew by 4.8 managed to reach economic security. percent annually compared to an overall annualized growth of 4.2 percent. Thus, the bottom 40 had 1.2 times Indonesia’s success in reducing poverty the average consumption growth, translating into a notwithstanding, this degree of economic insecurity “shared prosperity premium” of 0.6 percentage points. undermines productivity progress. Short spells of However, this reversal was not substantially due to higher lowered consumption can reduce productivity in the consumption growth of the bottom 40 (annualized at long run due to adverse effects on human capital 37 See for example Alderman, Hoddinott, and Kinsey 2006; Gubert and Robilliard 2007; Rosenzweig and Binswanger 1993; Klasen and Waibel 2013. 36 World Bank 2022b. 38 Elbers, Gunning, and Kinsey 2007. Pathways Towards Economic Security Indonesia Poverty Assessment 11 Poverty and Inequality Trends Box 2.2: Definition of economic insecurity Economic insecurity covers subsets of poor but also FIGURE BOX 2.2.1: Visualization (for 2019) of the extreme “non-poor” households not structurally poor but remain poor, poor and non-poor and its relationship to economic (in-) security susceptible to falling into poverty. Structurally poor is defined as being poor with a likelihood of more than 10 percent to remain poor the next year. Economic insecurity is Extreme poor defined as either: (i) being poor with a likelihood of less than Poor Non-poor - 10 percent to be poor next year, or (ii) being non-poor with a likelihood of more than 10 percent to be poor next year. Structurally poor Economically insecure Economically secure Thus, economically secure households are non-poor with a Poor in the current year, but unlikely to be poor in the next year likelihood of less than 10 percent to be poor next year. We While the original classi cation allows for non-poor to be chronic poor, thus categorize the three groups as (i) extreme poor, (ii) poor, the revised methodology ensures that non-poor are either economically insecure or secure, but cannot be chronically poor and (ii) non-poor (Figure Box 2.2.1).39 Source: Authors’ illustration The structurally “extreme poor”, economically insecure “poor”, and economically secure “non-poor” face different challenges. The extreme poor are not only coincidentally poor at the time of the survey (typically a fraction of the poor) but have high probability of being poor across time periods due to, for instance, missing human and physical assets to escape poverty. In contrast, the economically insecure poor may have such assets but are susceptible to falling into poverty due to a shock. The economically secure non-poor households are more resilient and not susceptible to falling into poverty even during shocks. The concept of economic insecurity roughly equivalent to “vulnerability to poverty” used previously in Indonesia, but is based on a different methodology. Previous work40 estimating vulnerability exploited SUSENAS (National Economic Survey) 2008 to 2010 panel household data to estimate the chance of a household of falling below the official poverty line in the next year based on its per capita consumption in the current year.41 The consumption level that corresponded to a 10 percent chance of being poor next year was defined as the “vulnerability line”.42 Given the unavailability of more recent panel data,43 we used a cross-sectional approach to estimate the probability of being poor next year for each household.44 Both approaches yield similar estimates of vulnerability/economic insecurity and its trends, but the cross- sectional approach has the advantage of allowing assessment of susceptibility of falling into poverty due to covariate and idiosyncratic shocks (see chapter on Shocks). 4.6 percent from 2006 to 2013) but came at the cost FIGURE 2.10: Share of population classified as structurally poor, economically insecure, and economically secure of the top 60, whose annualized growth dropped from 7.1 percent from 2006 to 2013 to 4.1 percent from 2014 100 to 2019. All of Indonesia’s peer countries, except for 80 Vietnam, have a positive—but often significantly larger— Percent 60 shared prosperity premium (Figure 2.11).39404142434445 40 39 See Estimating economic insecurity ‘Estimating economic insecurity’ in the Annex for details of the methodology. 40 The past approach was introduced by López-Calva and Ortiz-Juarez 2014 and 20 adopted by Wai-Poi (2014) for Indonesia, and used since in many country publications including World Bank 2019a. 0 41 Pritchett & Suryahadi, 2000; Wai-Poi, 2014; Chaudhuri, Jalan, and Suryahadi 2002; 2011 2019 2011 2019 World Bank 2019a. $3.20 $1.90 42 This consumption level was IDR 530,000 in 2016, equivalent to 1.5 times the poverty line. Structural poor Insecure Secure 43 The last round of the Indonesia Life Family Survey is from 2014, while the panel information for national surveys is currently not made available. Source: SUSENAS 2011 and 2019 44 Following Günther and Harttgen 2009. For its application to Indonesia, see Ali and Setiawan 2022. 45 Note that references cited within the Box applied Indonesia’s official poverty line while here the deflated international poverty lines are used. 12 Pathways Towards Economic Security Indonesia Poverty Assessment Poverty and Inequality Trends FIGURE 2.11: Consumption growth among the bottom 40 and FIGURE 2.12: Gini coefficient for Indonesia and its economic peers across the population, for Indonesia and peer countries 8 45 40 6 35 30 Percent 4 25 20 2 15 10 0 5 0 19 16 19 18 8 15 9 01 01 20 20 20 20 20 s2 a2 06 20 18 22 9 15 8 d p sia am ia 01 01 ne in an Re ys 20 20 20 20 20 ne Ch tn a2 s2 pi ala ail a, p d am sia ia do Vie ilip re ne Th in an re M ys ne In Ko Ch tn Ph pi ala ail a, do Vie ilip re Th M Ko In Bottom 40% All population Ph Source: Authors’ compilation based on World Development Indicators Source: Authors’ compilation based on World Development Indicators. Gini estimates for Republic of Korea, Malaysia, and the Philippines are based on income, while Thailand, Vietnam, Indonesia, and China are based on consumption FIGURE 2.13: Gini coefficient from 2002 to 2022 FIGURE 2.14: Unfair inequality in consumption and labor income, across years 0.5 60 0.4 50 40 0.3 Percent 30 0.2 20 0.1 10 0 0 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 Old Methodology New Methodology Labor income Consumption Source: Authors’ compilation based on SUSENAS Source: Authors’ calculation based on SUSENAS and SAKERNAS Notes: The underlying survey design and protocol changed from 2010 to 2011 by Notes: Results based on estimating general entropy of consumption/income predicted by improving the representativeness of the sample.43 Thus, the increase between those years a regression of log consumption/income on gender and location (urban/rural as well as might be attributable to a change in measurement. provinces) compared to predicted value using a regression with a full set of co-variates. Accordingly, inequality declined after peaking in 2014. from 2011 to 2013, inequality started to drop from .39 From 2002 to 2010, inequality increased considerably in 2014 to .37 in 2019, driven by the shared prosperity from a Gini coefficient of .32 to .36 points (Figure premium for the bottom 40. From 2019 to 2022, the Gini 2.13), due to unequal consumption growth during stayed constant at .37. Inequality in Indonesia is in the the commodity boom.46 The large increase of 3 points mid-range of its regional peers (Figure 2.12). between 2010 and 2011 was possibly due to a change in survey methodology.47 Following stagnating inequality Unfair inequality remains high, explaining between one-third to one-half of monetary inequality. 46 The Gini coefficient measures the inequality within a population, with respect to a measure like income or consumption. A Gini coefficient of 0 reflects perfect equality, Inequality can stem from two sources: (i) differences where all individuals have the same income / consumption, while the highest Gini coefficient of 1 (or 100%) reflects maximal inequality with one individual holding in preferences, abilities, and effort; and (ii) differences all income/consumption. Technically, the Gini is the area between the diagonal and the cumulative income/consumption distribution (Lorenz curve). The area is 0 in access to opportunities (“unfair” inequality). Unfair if each individual contributes the same to the population’s income/consumption. It is 1 if only 1 individual contributes everything. inequality due to gender and location explains 47 Note that the Gini time series is affected by a change of survey methodologies related to sampling and survey protocol, such that the increase from 2010 to about one-third of income inequality and one-half of 2011 should not be interpreted economically. The time-series until 2010 was based on the Economic Census 2010, while from 2011 onwards the master consumption inequality (Figure 2.14). Thus, place of birth sampling frame was taken from the Population Census 2010, resulting in new sampling weights. With an improved protocol to minimize unit non-response, and gender often determine access to opportunities the surveys from 2011 onwards were also able to interview richer households, which previously were more often replaced with other – often still non-poor but and disadvantages some vulnerable groups face for the less rich – households. This explains the stark impact on inequality measures with only limited impact on poverty measures (SUSENAS User Guide). duration of their lives. Pathways Towards Economic Security Indonesia Poverty Assessment 13 Poverty and Inequality Trends Sub-national over time while adjusting to changes in prices and Poverty rates across Indonesia have converged, but with living standards. In practice, the minimum standard some regions still lagging. of living represented by lines in Eastern Indonesia and other initially poorer regions increased by more than Urban and rural poverty rates have converged over in initially richer regions, obscuring the convergence the last two decades.48 While the rural poverty rate of these regions. They lag, in fact, by much less when remains higher, over time it has converging toward the using an absolute measure of poverty. urban poverty rate (Figure 2.15).49 Between 2002 and 2022, the poverty rate at US$ 3.20 2011 PPP fell from 73 to Most of Indonesia’s poor now live in urban areas, 16 percent in rural areas, almost completely closing the roughly in proportion to the share of the general gap with urban areas from 27 to less than 1 percentage population classified as urban. As Indonesia’s point (pp). Extreme poverty rates in urban and rural areas urbanization rate has gradually increased,51 so too has have been virtually indistinguishable since 2015, and in the share of poor living in urban areas, rising steadily 2022 stood at 1.5 percent. The depth (poverty gap) and from 30 to 53 percent, so that by 2022 a slight majority severity (squared poverty gap) of poverty continuously of the poor were residing in urban areas (Figure 24). In declined since 2002 and have similarly converged in absolute terms, the number of poor living in rural areas urban and rural areas. declined from 82 million in 2002 to 19 million in 2022, and in urban areas from 41 to 24 million respectively. Poverty rates in lagging regions remain higher than The share of extreme poor living in urban areas also elsewhere but are catching up, contrary to common rose from 39 to 60 percent over the same period. In perception. Underneath national aggregates, different 2022, 2.5 and 1.7 million extreme poor were living in regions in Indonesia’s vast archipelago face varied levels urban and rural areas respectively. A larger share of of deprivation (Figure 2.16). The so-called lagging regions the poor in urban areas has important implications for of Eastern Indonesia—namely the Maluku-Papua and service delivery and social protection, often making Nusa Tenggara island-regions—have had the highest service delivery less expensive. poverty rates in the country. However, since 2002, these regions have achieved impressive gains in poverty While the Java-Bali region is still home to the vast reduction, almost catching up with other regions. Poverty majority of the poor, an increasing share of the headcount rates declined from around 80 percent in extreme poor live in the lagging regions. Historically, 2002 to 26 and 27 percent in 2022 in Maluku-Papua and Indonesia has had a highly uneven distribution of poor, Nusa Tenggara respectively, compared to an average of and this remains the case today. Traditionally lagging 15 percent in the rest of country. Extreme poverty also regions have the country’s lowest populations, resulting reached closer to that in other regions but remained in the vast majority of poor Indonesians living in the about 5 pp higher in 2022. These estimates contrast populous, higher-density island regions of Java-Bali with the popular discourse around “lagging regions”, and Sumatera (Figure 2.18). Combined, they hold 76 which describes poverty rates in Eastern Indonesia as percent of Indonesia’s poor in 2022. The lagging regions, stagnating and not converging to the national poverty in contrast, were home to only 12 percent of the poor. rate.50 This view partly arises from the nature of the The geographical distribution of extreme poverty shows official poverty line methodology in Indonesia (Box a similar pattern, but with a notable difference: the share 2.3), under which poverty lines across provinces and of extreme poor in the Java-Bali and Sumatera regions within urban and rural areas increased at different rates declined from 79 to 64 percent between 2002 and 2022, while that of Nusa Tenggara and Maluku-Papua increased 48 See Box 2.3 for a thorough discussion of spatial deflation and its limitations. from 12 to 22 percent.52 In other words, extreme poverty 49 Note that urban and rural poverty trends are not independent of each other, as urbanization is partly driven by a re-classification of rural settlements becoming urban (see footnote 105 on page 31 for more details). Given converging rural- urban trends, this mixing effect is unlikely to explain the general urban and rural 51 Roberts, Gil Sander, and Tiwari 2019. trends for socio-economic indicators. 52 This was disproportionately high relatively to the share of these regions in 50 World Bank 2006; World Bank 2020b; Roberts, Gil Sander, and Tiwari 2019. Indonesia’s population (7 percent). 14 Pathways Towards Economic Security Indonesia Poverty Assessment Poverty and Inequality Trends became more concentrated in lagging regions, further to decline, those that remain extremely poor live in from areas that areas that are the economic engine of increasingly remote and difficult to reach areas. the country. As the number of extreme poor continues Box 2.3: The advantage of using absolute poverty lines to compare poverty across Indonesia’s provinces Indonesia’s poverty lines are annually adjusted for each province, separately for urban and rural areas. The poverty lines are defined by Indonesia’s national statistics office Badan Pusat Statistik (BPS) at the provincial urban/rural level. Each poverty line is updated annually as the amount of money required to obtain 2,100 calories per day, using as the reference group households that fall between the consumption percentile of last year’s poverty line and the next 20 percentiles. The methodology adds a small amount for basic non-food items. The methodology to estimate poverty lines leads to complications in comparing provincial urban to rural poverty, undermining the ability to compare trends across provinces. The lines are updated based on a reference group, which is above the last year’s poverty line. With improving living standards, the reference group will consume more expensive calories as they transition from a diet of necessity to a diet of choice. This increase of the minimum standard of living (Figure Box 2.3.1) can be conceptually desirable. However, moving the poverty lines annually as well as independently across provinces and urban/rural erodes comparability, especially between provinces and when comparing urban to rural areas. For example, the US$ 3.20 2011 PPP poverty line has doubled over the last 20 years due to prices, while the national poverty line nearly tripled due to prices and changes in diet. FIGURE BOX 2.3.1: National and international poverty FIGURE BOX 2.3.2: Poverty rates based on lines in nominal and constant terms (IDR) official poverty lines, by regions 600 40 500 35 400 30 Thousands 25 Percent 300 20 200 15 100 10 0 5 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 0 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 NPL NPL (Constant 2011) IPL 1.9 (Constant 2011) IPL 1.9 Sumatera Jawa-Bali Nusa Tenggara IPL 3.2 (Constant 2011) IPL 3.2 Kalimantan Sulawesi Maluku-Papua Source: Authors’ calculations based on SUSENAS FIGURE BOX 2.3.3: Poverty rates at US$ 1.90 2011 PPP FIGURE BOX 2.3.4: Poverty rates at US$ 1.90 2011 PPP without within-year spatial deflation, by region after applying within-year spatial deflation, by region 40 40 30 30 Percent Percent 20 20 10 10 0 0 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 Sumatera Jawa -Bali Sumatera Jawa-Bali Nusa Tenggara Kalimantan Nusa Tenggara Kalimantan Sulawesi Maluku -Papua Sulawesi Maluku-Papua Source: Authors’ calculations based on SUSENAS Pathways Towards Economic Security Indonesia Poverty Assessment 15 Poverty and Inequality Trends Box 2.3: The advantage of using absolute poverty lines to compare poverty across Indonesia’s provinces (contd) The international poverty lines are absolute lines lacking appropriate spatial deflation. In contrast, the international poverty lines for the extreme poor at US$ 1.90 2011 PPP, the poor at US$ 3.20 2011 PPP, and US$ 5.50 2011 PPP are absolute lines, with their values only updated every few years to reflect changes in purchasing power parity. For Indonesia, they are deflated for urban and rural areas in 2011 by using the PPP estimator split into urban/rural based on the spatial differences implicit in the national official lines and the sample distribution of the International Comparison Program (ICP) data collection.53 Temporal deflation across years is based on Indonesia’s urban-only Consumer Price Index (CPI). Thus, the estimated poverty rates implicitly assume parallel price trends in urban and rural areas. Since no deflation is applied beyond urban and rural, it is not recommended to use the resulting estimates to spatially disaggregate further.54 This Poverty Assessment introduces spatial differences across provinces, separately for urban and rural areas to the international poverty line to allow comparisons of trends across provinces. Separately for urban and rural areas, the food consumption aggregate is spatially deflated at the province-level.55 The spatial food deflators are survey-based, derived from unit prices reported in the corresponding consumption surveys (SUSENAS). Given the lack of comparable non-food prices at the province and urban/rural level, non-food consumption, whose share hovers around 40 percent of consumption in Indonesia, cannot be spatially deflated. The spatial deflation is anchored at the existing national urban and rural poverty rates. Thus, only the province-level poverty rates are revised. Contrary to when using national poverty lines, the new provincial estimates show a converging trend for lagging regions. Indonesia’s poverty lines based on the official methodology reveal a stagnating, parallel trend of poverty across island regions, with Maluku-Papua and Kalimantan around 20 percent, lagging other island regions at around 10 percent (at US$ 1.90 2011 PPP; Figure 2.3.2). In contrast, the non-spatially adjusted international poverty lines show a strongly converging trend with all island regions approaching 5 percent, and Nusa Tenggara as well as Sulawesi only slightly trailing (Figure2.3.3). The spatial deflation adds more nuance with a larger range of poverty rates across island regions, but still well below 10 percent (Figure 2.3.3). In addition, Indonesia’s official poverty lines indicate a slight uptick of poverty in 2021 across island regions due to COVID-19, while the spatially deflated international poverty line only shows a small increase in poverty for Java-Bali. The remainder of this Poverty Assessment will exclusively use the spatially deflated PPP poverty lines. FIGURE 2.15: Poverty rates based on official FIGURE 2.16: Poverty rates for Nusa Tenggara (NT), poverty lines, by regions Maluku Papua (MP) and other regions 100 100 80 80 Percent Percent 60 60 40 40 20 20 0 0 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022 2002 2005 2008 2011 2014 2017 2020 $1.90-urban $1.90-rural MP Rural $3.20 MP Urban $3.20 $3.20-urban $3.20-rural NT Rural $3.20 NT Urban $3.20 $5.50-urban $5.50-rural Rest Rural $3.20 Rest Urban $3.20 Source: Authors’ calculations based on SUSENAS 535455 53 Note that only 3 percent of Indonesia’s market prices for the PPP are collected in rural areas. Asian Development Bank 2020a: pages 215 and 340. 54 Indonesia’s poverty lines do not require spatial deflation as the province and urban/rural estimation already implicitly deflates for spatial differences in prices. 55 Deflation of the consumption aggregate is the inverse of the deflation of the poverty line. 16 Pathways Towards Economic Security Indonesia Poverty Assessment Poverty and Inequality Trends FIGURE 2.17: Share of the poor and the general population in FIGURE 2.18: Share of the (extreme) poor, by region rural and urban areas 100 100 90 31 80 42 80 50 56 56 70 57 Percent Percent 60 60 50 40 40 69 30 58 50 20 20 44 44 43 10 0 0 2003 2019 2022 2003 2019 2022 2003 2019 2022 2003 2019 2022 Poor Extreme poor Poor Population Java-Bali Sumatera Sulawesi Rural Urban Kalimantan Nusa-tenggara Maluku & Papua Source: Authors’ calculations based on SUSENAS Opportunities over the lifecycle owned a cell phone, and internet access steadily rose. In Compared to monetary poverty, progress in non-monetary contrast, provision of clean water and sanitation services dimensions of wellbeing was relatively muted and lagged remained a challenge. Indonesia still has one of the Indonesian peer countries. highest percentages of open defecation, with 29 percent among the rural population and 14 percent among the Progress was uneven in improving access to urban population. Poor sanitation and hygiene practices education and basic infrastructure services, important and unsafe water lead to high infectious disease rates, dimensions of wellbeing beyond monetary poverty. which contribute to chronic malnutrition and worse health Since 2002, the highest level of education completed outcomes. Overall, despite substantially increasing over rose both among young adults (Figure 2.19) and in the the last decade and closing important gaps, investment in adult population (Figure 2.20). Access to electricity and infrastructure remains quite inadequate; the gap between gas expanded in both urban and rural areas (Figure 2.21 resources allocated and actual needs is estimated to and Figure 2.22). In 2022, the vast majority of Indonesians exceed the size of the entire Indonesian economy.56 FIGURE 2.19: Educational attainment among young FIGURE 2.20: Educational attainment among Indonesian adults (19-25 yo) in Indonesia household heads 100 100 80 80 60 60 Percent Percent 40 40 20 20 0 0 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2017 2018 2019 2020 2021 2022 Primary or Below Junior Secondary Primary or Below Junior Secondary Senior Secondary Tertiary Senior Secondary Tertiary 56 World Bank 2020d. Between 2000 and 2013, Indonesia spent an average of 3.6 percent of GDP on infrastructure per year, compared with 17.7 percent in China, 11.3 percent in Malaysia and 6.3 percent in Thailand. The deficit in spending is estimated to be US$1.6 trillion compared to other emerging and developing economies. In water and sanitation, for example, Indonesia is among countries with the lowest public sector spending (0.2 percent of GDP). Pathways Towards Economic Security Indonesia Poverty Assessment 17 Poverty and Inequality Trends FIGURE 2.21: Access to infrastructure services in 2015 FIGURE 2.22: Access to infrastructure services in 2022 2015 2022 Electricity Electricity 1 1 0.8 0.8 0.6 0.6 Internet 0.4 Gas Internet 0.4 Gas 0.2 0.2 0 0 Cellphone Sanitation Cellphone Sanitation Urban Rural Urban Rural Despite expanded access to services, the foundations Childhood health outcomes are therefore an important of human capital remain weak. Human capital consists component of the HCI. In Indonesia, these outcomes of the knowledge, skills, and health people invest in have improved but lag regional peers (Figure 2.21 and and accumulate throughout their lives, enabling them Figure 2.22). In 2022, 21.6 percent of children under age to realize their potential as productive members of 5 were stunted,60 higher than among peers.61 Unlike the society. Indonesia’s investments in education and basic sustained, dramatic declines in extreme poverty that services have substantially expanded access to services have brought Indonesia very close to its target of zero that support human capital development. Still, human extreme poverty by 2024, progress in reducing stunting development outcomes have shown limited progress. was slow and only improved recently. This ECD deficit Indonesia’s Human Capital Index (HCI), a summary reflects weak foundations of human capital formation measure of the amount of human capital that an with long-term implications for productivity. Indonesians Indonesian child born today can expect to acquire by whose growth was stunted in childhood were shorter as age 18, grew only modestly from 0.50 in 2010 to 0.54 young adults, exhibited lower cognitive function, and in 2020.57 The value remains below average for the East spent fewer years in education, factors linked to lower Asia and Pacific region as well as upper middle-income earnings later in life.62 countries, mainly due to Indonesia’s relatively poor performance on child survival and nutrition outcomes, Stagnating learning outcomes at older ages indicate as well as on standardized test scores.58 that many Indonesians are not adequately prepared for the transition from school to work. While Early childhood development (ECD) outcomes show educational attainment has risen, the quality of learning limited progress and nutritional deficits are large. remained low, with the attained 12.4 years of schooling The period from a child’s birth to age 5 is a critical translating into 7.8 learning-adjusted years of schooling.63 time for shaping long-term skills and productivity.59 More than half (53 percent) of children aged 10 were unable to read and understand a short, age-appropriate 57 Ranging between 0 and 1, the index takes the value 1 only if a child born today can expect to achieve full health (defined as no stunting and survival up to at text.64 More broadly, learning outcomes among students least age 60) and achieve her formal education potential (defined as 14 years of high-quality school by age 18). A value of 0.54 implies that a child born in aged 15 changed little in the last two decades and remain Indonesia today will be 54 percent as productive as he/she could be if he/she enjoyed complete education and full health. 58 Human Capital Project 2020. 60 Kementerian Kesehatan Republik Indonesia 2022. 59 Heckman 2007. During this time, the brain develops rapidly to build the 61 Progress was slower even than in Laos and Cambodia, which exhibited similar foundation of skills needed to thrive not just in school, but also later in life as levels of stunting, as well as China and Mongolia, where stunting levels were an adult, for example, in the labor market. Analyses of the long-term impact lower by about a third. The pace of reduction in stunting has recently picked up. of early childhood interventions in the United States have helped quantify the Since rollout of Indonesia’s Stranas Stunting strategy, the stunting rate declined long-term benefits of targeted investments early in life (García et al. 2020). And by as much as 6.4 percentage points from 30.8 percent in 2018 to 24.4 percent cognitive ability in early childhood has been shown to influence labor market in 2021. outcomes in adult life (Case and Paxson, 2008). Small-scale studies in Guatemala, 62 Giles et al. 2017; Perkins et al. 2016; Alderman, Hoddinott, and Kinsey 2006. South Africa, and Jamaica show that children with low levels of cognitive development in early childhood do poorly in school. See Stith, Gorman, and 63 Human Capital Project 2020. Choudhury 2003, Liddell and Rae 2001, and Walker et al. 2005 respectively. 64 World Bank 2021b. 18 Pathways Towards Economic Security Indonesia Poverty Assessment Poverty and Inequality Trends much lower than in peer economies (Figure 2.23). This health contributes to poor fetal and child health and are low level of basic cognitive skills among youth in reading, associated with higher chances of intrapartum-related mathematics, and literacy provides a weak foundation complications, infections, defects, and a higher probability for skills needed for tertiary education, which remains of neonatal death. Poor nutrition and high adolescent relatively rare (Figure 2.19) and also lags regional peers. fertility contribute to the high child stunting rate.66 It also constrains capacity to build more advanced skills increasingly in demand in Indonesia, most prominently Progress in strengthening women’s agency—integral for the digital economy.65 to their health and wellbeing, as well as that of their households—was also limited. Women’s agency Maternal health—a key contributor to development matters for several reasons. First, a person’s ability to outcomes, especially in childhood—improved little make effective choices and exercise control over one’s over the last two decades. Indonesia’s maternal life is a key dimension of well-being. Further, women’s mortality rate has stagnated and remains significantly exercise of agency improves their children’s welfare.67 higher than peer countries (Figure 2.24). Poor maternal The median age at first marriage slowly increased FIGURE 2.23: Infant mortality for Indonesia and its FIGURE 2.24: Stunting levels for Indonesia and its economic peers economic peers 50 35 30 Philippines 40 25 20 30 15 Vietnam 20 10 Malaysia 5 10 Thailand 0 China a d ia am s sia ne in an ys ne 0 Ch tn pi ala ail do Vie ilip Th 2000 2020 M In Ph Indonesia FIGURE 2.25: Learning outcomes in Indonesia and its FIGURE 2.26: Maternal mortality in Indonesia and its economic peers economic peers 600 400 550 China* 300 Korea 500 Vietnam* 200 Philippines 450 Thailand Malaysia Malaysia Thailand 400 Indonesia 100 Vietnam Korea 350 Philippines China 0 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 300 2000 2003 2006 2009 2012 2015 2018 Indonesia 66 World Bank 2020a. Almost a third of girls entering pregnancy are undernourished and have micronutrient deficiencies. Gender differences in preferences are reflected in different patterns of 67 expenditure and consumption within the household, with women more 65 World Bank 2021a. strongly favoring investments in children’s human capital. Pathways Towards Economic Security Indonesia Poverty Assessment 19 Poverty and Inequality Trends over time68 but remained low at 21.8 years in 2017.69 from households in the bottom wealth quintile were Younger age at marriage is associated with lower intra- significantly less likely to have received basic vaccinations household bargaining power and higher risk of domestic (57 percent) than the richest (69 percent), and almost violence.70 Concerningly, attitudes toward wife beating twice as likely to receive no vaccines at all (8.4 versus 4.7 showed little change over time, with nearly one-third of percent) (Figure 2.27). They were also much less likely women—compared to 17 percent of married men71— to be fed adequately diverse diets; only 44 percent of agreeing that a husband was justified in beating his wife children were fed a minimally diverse diet, compared to in at least one of five specified circumstances (Figure 75 percent of the richest children.76 2.25), with less wealthy women more likely to agree.72 Given widespread violence against women and girls Wealth gaps in access to maternal health care services in Indonesia, this not a theoretical question, and has contribute to inequities in child outcomes. Richer physical and mental implications.73 While violence is women were significantly more likely to receive higher one extreme, decision power has many facets. Over quality prenatal care, such as a complete battery of one in five women did not exercise control over her essential diagnostic tests (Annex Figure A6).77 The total own earnings or was not involved in decisions related fertility rate (TFR) declined significantly among richer to household expenditures, and about one-third did women, but not among those from poorer households not participate in decisions regarding their own health (Figure 2.28). By 2017, the gap in TFR between women care, major household purchases, and visits to family or in the richest and poorest wealth quintiles rose to 0.6. A relatives. Women were also significantly less likely to own higher TFR among poorer women, compounded with property and land than men, and the gap was larger poorer quality of maternal care, contributes to poorer among the least wealthy.74 health outcomes during reproductive years as more pregnancies lead to more pregnancy-related illness and Children in poor households and lagging regions remained limit resources for routine care during each pregnancy. more likely to have an unhealthy start in life, a disadvantage In turn, children born to less healthy mothers have compounded by lower access to opportunities at all stages inferior health outcomes immediately after birth and of life. during infancy. Inequities in access to child health services contribute Gaps in education between the rich and poor were to persistent wealth gaps in health outcomes. Gaps slow to close. Preschool access remained low across the between the poor and the rich in childhood mortality population. Preschool enrollments improved somewhat rates reduced significantly since 2007 but remained since 2002 but remained quite low in 2022 across the twice as high among children in the bottom wealth population’ 33 and 40 percent of 10-year-old children quintile as those in the richest (Figure 2.26).75 Children from extreme poor and poor households are enrolled in preschool compared to 46 percent of the non-poor. With 68 Badan Pusat Statistik, National Population and Family Planning Board, and access to basic education nearly universal since 2015, Ministry of Health 2018. Among ever-married women aged 25-49, the median age of marriage increased from 17.7 years in 1991 to 21.8 years in net primary education enrollment rates among the poor 2017. 97% of employed women participate in decisions about the use of their earnings, 73% make decisions on their own, and 24% make decisions had already caught up with the non-poor by the turn jointly with their husbands. 69 Badan Pusat Statistik, National Population and Family Planning Board, and of the century. Catch-up was evident at the secondary Ministry of Health 2018. school level as well. However, while the rates reflected 70 World Bank 2020a. 71 The latest data on men are available only in the 2012 round of the survey. increasing access among the poor, they also showed 72 Badan Pusat Statistik, National Population and Family Planning Board, and Ministry of Health 2018. plateauing rates among the non-poor since 2015, 73 World Bank 2020a. 76 Estimates were for children aged 6-23 months living with their mother. 74 Among women, use of cellphones, computers and the internet rose steadily 77 Less than half of women in the poorest wealth quintile delivered in a health since 2018, and differences between men and women were relatively small facility compared to over 90 percent of the richest; the latter were over six times among the poor and non-poor, at most a few percentage points (authors’ as likely to access ante-natal care (ANC) through an obstetrician than the former calculations using SUSENAS 2018-2021). and the gap grew over time. The wealth gap in the share of share of births that 75 Badan Pusat Statistik, National Population and Family Planning Board, and received a post-natal check within two days of birth grew over time. Nearly half Ministry of Health 2008; Badan Pusat Statistik, National Population and Family of women in the poorest wealth quintile had a serious problem accessing the Planning Board, and Ministry of Health 2018. care they needed, compared to under a third of women in the richest. 20 Pathways Towards Economic Security Indonesia Poverty Assessment Poverty and Inequality Trends indicating system-wide stagnation. This is a concerning Gaps in health and education outcomes by wealth development in the context of increasing public status undermines Indonesia’s workforce productivity spending on education.78 Average years of schooling while exacerbating inequalities. Worse health and completed by heads extremely poor households rose education outcomes among the poor create barriers to somewhat faster than among the non-poor (1.9 vs. 0.1 entry into decent livelihoods and/or jobs that help secure years of schooling between 2003 and 2022). In 2022, a path of out of poverty. Worse, lags in these outcomes most household heads in the bottom quintile still only means lack of access to the same opportunities as the had completed primary education or less (Figure 2.29). better off, meaning that poor households, even with Only a tiny minority had completed tertiary education improvements, may never be able to catch up, increasing (2.9 percent compared to 25 percent in the top long-term income and consumption inequality. quintile). The overall slower progress among adults highlights the significant time for improvements Poor and extreme poor households benefited from In youth school enrollment to reflect in average increased access to basic infrastructure services, population educational attainment. but still lagged the non-poor by significant margins. FIGURE 2.27: Childhood mortality rates, by wealth FIGURE 2.28: Vaccination rates and dietary diversity among quintile children, by wealth quintile 100 All basic vaccines 80 80 60 60 40 40 20 20 0 0 Lowest Highest Lowest Highest Min. dietary No quintile quintile quintile quintile diversity vaccine Infant mortality Under 5 mortality 2007 2017 Lowest 2007 Lowest 2017 Highest 2017 FIGURE 2.29: Total fertility rates among women, by FIGURE 2.30: Educational attainment among the top 20 (top) and wealth quintile bottom 20 percent (bottom), household heads 3 100 50 2 0 100 80 Percent 60 1 40 20 0 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2017 2018 2019 2020 2021 2022 0 Lowest quintile Highest quintile Primary or Below Junior Secondary TFR 2007 TFR 2017 Senior Secondary Tertiary 78 World Bank 2020d. The share of education spending in total government spending increased from 11.3 percent in 2001 and has fluctuated between a sizable 17-20 percent since 2009. The share of public health spending in total government spending increased from 2.9 percent in 2001 to over 8 percent by 2017. Pathways Towards Economic Security Indonesia Poverty Assessment 21 Poverty and Inequality Trends Among the poor, access to electricity, gas, clean water, While sectors of employment among the poor diversified, and sanitation has been slowly catching up to that the poor remained more likely to be engaged in less among the non-poor, but gaps remain large, except in productive, insecure livelihoods. the case of electrification (Annex Figure A7). For example, in 2021, 45 of the extreme poor still lacked access to gas Labor force participation and employment rates and 50 percent lacked adequate sanitation services. stayed large constant over the last two decades.80 Labor force participation stayed relatively constant within Indonesia’s lagging regions exhibit severe deficits in a narrow range of 65 to 69 percent (Figure 2.32). With aggregate human capital. Subnational disaggregation job creation around 2 million new jobs per year between of the HCI presents a grim picture (Figure 2.30).79 Some 2014 and 2019, the labor market managed to absorb districts in the country, concentrated in the Nusa new labor entrants, and the employment rate remained Tenggara and Maluku-Papua island-regions (Box 2.4), near constant at around 94 percent (94.1 percent in 2015 have human capital levels comparable to Chad, Niger, and 94.7 percent in 2019). and Sierra Leone, while others regions are almost at par with countries like Vietnam and China. Differences in Under-employment remains high, squeezing income learning outcomes, as measured by harmonized test at the intensive margin. Under-employment (working scores, account for the largest share of variation in HCI less than 35 hours per week) continues to be high at scores. In addition to generally lower child survival rates around 40 percent for workers with no more than primary (Figure 2.31), early nutrition deficits are particularly education (Annex Figure A8). They also work fewer hours acute in eastern Indonesia: in provinces such as Nusa per week, dropping by 1 hour from 22 hours in 2014 to Tenggara Timur and Papua in 2022, children were 21 hours in 2019. At the same time, real wages increased over twice as likely to be stunted (nearly 35 percent from 2014 to 2019 by about 24 percent for workers of of children under age 5) compared to DKI Jakarta and no more than primary education and also with junior Bali (14.8 and 8 percent respectively). Poor maternal secondary degrees (Figure 2.33). health contributes to these deficits. Maternal mortality is significantly higher in the lagging regions than elsewhere (Annex Figure A9). FIGURE 2.31: Sub-national human capital index relative FIGURE 2.32: Childhood mortality rates in 2012 and 2020, to GPD per capita by region 60 50 40 2020 30 20 10 0 0 50 100 150 2012 Maluku-Papua Nusa Tenggara Rest of Indonesia Note: Mortality for children under 5 years. Y-axis = mortality rate in 2020 (Indonesia Population Census); x-axis = mortality rate in 2012 (MOH 2013). 80 Due to the lack of comprehensive labor data in the consumption household survey (Susenas) and no household information in the labor force survey (Sakernas), both datasets cannot be readily linked. Therefore, analyses on the labor force survey are conducted using educational attainment as a proxy for wealth status. Future work could build on the results presented here by 79 World Bank 2022c. imputing consumption and hence poverty status into the Sakernas. 22 Pathways Towards Economic Security Indonesia Poverty Assessment Poverty and Inequality Trends Box 2.4: Access to opportunities in Indonesia’s lagging regions Eastern Indonesia’s lagging regions have the highest poverty rates in the country and are home to an increasing share of the country’s extreme poor. The economies of these regions are different from the rest of the country. Per capita GDP of the Maluku-Papua region (RGDP) has remained close to that elsewhere in Indonesia (excluding Nusa Tenggara), in sharp contrast to the region’s poverty rate, which remains the highest (Figure Box 2.4.1). The Papua region has a higher share of output coming from the non-labor-intensive mining and quarrying sectors, included under the “other industries’ category” (Annex Figure A10). Contribution to output from low-value added (VA) services picked up only in recent years. Overall, the RGDP and labor employment shares by sector indicate a significantly lower level of diversification than elsewhere (Annex Figure A11). Half of workers are employed in agriculture. The Nusa Tenggara region, on the other hand, has the lowest RGDP per capita among Indonesia’s island regions, and a much higher dependence on agriculture and low-VA services compared to the rest of the country. While education attainment in Nusa Tenggara lags other parts of the country, the region is rapidly catching up. Over the last decade, educational attainment among adults increased in both urban and rural Nusa Tenggara, but remained lower than other parts of the country (Figure Box 2.4.2). However, completion rates among young adults have caught up. The share of the population aged 19-25 years in urban Nusa Tenggara that had completed senior secondary education rose from 27.7 percent in 2002 to 44.4 percent in 2022, close to that in the rest of urban Indonesia (44.4 percent).80 Likewise, in rural Nusa Tenggara, the share rose from 14.7 percent to 38.6 percent over the same period, almost closing the gap with other parts of rural Indonesia (40.9 percent). FIGURE BOX 2.4.1: Regional GDP per capita FIGURE BOX 2.4.2: Educational attainment, by urban and rural regions 50 11 40 9 30 7 20 5 10 3 0 2010 2012 2014 2016 2018 2020 2022 2013 2014 2015 2016 2017 2018 2019 Maluku-Papua Rural Maluku-Papua Urban Nusa Tenggara Rural Nusa Tenggara Urban Maluku-Papua Nusa Tenggara Rest of Indonesia Rest of Indonesia Rural Rest of Indonesia Urban FIGURE BOX 2.4.3: Access to basic infrastructure FIGURE BOX 2.4.4: Access to basic infrastructure services in urban areas in 2022, by region services in rural areas in 2022, by region Electricity Electricity Cellphone Sanitation Cellphone Sanitation Gas Water Gas Water MP NT RoI MP NT RoI 81 Notes: MP = Maluku-Papua; NT = Nusa Tenggara; RI = Rest of Indonesia 81 Catchup was also evident at the tertiary level. The share of young adults in urban Nusa Tenggara that had completed tertiary education rose from 5.3 percent in 2002 to 37.1 percent in 2022, higher than the share of 33.9 percent in the rest of urban Indonesia. Similarly in rural Nusa Tenggara, the share rose from a mere 2.4 percent to 21.4 percent, similar to the rest of rural Indonesia (20.4 percent). Pathways Towards Economic Security Indonesia Poverty Assessment 23 Poverty and Inequality Trends Box 2.4: Access to opportunities in Indonesia’s lagging regions (contd) Educational attainment in the rural Maluku-Papua region remains the lowest in the country, with relatively slow progress over time. Over the last decade, the average years of schooling completed among adults increased in rural Maluku-Papua but remained lower than anywhere else in Indonesia (Figure Box 2.4.2). While education completion rates among young adults were similar to those in rural Nusa Tenggara in the early 2000s, progress was significantly slower. The share of young adults that had completed senior secondary education stood at 33.7 percent in 2022, far below that in rural Nusa Tenggara and the rest of rural Indonesia. The urban-rural gap in education completion in the Maluku-Papua region is staggeringly large. Historically, urban Maluku-Papua has had the highest educational attainment in Indonesia (Figure Box 2.4.2). The urban-rural gap is largest among Indonesia’s island regions, at nearly four years of schooling, over twice the rest of the country. This is linked to presence of skilled workers who migrate to urban locations in the Papua region, many employed in mining and the extractives industry and services. With the notable exception of rural Maluku-Papua, lagging regions have caught up to other parts of Indonesia in access to many infrastructure services. By 2022, access to electricity, water, and sanitation services in urban areas of Maluku- Papua and Nusa Tenggara was similar to that elsewhere in urban Indonesia (Figure Box 2.4.3). Access in rural Nusa Tenggara was also very similar to rural areas elsewhere. Rural Maluku-Papua, however, was an outlier as barely half of households had access to adequate water and sanitation services, compared to over two-thirds in rural areas elsewhere. Nusa Tenggara made remarkable progress since 2012, starting with very low access like that in Maluku-Papua and rapidly catching up to the rest of Indonesia. In rural Maluku-Papua, however, progress was limited. The lagging regions also stand out from the rest of Indonesia in their slow transition from “dirty” energy sources for cooking to cleaner ones. But again, progress was impressive in Nusa Tenggara, with the share of households using gas as the primary cooking fuel increasing from under 10 percent to over 40 percent over the last decade. In contrast, the share stood at a mere 1 percent of households in Maluku- Papua in 2012, rising to only 2 percent a decade later. Agriculture still holds the largest share of workers from share of the rural poor will continue to be engaged in poor households, although the sector’s share declined agriculture in the coming decades. More broadly, sector over time. Employment sources diversified for all of employment is one of the key characteristics that workers, with the share of household heads in agriculture distinguishes the economically secure from the poor and declining, especially among the extreme poor, while the insecure; the secure have a significantly higher share of shares in industry and services increased, albeit slowly employment in services (50 percent) and a much lower (Figure 2.34). Given the pace of the transition, a large share in agriculture (17 percent; Annex Table A1). FIGURE 2.33: Labor force participation, by FIGURE 2.34: Real labor income, by education education and for women from 2001 to 2021 80 2,000,000 70 1,500,000 Percent 60 1,000,000 50 500,000 40 - 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 01 03 05 07 09 11 13 13 15 17 19 21 20 20 20 20 20 20 20 20 20 20 20 20 Primary or below Junior Secondary All Primary or below Junior Secondary Senior Secondary Women Senior Secondary All Source: Authors’ calculations based on SAKERNAS and SUSENAS 24 Pathways Towards Economic Security Indonesia Poverty Assessment Poverty and Inequality Trends FIGURE 2.35: Share of household heads by sector and type of FIGURE 2.36: Formalization by education, from 2001 to 2021 employment, by poverty status, for 2003 and 2021 80 80 60 60 Percent Percent 40 40 20 20 0 re y s ed ee r 0 ke ice str tu oy oy or du rv ul pl pl w Se 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 ric In Em em al Ag su lf- Ca Se 2003 Extreme poor 2021 Extreme poor Primary or below Junior Secondary 2003 Poor 2021 Poor Senior Secondary All Source: Authors’ calculations based on SAKERNAS and SUSENAS The share of poor working as employees increased Most women, poor, and non-poor, remained outside the somewhat, but engagement in casual work also labor force. increased. In 2019, heads of extreme poor, poor, and economically insecure households tended to Indonesia’s female labor force participation is low, concentrate in self-employment or casual work, with little change over the last two decades.85 Only while the economically secure were more likely to be half of women were employed or looked for work in engaged as employees, often in formal work (Annex 2019. This is a relatively low labor force participation Table A1). The share of poor working as employees rate compared to men (83 percent) as well as in the East rose slowly from 20 to 26 percent between 2003 and Asia and Pacific region (60 percent). Improvements in 2021, suggesting a shift toward more secure forms of women’s education outcomes have not lifted constraints employment. However, this positive trend was offset or changed preferences for entering the labor force. The by a moderate increase in the share in casual work or female disadvantage in adult educational attainment unpaid labor, from 11 to 17 percent.82 was small (less than a year of schooling) and declined over time (Figure 2.36). Among younger cohorts, the The labor market still exhibits high levels of gender gap has reversed, favoring women over the last informality despite a slow but steady trend towards couple of decades. This points to existence of multiple formalization. Formalization increased from around 30 barriers women face in joining the labor force arising percent in the 2000s reaching 44 percent in 2019 (Figure from sources other than lack of formal education alone. 2.35).83 Lower socioeconomic status was associated with a higher likelihood of being in informal work. In 2019, Lack of high-quality childcare is a critical barrier to workers with no more than primary education were women’s work. Marital status and presence of young much more likely to be in informal work, with only 23 children in the household are significant predictors of percent formally employed. The sizable gap between women’s presence in the workforce.86 Very few men the rich and poor in educational attainment described and women believe that a child suffers when a mother earlier likely contributes to these employment outcomes works for pay outside the home (Figure 2.37). Still, young by limiting the poor to move out of insecure, low- men feel that women can work outside the home until productivity employment.84 they become pregnant.87 Young women likewise find it difficult to reconcile family life with careers, describing 82 Trends among the poor were qualitatively similar to those among the extreme poor. secure, high-quality childcare as critical for working 83 Informality is measured using the BPS official definition using type and sector of outside the home once they have children. Others employment. Other definitions, e.g., ILO, can yield significantly higher estimates, e.g., 80 percent compared to 50 percent in 2019. 84 Indeed, while average years of schooling completed by heads of households that 85 World Bank 2020a; Cameron, Suarez, and Rowell 2018. were (extreme) poor, or economically insecure, were comparable to each other (annex Table A1), they were much lower than those among the economically 86 World Bank 2020a. secure, with the gap standing large at over three years of schooling. 87 Ririn Salwa Purnamasari et al. 2020. Pathways Towards Economic Security Indonesia Poverty Assessment 25 Poverty and Inequality Trends FIGURE 2.37: Gender gap in years of education completed FIGURE 2.38: Attitudes around women’s work, among among adults and 15-19-year-olds women and men 1.0 When a mother works for pay, the children su er 0.5 Being a housewife is just as ful lling as working for pay 0.0 Problem if a woman earns more than her husband 0.5 If jobs are scarce, men should have more right to a job 1.0 A university education is more 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2021 2022 important for a boy than a girl Adults EP Adults poor 0 20 40 60 80 100 15-19 years EP 15-19 years poor Women Men Notes: A negative value reflects a female disadvantage. EP = extreme poor. Adults = 15-64 Source: Haerpfer et al. 2020 years. The gender gap among the poor and extreme poor was not significantly different from that among the non-poor (not shown) prefer to leave children only in care of parents or nannies than half a day, not enough to facilitate employment and would stop working rather than rely on daycare of mothers outside of low-productivity, unpaid family they do not trust. This indicates that cultural and social and casual work.92 Occupational segregation has also norms around gender-based roles and responsibilities been linked to norms around professions designated in the household are prevalent. The limited progress in acceptable for men and women, which influences young women’s agency noted earlier also reflects this: neglect people’s educational choices.93 University students, for of children is the most reported justification for wife- example, feel pressured to conform to gender norms beating.88 However, there are signs of change. The in choosing fields of study; families and peers may propensity for women to participate in the labor force not support young men studying nursing or a woman increased across cohorts, and was notably higher for studying a STEM subject. those born in the 1980s and 1990s than birth cohorts in earlier decades.89 That the change mainly occurred in Married women in poor households had worse urban areas suggests increasing acceptance of women employment outcomes than those in non-poor into non-agricultural employment, which is a good households, suggesting that being poor may intensify sign. Expanding high-quality childcare is likely to further constraints women’s work productivity.94 The vast facilitate this transition: an additional preschool per majority of poor households in Indonesia are composed 1,000 eligible children increased maternal work by 4.8 of at least one married couple living with children (Figure percentage points.90 2.38).95 In 2019, these households were overrepresented 92 Halim, Johnson, and Perova 2022. Women workers concentrate in low-productivity 93 Ririn Salwa Purnamasari et al. 2020. sectors. Outside of agriculture, women remain 94 In contrast to married women living in married-couple households, women in poor households with no or only one married adult exhibited employment predominantly in informal jobs and low-productivity outcomes that connected in complex and varied ways with presence of dependents. For one, unmarried women in these households are much more service sector jobs such as retail, restaurants, and hotels. likely to work than married women when either no children are present or children alone are present (annex Table A2). This is likely a consequence of 91 Men dominate the construction, electricity, gas, water necessity of working as the vast majority of these households are female- headed, and as non-traditional households, unlikely to have access to informal supply, transport, and finance and business services and formal networks of support available to households with married couples. This is especially concerning because even when working-aged men are present, sectors. This segregation has been linked to inadequate they are much less likely to be working than other men in poor households (also see annex Table A3). Second, the forms of employment women in these childcare supply: preschools in Indonesia operate for less households were engaged in varied with the type of dependent present. With no dependents or with children only, just under half were in self-employment. This share fell to one-third in the presence of seniors. In contrast, the share in 88 Badan Pusat Statistik, National Population and Family Planning Board, and casual or unpaid work was significantly higher in the presence of seniors, rising Ministry of Health 2018. from one-quarter to one-third of working women. 89 Cameron, Suarez, and Rowell 2018. 95 Examining the relationship between the demographic composition of households and poverty rates can shed light on how gender differences in 90 Halim, Johnson, and Perova 2022. access to opportunities are associated with poverty. See for example Munoz 91 World Bank 2020a. Boudet et al. 2018. 26 Pathways Towards Economic Security Indonesia Poverty Assessment Poverty and Inequality Trends among the poor. Married women in poor households Deficits in women’s wellbeing, often larger were only slightly less likely to work than married women among poor women, can constrain women’s labor in non-poor households, and presence of children productivity both within and outside the home. was associated with a somewhat lower likelihood of Limited declines in fertility of women from low working (Annex Table A1).96 Presence of seniors seemed socioeconomic status backgrounds and stagnating to mitigate this; the share of women working was the health outcomes constrain women’s access to economic same as, or even slightly higher, in households where opportunities. Quality maternal care can ease health seniors were present than among households where challenges women face during childbearing years. only children were present. However, married women in Lowered burden of illness during the reproductive poor households were much more likely to be in low- years, bearing fewer children, and reduced childcare productivity and unpaid work than those in non-poor responsibilities at home can give women more time to households. A far larger share was in casual labor and invest in their own education and skills, and to engage unpaid work, where just under half of married women in in income-generating activities within and outside the poor households concentrated, compared to 25 percent home. Given health outcomes below peer countries, and of non-poor married women.97 Notably, while shares usually worse for poorer women, women’s productivity— across types of employment did not vary significantly and especially of poorer women—remains constrained. with presence of dependents, presence of children in poor households – but not seniors – was associated with Data on use of men and women’s time within and a higher likelihood of being in low-VA services and a outside the home, as well as associated norms lower likelihood of being in agriculture.98 and preferences, would help inform policies to lift constraints on women working. Women are often FIGURE 2.39: Poor and non-poor households in 2019, by demographic classification burdened with unpaid care work, limiting their time available for income generating activities.99 Women who Non-poor work may have to carry a double burden of working in a 2019 paid activity while continuing to provide care at home. Poor Deficits in basic infrastructure services such as water, Non-poor sanitation systems, and electricity can make the time needed to perform household work unnecessarily high. 2022 Poor Time-use surveys to measure the amount time people 0 20 40 60 80 100 spend doing various activities such as paid work and Percent household, family, and personal care can inform policies At least one married couple with children At least one married couple with seniors and children for freeing women’s time for economically productive At least one married couple-no dependence activities. Further, to understand time-use patterns, No or one married adult with seniors No or one married adult with children data are also needed on men and women’s preferences No or one married adult with seniors and children and cultural norms about time use, and the perceived and actual barriers to entry into fields of study and occupations of choice. 96 Among households with married couples, the likelihood of men and women working was in line with what could be expected given male and female labor force participation rates in the general population, regardless of the type of dependents present. 97 Wage employment, the largest employer of non-poor married women (around 40 percent), held only under a quarter of women from poor households. The shares employed in self-employment were similar across poor and non-poor households. 98 Among non-poor married women, the majority of whom were employed in low-VA services (nearly 60 percent), the differences across households with different types of dependents were muted in comparison. 99 World Bank 2020a. Pathways Towards Economic Security Indonesia Poverty Assessment 27 CHAPTER 3 DRIVERS FROM 2014 TO 2019 Photo: © Achmad/World Bank 28 Pathways Towards Economic Security Indonesia Poverty Assessment 3. DRIVERS FROM 2014 TO 2019 Poverty reduction from social and economic progress and benefit less Economic growth reached the poor more than the from growth. In addition, about one-fifth of the extreme extreme poor. Both urban and rural places offer pathways poor was not economically active. This emphasizes the out of poverty. importance of social assistance for these households to complement insufficient labor incomes. C onsumption growth across the entire population, rather than redistribution, characterized poverty reduction. In the period from 2006 to 2013, high Rural and urban areas’ contributions to poverty reduction reflected their share of poor, suggesting consumption growth translated into large poverty similar strengths of poverty reduction.103 Rural poverty reduction while rising inequality (redistribution) harmed reduction contributed two-thirds to overall poverty poor households and weakened poverty gains (Figure reduction from 2006 to 2013, and then dropped to just 3.1).100 A commodity boom drove economic growth in above half (55 percent) from 2014 to 2019 (Figure 3.4). this period, which led to higher household incomes, but Urban poverty reduction explained almost fully the also to capital gains accruing to wealthier households, remainder, since urbanization—or the population-shift increasing inequality. Since 2014, economic growth effect—contributed below 2 percent across periods. As slowed but poverty reduction managed to maintain rural poverty rates converged towards urban poverty its momentum. Solid macro-economic fundamentals rates, the share of rural poor among the poor declined; in combined with large job creation allowing a larger share 2006, about two-thirds of the poor were rural, dropping of economic growth to reach households and reduce to about one-half in 2014. poverty.101 Hence, consumption growth remained the dominant factor. Gains from labor income helped With urbanization continuing, removing constraints especially poorer households, such that redistribution for urban poverty reduction will become more critical. no longer slowed poverty reduction but accelerated it. Urbanization was mainly driven by re-classification of rural areas as urban due to urban sprawl, as well as the Economic growth reached the poor, but increasingly natural growth of the population in urban areas. Together, less so for the extreme poor. From 2014 to 2019, each these factors explain about 80 percent of urbanization percentage point of GDP per-capita growth reduced in Indonesia.104 Official re-classification reflects an actual poverty by 3 percent and extreme poverty by 5 percent transformation and not just a bureaucratic relabeling of (Figure 3.2). The larger reduction for extreme poverty settlements.105 Urbanization will remain an important is not surprising, given its significantly lower level.102 force, but is currently not delivering an urban premium In comparison to other countries, Indonesia’s poverty for poverty reduction, in part because most of the urban elasticity is slightly below its expected level, given its poor live in less productive urban peripheries rather than GDP per capita (Figure 3.3). In absolute terms though, productive and prosperous urban cores.106 Compared the semi-elasticities reveal that extreme poverty only to other countries, Indonesia’s poverty levels are higher dropped 0.2 percentage points while poverty dropped than expected given its level of urbanization (Figure 3.5). by 0.8 percentage points for each percentage point Thus, constraints to poverty reduction in urban places of GDP per-capita growth. Thus, economic growth will need to be addressed. started to have a diminishing role in extreme poverty reduction. With relatively low levels of extreme poverty, 103 The use of an absolute poverty line, rather than the weakly relative national poverty lines, is driving these results. With the convergence of poverty rates in the remaining extreme poor are more marginalized urban and rural areas, rural poverty is progressing faster than urban pove. 104 Roberts, Gil Sander, and Tiwari 2019. 105 Official re-classification is triggered when a settlement records a higher score 100 From 2002 to 2010, consumption growth was strongly biased towards richer on a composite index. The composite index measures progress towards higher households, but became equally distributed by 2011. population density, a transition away from agriculture, and more infrastructure, 101 World Bank 2020c. especially related to typically urban facilities. 102 Cuaresma, Klasen, and Wacker 2016. 106 Roberts, Gil Sander, and Tiwari 2019. Pathways Towards Economic Security Indonesia Poverty Assessment 29 Drivers from 2014 to 2019 FIGURE 3.1: Annualized contributions of growth and FIGURE 3.2: Elasticity (left) and semi-elasticity (right) of redistribution to poverty reduction poverty to per-capita growth 6.0 6.0 2 1 5.0 5.0 0 4.0 4.0 -1 3.0 3.0 -2 -3 2.0 2.0 -4 1.0 1.0 -5 0.0 0.0 1.9 3.2 1.9 3.2 1.9 3.2 9 9 5 5 3 01 3 01 00 00 01 01 -2 2002 -2005 2006 -2013 2014 -2019 -2 -2 -2 -2 -2 14 14 02 02 06 06 20 20 20 20 20 20 Growth Redistribution Total $1.90 $3.20 $1.90 $3.20 Note: Datt-Ravallion decomposition at US$ 1.90 PPP (2011) and US$ 3.20 PPP (2011) Notes: Elasticities reflect the relative change in the poverty rate (in percent) for each 1 percentage point of GDP per capita growth. Semi-elasticities reflect the absolute change in the poverty rate (in percentage points) Years 2020 and 2021 are omitted due to negative growth. FIGURE 3.3: Elasticity of poverty to per capita growth FIGURE 3.4: Ravallion-Huppi decomposition of poverty reduction, by urban/rural 12 1.0 10 Malaysia 0.0 8 -1.0 Korea 6 Thailand -2.0 China 4 Indonesia -3.0 2 Philippines Vietnam -4.0 0 2006-2013 2014-2019 1000 10000 100000 Rural Urban Log GDP per capita, 2019 Population-shift e ect Interaction e ect Notes: Elasticities estimated at $3.20-a-day (2011 PPP). Estimates for China, Indonesia and Thailand are between 2014-19; Korea: 2010-16; Malaysia: 2008-15; Philippines: 2012-2018; and Vietnam: 2013-18. Periods chosen on the basis of latest year for which poverty rates were available. Linear trend line Several factors contributed to poverty reduction the impact of household size on poverty), which is only from 2014 to 2019: demographics, employment, and significant in rural areas. Employment contributed to education as well as spatial factors, prices, and fiscal poverty reduction, though not due to a change in the policies. Using a simple framework (Box 3.1), we can level of employment, but because being employed analyze poverty reduction factors and quantify effects created higher returns for escaping poverty, especially from demography, employment, and education (Figure in rural areas. An increase in education contributed to 3.6).107 The observed, albeit small, reduction in fertility poverty reduction, particularly in rural areas. However, the contributed to poverty reduction from 2014 to 2019 returns from education were lower in 2019 compared to by reducing household sizes (Figure 3.6). The effect can 2014, such that the change in the pay-off slowed poverty be split into endowment (the drop in household size reduction, especially in urban areas. Thus, holding at least from 2014 to 2019) as well as the increased “return” of a secondary degree is becoming less of a distinguishing household size for poverty reduction (the change in feature between poor and non-poor households. Datt and Ravallion 1992. 107 30 Pathways Towards Economic Security Indonesia Poverty Assessment Drivers from 2014 to 2019 FIGURE 3.5: Poverty vs. urbanization rate by year FIGURE 3.6: Contributions of endowment and return to poverty reduction, by urban/rural 100 0.06 0.04 Poverty rate US$ 3.20 80 0.02 60 0 40 Rural Urban Rural Urban -0.02 Endowment Return 20 -0.04 -0.06 0 0 20 40 60 80 -0.08 Urbanization rate China Indonesia Vietnam Household size Employment Education Source: Authors’ compilation based on World Development Indicators Source: Authors’ calculation based on SUSENAS Notes: Oaxaca-Blinder decomposition of changes in poverty (by percentage points) from 2014 to 2019, distinguishing between changes in endowment (co-variates) and returns (coefficients). Box 3.1: Analytical framework for poverty reduction in Indonesia We use a simple framework to investigate poverty reduction drivers. Several factors critically influence real-per-capita household consumption, which—in conjunction with a poverty line— determines poverty status of a household. Demographics affect household size and composition; larger households share resources among a larger number of individuals. At the same time, additional working age adults can increase household income. Labor income is one key source of household income. Individuals with higher education often earn more, while less skilled workers might suffer from longer unemployment spells. Spatial factors—for example, if a household is residing in an urban or rural areas—affect the availability and type of employment opportunities. Real per-capita household consumption is also determined by prices; higher prices, if not offset by higher wages, erode purchasing power. Finally, taxes reduce household consumption, while subsidies and direct transfers (arguably a form of income) increase household budgets. FIGURE BOX 2.3.1: Framework to analyze drivers of poverty reduction in Indonesia Demography Prices Labor income / Real per-capita Spatial factors employment household consumption Taxes and public Education Poverty status spending Source: Authors’ visualization Other determinants are excluded for conceptional and data availability reasons. A multitude of additional factors influence poverty status, such as land ownership and more generally access to capital. However, gaps in data availability undermine the ability to include these factors into the framework. Public services including sanitation, health, and education, are also important factors but we exclude them from the framework because they are also consequences of escaping poverty. Pathways Towards Economic Security Indonesia Poverty Assessment 31 Drivers from 2014 to 2019 Demography and reach economic security. In developed countries, Fertility rates declined, reducing household sizes and this is achieved through higher labor productivity, driven contributing to poverty reduction. by better human capital and use of technology. This emphasizes the importance of making such investments Lower fertility contributed to poverty reduction while Indonesia’s economy still benefits from the through smaller household sizes, while the demographic dividend. dependency penalty dropped in the context of a shrinking demographic dividend. Even though fertility In addition, a gender poverty gap remains, with declined only slowly, it contributed to poverty reduction women—especially of child-bearing age—more likely through smaller households. The increased returns from to be poor, driven by the dependency penalty and for smaller rural households help them catch up with anticipatory fertility. A small gender gap has emerged urban households through avoiding a dependency particularly for rural women in their prime reproductive penalty, as improved family planning increases women’s years and old age (Figure 3.7 and Figure 3.8).110 The prime choices for work.108 The population share of households reproductive years are formative years to join the labor with fewer dependents and, to a lesser extent, more market and get trained with new remunerable skills to earners have increased, helping spur poverty reduction apply throughout life. Even after controlling for education among those groups (Figure 3.9). For rural households, and location, a small but persistent gender poverty gap closing this gap has contributed to poverty reduction by remained significant at around 2 percentage point for increasing the returns of smaller household sizes in rural the cohorts aged 19 to 24 and over 60 years, and a slightly areas (Figure 3.6). smaller gap persisted for women between those ages, but the gender gap is not significant for younger women However, the “demographic dividend” will soon be (Figure 3.10). The gender poverty gap was highest for exhausted. Indonesia’s share of working-age population married women with children, as well as when they were is expected to start falling between 2025 and 2030.109 older, often outliving men. A large part of the gap was At the same time, the number of old-age dependents driven by the dependency ratio, as the gap collapsed will increase, reversing gains in the dependency penalty for women aged 25 to 29 and above 60 years when through both demographic and economic factors. Thus, controlling for the dependency ratio. Interestingly, the workers will need to earn more to stay out of poverty gap for women aged 19 to 24 years remained, possibly FIGURE 3.7: Gender and age-cohort poverty rates for 2014 FIGURE 3.8: GDP growth (LHS) and GDP -per-capita (RHS) from 1990 to 2021 50 30 25 40 20 30 15 20 10 10 5 0 0 0-18 yo 19 -24 yo 25 -29 yo 30 -59 yo 60+ yo 0-18 yo 19 -24 yo 25-29 yo 30-59 yo 60+ yo Female rural Female urban Female rural Female urban Male rural Male urban Male rural Male urban Source: Authors’ calculations based on SUSENAS Poverty is estimated at the household level, because consumption is usually not 110 reported at the individual-level due to measurement constraints (methodologically 108 The dependency penalty comprises demographic factors (households with as well as conceptually). This analysis obtains gender-specific poverty estimates more dependents) and economic factors (households with fewer earners). by conservatively splitting consumption equally across household members. 109 Wihardja and Cunningham 2021. Equivalence scales and economies of scale affect estimates. 32 Pathways Towards Economic Security Indonesia Poverty Assessment Drivers from 2014 to 2019 FIGURE 3.9: Annualized changes in poverty rate and population FIGURE 3.10: Female effect on poverty status, with and without share from 2014 to 2019 for demographic and economic groups controlling for dependency ratio, for 2019 of households Economic 4 0.03 Demographic 2E 2 population share (in percent) 0.02 Annualized change in 1D 3E 0 2D 1E 0.01 3D -2 4D 0 5D -4 6+D -0.01 -6 -15 -10 -5 0 0-18 yo 19 -24 yo 25 -29 yo 30 -59 yo 60+ yo Annualized change in poverty (in percent) Excluding dependency Including dependency Notes: 1E to 3E indicate the number of earners in the household; 1D to 5D indicate the Note: Regression of individual poverty status on female dummy plus controls for highest number of dependent; 6+d corresponds to at least 6 dependents, using SUSENAS education in household, region and urban-rural, using SUSENAS. driven by anticipatory fertility: young women who have poverty reduction in this period was not explained by a married but not yet had children were not entering the change in endowment of employment. Nevertheless, workforce as they expected to have children soon.111 employment by itself contributed to poverty reduction112 despite little changes in employment 2014 Labor incomes to 2019. Bringing more women into the labor force can Employment drove poverty reduction, but women were not accelerate poverty reduction, while increasing women’s able to fully participate. While the workforce was becoming empowerment and child health.113 more educated, informal jobs in low-VA services limit FIGURE 3.11: Annualized real wages by sector workers’ earnings to reach economic security. An inward- 5.0 looking economy with limited exposure to international competition undermines the creation of high-productivity 4.0 jobs, exacerbated by challenges in skills matching. Urban centers were limited in producing positive agglomeration 3.0 Percent forces and making them available in rural areas, limiting 2.0 rural diversification, while subsistence farming was often insufficient to escape poverty. 1.0 0.0 Wage gains explain the increased returns of Agriculture Industry Services Total employment for poverty reduction, while increased 2002-2013 2014-2019 labor force participation—especially for women—can Source: Authors’ calculation based on SAKERNAS. accelerate poverty reduction. Real wages—a proxy for productivity—grew more strongly in the period However, unmet childcare needs in Indonesia from 2014 to 2019 compared to the earlier years of this prevented many women from working outside their century (Figure 3.11), reflecting the increased benefit of home. Care responsibilities posed a key barrier to female growth on labor incomes and, hence, poverty reduction. labor force participation.114 The absence of affordable Growth from 2014 to 2019 explains the increasing returns and high-quality care, especially for children, limited of employment for poverty reduction. In contrast, labor opportunities for women to participate in the labor force participation as well as employment stayed largely market.115 When this constraint was lifted, women were constant over the last two decades. It is not surprising that 112 Feriyanto, Aiyubbi, and Nurdany 2020. 113 Schaner and Das 2016; Majlesi 2016. 114 R. Purnamasari, Hambali, and Halim Forthcoming. Cameron, Suarez, and Rowell 2018. 111 115 Halim, Johnson, and Perova forthcoming. Pathways Towards Economic Security Indonesia Poverty Assessment 33 Drivers from 2014 to 2019 FIGURE 3.12: Wage premium for gender, formalization, location, FIGURE 3.13: Contributions of endowment and return to wages, education, and sector by gender, formalization, and location 120 10 100 80 0 Percent 60 40 -10 20 Percent 0 -20 High-VA services Male Formal Urban Junior Sec. Senior Src. Tertiary Manufacturing Industry Low-VA services -30 -40 Male vs. Formal vs. Urban vs. female informal rural Education Sector Endowment Return Note: OLS regression on log-transformed wage, including age, age-square and province Note: Oaxaca-Blinder decomposition based on Sakernas 2019, including age, education, dummies. Bars show exponentiated coefficients to reflect relative difference in wage. sector, type of work and location. Bars show the explained difference of observed wages. likelier to work.116 For example, opportunities in the worker characteristics did not explain the large gender digital economy provided more flexibility to align work wage gap. Instead, the observed difference in wages with care responsibilities and women sometimes use is completely explained by differences in returns to this as a first step to enter the labor market.117 endowments. Thus, women earned smaller premiums, for example on their education.119 This suggests that Women, however, faced additional barriers in the labor women might be disadvantaged in their quality of market, including cultural norms, as well as sectoral education, as well by employer discrimination. This segregation. Cultural norms were biased against women makes the labor market not only unfair but also less seeking employment, especially when in child-bearing attractive for women. age, but there are indications that norms are softening.118 An increase in women’s labor force participation, Informality limited worker’s ability to use their skills, however, was masked by sectoral segregation. Women and exposed them to additional risks. Differences in worked more often in agriculture, whose employment endowments (level of education, for example) explain is declining, putting downward pressures on female 24 percent of higher wages for formal work. Higher labor force participation. The agricultural sector also returns to endowments explain another 28 percent offered less opportunities to use education and skills, as of differences (Figure 3.12). Thus, formal workers were well as transitioning into higher productivity work. This able to put their endowments to better use, increasing counteracts the aggregate increasing rate of women their productivity. In addition, informal workers did not entering the labor force outside agriculture. benefit from formal regulations for workers, including access to subsidized contributory insurance schemes Women also earned significantly less than men, even such as for unemployment. Lack of unemployment after controlling for differences in individual and work and health insurance can further lower productivity. characteristics. On average, women earned 74 percent Unemployment insurance allows workers to be more of the wage of men in 2019. However, the difference selective in accepting job offers, contributing to better was not driven by lower education or other individual matching and, hence, productivity gains.120 Health characteristics. Instead, an average worker in Indonesia insurance makes workers miss fewer workdays, similarly earned 54 percent more just by being male (Figure 3.13). increasing their overall productivity.121 However, pushing Female and male workers had similar endowments, so towards formalization and its enforcement can also 119 It is unlikely that the full wage gap can be explained by differences in productivity 116 Halim, Johnson, and Perova 2022. due to unobserved characteristics, or better matching in the labor market. 117 World Bank 2021a. 120 Rujiwattanapong 2022. 118 Cameron, Suarez, and Rowell 2018. 121 Dizioli and Pinheiro 2016. 34 Pathways Towards Economic Security Indonesia Poverty Assessment Drivers from 2014 to 2019 destroy viable informal incomes and deter formal job exports. Due to a business-friendly climate, trade- and creation. A more nuanced approach is needed based financial openness, and a large labor surplus, Indonesia on a better understanding of the characteristics, costs, attracted investment in labor-intensive industries such and benefits of informal work in Indonesia, and whether as textiles, food, and beverages. In 1996, manufacturing workers and firms choose informality as a last resort or accounted for one-quarter of GDP and contributed to by preference. half of total exports. After the AFC, Indonesia shifted to a natural resource-based growth model during Even though employment was increasing in sectors the commodity boom. Sources of growth shifted with higher wages, most workers remained trapped towards commodity exports and services, especially in low-productivity sectors. Almost half of all workers in non-tradables. After the end of the commodity boom, 2019 were engaged informally in agriculture and low-VA the economy remained focused on often low-VA services. These sectors offered the lowest median wages, services without recovery in manufacturing, which only often insufficient to escape poverty and economic contributed 22 percent to GDP in 2019 compared to 28 insecurity (Figure 3.14). While employment in agriculture percent in 2002. was slowly declining, formal employment in the better paid sectors of manufacturing and high-VA services was The inward-looking economy missed out on growing. However, formal employment in manufacturing opportunities and integration into global value chains was growing at the same speed as informal employment due to export competition.123 Compared to Malaysia, in low-VA services. Furthermore, the combined share Thailand, and the Philippines, Indonesia had lower of formal employment in manufacturing and high-VA levels of exports (Figure 3.15 and Figure 3.16), and is services remained below 15 percent of total employment less competitive in manufacturing (Figure 3.17) with in 2019. At the same time, the low-VA service sector lower export sophistication (Figure 3.18). It also is less was the fastest growing sector in absolute numbers, integrated into global value chains.124 Its foreign direct adding almost 8.5 million (or 71 percent) of new workers investment focused on the extractive industries and between 2014 and 2019. access to local markets. Not surprisingly, Indonesia’s growth in labor productivity was low especially in the Indonesia’s premature deindustrialization explains the industrial sector, which includes manufacturing (Figure under-performance of high-productivity sectors.122 3.19). Despite low growth in labor cost per hour, its unit Before the Asian Financial Crisis (AFC) in 1997-98, labor costs were increasing significantly faster than for its Indonesia rapidly industrialized through manufacturing peers, diminishing its competitiveness (Figure 3.20). FIGURE 3.14: Annualized employment growth (2014 to 2019) by median sector wage and employment share (bubble size) in 2019 2,000,000 Manufacturing High VA services Median real wage 1,500,000 Other Other Low VA services 1,000,000 High VA services Agriculture Low VA services 500,000 Manufacturing Agriculture - -6 -4 -2 0 2 4 6 8 10 Employment share growth, percent Informal Formal Source: Authors’ calculation based on SAKERNAS. Wihardja and Cunningham 2021. 123 122 Wihardja and Cunningham 2021. World Bank 2022d. 124 Pathways Towards Economic Security Indonesia Poverty Assessment 35 Drivers from 2014 to 2019 FIGURE 3.15: Average volume and annual growth of FIGURE 3.16: Export to GDP ratio vs. change in GDP per capita exports of goods and services 15 8 8 Average annual change in exports/GDP 6 Average annual growth 6 India China 4 Volume (1980=1) 10 Thailand 4 2 Malaysia 0 5 2 -2 Indonesia -4 0 0 -6 1980 1983 1986 1989 1992 1995 1998 2001 2004 2007 2010 2013 2016 2019 -2 0 2 4 6 8 10 12 14 16 Indonesia World Malaysia Thailand Philippines Average annual change in GDP per capita Source: World Bank 2022d FIGURE 3.17: Manufacturing export competitiveness FIGURE 3.18: Export sophistication 2.0 10.0 9.9 1.5 9.8 1.0 9.7 9.6 0.5 9.5 0.0 9.4 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Indonesia Malaysia Indonesia Malaysia Thailand Vietnam Thailand Vietnam Source: World Bank 2022d FIGURE 3.19: Growth in labor productivity FIGURE 3.20: Average annual growth (2010 to 2019) of unit labor cost per output and labor cost per hour 7.0 6.00 6.0 4.00 5.0 4.0 2.00 3.0 0.02 2.0 0.00 1.0 0.0 -2.00 ia a d m ia s ne in an es ys na Ch pi ala n ail et -1.0 do ilip Th Agriculture Industry Services M Vi In Ph 2002-2009 2010-2013 2014-2019 Unit labor cost Labor costs per hour Source: BPS Source: Authors’ compilation based on Economist Intelligence Unit 36 Pathways Towards Economic Security Indonesia Poverty Assessment Drivers from 2014 to 2019 This inward focus did not create the quality and Low education, as well as inadequate levels of non- quantity of jobs needed to escape economic routine interpersonal and digital skills, further increase insecurity, and will exacerbate the risk of reversal of skills mismatch. Technical and vocational training the demographic dividend. Indonesia’s labor market (TVET) as well as better labor market information provided significantly fewer high-skilled jobs compared systems can help to reduce skills mismatch, as to other countries at similar levels of development. In addressed by the Government’s plan to revitalize fact, the share of high-VA services had dropped in the vocational education.126 longer-term from 2006 to 2014. Hence, workers needed to accept low-productivity jobs in the low-VA service With the lack of opportunities from structural sector. In addition, these jobs were often informal, transformation, low-VA services drove poverty without insurance coverage and other benefits, which reduction in urban areas, but limited returns to further reduced productivity (Figure 3.19). These jobs will education and corresponding productivity gains. not be able to mitigate the negative effects from reversal Urban workers were able to take advantage of broader of the demographic dividend expected before the end work opportunities, and in 2019 earned an average of the decade. This trend, however, can be reversed by 36 percent more than rural workers. Differences in attracting export-oriented foreign direct investment, worker characteristics explain half of this urban-to- which is associated with more innovation, and by rural wage gap. Urban workers were, on average, better integrating into global value chains, which increases educated than rural workers. The other half reflects productivity. advantages from being in an urban area and putting endowments to better use. This included better quality Inadequate and mismatching skills further education, better matching of job skills, and better contributed to the inability of workers to engage infrastructure, which increases productivity. Urban in higher-productivity jobs. Almost 80 percent of jobs often provide better social insurance. Hence, they employers faced difficulties in hiring high-skilled workers provided sustainable paths out of poverty. However, (managers, senior professionals), about two-thirds faced growth in the low-VA services sector provided the difficulties in finding mid-skilled workers (such as non- largest contribution to poverty reduction of about 1 production technicians, associate professionals, sales percentage point per year, and this also provided the representatives), and 40 percent could not find low- largest sectoral employment. skilled workers (unskilled, non-production workers).125 FIGURE 3.21: Sectoral contributions to poverty reduction FIGURE 3.22: Sectoral composition of household heads from 2006 to 2019, by poverty status and urban/rural 0 Non-poor (2019) Non-poor (2014) Urban Non-poor (2006) -20 Poor (2019) Poor (2014) Poor (2006) -40 Non-poor (2019) Non-poor (2014) Non-poor (2006) Rural -60 Poor (2019) Poor (2014) Poor (2006) Rural Urban -80 0 20 40 60 80 100 Agriculture Manufacturing Percent Low-VA services High-VA services Agriculture Manufacturing Low VA services Other Population-shift e ect High VA services Other Source: Authors’ calculation based on SUSENAS and SAKERNAS Source: Authors’ calculation based on SUSENAS and SAKERNAS Note: Ravallion-Huppi decomposition for 2014 to 2019. Visualization omits coefficient of households not working. Indonesia Enterprise Survey (2015). 125 Presidential Decree No. 68 of 2022. 126 Pathways Towards Economic Security Indonesia Poverty Assessment 37 Drivers from 2014 to 2019 Digital work is likely to play an increasingly important Especially for poor farmers, agriculture often suffers role for livelihoods, and potentially for poverty from low productivity due to crop choice, the quality reduction. Indonesia’s digital economy grew five-fold of agricultural extension services, and limited market between 2015 and 2019—a pace unmatched by any access.132 Poor farmers yielded about 1 ton less harvest other country in the region. In 2020, it stood at US$ 44 per hectare compared to non-poor farmers, hinting at billion, roughly four times as large as Malaysia’s and five potential avenues to overcome barriers specific to poor times as large as the Philippines’ and Singapore’s digital farmers. Poor farmers are often constrained to subsistence economies.127 It is expected to continue its stellar growth, and rice production even though productivity might possibly tripling in size by 2025. Almost 10 percent of be lower. A distortionary set of incentives and high informal workers in Indonesia were gig workers, whose food prices have contributed to slow diversification to jobs depended on digital platform’s intermediation. higher-value cash crops, for which in some areas the soil Indonesia’s gig workers were relatively young and better- might be more suited. Large differences exist between educated than other workers.128 They earned 6 percent poor and non-poor farmers, with the non-poor using more per hour, but also worked 10 hours more per fewer inorganic fertilizers (with long-term benefits) week.129 Almost two of three gig workers were providing and more often using irrigation and mechanization. location-based services in urban settings, explained by Agricultural extension services remained ineffective in the prevalence of ride-hailing, which requires a relatively increasing agricultural productivity, especially of poor concentrated market. Gig workers opted to do gig farmers. Market access is often difficult because of gaps work for flexibility, but about one in five gig workers in infrastructure, but also because middlemen drive a reported digital work as a buffer for income shocks, for wedge between farm gate and market prices. example from COVID-19.130 Thus, gig work is becoming an important opportunity for livelihoods. However, these Lack of land tenure further reduces livelihoods of poor opportunities are not open to all. They require digital farmers, investments into land, and access to credit. skills, good connectivity, and usually an urban setting. Indonesia has embarked on the largest land reform Thus, a growing digital divide can limit opportunities, program in the world, distributing and formalizing 12 particularly for the most disadvantaged workers.131 percent of the entire country (21.7 million hectares). Nevertheless, lack of land tenure continues to negatively The agriculture sector dominated rural poverty affect livelihoods and investments, especially in reduction, but many agricultural households remained agriculture. Poor farmers without land ownership lost poor. The agricultural sector contributed 53 percent to about 40 percent of their harvest due to sharecropping.133 rural poverty reduction, or about 2 percentage points In addition, lack of land tenure limits incentives for per year from 2014 to 2019 (Figure 3.21). The large investments in land and potentially encourages overuse benefit to poverty reduction is not surprising given of inorganic fertilizers. It also reduces access to credit as that the largest share of rural workers remained in land is often held as collateral. This constrains the ability agriculture (55 percent in 2014 and 53 percent in 2019), to invest into, for example, irrigation and mechanization, especially among the poor (64 percent in 2014 and 32 and has implications beyond agriculture. percent in 2019; Figure 3.22). Notably, many agricultural households continued to be poor even though a large Economic diversification in rural areas was slow share managed to escape poverty. outside Java-Bali, and mostly limited to low-VA jobs. Economic diversification is a typical pathway toward economic prosperity. A structural transformation characterized by a shift from low-productivity agriculture 127 Presidential Decree No. 68 of 2022. 128 World Bank forthcoming. to higher-productivity sectors such as industry and 129 World Bank 2021a. 130 World Bank forthcoming. World Bank 2020b. 132 131 World Bank 2021a. World Bank 2020b. 133 38 Pathways Towards Economic Security Indonesia Poverty Assessment Drivers from 2014 to 2019 FIGURE 3.23: Share of working-age population in high-skilled FIGURE 3.24: Changes in endowment and returns to education jobs vs. log GDP per capita (Indonesia highlighted) for real wages by income deciles, 2014 versus 2019 Tradable services 0.06 35 0.04 30 0.02 25 % of WAP 20 0 15 -0.02 10 -0.04 5 2019 1991 -0.06 0 4 6 8 10 12 -0.08 Log of GDP per capita (2017 international $) 1 2 3 4 5 6 7 8 9 IDN Endowments Returns Source: Forthcoming Jobs report based on ILO data Source: SAKERNAS Note: Recentered-Influence-functions (RIF) for real wages, including dummies for gender, formal work, sectors and urban, by thresholds between deciles (10, 20, …, 90) services can increase wages and improve aggregate Also, digitalization can play a significant role by over- productivity. However, manufacturing opportunities coming place-based disadvantages, but this requires were rare in rural areas, with only 5.3 percent of poor good digital connectivity, particularly in remote areas.136 and 6.7 percent of non-poor rural households taking advantage of them. Even in regions with large mining Education and quarrying sectors, employment opportunities in Returns to education were diminishing as workers suffered manufacturing or the broader industrial sector were very from low-quality education and continued to be trapped in limited, in large part due to the capital-intensive nature low-VA jobs. of mining and quarrying. Across all rural areas, mostly low-VA jobs represented the only real alternative to While changes in educational endowment played agricultural work, but few poor rural households could only a minor role, returns to education continued to take advantage of them. A notable exception was Java- increase incomes and, hence, contributed to poverty Bali where higher-quality off-farm opportunities started reduction. Increasing levels of education of household playing a more important role, enabled by increased heads did not play a significant role in poverty reduction. connectivity that encourages agglomeration and its Their endowment of education among the poor hardly spill-over effects.134 In addition to lack of demand for changed from 2014 to 2019, given that most households higher skilled workers, deficits in human capital suggests heads were already out of school. Being educated, that many poor, especially in rural areas, might not have however, still earned a substantial wage premium. the skills needed to enter types of employment offering Each additional year of education increased returns by higher wages and incomes outside agriculture. 6.9 percent, about the same as global middle-income countries but below middle-income countries in the Lack of spill-overs from urban areas limited economic region.137 An average worker with junior secondary diversification and creation of more and better rural education earned 17 percent more than a worker with opportunities, except for Java-Bali. Place-based no more than primary education (Figure 3.24). For senior advantages from agglomeration forces were drivers of secondary education, the premium increased to 48 productivity increases.135 Thus, opportunities were more percent and for tertiary education to 108 percent. Thus, likely to arise close to existing opportunities, naturally education continued to contribute to poverty reduction. feeding into spatial inequality. Improved connectivity However, returns to education played a less important role through better infrastructure, for example, can help bring in 2019 compared to 2014 due to diminishing returns to areas closer to each other and expand spill-over effects. education for better-off households. From 2014 to 2019, World Bank 2020b. 134 World Bank 2021a 136 Roberts, Gil Sander, and Tiwari 2019. 135 SAKERNAS 2019. 137 Pathways Towards Economic Security Indonesia Poverty Assessment 39 Drivers from 2014 to 2019 the overall level of education, or endowments, increased Spatial factors for workers (though less so for household heads) leading Urban areas offered opportunities but costs of living in cities to higher wages across the income distribution. Returns attenuated benefits on poverty reduction. to education, though still positive, diminished for the top 60.138 This explains the reduced contribution of Urban areas offered better services and higher returns to education for poverty reduction discussed connectivity, providing access to better opportunities. earlier. With lower returns to education among the top Urban areas provided better labor incomes, supported 60, the difference in returns to education between the by place-based agglomeration effects increasing poor and non-poor narrows. Thus, returns to education productivity, but also better labor market matching.141 contributed less (though still positively) to poverty Urban households also had access to better services, reduction in 2019 than in 2014. On other words, returns including electricity, sanitation, education, and health. of education are less of a distinguishing feature between Urban dwellers had cheaper and faster internet access, the poor and non-poor in 2019 compared to 2014. allowing them to access eCommerce to take advantage Increased supply of skills in the labor market as well as of lower prices, larger selection, and more comfort. Urban limited productivity gains from education may have workers enjoyed more opportunities to become digital contributed to this result. Indeed, the (low-VA) service gig workers and digital entrepreneurs, or to become more sector, which employs most workers outside agriculture, productive in their regular jobs when using digitalization. showed dismal productivity gains. Also, digital services were increasingly offered digitally, enabling urban citizens with better internet access to Low-quality education also contributed to low returns access them more easily and at lower cost.142 to education. Higher education helped to get better jobs in Indonesia, but it did not guarantee it. While job However, urban Indonesian households suffered from availability and skills match played important roles, so higher housing costs than rural households, as well as did quality of education. From 2006 to 2018, quality when compared internationally. Urban dwellers paid of education stagnated, with PISA scores settling just a larger share of their consumption in housing costs below 400. Not surprisingly, the increase in educational compared to rural households. While the difference attainment has not translated into higher skills.139 In fact, was larger for the top 60 at 3.4 percentage points, it was closing Indonesia’s quality gap in education would have still sizable for the bottom 40 at 1.6 percentage points and estimated seven times higher benefits for economic (Figure 3.25). Also compared internationally, many urban growth compared to closing the access gap.140 Indonesians—especially in the cities of Bandung (12.1 FIGURE 3.25: Consumption share for rent payments, by urban/ FIGURE 3.26: Hours in lost congestion by city per year rural and decile in 2019 20 160 147 140 15 120 100 Percent 10 80 5 60 59 56 40 36 0 1 2 3 4 5 6 7 8 9 10 20 US$ 3.2 Deciles 0 Rural Urban National Jakarta Denpasar Malang Surabaya Source: Authors’ calculation based on SUSENAS Source: Author’s compilation based on INRIX, 2019 138 Ferreira, Firpo, and Messina 2017; World Bank 2019c. 139 World Bank 2020d. Roberts, Gil Sander, and Tiwari 2019. 141 140 OECD 2015. World Bank 2021a. 142 40 Pathways Towards Economic Security Indonesia Poverty Assessment Drivers from 2014 to 2019 pp), Denpasar (11.9 pp), and Jakarta (10.3 pp)—paid a years in some pollution hotspots (at sustained 2016 levels higher share of their income on housing compared to of pollution).149 The poor often lived in more polluted Bangkok (7.7 pp), Singapore (4.8 pp), and Kuala Lumpur areas and did not have the means to protect themselves, (4.0 pp).143 Consequentially, about one-third of urban by using air filters, for instance. Not surprisingly, air households (31 percent in 2018) were estimated to live pollution is often identified as the most pressing urban in slums.144 More affordable housing can help to relieve environment issue.150 pressure on household incomes, especially for the poor, and help more households to access services and jobs in The substantial costs for living and working, especially urban areas. in core urban areas, triggered urban sprawl with households moving to urban peripheries, diminishing In addition, long commutes and congestion in urban urban agglomeration effects. Even though urban areas reduced productivity and cost 0.5 percent of households benefitted from access to more stable and GDP annually. High costs for housing in core urban areas better paid jobs and education, substantial costs arose extended commutes to jobs, often located in the center from living and working in urban areas. High housing of urban areas. Large ownership of private vehicles costs and long commutes as well as congestion created congestion, adding to commute times. About disincentivized households to move into urban core one-third of commuters in core Jakarta and one-half areas where agglomeration effects were strongest. in greater Jakarta spent more than 1 hour commuting Instead, they moved to urban peripheries connected to work.145 Accordingly, Jakarta was the third most to urban core areas. However, households in urban congested city among 18 megacities, adding 58 percent peripheries had less access to good public services, like of travel time to every trip.146 An average urban driver education and health, with its implications for human lost between 36 and 147 hours per year in congestion capital. It also reduced knowledge-spillovers and in the four most congested urban areas (Figure 3.26). prosperity-enhancing agglomeration effects.151 Lack of viable public transportation options made commutes longer and more expensive. Congestion Diminished agglomeration effects limited the urban reduced productivity and cost about 0.5 percent of premium for poverty reduction and positive spill- GDP annually147 while damaging health of citizens and overs to rural areas, while reducing incentives for the environment. Improved urban planning and better migration. Urban dwellers benefitted less from urban public transportation can reduce economic costs as well connectedness, but suffered from its high density. This as costs for households while improving urban wages helps explain the minuscule contribution of urbanization through increased productivity. on poverty reduction in Indonesia. Furthermore, it limited positive spill-over effects to rural areas as the Air pollution also reduced the quality of urban urban growth engine was not able to create demand life. Congestion, use of fuels for household power for higher-VA goods and services in the nearby rural generation, and industrial coal power plants in or near economy.152 It also disincentivized rural-urban migration, urban centers created air pollution. Indonesia ranked undermining urbanization acceleration that could 17th among countries with most polluted cities from unleash productivity gains from urban agglomeration 2018 to 2021, with Jakarta often the most polluted city.148 effects.153 In fact, rural-urban migration contributed less High air pollution is associated with illnesses that reduce than 20 percent to urban growth,154 with Java-Bali and productivity and decrease the life quality and length. especially Jakarta being less of a magnet for migrants as Pollution levels in Indonesia lower life expectancy by one might expect from a capital city.155 and estimated 1.2 years on average, with losses of over 4 149 Greenstone and Fan 2019. 143 Roberts, Gil Sander, and Tiwari 2019. 150 World Bank’s Urban Perception Survey, 2018. 144 World Development Indicators. 151 Roberts, Gil Sander, and Tiwari 2019. 145 Roberts, Gil Sander, and Tiwari 2019. 152 World Bank 2020b. 146 Tomtom Traffic Congestion Index. 153 World Bank 2020b. 147 Roberts, Gil Sander, and Tiwari 2019. 154 Roberts, Gil Sander, and Tiwari 2019. 148 IQAir World Pollution Index. 155 Wajdi, Mulder, and Adioetomo 2017; Pardede, McCann, and Venhorst 2020. Pathways Towards Economic Security Indonesia Poverty Assessment 41 Drivers from 2014 to 2019 Prices Sanitary and Phytosanitary Measures (SPS), pre-shipment Structurally high food prices limited purchasing power inspections, specific port of entry requirements, and for the poor, while producer subsidies were costly, largely import monopolies. While some measures are important ineffective, and did not target the poor. for food safety, lifting a subset of unnecessary NTMs and streamlining their implementation could lower food Food prices were increasing faster than inflation, prices by 8 to 55 percent, and reduce under-nourishment limiting purchasing power especially of the poor. by about 5 percent, or about 0.5 percentage points.159,160 Food price inflation from 2009 to 2019 has consistently It would also improve the use of imports as buffers outstripped general inflation by 1.8 percentage points on when domestic production drops, in the case of a shock average per year (Figure 3.27). This particularly affected for example. poor households as they spend a large share of their consumption on food (63 percent for the bottom 40 Low productivity and high distribution costs percent). Even though net producers of food benefitted hampered domestic agricultural production.161 Low from higher food prices, most agricultural households agricultural productivity is due to fragmented and in the bottom 40 were subsistence farmers producing labor-intensive production. Even though fertilizer use in insufficient quantities to cover their own consumption Indonesia increased over the past two decades, growth and remained net consumers. Only about 15 percent in yields remained low, especially compared to regional of households were net producers.156 Most of them had peers. Underfunded and institutionally fragmented R&D land holdings of more than 1 ha and, thus, were unlikely struggles to inform effective extension services. Poorly to be in the bottom 40. maintained rural infrastructure affects input prices as well as drives a wedge between farm gate and market Import trade barriers partly explain the high staple prices. More generally, restrictions on commercial activities food prices. The retail price for rice, the main staple food related to food and agriculture (for example, food retail in Indonesia, has remained highest among neighboring investments) and high logistics costs due to regulatory countries throughout the last decade (Figure 3.28).157 barriers to entry exacerbate market prices. Low resilience Burdensome, inefficient, and costly non-tariff measures to climate shocks contributes to price volatility. (NTM) contribute to the price gap.158 These include FIGURE 3.27: Annual inflation by product category FIGURE 3.28: Retail price for rice in US$ per kg, across countries 20 1.2 Indonesia 15 1.0 Retail prices, USD / kg 10 0.8 Thailand Philippines 5 0.6 Cambodia Myanmar 0.4 0 Vietnam 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 0.2 -5 Food Proc. Food, Bev., Tob. 0.0 Housing Transp., Comm., Fin. 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 General Note: Includes Processed food, beverages and tobacco (Proc. Food, Bev., Tob.); and Source: FAO GIEWS Transportation, Communication and Finance (Transp., Comm., Fin.) 159 Prevalence of Undernourishment is calculated based on the Minimum Dietary 156 Cali et al. 2021. Energy Requirement (MDER) according to the FAO guideline. 157 World Bank 2020 – IEP. 160 ali et al. 2021. 158 Townsend et al, 2016. 161 World Bank 2020b. 42 Pathways Towards Economic Security Indonesia Poverty Assessment Drivers from 2014 to 2019 Government producer subsidies to support food Taxes and public spending prices showed limited benefits. The GOI introduced Fiscal policies contributed to poverty reduction, but could price controls to dampen inflation and protect be better targeted to the poor. households from losses in purchasing power. However, price controls are expensive and distort consumer and Fiscal policy affects household consumption through producer choices.162 In Indonesia, the majority of central direct and indirect taxes as well as transfers. All government spending on agriculture, which fluctuated households pay some combination of taxes and benefit around 2 to 3 percent of GDP, subsidized irrigation and from some combination of direct transfers—often fertilizers (50 to 70 percent between 2005 and 2020). through social assistance—and other spending, such Despite massive spending, benefits on agricultural prices as subsidies or public health and education. The net were limited for multiple reasons. Rent seeking limited balance of contributions (taxes) and benefits (spending) benefits,163 while fertilizer subsidies, which accounted for for each household determines whether they are a 25-30 percent of the annual agricultural budget, were net fiscal contributor or beneficiary. The Commitment not only expensive but also poorly targeted, regressive, to Equity (CEQ) framework generates counterfactual subject to leakage, and not cost-effective.164 In addition, income distributions at various levels of taxes and the subsidies might have worsened productivity, public spending to compute poverty and inequality diversification, and competitiveness by “crowding out” indices (Box 3.2). This allows disentangling the different public spending for research, innovation, extension, components and programs to understand their specific diversification, processing, and marketing. 165 effects on poverty and inequality.166167168 Box 3.2: Commitment to Equity (CEQ) framework The Commitment to Equity (CEQ) framework helps assess how government fiscal programs affect poverty and inequality in Indonesia. Originally developed by researchers affiliated with the CEQ Institute at Tulane University, this methodology has become a standard and has now been applied to several countries across the world.166 In Indonesia, it has previously been applied in 2012 and 2017 while this report adds the analysis for 2019.167 While the framework provides general comparability across years, the available data in each year determines how indicators are estimated, which needs to be considered when comparing results across years. CEQ entails generating theoretical (counterfactual) effects on household income distributions and poverty and inequality based on application of different taxes and public spending options. The analysis starts with “market income”; that is, income earned either in the form of wages and salaries or profits from self-employment or as returns on capital. “Net market income” is what would be left after all relevant taxes, deductions, and withholdings paid. Any direct household government cash transfer is added to net market income to arrive at “disposable income”. At this stage, depending on the consumption basket of the household, various indirect taxes (VAT and excise) are added or subtracted to yield “consumable income” or “post-fiscal income”. Finally, accounting for any in-kind government benefits received for things like education and health, discounting associated co-payments and user fees, yields “final income”. Since Indonesian household surveys do not measure income but household consumption, the CEQ is slightly different in practice. The analysis proceeds by equating household consumption to disposable income and working backward and forward to determine the other income concepts. Once these income concepts have been calculated, the “impact” on poverty and inequality is essentially the difference between the relevant measures. For example, the impact of fiscal policy on “inequality” is the difference between the Gini of “market income” (arguably untainted by fiscal policy) and the Gini of “final income” when all relevant fiscal instruments have been applied.168 Source: Adopted from World Bank 2020f 166 Lustig 2018. 167 Jellema, Wai-Poi, and Afkar 2017; World Bank 2020f. 168 It is important to understand that the CEQ methodology is essentially a partial equilibrium analysis and the use of the word “impact” is more an accounting term than a true causal one. The correct interpretation of the difference in Gini 162 World Bank 2022e; World Bank 2022f. between market income and final income after evaluating a tax or spending 163 Krisnamurthi, Bayu and Utami, Anisa Dwi 2022. option would be the amount by which inequality would have been higher or lower without all existing fiscal policy instruments being applied to market 164 World Bank 2017. income. This is different from an interpretation that might suggest that that 165 World Bank 2020g. package of fiscal policies reduces inequality by the given amount. Pathways Towards Economic Security Indonesia Poverty Assessment 43 Drivers from 2014 to 2019 Fiscal policy continued to contribute to poverty 19 percent of GDP in 2012 to 16 percent in 2019, the reduction, but less so than in 2012. In 2019, fiscal policy lowest among all middle-income and emerging market contributed to poverty reduction by 1.1 percentage economies, which on average spent 30 percent (Figure points, compared to 4.3 percentage points in 2012 3.29). The main reason for the low expenditure ratio was (without considering health and education effects).169 low revenue collection, which dropped from 17 percent While lower poverty levels of 23 percent in 2019 of GDP in 2012 to only 14 percent in 2019 (Figure 3.30). compared to 40 percent in 2012 can make it harder to Only Pakistan and Sri Lanka collected less in 2019. reach the poor, several factors contributed to the decline Several reasons contributed to low revenue collection in effectiveness. in Indonesia: (i) the cyclical nature of revenues due to their linkage to commodity prices; (ii) the economic Indonesia’s fiscal expenditures did not increase structure with a reliance on resource-extraction and a commiserate with GDP, and generally remained large informal economy; (iii) tax administration capacity, constrained given limited revenue collection. limiting tax revenues to below half of potential; and (iv) Indonesia’s expenditure-to-GDP ratio dropped from sub-optimal tax policies, such as VAT exemptions.170 FIGURE 3.29: Government expenditure relative to GDP per capita FIGURE 3.30: Government revenues relative to GDP per capit 60 60 50 50 40 40 30 30 Malaysia Malaysia 20 20 Thailand Thailand 10 Indonesia 10 Indonesia Philippines Philippines 0 0 3 3.5 4 4.5 5 5.5 3 3.5 4 4.5 5 5.5 Log GDP per capita Log GDP per capita Source: IMF Fiscal Monitor showing data for all middle-income and emerging market economies for 2019 FIGURE 3.31: Absolute (left) and relative (right) incidence of indirect tax by income decile 30 10 25 8 20 6 Percent Percent 15 4 10 2 5 0 0 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 2012 2019 2019* 2012 2019 2019* Source: Authors’ calculations using SUSENAS (2002-2022) The Commitment to Equity (CEQ) framework, developed by the Commitment 169 to Equity Institute (CEQ Institute) at Tulane University, as well as implementation guidelines, applications, and software can be found in Lustig 2018. For a full set and discussion of 2012 Indonesia results, see World Bank 2015b; for 2017 results, World Bank 2020f. Note that the 2012 results presented for comparison in alongside the 2017 results differ from the original report due to revisions made to help ensure comparability of numbers over time. 170 World Bank 2020d. 44 Pathways Towards Economic Security Indonesia Poverty Assessment Drivers from 2014 to 2019 Nevertheless, indirect taxes worsened poverty, while FIGURE 3.33: Poverty impact, cost and efficiency of modeled direct transfers and subsidies VAT exemptions were ineffective in limiting poverty. Revenue collection includes direct taxes, such as a 10.0 personal income tax, as well as indirect taxes, including 8.0 VAT and excise. Revenue collection from personal income 6.0 taxes remained low at 0.9 percent of GDP in 2012 and 4.0 1.0 percent in 2019. They hardly affected poverty given high thresholds to pay personal income tax, as well as 2.0 the large informal sector.171 The largest share of revenue 0.0 2012 2019 2012 2019 2012 2019 was collected through indirect taxes, increasing poverty Poverty impact Cost (% GDP) E ciency by 4.2 percentage points in 2012 and 4.1 percentage (in pp) points in 2019. VAT has a statutory rate of 10 percent for Transfers Subsidies Total most goods and services, but with many exemptions Source: Authors’ calculations based on SUSENAS and administrative as well as budget data. Note: Marginal poverty impacts can differ from combined impacts. Cost only covers worth about two-thirds of a percent of GDP. While VAT programs included in the modeling, and can differ from the overall spending mentioned in the text. Efficiency calculated as poverty impact per cost of one percent of GDP. exemptions are granted mostly on equity grounds to help the poor, about half the value accrued to the expenditure reduction (reversed in 2021/20), while top 30 percent households (Figure 3.32).172 Removing spending on transfers increased from 0.3 to 0.7 percent exemptions and mitigating adverse impacts on the poor of GDP. Although overall efficiency increased with through targeted direct transfers can enhance fiscal the shift towards transfers (from 1.9 to 2.4 percentage space, given the higher efficiency of direct transfers. points of poverty reduction for each 1 percent of GDP in spending), the efficiency increase was insufficient to Direct transfers and energy subsidies reduced overall compensate the drop in budget. poverty, with a shift towards more efficient transfers. Direct transfers and subsidies reduced poverty by Governments often prefer price subsidies despite their 4.5 percentage points in 2019, significantly less than negative implications. Price subsidies distort consumers’ the 7.8 percentage points in 2012 (Figure 3.33). In and producers’ choices and are expensive, yet they are this period, fiscal expenditures on transfers and often used to dampen the negative effects of rising prices subsidies dropped from 4.2 percent of GDP in 2012 on household purchasing power. This may be because to 2.1 percent in 2019. A drop in subsidies from 3.9 price subsidies do not require complicated targeting percent in 2012 to 1.7 percent of GDP in 2019 drove mechanisms and are easy and quick to implement. FIGURE 3.32: Total consumption of exempted goods and services FIGURE 3.34: Efficiency of individual transfer and by consumption decile (2019) subsidy programs 25 3.50 3.00 20 2.50 2.00 15 1.50 Percent 1.00 10 0.50 - All LPG Diesel Premium Kerosene Electricity 5 0 1 2 3 4 5 6 7 8 9 10 Subsidies 2019 2012 2019 Source: Authors’ calculations. 171 Impact cannot be estimated for 2012 and 2019 due to unavailability of tax data. In 2017, the poverty impact was not significantly different from zero. 172 World Bank 2020f. Pathways Towards Economic Security Indonesia Poverty Assessment 45 Drivers from 2014 to 2019 Without specific targeting, they often benefit a large standing programs include the conditional cash transfer part of the population and, thus, attract broad support, Program Keluarga Harapan (PKH), a cash transfer for poor which also complicates removal of subsidies. The and vulnerable students Indonesia Pintar (PIP), a food population may even resist a promise to shift subsidies assistance program (BPNT/Sembako), and a subsidized towards a targeted social transfer, due to doubts about health insurance premium waiver (JKN-PBI).175 the commitment of government to follow through on the promise and concerns about corruption.173 Further improvements in targeting can increase the cost effectiveness of social assistance. Indonesia’s Indonesia’s subsidized energy prices are inefficient. The designed its social registry Data Terpadu Kesejahteraan Government subsidizes cooking gas (LPG and Kerosene), Sosial (DTKS) to include socioeconomic characteristics of electricity, and fuel (Diesel and Premium). LPG has low the poorest 40 percent of households.176 However, the and decreasing efficiency as wealthier households list was neither updated regularly nor comprehensively, benefit disproportionately from the subsidy, given a lack despite the mandated obligation to update twice per of targeting (Figure 3.34). While reform plans included a year. Constraints at both the national and local levels have closed distribution system, implementation of the reform left much of the database still with 2015 information. has been continuously delayed. Only Kerosene became Hence, errors of exclusion and inclusion are worsening as more efficient in reducing poverty, but inadvertently as time passes.177 As of late 2021, DTKS covered the poorest wealthier households stopped using kerosene to take 51.8 percent of Indonesia’s population due to a policy advantage of the LPG subsidy. The subsidy for diesel shift to expand coverage of the registry, including open did not play a big role given lack of consumption by online registration for individuals living in Jakarta.178 households and limited indirect effects. The subsidy for However, this recent expansion no longer included a premium fuel and electricity had similar efficiencies. Even set of socioeconomic variables.179 PKH is a case in point. though they implicitly targeted—for example, to low- Nationally, coverage of the program for the target volume retail consumers of electricity—many wealthier group of the bottom 10 increased from 9 to 27 percent households used them, thus increasing the cost for the between 2014 and 2019 and ranged from a minimum poor. Overall subsidy efficiency dropped from 2.0 in 2012 of 9 percent in rural Maluku-Papua to 47 percent in rural to 1.8 percentage points in 2019 for each 1 percent of Nusa Tenggara. Similarly, PIP covered only 18 percent of GDP in expenditure. The reduced efficiency emphasizes its targeted bottom 20 percent of households nationally the difficulty of targeting subsidies. While planned in 2019. In addition to low coverage of target groups, online registration to receive some fuel subsidies might convergence of social assistance programs was very low, improve targeting, it will remain sub-optimal, hard to despite theoretically targeted by the same DTKS registry. maintain, and potentially trigger work-arounds to game This reflects in part the need to improve institutional the system. coordination with subnational governments and between agencies responsible for key programs. While In contrast, the efficiency of transfers through social efforts are ongoing to prepare a socio-economic registry assistance programs improved significantly, reducing (Regsosek) intended to improve targeting, having poverty at less than one-third the cost of government multiple targeting databases without inter-operability subsidies (Figure 3.34). Indonesia’s social assistance will create challenges. programs evolved from providing temporary support 175 JKN-PBI is not included. during the Asian Financial Crisis to core permanent 176 The following programs use DTKS for targeting: BPNT-Sembako (Food Assistance programs with occasional temporary support during Program - Bantuan Pangan Non-Tunai); PKH Family Hope Program conditional cash transfer (Program Keluarga Harapan); PBI-JKN Subsidized Health Insurance periods of subsidy reform (for a comprehensive overview, (Penerima Bantuan Iuran - Jaminan Kesehatan Nasional); the PIP cash transfer for poor and vulnerable students (Program Indonesia Pintar), and BST Cash Transfer see Annex Table A4). Social programs thus evolved (Bantuan Sosial Tunai). 177 Holmemo et al. 2020. and expanded rapidly even before COVID-19.174 The 178 Based on the Decree of the Social Minister of the Republic of Indonesia Number 145 / HUK / 2021 issued on 26 November 2021 and the total population of 271,584,774 people from Susenas 2021. The number of families and individuals World Bank 2022f; World Bank 2022e. 173 recorded in DTKS was updated twice a year since 2017 through the Decree. Holmemo et al. 2020. 174 179 Hadiwidjaja, Williams, and Giannozzi 2022. 46 Pathways Towards Economic Security Indonesia Poverty Assessment Drivers from 2014 to 2019 Social assistance also had positive effect on health, Inequality education, and the environment. The conditional Long-term trend cash transfer program PKH has substantially increase Inequality reached its peak in Indonesia in 2010 and utilization of health and education services, helping declined after 2014 due to more inclusive growth. to reduce childhood stunting and improve nutritional intake, with positive behaviors sustained among those Inequality increased from 2002 to 2010 since who graduated from the program.180 The program also economic growth did not trickle down equally to the reduced deforestation in villages where households bottom 40. Even though all household benefitted from participated in the program.181 Cash transfers also the commodity boom during this period, consumption helped reduce yearly suicide rates by 18 percent (or 0.36 of the bottom 40 percent grew by only 1.8 percent per per 100,000 people) in Indonesia, possibly due to the year compared to 4.1 percent for the top 60 percent avoidance of depression.182 In contrast, energy subsidies (Figure 3.37). In this period, large investments in exacerbate negative externalities, contributing to GHG capital took place, increasing income from capital. The emissions and air pollution. unemployment rate dropped from a peak of 11 percent to 7 percent, while labor force participation stayed However, direct expenditures on education and health constant. The large pool of available workers limited remain low, contributing to unsatisfactory human pressures on wages, which grew by 1 percent annually. capital outcomes. Indonesia’s education expenditure Instead, growth translated more rapidly into capital gains ratio of 2.8 percent of GDP in 2019 was among the five accruing to wealthier households with financial assets, lowest ratios of middle-income and emerging market exacerbating inequality.183 economies (Figure 3.35). Similarly, its health expenditure ratio stood at 1.4 percent of GDP, with only Angola and From 2011 to 2014, the rise in inequality stopped with India spending less (Figure 3.36). The lack of funding, welfare gains becoming more equally distributed as well as its execution, explain the challenges around across the population. Consumption of the bottom improving human capital outcomes, including maternal 40 percent increased annually by 4.2 percent, similar to mortality, stunting rates, and learning outcomes. gains of the top 60 percent. The deteriorating terms of FIGURE 3.35: Government education expenditure relative FIGURE 3.36: Government health expenditure relative to to GDP per capita GDP per capita 8.0 7.0 7.0 6.0 6.0 5.0 5.0 4.0 4.0 3.0 Thailand 3.0 Malaysia 2.0 2.0 Thailand Malaysia 1.0 Indonesia 1.0 Indonesia Philippines Philippines 0.0 0.0 3 3.5 4 4.5 5 5.5 3 3.5 4 4.5 5 5.5 Log GDP per capita Log GDP per capita Source: IMF Fiscal Monitor showing data for all middle-income and emerging market economies for 2019 180 Syamsulhakim and Khadijah 2021. 181 Ferraro and Simorangkir 2020. 182 Christian, Hensel, and Roth 2019. World Bank 2016. 183 Pathways Towards Economic Security Indonesia Poverty Assessment 47 Drivers from 2014 to 2019 FIGURE 3.37: Consumption growth across periods, annualized FIGURE 3.38: Theil inequality decomposition between urban/ rural and provinces 6.0 0.40 5.0 0.30 4.0 0.20 Percent 3.0 0.10 2.0 0.00 1.0 2002 2005 2008 2011 2014 2017 2020 2002 2005 2008 2011 2014 2017 2020 0.0 1 6 Urban-Rural Province 11 16 21 26 31 36 41 46 51 56 61 66 71 76 81 86 91 96 2002-2010 2011-2014 2014-2019 Within Between Note: Due to changes in the sampling methodology of the underlying survey from 2010 to Note: Consumption aggregates are spatially deflated. 2011, periods are chosen to minimize its impact. trade in this period lowered capital investments. At the annually by more than 3 percent. The increase in labor same time, the unemployment rate reached its lowest incomes relative to capital gains helped stop the growth at 6 percent and remained at this level. With domestic of inequality. Thus, households, especially poorer ones, consumption increasingly contributing to economic were able to claim a larger share of the albeit slowing growth and less workers available, real wages increased economic growth. Box 3.3: Top incomes and measurement issues Missing incomes for the richest households in surveys is a common phenomenon in most countries.184 Household surveys are designed to be broadly nationally representative and are usually not stratified to capture the very top end of the income distribution. This problem does not significantly affect many survey-based estimates. For example, correcting for top incomes has relatively little effect on global poverty rates.185 However, it can have potentially large effects on inequality measures because income and wealth concentrate in the upper tail of the distribution. Indonesia exhibits a large gap of private consumption between survey-based and national accounts data, hinting at the problem of missing top incomes. One indication of missing top incomes is the gap between total consumption in household surveys and private consumption in national accounts.186 In Indonesia, total consumption in its household surveys (Susenas) accounts for only around 40 percent of private consumption in the national accounts. This gap of 60 percent is much larger than in many other countries; the developing country gap in 1998 was 23 percent and in East Asia 19 percent. The gap in Malaysia was 51 percent in 2019. Globally, per capita consumption is about 22 percent lower in surveys compared with national accounts.188 The effect of missing top-income data on inequality measures such as the Gini coefficient is significant, and might increase Gini coefficients by 20 percent in an average country. The bias from missing top-income data can be evaluated and somewhat corrected by merging survey data with administrative data from tax returns.189 For example, the share of national income going to the richest 1 percent of households in Chile increases from 14 percent in survey data to 17 percent once administrative tax data are included. This increases the Gini index by 8 percent from .64 to .69. In Brazil, the top 1 percent share increases from 10 percent to 24 percent, and the Gini index by 21 percent from .51 to .62.190 In Malaysia the top 1 percent income share increases from 8 percent in the survey data to 13 percent after including administrative tax data, and again to 15 percent after estimating undistributed corporate profits from national accounts.191 On average, observed Gini coefficients are 20 percent when adjusting the income distribution for missing top-incomes.192 184185186187188189190191192 184 Lustig 2019; Ravallion 2021. 185 Prydz, Jolliffe, and Serajuddin 2021. 186 See Prydz, Jolliffe, and Serajuddin 2021 for a discussion of other indications. 187 Bhalla 2002. 188 Prydz, Jolliffe, and Serajuddin 2021 based on comparisons involving 166 countries. 189 Piketty 2003; Piketty and Saez 2003 are the key reference works, while Atkinson and Piketty 2007 represent the main cross-country studies. 190 Blanchet, Flores, and Morgan 2022. 191 Khalid and Yang 2019. 192 Prydz, Jolliffe, and Serajuddin 2021. 48 Pathways Towards Economic Security Indonesia Poverty Assessment Drivers from 2014 to 2019 Inequality started to drop from 2014 to 2019 but Declines in fertility benefited the wealthy more than without significant relative gains for the bottom 40. the poor, helping to keep inequality high. Lower Annualized consumption growth of the bottom 40 fertility rates led to smaller households, contributing percent stood at 4.6 percent, while the top 60 percent to poverty reduction. However, wealthier households had slightly larger gains of 4.8 percent. However, were becoming smaller more quickly. This contributed consumption of the top 20 increased by only 3.7 percent. to inequality as it increased relative consumption of The stark drop in gains for the wealthiest households wealthier households. In fact, the Gini would have might partially be attributed to measurement problems been 1.3 points lower in 2019 at .364 instead of .371 if for top income quintiles (Box 3.3). Proper measurement household sizes would not have changed since 2014,193 would likely further exacerbate the relatively lower gains the same result as in 2014.194 Thus, fertility in poorer for the bottom 40 percent. households has not reduced to the point of wealthier households. Female education and family planning can Drivers since 2014 reduce fertility, especially among the poorest.195 This Lower fertility did not translate into lower inequality, while can create a virtuous cycle where women in smaller relative labor incomes benefitted the poor, but at the cost of households are better able to join the labor force while diminishing returns to education for top 60 workers. investing more in human capital, in turn further reducing fertility. With Indonesia’s fertility rate still relatively high Most inequality is within areas rather than differences at 2.3 in 2020, progress among poor households holds between areas. In 2002, only about 1.2 percent of potential to reduce poverty and inequality. inequality was explained between urban and rural areas, and 12.4 percent between provinces (Figure 3.38). In Despite the lack of relative gains in consumption 2019, the already small contribution of between-area for the bottom 40 percent, their relative labor inequality dropped even further to 0.8 percent for urban incomes have improved due to increasing real wages. and rural, and 3.3 percent for provinces. Thus, inequality Employment trends were relatively similar across was not as much a spatial phenomenon at the level of income groups in terms of labor force participation, urban, rural, or provinces but found within. In addition, the employment, and under-employment. However, real convergence of lagging regions contributed to an even wages for lower educated workers increased relative to smaller fraction of inequality explained between areas. workers with tertiary education (Figure 3.39). A worker However, many remote areas within provinces were still with no more than primary education in 2014 earned lagging, contributing to inequality within the province. only 30 percent of a worker with tertiary education, FIGURE 3.39: Relative real wages by education, compared FIGURE 3.40: Share of sector of employment of household heads to tertiary education 80 40 60 30 Percent 40 20 20 10 0 0 2006 2014 2019 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 Agriculture Manufacturing Primary or below Junior Secondary Low VA services High VA services Senior Secondary Other Source: SAKERNAS Source: SUSENAS 193 Based on a counter-factual simulation holding household sizes constant since 2014 at the decile. Estimated based on Susenas 2019. 194 World Bank 2016. 195 David E. Bloom et al. 2012. Pathways Towards Economic Security Indonesia Poverty Assessment 49 Drivers from 2014 to 2019 FIGURE 3.41: Share of household consumption by thus, more strongly affected consumption of the bottom category and decile, for 2019 40 relative to the top 60, increasing inequality. Solving 100 the structural constraints of high food prices would likely 80 reduce poverty and inequality. 60 Education inequality has dropped significantly, but Percent wealthier households receive better education. 40 Unequal access to education deprives the poor of future 20 economic opportunities, exacerbating inequality.196 Access to education, especially for the poor, has 0 1 2 3 4 5 6 7 8 9 10 significantly improved over the last years, with equal Food Energy Transport Other Non-Food enrollment rates for primary and junior secondary Source: SUSENAS. education between poor and non-poor students, but 37 percent (a relative increase of 24 percent) in and enrollment rates between the two groups have 2019. With poorer households having lower levels of converged significantly for senior secondary education. education, these relative gains reduced inequality. This has the potential to help improve inter-generational economic mobility. However, wealthier households were Real wages for lower-educated workers have increased often able to send children to better schools by paying because of sectoral shifts and lack of jobs for better fees, unaffordable for poorer households. In fact, the educated workers. Workers, especially those with lower average gap in learning outcomes between students education, continued moving from agriculture to services attending high versus low-performing schools within (Figure 3.40), which paid higher wages helping them to a district was equivalent to six years of education.197 begin catching up with more educated workers. In part, The Government reacted in 2021 to address some this is because relatively fewer educated workers were quality concerns. Civil service teachers are now able to take advantage of their higher education. With required to achieve a minimum score in a selection premature deindustrialization ongoing, the economy exam,198 helping curb teacher appointments based on was not able to offer a good number of skilled jobs while social connections instead of merit.199 Nevertheless, more educated workers enter the labor market, but not improving education quality remains an important always with the right skills. Due to a lack of opportunities issue, including across districts. and skills mismatch, they often ended up in low-VA, low- productivity service jobs. The increasing share of these An increasing digital divide can threaten inequality jobs further undermined returns to education, which in progress.200 Digitalization provides many opportunities, turn hurt real wages of better educated workers who for consumers, workers, and beneficiaries of public cannot find more productive jobs. While this reduced services. However, accessing these opportunities and inequality, it shows that the economy was not taking services requires good, affordable internet connection advantage of higher skills, limiting productivity gains and digital skills. In 2019, about two-thirds of urban and, hence, pathways to economically secure jobs. households had an internet connection, compared to about one-third of rural households. In Java Bali, Food prices increases above inflation exacerbated more than half of all households were connected, inequality. Higher food prices limited the purchasing compared to only one-third in Papua. Households power of the poor, weighing more heavily on in the top 10 had five times higher access to internet poor households given their larger shares of food 196 Wihardja and Cunningham 2021. 197 Dharmawan and Suryadarma 2021. consumption. The bottom 40 spent 63 percent of their 198 Peraturan Menteri Pendayagunaan Aparatur Negara dan Reformasi Birokrasi consumption on food, while the top 60 only spent 53 Republik Indonesia No. 28/2021 (MoSAUBR Regulation No. 28/2021). 199 Rosser and Fahmi 2016. percent (Figure 3.41). The relatively higher prices of food, 200 World Bank 2021a. 50 Pathways Towards Economic Security Indonesia Poverty Assessment Drivers from 2014 to 2019 compared to the bottom 10 (71 versus 14 percent). Energy subsidies benefitted all households, making Gaps in infrastructure and regulatory concerns reduced them ineffective in reducing inequality. Energy availability, affordability, and quality of internet access. subsidies had almost no effect on reducing inequality, Hence, a large segment of the population remained contributing only 0.1 points of the Gini index in 2012 excluded from taking advantage of digital opportunities. as well as in 2019.202 Only 25 percent of spending on This segment—rural, remote, and lower-educated energy subsidies goes to the bottom 40 (Figure 3.43). households—were already more likely to be poor, such The large drop in budget since 2012 reduced their that an increasing divide bears the risk of leaving them impact on market income across the income distribution further behind, exacerbating inequality. (Figure 3.44). They also became slightly more progressive in 2019, but still benefitted households across the Fiscal policies distribution, making them inefficient in reducing poverty Fiscal policies attenuated inequality but were expensive and inequality. given large, inefficient, subsidies, such as the ones for energy. In contrast, direct transfers reduced inequality, albeit Fiscal policy helped reduce inequality, although by in small magnitude given their relatively limited far less than in most other middle-income countries. coverage and benefits. Direct transfers from social In 2017 and 2019, the combination of taxes, transfers, assistance programs reduced the Gini Index by 1.0 point subsidies, and spending on health and education in 2019, an increase of 0.3 points since 2012.203 About 60 reduced the Gini Index by around 3 points. However, percent of social assistance benefits reached the bottom while Indonesia’s fiscal policy reduced inequality, it did 40 (Figure 3.44). Transfers substantially increased market so by much less than most other middle-income and income especially for the poorest 10 percent households, all high-income countries, which reduced inequality by in the same magnitude as subsidies did in 2012. They 5 to 15 points of Gini (Figure 3.42).201 In fact, Indonesia’s were also progressive with richer households benefitting 2017 result was the lowest ranked among upper-middle only marginally. However, they were low relative to income countries with data and better than just two market income, even though many poor households lower-middle income countries. One-third of the small received benefits through multiple programs. Even for number of lower-middle income countries with data the poorest decile of people with the lowest income, performed better. transfers were equivalent to only 9 percent of their pre- FIGURE 3.42: Impact of fiscal policy on Gini index, across countries 0 Change in Gini Index (points) -5 -10 -15 -20 Spain Uruguay Panama United States Croatia Mauritius Romania South Africa Argentina Brazil Mexico Namibia Georgia Venezuela Costa Rica Botswana Dominican Republic Colombia China Ecuador Peru Iran Turkey Belarus Malaysia Jordan Russia Albania Guatemala Paraguay Indonesia (2012) Indonesia (2017) Indonesia (2019) eSwatini Lesotho Zambia Tunisia Kenya Ukraine Honduras El Salvador Mongolia Bolivia India Nicaragua Moldova Egypt Tanzania Ghana Sri Lanka Comoros Ivory Coast Uganda Burkina Faso Togo Mali Ethiopia Niger Gambia Tajikistan Guinea -25 Low income High income Upper middle income Lower middle income Cash taxes and transfers In-kind spending on H+E Net scal impact Source: Authors’ calculations for Indonesia 2019; Indonesia 2017 from World Bank 2020f, other country results from the CEQ Data Center and World Bank databases, reported in World Bank 2022 202 Subsidies do not affect official measures of inequality in Indonesia which are based on consumption from Susenas. This measure of consumption is equivalent in the CEQ framework to Disposable Income, which is Market Income less direct taxes plus direct transfers, but before indirect taxes and subsidies. Nonetheless, the impact on household welfare is real, as is the opportunity cost of subsidy spending (for example, in terms of forgone social assistance or investments in health and education). 201 See World Bank 2022 for additional results with OECD countries. 203 World Bank 2020e. Pathways Towards Economic Security Indonesia Poverty Assessment 51 Drivers from 2014 to 2019 FIGURE 3.43: Concentration of direct transfers and subsidies, FIGURE 3.44: Share of benefits (direct transfers and subsidies) across consumption deciles, in 2017 relative to consumption, by consumption deciles 30 15 20 10 Percent Percent 10 5 0 0 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 2012 Subsidies 2019 Subsidies 2012 Subsidies 2019 Subsidies 2012 Transfers 2019 Transfers 2012 Transfers 2019 Transfers Source: Authors’ calculations, based on SUSENAS FIGURE 3.45: GIn-kind education benefit as share of FIGURE 3.46: In-kind health benefit as share of market income market income 22 20 17 15 Percent 12 Percent 10 7 5 2 0 -3 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Market Income Decile Market Income Decile 2012 2019 2012 2019 Source: Authors’ calculations, based on SUSENAS. fiscal income. The low level of overall benefits limited the health benefits have more than doubled since 2012, benefits on lowering inequality, but had a larger impact possibly due to the roll-out of universal health insurance on poverty because many poor households lived just increasing utilization rates. below the poverty line and could be lifted above the line with even small assistance. They need more help, However, decentralized service delivery exacerbated however, to bridge the gap to become economically geographic disparities, countering the decrease in secure households. inequality. Indonesia started to decentralize about two decades ago. Sub-national Governments (SNG) gained Health and education spending remained progressive, responsibilities for service delivery, including education reducing current and future inequality. In-kind and health care (Box 3.4), managing 40 percent of total education and health benefits204 were progressive government expenditure in these sectors. The quality both in 2012 and 2019, increasing relative consumable of subnational spending depends on allocative and incomes more for poorer than richer households (Figure technical efficiency. Allocative efficiency ensures that 3.45 and Figure 3.46). While education benefits had a allocations of SNG resources are aligned with needs, relatively larger impact of more than 15 percent for the including equity considerations. This mainly affected bottom 10 compared to health benefits at 8 percent, more populous districts, which received less budget than needed. Technical efficiency is the effectiveness 204 We monetize in-kind education and health in CEQ by applying government cost approach. We treated the benefit as the average cost government incurred per of SNG resources in producing service delivery results. utilization per beneficiaries. 52 Pathways Towards Economic Security Indonesia Poverty Assessment Drivers from 2014 to 2019 Sparsely populated districts struggled with technical funding. Both kinds of districts had worse outcomes, efficiency given a lack of capacity, leading to low quality while also having higher poverty rates, thus exacerbating of spending even205though206they received sufficient inequality. Box 3.4: Education and health services in the context of decentralization Services delivered by subnational governments (SNGs) are vital to reducing poverty and nurturing human capital. As a result of “Big Bang” decentralization reform two decades ago, SNGs are currently responsible for delivering most key services—including education and health care—while the central government plays a stewardship and quality assurance role.205 SNGs manage more than 40 percent of total government expenditure in Indonesia, including over 60 percent of health and education spending. Despite high reliance on transfers contributing two-thirds of their budget, SNGs also have wide-ranging autonomy in making spending decisions, with districts having discretion over the final use of about 85 percent of their total revenues. Despite major improvements, large gaps in access to, and quality of, services SNGs deliver still persist. Even though most districts improved access to services between 2010 and 2020, large disparities remain, such as in the share of households with access to safe sanitation or attended births. Relatedly, economic disparities also persist, and GDP per capita gaps between island groups are narrowing only gradually. To address human capital gaps, SNGs need to enhance the quality and efficiency of primary health care as well as the quality of primary and secondary education. The low quality of SNG spending creates gaps in service access and quality, and persistent geographic disparities. Indonesia’s transfer system does not adequately target fiscal resources to the SNGs that need them the most (allocative inefficiency). It especially underfinances the pressing service needs of populous urbanizing areas, which are key economic engines and home to many of the poor. In 2020, the most populous 20 percent of districts received less than one-fourth of revenues per citizen than the least populous 20 percent. This undermines the ability of more populous districts to provide high-quality services. At the same time, Indonesia’s main capital grant, the DAK Fisik, does not specifically target districts with the greatest infrastructure catch-up needs, perpetuating regional disparities in service infrastructure. SNG spending also inefficiently translates into service delivery results (technical inefficiency). Access to basic services, while much improved since decentralization, has not kept pace with growth in SNG spending. Whereas district real per capita spending, on average, increased by a factor of 2.4 between 2001 and 2020, access to basic services increased only by a factor of 1.7. Higher SNG spending, on average, translates into improved access to services, but by a smaller amount than perhaps possible; a 1 percent increase in transfer financed spending leads to only a 0.2 percentage point increase in an access to education and health (per a service index). Many districts—often the more remote and poorer ones—have insufficient capacity to spend the budget effectively or at all. Revenue outturn can vary between 5 to 80 percent, reflecting large differences in SNG public financial management (PFM) capacity on revenue forecast across districts. Only the DAU is significantly associated with improved service access. Low quality and inefficiency of SNG spending is particularly visible in the education sector, contributing to poor education outcomes. Along with the increasing trend of the national budget, the education budget has also increased, with a constitutionally mandated 20 percent earmark for education expenditures. The budget increase has financed significant expansion in student enrollment, but education outcomes measured by student learning still lags. SNGs, responsible for providing education service delivery, account for the bulk of education spending and differ in fiscal and administrative capacity to manage education performance. Indonesia should reassess the financial and technical capacity of SNGs. Various education programs implemented at subnational levels should be prioritized and consolidated to a smaller number proven to be effective in raising education outcomes. Three major cross-sectoral institutional gaps cause the low quality of SNG spending: an equity gap, an accountability gap, and a public financial management (PFM) capacity gap. First, the transfer system does not yet allocate transfers to those SNGs that need them the most. Second, SNGs are insufficiently accountable for achieving results to both citizens and central government. Third, subnational PFM policies, systems, and skills do not yet enable SNGs to spend effectively. Major SNG PFM challenges that constrain the quality of spending comprise: (i) a disconnect between SNGs’ policies and budgets, and low budget reliability; (ii) ineffective planning and execution of capital projects; and (iii) the low quality and transparency of spending information. PFM capacity also varies largely across districts. The central government has recently sought to reform SNG PFM policies to make support strategic SNG decision-making and effective and transparent spending.206 To be effective, however, these reforms need to be further defined and SNGs’ capacity to implement them strengthened. Bank forthcoming. 205 PP 12/2019 and Permendagri 77/2020. 206 Pathways Towards Economic Security Indonesia Poverty Assessment 53 Drivers from 2014 to 2019 CHAPTER 4 SHOCKS Photo: © Nugroho Sunjoyo/World Bank 54 Pathways Towards Economic Security Indonesia Poverty Assessment 4. SHOCKS Shocks threaten progress in poverty reduction, but social COVID-19 combined a large systemic shock with protection and financial inclusion— more flexible as well individual health shocks. as cost-effective than price subsidies—can help mitigate damaging effects from shocks. In the absence of a systemic crisis, household-level (idiosyncratic) rather than community-level (covariate) F rom the Asian Financial Crisis to a sick working household member, shocks affect livelihoods can push people into poverty. During the 1997- shocks drive economic insecurity. Idiosyncratic shocks are those a household experiences independent of other households in the community. Events such as 1998 financial crisis, Indonesia’s extreme poverty rate having an accident, being a victim of a crime, or suffering (measured at US$ 1.90 2011 PPP) jumped steeply from from a noncommunicable disease are usually household 44 percent in 1996 to 63 percent in 1998. Households specific. In contrast, covariate shocks, affect many were hurt by unemployment and rising food prices. Even households in a region or community at the same time. in the absence of a systemic shock, individual shocks Examples include droughts, floods, earthquakes, and can threaten household livelihoods. For example, a sick other natural disasters, as well as spikes in food prices household member can decrease income, especially if and epidemics. These are experienced simultaneously the person is employed in the informal sector without by most if not all households in a community. Both type sick-pay, while also triggering large health expenditures. of shocks can render households economically insecure. FIGURE 4.1: Ratio of idiosyncratic to covariate susceptibility to fall FIGURE 4.2: Account ownership among adults 15+ into poverty, for 2011 and 2019, by urban and rural 2.5 100 Thailand China 2.0 80 Malaysia Percent 1.5 60 Philippines 1.0 40 Vietnam 0.5 20 0.0 0 2011 2019 2011 2014 2017 2021 Rural Urban Bottom 40 Top 60 Total Source: Authors’ calculations using SUSENAS Source: Global Findex Database, 2021 FIGURE 4.3: Barriers to open accounts, 2021 FIGURE 4.4: Access to emergency funds and their sources, 2021 Percent 0 20 40 60 80 Overall Lack of money Family Family has one Work Savings Providers too expensive Borrow Providers too far Sell assets Lack of documentation Other Lack of trust 0 20 40 60 80 100 Religious reasons Percent Somewhat or not di cult Very di cult Not possible Source: Global Findex Database, 2021 Source: Global Findex Database, 2021 Pathways Towards Economic Security Indonesia Poverty Assessment 55 Shocks In the absence of a shock, the household remains out 56 percent of adults reported to be able to obtain of poverty but the shock may push it into poverty. emergency funds within 30 days, while 36 percent Idiosyncratic shocks were more than twice as important found it very difficult and 8 percent impossible (Figure in creating susceptibility to falling into poverty than 4.4). However, the main source for emergency funds – covariate shocks in Indonesia in 2019 (Figure 4.1). The family – can quickly be depleted or unavailable in the importance of idiosyncratic shocks increased, with rural case of large community-wide shocks, emphasizing catching up with urban areas. the importance of complementary sources, such as individual savings, insurance, and social assistance. Savings, insurance, and social assistance can buffer the negative effects of shocks and reduce their long- Health and employment term harm on households. Depending on the duration Health and employment shocks represent the main and severity of a shock, a combination of tools can help idiosyncratic household shocks. Employment shocks households cope. Savings can smooth consumption reduce household income. Similarly, health shocks can over the duration of a shock and replace lost assets, reduce household income due to the inability to work insurance can cover lost income or additional health or diminished productivity, but also because of required expenditures, and social assistance can complement care for another household member. In addition, they livelihoods. In the absence of savings, insurance, and/or can incur cost for health care.208 Thus, health shocks for social assistance, households can be forced to engage working adults are usually a larger burden for households in adverse coping strategies. These can include selling than employment shocks without illness. productive assets, but also reducing food consumption or removing children from school, adverse coming Indonesia is in the process of rolling out strategies that cause long-term harm to human capital unemployment insurance. Unemployment spells formation. In addition to the immediate impact of the can create lasting periods of poverty for households. shock, such coping strategies can further undermine They can be mitigated if households have access future livelihoods, making longer spells of poverty to insurance. The existing severance pay system in more likely. Indonesia provided only very limited protection to workers. Contributory insurance schemes are less costly Progress in access to financial services has helped for governments and can avoid higher spending on social build resilience to shocks, but access is still not assistance. In the past, no adequate unemployment available to all. Financial inclusion can facilitate insurance existed in Indonesia, leading to a large number savings, insurance, and social assistance through of early withdrawals from the broader old-age savings making saving devices and access to payment systems system, JHT, mis-using it as de-facto unemployment available.207 In particular, financial inclusion can allow insurance, thus undermining it as a pension system.209 governments to use digital payment systems for direct, In early 2021, the Government issued regulations for faster, and cheaper payments to beneficiaries. Financial new unemployment insurance, Jaminan Kehilangan inclusion has improved, with account ownership Pekerjaan (JKP).210 It holds the promise of protection significantly increasing, especially for the bottom 40, against income and employment shocks, such as those from 11 percent in 2011 to 47 percent in 2021. Among all experienced during the COVID-19 pandemic. It includes adults, it reached comparable levels with the Philippines cash benefits, access to jobs, and market information, as but lagged Malaysia, China, and Thailand (Figure 4.2). well as job training (for a comprehensive overview, see In addition to a lack of money and family accounts, the Annex Table A5). However, it is currently only available to main reasons for not owning an account were distance salaried, usually formal, workers. to, and costs of, financial institutions, as well as necessary 208 Khelfaoui et al. 2022. documentation (Figure 4.3). The lack of an account 209 Holmemo et al. 2020. decreased the ability to gather emergency funds; only 210 The JKP was stipulated under the Omnibus Law on Job Creation. The implementing regulations are Government Regulation No. 37/2021 on JKP and Ministry of Manpower Regulation no 7/2021 on Contribution Recomposition World Bank 2022g. 207 for JKP. 56 Pathways Towards Economic Security Indonesia Poverty Assessment Shocks FIGURE 4.5: Share of working age adults reporting FIGURE 4.6: Facilities visited for outpatient and inpatient sickness in bottom 40 and top 60 care, for bottom 40 and top 60 60 20 50 15 40 Percent Percent 10 30 20 5 10 0 tu l l r ic ) al rs ita ita 0 BM ne us in on he sp sp /P io Cl UK Ot iti 2014 2015 2016 2017 2018 2019 2020 2021 2022 tit ho ho as ad l( ac m ca ic te Tr es Pr bl iva Lo sk Pu B40 Sick T60 Sick Pr Pu B40 Outpatient care T60 Outpatient care B40 Outpatient T60 Outpatient B40 Inpatient care T60 Inpatient care B40 Inpatient T60 Inpatient FIGURE 4.7: Share of household members with insurance FIGURE 4.8: Type of health insurance for bottom 40 and top 60 coverage for bottom 40 and top 60 80 60 70 50 60 Percent 50 40 Percent 40 30 30 20 20 10 10 0 2015 2016 2017 2018 2019 2020 2021 2022 0 B40 >= 80% T60 >= 80% None BPJS-PBI BPJS Private O ce B40 >= 20% T60 >= 20% Bottom 40 Top 60 Source: Authors’ calculations based on SUSENAS 2014 to 2022 Smaller health shocks were relatively frequent and often.211 Bottom 40 households were also more likely to mostly dealt with by outpatient care, but less so attend local health centers (Pusekesmas), while top 60 for bottom 40 households. Sickness of working age households more often frequented private hospitals, adults that disturbed daily activities in the last month especially for inpatient care (Figure 4.6), pointing to fluctuated between 40 and 50 percent of households potential differences in quality of care. from 2014 to 2021 (Figure 4.5). Households in the bottom 40 suffered slightly more frequently (2.7 percentage Health insurance has been becoming more common, points) than households in the top 60. Households covering now more than 65 percent of all Indonesians. used outpatient care to deal with these shocks, except Reforms in 2014 rationalized the legal framework and in 2017 and 2020/2021 when outpatient care dropped institutional arrangements of Indonesia’s single national below the frequency of sickness, possibly because of health insurance scheme Jaminan Kesehatan Nasional mobility restrictions, risk of infection, and healthcare (JKN). The insurance is open to all Indonesians and capacity limitations. Notably, bottom 40 households covers health service fees.212 The insurance BPJS costs 5 used outpatient care less often (3.8 percentage points) percent of monthly income or IDR 42k for non-salaried than top 60 households, despite their higher frequency 211 Based on SUSENAS 2015 to 2017 only, because of limitations around of sickness. More serious health shocks occurred in questionnaire comparability. about 10 to 15 percent of households, and are dealt Health Insurance, managed by BPJS Health for all Indonesians, includes 212 workers and non-workers, civil service and military. Employment, managed with by inpatient care. As for less serious health shocks, by BP Jamsostek (BPJS Ketenagakerjaan) is for all workers except military and civil service. JKK/JKM, JP, JHT, and JKP schemes are offered to salaried workers (those whose contributions are deducted and paid for by an employer). To date, bottom 40 households were more often subject to non-salaried workers, however, can choose to participate in JKK/JKM and JHT. Employment insurance is managed by ASABRI for military and by PT TASPENT serious health shocks, but sought inpatient care less for the civil service. Pathways Towards Economic Security Indonesia Poverty Assessment 57 Shocks and IDR 160k for non-workers. Since 2015, the number of Long-term parental health shocks lowered girls’ people covered by health insurance increased from 130 educational attainment. Parental health shocks affect million 48 percent) to 175 million (65 percent) in 2021.213 children’s schooling through four channels: (i) the However, many households continued to not have income shock affects the household’s ability to afford all household members insured, particularly among education for children; (ii) it can push children to join the bottom 40 households. Only 52 percent of them had labor force early, or to cover unpaid work, including care coverage for at least 80 percent of household members for household members; (iii) it reduces time parents in 2021, compared to 65 percent of top 60 households, have to raise their children; (iv) illnesses can have with the gap failing to close (Figure 4.7). For poorer psychological implications for any household member. households, BPJS-PBI covers the insurance costs (Figure Chronic illness of fathers has led to significantly lower 4.8). Additional work-accident (JKK) and death (JKM) educational levels of girls, but not boys, in Indonesia, insurances are also available, but covered only 15 and 12 driven by the first two channels.217 percent of the working age population. Climate change and natural disasters Despite health insurance, health shocks still force Adaptation households to employ “scarring” coping strategies. Indonesia is highly exposed to natural disasters, While health insurances protect again catastrophic with climate change increasing their frequency. health expenditures in case of illness, they do not The country’s location on the “ring of fire” creates high compensate for lost labor income. Illness significantly risks of earthquakes, volcanic eruptions, tsunamis, and reduced earning of self-employed households, while landslides. Between 1990-2021, Indonesia experienced the number of working hours remains unchanged. more than 300 natural disasters (Figure 4.9), including However, working hours of casual households in 200 flooding events affecting more than 11 million agriculture significantly increase when a family member people.218 Not surprisingly, Indonesia ranks third globally is sick, but without increasing earnings.214 Accordingly, in terms of natural hazard risk.219 With climate change, even households with health insurance decreased temperatures and precipitation are both expected to consumption by 1.2 percent.215 They often coped rise by 2050, and will vary by geographic location. This with the shock by reducing their food, non-food, and will lead to more frequent and more extreme natural education expenditures, as well as selling assets and disasters. Climate-related disasters already accounted for relying on increased remittances. Particularly, having around 70 percent of total disasters from 1990 to 2021. disabilities affected food consumption of the poorest.216 With more than 17,500 islands, Indonesia is also prone FIGURE 4.9: Number of natural disasters FIGURE 4.10: Estimated agricultural yields in 2030 30 2 25 0 -2 20 -4 15 -6 10 -8 5 e rs e ps s e es Ric lse aiz be an ro bl Pu M tu rc lc ta ga ge Oi nd Su 0 ve sa 1990 1993 1996 1999 2002 2005 2008 2011 2014 2017 2020 d ot an Ro ts ui Fr Wild re Storm Landslide Flood Drought Irrigated Rainfed Source: Authors’ compilation based on EM-DAT data Source: Authors’ compilation based on IFPRI data 213 Only includes BPJS Kesehatan, BPJS-PBI Kesehatan and Jamkesda; excludes Asabri, as it is not covered in the SUSENAS questionnaires. 214 Santoso and Sriyana 2021. 217 S. S. Lim 2017. 215 Kolukuluri 2022. 218 World Bank 2022h. 216 Simeu and Mitra 2019. 219 World Bank 2019b. 58 Pathways Towards Economic Security Indonesia Poverty Assessment Shocks Box 4.1: Voices, formal and informal community mechanisms to mitigate shock impacts Formal and informal community mechanisms can support individual households affected by shocks. A women’s union, PEKKA, had established a loan and savings system based on small, tightly-knit affinity groups, which took responsibility for repayment of loans to individual members. When members suffered idiosyncratic shocks, the group usually provided small, unsecured loans to prevent the individual from resorting to the forced sale of clothing or basic household items. They also offered larger loans, secured by woven cloth, to enable the member to engage in weaving, trade, or other economic activities. When people suffered health shocks, they received adequate medical care at no or low cost in the acute treatment stage, although only limited support for long-term rehabilitation. In addition, people affected by idiosyncratic disasters often relied on gifts from extended family and community networks, including food or even cash, particularly for children’s education. Requests for “loans” (often gifts) to meet educational costs were considered most justified. With only a limited commercial market for agricultural produce, members commonly sharing food surpluses, with almost all households either receiving or providing free food, with a strong social obligation on richer individuals to support poorer relatives. These mechanisms were deemed necessary given the very limited provision of government safety nets. With the pandemic and natural disasters affecting the whole community, these mechanisms became dysfunctional. Natural disasters devastated agricultural production for a significant period, making it difficult to maintain formal and informal community mechanisms. With the wholesale destruction of assets belonging to PEKKA and its members, cash loans shrank almost to zero. In addition, informal mechanisms, such as food gifts, declined significantly as food production and distribution systems were disrupted. Households without gardens and weak family and social networks were most harmed. Kandida Abon “I became interested in weaving after I got involved with the PEKKA union [around 2012]. The facilitators convinced us that people from all around the world would buy the cloth at good prices. Outsiders often came to visit the center, and they almost always bought some cloth, at good prices. And PEKKA had good contacts to sell the cloth in Jakarta, too. The most important thing was that if a woman really needed some money, if a child was sick or needed money for school, she could use a piece of cloth that was finished or nearly finished to borrow money from the union. You paid it back when somebody bought the cloth.” “In 2012, PEKKA set up an informal, open-air school to teach young people how to weave, produce dyes, and make cloth. I’m proud to have been a teacher at the school. I don’t do it for the money, I do it because I want to do something useful. I feel proud to be able to contribute. If you help other people, then they also want to help you. Sometimes my students and the old people bring me corn and vegetables or a few eggs. If I need transport, people give me a lift on their motorbike.” “Since the disaster, I’ve been living in this shack on the land belonging to my husband’s family. There is no electricity, no telephone signal, and we have to either collect rain water or buy it from a truck for Rp 6000 per barrel. Rats are a terrible problem; they eat everything that isn’t stored in jars or boxes. My husband and I both work to grow corn and other crops for our own consumption. He hasn’t had any work that generates cash, our only source of income is my weaving. It’s hard to borrow money from the union now, because everyone is in the same position, everyone needs money. The members help each other as much as they can. If you grow more food than you can eat, you share it.” Source: Based on qualitative interviews conducted for the Poverty Assessment. to damage from sea-level rise. Without adaptation, the poor, households.221 Without resilience measures, total Indonesian population exposed to climate-related current agricultural yields are projected to drop dangers in 50 years could reach over 4.2 million people.220 significantly (Figure 4.10), while higher risk of crop failure will increase food price volatility. Second, poor The poor are disproportionally affected by climate households often live in areas prone to risks. Those change risks. Although climate change affects the in remote and fragile areas are reliant on natural whole population, the poor and economically insecure resources. In urban areas, the poor are more likely to are likely to carry a disproportionate burden for two live in densely built-up areas with limited capability reasons. First, higher variation in precipiation and to withstand natural hazards, along riverbanks temperature will particularly affect agriculture, which where flooding may happen or in areas particularly remained a key livelihood for many rural, and often exposed to air pollution. At the same time, poor and World Bank 2019b. 220 World Bank 2020b. 221 Pathways Towards Economic Security Indonesia Poverty Assessment 59 Shocks Box 4.2: Voices, impact of a health shock followed by a mudslide Olympius Reting “We’ve been living in this shelter for almost a year now, since the flash floods hit the village. It’s on our own family land, where we grow corn and vegetables. There’s no electricity, no regular water supplies, and only very patchy telephone signal. For water supplies, we collect rain water or carry the barrel to the main road. For lighting, we use some battery-charged lamps, if we need them. We cook over an open fire, using fire wood we collect in the garden.” “I’m almost seventy years old. I’m still strong and healthy, except for my legs. I fell out of a tree collecting lontar palm leaves some years ago, and I’ve had bad pain and trouble doing hard work ever since. Before that, I often worked as a builder, doing construction work in the village, or in the district capital. For a few years, I worked in Makassar, doing construction work there. When I came back to the village with savings, I bought a few goats, to raise them and breed them. I bought the first two more than 30 years ago, and I’d grown the flock to forty by the time the mudslides hit.” “The mudslides hit on Easter Sunday. It wasn’t until the next morning that I heard the news, that the rains had caused landslides and flash floods that had killed hundreds of people and destroyed almost the entire village. Our house had been completely destroyed, nothing left at all. Apart from our house, my wife lost all her cloth and her weaving equipment, and I lost forty goats that had been locked in a pen behind the house. Apart from the motorbike, my phone, and a few goats that were grazing elsewhere, all we had left was the clothes on our backs.” “In the first few weeks, we received a lot of assistance, some from the government, but most from NGOs and community groups. The government set up emergency medical posts near the fields around here, where many people were seeking refuge. They made it easy to replace identity cards and other documents. Later, the cooperatives agency provided my wife with thread and dyes and help to buy weaving equipment, so she could start weaving again. They also provided help for us to build a toilet here. But all the assistance has dried up now. We’re completely dependent on the food we grow, my wife’s weaving, and gifts and assistance from family members.” Source: Based on qualitative interviews conducted for the Poverty Assessment. economically insecure households have less resources In Sulawesi, for example, the share of children attending to protect assets against shocks and less savings to school dropped from 90 to 2 percent in the immediate recovery.222 aftermath of the disaster.225 Beyond disruption to livelihoods, destruction of Many poor and economically insecure households in physical and non-physical assets that can accompany Indonesia cope with covariate shocks by reducing food natural disasters also disproportionately harm the and non-food consumption, threatening to worsen poor and economically insecure. Natural disasters lead childhood malnutrition. Shocks can reduce long-term to direct loss of physical assets, including productive productivity if households respond by reducing human assets and business capital. In areas affected by the capital investments; for instance, by reducing nutrition, September 2018 earthquake in central Sulawesi, while taking children out of school to work, postponing or nearly all affected households lost assets directly as a neglecting health needs, or liquidating savings and result of the earthquake,223 over one in five households assets.226 This lowers the chances of securely escaping from the bottom 40 percent were still in temporary poverty in the long run. In the wake of the Sulawesi housing seven months later, compared to 13 percent of earthquake, poor and economically insecure households the top 20 percent.224 Covariate shocks threaten intangile in disaster-affected areas were much more like to cope assets as well, such as human capital, which poor and by adopting adverse coping strategies (Figure 4.11). economically insecure have less of even before crises hit. 222 Winsemius et al. 2018. 223 Assets included land, houses, livestock, vehicles, electronics, cash, savings, and gold/jewelry. 224 Authors’ calculations using World Bank Welfare Tracking in the Aftermath of World Bank 2021c. 225 Disaster (WelTrAC) Survey, Wave 1. Temporary housing included tents and See for example Alderman, Hoddinott, and Kinsey 2006; Klasen and Waibel 2013; 226 temporary shelters. Gubert and Robilliard 2007; Rosenzweig and Binswanger 1993. 60 Pathways Towards Economic Security Indonesia Poverty Assessment Shocks FIGURE 4.11: Share of households adopting adverse leads to inefficient use, contributing to inefficient strategies to cope with crises transport systems and urbanization patterns that further 100 increase high energy demand. 80 60 In 2021, the GOI committed to a substantial reduction 40 of GHG emissions. As part of the United Nations Climate 20 Change Conference (COP26) in 2021, the Government 0 introduced new commitments to lower carbon and its M40 energy transition, updating its Nationally Determined 7 mo. 2 years* 3 mo. 8 mo. 19 mo. Contribution (NDC), including a substantial reduction of Sulawesi earthquake COVID-19 pandemic GHG emissions. Indonesia’s Enhanced NDC, released in (October 2018) (March 2020) September 2022, sets out an unconditional 31.9 percent Source: Authors’ calculations using World Bank Welfare Tracking in the Aftermath of reduction in emissions against business-as-usual (BAU) Disaster (WelTrAC) Survey, Waves 1 and X, and World Bank Indonesia High-Frequency (HiFy) Phone Survey of Monitoring COVID-19 Impacts Rounds 1,3 and 5 projections by 2030, and up to a 43.2 percent reduction Note: B40 = bottom 40 percent of households; M40 = middle 40 percent; T20 = top 20 percent. Mo = months. conditional on international support. It also committed to reach net-zero emission by 2060. Three important Although covariate shocks were not the main driver of contributors to these goals are: (i) implementation of the economic insecurity, resilience and social protection Forestry and Other Land Use (FOLU) Net Sink 2030 policy measures remain essential to mitigate harm from to make forests a net carbon sink by 2030 through forests natural disasters. Idiosyncratic shocks dominating and peatlands restoration and deforestation avoidance; economic insecurity. With the increase in frequency (ii) reducing the pipeline of new coal plants to 3.9 GW; and severity of natural disasters, however, economic and (iii) introducing a carbon tax of US$ 2.10 per ton insecurity due to covariate shocks is likely to increase combined with an emissions trading scheme for coal if mitigation measures are not implemented. Currently, power plants. Indonesia‘s risk remains high, ranking 100 of 191 countries in climate change resilience, compared to Phasing out coal will particularly affect coal-producing Thailand (67th) and China (62nd).227 communities. Indonesia is the world’s largest coal exporter, with coal exports representing 2 percent Mitigation of GDP, or 13 percent of total goods exports. Coal Alongside Indonesia’s development over the last mining workers represented 0.2 percent of total formal decades, greenhouse gas emissions rose sharply, employees in 2018. However, a larger number of people making Indonesia the seventh biggest emitter in work informally in coal. With coal mining generally the world.228 In the past quarter century, Indonesia concentrated in specific areas and communities, has experienced an important development transition phasing-out coal will decrease some employment in making important infrastructure, such as access directly, but also indirectly threaten firms depending on to electricity, almost universally available. However, the coal mining business and their workers. More detailed this transition contributed to Greenhouse Gas (GHG) analysis will be needed to better understand potential emissions. Between 1990 and 2018, Indonesia’s coal- effects on reducing coal mining for communities and to powered capacity almost doubled while emissions design mitigation measures. increased by 140 percent. The high supply of, and domestic and international demand for, carbon-intensive Two scenarios are used to estimate the long-term resources contributed to the country’s emissions distributional impacts of climate mitigation policies, a intensity. Their often low price, especially domestically, “low-ambition” model and a “medium-ambition” model 227 The ND-GAIN country index score provides a summary of a country’s (Box 4.1). The two scenarios model gradual levels of vulnerability to climate change and other global challenges combined with its readiness to improve resilience. ambition to assess costs and benefits of decarbonization 228 OECD.Stat GHG excluding LULUCF for latest available year. Pathways Towards Economic Security Indonesia Poverty Assessment 61 Shocks FIGURE 4.12: Contributions to consumption growth for low- FIGURE 4.13: Contributions to consumption growth for medium- ambition scenario, by income decile ambition scenario, by income decile 10 10 8 8 6 6 Percent Percent 4 4 2 2 0 0 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 -2 -2 Purchasing power Social assistance Labour income Source: Box 4.3: Modeling parameters for distributional climate impact simulations The specific parameters for the two climate mitigation scenarios are: (i) Low-ambition: Removal of electricity and fuel subsidies with savings used for investment. No land or energy policies are assumed. (ii) Medium-ambition: Nationally Determined Contribution (NDC) adds land and energy policies and a carbon tax to the low-ambition scenario. Land policies includes an NDC-consistent energy sector plan for 2021–2030, coal phaseout (plants retiring about 8-12 years earlier), and decarbonization through a cap on emissions to drive a 70 percent power sector reduction. Energy policies include peatland restoration, extended forest and peatland moratoria, and a land-based emissions tax with redistribution (US$ 5/tCO2eq). The carbon tax will reach US$40/tCO2 by 2040 applied to all sectors and GHG emissions except for agriculture. Revenues from the carbon tax are proportionally allocated across public expenditures, including for social protection spending and investment in low-carbon equipment. It is assumed that replacing stranded fossil fuel assets accounts for 25 percent of the new investment. The model outlines economic and social impacts of climate mitigation in three stages: • Stage one: two separate land and energy models are used to assess the effects of sector-specific policies on sector emissions and other outcomes. • Stage two: land and energy policies are brought together and complemented with a range of fiscal policies—such as elimination of fossil fuel subsidies and carbon taxes—to estimate economy-wide effects using a Computable General Equilibrium (CGE) model. • Stage three: uses the outputs of the macro-CGE model as inputs for a micro-economic simulation model to assess household level effects from climate mitigation. The reference is a Business-As-Usual (BAU) case in which only currently enacted climate policies are included without any new policies. Over the time horizon of the CGE, certain parameters are adjusted. The population parameters are adjusted based on UN population growth projections, with education levels adjusted based on the aging of the youngest cohorts. The CGE models wages separately for low-skilled and high-skilled workers across sectors. The micro-economic simulations use an occupational choice model to re-allocate workers based on the outputs of the CGE model. Income is transformed into consumption based on the marginal propensity to consume. Consumption shares are kept constant assuming that households do not adapt their behavior in response to changes in prices. until 2040, based on a general-equilibrium model linked (consistent with an 80 percent reduction of energy with micro-economic simulations.229 The low-ambition sector emissions by 2040) as well as a moderate carbon scenario removes fuel subsidies, while the medium- tax reaching 40 US$/t, but excluding the agricultural ambition scenario adds land and energy policies sector. Pape et al. 2023. 229 62 Pathways Towards Economic Security Indonesia Poverty Assessment Shocks Poor households will gain in absolute and relative Prices terms if agriculture is excluded from the carbon tax Russia’s invasion of Ukraine has triggered an evolving and government uses the tax revenue to boost social price shock that is decreasing households’ purchasing assistance. The agricultural sector will expand, given power and exacerbating poverty. Russia’s invasion of its relatively lower cost stemming from the carbon Ukraine contributed to supply shortages, generating tax exclusion. Accordingly, agricultural labor incomes significant price pressures. Indonesia fared relatively will increase, which should benefit the poor more as well given its own exports, being able to take advantage they are more often engaged in agriculture (Figure of improved terms of trade.231 However, consumers 4.14 and Figure 4.15). This will reduce relative prices were still suffered price increases. Food, for example, of food (ranging from 1 to 4 percent across scenarios) increased by 9.3 percent year-on-year in July 2022, but countered by an increase in energy prices (ranging almost double general inflation (Figure 4.14). In ceteris from 16 to 165 percent). Overall purchasing power paribus, consumers in the bottom 40 lost 4.7 percent in will decline but will remain relatively constant for all purchasing power, while the top 20 lost only 3.7 percent deciles. In the low-ambition scenario, increased labor (Figure 4.15). The combined price shock for food, fuel, and incomes and reduced purchasing power cancel out, electricity, as well as transportation can, if not mitigated, muting overall consumption and poverty impacts. For push an additional 6.6 million Indonesians (or 2.4 percent the medium-ambition scenario, however, labor incomes of the population) into poverty (Figure 4.16).232 significantly exceed the decrease of purchasing power. In addition, social assistance will increase substantially Price subsidies can mitigate poverty but are due to higher government revenues, assuming they expensive; targeted social assistance is vastly more are spent proportionally across public expenditures. efficient and progressive. A theoretical food subsidy to Even without increased labor incomes (for example, fully compensate for the increase in food prices would if agriculture were not excluded from the carbon tax), have cost 2.2 percent of GDP. While this would eliminate social assistance in the medium-ambition scenarios all negative poverty effects, it is unrealistically expensive. would be sufficient to compensate for consumption However, targeted social assistance would cost only 0.5 losses for the bottom 40. Accordingly, the low-ambition percent of GDP and be able to fully compensate the scenario is largely neutral, while the medium-ambition bottom 40. At the same time, it would also be strongly scenario will slightly reduce poverty.230 progressive, reducing the Gini by a significant 1.4 FIGURE 4.14: Year-on-year inflation for 2022, by item group FIGURE 4.15: Loss of purchasing power due to shock in prices, as in Scenario A 10 7 6 8 5 Percent 6 4 Percent 3 4 2 2 1 0 0 Jan Feb Mar Apr May Jun Jul Aug 1 2 3 4 5 6 7 8 9 10 Food, drinks, tobacco Transportation Food Household energy price (Fuel and Electricity) Electricity + fuel Overall Transportation All three categories Source: BPS Source: SUSENAS 2021 231 World Bank 2022e. 232 Results are based on partial equilibrium simulations, which do not incorporate At US$ 5.50 PPP 2011, poverty will drop by 0.5 percentage points in 2040. Poverty 230 transmission into wages. However, indirect effects as well as behavioral effects at US$ 1.90 and US$ 3.20 will already be indistinguishable from zero in 2040. are considered. Pathways Towards Economic Security Indonesia Poverty Assessment 63 Shocks FIGURE 4.16: Unmitigated poverty impact (in pp) FIGURE 4.17: Change of consumption after compensating for of price shocks price shocks, by decile, for Scenario A 10.0 20 8.0 15 10 6.0 5 4.0 0 2.0 -5 0.0 -10 Scenario A Scenario B 1 2 3 4 5 6 7 8 9 10 Food Energy Transportation Loss in Consumption Compensation Net Impact Note: Scenario A assumes price increases as observed in July 2022 for food (9.3 percent), Source: SUSENAS 2021 energy (4.9 percent) and transportation (6.6 percent). Scenario B assumes a 30 percent increase for the same groups. Based on SUSENAS 2021. points (while the food subsidy would have no impact FIGURE 4.18: Google mobility and Oxford stringency index as well as number of new COVID-19 cases on the Gini).233 However, removing price subsidies is Mobility Index Stringency Index complicated by political economy issues. R1R2 R3 R4 R5 R6 R7 40 100 Stringency Index (2) Mobility Index (1) 20 80 COVID-19 0 60 -20 40 The COVID-19 pandemic, another large covariate -40 20 shock, pushed the Indonesian economy into recession -60 0 60000 for the first time since the Asian Financial Crisis, before Daily cases (3) 40000 rebounding in 2021. The COVID-19 pandemic forced 20000 governments around the world to restrict mobility. 0 Indonesia introduced restrictions in March 2020, followed Jan20 Apr20 Jul20 Oct20 Jan21 Apr21 Jul21 Oct21 Jan22 Apr22 Jul22 by several periods of gradual relaxations and reversals, Daily cases especially with the arrival of the COVID-19 delta wave Source: Google Mobility Data; Oxford COVID-19 Government Response Tracker; COVID-19 Data Repository by Johns Hopkins University (Figure 4.18). Since early 2022, the number of new cases Note: Baseline value for Google Mobility Data is the 5-week period from January 3rd to February 6th 2020. The mobility index represents the average change in time spent outside remained low while mobility restrictions continuously home. Trendline is using the Lowess function. Daily cases smoothed by 7-day average. R1 to R7 indicate the rounds of the HiFy survey. eased. The economy suffered through both domestic and Labor incomes external channels. Domestically, mobility restrictions and With the pandemic onset, workers—especially in social distancing depressed economic activity. Externally, urban areas—lost their jobs in the first year of the trade for goods and services such as tourism decreased pandemic, while others had to work less hours and as did investment flows.234 Accordingly, Indonesia earned less. In the first six months of the pandemic, experienced its deepest contraction in two decades more than 5 million Indonesians (2.5 percent of the during the second quarter of 2020. In 2021, the economy working age population) lost their jobs, and 24 million rebounded, albeit more slowly than expected because workers (11.8 percent of the working age population) of the delta wave. Relaxed mobility restrictions, as well as had to work reduced hours, with urban areas being most stronger external demand and higher commodity prices, affected. The unemployment rate rose by 1.8 percentage buoyed exports and manufacturing activity.235 points to 7.1 percent in the third quarter of 2020 (compared to the previous year), mainly driven by urban unemployment increasing from 6.3 to 9 percent, while 233 The simulation assumes PMT targeting. 234 World Bank 2020g. rural unemployment only increased from 4 to 4.7 percent. 235 World Bank 2021d. 64 Pathways Towards Economic Security Indonesia Poverty Assessment Shocks Underemployment increased by 7.3 percentage points in 2020—substantially more than the previous five- to 28.9 percent. COVID-19 exacerbated the intensity of year average of 1.1 million women. At the same time, under-employment, reducing average working hours by 330,000 men left the labor force, in stark contrast to a 0.5 hours and earnings by about 5 percent. usual 1.3 million men per year entering the labor force. Jobs became less secure, with 4.5 million formal jobs For 3.4 million new labor market entrants, the vanishing, while 6.4 million informal jobs were created economy added only 1.9 million jobs in 2020, while (Figure 4.19). Some of them were in the digital economy jobs became less secure. To compensate for losses as gig workers.236 However, 4 million of the new jobs were in household livelihoods and the loss of jobs for men, unpaid family work (Figure 4.20) and often in agriculture, women entering the labor force increased by 2.8 million mostly filled by237women (4.21). Net job losses among Box 4.4: Voices, residents in Kediri on COVID-19 effects Zooming into the lives of residents in the town Kediri reveals the devastating effects of COVID-19. Prior to the pandemic, many residents in Kediri were involved in commerce, often through traditional markets. The two main challenges were regulations closing the markets or limiting their operations and reduced demand for their goods and services. Often, they continued to operate until they had depleted their accumulated capital, forcing them to abandon their livelihood. Providers of face-to-face services were particularly affected, such as fitness club operators and laundries. Food vendors lost their businesses when restrictions curtailed activities of their customers; for instance, vendors catering to students or travelers. Workers in transport services faced almost complete collapse in demand. Some domestic workers lost their jobs due to employers’ fears of becoming infected. Coping became a challenge, affecting incomes, food consumption, and education. Most residents responded by engaging in alternative but lower-paying activities, including agricultural labor, subsistence farming, petty trade, or foraging in the peri-urban area. In many cases, their lack of experience in these activities lowered wages and yields, in already generally unrewarding areas. While some residents received government assistance, it was often very limited, delivered only for short periods, and with opaque eligibility requirements. Many reduced expenditures on food, often consuming only rice and vegetables and sometimes skipping meals. Effective, large-scale community initiatives did not exist to help, but individual acts of charity or cooperation were evident. Children’s participation in online education was fitful and patchy, with a lack of motivation and limited feedback and interaction with teachers, and parents’ limited ability to help. Expenditure on internet access became a severe constraint, sometimes consuming up to one-quarter of a family’s income. Umi Kalsum Saban “When I was around twenty, I went to Malaysia as a domestic worker, to care for my boss’s children. I sent all my earnings back to support my family. When I got back, I didn’t have any savings. My parents used most of the money to rebuild our house. A year after I got back, I got married to my husband. After we’d had two children, he left to find work outside the island. He has never sent any money and I never heard any news from him.” “To support myself and my children and parents, I set up a kiosk selling basic goods (sembako), cigarettes and sweets. I took out a small loan from BRI. My parents have a clear ownership title (hak milik)237 to the land around our house, so I could use that as collateral for the loan. The kiosk was well located, on a main inter-village road, in a busy hamlet, so I could make enough to pay off the loan and support my family.” Throughout the pandemic, the only form of assistance that I received from the government was through Bantuan Langsung Tunai. For more than a year, we received Rp 300 thousand a month (from Village Funds). There were often delays, so maybe we only got it after a three-month wait (with accumulated back pay). The government also provided some building materials and other assistance to people living in emergency housing after the floods and the eruption, but most of the aid was from NGOs and big organizations, not the government.” “Life is still very difficult. I still keep the kiosk running, but my sales are about a quarter of what they used to be, and I still owe money for the loans I took. About three-quarters of the people in this area haven’t rebuilt their homes or returned to their village. They are either still living in emergency shelters in the fields, or with family members elsewhere.” World Bank forthcoming. 236 Holding a hak milik title to land is somewhat unusual. Most farm land is classed 237 as tanah ulayat, or land to which families have customary rights. This land can’t usually be bought or sold, or used as collateral. Pathways Towards Economic Security Indonesia Poverty Assessment 65 Shocks men were particularly prevalent in manufacturing, while in school can diminish learning, with life-long while service jobs became informal.238 Overall, COVID-19 consequences. Less work opportunities for youth once increased women’s employment in 2020, compensating out of school delays crucial work experience, and can for employment losses by men. trap workers in low-productivity, often informal, jobs.240 COVID-19 also affected youth by leading to school In 2021, the labor market rebounded, with better drop-outs and part-time work, while diminishing work jobs returning. With relaxed mobility restrictions opportunities for older youth.239 Younger youth (aged and COVID-19 better controlled, economies around 15 to 18) were more likely due to COVID-19 to work the world rebounded, as did Indonesia. In the labor while being in school, more often above 20 hours per market, some of the less secure jobs in agriculture week. In addition, the pandemic increased the number disappeared, with additional, mostly formal, jobs of younger youth neither in school nor employed returning in manufacturing as well as low-VA services (NEET). For the older cohort of youth aged 19 to 24, the (Figure 4.19). However, many unpaid jobs remained in pandemic similarly led to school drop-outs, but led to the economy (Figure 4.20). The unemployment rate less work opportunities with a larger share being NEET. reversed from 7.1 percent in 2020 to 6.5 percent in 2021. Early drop-outs and long hours in partial employment Nevertheless, frictions remain in the labor market. Spells FIGURE 4.19: Number of added/lost workers since previous year, FIGURE 4.20: Number of added/lost workers since previous year, by sector and informal/formal by type of employment 4 6 3 4 2 2 Millions 1 Millions 0 0 Own account Employer+temp Employer+perm. Employee Casual Ag Casual non-Ag. Unpaid family worker -2 Other Other Agriculture Manufacturing Agriculture Manufacturing Low-VA Services High-VA Services Low-VA Services High-VA Services -1 -2 -4 -6 2020 2021 Sector Informal Formal 2020 2021 FIGURE 4.21: Number of added/lost workers since previous year, FIGURE 4.22: Likelihood of employment in August 2020, after by gender and informal/formal controlling for individual characteristics 4 4 Post-COVID-19 impacts (p.p.) 2 2 0 Millions 0 -2 Women Men Women Men 2020 2021 -4 -2 -6 All 19-29 30-44 45-54 55-64 (15-64) -4 Age groups Informal Formal Male Female Source: Authors’ calculations based on SAKERNAS; except for Figure 4.22: Halim, Hambali, Note: Each dot represents the coefficient of the COVID-19 dummy variable, which is 1 and Purnamasari 2022 for August 2020. Vertical lines indicate 90% confidence intervals. Regressions control for individual characteristics. 238 Putra, Ovsiannikov, and Kotani 2022. 239 Halim, Hambali, and Purnamasari 2022. 240 Duryea, Lam, and Levison 2007; Naidoo, Packard, and Auwalin 2015. 66 Pathways Towards Economic Security Indonesia Poverty Assessment Shocks of unemployment last longer, while fresh labor market assistance programs. Most notable among the measures entrants face difficulties finding jobs.241 introduced was a substantial short-term increase in the target number of beneficiaries as well as the benefit Despite the rebound, poorer and urban households level offered under PKH and Program Sembako. A new suffer lingering negative employment and income unconditional cash transfer program was also introduced effects from COVID-19. While negative employment to cover approximately nine million additional and income effects were widespread, those already households not previously eligible for PKH or Sembako. economically insecure prior to the pandemic were Rollout of some of these measures (for example, BLT hit harder. Among primary breadwinners, household Dana Desa), however, required manual enrollment at heads with lower education were initially more likely to the local level to reach affected households for which experience work stoppages.242 Those and households information was not available in DTKS. This reflects the with young children more often reported an income somewhat static nature of Indonesia’s social protection shock. Two years into the pandemic, these and delivery systems, a challenge many countries faced in households in the bottom 40 percent were recovering responding to the pandemic crisis, which often delayed more slowly than other households. The lack of resilience delivery of assistance.243 of bottom 40 households exacerbated pandemic negative effects and pre-existing inequalities. COVID-19 response scale-up managed to meet the announced numerical targets by 2021, and helped Mitigation measures reduce food insecurity. The Government spent almost The Government quickly scaled-up social assistance all funds allocated for the social assistance components in response to the COVID-19 pandemic. To help of the pandemic response both in 2020 and 2021, with households cope with employment and income budget realization rates for most programs exceeding 90 shocks during COVID-19, the GOI quickly launched percent. By 2021, most assistance programs, including in 2020 an impressive array of social assistance, jobs/ Sembako and PKH, had met announced numerical targets. skills, and social insurance measures, as well as support Receiving PKH and Sembako reduced the probability of for firms, for example, through the Pemulihan Ekonomi households assessing themselves as experiencing food Nasional (PEN) Program. Expenditure on social assistance shortages or eating less due to insufficient resources. The expanded to 1.6 percent of GDP in 2020 and 1.5 percent differential effect was stronger earlier on in the pandemic in 2021, essentially doubling spending on core social but declined in magnitude as the crisis eased in 2021.244 FIGURE 4.23: Share of social protection beneficiaries in FIGURE 4.24: Share of program beneficiaries assessing March 2021 benefits as adequate 50 40 Pre-pandemic 30 20 10 Last three months 0 Bottom 40% Middle 40% Top 20% Bottom 40% Middle 40% Top 20% Bottom 40% Middle 40% Top 20% 0 20 40 60 80 100 Percent Sembako Card PKH Bansos Tunai Not at all Partially Mostly Completely Note: Sembako is targeted at the bottom 30, PKH at the bottom 20 and Bansos Tunai at the 4th decile. Share of households that ever received any benefit since the onset of the pandemic until March 2021. Halim, Hambali, and Purnamasari 2022. 241 World Bank 2020g. 243 Sari, Purnamasari, and Febriady 2022. 242 Sjahrir and Wibisono 2022. 244 Pathways Towards Economic Security Indonesia Poverty Assessment 67 Shocks Box 4.5: Voices, receiving COVID social assistance Most formal assistance was received in urban areas. Around half of the informants in around Kediri received some form of financial government assistance, usually PKH if the household had school-aged children. Many also received food handouts. However, none of the informants in remote rural areas had received these forms of assistance. The only exception was immediate post-disaster relief programs in some areas when food assistance was provided to people in emergency housing. However, selection criteria remained unclear. No household was able to explain the factors determining their selection or rejection. Government officials usually provided them with a list of conditions that were necessary conditions but not sufficient. If people met the criteria, they might be selected. All possibly eligible households felt that it was natural and normal for decisions to be arbitrary and based on opaque considerations. Many believed that if one had been selected for a program in the past, it was unlikely that they would be selected again soon, regardless of individual program eligibility conditions. The explanation was that this would be “unfair,” and that someone who had never received assistance should benefit instead. Single women, even though among the poorest community members, hardly received assistance. Single women are particularly poor. Nevertheless, women rarely received assistance unless they were married to, or the daughter of, an eligible man, who received it on their behalf. Women heads of households often do not have their own Family Card (Kartu Keluarga), a document representing a vital requirement to receive almost all forms of government assistance. Instead, they are listed as “dependents” of a male relative, either a father or a brother, or perhaps a long-absent husband—even when these relatives do not provide any support. While various government initiatives have been conducted to address this, officials still find it difficult to accept that a woman household head can be eligible, with women themselves often thus reluctant even to try to apply. Titien “Our family receives assistance from the PKH program. It’s about Rp 400,000 per month, paid once every three months. But it’s often late, and there are sometimes problems. If the names on the forms aren’t exactly the same as on the birth certificates and other documents, the payment is delayed. But sometimes the names are spelled differently on different documents, which makes it confusing. I get Rp 250,000 for my boy, and Rp 150,000 for his sister. We use all the money for our girl’s school and education costs. As part of the PKH program, we also receive a food package once a month, usually with rice, oil, sometimes some eggs. This month, there were some bananas. It varies.” However, Indonesia’s targeting system could have sources of support to meet daily needs during the been more effective in response to the shock. Even pandemic. Among the bottom 40 percent of households, among the bottom 40, less than 40 percent received for example, nearly half coped by either taking loans or benefits for each of the expanded programs (Figure receiving assistance from friends, family, or relatives. 4.23). Also, wealthier households received benefits despite not being targeting, a result of inclusion errors Poverty and infrequent DTKS updating. The DTKS registry is not COVID-19 decreased consumption of the better- designed to allow targeting in response to a disaster and off—but not the richest—households most severely, excludes most economically insecure households, which especially in urban areas. Consumption growth less for could become eligible in response to a shock such as the urban population at 0.7 percent, given the larger COVID-19. COVID-19 effects, compared to the rural population at 2.6 percent (Figure 4.25). In urban areas, the median up Also, adequacy of benefits was low relative to to the top 20 percent households lost income, given beneficiaries’ needs during the pandemic. While self- their higher reliance on formal jobs, most affected reported benefit levels of social protection programs by COVID-19. The poorer and wealthiest households matched the official amounts announced, less than half were better insulated against the shock, with the poor of program beneficiaries self-assessed the benefits of more often engaged in agriculture and the wealthiest current programs as adequate (Figure 4.24). Consistent being able to smooth consumption. In rural areas, a with this, many Indonesians relied on non-government similar pattern emerged but with consumption losses 68 Pathways Towards Economic Security Indonesia Poverty Assessment Shocks FIGURE 4.25: Consumption growth Incidence curve for FIGURE 4.26: Ravallion-Huppi decomposition for 2020 to 2021 2020 to 2021, by urban/rural 6 20 0 4 Percent -20 Percent 2 -40 0 1 7 13 19 25 31 37 43 49 55 61 67 73 79 85 91 97 -60 Rural Urban Agriculture Manufacturing -2 Low-VA Services High-VA Services Urban Rural Other Population-shift e ect Source: SUSENAS March 2020 and 2021 Source: SUSENAS March 2020 and 2021 Note: 5-point moving average. Dotted lines show overall averages. Note: Annualized coefficients, omitting households not working. Given the timing of the survey, the 2020 March round captures the pre-COVID situation. concentrated at better-off, though still not the richest, 24 percent of children under age 5 were stunted in 2021, households, given the large share of the population with long-term impacts on health and education. engaged in agriculture. FIGURE 4.27: Coping strategies of households in the bottom 40 and top 20 Poverty reduction was a almost completely a result of 100 a poverty decrease in rural agricultural households. Poverty reduction continued to drop from 20 percent 80 in 2019 to 18 percent 2021, albeit at a smaller pace. 60 Poverty reduction was completely a result of rural areas 40 dropping from 22 percent in 2019 to 19 percent in 2021, 20 while urban poverty slightly increased from 18 percent to 19 percent, reflecting the stronger negative COVID-19 0 Jun-20 Nov-20 Oct-21 Jun-20 Nov-20 Oct-21 Jun-20 Nov-20 Oct-21 effects on urban areas. Agricultural households in rural areas contributed the most to poverty reduction, while Scarring Short-term xes Social capital households shifting between sectors contributed to B40 T20 poverty in both urban and rural areas (Figure 4.26). Source: HiFy Rounds 2, 4 and 6 Note: Scarring strategies including reducing food consumptions, selling assets, withdraw children from school, and stop medication; Short-term fixes includes taking loans, delayed payments, and credited purchases; Social capital includes taking loans or receiving The bottom 40 used more adverse coping strategies assistance from friends, family or relatives. to cope with COVID-19. Households had to cope with large, short-term income shocks. While some of Learning losses 245 the wealthiest households used savings, this strategy Between 2020 and 2021, at the height of the COVID-19 remained unavailable to most households. While almost pandemic, schools nationwide closed for face-to-face all households avoided sale of productive assets, 73 learning. The Government closed all schools in March percent of households in the bottom 40 used scarring 2020, but soon after changed to a more decentralized coping strategies at the onset of the shock, compared approach allowing green zones to reopen schools. to 55 percent of households in the top 20 (Figure 4.27). In November 2020, less than one-third of students Among the scarring coping strategies, reduced food attended face-to-face learning. Given the decentralized consumption was most used (above 95 percent). Food approach, and a larger number of COVID-19 infections insecurity is particularly problematic if it affects children. in urban areas, students in DKI Jakrta were most affected Depending on the within-household distribution of the by school closures, with only 10 percent attending face- burden, children can be affected. Even before COVID-19, 245 This section draws from R.S. Purnamasari et al. 2022. Pathways Towards Economic Security Indonesia Poverty Assessment 69 Shocks to-face larning (Figure 4.28). The situation rebounded The Government enacted emergency measures to significantly only by April 2022 when over 90 percent of mitigate potential learning losses, including the students attended face-to-face learning. provision of free internet credit.247 The Government introduced an emergency curriculum and supported Digital technology mitigated learning deficits from different modes of distance learning. Most notably, it has school closures, but only when teachers were provided free internet credit to students since August involved and only for households with internet 2020. In November 2020, 51 percent of students received access and digital skills, exacerbating inequities. the credit. However, access to the internet credit was not A large share of schooling transitioned to distance equal, with students outside Java and in the bottom learning, creating challenges for teachers to adapt new 40less likely to receive them. pedogogical concepts and using digital technology.246 Also students and parents had to use their own Nevertheless, substantial learning losses accrued resources, limited the reach of distance learning. In and exacerbated learning inequalities. COVID-19 is November 2020, only 38 percent of students in the estimated to have lowered educational attainment by bottom 40 were able to use mobile learning apps or 0.9 to 1.2 learning adjusted years, even after taking into online schooling, compared to 66 percent of the top 20 account Government’s mitigation measures. Reading (Figure 4.29). Rural students did not suffer significantly scores are estimated to drop from 371 in 2018 to about less access to online learning compared to urban 340.248 These learning losses translate to an estiamed 7 areas. Even among students managing to access to 10 percent of losses in annual earnings per individual. distance learning, three in four expressed challenges, Furthermore, the pre-existing gap of 57 PISA points including internet access and problems with focusing between the bottom 20 and the top 20 is estimated to have and concentrating. To improve resilience for future increased to 64 points.249 Simulations using global data shocks, schools, teachers, and students all need to suggest even longer-term negative effects due to reduced be better digitally equipped. A digitalization strategy inter-gnerational educational mobility.250 Targeted actions can include a national real-time platform to monitor are needed to recover learning losses. An assessment of school conditions, but also track student learning to children’s learning level can inform teacher instructions, facilitate a student-centered, interdisciplinary, project- while teachers might require support to develop based, and collaborative focus. personalized catch-up strategies and more individualized learning to help students overcome learning gaps. FIGURE 4.28: Share of students attending face-to-face FIGURE 4.29: Share of students using mobile learning apps or learning online schooling in November 2020 100 100 90 80 80 70 60 60 50 40 40 30 20 20 0 10 0 Primary Jr Secondary Sr Secondary DKI Jakarta Java (Non-DKI) O -Java Rural Urban Bottom 40 Middle 40 Top 20 Primary Jr Secondary Sr Secondary DKI Jakarta Java (Non-DKI) O -Java Rural Urban Bottom 40 Middle 40 Top 20 Nov-20 Apr-22 Source: HiFy Round 6 and 7. Source: HiFy Round 6. 247 R.S. Purnamasari et al. 2022. 248 Afkar and Yarrow 2021. 249 Yarrow, Masood, and Afkar 2020. D. Lim et al. 2022 246 250 Azevedo et al. 2022. 70 Pathways Towards Economic Security Indonesia Poverty Assessment Shocks Box 4.6: Voices, challenges of online learning Titien “My girl is in the first class of senior high school. For two years, she’s mostly been doing online classes. She’s back at school a few days each week now. I’m glad. I wasn’t happy about her doing her school online, at home all the time. I don’t think she can learn anything. There is no feedback, the teacher doesn’t check if she really understands. The teacher just gives her assignments, then gives her a grade. I’m worried that she’s falling behind. “She has her own mobile phone for school. We bought a cheap smartphone for her when the restrictions came in. It cost less than a million rupiah. She knows that the family doesn’t have much money, so she only uses it for school work, she doesn’t spend time on Facebook or playing games. Every time I buy internet credits, it costs Rp 68,000. That’s enough for about two weeks, maybe a bit more. I can’t really help her with her school work, it’s too advanced for me. When I was young, I really wanted to finish school, but my parents couldn’t afford it.” “For me, education for my children is the most important thing. It’s the only way that they’ll be able to get a good job. But I worry that these days, just graduating from high school isn’t enough. When I was younger, you could get a good job with a high school certificate. These days, most jobs require a tertiary diploma. That makes it hard for children from poor families. I don’t think we’ll be able to send our girl to university. The government should do more to make it possible for young people without diplomas to get good, steady jobs.” Pathways Towards Economic Security Indonesia Poverty Assessment 71 Shocks CHAPTER 5 POLICY RECOMMENDATIONS Photo: © Josh Estey/World Bank 72 Pathways Towards Economic Security Indonesia Poverty Assessment 5. POLICY RECOMMENDATIONS In the aftermath of COVID-19 and in the context of climate extreme poverty, with a small amount of frictional change and growing global uncertainties, Indonesia’s poverty likely to persist. Given Indonesia’s development, inclusive growth policies can be adapted to help more a broader definition of poverty, for example around households escape poverty and reach economic security. the US$ 3.20 2011 PPP poverty line, would be more adequate to allow formulation of meaningful anti- Indonesia’s inclusive growth needs to plan for a low- emission future. Indonesia still relies on export of carbon-intensive commodities. Despite currently higher poverty programs reaching a significant fraction of the population. These programs can be improved to include economically insecure households and sustain poverty prices also for carbon-intensive commodities triggered reduction gains. Better opportunities are needed in low- by Russia’s invasion of Ukraine, the long-term future carbon sectors with high productivity growth to boost will penalize countries relying on their production and incomes and reduce poverty, while taking advantage exports.251 This is a formidable challenge for Indonesia of digital opportunities. However, shocks are inevitable and will require structural transition of its economy.252 and will become more frequent with climate change, This transition will need to be inclusive and allow workers but resilience can be fostered to minimize their harm. to put their skills to higher productive use for growth to With about one-half of the non-poor population eradicate poverty and foster economic security. susceptible to falling back into poverty, better resilience and protection are needed. These measures will require Policies need to create better opportunities, protect public investments in a fiscally tight space. Policies need households against poverty, and focus fiscal resources to ensure cost-effective design while raising revenues on pro-poor investments, while promoting better and lift constraints to improve human capital equitably information and evidence for decision making. across the country. Finally, policy makers need to close Indonesia made impressive poverty reduction and remaining data and knowledge gaps to inform more shared prosperity gains. The country nearly eradicated effective policies (Figure 5.1). FIGURE 5.1: GDP growth (LHS) and GDP -per-capita (RHS) from 1990 to 2021 Towards economic security Livelihoods Resilience to shocks SA agility Infrastructure investments to SA coverage and bene ts create resilience Protecting Improve a ordability and Social insurance covering all workers Better and more agile social against quality of childcare Increase nancial inclusion assistance poverty Enable high- productivity and Improve sub-national administrative Productivity low-carbon capacity sectors Increase agricultural Human capital outcomes productivity and geographic disparities Make urban areas Education and health spending Social protection spending engines of growth Government revenues Re-examine the use of VAT exemptions Financing Increase taxes on alcohol, tobacco, sugar and carbon pro-poor Creating better opportunities Remove energy and agricultural subsidies investments Strengthen o cial statistics Enable data use Close analytical gaps Improving future policies Source: Authors’ calculations using SUSENAS (2002-2022) 251 World Bank 2022h. 252 World Bank 2021d Pathways Towards Economic Security Indonesia Poverty Assessment 73 Policy Recommendations Creating better opportunities Indonesia’s cities do not fully exploit the advantage of Indonesia’s economy can increase agricultural productivity, connectedness because of congestion. In addition, it is make urban areas engines of growth, enable high- costly to live in cities, while air pollution diminishes life productivity and low-carbon sectors while improving quality. As a result, while urban areas provide a wage affordable and high-quality childcare. premium, in Indonesia they do not provide a poverty- reduction premium. Attracting more workers to urban Continuing poverty reduction requires creation of places can steer them toward higher-productivity work, better work opportunities. Strong economic growth is especially in a context of job growth in high productivity reducing poverty in Indonesia at a similar pace compared and low-carbon sectors. Public investments in better to other countries. Poverty in urban and rural areas urban infrastructure and fostering affordable housing have converged, with most poor Indonesians living in will lower costs of living in cities, helping transform them urban areas now. Work, however, is often insufficient to into centers of connectedness rather than just high escape poverty and reach economic security. Low labor density. Infrastructure planning will need to take account productivity keeps wages low, while low agricultural for climate adaptation into and making cities safer.254 productivity limits incomes especially for rural households. Low female labor force participation limits Policies to improve structural competitiveness and available labor and hinders gender equity and women’s inclusion in global value chains can foster job growth empowerment. Better opportunities in urban and rural in high-productivity, low-carbon sectors. While the areas, especially for women, can help sustainable poverty labor market contributes to poverty reduction, it does reduction. not provide opportunities commensurate with high- income ambitions. The low-VA sector is sufficient Increasing agricultural productivity will remain for some to escape poverty but does not increase important for rural households to escape poverty. workers’ skills, depressing productivity. Unlocking the Most rural households – especially in remote areas – are entry and growth of new firms can be fostered by less still primarily engaged in agriculture. While diversification restrictive trade and foreign direct investment policies of income with non-agriculture income helps, including as well as more effective anti-competitive policies. This to mitigate negative shock effects, it is not an option for – together with boosting the digital economy – can spur everyone, and is often limited to low-VA jobs. Improved competition, innovation, and productivity contributing agricultural extension services promoting climate- to job growth.255 Integration into global value chains can smart approaches and better market access can boost enhance productivity, including attracting foreign direct agriculture productivity. Reduction of often badly investment for exporting industries and promoting targeted producer subsidies can improve competition investment in specific areas to encourage workers toward in the agricultural sector, as can the removal of import higher-productivity sectors. With a view to the future, barriers such as non-tariff measures (NTMs). This has the these policies should be focused on high productivity added advantage of reducing food prices.253 Reducing and low-carbon sectors, and promote eco industrial subsidies for rice production can shift production from parks and circular economy solutions. Improvements rice to high-value cash crops, for which the soil is often in skills and better matching them to labor market better suited. These policies can help to increase rural requirements can complement such policies; for incomes and provide pathways out of poverty. example, by improving educational quality, improving TVET, and integrating labor market information systems. Investments in urban infrastructure and affordable housing are needed to make urban areas engines Offering affordable childcare can create jobs, foster of growth and attract more workers to higher- female labor force participation, and improve productivity jobs, which can drive poverty reduction. productivity. Women often remain excluded from the World Bank 2022h. 254 Cali et al. 2021. 253 Wihardja and Cunningham 2021. 255 74 Pathways Towards Economic Security Indonesia Poverty Assessment Policy Recommendations labor force, especially when anticipating child-bearing The poor especially need protection, as they most at and during child-bearing age. This “dependency penalty” risk and have limited coping strategies. Thus, creating contributes to women being poorer than men. When resilience also mitigates inequality. affordable childcare is available, women have the option of shifting from unpaid to higher-productivity work,256 Social assistance can be better targeted and be made improving labor market skills and firm productivity.257 more agile. A shock can quickly trap a household in It helps to close the gender wage gap, which is still poverty if not mitigated. Social assistance plays a very substantial in Indonesia. In addition, it creates jobs. important role, as COVID-19 proved. However, the Among the many benefits of subsidizing childcare, pandemic also showed that fast expansion of social it fosters early childhood learning, with long-term assistance comes with beneficiary targeting challenges. positive effects on economic productivity.258 Finally, it For example, expanding PKH coverage during COVID-19 is relatively inexpensive and even raises government has been accompanied by reduced beneficiary personal income tax (PIT) and VAT revenues. Despite incidence among the poorest 20 percent, reaching only improvements in the availability of childcare and PHK 39 percent of eligible recipients in 2019. Indonesia’s cash transfers conditioned on enrolling children in social assistance can be improved in several dimensions, preschool (PAUD), Indonesian families do not widely including investing in coverage and data quality. use it, especially for younger children; only 17 percent Accuracy of targeting can also be improved; for example, of bottom 40 children and 22 percent of top 60 children through regular updating of the targeting database and were in preschool in 2019. Ideally, incentivizing use of calibrating eligibility criteria to reflect updated poverty supply of childcare would include education, information, definitions. A broader, and more frequently updated, and communication strategies to tackle cultural norms social registry can facilitate faster shock responses to and promote their use. affected households.259 This can be achieved by adopting a modular delivery system that is interoperable and Protecting against poverty supports open data standards, underpinned by a clear Social assistance needs to be better targeted and become data protection and consent framework. more agile, social insurance should be expanded to cover all workers, financial inclusion should be improved, Expanding coverage of contributory social insurance while infrastructure investments can increase disaster can be more systematic. While it is too early to expand preparedness. coverage of Indonesia’s new unemployment insurance to informal workers, more can be done to increase informal Protecting households safeguards poverty reduction workers’ participation in schemes that protect against progress. Despite poverty reduction progress, almost employment-related income shocks. This includes one-fifth of the population remains in poverty. Most of subsidization of premiums for poor and economically the poor are now in urban areas. In addition, about half of insecure workers, or adjustments of existing schemes the non-poor (one-third of the population) is susceptible to better accommodate informal work characteristics, to falling into poverty. The susceptibility is largely due to including flexibility in payment contributions and shocks rather than structural causes. Idiosyncratic shocks, in accessing benefits. Expanding contributory social which affect individual households—for example, a spell insurance coverage to all workers can help households of unemployment—are the largest source of risk. Co- buffer shocks and avoid poverty spells, while helping variate shocks, affecting multiple households at the same mitigate inequality,260 and even contributing to increased time—for instance, natural disasters—are increasing, a labor productivity.261 trend likely to continue given increased climate change. 256 Halim, Johnson, and Perova forthcoming. 259 Holmemo et al. 2020. 257 Cali et al. 2022. 260 Holmemo et al. 2020. 258 Heckman and Masterov 2007; García et al. 2020. 261 Rujiwattanapong 2022. Pathways Towards Economic Security Indonesia Poverty Assessment 75 Policy Recommendations Financial services play a critical role in creating for poorer sub-national governments. Policy design must resilience to shocks and reducing poverty. Financial aim to leverage expenditures to have the largest poverty services are not yet available to all Indonesians, limiting reduction and human capital benefits, while ensuring the ability to save and borrow, as well as receive digital equity. At the same time, creation of fiscal space cannot payments from government programs, for example. endanger economic growth nor unduly burden the Without being able to smooth consumption, households poor, while maintaining a fiscal buffer to draw from in can be forced to resort to adverse coping strategies, the case of future shocks. such as selling assets or restricting consumption for children and adults. Digital financial services can enable In the short-term, the Government can increase more households to enter the financial system and can revenues from indirect taxes, including reexamination be expanded by establishing a well-functioning, fully of VAT exemptions. Direct taxes such as personal interoperable payment system, creating digital IDs, and income tax (PIT) and corporate income tax (CIT) are promoting open banking policies. more progressive than indirect taxes but require greater tax administrative capacity and a high degree of Infrastructure investments are needed to improve formalization in the economy. Only developed countries resilience against natural disasters. Large areas of rely upon direct taxes for most tax revenue. In the near- Indonesia are prone to disaster risks. Once a disaster term, increases in revenues in Indonesia are more feasible strikes, people can be killed or injured and families through indirect taxes. A practical way to quickly increase can lose their homes and assets. In addition, costs to revenues is to eliminate VAT exemptions and preferred cover large-scale damages can be exorbitant. Creating rates for various goods and services. While these items resilience also requires investments, but at a fraction often represent a greater share of poorer household of the cost of post-disaster recovery. Thus, investing consumption, they are also consumed by richer early into resilient infrastructure, including housing and households in greater amounts, meaning that—like transportation, will be crucial to avoid setbacks on the traditional energy or food subsidies—most spending (or path to being a high-income country.262 Also, investments revenue forgone) does not go to the poor. Approximately in climate-smart agricultural are key to protect farmers one-third of potential VAT revenues in Indonesia are lost from harvest losses. through the current exemptions structure, or IDR 91 trillion (0.67 percent of GDP), about the same amount as Financing pro-poor investments the entire expanded social assistance budget in 2019. Government can increase tax revenues through indirect taxation strategies, such as reducing VAT exemptions Indirect taxes can be further increased through or ineffective energy subsidies or by increasing taxes on higher taxes on tobacco, alcohol, sugar-sweetened tobacco, alcohol, sugar-sweetened beverages, and carbon. beverages, and carbon. Tobacco, alcohol, and sugar- At the same time, improving sub-national administrative sweetened beverages have adverse health effects, capacity can help improve human capital outcomes more with large cost implications for public health systems. equitably across the country. Increasing tax on these goods will reduce their consumption, saving public health system funds while Pro-poor public investments are needed, but must generating government revenue. While these taxes be designed carefully in the context of limited fiscal often are regressive, changes in behavior often mitigate space. Creating better opportunities, protecting against implications for poverty. Individuals choose to consume poverty, and achieving higher and more equitable less, unleashing health benefits, while producers change human capital outcomes requires public investment ingredients, such as sugar levels, to avoid taxes and price and sub-national administrative capacity. However, fiscal increases. While decreasing gains in direct government space and administrative capacity are limited, especially revenue, it still creates savings through health benefits. Similarly, a carbon tax can incentive transition towards World Bank 2022h. 262 76 Pathways Towards Economic Security Indonesia Poverty Assessment Policy Recommendations a low-carbon economy while reducing air pollution be addressed. Ineffective planning and execution of and generating tax revenues. The price for emissions, capital projects can be improved. The low quality and however, is a key parameter for the success of the tax, transparency of spending information can be enhanced, and needs to make crucial trade-offs between revenues also to create better accountability, with a feedback and socio-economic effects. mechanism to improve service delivery quality. Expensive and often ineffective and poorly targeted Improving future policies subsidies for energy and agriculture can be removed. The GOI can strengthen official statistics, better encourage Energy subsidies can reduce poverty but suffer from data use, and close knowledge gaps, specifically around targeting errors, including non-poor households structural transformation and informality. while excluding poor ones. At the same time, these subsidies distort the market and increase consumption The knowledge base to inform policies needs to be of subsidized products. Especially in the case of fuel expanded, as not all challenges are yet fully understood. and electricity, this is problematic given their negative Indonesia has an impressive evidence base, thanks to a climate implications. Agricultural subsidies are large number of high-quality surveys conducted with expensive, distort the market, and are not helping poor higher frequency than in many countries. However, the farmers. The Government spends 2 to 3 percent of GDP country can still strengthen official statistics, some of on agriculture, most of it on subsidies for agricultural which only require small modifications in methodologies, producers. Revisiting agricultural expenditures towards while others demand new data collection efforts and enhancing competitiveness and productivity can lead to improved capacity. In addition, data use can enable large fiscal savings. Coupling the elimination of subsidies more research and analysis to address new challenges as with measures to raise productivity, such as enhancing Indonesia requires more careful study to inform policies. agricultural extension services and opening the Hard to compare poverty measures need to be replaced agricultural market to more competition and imports, by an absolute measure of poverty that includes spatial can mitigate any food price increases. In addition, price inflation. The “weakly relative” nature of Indonesia’s social assistance can substitute for subsidies to mitigate poverty line, updated yearly at the province-level, negative effects on the poor. Scaling-up social assistance undermines comparability of trends across provinces. is more cost-effective, better targeted than direct or Indonesia needs a new absolute measure of poverty— indirect subsidies, and is more sustainable. similar to international poverty lines—to fill this gap. Such a poverty line can define poverty more broadly to Improving administrative capacity for sub-national ensure that anti-poverty programs improve livelihoods governments will be critical to achieve better and for a meaningful fraction of the population. Given more equitable human capital outcomes, spurring large price differences across Indonesia, spatial price future productivity and reducing long-term inequality. deflation is important but requires a reliable rural CPI Subnational public financial management policies, to complement the current urban-only CPI. This will systems, and skills do not yet enable subnational also help measure poverty reduction progress using governments to spend effectively. Recent subnational the international poverty lines. In addition, revisiting public financial management reforms263 support data needs would help record key data, either strategic decision-making. Effective and transparent through existing or new surveys. For example, missing spending need to be further defined, in addition farmers’ incomes severely limit the evidence needed to strengthening the implementation capacity of to understand rural livelihoods; and time-use data, subnational governments in general, but particularly especially for women, is not available. Also, data quality the ones with especially low capacity. Specifically, the can be improved, for example, by closing the large disconnect between subnational government policies gap between national accounts and household survey and budgets and low budget reliability will need to private consumption data. PP 12/2019 and Permendagri 77/2020. 263 Pathways Towards Economic Security Indonesia Poverty Assessment 77 Drivers from 2014 to 2019 Shifting policies to a more “enabling” perspective, New challenges will need new policies based on while addressing data and quality gaps, would new evidence, including for key issues such as increase the use of Indonesia’s data. Two surveys, structural transformation and informality. Low work Susenas for household consumption and Sakernas for the productivity has far-ranging consequences, limiting labor force, providing a wealth of data. However, some livelihoods and economic growth. Better understanding analysis remains constrained due to limitation in making of the current structural transformation can improve data available. Currently, geographic identifiers are not recommendations in the context of climate change shared beyond the province-level, diminishing its use in and global uncertainties. Also, the high rate of work conjunction with big data. Also, household information informality of Indonesia’s economy is closely linked to the is currently not available in Sakernas, undermining the large low-VA service sector, which offers employment ability to link it statistically with Susenas—for instance, to and an escape from poverty but limits productivity, thus impute welfare. With a more enabling approach, more trapping workers in economic insecurity. More detailed researchers can engage with Indonesian data to inform studies are needed to shed light on the informal sector— future policies. including the level of informality, its different aspects, costs, and benefits—and why firms and household enterprises decide for or against formalization. 78 Pathways Towards Economic Security Indonesia Poverty Assessment REFERENCES Afkar, Rythia, and Noah Yarrow. 2021. “Rewrite the Future: How Indonesia’s Education System Can Overcome the Losses from the Covid-19 Pandemic and Raise Learning Outcomes for All.” Washington, D.C.: World Bank. Alderman, Harold, John Hoddinott, and Bill Kinsey. 2006. “Long Term Consequences of Early Childhood Malnutrition.” Oxford Economic Papers 58, no. 3: 450–74. Ali, R., and I. Setiawan. 2022. “Looking Beyond Aggregate Poverty and Vulnerability Rates in Indonesia: Aligning Social Protection with Sources of Vulnerability to Poverty.” Washington, DC: World Bank. Asian Development Bank. 2007. “Asian Development Outlook 2007 Update.” Philippines: Asian Development Bank. At https://wedocs.unep.org/handle/20.500.11822/8908. ———. 2019. “Policies to Support Investment Requirements of Indonesia’s Food and Agriculture Development during 2020-2045.” Asian Development Bank. Indonesia. At https://www.adb.org/publications/indonesia-food- agriculture-development-2020-2045. ———. 2020a. “2017 International Comparison Program for Asia and the Pacific: Purchasing Power Parities and Real Expenditures Results and Methodology.” Manila: Asian Development Bank. ———. 2020b. “Quality Jobs and the Future of Work: Technical Assistance Report.” Technical Assistance Report 54100– 001 Asian Development Bank. ———. 2022. “Building Resilience of the Urban Poor in Indonesia.” Philippines: Asian Development Bank. Indonesia. At https://www.adb.org/publications/building-resilience-urban-poor-indonesia. Atkinson, A. B., and Thomas Piketty, eds. 2007. Top Incomes over the Twentieth Century: A Contrast between European and English-Speaking Countries. 1 edition Oxford ; New York: Oxford University Press. Azevedo, João Pedro, Alexandru Cojocaru, Veronica Montalva Talledo, and Ambar Narayan. 2022. “COVID-19 School Closures, Learning Losses and Inter-Generational Mobility.” The World Bank: 31. Badan Pusat Statistik, National Population and Family Planning Board, and Ministry of Health. 2008. “Indonesia Demographic and Health Survey 2007.” At https://dhsprogram.com/publications/publication-fr218-dhs-final- reports.cfm. ———. 2018. “Indonesia Demographic and Health Survey 2017.” At https://dhsprogram.com/publications/publication- fr342-dhs-final-reports.cfm, accessed September 13, 2022. Bank Indonesia. 2022. “Monetary Policy Report - Quarter II 2022.” Monetary Policy Report Economic and Monetary Policy Department, Bank Indoensia. At https://www.bi.go.id/en/publikasi/laporan/Pages/Laporan-Kebijakan-Moneter- Triwulan-II-2022.aspx. Bhalla, Surjit. 2002. “Imagine There’s No Country: Poverty, Inequality, and Growth in the Era of Globalization.” Peterson Institute Press: All Books Peterson Institute for International Economics. At https://econpapers.repec.org/ bookchap/iieppress/348.htm. Blanchet, Thomas, Ignacio Flores, and Marc Morgan. 2022. “The Weight of the Rich: Improving Surveys Using Tax Data.” The Journal of Economic Inequality 20, no. 1: 119–50. Bulman, D., M. Eden, and H. Nguyen. 2016. “Transitioning from Low-Income Growth to High-Income Growth: Is There a Middle-Income Trap?” 646 ADBI Working Paper Tokyo: Asian Development Bank Institute. Cali, Massimiliano, Hillary C. Johnson, Elizaveta Perova, and Nabil R. Ryandiansyah. 2022. “Caring for Children and Firms? The Impact of Preschool Expansion on Firm Productivity.” 10193 Policy Research Working Paper Washington, D.C.: World Bank. Cali, Massimiliano, Mochamad Pasha, Francis Addeah Darko, Sudarno Sumarto, Taufik Hidayat, and Sailesh Tiwari. 2021. “Nutritional Impacts of Trade Policies: Evidence from Indonesia.” Washington, DC: World Bank. Cameron, Lisa, Diana Contreras Suarez, and William Rowell. 2018. “Female Labour Force Participation in Indonesia: Why Has It Stalled?” Bulletin of Indonesian Economic Studies 55, no. 2: 157–92. Chaudhuri, Shubham, Jyotsna Jalan, and Asep Suryahadi. 2002. “Assessing Household Vulnerability to Poverty from Cross- Sectional Data: A Methodology and Estimates from Indonesia.” At https://doi.org/10.7916/D85149GF. Christian, Cornelius, Lukas Hensel, and Christopher Roth. 2019. “Income Shocks and Suicides: Causal Evidence From Indonesia.” The Review of Economics and Statistics 101, no. 5: 905–20. Credit Suisse. 2019. “Global Wealth Report 2019.” Credit Suisse. At https://www.credit-suisse.com/about-us/en/reports- research/global-wealth-report.html. Pathways Towards Economic Security Indonesia Poverty Assessment 79 References Cuaresma, Jesus Crespo, Stephan Klasen, and Konstantin M. Wacker. 2016. “There Is Poverty Convergence.” SSRN Scholarly Paper Rochester, NY. At https://papers.ssrn.com/abstract=2718720, accessed September 12, 2022. Datt, Gaurav, and Martin Ravallion. 1992. “Growth and Redistribution Components of Changes in Poverty Measures.” Journal of Development Economics 38, no. 2: 275–95. David E. Bloom, David Canning, Günther Fink, and Jocelyn E. Finlay. 2012. “Microeconomic Foundations of the Demographic Dividend.” Program on the Global Demography of Aging. At https://ideas.repec.org/p/gdm/wpaper/9312.html. Dharmawan, G., and D. Suryadarma. 2021. “Education Quality Across Indonesia’s Districts: Estimations from a Policy Experiment.” Journal of Southeast Asian Economies 38, no. 3: 401–25. Dizioli, Allan, and Roberto Pinheiro. 2016. “Health Insurance as a Productive Factor.” Labour Economics 40: 1–24. Duryea, S., D. Lam, and D. Levison. 2007. “Effects of Economic Shocks on Children’s Employment and Schooling in Brazil.” Journal of Development Economics 84, no. 1: 188–214. Elbers, Chris, Jan Willem Gunning, and Bill Kinsey. 2007. “Growth and Risk: Methodology and Micro Evidence.” World Bank Economic Review 21, no. 1: 1–19. Feriyanto, Nur, Dityawarman El Aiyubbi, and Achmad Nurdany. 2020. “The Impact of Unemployment, Minimum Wage, and Real Gross Regional Domestic Product on Poverty Reduction in Provinces of Indonesia.” Asian Economic and Financial Review 10, no. 10: 1088–99. Ferraro, Paul J., and Rhita Simorangkir. 2020. “Conditional Cash Transfers to Alleviate Poverty Also Reduced Deforestation in Indonesia.” Science Advances 6, no. 24. At https://www.science.org/doi/10.1126/sciadv.aaz1298. Ferreira, Francisco H. G., Sergio P. Firpo, and Julian Messina. 2017. “Ageing Poorly?: Accounting for the Decline in Earnings Inequality in Brazil, 1995-2012.” At http://hdl.handle.net/10986/26359. Gao, Jia, Katja Vinha, and Emmanuel Skoufias. 2020. “Vulnerability Tool A User’s Manual.” Washington, DC: World Bank. García, Jorge Luis, James J. Heckman, Duncan Ermini Leaf, and María José Prados. 2020. “Quantifying the Life-Cycle Benefits of an Influential Early-Childhood Program.” Journal of Political Economy 128, no. 7 The University of Chicago Press: 2502–41. Giles, Gordon Muir, Dorothy Farrar Edwards, M. Tracy Morrison, Carolyn Baum, and Timothy J. Wolf. 2017. “Screening for Functional Cognition in Postacute Care and the Improving Medicare Post-Acute Care Transformation (IMPACT) Act of 2014.” The American Journal of Occupational Therapy 71, no. 5: 7105090010p1–6. Goldstein, Harvey. 2010. Multilevel Statistical Models (4th Edition). 1st ed John Wiley & Sons, Ltd Wiley. At https:// onlinelibrary.wiley.com/doi/book/10.1002/9780470973394, accessed September 15, 2022. Greenstone, M., and Q. Fan. 2019. “Air Quality Life Index: Indonesia’s Worsening Air Quality and Its Impact on Life Expectancy.” Chicago, IL: University of Chicago. Gubert, Flore, and Anne-Sophie Robilliard. 2007. “Risk and Schooling Decisions in Rural Madagascar: A Panel Data- Analysis.” Journal of African Economies Rochester, NY. At https://papers.ssrn.com/abstract=1153288. Günther, Isabel, and Kenneth Harttgen. 2009. “Estimating Households Vulnerability to Idiosyncratic and Covariate Shocks: A Novel Method Applied in Madagascar.” World Development 37, no. 7: 1222–34. Hadiwidjaja, Gracia, Asha Williams, and Sara Giannozzi. 2022. “Improving Data Quality for an Effective Social Registry in Indonesia.” Washington, D.C.: World Bank. Haerpfer, C., R. Inglehard, A. Moreno, C. Welzel, K. Kizilova, J. Diez-Medrano, M. Lagos, P. Norris, P. Ponarin, and P. Puranen. 2020. “World Values Survey: Round Seven – Country-Pooled Datafile.” Madrid, Spain & Vienna, Austria: JD Systems Institute & WVSA Secretariat. Halim, Daniel, Sean Hambali, and Ririn Salwa Purnamasari. 2022. “The Initial Gendered Impacts of the COVID-19 Pandemic on Indonesia’s Labor Market: How Does It Shape the Future of Women’s Work?” World Bank. Halim, Daniel, Hillary C. Johnson, and Elizaveta Perova. 2022. “Preschool Availability and Women’s Employment: Evidence from Indonesia.” Economic Development and Cultural Change 71, no. 1: 39–61. ———. forthcoming. “Preschool Availability and Women’s Employment: Evidence from Indonesia.” Economic Development and Cultural Change. Heckman, James J. 2007. “The Economics, Technology, and Neuroscience of Human Capability Formation.” Proceedings of the National Academy of Sciences 104, no. 33: 13250–55. Heckman, James J., and Dimitriy V. Masterov. 2007. “The Productivity Argument for Investing in Young Children.” NBER Working Paper no. 13016. Hill, Hal. 2021. “What’s Happened to Poverty and Inequality in Indonesia over Half a Century?” Asian Development Review 38, no. 1: 68–97. 80 Pathways Towards Economic Security Indonesia Poverty Assessment References Holmemo, Camilla, Pablo Acosta, Tina George, Robert J. Palacios, Juul Pinxten, Shonali Sen, and Sailesh Tiwari. 2020. “Investing in People: Social Protection for Indonesia’s 2045 Vision.” Washington, DC: World Bank. At https://openknowledge.worldbank.org/handle/10986/33767. Human Capital Project. 2020. “Indonesia Human Capital Index 2020.” Washington, D.C.: World Bank. International Monetary Fund. 2022. “IMF Country Report: Indonesia.” IMF Country Report No. 22/84 International Monetary Fund. Jellema, Jon, Matthew Wai-Poi, and Rythia Afkar. 2017. “The Distributional Impact of Fiscal Policy in Indonesia.” Directions in Development - Poverty and Inequality Washington, D.C.: World Bank. Kementerian Kesehatan Republik Indonesia. 2022. Hasil Survei Status Gizi Indonesia (SSGI) 2022. Indonesia: Kementerian Kesehatan Republik Indonesia. Kementerian PPN (Bappenas). 2021. “Sustainable and Resilient Recovery from the COVID-19 Pandemic for the Achievement of the 2030 Agenda.” Indonesia: Ministry of National Development Planning of the Republic of Indonesia. At https://sdgs.bappenas.go.id/laporan-voluntary-national-review-vnr-indonesia-2021/. Khalid, Muhammed Abdul, and Li Yang. 2019. “Income Inequality and Ethnic Cleavages in Malaysia: Evidence from Distributional National Accounts (1984–2014).” Journal of Asian Economics 72: 101252. Khelfaoui, Issam, Yuantao Xie, Muhammad Hafeez, Danish Ahmed, Houssem Eddine Degha, and Hicham Meskher. 2022. “Effects of Health Shocks, Insurance, and Education on Income: Fresh Analysis Using CHNS Panel Data.” International Journal of Environmental Research and Public Health 19, no. 14: 8298. Klasen, Stephan, and Hermann Waibel. 2013. Vulnerability to Poverty: Theory, Measurement and Determinants, with Case Studies from Thailand and Vietnam. Basingstoke, England: Palgrave Macmillan. Kolukuluri, Kalyan. 2022. “Adverse Health Shocks, Social Insurance and Household Consumption: Evidence from Indonesia’s Askeskin Program.” International Journal of Health Economics and Management. At https://doi. org/10.1007/s10754-022-09329-6. Krisnamurthi, Bayu and Utami, Anisa Dwi. 2022. “The Effect of Price Policy on Price Dynamics: Empirical Evidence in Indonesian Rice Market at Wholesale Level.” Journal of Agribusiness and Rural Development Research 8, no. 1: 34–45. Liddell, Christine, and Gordon Rae. 2001. “Predicting Early Grade Retention: A Longitudinal Investigation of Primary School Progress in a Sample of Rural South African Children.” British Journal of Educational Psychology 71, no. 3: 413–28. Lim, D., N. Rarasati, F. Tresnatri, and A.R. Barasa. 2022. “Learning Loss or Learning Gain? A Potential Silver Lining to School Closures in Indonesia.” RISE Insight Series RISE. Lim, Sung Soo. 2017. “In Times of Sickness: Intra-Household Labour Substitution in Rural Indonesian Households.” The Journal of Development Studies 53, no. 6: 788–804. López-Calva, Luis F., and Eduardo Ortiz-Juarez. 2014. “A Vulnerability Approach to the Definition of the Middle Class.” The Journal of Economic Inequality 12, no. 1: 23–47. Lustig, Nora. 2018. Commitment to Equity Handbook: Estimating The Impact of Fiscal Policy on Inequality and Poverty. Brookings Institution Press. ———. 2019. “The ‘Missing Rich’ in Household Surveys: Causes and Correction Approaches.” Preprint SocArXiv. At https:// osf.io/j23pn. Majlesi, Kaveh. 2016. “Labor Market Opportunities and Women’s Decision Making Power within Households.” Journal of Development Economics 119: 34–47. Munoz Boudet, Ana Maria, Paola Buitrago, Benedicte Leroy De La Briere, David Newhouse, Eliana Rubiano Matulevich, Kinnon Scott, and Pablo Suarez-Becerra. 2018. “Gender Differences in Poverty and Household Composition through the Life-Cycle: A Global Perspective.” Working Paper Washington, DC: World Bank. At https:// openknowledge.worldbank.org/handle/10986/29426. Naidoo, Darian, Truman Packard, and Ilmiawan Auwalin. 2015. “Mobility, Scarring and Job Quality in Indonesia’s Labor Market.” World Bank. At http://elibrary.worldbank.org/doi/book/10.1596/1813-9450-7484. Nayyar, Gaurav, Mary Hallward-Driemeier, and Elwyn Davies. 2021. “At Your Service? The Promise of Services-Led Development.” Washington, D.C.: World Bank. OECD. 2015. “Universal Basic Skills: What Countries Stand to Gain.” OECD Publishing. ———. 2021. “OECD Economic Surveys Indonesia.” East Asia and Pacific Region: OECD. Pape, Utz, England Can, Citra Kumala, Anissa Rahmawati, Virgi Sari, and Imam Setiawan. 2023. “Distributional Impacts of Climate Change Mitigation in Indonesia: A Background Note for the CCDR in Indonesia.” Pathways Towards Economic Security Indonesia Poverty Assessment 81 References Pardede, Elda Luciana, Philip McCann, and Viktor A. Venhorst. 2020. “Internal Migration in Indonesia: New Insights from Longitudinal Data.” Asian Population Studies 16, no. 3 Routledge: 287–309. Perkins, Jessica M., S.V. Subramanian, George Davey Smith, and Emre Özaltin. 2016. “Adult Height, Nutrition, and Population Health.” Nutrition Reviews 74, no. 3: 149–65. Piketty, Thomas. 2003. “Income Inequality in France, 1901-1998.” SSRN Scholarly Paper Rochester, NY. At https://papers. ssrn.com/abstract=441962. Piketty, Thomas, and Emmanuel Saez. 2003. “Income Inequality in the United States, 1913–1998.” The Quarterly Journal of Economics 118, no. 1: 1–41. Pratomo, Devanto Shasta, Wildan Syafitri, and Clarissa Sekar Anindya. 2020. “Expanding Middle Class in Indonesia.” The Journal of Indonesia Sustainable Development Planning; Kementerian PPN/Bappenas 1, no. 3: 307–12. Prydz, Espen Beer, Dean Jolliffe, and Umar Serajuddin. 2021. “Mind the Gap: Disparities in Assessments of Living Standards Using National Accounts and Household Surveys.” Working Paper Washington, DC: World Bank. At https://openknowledge.worldbank.org/handle/10986/36306, accessed September 13, 2022. Purnamasari, R., S. Hambali, and D. Halim. Forthcoming. “Narrowing of Gender Gaps in Employment during the Onset of COVID-19 in Indonesia.” Asian Development Review. Purnamasari, Ririn Salwa, Nina Weimann-Sandig, Resmi Setia Milawati, and Sheila Town. 2020. “What’s Holding Women Back?” Washington, DC: The World Bank. Purnamasari, R.S., P.H. Satyaka, Suharti, Matthew Wai-Poi, and Rythia Afkar. 2022. “COVID-19 and Education in Indonesia: When School Reopening Is Not Enough to Avert Losing a Generation.” In S.M. Indrawati, S. Nazara, T. Anas, C.F. Ananda, and K. Verico, eds. Keeping Indonesia Safe from COVID-19 Pandemic: Lessons Learnt from the National Economic Recovery Programme Singapore: ISEAS. Putra, Rendra A., Kostiantyn Ovsiannikov, and Koji Kotani. 2022. “COVID-19-Associated Income Loss and Job Loss: Evidence from Indonesia.” SDES-2022-4 Working Papers Kochi University of Technology, School of Economics and Management. At https://ideas.repec.org/p/kch/wpaper/sdes-2022-4.html. Ravallion, Martin. 2021. “Missing Top Income Recipients.” Working Paper Working Paper Series National Bureau of Economic Research. At https://www.nber.org/papers/w28890. Roberts, Mark, Frederico Gil Sander, and Sailesh Tiwari. 2019. Time to ACT: Realizing Indonesia’s Urban Potential. Washington, DC: World Bank. At https://openknowledge.worldbank.org/handle/10986/31304. Rodrik, Daniel. 2015. “Premature Deindustrialization.” Journal of Economic Growth 21: 1–33. Rosenzweig, Mark R., and Hans P. Binswanger. 1993. “Wealth, Weather Risk and the Composition and Profitability of Agricultural Investments.” The Economic Journal 103, no. 416: 56. Rosser, Andrew J., and Mohamad Fahmi. 2016. “The Political Economy of Teacher Management in Decentralized Indonesia.” 7918 World Bank Policy Research Working Paper Washington, D.C.: World Bank. Rujiwattanapong, W. Similan. 2022. “Unemployment Insurance and Labour Productivity over the Business Cycle.” Review of Economic Dynamics 46: 196–223. Santoso, Rokhedi Priyo, and Jaka Sriyana. 2021. “The Effect of Idiosyncratic Shocks on Labor Market Outcomes of Informal Households in Indonesia.” Jurnal Ekonomi Malaysia no. Jurnal Ekonomi Malaysia: 15. Sari, Virgi Agita, Ririn Salwa Purnamasari, and Ade Febriady. 2022. “Households’ Exposure to, and Recovery from, COVID-19 Socio-Economic Shocks in Indonesia: Insights from a Year of High-Frequency Monitoring.”: 38. Schaner, Simone, and Smita Das. 2016. “Female Labor Force Participation in Asia: Indonesia Country Study.” 474 Economics Working Paper Series Manila: Asian Development Bank. Simeu, Natalie, and Sophie Mitra. 2019. “Disability and Household Economic Wellbeing: Evidence from Indonesian Longitudinal Data.” Oxford Development Studies 47, no. 3: 275–88. Sjahrir, B. S., and I. D. Wibisono. 2022. “Food Insecurity in Indonesia During COVID-19 Pandemic: What Do We Learn from High-Frequency Monitoring of Households?” Washington, DC: World Bank. Skoufias, Emmanuel, Katja Vinha, and Berhe Mekonnen Beyene. 2021. “Quantifying Vulnerability to Poverty in the Drought-Prone Lowlands of Ethiopia.” Working Paper Washington, DC: World Bank. At https://openknowledge. worldbank.org/handle/10986/35107. Stith, Adrienne Y., Kathleen S. Gorman, and Naseem Choudhury. 2003. “The Effects of Psychosocial Risk and Gender on School Attainment in Guatemala.” Applied Psychology 52, no. 4: 614–29. Syamsulhakim, Ekki, and Nurzanty Khadijah. 2021. “Graduating from a Conditional Cash Transfer Program in Indonesia: Results of a Household Survey of Prosperous-Independent Graduates of the Family Hope Program in 2020.” Washington, DC: World Bank. At https://openknowledge.worldbank.org/handle/10986/36784. 82 Pathways Towards Economic Security Indonesia Poverty Assessment References Tesliuc, Emil D., and Kathy Lindert. 2004. “Risk and Vulnerability in Guatemala: A Quantitative and Qualitative Assessment.” 30154 Social Protection Discussion Papers and Notes The World Bank. At https://ideas.repec.org/p/wbk/ hdnspu/30154.html. Wajdi, Nashrul, Clara H. Mulder, and Sri M. Adioetomo. 2017. “Inter-Regional Migration in Indonesia: A Micro Approach.” Journal of Population Research 34, no. 3: 253–77. Walker, Susan P, Susan M Chang, Christine A Powell, and Sally M Grantham-McGregor. 2005. “Effects of Early Childhood Psychosocial Stimulation and Nutritional Supplementation on Cognition and Education in Growth-Stunted Jamaican Children: Prospective Cohort Study.” The Lancet 366, no. 9499: 1804–7. Widianto, Bambang, and Widjajanti Isdijoso. 2020. “The Situation of the Elderly in Indonesia and Access to Social Protection Programs: Secondary Data Analysis.” SMERU Research Report Indonesia: The National Team for the Acceleration of Poverty Reduction, The SMERU Research Institute. Wihardja, Maria Monica, and Wendy Cunningham. 2021. “Pathways to Middle-Class Jobs in Indonesia.” Technical Report Washington DC: The World Bank. Winsemius, Hessel C., Brenden Jongman, Ted I. E. Veldkamp, Stephane Hallegatte, Mook Bangalore, and Philip J. Ward. 2018. “Disaster Risk, Climate Change, and Poverty: Assessing the Global Exposure of Poor People to Floods and Droughts.” Environment and Development Economics 23, no. 3: 328–48. World Bank. 1993. “The East Asia Miracle.” Washington, D.C.: World Bank. ———. 2006. “Making the New Indonesia Work for the Poor.” Washington, DC: World Bank. At https://openknowledge. worldbank.org/handle/10986/8172. ———. 2015a. “Live Long and Prosper: Aging in East Asia and the Pacific.” Washington, D.C.: World Bank. ———. 2015b. “Taxes and Public Spending in Indonesia : Who Pays and Who Benefits.” Washington, D.C.: World Bank. ———. 2015c. “Indonesia Systematic Country Diagnostic: Connecting the Bottom 40 Percent to the Prosperity Generation.” Washington, DC: World Bank. At https://openknowledge.worldbank.org/handle/10986/23117. ———. 2016. “Indonesia’s Rising Divide.” Working Paper Washington, DC: World Bank. At https://openknowledge. worldbank.org/handle/10986/24765. ———. 2017. “The Fertiliser Sector and Subsidy Policy in Indonesia - Key Takeaways.” Washington, D.C.: World Bank. ———. 2018a. “Riding the Wave: An East Asian Miracle for the 21st Century.” Washington, D.C.: World Bank. ———. 2018b. “Economic and Social Mobility in Urbanizing Indonesia.” Indonesia: World Bank. ———. 2019a. “Aspiring Indonesia—Expanding the Middle Class.” Washington, DC: World Bank. At https:// openknowledge.worldbank.org/handle/10986/33237. ———. 2019b. “Strengthening the Disaster Resilience of Indonesian Cities: A Policy Note.” Working Paper Washington, DC: World Bank. At https://openknowledge.worldbank.org/handle/10986/32459. ———. 2019c. “Bangladesh Poverty Assessment: Facing Old and New Frontiers in Poverty Reduction.” Washington, DC: World Bank. ———. 2020a. “Indonesia Country Gender Assessment: Investing in Opportunities for Women.” Washington, DC: World Bank. At https://openknowledge.worldbank.org/handle/10986/35310. ———. 2020b. “No One Left Behind: Rural Poverty in Indonesia.” Washington, DC: World Bank. At https://openknowledge. worldbank.org/handle/10986/34163. ———. 2020c. “Indonesia Systematic Country Diagnostic.” East Asia and Pacific Region. ———. 2020d. “Indonesia Public Expenditure Review 2020: Spending for Better Results.” Washington, DC: World Bank. At https://openknowledge.worldbank.org/handle/10986/33954. ———. 2020e. “Indonesia Economic Prospects, July 2020: The Long Road to Recovery.” Washington, DC: World Bank. At https://openknowledge.worldbank.org/handle/10986/34123. ———. 2020f. “Revisiting the Impact of Government Spending & Taxes on Poverty & Inequality in Indonesia.” East Asia and Pacific Region: World Bank. ———. 2020g. “Indonesia Economic Prospects, December 2020: Towards a Secure and Fast Recovery.” Washington, DC: World Bank. At https://openknowledge.worldbank.org/handle/10986/34930. ———. 2021a. “Beyond Unicorns : Harnessing Digital Technologies for Inclusion in Indonesia.” Washington, D.C.: World Bank. ———. 2021b. “Indonesia Learning Poverty Brief.” Washington, D.C.: World Bank. ———. 2021c. “Welfare Tracking in the Aftermath of Crisis: The Central Sulawesi Disaster Response.” World Bank. Pathways Towards Economic Security Indonesia Poverty Assessment 83 References ———. 2021d. “Indonesia Economic Prospects, December 2021: A Green Horizon - Toward a High Growth and Low Carbon Economy.” Washington, DC: World Bank. At https://openknowledge.worldbank.org/handle/10986/36732. ———. 2022a. “Commodity Markets Outlook: The Impact of the War in Ukraine on Commodity Markets.” Washington, D.C.: World Bank. ———. 2022b. Poverty and Shared Prosperity 2022: Correcting Course. Washington, D.C.: World Bank. ———. 2022c. “The Geography of Human Capital: Insights from the Sub-National Human Capital Index in Indonesia.” World Bank. ———. 2022d. “Indonesia Economic Prospects, December 2022: Trade for Growth and Economic Transformation.” Washington, DC: World Bank. ———. 2022e. “Reforms for Recovery.” October 2022 East Asia and Pacific Economic Update Washington, D.C.: World Bank. ———. 2022f. “Braving the Storms.” April 2022 East Asia and Pacific Economic Update Washington, D.C.: World Bank. ———. 2022g. “Finance for an Equitable Recovery.” World Development Report Washington, D.C.: World Bank. ———. 2022h. “Indonesia: Country Climate and Development Report.” World Bank. ———. 2022i. “Indonesia Economic Prospects, June 2022: Financial Deepening for Stronger Growth and Sustainable Recovery.” Washington, DC: World Bank. At https://openknowledge.worldbank.org/handle/10986/37584. ———. 2023. “Jobs Flagship Report: Jobs After Covid-19.” Washington, D.C.: World Bank. ———. forthcoming. “Expanding Social Insurance Coverage to the Informal Sector in Indonesia.” Washington, D.C.: World Bank. ———. forthcoming. “Public Expenditure Review for Indonesia.” Washington, D.C.: World Bank. Yarrow, Noah, Eema Masood, and Rythia Afkar. 2020. “Estimates of Covid-19 Impacts on Learning and Earning in Indonesia: How to Turn Tide.” Washington, D.C.: World Bank. 84 Pathways Towards Economic Security Indonesia Poverty Assessment ANNEXES Annex Estimating economic insecurity 264 Günther and Harttgen 2009; Chaudhuri, Jalan, and Suryahadi 2002; Goldstein 2010 265 The model also allows for heteroscedasticity (unequal variance of error terms across households) and efficient estimators given the hierarchical data structure (household & community variables used simultaneously). There is no need for lengthy panel data or information on specific shocks. See Gunther & Harttgen (2008) for details. Ideally, this estimation would entail a repeated observation of the same household’s consumption at different periods over time. While Indonesia is fortunate to have the Indonesia Family Life Survey panel data (IFLS) that enumerated each seven years since 1993, and Panel Indonesia Socio-Economic Survey that tracked a shorter period consumption variation from the same households. Unfortunately, the last wave of these data is only available for 2010 and 2014 which creates additional challenges in measuring vulnerability to poverty for a more recent time. The IFLS is an on-going longitudinal survey, conducted once every seven years in Indonesia starting from 1993. The survey has sample that is representative of about 83% of the Indonesian population (except Maluku and Papua) and has track over 30,000 individuals living in half of the provinces in the country for over 21 years (1993-2014). 265 Panel SUSENAS is specialized years of Indonesia socio-economic survey data where instead of surveying different households in different wave of the survey, the SUSENAS track the information of the same household for several years of time. The last time this nationally representative survey capture panel information was between the period of 2008-2010. 86 Pathways Towards Economic Security Indonesia Poverty Assessment Annex 270 267 Following Günther and Harttgen 2009. 268 See, e.g., Chaudhuri, Jalan, and Suryahadi 2002; Tesliuc and Lindert 2004 269 For empirical applications using the 29% threshold, see for Madagascar Günther and Harttgen 2009; Skoufias, Vinha, and Beyene 2021; Gao, Vinha, and Skoufias 2020. 270 Notably, World Bank 2019a, implemented this following López-Calva and Ortiz-Juarez 2014 and Pathways Towards Economic Security Indonesia Poverty Assessment 87 Annex FIGURE A1: Permanent low consumption prospects vs. high consumption volatility ! Definitions Extending the above analysis, the Assessment defines the concepts of structural poverty, economic insecurity and economic security in the following manner: • The structurally poor are those who are poor in the current period and identified as vulnerable to falling into poverty, based on the methods described above; • Economically insecure consist of two groups: (i) those currently poor but not identified as vulnerable to poverty in the next period; and (ii) those not currently poor but identified as vulnerable above; and • Economically secure: neither poor in the current period, nor vulnerable in the next. The analyses of idiosyncratic and covariate susceptibility (vulnerability) is carried out as described above, and reported for the economically insecure households. 88 Pathways Towards Economic Security Indonesia Poverty Assessment Annex Voices: Background Information Lembata Lembata is a district in the province of Nusa Tenggara Timor, in Eastern Indonesia. It consists of a small island off the east coast of much larger Flores and has a population of around 135,000, most of whom are engaged in rain- fed, largely subsistence agriculture, growing vegetables and corn, or in fishing. With poor transport infrastructure, farmers and fishers have limited access to commercial markets for their produce, the majority of which is consumed within the island. Small-holder farmers in particular are heavily dependent on the sale of fine hand-woven cloth for cash income, with this cloth produced almost exclusively by women. In addition, both men and women frequently seek work on neighboring islands or further afield, often sending cash back to their families. In recent years, PEKKA, a women’s union, has been active on the island, organizing savings and loans groups, informal educational activities, and other types of support, enabling poor women to borrow money, using their woven cloth as a form of security. In 2017, an earthquake caused major damage in several sub-districts, including Ile Api, where most of the informants lived. This was followed by droughts, and then, with the pandemic raging across Indonesia, a cyclone (2020) and mudslides (2021) that resulted in hundreds of deaths and in the loss of thousands of homes, livestock, woven cloth, and other productive assets. While informants stated that government agencies played a very active role in the initial post-disaster periods, providing food assistance and temporary housing and rebuilding access roads and other public facilities, this support dried up quite quickly, with NGOs and community organizations playing a greater role. There appears to have been almost no agricultural extension services available, even prior to the disasters, and little or no support in the post-disaster period for agricultural activities. One fisherman stated that over the years, his livelihood had improved as a result of improved road and transport infrastructure, with the price of fish now roughly the same across the island. Some women stated that district agencies had provided grants and other support to enable them to buy looms and weaving materials, and thus to generate cash incomes. Kediri Kediri is a small city (population around 288,000) in the province of East Java. It has large cigarette and sugar production industries, with both tobacco and sugar widely cultivated in surrounding rural areas. Most of the urban poor seek their livelihoods as waged laborers, domestic workers, small time traders, and providers of transport and other services. Many maintain some connections with surrounding villages, where some continue to own agricultural land, although this is often insufficient to sustain them. With good road and rail connections to Surabaya and Malang, both large, industrial cities, many also travelled or migrated to these centers to seek work or business opportunities, at least prior to the pandemic. The city government has introduced programs to ensure that low- income residents have adequate access to medical care, at least for the treatment of emergencies and acute conditions, with informants generally stating that they were satisfied with these services. Prior to the pandemic, many – but not all – informants stated that they had received some form of food and non-food assistance from government agencies, including through the PKH program for families with school-aged children. Informants were often confused regarding their eligibility for these programs, generally waiting for sub-district officials to contact them to invite them to participate. A government urban revitalization project, KOTAKU, is active in the city and conducts activities to improve physical living conditions in the slum areas through the construction of community bathing and toilet facilities and other infrastructure. Pathways Towards Economic Security Indonesia Poverty Assessment 89 Annex Additional Graphs and Tables FIGURE A2: Depth and severity of extreme poverty declined FIGURE A3: Depth and severity of poverty declined continuously, and continuously, with rural rates converging to urban ones since 2014 rural areas have largely caught up with urban ones 0.08 0.30 0.06 0.20 0.04 0.10 0.02 0.00 0.00 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 Rural gap Rural severity Rural gap Rural severity Urban gap Urban severity Urban gap Urban severity FIGURE A4: The vast majority of the poor, insecure and secure reside in FIGURE A5: Except in Java-Bali, the extreme poor are concentrated in the most populous island-regions of Java-Bali and Sumatera rural areas while the economically secure are mainly in urban areas 100 100 80 80 60 Percent 60 Percent 40 40 20 20 0 Extreme Poor Economic Economic Poor Insecure Secure 0 Java -Bali rural Java -Bali urban Extreme Poor Economic Economic Sumatera rural Sumatera urban poor insecure secure Kalimantan rural Kalimantan urban Java-Bali Sumatera Sulawesi Sulawesi rural Sulawesi urban Nusa -tenggara rural Nusa -tenggara urban Kalimantan Nusa-tenggara Maluku & Papua Maluku & Papua rural Maluku & Papua urban 90 Pathways Towards Economic Security Indonesia Poverty Assessment Annex FIGURE A6: Access to quality maternal health care among pregnant FIGURE A7: Share of households with access to women, by wealth quintile services, by poverty status Obstetrician care Electricity 100 1 80 0.8 60 0.6 Problem accessing Urine sample 40 care Internet 0.4 Gas 20 0.2 0 0 PNC Blood sample Facility birth Cellphone Sanitation Lowest 2007 Lowest 2017 Highest 2017 Extreme poor Poor Non poor FIGURE A8: Employment rates by education, from 2001 to 2021 FIGURE A9: Maternal mortality rate and total fertility rates in 2020, by region 100 2.8 400 Maternal Mortality / 100,000 living birth 95 Total fertility rate (15-49 y.o) 2.6 300 Percent 90 2.4 200 85 2.2 100 80 2.0 0 ua ra sia ga ap ne 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 ng -P do ku Te In alu sa of Nu M st Re Primary or below Junior Secondary Senior Secondary All Total fertility rate (15-49 y.o) Maternal mortality / 100,000 Pathways Towards Economic Security Indonesia Poverty Assessment 91 Annex FIGURE A10: Share of sectoral RGDP, by region FIGURE A11: Share of employment by sector and by region Maluku-Papua Maluku-Papua 100 100 80 80 60 60 Percent Percent 40 40 20 20 0 0 Agriculture Manufacturing Low VA services Agriculture Manufacturing Low VA services High VA services Other Industries High VA services Other Industries Nusa Tenggara Nusa Tenggara 100 100 80 80 60 60 Percent 40 Percent 40 20 20 0 0 Agriculture Manufacturing Low VA services Agriculture Manufacturing Low VA services High VA services Other Industries High VA services Other Industries Rest of Indonesia Rest of Indonesia 100 100 80 80 60 60 Percent Percent 40 40 20 20 0 0 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003 2006 2009 2012 2015 2018 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 Agriculture Manufacturing Low VA services Agriculture Manufacturing Low VA services High VA services Other Industries High VA services Other Industries 92 Pathways Towards Economic Security Indonesia Poverty Assessment Annex Table A1: Characteristics of the extreme poor, poor and the economically insecure and secure, 2019 Extreme Poor Economically Economically poor insecure secure Number 7,202,731 53,307,242 104,419,620 120,770,301 Household Age (years) 49 49 50 47 head …was male 86% 88% 86% 84% Size 6.0 5.3 4.1 3.2 Demo- Dependency ratio 0.98 0.81 0.58 0.44 graphic Household Ratio of children to total members 0.38 0.34 0.24 0.19 character- istics Ratio of female to total members 0.51 0.51 0.50 0.50 Ratio of elderly to total members 0.06 0.05 0.10 0.08 Educational attainment of head (years) 5.9 6.6 6.4 10.1 No. of working members 2.1 2.1 1.8 1.7 Working to total working age members 0.67 0.68 0.70 0.74 Ratio of… Employed female to total working age 0.47 0.47 0.46 0.57 female members Employed 87% 89% 88% 86% Education In agriculture 50% 44% 43% 17% & labor In industry 15% 19% 19% 19% market HH head In services 22% 27% 27% 50% was… Self-employed or employer 54% 49% 49% 37% An employee 20% 26% 25% 44% A Casual worker or unpaid family worker 14% 15% 14% 5% Protected water source 79% 83% 86% 94% Electricity 92% 97% 98% 100% Access to HH had Proper sanitation 46% 56% 59% 84% Services access to… Gas for cooking 51% 66% 74% 89% Internet 10% 16% 25% 55% Note: Differences between groups for all estimates were statistically significant at the 95 percent level. Characteristics examined but not shown here included share of household heads that were migrants. Pathways Towards Economic Security Indonesia Poverty Assessment 93 Annex Table A2: Employment and other outcomes among women, by household demographic category Share of Share of working married women by sector Share of working married women married… of employment by type of employment Manufacturing Unpaid/Family Self-employed Share FHHH Agriculture … women Employee Employee working working services services High VA … men Low VA Worker Others Wage Wage No Poor 89% 56% 58% 12% 29% 0% 1% 28% 21% 15% 36% 2% dependents NP 92% 56% 28% 13% 56% 3% 1% 34% 37% 5% 23% 3% Households with >= one >=1 child Poor 97% 47% 44% 15% 39% 1% 1% 35% 24% 10% 31% 4% married couple but no seniors NP 98% 53% 21% 15% 60% 3% 1% 38% 39% 5% 18% 3% >=1 senior Poor 97% 64% 53% 16% 31% 0% 0% 33% 23% 10% 34% 14% but no children NP 95% 63% 25% 12% 59% 3% 1% 33% 42% 5% 20% 14% Seniors & Poor 97% 56% 41% 18% 40% 1% 1% 34% 26% 8% 32% 12% children NP 97% 63% 22% 15% 59% 4% 1% 35% 42% 4% 19% 13% Share of working women by sector of Share of working women by type of Share of… employment employment High VA services … men working Wage Employee Wage Employee Low VA services Manufacturing Unpaid/Family Self-employed Agriculture Share … women FHHH working Worker Others Households with no or at most one No Poor 66% 60% 48% 16% 34% 1% 1% 49% 25% 19% 7% 73% dependents married member NP 79% 65% 23% 15% 55% 5% 1% 43% 48% 7% 3% 54% >=1 child Poor 58% 61% 40% 13% 45% 1% 1% 48% 31% 14% 7% 81% but no seniors NP 65% 60% 23% 14% 58% 4% 1% 47% 42% 8% 4% 71% >=1 senior Poor 69% 55% 52% 15% 33% 0% 0% 36% 25% 13% 26% 35% but no children NP 80% 55% 29% 12% 53% 5% 1% 34% 39% 6% 21% 36% >=1 senior Poor 56% 51% 36% 16% 45% 2% 1% 36% 37% 11% 17% 38% but no children NP 59% 51% 22% 15% 58% 3% 1% 35% 43% 5% 17% 34% Note: Estimates reported for working aged men and women in 2019. Self-employment includes the self-employed as well as employers of temporary and permanent workers. NP = non-poor. FHHH = female headed households. 94 Pathways Towards Economic Security Indonesia Poverty Assessment Annex Table A3: Employment and other outcomes among men and women in male- and female-headed households Share Poverty Share of… Share of working women by sector Share of working women by type of rate of employment employment High VA services Wage Employee Casual Worker …men working Low VA services Manufacturing Unpaid/Family Self-employed Agriculture …women working Worker Other MHHH 84% 22% 86% 48% 48% 12% 39% 1% 1% 33% 26% 7% 34% Rural FHHH 16% 25% 74% 63% 49% 11% 39% 1% 1% 55% 26% 11% 8% MHHH 85% 18% 80% 46% 6% 18% 70% 5% 1% 33% 54% 4% 10% Urban FHHH 15% 18% 72% 60% 6% 17% 69% 7% 1% 37% 56% 5% 3% MHHH 85% 83% 47% 25% 15% 56% 3% 1% 33% 41% 5% 21% National FHHH 15% 73% 61% 25% 14% 56% 4% 1% 45% 43% 8% 5% Note: Estimates reported for working aged men and women in 2019. Self-employment includes the self-employed as well as employers of temporary and permanent workers. FHHH = female headed households. MHHH =male headed households. Pathways Towards Economic Security Indonesia Poverty Assessment 95 96 Table A4: Main social assistance programs in 2019 and 2021, using US$ 3.20 2011 PPP as poverty measure Program Purpose Targeted Cover- Budget (IDR Average Benefit Incident as share Coverage of the Beneficiaries Poverty Re- Inequality Reduc- Budget as Share Poverty Effective- Inequality Effec- age (in million Trillion) (per month and of consumption poorest 10% incidence to the duction Impact tion Impact of GDP (%) ness Index tiveness Index benef.) beneficiary) from poorest poorest eligible (in pp) 10% 10% ‘19 ‘21 ‘19 ‘21 ‘19 ‘21 ‘19 ‘21 ‘19 ‘21 ‘19 ‘21 ‘19 ‘21 ‘19 ‘21 ‘19 ‘21 ‘19 ‘21 ‘19 ‘21 Poverty reduction, PKH 10 10 34.3 28.3 285,833 235,833 4.3% 2.5% 47% 36% 24% 17% 1.0 1.5 0.004 0.006 0.2 0.2 4.60 8.87 0.02 0.03 strengthened human capital Lower cost of PIP 18 20 9.6 10.9 44,000 45,416 1.1% 2.8% 33% 59% 19% 19% 0.4 0.2 0.001 -7E-05 0.1 0.1 6.59 2.49 0.02 0.00 attending school Boost food Sembako 16 19 20.8 49.9 111,000 221,144 0.6% 3.6% 47% 53% 17% 15% 1.0 2.3 0.004 0.009 0.1 0.3 7.86 7.69 0.03 0.03 security Preventing health PBI-JKN 92 97 26.7 48.8 24,080 42,000 4.0% 64% 15% 15% 0.2 -7E-05 0.2 0.3 0.56 0.00 shocks Uncondi- Supplementary tional cash Income buffer 10 17.5 145,500 1.4% 31% 18% 0.8 0.003 0.1 7.97 0.02 transfer from during pandemic MoSA Pre-Em- Training and ployment cash for those Program unemployed 6 20.0 212,500 0.2% 2% 2% 0.4 5E-04 0.1 3.82 0.00 (Kartu Pra or entering the Kerja) labor force Supplementary UCT Village Income buffer 8 28.8 300,000 1.0% 9% 6% 0.9 0.003 0.2 5.42 0.02 Fund during pandemic 877/ KwH Electricity bill for 450 Electricity waivers for the 33 9.5 VA and 0.9% 50% 7% 0.9 0.002 0.1 15.56 0.04 subsidy poor 456/ Kwh for 900 V Wage Wage subsidy for 9 8.8 83,333 0.2% 9% 6% 0.4 6E-04 0.1 7.53 0.01 Subsidy the poor Bulog Rice support Rice assistance 29 3.6 10,225 0.3% 86% 17% 0.3 5E-04 0.0 14.22 0.02 (Rice PKH in during pandemic 2020) Annex Pathways Towards Economic Security Indonesia Poverty Assessment Annex Table A4: Main social assistance programs in 2019 and 2021, using US$ 3.20 2011 PPP as poverty measure Program Purpose Targeted Cover- Budget (IDR Average Benefit Incident as share Coverage of the Beneficiaries Poverty Reduc- Inequality Budget as Share Poverty Effective- Inequality Effec- Pathways Towards Economic Security Indonesia Poverty Assessment age (in million Trillion) (per month and of consumption poorest 10% incidence to the tion Impact Reduction Impact of GDP (%) ness Index tiveness Index benef.) beneficiary) from poorest poorest eligible (in pp) 10% 10% PPKM Staple Staple food Food Card assistance during (Sembako pandemic for 6 7.1 100,000 0.1% 2% 2% 0.2 -5E-06 0.0 5.51 0.00 Jabode- the poor living in tabek in Jabodetabek 2020) Assis- tance for Cash Transfer for Micro-En- micro enterprise 13 15.4 100,000 0.0% 53% 8% 1.4 0.004 0.1 15.69 0.05 treprises during the (Banpres pandemic Produktif) Internet subsidy Internet for students 38 8.5 22,936 0.2% 33% 6% 0.3 4E-04 0.1 6.96 0.01 Subsidy during the pandemic 97 Annex Table A5: Main social insurance programs in 2019 and 2021 Targeted Coverage of the Amount of coverage (million eligible (million Benefit Program Broad purpose contribution individual) individual) description 2019 2021 2019 2021 2019 2021 JKN (social health Preventing health 268 273 233 236 5% of monthly income Class 1: IDR 150,000, Health service fee insurance) shock (salaried) or IDR 42,000 Class 2: IDR 100,000, waiver (non salaried and non Class 3: 35,000 worker) JKK (works Health service ad cash 126 131 29.9 30.6 0.24-1.75% of monthly 0.24-1.75% of Medical treatment, accident benefit) benefit for work related income depending on monthly income home care injury and/or fatality works risk depending on services, and cash works risk benefit JKM (death Cash benefit to 29.9 30.6 0.24-1.75% of monthly 0.3% of monthly Death grant and benefit) beneficiary in event of income depending on income (salaried) funeral grant of death of participant works risk or IDR 6800 (non IDR 42 million and salaried) children schol- arship up to 174 million JHT (old age Ensuring participant 36.5 16.6 5.7% of monthly 5.7% of monthly Lump sum cash saving) has saving when income (salaried) income (salaried) payment on entering retirement or around 2% (non or around 2% (non retirement or in even permanent salaried) salaried) disability JP (pension) Ensuring decent 45 42 16.4 13.3 3% of monthly 3% of monthly Monthly cash living condition for income income payment participant after retirement or during permanent disability 98 Pathways Towards Economic Security Indonesia Poverty Assessment