Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized INDONESIA SPEND HEALTH FINANCING SYSTEM ASSESSMENT MORE, BETTER RIGHT & October 28, 2016 INDONESIA HEALTH FINANCING SYSTEM ASSESSMENT SPEND MORE, RIGHT & BETTER CONTENTS ACKNOWLEDGMENTS v ABBREVIATIONS AND ACRONYMS vi EXECUTIVE SUMMARY 1 section one. INTRODUCTION 9 section two. BACKGROUND 15 Economic Growth, Poverty and Shared Prosperity 17 Macrofiscal Context 23 section three. HEALTH AND UHC OUTCOMES 29 Demographics and Population Health Outcomes 31 Universal Health Coverage 38 section four. HEALTH SYSTEM 43 Health Care Organization and Delivery 45 Human Resources 48 section five. HEALTH FINANCING 55 Government Budgetary Expenditure on Health 61 Social Health Insurance 66 Out-of-Pocket Spending for Health 74 External Financing for Health 80 Efficiency 84 section six. A CASE STUDY ON IMMUNIZATION 89 Outcomes and Determinants 93 Service Delivery 96 Financing and JKN 100 Introduction of New Vaccines 104 section seven. DISCUSSION AND POLICY OPTIONS 107 Policy Options 113 APPENDIXES 117 ACKNOWLEDGMENTS This report was prepared by a team led by Ajay Cernuschi (WHO), Marisa Ricardo (UNICEF, Jakarta), Tandon (Lead Economist, GHN02) and comprising Logan Brenzel (Bill and Melinda Gates Foundation), Eko Pambudi (Research Analyst, GHN02), Pandu Mike McQuestion (SABIN), and Michael Borowitz, Gail Harimurti (Senior Health Specialist, GHN02), Emiko Steckley, and Di Wu from the Global Fund team. Masaki (Senior Economist, GHN05), Ali Subandoro (Health Specialist, GHN02), Puti Marzoeki (Senior The team benefited from valuable insights provided Health Specialist, GHN02), Vikram Rajan (Senior by representatives of the Indonesian Government, Health Specialist, GHN02), Darren W. Dorkin (Program including Donald Pardede (Senior Adviser to the Coordinator, GOKMU), Amit Chandra (Consultant), Minister of Health), Kalsum Komaryani and Widiarti Chantelle Boudreaux (Consultant), Melissa Chew (Center for Health Financing and Insurance - P2JK), (Consultant), and Nugroho Suharno (Consultant). The Pungkas B. Ali (Bappenas) and the Ministry of team received tremendous support from Christina Finance, especially from Ronald Pasaribu and Sukmawati in completing this report Nasrudin Djoko Sarono (Fiscal Policy Board). v The team would like to thank colleagues from During the report writing, the study results have been the MFM GP consisting of Yue Man Lee (Senior presented on several occasions and received useful Economist, GMF10), David Elmaleh (YP, GMF01), feedback from the audience. The team is grateful Ruth Nikijuluw (Consultant, GMF10), Indira Maulani to Prastuti Soewondo and her team from TNP2K, Hapsari (Consultant, GMF10), Yus Medina Pakpahan and Laksono Trisnantoro and his team from School (Consultant, GMF10), and Dwi Endah Abriningrum of Medicine, Gadjah Mada University in Yogyakarta. (Research Analyst, GMF10) for data and preliminary In addition, Indonesian academics and researchers, analysis. Key contributions for the tobacco section Amila Megraini and Ratih Oktarina from University were received from Patricio V. Marquez (Lead Health of Indonesia health research institutes contributed Specialist, GHN03) and Violeta Vulovic (Consultant, through data and useful comments. The report was GMF10). The team is grateful for the guidance and edited by Chris Stewart. The cover design and layout resources provided by the Global Solution group on was done by Indra Irnawan. health financing team led by Christoph Kurowski (GSG Lead Health Financing, GHNDR), and consisting of The team would also like to acknowledge that this Sarah Alkenbrack (GHN05), and Reem Hafez (GHN05). work is funded by PFM MDTF supported by (Canada, European Union, and Switzerland). Additional financial The team also received substantive inputs during the support for this report was provided by the Australian decision review process from peer reviewers: John Government’s Department of Foreign Affairs and Leigh (DFAT), Mardiati Nadjib (University of Indonesia), Trade, and Gavi, The Vaccine Alliance. Mirja Channa Sjoblom (Senior Economist, GHNGF), Aparnaa Somanathan (Senior Economist, GHN03), Overall guidance was provided by Rodrigo A. Chaves Kathleen A. Wimp (Senior Public Sector Specialist, (World Bank Indonesia Country Director, EACIF), GGO14), Cut Dian Agustina (Economist, GGO14), Cristobal Ridao-Cano (Program Leader, EACIF), and Robert Wrobel (Senior Social Development Specialist, Toomas Palu (Practice Manager, GHN02). GSU03), Karrie McLaughlin (Consultant, GSUID), Lluis Vinyals Torres and Phylida Travis (WHO SEARO), Tania In memory of Elif Yavuz. Abbreviations & Acronyms AEFI Adverse Events Following Immunization DFAT Department of Foreign Affairs and Trade ANC Antenatal Care DHO District Health Office APBD Anggaran Pendapatan dan Belanja DOEN Daftar Obat Esensial Nasional (National Daerah (Local Government Budget) Essential Drug List) APBN Anggaran Pendapatan dan Belanja DPT Diphtheria Pertussis Tetanus Nasional (Central Government Budget) DQS Data Quality Self-assessment APBN-P Anggaran Pendapatan dan Belanja EAP East Asia and Pacific Nasional Perubahan (Revised Central Government Budget) EPI Expanded Program for Immunization Askes Asuransi Kesehatan (social health EVM Effective Vaccine Management insurance scheme for civil servants, FCTC Framework Convention on Tobacco now merged into JKN) Control ARV Antiretro Viral FORNAS Formularium Nasional (National Badan POM Badan Pengawasan Obat dan Makanan Formularium) also referred to as BPOM (National Food Gavi Global Alliance for Vaccines and and Drug Control Agency) Immunization (Gavi, The Vaccine Bappenas Badan Perencanaan Pembangunan Alliance) Nasional (National Development GDP Gross Domestic Product Planning Agency) GNI Gross National Income BCG vaccine Bacillus Calmette-Guérin vaccine GoI Government of Indonesia BEONC Basic Emergency Obstetric and Neonatal Care GST General Sales Tax BIAS Bulan Imunisasi Anak Sekolah (School HDI Human Development Index Children Immunization Month) HIV/AIDS Human Immunodeficiency Virus/ BKKBN Badan Koordinasi Kependudukan dan Acquired Immune Deficiency Syndrome Keluarga Berencana Nasional (National HPV Human Papilloma Virus Population and Family Planning HRH Human Resources for Health Coordination Board) IDA International Development Association BPJS Badan Penyelenggara Jaminan Sosial (Social Security Management Agency) IDHS Indonesia Demographic and Health Survey BRICS Brazil, Russia, India, China, South Africa IDR Indonesian Rupiah CBR Crude Birth Rate IFLS Indonesia Family Life Survey DAU Dana Alokasi Umum (General Allocation Fund) IHME Institute of Health Metrics and Evaluation DAK Dana Alokasi Khusus (Special Allocation Fund) IMF WEO International Monetary Fund World Economic Outlook DALYs Disability-adjusted Life Years INA-CBG Indonesia Case-based Groups DBH Dana Bagi Hasil (Revenue Sharing Fund) (diagnosis-related groups) Decon Deconcentration Funds IPV Inactivated Polio Vaccine DJPK Direktorat Jenderal Perimbangan ITAGI Indonesian Technical Advisory Group on Keuangan (Directorate-General of Fiscal Immunization Balance) Jamkesmas Jaminan Kesehatan Masyarakat PER Public Expenditure Review (Community Health Insurance Scheme; PHO Provincial Health Office government-financed health coverage for poor and near poor, now merged PMK Peraturan Menteri Keuangan (Ministry of into JKN) Finance Regulation) Jamkesda Jaminan Kesehatan Daerah (local PNS Pegawai Negeri Sipil (Civil Servant) government-financed health coverage) PPP Purchasing Power Parity Jamsostek Jaminan Sosial Tenaga Kerja (social PTT Pegawai Tidak Tetap (Temporary/ health insurance for private sector Contracted Civil Servant) employees, now merged into JKN) Puskesmas Pusat Kesehatan Masyarakat JE Japanese Encephalitis (Community Health Center) JKN Jaminan Kesehatan Nasional (National Poskesdes Pos Kesehatan Desa (Village Health Health Insurance) Post) LKPP Lembaga Kebijakan Pengadaan Posyandu Pos Pelayanan Terpadu (Integrated Barang/Jasa Pemerintah (Government Health Services Post) Goods and Services Procurement Policy Institute) Polindes Pondok Bersalin Desa (Village Maternity Clinic) MDG Millennium Development Goals Rifaskes Riset Fasilitas Kesehatan (Health Facility MDR-TB Multidrug-Resistant Tuberculosis Research) MDTF Multi-donor Trust Fund RSUD Rumah Sakit Umum Daerah (Local MENPAN Kementerian Negara Pendayagunaan General Hospital) Aparatur Negara (Ministry of SARA Service Availability and Readiness Administrative and Bureaucratic Assessment Reform) SDG Sustainable Development Goals MoH Ministry of Health SIKD Sistem Informasi Keuangan Daerah MMR Maternal Mortality Ratio (Subnational Financing Information MR Measles Rubella System) MSS Minimum Service Standards SJSN Sistem Jaminan Sosial Nasional (National Social Security System) MTEF Medium-term Expenditure Framework Susenas Survei Sosial Ekonomi Nasional NCD Noncommunicable Diseases (National Socioeconomic Survey) NGO Nongovernment Organization TB Tuberculosis NHA National Health Accounts THE Total Health Expenditure NIHRD National Institute for Health Research TNP2K Tim Nasional Percepatan and Development Penanggulangan Kemiskinan (National NIP National Immunization Program Team for the Acceleration of Poverty NTP National Tuberculosis Program Reduction) OECD Organisation for Economic Co-operation TP Tugas Pembantuan (Co-Administered and Development Tasks) OECD CRS OECD Creditor Reporting System UHC Universal Health Coverage OOP Out-of-pocket UN United Nations OPV Oral Polio Vaccine UNICEF United Nations Children’s Fund PAD Pendapatan Asli Daerah (Local USAID United States Agency for International Government Own-source Revenue) Development PBI Penerima Bantuan Iuran (Recipient of VAT Value-added Tax government paid preimum) VPD Vaccine-preventable Disease PEFA Public Expenditure and Financial WB World Bank Accountability WEO World Economic Outlook Pentavalent Diphtheria Pertussis Tetanus - Hepatitis WHO World Health Organization B - Haemophilus Influenzae Type B vaccine HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better viii EXECUTIVE SUMMARY 1 EXECUTIVE SUMMARY HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better Health financing is not only about assessing the sufficiency of resources, but also about how resources are equitably and efficiently raised, pooled, and allocated to make progress towards UHC. This report assesses Indonesia’s health financing Indonesians have undoubtedly become healthier in system. As an intrinsic and necessary element of recent decades and important progress has been universal health coverage (UHC), health financing is made on key health indicators. Life expectancy at birth not only about assessing the sufficiency of resources, has steadily increased to 69 years in 2014, up from 63 but also about how equitably and efficiently resources years in 1990 and only 49 years in 1960. The under- are raised, pooled, and allocated to make progress five mortality rate has declined from 222 per 1,000 towards UHC. live births in 1960 to 85 in 1990 and 27 in 2015; Infant mortality has declined six-fold since 1960, down to 23 Indonesia has made key strides towards attaining per 1,000 live births in 2015. UHC in terms of population coverage. In 2015, nearly 160 million individuals, or more than 60 percent of the Nevertheless, key challenges remain, especially population, have been covered by Jaminan Kesehatan with regard to maternal health, malnutrition, as Nasional, or JKN, one of the largest single-payer well as persistent inequalities in health outcomes. 2 social health insurance (SHI) programs in the world; The maternal mortality ratio (MMR) remains high by 2019, everyone in Indonesia is expected to have at 126/100,000 live births, far above the 2030 SDG coverage under JKN. Nevertheless, Indonesia faces target of less than 70 per 100,000 live births. At the key challenges in order to meet its 2019 population same time, 37 percent of under-five children are coverage target as well as on other, arguably more stunted, while 12 percent are wasted. Large regional important, dimensions of UHC, including service and income-related inequalities in health outcomes coverage and financial protection. remain across the country. Albeit still relatively low, EXECUTIVE SUMMARY Years Mortality 250 75 rate per 1,000 Under-five mortality (left axis) Life expectancy (right axis) 70 live 150 births 65 100 Infant mortality (left axis) 60 50 55 50 25 45 1960 1970 1980 1990 2000 2010 2015 Source World Development Indicators database Note y-scales logged the prevalence of HIV and AIDS is growing; and population lived on $3.1-a-day and $1.9-a-day pockets of the country continue to face challenges of (respectively) in 2010, down from 82 percent and 48 communicable diseases such as malaria and TB. percent (respectively) in 1999. 3 New challenges are rapidly emerging with an Despite impressive gains in poverty reduction, ageing population and a rising prevalence of chronic the level of informality in the labor market has diseases which the health system is ill-equipped remained persistently high in Indonesia and income to address. A rapid increase in the share of the inequality is raising rapidly. The bottom 40 percent population ages 65 and above is expected to occur of the population saw an average growth in real per beginning around 2015. Almost 70 percent of the capita consumption of only 1-2 percent per year disease burden is now due to noncommunicable over the period 2003-10; by way of contrast, the diseases (NCDs) and this is expected to grow top 20 percent increased their consumption by 5-6 rapidly in coming years as Indonesia completes its percent per year. This has resulted in a dramatic rise epidemiological transition. Supply-side readiness in income inequality, one of the largest increases is a key challenge, especially in the eastern part of in the East Asia and Pacific (EAP) region. Over 60 the country. Managing, regulating, and integrating percent of those employed continue to be classified a growing private sector under the UHC umbrella is as nonsalaried workers. Given declining poverty another key issue. rates, this indicates a growing share of the nonpoor informal sector in the population. Indonesia is currently classified as a lower-middle- income country with GNI per capita of US$3,238 Health services in Indonesia are delivered through (US$10,680 in PPP terms) in 2015. Indonesia first both public and private providers. The public sector transitioned from low-income to lower-middle- generally takes a dominant role in rural areas and for income status in 1992. The country was, however, secondary levels of care, but this is not necessarily reclassified as a low-income country in 1998 the case across all health services. Private provision as a result of the 1997-98 Asian financial crisis, has been increasing rapidly in recent years, including but regained its lower-middle-income status for primary care. The country has 34 provinces, 514 in 2003. Indonesia’s relatively strong economic districts/cities, and some 72,000 villages. Public growth (5.5 percent per year since 2000) has been provision is decentralized to the district/city level. As a accompanied by a sustained decline in poverty country with over 6,000 inhabited islands, geography rates: about 46 percent and 16 percent of the poses a significant obstacle to service delivery. HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better From a health financing redistributive capacity of the health-financing system and are, therefore, undesirable. Although a relatively perspective, the small share (1 percent) of households face OOP health expenditures that are deemed “catastrophic” (that is, in relatively low quantum of excess of 25 percent of total household expenditure), 8 percent of all households (an estimated 7 million overall health spending households) are either impoverished or pushed deeper into poverty as a result of high OOP spending in Indonesia is one of the using a national poverty line. key bottlenecks toward The OOP spending share of THE has remained achieving UHC. in excess of 45 percent since 1995. Despite the increasing population coverage by SHI from 15 percent to 60 percent, OOP spending on health is unlikely to decline unless there is significant expansion From a health financing perspective, the relatively in JKN coverage, an improvement in supply-side low quantum of overall health spending in readiness at health facilities, and increased inclusion Indonesia is one of the key bottlenecks toward of branded drugs in the JKN benefits package. Use achieving UHC. This is a result of a combination of of branded pharmaceuticals that are not covered factors, including relatively low overall government by the JKN package is one of the key drivers of OOP revenue generation, low prioritization for health, spending even among those covered. high levels of informality, and low utilization rates. Global and regional benchmarks indicate that Government budgetary supply-side spending is the 4 Indonesia’s health system remains significantly second largest component of health financing in under resourced. At 3.1 percent of GDP, Indonesia’s Indonesia. Despite recent increases, the overall level total health expenditure (THE) levels are among the of public financing remains very low at 1.5 percent lowest in the world, and are particularly low when of GDP, one of the lowest in the world. This is partly benchmarked against other lower-middle-income a result of low revenue generation capacity of the countries and across the EAP region. country: Indonesia’s revenue share of GDP was only 15 percent in 2015, far lower than the average for Although Indonesia is following an SHI model for lower-middle-income countries (28 percent) and attaining UHC in principle, in reality, the health system less than one-half the average for the EAP region is financed through a combination of sources and (38 percent). Health is accorded a generally low disparate flows. The four primary sources of health priority as reflected in the small share of the national financing in the country include OOP spending by budget. Health’s share of the central government households, government budgetary supply-side budget has remained less than 3 percent and only health spending (both at the central and subnational in 2016 did it increase to 5 percent. The health share levels), SHI, and external financing. Despite increases of total government budget at 4.7 percent is low in public financing in recent years, the fundamental in comparison to several countries in the region, structure of health financing has remained largely including the Philippines, China and Thailand. unchanged in Indonesia because of concomitant increases in OOP spending for health. SHI is the third largest source of health financing– although JKN covers more than 60 percent of the OOP spending by households–a generally inefficient population, it accounts for 13 percent of THE. Almost and inequitable modality–remains the largest source one-half of JKN expenditure is currently sourced of financing for health. Although there is evidence from the central government in the form of premium that OOP spending on health is relatively progressive payments for the poor/near-poor and significant in Indonesia as the rich paid a higher share of total cofinancing from government budgetary expenditure expenditures as OOP spending, the high levels of remains at public facilities. Although Indonesia has OOP spending deter utilization by the poor. Moreover, successfully instituted a single-payer SHI system, high levels of OOP spending reduce the potential contribution collection among nonpoor informal EXECUTIVE SUMMARY workers has been difficult (under current regulations, this group must contribute in order to enroll in JKN), Cross-cutting Issues thus JKN coverage for this population group has been limited. Few nonpoor informal participants have affecting the overall enrolled to date and those who have are adversely performance of selected, undermining equity, and threatening financial sustainability of JKN. Provider payment Indonesia’s health mechanisms under JKN are “passive” in that there are no explicit linkages with outputs/outcomes. JKN system. offers comprehensive benefits, yet JKN’s current reimbursements do not cover the full cost of care. Although external sources, the fourth largest source of financing, account for only 1 percent of THE, care (50 percent) and outpatient care (15 percent). they remain an important source of financing and About 20 percent of the expenditure was on capitated technical assistance for immunization, HIV, TB, and primary care at puskesmas and empaneled private malaria programs. In 2015, MOH estimated that the clinics. A very small amount–less than 1 percent–went external share of the total program spending was towards preventive and promotive activities. high as 60 percent for tuberculosis (TB) , through reduced from around 70 percent in 2014; it is lower Complex and fragmented interfiscal government for immunization programs at around 10-15 percent, transfers in a decentralized system resulted in wide with an increasing trend with the introduction of new variations of health spending across districts. While vaccines financed by Gavi. The smooth transition of the bulk of government health expenditure occurs externally financed health programs, such as HIV, at the district level, the central government remains 5 TB, malaria and immunization, is crucial to that gains the dominant source of revenues. In Indonesia’s made in recent years are sustained. decentralized context, interfiscal government transfer is significant, yet the system of intergovernmental The integration of service delivery of externally transfers to districts is complex and fragmented: some financed and vertically managed programs into JKN earmarked for inputs while, for the remainder, district in a decentralized setting has become one of the key governments have discretion over how budgets are policy discussions to ensure the sustainability of these allocated and the amount to be spent on health. programs. Integrating these programs into the health system, including JKN, will entail more than addressing The central government does not have mechanisms actuarial matters related to which services should be to incentivize generation of outputs/outcomes included. It needs to be discussed within the overall from use of resources, nor does it have clear policy health system context and take into account all the levers to influence the allocation of resources at the health system pillars. This includes: (i) preparedness subnational level. There are wide variations in district- to provide included services; (ii) better responsiveness level health expenditures although, in aggregate, and sensitivity to the needs of specific target population districts do spend 10 percent of their budget on health groups; and (iii) provider-payment mechanisms that (as required by law). Some districts view health as incentivize providers to reach out to target beneficiaries a revenue-generating sector and have targets for and retain them in the treatment cascade. resources generated by user fees at public health facilities; these are pooled with other sources of As Indonesia’s health system develops, the key is to revenues and allocated across sectors. improve efficiency of its system and to ensure that health expenditures lead to the greatest improvement There is a fundamental disconnect between in health outcomes. Health expenditures are largely the level and geographic distribution of public focused on curative and rehabilitative care. Hospital financing for health and JKN benefits offered, accounts for the largest share of THE, followed by leading to inequities in the incidence of social health providers of ambulatory care. More than 65 percent expenditure and to implicit rationing. Important of JKN expenditures were for hospital-based inpatient challenges remain with regard to mistargeting HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better of the poor and nonpoor, covering nonpoor to ensure the adequacy of service and financing. informal workers and, more generally, the financial In the absence of an explicitly defined benefit sustainability of the JKN program. package, providers refer to various national clinical guidelines. As a result, there are variations The integration of service delivery of externally in standards of practice and case management financed and vertically managed programs into JKN that ultimately results in inefficiency of service in a decentralized setting has become one of the delivery. Furthermore, the JKN benefits package key policy discussions to ensure the sustainability needs to be adjusted so that it is commensurate of these programs. Integrating these programs into with current public financing resources, economic the health system, including JKN, will entail more growth and projected macrofiscal trajectory, and than addressing actuarial matters related to which service delivery capacity. Service- readiness, and the services should be included. It needs to be discussed capacity of health facilities to provide interventions within the overall health system context and take into in key program areas remains major challenge in account all the health system pillars. This includes: deliver services under the JKN benefits package. (i) preparedness to provide included services; (ii) Some 30 percent of puskesmas lacked the ability better responsiveness and sensitivity to the needs of to do hemoglobin tests and about 50 percent of specific target population groups; and (iii) provider- puskesmas lacked the ability to do urine tests. payment mechanisms that incentivize providers to reach out to target beneficiaries and retain them in Ensure adequate public financing for UHC. Given the the treatment cascade. very low level of government health spending at 1.5 percent of GDP, it is crucial to increase government In moving forward, Indonesia can seek several health spending as a necessary and critical but not opportunities that exist for improving the efficiency, sufficient, condition to progress towards achieving 6 effectiveness and sustainability of the health UHC. In acknowledging the challenges of increasing system. From a health-financing perspective, the fiscal space for public financing for UHC, key some key policy recommendations include: (i) options that can address this include a combination making the benefit package explicit; (ii) improving of: (i) increasing overall government revenues through supply-side readiness; (iii) strengthening primary improved tax collection; (ii) reprioritization of health care; (iv) reducing OOP payments by expanding in the government’s budget; (iii) efficiency gains; and deepening coverage; (vi) enhancing the (iv) earmarked tobacco taxes; (v) complementary effectiveness of intergovernmental fiscal transfers; subnational financing; (vi) increasing enrollment of (vii) strengthening JKN linkages with externally the remaining formal sector; and (vii) incentives to financed health programs; and (viii) enhancing cross- formalize the informal sector. subsidization from prepaid/pooled health resources. Explicitly defining the benefit package is crucial Increase focus on primary health care, including prevention and promotion. There are concerns that the focus on UHC is for curative and rehabilitative In order to accelerate care and is distracting from the focus on improving primary health care and population/public health progress toward interventions. Most cost-effective interventions are usually delivered at the population level as well as the UHC and meeting its primary-care level. population coverage Cover the nonpoor and eliminate mistargeting. Given target by 2019, Indonesia challenges of public financing, supply-side readiness, equity in, and financial sustainability of, social would have to Spend health expenditure incidence, and implicit rationing, availability of benefits, enrollment of the nonpoor More, Spend Right and into JKN continues to be a challenge. Expanding the coverage for the nonpoor informal sector and the Spend Better. elimination of mistargeting need to be key priorities. EXECUTIVE SUMMARY Increase effectiveness of intergovernmental fiscal of financing required, but also on the governance transfers by improving local government capacity, and service-delivery mechanisms in place to deliver ensuring accountability, and incentivizing results. these services. As JKN expands coverage, the key to Efficiency and effectiveness of Indonesia’s health financial and institutional sustainability will be for these system can be enhanced by improving local externally financed health programs to be better government’s (provincial and district) capacity to integrated within the context of UHC. prioritize, mobilize, plan, budget, and effectively utilize both supply- and demand-side financing; Leveraging JKN provider payment mechanisms strengthening the monitoring and evaluation system to incentivize preventive/promotive services for to make local governments more accountable; and results. Improved socialization of guidelines on introducing nonfinancial and financial incentives tied use of JKN capitation payments would help as to achievement of results. would other mechanisms such as introduction of “strategic” purchasing, e.g., to better integrate Stronger and clearer links to JKN is key to the JKN provider payment mechanisms with provision sustainability of externally financed health programs. of preventive/promotive care so as to improve To transition from externally financing smoothly, the efficiency and financial sustainability of public Indonesia, needs to focus not only on the quantum financing for UHC in Indonesia. Policy Options 7 SPEND MORE • Make the JKN benefits package explicit so that current public financing gaps can be clearly identified and estimated. • Adjust the JKN benefits package so that it is commensurate with current public financing resources, economic growth and projected macrofiscal trajectory, and service delivery capacity. • Raise additional public financing for health by: (i) increasing overall government revenues through improved tax collection and introduction of higher “sin” taxes including on tobacco; (ii) encouraging labor formality; (iii) reprioritization for health in the government’s budget; and (iv) increasing enrollment of the remaining formal sector. • Increase and expand coverage of the nonpoor informal sector. SPEND RIGHT • Focus on primary health care including prevention and promotion. • Reduce mistargeting for the poor and nonpoor and ensure subsidies are spent on the right people through better targeting. • Integrate supply-side and demand-side financing to improve public and private provider supply- side readiness. SPEND BETTER • Increase effectiveness of intergovernmental fiscal transfers by improving subnational government capacity, ensuring accountability, and incentivizing results. • Adjust JKN provider payment mechanisms to incentivize preventive/promotive services for results. HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better 8 section one. INTRODUCTION section 1 . 9 INTRODUCTION HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better This policy paper assesses Indonesia’s health about the extent of financial protection accorded financing system in light of recent reforms and the by the health systems.2 Health financing refers to government’s commitment to attaining universal the “function of a health system concerned with the health coverage (UHC) for its population by 2019. mobilization, accumulation, and allocation of money Landmark legislation in 2004 and 2011 has helped to cover the health needs of the people, individually realize a potential pathway to UHC in Indonesia and and collectively, in the health system…the purpose of social health insurance (SHI) coverage rates have health financing is to make funding available, as well increased significantly in recent years. As of 2014, as to set the right financial incentives to providers and when Indonesia merged its various SHI schemes, to ensure that all individuals have access to effective the country has one of the largest single-payer public health and personal health care” (WHO 2000). population coverage programs in the world (Jaminan Kesehatan Nasional, JKN). By 2019, everyone in All health financing approaches should try to fulfill Indonesia is intended to have coverage under JKN. three basic principles of public finance: (i) raise enough revenues to provide individuals with the The overarching goal of this assessment is to intended packages of health services that assure identify critical constraints and opportunities facing health and financial protection against catastrophic Indonesia’s health financing system in order to help medical expenses caused by illness and injury in accelerate and sustain progress towards UHC. As the an equitable, efficient, and financially sustainable country gears towards attaining UHC and prepares manner; (ii) manage these revenues to pool health to gain upper-middle income status, it is also risks equitably and efficiently; and (iii) ensure transitioning away from, and integrating, traditional the payment for, or purchase of, health services external-financed health programs, such as those for is carried out in ways that are allocatively and immunization, Human Immunodeficiency Virus (HIV), technically efficient.3 In doing so, health financing 10 tuberculosis (TB), and malaria. In order to identify focuses specifically on two dimensions of the UHC these constraints and opportunities, as well as to ‘cube’: (i) the height of the cube, representing the assess Indonesia’s overall health financing system, extent of financial protection accorded by direct the paper also includes an in-depth examination, costs at the time and point of seeking care; (ii) the using immunization as a case study, both for the volume of the inner cube, representing the extent context of UHC reforms as well as the country’s of prepaid/pooled financing (a function of all three impending exit from donor financing.1 dimensions of the UHC cube, that is, of the number of people covered, the services covered, and of Health financing is instrumental for, and intrinsic the extent of financial coverage provided by health to, UHC. UHC can be conceptualized as consisting systems); and (iii) the volume of the outer cube of three key dimensions: (i) population coverage representing the aggregate amount of total health (“breadth” of coverage); (ii) service coverage (“depth” expenditures (THE) in the country. of coverage); and (iii) cost coverage (“height” of coverage)(Figure 1.1). UHC is not only about Another way to conceptualize the link between increasing the number of people having access to health financing and UHC is by using WHO’s health services, although this is clearly one important “building blocks” framework. WHO defines a health dimension of UHC, but also about ensuring that system as “…the sum total of all the organizations, services are available and of sufficient quality and institutions, and resources whose primary purpose is 1 A subsequent policy paper will examine HIV, malaria, and TB in the same contexts, building on previous work done on HIV and reflecting new service delivery data that are being collected from a national sample of public and private primary-care facilities. 2 The three dimensions of UHC (“depth”, “breadth”, and “height”) are neither independent nor mutually exclusive: ensuring depth of coverage has implications for the breadth and height of UHC as well. Universal availability of the benefit package for all–not just those who are well-off and live in urban areas–is a key aspect in ensuring that UHC is not a hypothetical aspiration but a realized policy designed to enhance health and improve social protection. High out-of-pocket (OOP) payments–that is, the low height of UHC–can (and is) often a result of poor depth of coverage if patients have to pay OOP for drugs or seek care elsewhere in private facilities that are outside the network. 3 World Bank, “Financing”, World Bank, http:/ /go.worldbank.org/I9NCO1V9N0, accessed January 4, 2016. section one. INTRODUCTION to improve health” (WHO 2014). WHO conceptualizes (as well as other objectives such as improved health systems as comprising six core “building responsiveness and enhanced health security) blocks”: (i) service delivery; (ii) health workforce; (iii) (Figure 1.2)(WHO 2013a). health information systems; (iv) access to essential medicines; (v) financing; and (vi) leadership/ Sufficiency of financing for UHC is typically a governance (WHO 2010a). These six “building prominent policy consideration across many blocks” represent inputs and processes that, when developing countries, including Indonesia. Resources combined, generate outputs, outcomes, and impact needed for financing UHC depend in large part on for attainment of desired objectives such as UHC country context, the extent of population coverage, Figure 1.1 Three Dimensions of UHC4 DIRECT COST PROPORTION OF THE COST COVERED REDUCE COST SHARING INCLUDE AND FEES OTHER SERVICES 11 CURRENT POOLED EXTEND TO FUNDS NON - COVERED SERVICES: WHICH SERVICES ARE COVERED? POPULATION: WHO IS COVERED? Figure 1.2 Results Chain From “Building Blocks” to UHC and Other Health Systems Objective BUILDING BLOCKS OUTPUTS OUTCOMES IMPACT Service Delivery Service access and readiness Health workforce Improved health status Service quality and safety Coverage of interventions Health information systems Improved financial well-being Service utilization Financial risk protection Access to essential medicines Increased responsiveness Financial resources pooled Risk factor mitigation Financing Increased health security Crisis readiness Leadership / Governance QUANTITY, QUALITY AND EQUITY OF SERVICES SOCIAL DETERMINANTS Source Adapted from WHO (2013) 4 WHO. 2010a. HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better the risk profile of beneficiaries and their utilization rates, the costs of inputs, the nature and extent of benefits provided, and how health systems are organized and financed to deliver services. At the same time, resource availability for financing UHC is dependent on the willingness and ability of beneficiaries to contribute, the administrative capacity of countries to collect contributions, the fiscal capacity of governments to subsidize coverage for those who are not able to contribute, and the extent of cross-subsidization from richer to poorer beneficiaries, among other factors. Health financing, however, is not just about the sufficiency of resources; it is also about the efficiency, equity, and effectiveness of how such resources are raised, pooled, allocated, and utilized to attain the desired health system outcomes, such as UHC (Hsiao 2007). Implications of health-financing strategies can also include assessments related to financial sustainability and the impact of reforms on the broader economy. UHC-related health-financing reforms can potentially improve health outcomes, 12 mitigate household vulnerability, and reduce the risk of impoverishment from catastrophic health spending. Health financing reforms can, however, also have unintended consequences, for example, policies to improve revenue collection may result in increasing labor costs, encouraging informality, as well as unduly raising the fiscal burden on governments (Wagstaff 2010). Rising health care costs, if not mitigated by strategic purchasing and efficiency improvements, can threaten the financial sustainability of health care reforms. With implementation of UHC in countries that have externally financed programs, there are additional challenges related to whether or not benefits packages adequately stipulate and deliver UHC. These subareas of focus are not necessarily comparable services to those that were previously mutually exclusive, and some of the trade-offs and externally financed, and to what extent some health complementarities across the different subareas programs continue to be managed separately from are acknowledged and addressed as they come UHC implementation modalities. up. Equity and efficiency considerations underpin all subareas of focus and are cross-cutting themes Given this backdrop, the remainder of the paper throughout the assessment. focuses on describing and analyzing three broad questions and subareas of focus in assessing The paper is structured into seven sections, including Indonesia’s health-financing system, namely, how this introduction. Section Two (Background) begins equitable and efficient the health system is in raising, with a summary on Indonesia’s country context, pooling, and allocating resources to purchase including economic growth, poverty, shared health services in Indonesia’s quest to attaining prosperity, and a discussion of the macrofiscal section one. INTRODUCTION 13 environment within which the country’s health- agents of health financing in the country: government financing system operates. This is followed by an budgetary expenditures (both at the central and overview of Indonesia’s attainment of key population subnational levels); SHI expenditures; OOP spending health outcomes and progress towards UHC on health by households; and external financing. (Section Three). Section Four examines health care Section Six takes a close look at immunization as a organization, delivery and resources. In Section disease-specific context for health financing. Section Five, the paper recaps Indonesia’s health-financing Seven concludes the paper with a discussion and system, with a focus on four of the largest sources and some policy options for consideration by.5 5 Whenever the term “national” is used with regard to government-related indicators, this is a reference to both central and subnational taken together HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better 14 section two. BACKGROUND section 2 . 15 BACKGROUND HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better In Summary 1. Indonesia has made significant progress and has a positive macroeconomic outlook with economic growth projected at a respectable 5-6 percent annually for the next five years. 2. Inflation and unemployment are low and Indonesia has manageable levels of debt and fiscal deficit. 3. There has been an overall decline in poverty, but rising income inequality and a persistence in labor market informality. 4. National (central plus subnational) spending is low relative to other countries with comparable income level, and the national revenue collection is also low. 5. Revenue collection is largely centralized, while expenditure and service delivery has been decentralized to the district level. There is a clear disconnect between revenue collection and expenditures. 6. Indonesia’s intergovernmental fiscal transfers of revenues are large, fragmented, and complex. 7. Conditional, earmarked capital grants are allocated by central government as an equalizing mechanism to prioritize some service sectors, including health. 16 section two. BACKGROUND Economic Growth, Poverty and Shared Prosperity Indonesia has made significant progress since the Indonesia’s GDP grew at an average annual rate of 1997-98 Asian financial crisis which resulted in a 4.1 percent over the 1995-2015 period, slightly better, decline in GDP of over 15 percent. With GNI per but relatively more volatile, than its regional peers, capita of US$3,238 (US$10,680 in PPP terms) in 2015, with an average economic growth of 2.8 percent in Indonesia is currently classified as a lower-middle- per capita terms over the same period (Table 2.1). income country and ranks next to Swaziland, Bolivia, Indonesia’s relatively strong economic growth post Philippines and Egypt (and to Dominica, Bosnia, 1997-98 Asian financial crisis (5.5 percent per year and Egypt in PPP terms). Indonesia first transitioned since 2000 - Figure 2.2) has been accompanied by a from low-income to lower-middle-income status sustained decline in poverty rates: about 46 percent in 1992. The country, however, was reclassified as and 16 percent of the population lived on $3.1-a-day a low-income country in 1998 as a result of the and $1.9-a-day (respectively) in 2010, down from 82 1997-98 Asian financial crisis, but regained its lower- percent and 48 percent (respectively) in 1999. About middle-income status in 2003 (Figure 2.1). In parallel, 10 percent of the global share of people subsisting Indonesia became reeligible for concessionary IDA on less than $1.9 a day are in 2010 estimated to live 17 credits in 1999 as a “blend” country and regained full in Indonesia. The national poverty rate stands at 11 status with the International Bank for Reconstruction percent in 2014. and Development in 2008.6 Figure 2.1 GDP per capita and Poverty Trends in Indonesia (1995-2015) GDP 3,500 100 Share of per population capita (%) US$ $3.1 a-day poverty (right axis) 80 3,000 IBRD 60 2,500 $1.9 a-day poverty (right axis) 40 Blend IBRD 2,000 20 LOWER LOWER MIDDLE LOW MIDDLE INCOME INCOME INCOME 1,500 0 1993 1996 1999 2002 2005 2008 2011 2015 Source World Development Indicators 2016. Note GDP per capita in 2015 constant US$. 6 The World Bank defines ‘blend’ countries as being IDA-eligible based on per capita income levels and are also creditworthy for some IBRD borrowing HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better Figure 2.2 Year-on-year Economic Growth (1995-2015) percentage (%) 10 5 0 -5 -10 -15 1995 2000 2005 2010 2015 Source World Development Indicators 2016.. 18 Figure 2.3 Inflation and Exchange Rate (1995-2015) Inflation (%) IDR per US$ 60 12,000 50 10,000 40 8,000 Exchange rate (right axis) 30 6,000 20 Inflation (leftaxis) 4,000 10 0 2,000 1995 2000 2005 2010 2015 Source World Development Indicators 2016. section two. BACKGROUND A broadly conducive macroeconomic environment and then the gap between revenue and expenditure is expected over the next five years, with economic has begun to widen, with some periods of decline growth projected at a respectable 5-6 percent per year. resulting from economic slowdowns (Figure 2.4). Unlike A similarly positive outlook is expected for other key its peer countries by income, such as India and Sri macrofiscal variables (World Bank 2015e). While inflation Lanka, Indonesia’s debt and deficit levels are relatively has been slowly trending downwards (to 6 percent in low (in fact, lower than the European Union Maastricht 2015), the exchange rate has been slowly devaluing in Treaty benchmarks of 60 percent and 3 percent of GDP recent years (Figure 2.3). With continued and sustained respectively). The primary deficit has generally been economic growth, Indonesia is likely to transition to below 1 percent of GDP in recent years (Figure 2.5). upper-middle-income status within the next few years. This is, in part, because both central and subnational Indonesia’s overall debt and deficit levels appear to budgets must adhere to the fiscal rule,7 which sets be at manageable levels. Expenditures have generally a maximum annual deficit at 3 percent of GDP and tracked increasing revenues in real terms up until 2012 maximum accumulated debt at 60 percent of GDP. Table 2.1 Average Annual Economic Growth (1995-2015) GDP growth GDP per capita growth COUNTRY Mean SD Mean SD Brazil 2.6% 2.5% 1.4% 2.5% Cambodia 7.0% 2.4% 5.2% 2.6% China 8.6% 1.5% 8.0% 1.5% India 6.5% 1.8% 4.9% 1.9% 19 Indonesia 4.1% 4.6% 2.8% 4.6% Lao PDR 6.5% 1.1% 4.7% 1.1% Malaysia 4.7% 3.8% 2.7% 3.9% Philippines 4.5% 1.9% 2.6% 2.0% Russia 2.5% 5.0% 2.6% 5.1% South Africa 2.9% 1.6% 1.4% 1.7% Sri Lanka 5.3% 3.1% 4.4% 3.3% Thailand 3.2% 3.8% 2.5% 3.8% Vietnam 6.3% 1.1% 5.2% 1.1% East Asia and Pacific 3.9% 4.3% 2.7% 5.6% Lower-middle-income 4.2% 5.1% 2.8% 5.1% Source World Development Indicators 2016. Note SD: Standard deviation 7 Indonesia’s fiscal rule was enacted under Peraturan Pemerintah No. 23/2003. Pengendalian Jumlah Kumulatif Defisit Anggaran Pendapatan Dan Belanja Negara, Dan Anggaran Pendapatan Dan Belanja Daerah, Serta Jumlah Kumulatif Pemerintah Pusat Dan Pemerintah Daerah. HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better Figure 2.4 Real Revenues and Expenditures (1995-2015) IDR Trillions 2,000 Revenues Expenditures 1,500 1,000 500 0 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Source World Development Indicators 2016. Note Data are in 2015 constant local currency units 20 Figure 2.5 Fiscal Deficit and Debt Ratio (2012-15) Fiscal deficit to GDP ration (%) 20 10 Solomon Islands Philippines Rusia China 0 Indonesia Thailand Lao PDR Cambodia Nigeria South Africa Papua New Guinea Malaysia Brazil Sri Lanka Vietnam India -10 Ghana -20 0 25 60 100 150 Debt to GDP ratio (%) Source IMF World Economic Outlook database section two. BACKGROUND Although the industrial sector and service construction, electricity, water, and gas) rose from 6 21 sector are the two biggest contributors to GDP, percent to 19 percent (for a total share of industry in the service sector remains the biggest source GDP that went from 15 percent to 40 percent) (Table of employment in the country, followed by 2.2). The employment shares in 2014 across the three agriculture sector. The composition of Indonesia’s subsectors are different from those of GDP due to GDP has changed significantly in recent decades. differences in value addition: services is 45 percent, Agriculture’s share in GDP fell from more than 51 agriculture is 34 percent, and industry is 21 percent. percent in 1960 to just 14 percent in 2015. Over At 6 percent in 2014, the contribution of natural the same period, the services share of GDP rose resources to GDP was relatively low in Indonesia, from 33 percent to 43 percent, manufacturing in between the average for EAP (5 percent) and for rose from under 9 percent to 21 percent, and lower-middle-income countries (8 percent) (Word the nonmanufacturing industry share (mining, Development Indicator, 2016). Table 2.2 Economic Sector as Share of GDP and Employment GDP Share (%) Employment Share (%) ECONOMIC SECTOR 1960 2015 2014 Agriculture 51 14 34 Services 33 43 45 Manufacturing 9 21 21 Nonmanufacturing Industry 6 19 TOTAL 100 1008 100 Source World Development Indicators 2016. Note (i) Employment share for manufacturing includes nonmanufacturing industry. (ii) Total may not necessarily be 100% due to effects of rounding. 8 The authors make no representation regarding the completeness or accuracy of the data presented in this table. HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better The bottom 40 percent of Indonesia’s population Despite impressive gains in poverty reduction, the has not gained as much from recent economic level of informality in the labor market has remained growth, resulting in a rise in income inequality and a persistently high in Indonesia. Over 60 percent persistence in labor market informality. The bottom of those employed continue to be classified as 40 percent of the population saw an average growth nonsalaried workers (Figure 2-6). Given declining in real per capita consumption of only 1-2 percent poverty rates, this indicates a growing share of the per year over the period 2003-10; by way of contrast, nonpoor informal sector in the population. The the top 20 percent increased their consumption by labor force participation rate is almost 70 percent, 5-6 percent per year. This has resulted in a dramatic with unemployment at a relatively low 6 percent. rise in income inequality, one of the largest increases The average educational attainment in the adult in the EAP region (after China)(World Bank 2015b, population is 7.5 years (8 years for males and 7 2016b). The bottom 40 percent of the population years for females), about the average for what remains highly vulnerable to shocks–including might be expected for Indonesia’s economic status. those related to health–and tends to work in low- Educational attainment among the bottom 40 productivity, low-pay, nontradable sectors. While percent, however, remains relatively low and of poor 28 million Indonesias live below the poverty line, a quality and the enrollment rate among them drops further 68 million live less than 50 percent above it significantly after age 15 (World Bank 2013, 2015c). (World Bank 2016b). The extent of fiscal redistribution is limited with only a marginal difference between pre- and posttax Gini coefficients (Lustig 2015). 22 Figure 2.6 Informal (Nonsalaried) Workers as Share of Employed Population (1995-2013) Share of employed population (%) 80 60 40 20 0 1995 2000 2005 2010 2013 Source World Development Indicators database 2016. section two. BACKGROUND Macrofiscal Context National and subnational government budgetary level, and the remaining 13 percent at the provincial expenditures9 were a relatively low 17 percent of GDP level (World Bank 2012a). The legal framework for in 2015 (WEO 2016). In terms of size of government– central budgeting was introduced in the wake of the as measured by its share in GDP–Indonesia is a global 1997-98 Asian financial crisis and served to introduce outlier. The average government expenditure share transparency and accountability into what had been of GDP in 2015 was 32 percent among lower-middle- an opaque and unresponsive process, reflecting income countries. Indonesia’s government share of the heavy influence of the Dutch colonial budgeting GDP is low even in comparison with the average for system, in which the budget was conducted low-income countries (26 percent) and far lower than internally by the executive branch with little oversight that for lower-middle-income countries. As discussed or accountability to either the implementing below, this is in part a result of relatively low revenue- partners or taxpayers. Between 2000 and 2005, raising efforts in the country (Figure 2.7). the country made a concerted effort to change this and introduced a series of laws establishing a legal Indonesia has decentralized the provision of framework for the process, as well as a timeline and 23 government services to the district level, but a large mandatory milestones for the annual preparation, share of national government expenditure still occurs approval, and adoption of the budget. These laws at the central level. About 50 percent of all national also helped to establish the financial relationship government expenditures occurred at the central between various government agencies, and the roles level, followed by roughly 38 percent at the district and responsibilities of regional governments. Figure 2.7 Government Revenues and Expenditures as Share of GDP (2015) Percentage of GDP (%) 40 Revenues Expenditures 30 20 10 0 Sri Lanka Cambodia India Malaysia Lao Lower China South Brazil Rusia East PDR Middle Africa Asia & Indonesia Philippines Vietnam Thailand Income Pacific Source IMF World Economic Outlook database 9 The national government budget (APBN - Anggaran Pendapatan dan Belanja Negara) and subnational (APBD - Anggaran Pendapatan dan Belanja Daerah). HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better Table 2.3 All Government Expenditures by Category (2014) Share (%) Expenditure Category Central Provinces11 Districts12 Personnel 36 21 47 Goods and services 26 29 21 Capital 22 30 26 Social assistance 14 0 0 Other 2 19 6 Total 100 100 100 Source World Bank (2015) and Consolidated fiscal database (March 2016). Note Total may not necessarily be 100% due to effects of rounding. To facilitate medium-term planning, both national On the other hand, the discretionary expenditure and regional governments develop five-year plans to share of district expenditures was relatively low: coincide with presidential and district heads’ (bupati) personnel accounted for almost one-half of all district terms of office. These five-year plans tend to be expenditures, with goods and services accounting broader and more future-oriented than the annual for 21 percent and capital 26 percent. At the province budgets, but outline both the regulatory measures and level, capital and goods and services were the largest the budget needed to achieve the stated goals. The shares of expenditure.. 24 five-year plan is operationalized by annual work plans, which are linked to the annual budget. A Medium- Revenue-raising efforts are relatively modest in term Expenditure Framework (MTEF) supplements Indonesia. Indonesia’s national government revenue these work plans with three-year forecasts at the share of GDP was only about 14.8 percent of GDP in national level. The government plans to introduce 2015, far lower than the average for lower-middle- sector-specific MTEFs in the coming years. A 2011 income countries (28 percent) and less than one-half Public Expenditure and Financial Accountability (PEFA) the average for the EAP region (38 percent) (Figure assessment found public financial management 2.7). Somewhat surprisingly, Indonesia’s revenues to be strong in areas related to transparency and are even lower than the average for low-income comprehensiveness of budget documentation, having countries (22 percent): Cambodia, India, and Lao a well-defined and timely-executed budget process, PDR are all poorer than Indonesia, but have higher and a budget classification system that complied with government revenue shares of GDP (Fenochietto and international standards. It also found that weaknesses Pessino 2013).13 remained with regard to budget execution, financial reporting, and variations between allocations and Almost 90 percent of national government revenues expenditures (World Bank 2012b). was raised by the central government in 2013. The remaining 6 percent came from provincial own- In terms of expenditure categories, personnel source (Pendapatan Asli Daerah, PAD) revenue and expenditures accounted for 36 percent of all central only 4 percent was district own-source revenue. government expenditures, followed by 26 percent for Under Law No. 28/2009 concerning local taxes and goods and services spending, 22 percent for capital, charges, provincial governments are allowed to and 14 percent for social assistance (Table 2.3).10 collect the following taxes: motor-vehicle tax; excise 10 The shares are calculated by excluding expenditures on subsidies and interest payments. 11 Excludes intergovernmental transfers to districts, based on budget allocation data. 12 Based on 2015 budget allocation data. 13 Revenue generated through tax collection (“Tax effort”) is estimated to be only about 50 percent. 14 Government of Indonesia (GoI). Law 28/2009: Local Taxation and Charges in Chapter 2 – Article 2 - 93. section two. BACKGROUND Table 2.4 Central Government Revenues (2013-15) 2013 2014 2015 Revenues and Grants IDR trillion Share (%) IDR trillion Share (%) IDR trillion Share (%) Income tax (nonoil and gas) 418 29.0 453 29.2 630 35.7 VAT 385 26.8 409 26.4 577 32.7 Excise 109 7.6 118 7.6 146 8.3 Tobacco tax 104 7.2 113 7.3 121 6.9 Oil and gas tax (income and nonincome) 293 20.3 304 19.6 131 7.4 Other 227 15.8 262 16.9 276 15.7 Grants 7 0.5 5 0.3 3 0.2 Total 1,439 100 1,551 100 1,763 100 Source LKPP 2013-2014 and APBN-P 2015. tax for transfer of ownership of motor vehicles; motor 2.5). Provisional estimates for 2015 indicate that these vehicle fuel tax; surface water tax; and cigarette taxes. transfer amounts will remain about the same as a District governments, on the other hand, collect taxes share of GDP. Several modalities of intergovernmental on: hotels; restaurants; entertainment; advertising; fiscal transfers exist in Indonesia. Prominent among street lighting; nonmetal mineral and rock; parking; these are “fiscal balance” transfers comprising three and land and buildings; as well as acquisition rights primary components: general allocation funds (Dana on land and building, among others. Alokasi Umum, DAU), revenue sharing (Dana Bagi Hasil, DBH), and special allocation funds (Dana 25 Income and value-added taxes (VAT) were the largest Alokasi Khusus, DAK). sources of revenues for the central government. In 2014, nonoil and gas income tax revenues were 29 DAU represented the largest share (60 percent) of total percent of central government revenues followed resources transferred to subnational governments by VAT which was about 26 percent. Oil and gas in 2014.15 DAU is the unconditional equalizing grant revenues comprised 20 percent of central government from the center to provinces and districts in the form revenues; grants were less than 1 percent (Table 2.4). of a “basic allocation” (based on the total salary of Estimated numbers for 2015 from budget data shows subnational civil servants) and a “fiscal gap” (based an increase in the share of revenues from income on the difference between fiscal requirements and tax, which is about 36 percent, followed by VAT at 33 fiscal capacity). Fiscal requirements are determined percent. In 2015, the decline in the share of revenues based on population, land/sea area, a “construction from oil and gas taxes is notable. Excise tobacco taxes expensiveness index”, the human development index are 7 percent of central government revenues. (HDI), and gross regional domestic product while fiscal capacity is based on PAD and DBH revenues.16 Intergovernmental fiscal transfers of revenues Districts receive 90 percent of DAU, with the remaining are large, fragmented, and complex. Given the 10 percent going to provinces. Districts have discretion disconnect between largely centralized revenue over how DAU resources that are not tied to civil collection and decentralized expenditures across servant salaries are allocated. levels of government, approximately IDR 574 trillion, almost 6 percent of GDP, was transferred from the DBH grants totaled 18 percent of intergovernmental central to subnational governments in 2014 (Table fiscal transfers in 2014. These are unconditional 15 Provisional numbers for 2015 indicate DAU being 55 percent of all intergovernmental fiscal transfers. 16 Morespecifically, a district’s fiscal capacity is determined by the sum of revenues from PAD, DBH, and DAU minus personnel expenditures divided by the number of poor people in the district. HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better Table 2.5 DAK and DAK for Health as Share of all Intergovernmental Transfers (2011-15) DAK (IDR trillion) 2011 2012 2013 2014* 2015** All intergovernmental fiscal transfers 411 481 513 574 644 DAK total 25 26 31 32 59 Share of DAK in all intergovernmental fiscal transfers 6.1% 5.4% 6.0% 5.6% 9.2% DAK for health 3 3 3 3 6 Share of health in DAK 12.0% 11.5% 9.7% 9.4% 10.2% Source LKPP audited year 2011-2014 Note * Revised APBN 2015 ** PMK DAK Allocation 2011-2014. Table 2.6 Subnational Government Revenues (2013) IDR trillion Share (%) IDR trillion Share (%) Revenues and Grants Districts Provinces DAU 284 53.7 31 14.9 Other 88 16.6 40 19.3 DBH 68 12.9 32 15.4 Own-source (PAD) 58 11.0 102 49.4 DAK 30 5.7 2 1.0 26 Total 528 100 207 100 Source SIKD, DJPK-MoF. revenue-sharing transfers from the center to districts (although there is some discussion that this provinces and districts of taxes on income, property, requirement is to be eliminated). DAK for health can and natural resources with predefined shares being be used to procure infrastructure and equipment at returned to originating jurisdictions.17 About 2 percent public health facilities, including basic emergency of DBH grants represent tobacco revenue sharing. obstetric and neonatal care (BEONC) equipment, Subnational distributions are by provincial point of immunization equipment, laboratory equipment, origin; producing districts within provinces receive health promotion equipment, mobile health centers, larger proportions than nonproducing districts. and power sources (generators). Subnational governments have total discretion over the use of allocated funds. Provisional estimates for 2015 indicate that DAK’s share of all intergovernmental fiscal transfers will DAK resources that represented 6 percent of increase to more than 9 percent, up from around central government transfers in 2014 are key for 6 percent in 2013 and 2014 (Table 2.5). DAK for the government health sector. DAK allocations are health has almost doubled from 2014 to 2015 (and conditional, earmarked capital grants for prioritizing is expected to more than double again in 2016). some sectors (including health, which received 10 There are plans to convert DAK from a formula- percent of all DAK financing in 2014). DAK resources based to proposal-based allocation and to allow are designed to provide additional resources to for financing of noncapital expenditures. In general, districts that are underdeveloped, vulnerable, DAK is a substantial funding source for the districts and have low financial capacity. DAK allocations and its importance will increase in light of reforms. generally also require a 10 percent cofinancing from The possibility for financing infrastructure from 17 Provisional numbers for 2015 indicate a slight increase in DBH share of all intergovernmental fiscal transfers to 20 percent. section two. BACKGROUND DAK, for example, is quite significant, including for As sources of revenue, DAU is the largest for districts construction and upgrading of public health facilities. and own-source (PAD) is the largest for provinces. Over one-half of district financing comes from Other intergovernmental transfers include DAU allocations (Table 2-6). In aggregate across resources provided to special autonomous all districts, DAK’s share of district revenues is less regions and transfers to villages. For example, than 6 percent (although this is likely to be higher Law No. 06/2014 (or the “Village Law”), that was in districts with low fiscal capacity). In 2013, about ratified in early 2014 mandates an annual transfer 85 percent of all districts received DAK transfers of approximately US$140,000 from central and earmarked specifically for health. Unlike districts, subnational government budgets to every village where PAD revenues accounted for only 11 percent of in the country (amounting to about 1 percent of total district revenues, PAD revenues accounted for all intergovernmental fiscal transfers in 2015). almost one-half of all provincial revenues. The government is drafting the implementing regulations and ministerial decrees needed Many district governments view health as a revenue- to implement the Village Law. Village Law generating sector. Some have explicit targets on implementation provides a major opportunity for amounts of resources raised from health-user charges village governments to substantially increase that are then pooled at the district treasury level along investments in local development priorities. There with other revenue sources and allocated across is, however, a concern that Village Law financing sectors. A rapid assessment of 44 districts showed that, needs will crowd out already low levels of district in 2013, the biggest source of revenue was from DAU government expenditures for health worker outreach, (58 percent) and about 10 percent came from PAD preventative, or promotive care, which village revenue (consistent with aggregated numbers reported governments have no obligation to replace, or lack in Table 2.6). Over 40 percent of the latter came from the the capacity to procure and maintain. health sector (with one-half from public hospitals). 27 HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better 28 section three. HEALTH AND UHC OUTCOMES section 3 . 29 HEALTH AND UHC OUTCOMES HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better In Summary 1. Indonesia’s health status has improved significantly–life expectancy has increased steadily, infant and under-five child mortality rates have declined, and fertility and population growth rates have fallen. 2. The country is facing challenges due to: • Demographic transition: population aged 65 years and above is currently around 5 percent and is projected to double by 2030 and to reach 25 percent in 2070. • Epidemiological transition: the cause of disease has shifted to NCDs, and the emergence of overnutrition, while maternal mortality and stunting remain persistently high. • Large inequality: The national average masks regional and widespread income-related disparity. • Persistent health challenges: especially in maternal health and childhood nutrition. 3. The implementation of national Social Health Insurance (JKN) that aims to cover everyone by 2019 is one of the instruments for Indonesia to attain UHC. 4. Using the WHO-WB UHC monitoring framework to assess country progress towards UHC, Indonesia’s performance is mixed when it comes to preventive/promotive/treatment service coverage and financial coverage indicators. 5. OOP expenditures on health have pushed a significant percentage of the population into poverty or further into poverty. 6. Performance-based financing can serve as a tool to incentivize health systems and health 30 providers to move towards UHC. section three. HEALTH AND UHC OUTCOMES Demographics and Population Health Outcomes With a population of 257 million in 2015, Indonesia Indonesians have become healthier over the past is currently the fourth most-populous country in the several decades. Life expectancy at birth has world. Fertility rate and population growth rates have steadily increased to 69 years in 2014, up from 63 been steadily declining over the past few decades. years in 1990 and only 49 years in 1960 (Figure 3.3). The total fertility rate in 2014 was only 2.5 and the The under-five mortality rate has declined from 222 population growth rate in 2015 was 1.2 percent. UN per 1,000 live births in 1960 to 85 in 1990 and 27 in population projections estimate that Indonesia’s 2015, thereby meeting the MDG under-five mortality population will be almost 300 million in 2030, peaking rate of 28 per 1,000 by 2015 (UNICEF et al 2014). at 325 million by 2070, following which it is projected Nonetheless, Indonesia needs further improvement to decline (United Nations 2015). The age distribution to meet the SDG target in reducing under-five of the population is an important factor influencing mortality below 25 per 1,000 live births by 2030. Infant the utilization of health services: younger and older mortality has declined six-fold since 1960, down to subgroups tend to have much higher utilization rates 23 per 1,000 live births in 2015. Both life expectancy in general. Approximately 29 percent of Indonesians and infant mortality rates are about the average of 31 are below 15 years of age and the median age is what might be expected for Indonesia’s income around 28. While only 5 percent are 65 years of age level (Figure 3.4). Indonesia’s outcomes compare and above in 2015, this share is expected to increase unfavorably to those in better-performing countries sharply beginning in 2015, reaching 10 percent of the such as Vietnam and Sri Lanka. population by 2030 and 25 percent of the population by 2070 (Figure 3.2). Figure 3.1 Total Fertility Rate and Population Growth Rate (1960-2015) Average 6 3 Percentage number of (%) children 5 2.5 Population growth rate (right axis) 4 2 Total fertility rate (left axis) 3 1.5 2 1 1960 1970 1980 1990 2000 2010 2015 Source World Development Indicators 2016. HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better Figure 3.2 Share of Population Aged 65 and Above (1950-2070) Share of 30 total China population Thailand (%) Indonesia Vietnam Philippines 20 10 0 1950 1970 1990 2010 2030 2050 2070 Source UN population projection 2015 32 Figure 3.3 Key Population Health Outcomes (1960-2015) Years Mortality 250 75 rate per 1,000 Under-five mortality (left axis) Life expectancy (right axis) 70 live 150 births 65 100 Infant mortality (left axis) 60 50 55 50 25 45 1960 1970 1980 1990 2000 2010 2015 Source World Development Indicators 2016. Note y axis in log scale. section three. HEALTH AND UHC OUTCOMES Figure 3.4 Life Expectancy and Infant Mortality Relative to Income (2014) years Life Expectancy rate per 1,000 live births Infant Mortality 85 125 75 Nigeria China Vietnam Sri Lanka Lao PDR 75 Malaysia 50 Ghana Papua New Guinea Thailand Brazil India South Africa Indonesia Rusia Cambodia Indonesia Cambodia 25 Philippines Solomon Islands Philippines India Lao PDR Solomon Islands 65 Vietnam Papua New Guinea Brazil Ghana 10 Thailand Sri Lanka China Rusia South Africa Malaysia 5 55 Nigeria 2 LOWER UPPER LOWER UPPER MIDDLE MIDDLE MIDDLE MIDDLE LOW INCOME INCOME INCOME HIGH INCOME LOW INCOME INCOME INCOME HIGH INCOME 45 1 250 500 1000 2500 10000 35000 100000 250 500 1000 2500 10000 35000 100000 GNI per capita, US$ GNI per capita, US$ 33 Source World Development Indicators 2016.. Note Both x and y axis in log scale. Figure 3.5 Burden of Disease by Cause (1990-2015) 7% 9% 9% 8% 1990 2000 2010 2015 27% 37% 49% 58% 66% 56% 33% 43% Injuries Noncommunicable Communicable Source Institute of Health Metrics and Evaluation database (IHME) 2015 HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better Table 3.1 Top Ten Causes of Morbidity and Premature Mortality (1990-2015) DALYs lost share (%) Rank in 2015 Disease/Condition 1990 2000 2010 2015 1 Cerebrovascular disease 4.2 6.4 7.6 8.6 2 Ischemic heart disease 3.6 5.4 6.8 7.4 3 Diabetes mellitus 2.0 3.2 4.5 5.3 4 Tuberculosis 5.7 5.5 4.9 4.3 5 Road injuries 3.4 3.7 3.5 3.4 6 Lower back and neck pain 1.8 2.5 3.0 3.3 7 Neonatal preterm birth 5.6 4.6 3.5 3.0 8 Sense organ diseases 1.5 2.1 2.5 2.8 9 Diarrheal diseases 8.0 5.1 3.4 2.7 10 Lower respiratory infections 8.9 5.6 3.7 2.6 DALYs per 100,000 population 45,138 34,725 30,681 29,217 Source Institute of Health Metrics and Evaluation database (IHME) 2015. 34 Indonesia is undergoing a rapid epidemiological prevalence of NCDs among younger age groups transition. Noncommunicable diseases (NCDs) in Indonesia is also increasing. Physical inactivity, now account for the largest share of the burden of unhealthy diets, tobacco use, and child and disease in Indonesia. Whereas in 1990 only about 37 maternal malnutrition are key risk factors for NCDs. percent of morbidity and mortality in Indonesia was Several of these risk factors–including dietary risks, due to NCDs, by 2015 this number had risen to 66 hypertension, smoking, high fasting plasma glucose percent (Figure 3.5)(Institute of Health Metrics and level, and physical inactivity–are prominent among Evaluation 2016). This trend is expected to continue the top ten risk factors contributing to the overall in the coming years. Cerebrovascular diseases disease burden in the country (Table 3.2). The share of were responsible for the largest share of the overall dietary risks and high blood pressure as contributors disease burden in Indonesia, causing 8.6 percent to DALYs lost has more than doubled over the period of all disability-adjusted life years (DALYs) lost due of 1990-2015. Tobacco use is rising and constitutes to morbidity and premature mortality in 2015 (Table one of the most significant public health threats. The 3.1).18 Other NCDs such as ischemic heart disease and government has embarked on the tobacco taxation diabetes have more than doubled as a share of the reform with the motivation to increase revenue from disease burden in Indonesia over the period 1990- excise tax and, at the same time, curbing smoking 2015. Tuberculosis remains a prominent contributor prevalence, although the government is facing to the overall burden of disease in the country; challenges in its implementation (See Appendix D). however, its share of the overall burden dropped between 2010-15 making it the fourth highest source Large regional and income-related inequalities of morbidity and mortality. remain across the country. The infant mortality rate   in West Sulawesi, for example, is two to three times The rise in NCDs in Indonesia is a result of changes higher than that in some other provinces (Figure 3.6). in several sociodemographic and lifestyle factors. Moreover, infant and child mortality rates among the Ageing is one contributory factor, although the poorest wealth quintile of households are more than 18 DALYs refer to aggregated healthy years of time lost at the population level as a result of disease-related morbidity and premature mortality. section three. HEALTH AND UHC OUTCOMES Table 3.2 Top Ten Risk Factors (1990-2015) DALYs lost share (%) Rank in 2013 Risk Factors 1990 2000 2010 2013 1 Dietary risks 6.8 10.5 13.4 15.1 2 High systolic blood pressure 6.0 9.2 11.7 12.9 3 High fasting plasma glucose 3.9 6.1 8.7 10.0 4 Tobacco smoke 5.6 6.6 8.1 8.7 5 High body-mass index 1.5 2.9 5.5 6.9 6 Child and maternal malnutrition 20.7 11.4 7.0 5.2 7 Air pollution 6.8 5.9 5.5 4.9 8 High total cholesterol 1.8 2.8 3.9 4.3 9 Unsafe water, sanitation, and handwashing 10.4 6.7 4.6 3.6 10 Low glomerular filtration rate 1.4 2.0 2.5 2.6 Source Institute of Health Metrics and Evaluation database (IHME) 2015. double those in the richest. The variation of health far from reaching the SDG target of MMR less than 35 outcomes and outputs, such as life expectancy, 70 per 100.000 live births by 2030. Unlike some of antenatal care, institutional delivery rates, and the other key health outcomes, Indonesia’s MMR stunting across districts is even more pronounced. is one of the highest in the region, much worse For example, some districts have no births in health than what might be expected given its income facilities whereas others have a 100 percent rate and comparable to estimates from lower-income (Figure 3.7). In general, there is an economic gradient countries such as India. Furthermore, Indonesian to health outcomes. For example, the average value children suffer from high rates of malnutrition with a of MoH’s combined public health index is higher for prevalence of stunting at 37 percent and of wasting districts that are in richer economic deciles (the latter at 12 percent. There is also wide variation in the measured by average consumption per capita of prevalence of malnutrition across provinces within households in the district) (Figure 3.8).19 Indonesia (Figure 3.9). With over 8 million children affected, Indonesia has the fifth-highest number of Despite notable progress on some key health stunted children in the world (Millennium Challenge outcomes, several challenges remain, especially Account-Indonesia 2015). Stunting in the first with regard to maternal health and nutrition. At 126 two years of life can lead to irreversible damage, per 100,000 live births, the maternal mortality ratio including shorter adult height, lower schooling (MMR) remains high and Indonesia has not met attainment, reduced adult income, and increased the maternal health MDG target of 102.20 It is still incidence of morbidity in later life. 19 The public health index (IPKM) for 2013 is generated by MoH using 24 indicators covering community health indicators (for example, handwashing, access to sanitation, access to water), individual health indicators (percentage of population with diarrhea, pneumonia, hypertension, diabetes), health inputs (ratio of midwives to villages, ratio of doctors to puskesmas), maternal health services (skilled birth attendance), and nutrition status (underweight, stunting, and overweight) 20 Trends in maternal mortality: 1990 to 2015: estimates by WHO, UNICEF, UNFPA, World Bank Group and the United Nations Population Division (WHO 2015). There is some uncertainty about the exact level of MMR in Indonesia; the Institute of Health Metrics and Evaluation (IHME) model estimated an MMR of 189 in 2011; IDHS 2012 estimates based on sibling-survival data indicate an MMR of 359, although it is important to note that this latter estimate is derived from a sample occurrence of only 92 maternal deaths over a five-year period; the 2010 Indonesian census indicates an MMR of 278. HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better Figure 3.6 Infant Mortality (by province) Rate 80 per 1,000 live births 60 40 Indonesia average 20 0 Riau Islands DI Yogyakarta Bangka Belitung Riau East Java West Java Central Java Maluku Central Kalimantan Papua East Kalimantan DKI Jakarta South Sulawesi West Sumatra Bali Bengkulu South Sumatra Lampung West Kalimantan Banten North Sulawesi Jambi North Sumatra South Kalimantan Southeast Sulawesi Nanggroe Aceh West Nusa Tenggara Central Sulawesi West Sulawesi North Maluku Gorontalo West Papua East Nusa Tenggara Source IDHS 2012 36 Figure 3.7 Distribution of Key Health Indicators Across Districts (2013) Percentage of districts (%) Percentage of districts (%) 60 65 70 75 0 20 40 60 80 100 Life expectancy at birth (year) 4+ ANC visits (%) Percentage of districts (%) Percentage of districts (%) 0 20 40 60 80 100 0 20 40 60 80 Institutional delivery rate (%) Stunting rate (%) Source Indeks Pembangunan Kesehatan Masyarakat (IPKM), MoH. 2013; Indeks Pembangunan Manusia (IPM), Indonesia Statistic-2013 Figure 3.9 Figure 3.8 (%) Percentage 10 30 40 50 0 20 (higher is better) Riau Islands Source Riskesdas 2013 DI Yogyakarta 0 .3 .4 .5 Health index . 6 .1 .2 DKI Jakarta East Kalimantan Bangka Belitung Source World Bank staff calculation Poorest Bali Banten North Sulawesi Indonesia average West Java East Java Central Java 3rd Riau South Sumatra Jambi West Kalimantan Health Index by District Economic Deciles (2013) Gorontalo Stunting Among Under-five Children (by province West Sumatra 5th Bengkulu Papua Maluku Economic deciles South Sulawesi North Maluku Central Sulawesi 7th Central Kalimantan Nanggroe Aceh North Sumatra Lampung Southeast Sulawesi South Kalimantan West Papua West Nusa Tenggara West Sulawesi East Nusa Tenggara Richest section three. HEALTH AND UHC OUTCOMES 37 HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better Universal Health Coverage Indonesia plans to attain UHC with everyone single-payer umbrella beginning in 2014. Following covered under its newly unified SHI program, institutionalization of the single-payer insurance JKN, by 2019. SHI has undergone major reforms administrator (BPJS) and the new unified health in Indonesia in recent years. The universal right insurance program (JKN) in 2014, the government to health care was included as an amendment to plans to incrementally extend coverage to the entire Indonesia’s constitution in 1999. The impetus for population by 2019. expansion of SHI came a few years later, however, in a piece of landmark legislation in 2004–the The WHO-WB’s monitoring framework recommends Sistem Jaminan Sosial Nasional (SJSN) law–that tracking a mix of preventive/ promotive/ treatment formed the legal basis for attaining several social service coverage and financial coverage indicators protection objectives in the country. In 2011, the to assess country progress towards UHC. government of Indonesia passed a ground-breaking Recommendations under preventive/promotive follow-up law (Law No. 24/2011) that defined the coverage include family planning coverage 38 administrative and implementation arrangements– with modern methods, antenatal care, skilled the Badan Penyelenggara Jaminan Sosial (BPJS) birth attendance, DPT3 immunization coverage, law–which stipulated that all existing contributory nonprevalence of tobacco smoking, access to and noncontributory SHI schemes be merged improved water sources, access to improved to provide streamlined uniform benefits under a sanitation, and preventive chemotherapy coverage against neglected tropical diseases. Recommended treatment interventions include antiretroviral therapy (ARV) coverage, tuberculosis, hypertension, diabetes, and cataract surgical coverage (WHO and World Bank 2015). Recommended financial coverage indicators include those derived from levels of OOP health expenditures as a share of total expenditure, as a share of capacity to pay, and as a share of nonfood expenditure. In addition, the UHC framework recommended that financial coverage also be assessed by looking at the share of the population not pushed into poverty (that is, with expenditures net and gross of OOP above an international poverty line/level of subsistence food consumption/multiple poverty lines), share of the population not further pushed into poverty (that is, with expenses below an international poverty line/ level of subsistence food consumption/multiple poverty lines), and no OOP, as well as share of the population that are neither pushed nor further pushed into poverty. section three. HEALTH AND UHC OUTCOMES Table 3.3 UHC Indicators: Preventive, Promotive and Treatment (%) (2010-15)21 Preventive/promotive Treatment Country Family Skilled birth Tobacco ANC DPT3 Water Sanitation ARV TB planning attendance nonuse Brazil 80 96 99 93 83 98 81 46 59 Cambodia 51 89 71 97 76 71 37 71 59 China 85 95 100 99 75 92 65 52 85 India 55 75 67 83 87 93 36 36 50 Indonesia 62 96 83 78 62 85 59 8 28 Lao PDR 50 53 40 88 65 72 65 30 28 Malaysia 49 97 99 97 77 100 96 21 62 Philippines 49 95 73 79 73 92 74 24 73 Russia 68 100 100 97 59 97 70 29 56 South Africa 60 97 94 70 80 95 74 45 53 Sri Lanka 68 99 99 99 85 94 92 19 59 Thailand 79 98 100 99 78 96 93 61 45 Vietnam 78 96 94 95 76 95 75 37 68 East Asia and Pacific 48 90 83 86 71 87 67 38 60 Lower-middle income 46 86 74 86 78 83 59 29 56 Source World Development Indicators database 2016. Note Attainment less than 80 percent is highlighted in blue. 39 On the basis of the WHO-WB UHC monitoring to move towards UHC. In an attempt to better framework, and based on available data on coverage, incentivize providers to attain UHC outcomes such as Indonesia’s performance is mixed. In terms of immunization, many countries have amended their preventive/promotive indicators, deficiencies provider payment mechanisms to make the UHC- are notable in: access to modern family planning immunization links more explicit. Some examples methods, DPT3 immunization coverage, tobacco from Argentina, Estonia, New Zealand, and the UK nonuse, and access to improved sanitation. Tobacco that are summarized in Box 3.1 below can provide nonuse is particularly low, almost as low as in relevant experiences and lessons for Indonesia. Russia. Whereas coverage of preventive/promotive interventions is higher in Indonesia than in Cambodia, With regard to financial protection, even though Lao PDR, and India, it is far below that of some of the prepaid/pooled share of THE is relatively low the BRICS (Brazil, Russia, China, and South Africa) in Indonesia, the incidence of OOP expenditures countries and is notably lower than that of Vietnam being greater than 25 percent of total household (Table 3.3). For treatment indicators, although TB expenditure is only 1 percent. Nevertheless, because detection and treatment rates are relatively high, of the bunching of population just above the poverty ARV treatment and diabetes treatment rates are line and the incidence of OOP expenditures among exceedingly low (although comparable to those in those below the poverty line, 18 percent of the other lower-middle-income and EAP countries). population was either pushed into poverty or further impoverished as a result of high OOP spending on Performance-based financing can serve as a tool health (Table 3.4). There is more discussion on this to incentivize health systems and health providers later under the section on OOP spending. 21 Although the WHO-WB recommended UHC indicator for ANC refers to at least four visits during pregnancy, because of limited availability, the data reported in the table are for at least one ANC visit during pregnancy. The TB tracer indicator is a multiplication of two indicators; the treatment success rate and case detection rate in a given year. HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better 40 Table 3.4 UHC Indicators: Financial Protection22 Prepaid/pooled share OOP<25% Total household Neither pushed nor further Country of THE (%) Consumption (%) pushed into poverty (%) Brazil 70 97 97 Cambodia 40 97 83 China 66 87 90 India 42 99 72 Indonesia 54 99 82 Lao PDR 60 100 93 Malaysia 64 100 99 Philippines 43 100 78 Russia 52 100 100 South Africa 93 100 93 Sri Lanka 53 100 99 Thailand 89 100 100 Vietnam 51 95 75 East Asia and Pacific 76 98 87 Lower-middle-income 60 97 84 Source World Development Indicators database 2016. Note Attainment less than 80 percent is highlighted in blue. 22 Theprepaid/pooled share of total health expenditure is not a WHO-WB recommended financial protection indicator; nevertheless, this is included in the table because it is generally highly correlated with the incidence of catastrophic health spending. section three. HEALTH AND UHC OUTCOMES Box 3.1 Improving Performance for UHC through Provider Payments: Some Global Experience ESTONIA Through the Quality Bonus Scheme (QBS)–a joint initiative between the Estonian Health Insurance Fund and the Estonian Family Physician Association–primary care providers receive “points” for achieving coverage targets across weighted domains of: (i) disease prevention (including child immunization, child preventive care, and cardiovascular prevention); (ii) chronic disease management (for example, diabetes and hypertension); and (iii) specific additional activities (for example, primary care provider training, maternity care). Achievement of at least 80 percent of the points allows the providers to receive a pro-rataed lump sum (negotiated annually) in addition to other payment sources (that is, capitation, travel allowance, and fee-for-service payments for diagnostic procedures). Primary-care physicians determine how the bonus is distributed to nurses and other staff. ARGENTINA Argentina’s Plan Nacer was initiated in 2004 to provide coverage for the poor in provinces located in the northern part of the country. The program is designed to provide results-based financing to provincial governments based on the number of enrollees in the program, as well as performance on a set of basic health indicators. About 60 percent of intergovernmental fiscal transfers from the central government to the provincial governments are based on the number of enrollees and the remaining 40 percent is tied to attainment of ten tracer indicators, such as immunization rates and average 41 weight at birth of newborns. Service delivery is contracted out by the provincial governments to certified public and private providers, with patients free to choose among the providers. The program finances a conditional matching grant from the central government to provinces that pays one-half the average per capita cost of a basic benefit package covering 80 cost-effective maternal and child- health interventions to uninsured mothers and children up to six years of age. The program has built-in incentives for increasing enrollment rates as well as for provision of quality care. Capitation-based and unit-costed payments encourage negotiation with providers and efficiency in delivery of services. NEW ZEALAND The Primary Health Organisation (PHO) Performance Programme–which is transitioning into the Integrated Performance and Incentive Framework (IPIF)–incentivizes eligible PHOs to achieve population health and inequality priorities measured via clinical indicators (for example, childhood vaccination; influenza vaccination in the elderly; cervical and breast cancer screening; cardiovascular risk assessment); process/capacity indicators (for example, progress against performance plan); and financial indicators (for example, pharmaceutical and laboratory expenditure against benchmarks) by paying a flat-rate bonus every six months on the basis of percentage attainment of targets. UNITED KINGDOM Providers receive quarterly payments when they can prove that at least 70 percent of cohorts of children aged two years registered under the providers have completed immunization for certain types of vaccines. Achievement of a 90 percent target enables providers to receive three times the amount providers would have been eligible for if they had achieved a 70 percent target. Providers also receive an additional payment per child, if the registered child has completed all rotavirus, pneumococcal and meningitis C/HiB booster doses. Recent large intergovernmental transfer programs such as Village Fund also provide an opportunity for performance-based mechanisms to improve immunization coverage at the local level. HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better 42 section four. HEALTH SYSTEM section 4 . 43 HEALTH SYSTEM HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better In Summary 1. Outpatient and inpatient utilization have increased, especially among the bottom 40 percent and at inpatient private facilities, although geographic variation remains high. 2. Utilization patterns at facilities suggest that puskesmas were generally pro-poor while public hospitals were pro-rich. 3. The number of hospitals has doubled over the past decade, more than one-half were private hospitals. 4. The bed-density ratio has increased although it is still below the WHO standard of 2.5 per 1,000, and there is maldistribution of beds across the country. 5. There has been little improvement in the readiness to provide key health services since the 2011 health facility census (Rifaskes). 6. Many Indonesians face significant physical and time barriers to accessing health care. 44 section four. HEALTH SYSTEM Health Care Organization and Delivery Indonesia has mixed public-private provision of utilizing outpatient services in the last 30 days and health services and dual practice is legal. The almost 4 percent reported utilizing inpatient services public sector generally has a dominant role in rural in the past 12 months (SUSENAS 2015). These areas and for secondary levels of care, but this is numbers have increased in recent years following a not necessarily the case across all health services. period of decline during and after the 1997-98 Asian Private provision has been increasing rapidly in recent financial crisis (Figure 4.1). IDHS data indicates that years, including for primary care. The country has the number of caesarean sections–another indicator 34 provinces, 514 districts/cities, and some 72,000 capturing improved access to high-end maternal villages, with public provision decentralized to the health services–has tripled: from 4 per 100 deliveries district/city level. As a country with over 6,000 in 1997-2002 to 12 per 100 deliveries in 2012. inhabited islands, geography poses a significant obstacle to service delivery. The data indicates an increasing trend of outpatient utilization in all type of facilities, although almost Outpatient and inpatient utilization rates have risen one-half of all outpatient utilization occurred 45 steadily, especially among the bottom 40 percent at private facilities in 2015. On the other hand, of the population and at private facilities. In 2015, there is an increasing trend, albeit small, for approximately 17 percent of the population reported inpatient services in the private sector (Table 4.1).23 Figure 4.1 Inpatient and Outpatient Utilization Rates (1995-2015) Percentage 20 (%) 15 Outpatient 10 5 Inpatient 0 1995 1998 2004 2004 2007 2010 2015 Source SUSENAS (various years). 23 Inpatient utilization rates refer to the proportion of the population that utilized inpatient care in the past 12 months (SUSENAS). HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better Nevertheless, the annual inpatient admission rate the major actors. Other ministries and public remains one of the lowest in the region. There are entities involved in the health sector include the wide variations in utilization rates across the country Ministry of Home Affairs, the Ministry of National with provinces in the Java-Bali region generally Development Planning/the National Development having much higher utilization rates compared with Planning Agency (Bappenas), BPJS, the National other provinces; Maluku, Papua, and North Maluku Food and Drug Control Agency (BPOM), the National have some of the lowest utilization rates in the Population and Family Planning Board (BKKBN), and country (Figure 4.2). the Ministry of Villages, Disadvantaged Regions, and Transmigration (Kemendesa). Provincial Health Offices Figure 4.3 summarizes the organization of Indonesia’s (PHOs) run provincial hospitals and coordinate cross- health system and reflects the relationships among district issues. All other public facilities are managed Table 4.1 Inpatient and Outpatient Utilization Rates (by economic status and at public/private facilities) (2012-15) FUNCTION 2012 2013 2014 2015 National 12.9% 13.5% 15.4% 17.0% Outpatient utilization (all) Bottom 40% 11.7% 12.2% 13.9% 16.0% National 8.1% 8.7% 10.4% 8.7% Outpatient utilization (private) Bottom 40% 6.4% 7.1% 8.5% 7.6% National 1.9% 2.3% 2.5% 3.6% Inpatient utilization (all) Bottom 40% 1.3% 1.6% 1.8% 2.6% 46 National 0.8% 1.0% 1.1% 1.7% Inpatient utilization (private) Bottom 40% 0.4% 0.5% 0.6% 0.9% Source SUSENAS (2012-2015). Figure 4.2 Utilization Rates (by province) (2015) Percentage 25 (%) outpatient inpatient 20 15 10 5 0 Riau Islands Bangka Belitung DI Yogyakarta Maluku Papua Central Kalimantan Riau West Java East Java Central Java North Maluku West Papua Southeast Sulawesi Jambi West Kalimantan North Sumatra East Kalimantan West Sulawesi Central Sulawesi South Sulawesi Bengkulu South Sumatra Lampung Banten North Sulawesi South Kalimantan Gorontalo West Sumatra West Nusa Tenggara Nanggroe Aceh DKI Jakarta Bali East Nusa Tenggara Source SUSENAS 2015 section four. HEALTH SYSTEM by District Health Offices (DHOs), under the overall purview of district governments. The central MoH operates some tertiary and specialist hospitals but, otherwise, plays more of a stewardship role in terms of regulation and supervision of the health system. Figure 4.3 Organization of Indonesia’s Health System Central Government & BPJS Parliament Ministry of Ministry of Health Central Hospital Home Affairs 47 Provincial Provincial Health Office Government & Regulation Parliament Provincial Hierarchical hospitals District Public primary District Health Office care facilities Government & Parliament District Private clinics hospitals and practice Source Asia Pacific Observatory and WHO 2015 HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better Physical Resources Puskesmas are the backbone of Indonesia’s public (IFLS 2014) data show that the turnover of kader has health system, each serving a catchment area of been high, as demonstrated by the fact that more 25,000-30,000 individuals, and providing primary than 40 percent have less than five years working at a care services. There were 9,731 puskesmas in posyandu. Moreover, only 6 percent reported having 2014, with almost one-third having inpatient beds no problems with financial support, human resources, (MoH 2015). As mentioned above, private clinics and supplies. increasingly provide primary care but there is no systematic information available at the central Indonesia has a mix of public and private hospitals level on their numbers and distribution. The public for secondary and tertiary care.25 Indonesia’s primary care system also includes 23,000 auxiliary MoH Regulation No. 340/2010 classifies general puskesmas (pustu) for outreach activities in remote hospitals, both public and private, as types A, B, C, regions, village-level delivery posts (polindes - often and D according to services provided (this excludes the home of the village midwife), and village health puskesmas with beds).26 In 2014, MoH recorded 60 48 posts (poskesdes). In addition, community-level Type A, 308 Type B, 803 Type C, and 537 Type D participation is active in maternal and child-health hospitals, and 700 hospitals were not classified. promotion activities at around 289,635 integrated There are at least 17 types of specialty hospitals, health services posts (posyandu).24 of which the largest numbers were mother and child hospitals, followed by maternity and mental While puskesmas are the backbone of the Indonesian hospitals. The number of hospitals has almost public health system, the kaders who implement doubled over the past decade to an estimated 2,228 the posyandu and other community-based health in 2014, with more than one-half of all hospitals now activities, are the outreach linchpin from the health being private (MoH 2015). sector to the community. While international experience shows the significant value and The number of beds per capita in Indonesia stands contribution of these community health workers to at 1.07 per 1,000 population and,27 despite a rise in disease prevention and health promotion, as well as the bed-density ratio in recent years, this number follow-up and compliance to TB control for example, remains far below WHO’s norm/recommended ratio kader and posyandu are not under the control of of 2.5 per 1,000 (MoH 2015). Indonesia’s numbers the Ministry of Health (MoH) and their capacity to are still much lower than comparator countries in provide services is very limited. World Bank staff the region, including Thailand, Malaysia, Sri Lanka, calculations using the Indonesia Family Life Survey China, and Vietnam. Key issues are the lack of 24 Puskesmas (Pusat Kesehatan Masyarakat): Community Health Center. Pustu (Puskesmas Pembantu): Subhealth Center. Polindes(Pondok Bersalin Desa): Village Maternity Clinic. Poskesdes (Pos Kesehatan Desa): Village Health Post. Posyandu (Pos Pelayanan Terpadu): Integrated Health Services Post 25 Additional details on this are provided later in the document. 26 Type A provides, at a minimum, four basic specialist services (internal medicine, pediatrics, surgery, obstetrics-gynecology), five medical support specialist services (four medical diagnostics and anesthesia), twelve other specialist services, and thirteen subspecialist services; Type B provides, at a minimum, four basic specialist services, four medical support specialist services, eight other specialist services, and two subspecialist services; Type C provides, at a minimum, four basic specialist services, and four medical support specialist services; Type D provides, at a minimum, two basic specialist services. 27 This number does not include beds in private clinics. section four. HEALTH SYSTEM systematic information on the number of hospital Pharmaceutical production is dominated by domestic beds in private clinics and the maldistribution firms and price-regulated unbranded generics are of beds across the country. There is a four-fold widely used by the government as a means of cost difference in the bed-density ratio across the containment. Pharmaceutical expenditures are 33-44 country: from a high of 2.8 beds per 1,000 in DI percent of THE and domestic firms hold an estimated Yogyakarta to a low of 0.71 per 1,000 in Lampung. 75 percent of the pharmaceutical market share, Thirteen provinces had a bed-density ratio below with the remainder being multinational firms (BMI the Indonesian average (MoH 2015). Research 2015). The National Medicines Policy, last updated in 2006, provides guidance on key issues Many Indonesians face significant physical and time and priority health problems such as medicines barriers to accessing health care. This is particularly financing, availability, affordability, selection of true in the eastern provinces, resulting in higher essential medicines, and rational use of medicines. morbidity and mortality rates and inefficient use of potentially productive time by patients as well as BPOM provides regulatory and policy oversight for accompanying family members and friends (Schoeps medicines, traditional medicines, cosmetics, and et al 2011).28 Although the median distance to a supplements. The agency reports directly to the health facility in Indonesia is only five kilometers, the president and works closely with MoH. In addition median distance in provinces such as West Papua, to overseeing the registration of pharmaceutical Papua, and Maluku was over 30 kilometers. Widely products in the country, BPOM is also responsible for divergent geographic accessibility is correlated pre-marketing and post-marketing assessment of the with the time ranges that Indonesians experience to quality of all drugs. Since 1978, Indonesia has had a reach public health facilities. On average, more than national list of essential medicines (DOEN)30 which is 18 percent of Indonesians took more than one hour updated every three to five years. Brands and prices to reach a public hospital (using any travel means), of drugs are based on supplier bids and are listed in 49 more than 40 percent of people in West Sulawesi, an e-catalog that is used for ordering and procuring Maluku, and West Kalimantan faced this barrier to drugs with the help of LKPP.31 PHOs, DHOs, and public access (National Institute for Health Research and hospitals use the e-catalog for procuring drugs. More Development 2013). Measured in time, puskesmas than 90 percent of the drugs included in the e-catalog were more accessible, as only 2 percent of the are generic (branded and unbranded). The use of national population took more than one hour to reach e-catalog means that pharmaceutical prices are a puskesmas, but the proportion of the population bargained nationally through an open tender. facing this travel time was much higher in Papua (28 percent), East Nusa Tenggara (11 percent), and West In a recent assessment based on the analysis of facility Kalimantan (11 percent) (United Nations 2003).29 data, the general service readiness of health facilities to provide basic health services at minimum standards Utilization patterns at facilities suggest that was found to be highly variable across provinces puskesmas are generally pro-poor whereas public (MoH-World Bank 2014a and MoH-World Bank 2014b). hospitals are pro-rich. Of all persons who sought There are notable weaknesses in some of the eastern care at puskesmas, either for outpatient or inpatient provinces such as Papua, Maluku, West Papua, West care, a higher percentage were from poorer income Sulawesi, and North Maluku. The readiness to provide deciles as opposed to the richer deciles (Table 4.2). basic services was measured by a set of 38 indicators With regard to utilization at public hospitals, however, that were collected as part of the 2011 health facility richer deciles had generally higher utilization patterns census (Rifaskes) across five domains: basic amenities, as compared to poorer deciles. Utilization patterns at basic equipment, standard precautions for infection private facilities were generally pro-rich. prevention, diagnostic capacity, and essential 28 See also: Abhimanyu et al. (2011) and Mulholland et al. (2008). 29 It is noted that the time to walk to a private health facility or drug outlet to access affordable essential drugs on a sustainable basis is a key indicator used for MDG tracking, with one hour identified as the benchmark. 30 DOEN: Daftar Obat Esensial Nasional: National List of Essential Medicines. 31 LKPP: Lembaga Kebijakan Pengadaan Barang/Jasa Pemerintah: Government Goods and Services Procurement Policy Institute. HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better medicines (WHO 2013b).32 Across Indonesia, not even obstetric care, routine childhood immunization, one puskesmas reported meeting all 38 indicators malaria, tuberculosis, diabetes, basic surgery, available for general service readiness (World Bank blood transfusion, and comprehensive surgery. In 2014).33 While puskesmas, on average, met more particular, Table 4.3 highlights a snapshot of the than 80 percent of the 38 indicators available in DI deficiencies and variation in provision of key services Yogyakarta, East Java, and Central Java, only about provided in an analysis of Rifaskes facility data and one-half reported the same level of achievement in other sources. These deficiencies are reflective of Papua and Maluku. significant variations in the availability of the JKN benefits package, especially in the eastern parts of Numerous challenges remain with regard to service- the country (MoH-World Bank 2014a and MoH-World specific readiness, and the capacity of health facilities Bank 2014b). More recent data (IFLS 2014) reported to provide interventions in key program areas. This that there has been little improvement in the includes family planning, antenatal care, basic readiness to provide key health services (Table 4.3). Table 4.2 Participation Incidence for Utilization at Public and Private Facilities (2015) Share OUTPATIENT utilization by income decile (%)(all) Type of facility Total Poorest 2nd 3rd 4th 5th 6th 7th 8th 9th Richest Public facility 13.2 12.2 13.0 11.4 10.4 9.9 9.4 8.1 7.2 5.2 100.0 Public hospital 6.9 7.8 8.8 8.2 9.1 9.3 10.3 11.4 13.0 15.2 100.0 Puskesmas 14.5 13.1 13.8 12.1 10.7 10.0 9.3 7.5 5.9 3.1 100.0 50 Private facility 8.6 9.7 9.9 10.4 10.3 10.3 10.3 10.4 10.2 9.9 100.0 Private hospital 3.0 4.2 4.8 5.9 6.4 8.8 10.2 13.1 17.2 26.4 100.0 Private clinic 9.1 10.3 10.4 11.0 10.7 10.5 10.3 10.1 9.6 8.0 100.0 Share INPATIENT utilization by income decile (%) (all) Public facility 9.2 9.8 10.0 9.6 10.1 10.3 10.7 10.2 10.4 9.7 100.0 Public hospital 8.0 8.3 8.7 9.1 9.9 10.0 11.3 11.2 11.8 11.7 100.0 Puskesmas 13.3 14.3 13.8 11.3 10.7 11.1 8.6 7.7 5.9 3.3 100.0 Private facility 5.0 5.7 5.8 7.1 9.2 9.1 11.3 12.0 15.0 19.8 100.0 Private hospital 3.6 4.8 4.5 6.1 8.0 9.0 11.1 12.3 16.8 23.8 100.0 Private clinic 8.9 8.8 10.0 10.3 13.0 9.1 11.7 11.0 9.5 7.7 100.0 Share INPATIENT utilization by income decile (%) (insured) Public facility 9.8 9.8 9.7 9.6 9.3 9.9 10.7 10.1 10.8 10.3 100.0 Public hospital 8.4 8.3 8.7 9.2 9.1 9.8 11.3 11.0 12.0 12.2 100.0 Puskesmas 15.5 15.4 13.8 11.4 9.8 10.5 8.7 6.4 5.8 2.7 100.0 Private facility 5.1 5.4 5.2 6.4 8.8 8.4 10.8 11.6 15.8 22.5 100.0 Private hospital 3.9 4.4 4.1 5.6 7.7 8.3 10.8 11.8 17.4 26.0 100.0 Private clinic 9.8 9.4 9.1 9.2 13.7 8.6 10.9 10.8 9.6 8.9 100.0 Source SUSENAS (2015). 32 WHO’s SARA Reference Manual lists 50 indicators for general service readiness while Rifaskes collected data on 38 related indicators. 33 For more information on geographical disparities in public services, see World Bank 2012d. section four. HEALTH SYSTEM Table 4.3 Deficiencies and Regional Variation in Provision of Key Health Services Key Health Health Facility Census (2011) Indonesia Family Life Survey (2014) Service Some 80% of puskesmas lacked one staff Some 42% of puskesmas lacked one staff member trained in the previous two member trained in the previous one year in years in family planning services, and 38% lacked family planning guidelines family planning services. Family planning available at the facility. About 60% of private clinics lacked combined oral About 40% of private clinics lacked contraceptive pills and about 35% lacked injectable contraceptives. combined oral contraceptive pills and about 20% lacked injectable contraceptives. In North Sulawesi, Maluku, and Papua, less than 60% of puskesmas were able Some 30% of puskesmas lacked the ability to diagnose anemia with hemoglobin testing, while urine tests were almost to do hemoglobin tests and about 50% of completely unavailable in Gorontalo, North Sulawesi, and Maluku. Only 14% of puskesmas lacked the ability to do urine Antenatal care the 30 private hospitals and 15% of private clinics surveyed were able to conduct tests. A total of 90% of private primary care hemoglobin or urine tests. This largely explains why only 25% of public hospitals, facilities lacked the ability to do urine tests and none of the 30 private hospitals surveyed maintained all eight antenatal care and only one-half of private facilities were tracer items. able to do hemoglobin tests. Only about 20% of puskesmas had at least one of their staff trained in safe delivery in Only 62% of puskesmas mandated to provide BEONC treatment had at least one the last one year, about 30% lack delivery Basic obstetric staff trained in this area in the previous two years. Only 39% of public hospitals, sets, about one-half of puskesmas and care and 3% of the 30 private hospitals surveyed, maintained all 23 basic obstetric care less than one-half of private clinics have tracer items. uterotonic agents such as oxytocin or ergometrine. More than 20% of puskesmas reported More than 20% of puskesmas in Papua, West Papua, and Maluku reported that they that they did not have measles, DPT, polio did not have measles, DPT, polio, and BCG vaccines, while only about one-quarter Immunization and BCG vaccines,34 while only about of private facilities, and less than 10% of those in the eastern provinces, reported one-quarter of private facilities reported availability of these vaccines. availability of these vaccines. Antimalarial medicine was not available in 38% of puskesmas and malaria blood Antimalarial medicine was not available in Malaria tests were not available in 29% of puskesmas in the 10 provinces with the highest about 60% of puskesmas in areas with the malaria prevalence rates. highest malaria prevalence rates. 51 Some 40% of puskesmas did not have A total of 35% of puskesmas did not have staff trained in TB management, and 27% the ability to diagnose TB from sputum TB did not have the capacity to diagnose TB from sputum samples, while crucial first- samples, and 35% did not have anti-TB line treatment was not widely available either in puskesmas or public hospitals. medicines. Only 66% of public hospitals, and 27% of the 30 private hospitals surveyed, Only 70% of all puskesmas reported the maintained all seven diabetes tracer items. Only 54% of all puskesmas reported the ability to do blood glucose tests, and about Diabetes ability to test for blood glucose–a crucial aspect in the management of diabetes– 65% have medicines such as metformin to and only 47% reported the ability to test urine, with availability of each test well control blood sugar. below 20% in six eastern provinces. Very low availability was evidenced for provision of many key basic surgery items, including nasogastric tubes (16%), guidelines (21%), trained staff (29%), adult Basic surgery and pediatric resuscitators (47%), oxygen (53%), and scalpel handle with blade Not available (56%). Only 53% of public hospitals and 60% of the 30 private hospitals surveyed maintained all 12 basic surgery tracer items. Only 20% of all public hospitals, and none of the 30 private hospitals surveyed, maintained all six blood transfusion items. Only DI Yogyakarta (47%) and West Sumatra (41%) had more than 40% of public hospitals with all items, while eight Blood provinces had less than 10% of public hospitals with all items. Blood typing Not available transfusion capacity was largely unavailable in private hospitals (11%). Hospitals scored very low on sufficient blood supply (public, 41%; private, 13%) and blood supply safety (public, 44%; private 37%). Only 18% of all public hospitals, and 33% of the 30 private hospitals surveyed maintained all nine comprehensive surgery items. Only Bali (62%), DKI Jakarta (47%), Comprehensive and Banten (44%) had more than 40% of public hospitals with all items. In contrast, Not available surgery a large majority of provinces (25 out of 3335 ) had less than 30% of public hospitals with all items, including eight provinces with zero hospitals reaching this target. Source Rifakes (Health Facility Cencus) 2011 and IFLS 2014. 34 The question and observation noted the unavailability of vaccines for the past 30 days in the facility. 35 There are currently 34 provinces, North Kalimantan province has been created since this census was undertaken. HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better Human Resources Indonesia has rapidly increased the supply of core MoH 2012 data reported large numbers of unfilled human resources for health (HRH) in recent years. posts of nurses at puskesmas (more than 10,000) The core HRH to population ratio in 2013 was and hospitals (close to 90,000) (World Bank 2014b). estimated at 2.3 per 1,000, equal to the minimum Although midwife and physician availability were recommended by WHO as necessary to attain an 80 similar at around 0.5 per 1,000 population, midwife percent skilled birth attendance rate.36 Of the 2.3 HRH distribution was much better because of the workers, physicians were 0.5, nurses were 1.3, and government policy to deploy midwives down to midwives were 0.5 per 1,000. The nurse-to-physician the villages to improve access to maternal health ratio was 2.6, close to the average observed across services. On the other hand, the challenge on the OECD countries. Most of the recent rise in HRH has availability and distribution of nutritionists and come from increased output of private universities. laboratory technicians at the puskesmas level was Indonesia has also made significant investments in still quite significant. Competency of HRH workers is improving the quality assurance system of health generally low and variable: evidence from vignette 52 professional education by strengthening the school responses indicates poor knowledge and awareness accreditation system and introducing nationally of diagnosis and treatment options in several parts of standardized competency testing for graduate the country. certification. This was much needed because of the rapidly growing number of health professional A large proportion of physicians and midwives are schools, especially those that are privately managed. employed in the public sector. Public HRH staff can either be permanent civil servants (PNS) or contract Despite having attained the minimum WHO norm employees (PTT),38 the latter being either physicians in terms of aggregate numbers of workers, HRH or midwives. Despite shortages of nurses in public remains a key challenge for Indonesia’s health health facilities, recruitment of nurses, either as PNS sector. Key issues include maldistribution, a shortage or PTT, appears not to be a government priority of specialists, and poor skills of health workers. yet. In principle, deployment of HRH is determined Inequalities in the distribution of HRH between based on a combination of subnational proposals and geographical regions and provinces, and between centralized allocations based on norms. In practice, urban and rural areas, are stark. For example, HRH PNS allocations are based on available slots the physician-to-population ratio in Kalimantan (formasi) and, since the 1990s, the government has and Maluku-NTT-Papua is, respectively, one-half had a zero-growth policy for the civil service and and one-third of that in the Java-Bali region. The current allocations for PNS HRH are determined geographic maldistribution for specialists is even largely by central Ministry of Finance (MoF) resources worse than for physicians. Across Indonesia there allocated for this purpose channeled through DAU. is an acute shortage of nurses in puskesmas and Formasi in each district is based on attrition only. hospitals compared to MoH standards.37 36 Indonesia Medical Council and Badan Pengembangan dan Pemberdayaan Sumber Daya Manusia Kesehatan (Badan PPSDMK) (MoH) 2013 data. 37 MoH standards require employment of six nurses for each regular puskesmas and 10 nurses for each puskesmas with beds, while for type A and B hospitals, the standard is one nurse for every bed and for types C and D hospitals, two nurses for every three beds. 38 PNS: Pegawai Negeri Sipil; PTT: Pegawai Tidak Tetap. section four. HEALTH SYSTEM If a district has special needs (for example, specialists Since 2013, freshly graduated physicians have to go are urgently needed), it can negotiate with the through an internship with four months at puskesmas Ministry of Administrative and Bureaucratic Reform and eight months in public hospitals. The MoH has (MENPAN) in which case MENPAN will check with considered task shifting as an option to address MoF regarding resource availability. If, however, the HRH availability challenges but this has not yet been total salary allocation in a given district is already formally endorsed/implemented. Article 73 of the above 55 percent of DAU transfers, not all formasi will 2004 Medical Practice Act, for example, makes it be filled. PTT are proposal-based contractual staff possible for nurses and midwives to perform medical deployed at subnational levels and paid for either out practices as long as they are authorized by regulation. of the MoH budget or–in the case of PTT physicians– Aside from physicians, family planning and counseling from the APBD budget. Presidential Regulation in Indonesia are also provided by midwives. No. 81/2004 includes a formula to guide regions in calculating staff need based on workload, but this Indonesia has a long-term strategy for HRH covering has never been used. the period of 2011-25. Under the overarching objective for everyone to have access to qualified Most village midwives have been contracted and health workers, the strategy has four objectives: deployed by the central government, however the (i) strengthening regulation and planning for HRH government is considering a plan to change this so development; (ii) improving the production/education that current contracted midwives would, become civil of HRH to meet service delivery needs; (iii) assuring servants paid out of APBD (this is a one-off change). the equitable distribution, utilization, and development Future contract midwives would be paid out of APBD of HRH; and (iv) improving supervision and quality without any guarantees for conversion to civil servant control of HRH. The long-term plan sets strategic status. These changes could have a significant impact goals for HRH indicators, including ensuring that there on frontline delivery of health services: on the one are 0.96 general physicians per 1,000 population by 53 hand, this can increase flexibility of districts to hire and 2019 and 1.12 by 2025; similarly, medical specialists are deploy village midwives without being constrained to increase to 0.24 per 1,000 in 2019 and to 0.28 per by central dictates. On the other hand, this may result 1,000 in 2025 (MoH 2011). in an exacerbation of inequalities given variations in district-level fiscal and managerial capacity as well as Despite improvements in coverage and access, the in midwife per capita ratios. quality of HRH has tended to be low and stagnant in Indonesia. Although some improvements can be Dual practice is legally allowed in Indonesia, observed from comparisons between diagnostic and almost 70 percent of physicians and over 90 vignettes from the 1997 and 2007 Indonesia Family Life percent of midwives in puskesmas reported as Survey (IFLS), the changes are marginal and overall doing so (Rokx et al. 2010).39 Allowing dual practice quality of services remains low, with only around represents practical challenges, especially when one-half of the health workers responding correctly to the system is largely unregulated and unsupervised. standard questions and procedural vignettes. Physicians spending more time in private practice is often reported and is an important reason for absenteeism in public facilities. The basic salary for HRH is generally low, but allowances are relatively high. The dual-practice policy also contributes to difficulties in deploying physicians to rural areas where there are less opportunities to earn extra income from private practice. 39 See also World Bank (2008). HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better 54 section five. HEALTH FINANCING section 5 . 55 HEALTH FINANCING HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better In Summary 1. Total health spending, and government health spending has increased for the past few years and is expected to increase further to meet the government’s target of 5 percent of General Government Expenditures (GGE) in 2016, however, it remains one of the lowest in the world. The low spending is the result of low prioritization and ability to generate revenue. 2. OOP spending continues to be the largest share of THE, around 45 percent in 2014; partly due to a large population that is still uncovered. The vulnerability to be pushed into poverty due to health shocks remains high. 3. Government spending is around one-third of THE; more than 60 percent spending occurred at the subnational level with complex intergovernmental transfer. 4. JKN is one of the largest SHI schemes in the world covering 57 percent of Indonesia’s population, while it accounts for a small fraction of THE. Challenges of mistargeting and covering nonsalaried, nonpoor workers remain. 5. Comprehensive benefit package without adequate financing leads to limited service availability. 6. Indonesia spends two-thirds of THE on curative care; more than 65 percent of JKN expenditures were for hospital-based inpatient and outpatient care. 7. External financing for health remains at a low level (1 percent of THE), but it continues to play a significant role for several key health programs. 56 section five. HEALTH FINANCING THE per capita was US$126 in 2014, about 3.6 health as a share of GDP have been outpacing GDP percent of GDP. About 41.4 percent of THE was growth rates since around 2000, with growth in the public (composing government budgetary and social former generally outpacing the latter (Figure 5.2). insurance expenditures) with the remainder being private (three-fourths of this is OOP spending by Projected economic growth is likely to increase households; the remainder being private/corporate overall public spending on health. Over 1995-2014, insurance and spending by NGOs). Indonesia the elasticity of public spending on health (including is a significant outlier when it comes to health central, subnational, and SHI) to GDP per capita has expenditures: its total and public spending share of been about 1.2, implying that for every 1 percent GDP is one of the lowest in the world, far below what change in GDP per capita, public spending changed might be expected for its income level and when by 1.2 percent on average (Figure 5.3). With an compared with regional peers (Figure 5.1). average annual growth rate of GDP per capita of 8.1 percent expected over 2017-21, and assuming the In 2014, total (public) health expenditures as a elasticity follows the same trend as it has over the share of GDP was low at 3.6 percent (1.5 percent) period 1994-2014, this would imply an increase of in Indonesia, compared to 5.9 percent (3.3 percent) almost 10 percent per year in public spending on among lower-middle-income countries and health per capita over the next five years. 6.6 percent (4.98 percent) in the EAP region. A combination of low levels of national government As might be expected, elasticity of budgetary health revenues/expenditures (mentioned earlier), low spending varies between central and subnational prioritization of health in the government budget governments. While the average elasticity for central (discussed in more detail below), high levels of government health spending to GDP per capita has informality, and relatively low levels of utilization of only been 0.85, implying that a 1 percent increase health care services (discussed earlier) help explain in GDP per capita has led to an increase in central 57 the low levels of total and public health expenditures government health spending of only 0.85 percent in Indonesia. Both total and public expenditure on on average, post decentralization subnational Figure 5.1 Total and Public Expenditure on Health as Share of GDP vs Income (2014) Share Share of GDP Total Health Expenditure of GDP Public Health Expenditure (%) (%) 20 20 15 15 10 10 South Africa Brazil Vietnam Rusia Thailand Cambodia Thailand China Solomon Islands 5 Solomon Islands Philippines 5 India South Africa Papua New Guinea Nigeria Malaysia Brazil Vietnam Indonesia Papua New Guinea Rusia Ghana Sri Lanka China Ghana Malaysia 2 2 Sri Lanka Lao PDR Philippines India Indonesia Cambodia 1 1 Lao PDR Nigeria LOWER UPPER LOWER UPPER MIDDLE MIDDLE MIDDLE MIDDLE LOW INCOME INCOME INCOME HIGH INCOME LOW INCOME INCOME INCOME HIGH INCOME .5 .5 250 500 1000 2500 10000 35000 100000 250 500 1000 2500 10000 35000 100000 GNI per capita, US$ GNI per capita, US$ Source World Development Indicators database 2016 Note : (i) Indonesia 2014 figure based on NHA country report. (ii) Both x and y axes in log scale. HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better Figure 5.2 Total and Public Expenditure on Health as Share of GDP (1995-2014) IDR 1,600,000 4 Share per of GDP capita (%) 3.5 1,200,000 3 Total as % share of GDP (right axis) 2.5 800,000 2 Public as % share of GDP (right axis) Total per capita (left axis) 1.5 400,000 1 .5 Public per capita (left axis) 0 0 1995 2000 2005 2010 2014 Source World Development Indicators database 2016 Note Total and public spending is in 2014 constant IDR. 58 Figure 5.3 Elasticity of Public Spending on Health (1995-2014) Public .6 expenditure on health .4 per capita .2 Elasticity=1.2 .1 .05 2 3 4 6 10 15 20 30 50 GDP per capita Note data in IDR millions section five. HEALTH FINANCING Figure 5.4 Health Financing Flows SOURCES OF AGENTS PROVIDERS SPENDING Central Government Ministry of Health Puskesmas Provincial Provincial Health Office Public Hospitals Government District District Health Office General Practitioners Government 59 Private Firms BPJS Private Clinics Households Jamkesda Private Hospitals External Resources Private Insurance Pharmacies HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better government elasticity to GDP has been higher–at 1.3. collection of revenues from households, private firms, At 45.3 percent of THE (1.6 percent of GDP), OOP and external sources and transfers to subnational spending by households remains the largest source governments and health facilities. BPJS is the SHI of health financing for Indonesia. This was followed administrator that collects contributions from the by general government budgetary expenditures (41.4 government, private firms, and households into a single percent of THE or 1.5 percent of GDP). SHI accounted for national pool and purchases health services from public 13 percent of THE (0.5 percent of GDP) in 2014. As with and empaneled private providers (in orange). Private other countries in the region such as the Philippines flows include OOP spending from households and firms and Vietnam, in addition to large levels of OOP health at public and private facilities (in blue). Over the past spending, Indonesia’s public health financing system five years, only about 1 percent of THE has come from is characterized by the coexistence of traditional external sources. From the perspective of per capital government budgetary supply-side health financing spending on health, among comparators, Indonesia’s and demand-side SHI financing. It is not clear why this health financing situation most closely mirrors that of dual cofinancing modality remains and whether this will the Philippines and Vietnam (Table 5.1). change in the near- to medium term. The next subsections provide an overview on each of Figure 5.4 summarizes the prominent financing flows the four major health financing sources: government in Indonesia’s health system. Government budgetary budgetary spending, SHI, OOP spending, and funding flows (in black) include the process of external financing. Table 5.1 Key Health Financing Indicators (2014) 60 THE expenditure Share of Public SHI share OOP share External share Country per capita GDP (%) share (%) (%) (%) (%) Brazil US$947 8.3 46.0 0.0 25.5 0.0 Cambodia US$61 5.7 22.0 0.0 74.2 16.3 China US$420 5.5 55.8 37.7 32.0 0.0 India US$75 4.7 30.0 1.7 62.4 1.0 Indonesia* US$126 3.6 41.4 13.0 45.3 0.8 Lao PDR US$33 1.9 50.5 1.6 39.0 31.8 Malaysia US$456 4.2 55.2 0.6 35.3 0.0 Philippines US$135 4.7 34.3 14.0 53.7 1.4 Russia US$893 7.1 52.2 27.7 45.8 0.0 South Africa US$570 8.8 48.2 1.2 6.5 1.8 Sri Lanka US$127 3.5 56.1 0.0 42.1 1.3 Thailand US$360 6.5 86.0 5.1 7.9 0.0 Vietnam US$142 7.1 54.1 24.1 36.8 2.7 East Asia and Pacific US$217 4.9 49.9 12.1 40.5 6.6 Lower-middle-income US$106 4.2 44.4 8.6 46.5 6.5 Source World Development Indicators database 2016. * Indonesia data is based on the NHA country report, 2014 (Ministry of Health -Center for Health Economic Policy Studies-AIPHSS. 2015) section five. HEALTH FINANCING Government Budgetary Expenditure on Health National government budgetary expenditures on 2008. Provisional estimates indicate a decline in health amounted to IDR 467,959 (~US$39) in per 2015, however, it remains to be seen if this trend is capita terms in 2014 and are the second-largest realized (Figure 5.5). source of financing for health in Indonesia. In the same year, in aggregate, national government Health’s share of the national (that is, combined expenditures on health were IDR 118.3 trillion central and subnational) budget is relatively small (~US$9.9 billion) in 2014, about 1.1 percent of GDP40 in Indonesia. WHO data indicate that Indonesia’s based COFIS data and 1.5 percent of based on prioritization for health is on the lower side in the data from the Indonesia NHA country report. global comparisons: several countries including the Aggregate national expenditures on health have Philippines, China, South Africa, and Thailand devote increased by an average of 7 percent per year since a much larger share of the budget to health (Figure the advent of decentralization in 2001. National 5.6).41 At 4.7 percent, health’s share of the national government health expenditure has also been rising budget is small relative to that of general government as a share of GDP and as a share of total national administration (~20 percent), subsidies (~20 percent), 61 government expenditures since the turn of the education (~20 percent), and infrastructure (~10 century, albeit at a somewhat slower pace since percent). The combination of a relatively small overall Figure 5.5 National Government Budgetary Expenditures on Health (1995-2015) Share (%) 6 Share of national government expenditure 5 4 3 2 Share of GDP 1 0 1995 2000 2005 2010 2015 Source Indonesia COFIS database 2016. Note 2015 numbers are provisional. 40 Data was generated using The Indonesia Consolidated Fiscal dataset (COFIS). The database has been developed by the World Bank COFIS team and contains data on expenditure from the central and subnational (provinces, districts) governments. The data comes from publicly available data sources, managed by the Government of Indonesia (GoI) and, unless indicated otherwise, is audited realized expenditure data. 41 In WHO data, this share is calculated by combining government budgetary expenditures and social health expenditures HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better Figure 5.6 Health’s Share of National Budget (2014) Percentage (%) 30 25 Thailand 20 South Africa 15 Vietnam China Russia Sri Lanka 10 Philippines Malaysia Brazil India Cambodia 5 Indonesia Lao PDR 0 Source World Development Indicators database 62 Indonesia using COFIS databse Figure 5.7 Trends in Government Health Spending by Level of Government (1995-2013) Rp trillion 1000 800 CENTRAL 600 PROVINCE 400 200 DISTRICT 0 1995 2000 2005 2010 2013 Source Indonesia COFIS database Note Data in 2013 constant IDR section five. HEALTH FINANCING share of government spending as a share of GDP and the formal approval of the plan by the parliament or relatively low prioritization given to health are two the local council for the national and local budgets, reasons why Indonesia’s health expenditure share of respectively. For the past ten years, expenditures for GDP is one of the lowest in the world. health have closely tracked allocations.   District governments have taken an increasingly There are some expenditures that occur at the dominant role in government health spending subnational level but are financed centrally and postdecentralization in 2001. Over one-half of national are recorded under APBN and not under APBD. government expenditures on health now occur at the For example, deconcentration funds (DEKON) district level, up from an average of less than 10 percent are allocated by line ministries under APBN (for predecentralization (Figure 5.7). The provincial share example, to MoH) but are administered by provincial of government health expenditures has also declined: governments (so for health these are administered from an average of over 30 percent predecentralization by PHOs) and are used to finance nonphysical to just over 15 percent postdecentralization. The activities, for example for technical assistance, level of decentralization as reflected in government training, supervision, research, and promotion. expenditures for health is similar to that for education Co-Administered Tasks (Tugas Pembantuan, TP) for which, in 2013, 57 percent of spending occurred at are allocated in line ministries for in-kind grants the district level, 36 percent at the central level, and 7 to districts for vaccines, drugs, and supplies. MoH percent at the provincial level. also pays for the salary of contract physicians and contract midwives (PTT) employed by districts. Annual planning and budgeting occurs in parallel PTT physicians work in the puskesmas, while PTT top-down and bottom-up streams. The top-down midwives are usually based at the village (some stream creates a national budget and seeks to districts recruit additional physicians/midwives under anticipate and prepare for the financial needs from PTT using their own resources). 63 the central budget, whether for core or noncore functions. To this end, stakeholders must consider There are budgetary benchmarks for health the overall government budget and the relative spending. In 2009, the DPR enacted Law No. 36/2009 importance of various priorities, including health, stipulating that at least 5 percent of the central in the budget. These competing priorities are budget (APBN) and 10 percent of the district budget considered in the context of revenue forecasts (APBD), excluding salaries, be allocated for health. In established by the Fiscal Policy Office of the MoF. addition, the law states that at least two-thirds of the Bottom-up planning, on the other hand, considers health budget from the central and district budgets local needs and ability to address these needs in the should be prioritized for public services, in particular development of local plans and budgets. health services benefitting the poor, elderly, and disadvantaged children. The process undertaken by the health unit, for example, considers the health profile of the population By “function”,42 health received less than 2 percent of in a given area, and seeks to put into place a plan the central government budget in 2013. The largest to address these needs. This plan considers that share of central government expenditure was for fuel current status and additional needs of the health and other subsidies (Table 5-2). Premium payments for system inputs, including the number and mix of health insurance for the poor and near-poor–which are health workers, drugs, equipment/supplies, and functionally classified under “general public services”– infrastructure availability. These plans are then were less than 1 percent of all central government integrated with similar plans from other sectors, for expenditures. A key policy change is that planned example, education and infrastructure, and compiled expenditures for fuel and other subsidies have declined into a single, integrated district plan. District plans are significantly for 2015: from almost one-fifth of the collated and merged to create provincial plans. In both budget in 2013 and 2014 to only 5 percent in 2015. This cases, planning and budgeting follows a specified has not, however, had an impact on health’s share of the schedule and culminates in October each year with government budget that remains at less than 2 percent 42 The government categorizes budget expenditures into 11 functions. HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better Table 5.2 Central Government Expenditures by Function (2013-15)43 2013 2014 2015 Expenditure Category IDR trillion Share (%) IDR trillion Share (%) IDR trillion Share (%) General public services 706 62.0 798 66.2 695 52.7 Fuel subsidies 210 18.5 240 19.9 65 4.9 Electricity subsidies 100 8.8 102 8.5 73 5.5 Nonenergy subsidies 45 4.0 50 4.2 74 5.6 Interest payments 113 9.9 133 11.1 156 11.8 Premiums for poor and near-poor 8 0.7 20 1.7 20 1.5 Economic affairs 108 9.5 97 8.1 216 16.4 Defense 88 7.7 86 7.2 102 7.8 Education 115 10.2 123 10.2 156 11.8 Health 18 1.5 11 0.9 24 1.8 Social protection 17 1.5 13 1.1 23 1.7 Other 86 7.6 76 6.3 103 7.8 Total 1,138 100 1,204 100 1,319 100 Source LKPP 2013-2014 Audited. Note Budget revised memorandum 2015. for 2015. The largest beneficiary of the decline in fuel Within the health sector, expenditure allocations 64 and other subsidies appears to be the economic affairs can be assessed from an economic as well as function and, to a lesser extent, education. a functional perspective. Functional allocations   delineate expenditures based on the purpose By “sector”,44 health’s share of the central government towards which funding is targeted (for example, budget was 3 percent in 2013. The government’s individual versus community health care), while an sectoral classification for health includes health- economic classification focuses on the economic related expenditures undertaken by nonhealth line characterization of spending (such as, capital ministries, premium payments for health insurance for versus recurrent). Less than 15 percent of central the poor and near-poor, as well as interfiscal transfers government expenditures for health were for related to health (for example, DAK). In the 2016 personnel costs, over 51 percent was for goods and budget and 2017 budget plan, the health share of the equipment, roughly 11 percent for capital, and 25 central budget reached and stabilized at 5 percent, the percent for social assistance. Although there are legally mandated amount for the sectoral health share no global optimal norms for assessing economic of central government expenditures. The changes in classification shares, comparisons with other 2017 will include another increase of DAK for health countries suggest that both the personnel cost following last year’s significant increase that more share of government health expenditures and for than doubled from 2015. While the 2017 budget plan goods and equipment are on the low side for central maintains the 2016 health share of the central budget government expenditures. By function, most (58 at 5 percent, it is against a reduced overall revenue. percent) of central government health spending was The budget envelope for the health sector may for individual health care, 10 percent was for drugs experience a slight decline in nominal terms in 2017, and medical supplies, 10 percent for community IDR 103.5 trillion, compared to IDR 104.1 trillion in 2016. health, and 9 percent for family planning. 43 This excludes intergovernmental fiscal transfers. 44 Since 2011, some health spending on goods, services and capital items from nonprofit public service agencies (Badan Layanan Umum), puskesmas, and regional public hospitals (RSUD) has been reclassified into the “General public services” function instead of “Health function”. To make the definition consistent across time, the World Bank developed a new classification of “Sector” that put the above spending back into the health function. section five. HEALTH FINANCING Health represents about 10 percent of both district Aggregate numbers mask huge variations across and provincial government expenditures. At least districts in government budgetary health spending, in aggregate across districts, health meets the both in levels and as a share of district expenditures. A legally mandated minimum requirement for health rapid assessment across 44 districts showed health’s expenditures. Education was the biggest share of share of the district budget varies from 3 percent to district expenditures (accounting for more than over 18 percent, with an average of 10 percent in 2013 one-third of the spending). General government (Figure 5.8). This translated into fairly large variations in administration represented the greatest share of per capita terms. provincial expenditures (Table 5.3). Table 5.3 Subnational Government Expenditures (2013) IDR trillion Share (%) IDR trillion Share (%) Expenditure Category Districts Provinces45 Education 179 34.1 21 12.1 General government administration 137 26.1 70 40.5 Infrastructure 88 16.8 36 20.8 Health 53 10.0 18 10.4 Other 68 13.0 28 16.2 Total 525 100 173 100 65 Source COFIS database; World Bank staff calculations based on MoF data. Figure 5.8 Health’s Share of Government Budget and Per Capita Spending Across 44 Districts (2013) Percentage (%) IDR thousands 20 800 Health share of district government budget (left axis) Government health expenditure per capita (right axis) 15 600 10 400 5 200 0 0 Kab. Indragiri Hilir Kab. Lampung Timur Kab. Ogan Komering Ilir Kab. Kaur Kab. Muara Enim Kab. Luwu Kab. Nias Selatan Kab. Labuhan Batu Kab. Deli Serdang Kab. Tasikmalaya Kab. Kediri Kab. Bandung Kab. Banyuwangi Kota Banjar Baru Kab. Pesawaran Kota Bandung Kab. Garut Kab. Tuban Bengkulu Kota Banjarmasin Kota Bandar Lampung Kab. Langkat Kab. Kendal Kab. Lampung Barat Kab. Majalengka Kab. Labuhan Batu Utara Kota Surabaya Kota Lhokseumawe Kab.Lahat Kab. Wajo Kab. Jeneponto Kota Cilegon Kab. Bangka Barat Kab. Aceh Jaya Kab. Maluku Tenggara Barat Kab. Simeulue Kota Pare-Pare Kota Padang Kab. Pemalang Kab. Batu Bara Kab. Malang Kab. Jombang Kota Pasuruan Kab. Lampung Utara Kab. Pekalongan 45 Excludes intergovernmental transfers to districts. HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better Social Health Insurance (SHI) SHI expenditures are the third-largest source of program). Salaried workers employed in the public financing for the health sector in Indonesia, accounting and private sector pay 5 percent of their salary (3 for 13 percent of THE. BPJS revenue from contributions percent employer and 2 percent employee for public amounted to almost IDR 52.8 trillion (~US$3.96 billion) sector workers; 4 percent employer and 1 percent in 2015, about 40 percent of national government employee for private sector workers). This group budgetary expenditure on health. Official BPJS reports includes those who were previously covered under indicate that JKN covered 156.8 million individuals Askes and Jamsostek, respectively. Nonsalaried, (~60 percent of the population) in 2015. This represents nonpoor workers in the informal sector are expected an average expenditure of IDR 336,735 (~US$25) per to pay a voluntary fixed premium contribution member per year. (ranging from ~US$36-72 per year) upon enrollment in the program. The local health insurance schemes BPJS pools contributions from three broad categories are expected to fold into the national scheme, JKN, of people: (i) the poor and near-poor; (ii) salaried in 2016 which will reduce the opportunity to use local 66 workers in the formal sector; and (iii) nonsalaried, health insurance schemes as a political influence for nonpoor workers in the informal sector. Fixed local election (Pisani, Kok, Nugroho, 2016). premium contributions of IDR 23,000 (~US$2) per person per month are paid for entirely by the central JKN benefits are unified, except for hoteling government for the poor and near-poor. This group entitlements which vary by level and type of was previously covered under the Jamkesmas contribution. JKN membership includes 87.8 million section five. HEALTH FINANCING (56 percent of those covered) noncontributory BPJS (the latter are slated for phase-out by 2017) central government-financed poor and near- (World Bank 2015d). Salaried workers contribute poor, 37.9 million (24 percent) public and private a larger share to the overall revenues (49 percent) sector contributory salaried workers, 20 million than their share of membership in JKN (Table 5.4). (13 percent) contributory nonsalaried, nonpoor Contribution collection from nonsalaried (informal individuals, and 11.2 million (7 percent) covered under and unemployed) workers is disproportionately small subnational Jamkesda programs administered by relative to their share of membership. Table 5.4 JKN Membership and Contributions by Type (2015) Membership Contributions Classification of Member Number Share of total Amount (IDR Share of total (%) (million) (%) trillion) Salaried 37.9 24.2 25.8 48.8 Public – – 15.0 28.4 Private – – 10.8 20.5 Nonsalaried 20.0 12.8 4.7 8.9 Informal 15.0 9.6 – – Unemployed 5.0 3.2 – – 67 Poor and near-poor 99.0 63.1 22.3 42.2 Central government-financed 87.8 56.0 19.9 37.7 Subnational government financed 11.2 7.1 2.4 4.5 JKN (total) 156.9 100 52.8 100 Source BPJS 2015. Table 5.5 SHI Expenditure Pre- and Postunification (2013-15)47 Expenditure per member (Average IDR) SHI Program 2013* 2014** 2015** Askes 500,000 – – Jamsostek 60,000 – – Jamkesmas 100,000 – – Askes+Jamkesmas+Jamsostek 132,000 – – JKN (PBI) – 94,098 100,455 JKN (Non-PBI) – 635,318 539,668 JKN (total) – 249,281 262,344 Source *Annual reports; **Author estimates. 46 Authorestimates. It does not include capitation and noncapitation amounts at the primary care level. (Based on Ministry of Health presentation: “Introduction to Constructive Dialog for JKN Improvement”. Jakarta, May 30, 2016 HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better In per-member terms, SHI expenditures have almost fall under promotive, preventive, curative and doubled between 2013 and 2015.47 While in 2013 rehabilitative services. All registered JKN members Indonesia had several fragmented programs–Askes are entitled to a range of medical services, including for the formal public sector, Jamsostek for the formal consultation and treatment at a primary health care private sector, and Jamkesmas for the poor and near- center (puskesmas or empaneled private clinic). poor, each with different benefits and expenditure Primary care facilities act as gatekeepers and manage patterns–implementation of the unified JKN program access to higher level services. Service coverage as of 2014 appears to have resulted in consolidation categories are outlined below in Table 5.6. and a big increase in financing from SHI in Indonesia’s health sector (Table 5.5). JKN has a “negative list” that specifies what is not covered. The negative list includes: (i) health services JKN membership coverage has increased significantly. that do not follow stipulated procedures, including By the end of April 2016, BPJS Health reported 165 referrals; (ii) health services in facilities that are not million of the population have been covered, a 24 contracted by BPJS, except under emergencies; (iii) percent increase compared with the end of 2014. health services that are covered by occupational The insurance coverage is relatively high among accidental insurance; (iv) health services abroad, low-income and high-income groups, but it remains cosmetic procedures, health services for infertility, and relatively low amongst nonpoor informal sector orthodontic services; (v) health disorders/diseases workers (only about 7 percent of the nonpoor informal caused by drug addiction and/or alcohol; (vi) health sector population currently has JKN coverage); hence, problems caused by self-harm; (vii) complementary Indonesia faces a “missing middle” problem. treatment using alternative/traditional medicine, unless deemed effective by health technology JKN’s benefits package is comprehensive and is set assessments; (viii) experimental procedures, health 68 and updated by MoH, not BPJS. The benefits package equipment for households, contraceptives, baby food, is not explicit in that all medically necessary coverage and milk; and (ix) health services for disaster situations. is automatically deemed to be covered without any copayments, balanced billing, or expenditure caps. An explicit benefit package is crucial to ensure the JKN benefits include both medical and nonmedical adequacy of service and financing. In the absence of benefits. Medical benefits include comprehensive an explicit benefit package, providers refer to various health services at the primary, secondary, and tertiary national clinical guidelines and from drugs that are levels; nonmedical benefits include accommodation included in the national formulary (FORNAS) as and emergency transportation to health facilities. JKN’s ‘positive list’. As a result, there are variations in Medical services include a range of services that standards of practice and case management, which in Table 5.6 JKN Service Coverage LEVEL OF CARE TYPE OF SERVICE Primary care Primary care coverage includes: (i) administration services; (ii) promotive/preventive services; (iii) examination, treatment, and medical consultation; (iv) nonspecialist medical treatment, both operative and nonoperative; (v) drug services, medical consumables and materials; (vi) blood transfusion in accordance with medical needs; (vii) laboratory diagnostic primary level; and (viii) primary hospitalization in accordance with medical indications. Secondary and Secondary and tertiary care coverage includes: (i) administration services; (ii) examination, tertiary care treatment and specialist consultation by a specialist and subspecialty; (iii) specialist medical treatment in accordance with the medical indications; (iv) drug services, medical consumables and materials; (v) advanced diagnostic services in accordance with medical indications; (vi) medical rehabilitation; (vii) blood services; (viii) forensic medical services; (ix) corpse in health facilities; and (x) nonintensive inpatient care; and (xi) hospitalization in intensive care. 47 The 2013 numbers exclude Jamkesda outlays as this information was not available. section five. HEALTH FINANCING Box 5.1 Both the noncontributory and contributory approaches Covering the Informal Sector: raise issues. Challenges to the former include the fiscal Lessons from Global Experiences space implications of general revenue financing. While this strategy enables a rapid expansion to the noninsured population, lower-income countries in particular often The path to expanded health coverage in lower-middle- have very large informal sector employment and may income countries generally begins with the dual strategy not have the capacity to do so. Unless new taxes are of enrolling formal-sector workers into contributory introduced to cover the informal sector, the budget impact schemes while the government fully subsidizes health is immediate, forcing trade-offs within the health sector or care for those who qualify as poor. This typical pathway across sectors. This approach may, therefore, work best in omits nonpoor informal workers, who can be difficult countries with relatively small nonpoor informal sectors. to identify and whose income is both uncertain and Other issues to consider include the concern that general often impossible to verify. This has led to the so-called revenue financing may encourage informality (reported problem of covering the “missing middle” for countries in Mexico) and misreporting of income. The latter was a seeking to achieve UHC. serious issue with the Chilean system, with an audit finding that up to 400,000 persons had misreported their income Global experience suggests two basic approaches to in order to avoid paying contributions. providing coverage for nonpoor individuals working in the informal sector: (i) noncontributory schemes in which Important challenges to the contributory scheme resources for the poor are extended to the informal include the difficulty of establishing the income of sector (as in Thailand); and (ii) contributory schemes, in informal sector workers, and the costs associated with which schemes targeting the formal sector are extended developing the infrastructure to routinely report and to the informal sector (as in Indonesia), generally in some monitor income. Identification of eligible individuals can tiered form according to ability to pay. Whether a country be a challenge in any event, and is exacerbated when takes the first, the second, or a mix of the two approaches potential beneficiaries may seek to avoid contributions. 69 generally depends on political and economic factors The administrative costs of maintaining a contributory within the country. These include fiscal space capacity scheme can be quite high, especially as informal workers and constraints to expanding coverage, the size and frequently have fluctuating income and their eligibility make-up of the informal sector within the country, and for any subsidies must be regularly reviewed. When no the institutional capacity to identify and verify the income additional budget is provided, however, contributions can of informal sector workers. help to pay for the scheme, and contributory schemes may help to encourage a sense of entitlement, leading Turkey provides health coverage for the nonpoor informal patients to advocate for better services. sector through its Green Card program, which was initially launched in 1992 as a noncontributory health insurance In addition to considering the premium, insurance scheme for the poor. Through a comprehensive schemes must consider whether they will rely on health reform–the “Health Transformation Project”–in mandatory or voluntary enrollment. This is a particularly 2003, Turkey subsequently merged all existing health important consideration for contributory schemes. insurance schemes, including the Green Card program Mandatory contributions can be both challenging and in 2012, into a Universal Health Insurance Scheme costly to implement, while voluntary enrollment schemes managed by the Social Security Institution. Although typically have low uptake and result in substantial targeting of the informal sector has historically been adverse selection. Literature suggests that “…successful difficult, Turkey’s efforts to expand benefits, and improve initiatives to cover this population group are the ones supply-side readiness, coupled with the establishment where the government has abandoned its expectations of a sophisticated, responsive system to determine to derive relatively substantial revenue from it,” and contributions from the informal sector (dependent on typically offer informal workers a smaller benefit package household income, value or size of property occupied, than that offered to formal-sector employees, but that is as well as size and age of car owned) has resulted in accessed at a far lower premium. Although they come expanded coverage for the informal sector. From 2003 to at the cost of a fully equitable universal health insurance 2008, targeting performance of the Green Card improved system, these tiered schemes are designed to encourage and about 70 percent of benefits reached the lowest accurate income reporting and, thereby, strengthen the quintile in 2008, from just 55 percent in 2003, highlighting financial sustainability of the system. the effective targeting of the program and improving levels of financial protection and equity in Turkey. Source: Bitran 2014. HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better the end leads to inefficiency in service delivery. There public sector is the purview of DHOs. MoH has a are several examples of countries (such as Chile) that health technology assessment unit that determines have moved from a model of open-ended “everything changes in medical technology. is covered” to one in which a basic set of benefits is explicitly covered and guaranteed with adequate There is a national formulary (FORNAS) financing from public sources (via government recommended by an expert group managed by MoH budgetary supply-side expenditures and/or social that was adopted in 2013. It is an expanded version health expenditures). of the DOEN that forms the basis for the provision of drugs under JKN. The FORNAS lists the generic BPJS provides coverage in public and empaneled name, usage, and formulation of drugs but does not private facilities and reimburses claims based on list brand or price. Private clinics that are empaneled tariffs set by MoH. BPJS reimbursements do not by BPJS can also use the abovementioned e-catalog cover the full cost of care and there is significant services, however, general practitioners empaneled cofinancing by supply-side government budgetary by BPJS cannot use the e-catalog and must rely on expenditures in the public sector (nevertheless, private pharmacies. Not all of the drugs in FORNAS reimbursements are the same for both public and were available through the e-catalog and the private facilities). Primary care is paid by capitation e-catalog can list more than one supplier for a drug. and reimbursements to hospitals are based on diagnosis-related groups (known as INA-CBG). According to the NHA (2014), more than 65 percent of Procurement of medicines and equipment in the JKN expenditures were for hospital-based inpatient Table 5.7 Top-ten JKN Outpatient and Inpatient Claims (2014 and 2015) 70 2014 2015 OUTPATIENT Other minor chronic disease Other minor chronic disease Dialysis Dialysis Cataract Other minor acute disease Rehabilitation procedure Rehabilitation procedure Physical therapy and minor musculoskeletal procedure Radiotherapy procedure Ultrasound gynecology Wound treatment Other minor acute disease Physical therapy and minor musculoskeletal procedure Radiotherapy procedure Other major chronic disease Wound treatment Cataract Other ultrasound procedure Ultrasound gynecology INPATIENT Cesarean section Cesarean section Other digestive system diagnosis Bacterial and parasitic infection disease Cardiac failure Other digestive system diagnosis Bacterial and parasitic infection disease Cardiac failure Vaginal delivery Nonbacterial infection Abdominal pain and other gastroenteritis Abdominal pain and other gastroenteritis Hypertension Vaginal delivery Simple pneumonia and whooping cough Percutaneous cardiovascular procedures Respiratory infection and inflammation Simple pneumonia and whooping cough Bacterial infection Hypertension Source Siallagan 2015. section five. HEALTH FINANCING (50 percent) and outpatient care (15 percent). About poorest 40 percent–all of whom should have central 20 percent of the expenditure was on capitated government-financed poor and near-poor coverage– primary care at puskesmas and empaneled private reported having so, indicating significant mistargeting clinics. The remainder was for noncapitated INA- and capture by noneligible subgroups. CBG and some limited fee-for-service payments to facilities. A very small amount–less than 1 percent– In 2011, a new list of the poor and near-poor was went towards preventive and promotive activities. formulated to cover over 40 percent of all households MoH Regulation No. 19/2014 specifies that capitation in Indonesia and is now being used as the basis for payments are to be split between financing health a unified registry of potential beneficiaries for all services (60 percent) and supporting operational social assistance programs. The poor and near-poor expenses (40 percent). Operational expenses are now being targeted on the basis of household include medicines, medical devices, and medical per-capita consumption. This is done with a mixture disposables. In addition, while there is no specific of geographic and proxy means-testing methods. regulation stipulating the use of capitation funds for Proxy means-testing indicators are collected on immunization, vaccination services are also generally all households, and these are used to generate a financed by this budget. Some of the largest claims consumption estimate using standard proxy means- for diseases/conditions in 2014 and 2015 are listed testing methods. The consumption estimate is in Table 5.7 and include reimbursements for dialysis, used to select beneficiaries, but this is done on a cesarean births, and vaginal deliveries. district-by-district basis, with a quota set for each district based on poverty rates from the national Significant mistargeting appears to exist under JKN. socioeconomic survey (SUSENAS). Poor and near- Household survey data estimates also indicate poor targeting identifies eligible households, but that 57 percent of households had some form of membership is individual, with each household SHI coverage in 2015 (SUSENAS 2015). Reported member entitled to receive a JKN card. 71 coverage rates tend to be highest among the richest economic deciles and lowest among the middle- One reason behind the suboptimal performance of income groups (Figure 5.9). Only 53 percent of the JKN in terms of targeting is likely to be a variation in Figure 5.9 Coverage by SHI Programs (2015) Percentage 70 (%) Poor and near-poor 60 50 40 30 20 10 0 Poorest 3rd 5th 7th Highest Economic deciles Source SUSENAS 2015. HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better the proxy-means-testing criteria used across districts. costs of around 6 percent of total premiums collected The criteria used to identify household characteristics appears to be reasonable, especially given that JKN is vary across districts; in some districts, village midwives a relatively new SHI program. and subdistrict health center officials often distribute health cards according to their own criteria, regardless There are significant geographic deficiencies in the of economic status (World Bank 2012c). There are no availability and quality of the basic benefits package, specific incentives in the system to either maximize especially for those living in relatively remote enrollment or minimize mistargeting. There is some and rural locations of the country, and this limits anecdotal evidence and allegations of fraud and the effective availability of benefits for many JKN political clientelism, but only a few cases have been beneficiaries. In addition, the architecture of JKN is reported. Since enrollment of the poor and near-poor such that it effectively functions more like a demand- is not mandatory, there is some evidence that the side “top-up” of essentially a (constrained) supply- target beneficiaries enroll only when they need to side system rather than a full-fledged SHI program. use health services. The list of eligible beneficiaries JKN does not reimburse the full cost of care: salaries, compiled by district officials is not subject to capital, and some of the operating costs at public validation from the central government, resulting in facilities continue to be paid for by the government mismatching, poor coverage, and leakage of health (central, provincial, or district, depending on the insurance benefits to the nonpoor. Furthermore, poor type of public facility). Estimates suggest that these and near-poor households that were denied the card subsidies account for upwards of one-half of the full despite being eligible do not have a clear recourse. cost of providing care under JKN. Adverse selection among the nonpoor informal The combination of supply-side constraints and sector is a significant challenge, as are inequities in supply-side subsidies reduces the program’s 72 expenditures. JKN’s overall claims ratio (that is, the overall effectiveness and will likely impact its future ratio of expenditures to revenues) was about 104 sustainability. Supply-side constraints comprise percent in 2014 (Table 5.8) (World Bank 2015d). There all the factors that limit health care delivery at the were, however, wide discrepancies by membership, point of service, including the number of doctors, with the claims ratio among nonsalaried workers being nurses, and midwives; the number of beds; medical more than 600 percent due to adverse selection, equipment and technology; medicine supplies; and while that for the poor and near-poor and others was other basic amenities. Given Indonesia’s geography, closer to 80 percent (Pradiptyo 2015). Even though supply-side constraints reflect not only shortages in expenditure per member for the poor and near-poor overall numbers, but also in distribution. Rural and has almost doubled, large inequities remain across the remote areas are disadvantaged in that they not different subgroups under JKN. There are indications only have fewer health facilities, but also face the of geographic inequities in claims as well, which is difficulties associated with the retention of health not surprising given the maldistribution of facilities personnel, especially doctors. and specialist care across the country. Administrative Table 5.8 Claims Ratio for Nonsalaried Workers vs Others (2014) Aspect Nonsalaried workers All Others Total Membership 9.1 million 124.4 million 133.5 million Contributions IDR 1.9 trillion IDR 38.8 trillion IDR 40.7 trillion Expenditures IDR 11.6 trillion IDR 31.0 trillion IDR 42.6 trillion Claims ratio (%) 617.4 79.9 104.7 Source BPJS section five. HEALTH FINANCING Box 5.2 they relied on public or private insurance. At its launch, AUGE guaranteed access to 56 explicitly defined Closing the Gap Between What is services for priority problems, with a mechanism for the Medically Possible versus What is package to expand over time (in 2010, it went up to 69, Financially Feasible: Chile’s AUGE and then up to 80). Expansions are done following joint consultations between MoH and MoF. Designation of Reforms access under AUGE defines the treatment protocol with an explicit definition of interventions to be guaranteed, and all information is made publicly available on the There is an increasing trend for countries to make their AUGE website. benefits packages (a set of services or health conditions covered by a health financing arrangement, such as health AUGE establishes detailed clinical protocols for each of insurance) much more explicit. The motivation for adopting the 80 conditions covered under AUGE. These protocols explicit benefits packages varies across countries and begin with clinical guidelines on diagnosis and outline includes: (i) to reconcile constitutional rights to health or the appropriate screening, diagnosis, treatment, and government commitments to universal coverage with education procedures. They detail who should be available resources; and (ii) to increase funding envelopes screened and how often, what diagnostic tools are for health by linking budget decisions to entitlements. appropriate for use, and the appropriate treatment by Independent of the immediate financing rationale, these diagnostic outcome. Therapies are described in detail, reforms aim more generally to reduce inequalities in including the make and manufacture of drugs covered access to services, enhance the allocative efficiency of under AUGE and the maximum wait time patients can health systems, and improve financial protection. queue for services. In addition to ensuring treatment, the reform also put caps on waiting time and OOP payments Chile provides UHC to its 17 million people using a mixed for treatment. Copayments range between 0-20 percent, public-private SHI modality. Its SHI system comprises a depending on the type of beneficiary, and annual limits large public insurer (the National Health Fund, or Fonasa) cap copayments at two months’ salary within a given 73 covering three-fourths of the population, including the 12-month period. Beyond this, Fonasa or Isapres are indigent and low- and middle-income citizens, providing required to cover all remaining costs associated with health services mostly through public providers; and eligible services. several for-profit private insurers (Isapres) that cover the better-off population, comprising about one-sixth of the While services not included in AUGE are not guaranteed, total population, providing services almost exclusively they are also not excluded from care. More than one-half in the private sector. Until 2005, the system lacked an of the Fonasa budget goes to non-AUGE services, and explicit benefits package: as a result, large differences in expectations around access to nonguaranteed services the content and quality of services between Fonasa and are a serious challenge to Chile’s health system. In part the Isapres emerged. Combined with limited financing, due to the reform, Chile has seen improved access the implicit nature of benefits resulted in rationing by to services for all citizens. While AUGE establishes a queues, (sometimes unofficial) user fees, and poor minimum standard for all beneficiaries, Isapres is working quality. Rationing in the form of denials and deflection to increase its share of coverage, and offers additional were the prime mechanisms to contain demand. benefits that vary by provider and premium. This has resulted in substantial tension between Fonasa and In response to this situation, the 2005 reform (Universal Isapres and raised concerns that differentials will lead Access with Explicit Guarantees, acronym AUGE in to an “arms race” in which political pressure to increase Spanish) defined an explicit benefit package for all, the AUGE-guaranteed package will undermine the fiscal whether they were enrolled in Fonasa or Isapres. This sustainability of the program, resulting in rationing/ reform was introduced in response to the fact that queues and reversion to the old system. benefits were undefined/implicit and served to put in place a coverage floor for all SHI beneficiaries, whether Source: Bitran 2013; Missoni 2010. HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better OOP Spending for Health Despite significant reforms in recent years to its health In 2015, about 40 percent of the population had financing system, OOP remains the predominant source no SHI coverage. Household survey data indicate of financing for health in Indonesia. In terms of trends, annual OOP health spending to be a 2.1 percent over the period 1995-2014 and, despite rising SHI share of total household consumption expenditure coverage, the OOP share of total health spending has for all households. Some of the persistence in OOP remained substantially unchanged, while OOP spending spending for health can be explained by the fact that per capita has risen in real terms (Figure 5.10).48 There are 43 percent of households reported having no form of three prominent reasons for the continued dominance SHI coverage. About 30 percent of all OOP spending of OOP spending as a source of health financing in reported came from these households. Almost 70 Indonesia: (i) low levels of public health spending on percent of households that reported no SHI coverage health; (ii) incomplete breadth of coverage under JKN; were headed by individuals working in the informal and (iii) poor supply-side readiness; and (iv) a preference sector, 50 percent of whom were in agriculture. for branded pharmaceuticals (which are not included in 74 the JKN package). The data also indicate, however, that 70 percent of all OOP spending was incurred by the 57 percent of Increases in government budgetary health spending all households that reported having some insurance and SHI expenditure in recent years have been coverage. This is despite the fact that JKN does not matched by increases in OOP spending in Indonesia. have any copayment or balanced billing stipulations. Even though publicly financed prepaid/pooled A recent study found that respondents with insurance health expenditures have risen, Indonesia has barely coverage reported higher rates of OOP spending due made any progress in its “health financing transition”– to the unavailability of medicines at health facilities.50 where countries experience both an increase in As a share of total household expenditure, OOP their total health spending per capita as well as in spending was higher among those with coverage the share that is prepaid/pooled as their economies than those without (Table 5-9). The latter is likely, at grow and develop (Savedoff et al. 2012).49 The share least in part, due to higher utilization rates among of OOP in THE remained roughly stable because the those insured. growth in coverage as well as prepaid and pooled public financing for health was accompanied by an Although OOP health spending is generally almost identical increase in OOP health spending regressive, this is not the case for Indonesia. Most of per capita. By way of contrast, countries such as the OOP spending is incurred by the rich in Indonesia, Thailand, China, Vietnam, and Brazil have made faster and the rich paid a higher share of total consumption progress in their health financing transitions. Thailand, expenditure as OOP health spending, likely the in particular, has been a clear outlier in terms of the result of access to, and utilization of, private care in speed with which it has realized its health financing urban areas. While the poor and near-poor are 40 transition (Figure 5.11). percent of the population, their share of total OOP 48 The methodology used for national health accounts was different prior to 2005 49 Implicit prepayment and pooling underlies government budgetary expenditures, and social health expenditures are explicitly prepaid for and pooled. 50 Financial sustainability and effectiveness of JKN program coverage: First year assessment” study managed by DJSN, conducted by CHAMPS, UI and funded by GIZ SPP in 2015 section five. HEALTH FINANCING Figure 5.10 SHI Coverage and OOP Share of Total Health Spending (1995-2014) OOP 800,000 60 Percentage spending (%) per capita OOP spending share of THE (right axis) (IDR) 600,000 40 400,000 SHI coverage (right axis) 20 200,000 OOP spending per capita (left axis) 0 0 1995 2000 2005 2010 2014 Source World Development Indicators database and SUSENAS (various years). Note OOP spending is in 2014 constant IDR. 75 Figure 5.11 Health Financing Transition (1995-2014) Annual change 20 in OOP/capita 45 degree line health spending 15 (%) REGRESSING China Indonesia 10 Philippines Sri Lanka Vietnam Malaysia India Lao PDR Cambodia 5 Brazil SLOW TRANSITION 0 Thailand RAPID TRANSITION -5 -10 -5 0 5 10 15 20 Annual change in pooled/capita health spending (%) Source World Development Indicators database Note Data are for countries with OOP share of 20% in 1995. HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better expenditures was only 11 percent; the top 20 percent health expenditure per visit. In addition, households of the population, on the other hand, accounted for reported outpatient utilization rates in the previous 53 percent of the total OOP spending in the country month at the time of the survey, whereas inpatient (Figure 5.12). (Figure 5.13 shows the “Pen’s Parade”51 utilization rates were reported over the previous graph for OOP health spending: the x-axis represents year. Given these data constraints, a more general households ranked in terms of consumption per relationship can be inferred between OOP health capita; the y-axis represents consumption before and spending at the household level over the past year after health spending. As can be seen, most of the and aggregate all-member outpatient and inpatient impoverishing effects of health spending occur right utilization numbers. above the poverty line among the near-poor. OOP health expenditure per capita in households that As might be expected, OOP health expenditures are reported no outpatient or inpatient visits was about related to the extent of household-level outpatient IDR 103,339 (~US$8; 0.1 percent of total consumption and inpatient utilization rates, especially the latter. expenditure). This increased to IDR 209,576 (~US$16; SUSENAS data collect information on utilization 1.6 percent of total consumption expenditure) for rates at the individual level. OOP health expenditures those that reported at least one outpatient visit are, however, reported only at the household level, but no inpatient visits. Those households reporting making it difficult to make a direct link between OOP no outpatient visits but at least one inpatient visit Table 5.9 OOP Share of Total Consumption Expenditure (2015) OOP health as share of total 76 Outpatient utilization (%) Inpatient utilization (%) Economic Coverage expenditure (%) status (%) With Without With Without With Without All All All coverage coverage coverage coverage coverage coverage Bottom 40% 56 17.2 14.3 16.0 3.2 1.8 2.6 1.6 1.4 1.5 Middle 40% 54 18.3 16.7 17.6 4.7 2.8 3.9 2.3 1.9 2.1 Top 20% 65 18.3 17.9 18.2 6.3 4.4 5.7 3.2 2.7 3.0 All 57 17.8 15.7 16.9 4.3 2.5 3.6 2.3 1.8 2.1 Source SUSENAS 2015. Table 5.10 OOP Spending Share of Consumption Expenditure for Those With at Least One Inpatient Visit in Past Year OOP health as share of total expenditure (%) Economic status With coverage Without coverage All Bottom 40% 6.5 7.7 6.9 Middle 40% 10.5 11.5 10.8 Top 20% 14.9 17.9 15.6 All 10.7 12.1 11.1 Source SUSENAS 2015. 51 Pen’s Parade is a plot of ordered values versus their ranks; it is useful as a means of showing how incomes, and income distribution change over time. section five. HEALTH FINANCING Figure 5.12 OOP Spending on Health by Economic Decile (2015) percentage 50 (%) 40 30 Decile share of total OOP health spending 20 OOP health spending share of total consumption 10 0 Poorest 3rd 5th 7th Highest Economic deciles Source SUSENAS 2015 Figure 5.13 Pen’s Parade (2015) 77 Household Comsumption per capita 25,000 10,000 5,000 Poverty line 2,500 Consumption post OOP health spending 1,000 Note Poverty line based on Statistics Indonesia (BPS-March 2015 period) HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better incurred OOP health expenditures per capita of The OOP share of THE should generally be below about IDR 1,414,947 (~US$109; 8 percent of total 20-30 percent, in line with levels observed in most consumption expenditure). Not surprisingly, the high-income OECD countries. Higher levels of highest OOP health spending per capita levels were the OOP share are strongly correlated with higher among those households that reported at least incidences of catastrophic health expenditures one outpatient and at least one inpatient visits: IDR and of resulting impoverishment rates in the 1,621,736 (~US$124; 9 percent of total consumption population (OECD 2011). Globally, an increase in expenditure) (Figure 5.14). the public spending on health share of GDP–either general revenue-financed and/or financed by Financial protection from SHI coverage is evident at expansion in SHI–tends to be associated with a all levels, including among the poorest 40 percent decline in the OOP share of THE. It remains to who utilized inpatient services. OOP health spending be seen if recent increases in public spending in as a share of total consumption expenditure is Indonesia and increases in JKN coverage will be slightly higher among uninsured households that had associated with a decline in the OOP share of THE inpatient utilization in the previous year. in future (Figure 5.15). 78 section five. HEALTH FINANCING Figure 5.14 OOP Health Expenditure by Utilization Pattern (2015) OOP Health Expenditure As Share of Total Consumption 1,500,000 10 OOP health as OOP health share of total expenditure consumption (%) per capita (Rupiah) 8 1,000,000 6 4 500,000 2 0 0 00 10 01 11 00 10 01 11 Utilization pattern Utilization pattern 79 Source SUSENAS 2015. Note Utilization pattern: 00=0 outpatient and 0 inpatient visits; 10=1 or more outpatient and 0 inpatient visits 01=0 outpatient and 1 or more inpatient visits; 11=1 or more outpatient and 1 or more inpatient visits Figure 5.15 OOP versus Public Spending on Health (2015) OOP share of 80 total health Cambodia spending (%) India 60 Philippines Indonesia Rusia Sri Lanka 40 Vietnam Lao PDR Malaysia China Brazil 20 Thailand South Africa 0 0 3 6 9 12 15 Government health spending share of GDP (%) Source World Development Indicators database HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better External Financing for Health Over the past five years, only about 1 percent of THE Even though external financing is a relatively small has come from external sources in Indonesia (Figure share of total health spending, donors provide a 5.16). This proportion–following an increase in the significant share of resources for key priority areas post-crisis period of 1997-2000–has generally been such as TB. For communicable disease programs, the declining over time in the past decade or so (Figure external share of the total program spending can be as 5.17). Indonesia’s low and declining dependence on high as 60 percent for TB; it is lower for immunization external sources is not surprising given its economic programs at around 10-15 percent. While the country status as a country on the verge of transitioning is eligible for funding from the Global Fund for the to upper-middle-income status. OECD-CRS data next funding cycle until 2020, Indonesia is slated to on external financing for health indicate average “graduate” from Gavi financing in 2016: this will imply a annual health-related disbursements of only about loss in financial resources for immunization as well as US$270 million over the period of 2011-13 going to of relevant technical assistance.52 Indonesia. Australia, USA, Gavi, and the Global Fund 80 are some of the biggest donors to the health sector, In contrast to relatively good performance in the with disbursements from the Global Fund and Gavi absorption of government budget, the absorptive accounting for about 29 percent and 6 percent, capacity for external financing of the public respectively, of all external financing for health. sector is low. According to the BAPPENAS report Figure 5.16 External Share of THE (2014) Share of 100 total health expenditure Solomon Islands (%) 50 Lao PDR Papua New Gguinea 20 Cambodia Ghana 10 Nigeria 5 Sri Lanka South Africa 2 Vietnam Philippines 1 Indonesia India China UPPER LOWER Malaysia MIDDLE MIDDLEBrazil LOW INCOME INCOME Thailand INCOME Rusia HIGH INCOME 250 500 1000 2500 10,000 35,000 100,000 GNI per capita, US$ Source World Development Indicators database 52 In January 2011, Gavi established a country eligibility threshold of US$1,500 GNI per capita. The first year of ileligibility is considered a grace year and no change is made in the cofinancing requirement. Once a country enters the graduation process, its cofinancing requirement increases rapidly to reach 100 percent by the fifth year, where countries will fully finance their vaccines. section five. HEALTH FINANCING Figure 5.17 External Share of THE (1995-2014) Share of total health expenditure (%) 15 10 5 0 1995 2000 2005 2010 2014 Source World Development Indicators database 81 Figure 5.18 Development Assistance for Health in Indonesia (2002-14) Millions of 2013 US$ 350 300 World Health Organisation [WHO] WFP UNICEF UNFPA 250 UNDP UNAIDS International Development Association [IDA] 200 Global Fund Global Alliance for Vaccines and Immunization [GAVI] Food and Agriculture Organisation [FAO] EU Institutions Korea 150 AsDB Special Funds Japan United States Italy United Kingdom Ireland 100 Switzerland Germany Sweden France Spain Finland Poland Denmark 50 Norway Canada New Zealand Belgium Netherlands Austria 0 Luxembourg Australia 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 Source OECD-CRS Data extracted on 21 Jan 2016 Data include ODA for population & reproductive health HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better on Loans and/or Grants Performance for the HIV, US$85 million for TB, US$43 million for malaria third trimester of 2015, the actual disbursement and US$8 million for Health Systems Strengthening). for lending programs at the end of September Gavi has disbursed US$121 million since 2002 and 2015 was only around 37 percent of the annual US$44 million was spent on the introduction of disbursement target (BAPPENAS 2015). The other pentavalent vaccine during 2013-15. measurement that is often used to describe slow disbursement is the amount of commitment fee National expenditures on HIV and AIDS have been the borrower has to pay for undisbursed funds, rising steadily, led by an accelerating rate of domestic which increased in 2014 compared to 2013. Several expenditure. HIV-related expenditures financed factors have been identified as the major drivers for by external sources have declined steadily as a slow disbursement, which include overoptimistic percentage of total expenditures from 2009-14 (Figure planning, unmet readiness criteria at the time of 5-19). This trend will have to continue and accelerate. project implementation, and slow procurement The total HIV program budget in 2014 has increased 8 processes. For grants, 22 grants for MoH were percent from the previous year, and overall there has recorded in the report with total value of US$121.9 been an average 12 percent annual increase over the million, and only US$2.4 million disbursed (by the period 2009 to 2014. The proportion of domestic and end of the third trimester). international funding has switched from predominantly external to domestic, from 35 percent in 2009 to 64 The Global Fund is the largest external donor for percent in 2014. The increase of domestic funding was health in Indonesia, followed by the Australian mainly driven by an increase in central government government and the US government. Between expenses, especially for ARV, while subnational and 2002 and 2015, the Global Fund committed US$729 private contributions continue to be small. million, of which US$617 million has been disbursed 82 to date. The latest Global Fund grant was signed The share of national (a combination of central and for the amount of US$218 million (US$82 million for subnational government) spending, for the National Figure 5.19 Annual HIV Expenditures by Source of Funds (2009-14) HIV Spending 2009 - 2014 External Funding for HIV Programs in US$ millions, current value 2009 - 2014 60 60 70 % external share to Domestic International Total % total HIV 60 spending 50 50 International 50 40 40 40 30 30 30 20 20 20 10 10 10 0 0 0 2009 2010 2011 2012 2013 2014 2009 2010 2011 2012 2013 2014 Source NASA multiple years. section five. HEALTH FINANCING TB Program has significantly increased between regulatory agencies, and technical capacity for 2009 and 2014. The share of external financing vaccine planning and advocacy (Saxenian et al. 2014). continued to be significant, but the proportion of national budget compared with external sources has Integrating vertical programs such as previously changed, from a ratio of approximately 65:35 in 2009, externally funded and vertically managed HIV, TB, to close to 50:50 in 2014. The data are, at best, patchy malaria and immunization programs into JKN will and incomplete as the contribution at the subnational entail more than addressing actuarial matters related level is underrecorded, and the contribution of other to which services should be included, but will also development partners and the private sector were have to take into account all the health system pillars. not well recorded. The main driver for the increase This includes: (i) preparedness to provide included was that central government started to fully finance services; (ii) being more responsive and sensitive to the provision of first-line antituberculosis drugs, the needs of specific target population groups; and reagents, and laboratory supplies and consumables. (iii) provider-payment mechanisms that incentivize The Global Fund continues to be the main source providers to reach out to target beneficiaries and retain of external finance and has committed US$61.3 them in the treatment cascade. Other issues that will million for 2016-17, followed by USAID for around have to be addressed include: (i) the existing clause on US$10.5 million per annum for the period of 2015- excluding services for self-inflicted medical conditions; 17. Unfortunately, no information is available for the (ii) different interpretations of the benefit package National Malaria Program. at different service delivery points and JKN branch offices; and (iii) fragmented coverage that discourages From an external financing perspective, one of use of certain services (for instance, diagnostic tests the key challenges facing Indonesia is that of and pre ARV services in HIV program). strengthening its health and financing systems to accelerate and sustain progress towards outputs Transition will be a challenge for all program areas 83 such as HIV, TB and immunization–key WHO-WB that are largely donor dependent, but for different recommended tracer indicators of UHC–while reasons. The extent of financial transition required is effectively managing the transition from external the least challenging for the immunization program, financing. This implies ensuring not just adequacy given that 85-90 percent of program costs are in terms of levels of domestic-sourced replacement currently being financed via domestic sources, financing for these programs, but also of governance, but the transition must occur more quickly as service delivery, and complementarity or even Indonesia will graduate from Gavi in 2016. As noted integration with JKN. earlier, the transition will entail not only replacing Gavi funds, but securing the additional resources In addition to financing, donors also provide needed to be able to push through with upgrading substantial technical assistance to the program that the vaccines being used in the national program as needs to be incorporated into the health system per the current strategic plan. Financial transition functions for programmatic sustainability. Global may not, however, be the biggest challenge for any organizations such as Gavi also supported countries of the four programs. Instead, getting more than with technical assistance, including immunization 500 local governments to implement the policies planning, surveillance, communication, Effective and programs the way that they were designed Vaccine Management (EVM) and National Regulatory at central level in Jakarta may be the biggest Authority (NRA) development. This implies that these challenge. The other programs have larger financial gains from technical assistance need to be sustained transitions to make, but a longer period of time in through local advocacy efforts and strengthened which to accomplish the transition insofar as they national technical skills. A recent systematic will be eligible for at least one additional round of assessment on 14 graduating countries suggested the Global Fund grants. that, in addition to financial sustainability, a number of challenges could impact the ability of countries to successfully graduate from Gavi support. These include planning and budgeting for vaccine, national procurement practices, performance of national HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better Efficiency As Indonesia’s health system develops, the key is In order to further assess efficiency of Indonesia’s for it to ensure that health expenditures lead to the health system, the research team used Data maximum possible increases in health-adjusted Envelopment Analysis (DEA) to derive estimates life expectancy rates. This should be heading in (called scores) of relative technical efficiency in a trajectory towards China, Thailand, Sri Lanka, transforming inputs into outputs. Healthy Life and Vietnam–rather than following a less-efficient Expectancy at Birth was used as an output indicator expansion route (heading on trajectories towards as it conversely represents narrow indicators more Brazil, Russia, and South Africa) (Figure 5.20). directly linked to health institutions and policies, thereby being potentially more relevant for policy Table 5.11 shows several countries in the period of recommendation,53 and THE per capita (constant 2013-15 that spent less on health care than Indonesia US$) was defined as an input. but attained higher DPT3 coverage rates and had lower MMRs. Clearly, it does not necessarily show Results derived from the DEA analysis suggest 84 that Uzbekistan’s health system is more efficient than that Indonesia can further improve the efficiency Indonesia’s. It could suggest, however, that there of its health system. Although there has been an might be macrolevel technical and/or allocative increase in total health spending per capita, it seems efficiency-related problems in Indonesia that are that there is little improvement in Healthy Life manifest in its relatively poor performance on key Expectancy at Birth (Figure 5.21), and the technical indicators such as DPT3 immunization rates and MMR efficiency score has declined from 0.30 in 2000 to in light of resources expended. 0.16 in 2013. Table 5.11 Countries With Lower Health Spending, Higher DPT3, and Lower MMRs than Indonesia, (2013-15) Country THE per capita MMR DPT3 (%) Indonesia $126 126 78 Uzbekistan $124 36 99 Solomon Islands $102 114 88 Kyrgyz Republic $82 76 96 Tajikistan $76 32 97 53 Efficiency estimates of health care systems, Economic Papers 549 | June 2015, European Union-2015. section five. HEALTH FINANCING Figure 5.20 Health-adjusted Life Expectancy vs THE (2013) Health-adjustment life expectancy 80 70 China Vietnam Sri Lanka Thailand Brazil Malaysia Cambodia Indonesia Rusia 60 Philippines Lao PDR India South Africa 50 40 25 50 100 250 1,000 2,500 7,500 Total health expenditure per capita, US$ Source World Development Indicators database Note x axis in log scale 85 Figure 5.21 Health-adjusted Life Expectancy vs THE (2000 and 2013) HALE Country ranked HALE Country ranked (0=lowest ; 200=highest) 2000 (0=lowest ; 200=highest) 2013 200 200 150 150 China Vietnam China Vietnam Sri Lanka Thailand Brazil Malaysia 100 100 Malaysia Sri Lanka Thailand Brazil Philippines Indonesia Rusia Indonesia Cambodia Rusia Philippines India India 50 50 Lao PDR South Africa Lao PDR Cambodia South Africa 0 0 0 50 100 150 200 0 50 100 150 200 THE Percapita Country Ranked THE Percapita Country Ranked (0=lowest ; 200=highest) (0=lowest ; 200=highest) Source World Development Indicators database Note y=axis: HALE-Healthy life Expectancy at Births (years) x axis: THE Percapita in 2013 Constant US$ Red line = DEA line; Green lline = Distance to DEA line HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better Despite a rise in the bed-density ratio in recent populated provinces. Occupancy rates in both years, this number remains far below WHO’s norm/ public and private facilities are low at 55-65 percent, recommendation of 2.5 per 1,000 and Indonesia’s approximately 25 percent lower than occupancy rates numbers remain far below that of comparator in other countries in the region. The average length countries in the region including Thailand, Malaysia, of stay has been trending upwards and is about six Sri Lanka, China, and Vietnam. Key issues are the lack days. The density of health centers in Indonesia of of systematic information on the number of hospital one puskesmas per 26,000 inhabitants is aligned with beds in private clinics and the maldistribution of beds other low- and middle-income countries. A health across the country. There is a four-fold difference in center in Nigeria covers 20,000 inhabitants (Ujoh 2014); the bed-density ratio across the country: from a high in Maharashtra, India a health center covers 30,000 of 3.2 beds per 1,000 in DI Yogyakarta to a low of 0.8 inhabitants and in Liberia, a health center covers a per 1,000 in West Java. Two additional provinces– population of between 25,000 and 40,000 inhabitants North Sulawesi and West Papua–exceed the WHO (Ministry of Health and Social Affairs 2008). norm of 2.5. In addition to the overall health system performance in Thirteen provinces had a bed-density ratio below the Indonesia, the country also faces efficiency challenges Indonesian average. Higher bed-density scores were in areas that are identified by the WHO as common evident in both moderate to large size and sparsely sources of health system inefficiency (Table 5.12). 86 Box 5.3 Measurement of health system efficiency is complex. Health System Efficiency Broader macrolevel analyses of efficiency of health systems often use countries or subnational administrative units as DMUs. In such cases, outputs are often specified Efficiency, broadly defined for any generic production in terms of population health indicators such as the system, typically implies getting the most out of limited MMR, health-adjusted life expectancy, or as a set of resources. Two components of efficiency are generally intermediate outputs such as immunization rates and differentiated: technical efficiency implies attaining the other health service coverage rates. The latter are, most output from a given set of inputs; and allocative arguably, a more direct measure of the output of a health efficiency implies choosing the optimal set of inputs, system. Broader population health indicators such as given their prices, to maximize output and minimize cost. the MMR is often more of a function of additional factors Subsumed under technical and allocative efficiencies (for example, female education, infrastructure, water are possible efficiencies related to scale and scope in and sanitation, and so forth) that are generally outside the health system. Taken together, inefficiencies can be the purview of health systems. Microlevel analyses of a result of waste (technical inefficiency) and/or due to efficiency usually look at case mix-adjusted unit costs in a suboptimal choice of inputs (allocative inefficiency). hospitals and health centers as DMUs, with outputs and In this regard, technical efficiency is often referred to as input indicators reflecting the functions of the specified “doing things right” and allocative efficiency as “doing the DMU. Hospital-level efficiency analyses often look at right things”. Measuring efficiency requires defining the benchmark comparisons of bed occupancy and turnover appropriate decision-making unit (DMU) so as to specify rates. appropriate outputs and inputs. Source: Hollingsworth and Peacock (2008). section five. HEALTH FINANCING Table. 5.12 Ten Major Sources of Inefficiency in Health Systems Worldwide EFFICIENCY CHALLENGE RELEVANCE TO INDONESIA Under use of generic Regulations require government and JKN-affiliated health facilities to use generics. Although drugs and higher-than- poor supply-side readiness and preference for branded medicines not covered by JKN has led to necessary prices for high OOP spending, prices of branded and patented medicines in Indonesia are higher compared medicine. with international reference prices. Local production (most of which are generics) dominates the Indonesian pharmaceutical market. Use of substandard and • Much of the financial burden (and health hazards) of substandard and counterfeit medicines is counterfeit medicine. believed to be borne by consumers, however, little data exists documenting this.54 Counterfeit vaccines, including vaccines for routine childhood immunization, have been found being sold in private hospitals and facilities, leading to public health concerns over the government and BPOM’s ability to effectively regulate vaccines and medicines in the country. Commonly counterfeited medicines include antibiotics, antimalarials, painkillers, anesthesia, vaccines and erectile dysfunction medicine.55 • In 2016, a nationwide substandard and counterfeit medicines operation (Pangea IX) seized US$4.2 million worth of substandard and counterfeit medicines across 32 provinces in the country.56 Inappropriate and • Inappropriate, ineffective use and self medication of prescription medicines, especially ineffective use of antibiotics, remains widespread in public and private health facilities, and pharmacies, burdening medicine. both the government budget and OOP spending.57 • Under JKN, there is evidence of lower number of drug per prescription and higher number of generic drugs prescribed compared with those uninsured; but at the same time an increase in the prescription of nonformulary drugs. Overuse and oversupply of Under JKN scheme, the increase of Cesarean Section has been observed in its two-year equipment, investigations, implementation. Of 1.5 million delivery claims, more than one-half (54 percent) were by and procedures. Cesarean Section.58 Although no baseline figure is available, this is much higher than the WHO’s recommended upper limit which is 15%.59 Inappropriate or costly Key issues in HRH incude maldistribution, a shortage of specialists, and poor skills of health workers. staff mix, unmotivated Stark inequalities in the distribution of HRH between geographical regions and provinces, and between 87 health workers. urban and rural areas have become one of the contributing factors to variable health outcomes. Inappropriate hospital For JKN members, readmission is around 10% for hospital inpatient and around 40% of those are admissions and length of questionable; there is also high readmission for outpatient services. Length of stay has increased stay. from four days in 2009 to six days in 2015. Inappropriate hospital size Despite rapid growth of hospital numbers over the past decade, the total hospital bed to population (low use of infrastructure). ratio remains low (ranging from 1.07/1,000 compared with a global average of 2.5/1,000). Medical errors and A study on maternal deaths verbal autopsy found that almost 40% of the decision to refer was made too suboptimal quality. late and, in more than one-half of the cases, clinical decision making was conducted inappopriately.60 Waste, corruption, and Potential fraud practices in JKN claims, including upcoding, unbundling, prescribing drugs outside fraud. of catalogue, and false claims are exacerbated by lack of supervision.61 Insufficient mix of health The implementation of JKN has raised concerns that the system is focusing more, by spending interventions (for example, more, on curative rather than preventive practices. In the absence of Certification of Need (CON) between prevention and requirements for providing advanced technology medical equipment and more expensive services treatment, high vs low (for example, diagnostic equipment), has led to supply-induced demand which in the end drove value). the medical cost up. Source WHO 2010c (left column)62 and World Bank staff (right column). 54 This is implied from the fact that most types of counterfeit medicines include OTC and self-prescribed medicines, and operation Pangea IX that includes medicines seized from social media and the internet. 55 Badan POM. 2016 56 Jakarta Globe, 2016. 57 Sources: 1. Widawati et al 2011 2. Hadi et al 2010 3. Puspita sari et al 2011 4. Hadi et al 2008 58 Hidayat 2016 59 Gibbons et al. 2010. 60 JHPIEGO Study 2016. 61 Pradiptyo 2015. 62 Chisholm 2010 HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better 88 section six. A CASE STUDY ON IMMUNIZATION section 6 . 89 A CASE STUDY ON IMMUNIZATION HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better In Summary 1. Financial and technical support from external resources is important for the immunization program in Indonesia, however, Indonesia needs to be ready for transitioning from external financing. 2. Indonesia has made progress in increasing coverage rates, although large inequalities across regions and economic status still exist. 3. Vaccine availability in puskesmas was generally good, but several issues remain, such as human resources (availability and expertise) and cold-chain management. 4. At the household level, physical and time barriers pose a challenge to accessing immunization services. 5. The total cost of immunization in Indonesia must also take into account all nonoperational immunization activities. 6. JKN’s benefit package covers routine immunization, but there are concerns about the usage of capitation payments to improve immunization service delivery. 7. New vaccines are currently being planned for introduction and several key factors, such as scientific evidence, financing, production and political support, must be addressed. 90 section six. A CASE STUDY ON IMMUNIZATION This section will apply the HFSA approach to examine Indonesia’s current immunization schedule covers all of the disease-specific context, funding outlook, and WHO’s recommendations for traditional vaccines, except essential program functions needed to ensure the for the rubella vaccine. The schedule initially included financial and institutional sustainability of externally BCG and smallpox, and was later expanded to include financed programs within the broader health system. DPT. The schedule has undergone several changes The immunization program was chosen as an example, since, first with the addition of polio and measles into the particularly given the context of Indonesia’s graduation program and then, in 1997, Hepatitis B. More recently, from Gavi by the end of 2016. It will focus on assessing in 2013, the pentavalent (DPT-HB-Hib) vaccine was the current system and identifying bottlenecks and introduced and is set to replace the DPT-HB combination future challenges for transition of a program. This in three staggered phases. Full roll-out of the pentavalent would inform the transition planning by formulating the vaccine in all 33 provinces64 was completed in early 2015. necessary activities in a posttransition environment and Indonesia is planning to launch a combination measles/ assessing the government’s capacity to conduct these rubella vaccine in 2017 as a supplementary immunization activities in the absence of donor support. activity and in 2018 this should be part of the routine immunization package. The schedule is revised annually Indonesia’s population ages 0-1 year–the primary target based on recommendations made by an independent group for immunization–was 4.3 million in 2013. This advisory body, the National Immunization Technical number has been declining ever since it peaked at Advisory Group. around 4.7 million in 1998. It is projected to decline to 3.9 million by 2030.63 As a consequence, the share of Indonesia has not yet adopted WHO’s new vaccine the population ages 0-1 year has declined steadily from recommendations but is in the process of doing so. around 2.4 percent in 1995 to 1.6 percent in 2013 (Figure This is similar to the situation in most comparable 6.1). This share is projected to continue to decline to countries (Table 6.1). Indonesia has recently introduced only about 1.3 percent by 2030. This implies that the IPV (Inactivated Polio Vaccine) into its routine 91 financing requirement for the immunization program, immunization program in accordance with World Health given no change in the basic immunization package, Assembly resolutions related to polio eradication. will decline. The government plans, however, to Other new vaccines such as Japanese Encephalitis introduce several new vaccines in the future, which has (JE), pneumococcal, HPV and rotavirus vaccines are implications for an increased resources requirement. scheduled for introduction by 2019 (MoH 2013). Figure 6.1 Population Ages 0-1 Year (1995-2030) millions 4.8 2.5 percentage (%) 2.3 4.6 0-1 share of total population (right axis) 2.0 4.4 1.8 4.2 1.5 4 population 1.3 Population projection 3.8 1.0 1995 2000 2005 2010 2015 2020 2025 2030 Source www.census.gov 63 Indonesia age-specific population growth over the years; www.census.gov. 64 The 34th province (North Kalimantan) was established in 2012. HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better The burden of disease (BOD) for vaccine-preventable rotavirus vaccination. Evidence in Indonesia points diseases (VPD) for children under five remains to rotavirus as being responsible for at least 60 high for diseases not yet covered by the routine percent of hospitalization in children due to diarrhea immunization programs in Indonesia. For example, in 2006 (Soenarto et al. 2009). At the same time, the Table 6-2 indicates that the BOD of diarrheal disease persistently high BOD of other VPDs already covered has been continuously high since the 1990s and by the routine immunization program indicates remains one of the most pressing VPDs that could the need for routine immunization services to be be easily addressed in Indonesia with a simple continuously supported, improved, and sustained. Table 6.1 National Immunization Schedules (2015) WHO Recommended Routine Vaccines Other Traditional vaccines New vaccines Vitamin BCG HepB Polio DTP Hib Measles Rubella Rotavirus Pneumococcal HPV HepA Influenza JapEnc Meningococcal A Brazil * * ** * * * * * * * * * * Cambodia * * * * * * * * China * * * * * * * * * India * * * * * * * Indonesia * * * * * * * Lao PDR * * * * * * * * * * * * 92 Malaysia * * * * * * * * * * * Philippines * * ** * * * * * * * * * Russia * * ** * * * * * * * * South Africa * * * * * * * * * * * Sri Lanka * * * * * * * * * Thailand * * * * * * * * * * Vietnam * * * * * * * * Source WHO. Note Two stars under polio column indicates IPV has been introduced. Table 6.2 Burden of Disease for VPDs (2013) DALYs lost share in Under 5s (%) Rank in 2013 VPDs in Under 5s in 2013 1990 2000 2010 2013 1 Diarrheal diseases 10.51 6.70 7.28 6.40 2 Measles 6.00 3.71 2.39 4.06 3 Pertussis 1.40 1.70 1.93 2.13 4 Haemophilus influenza type b meningitis 1.72 1.49 1.37 1.17 5 Pneumococcal meningitis 1.13 0.84 0.87 0.75 6 Encephalitis 0.56 0.61 0.69 0.69 7 Tetanus 3.65 1.00 0.40 0.31 8 Meningococcal meningitis 0.31 0.31 0.23 0.20 9 Tuberculosis 0.31 0.19 0.14 0.12 10 Diphtheria 0.03 0.03 0.04 0.03 DALYs lost per 100,000 population 161,487 101,400 69,494 58,618 Source Institute of Health Metrics and Evaluation database (2015). section six. A CASE STUDY ON IMMUNIZATION Outcomes and Determinants Immunization coverage for Indonesia has increased Despite increases in coverage rates in recent over the years but the dropout rate remains an issue. decades, Indonesia does not compare favorably to There are a variety of estimates of immunization its peers, and large inequalities across regions and coverage for Indonesia. The latest IDHS data indicate economic status exist. For example, Indonesia is that 66 percent of children 12-23 months were richer than Cambodia, the Philippines, and Vietnam, fully immunized in 2012.65 Over the period of 2012- but has significantly lower coverage rates for DPT3 14, and depending on source, estimates of BCG and measles immunization (Figure 6.2). There is as immunization rates ranged from 89 to 97 percent; much as a three-fold difference in immunization DPT ranges from 72 to 82 percent; polio immunization coverage rates across provinces in Indonesia. rates ranged from 74 to 83 percent; and measles DPT3 immunization rates, for example, are almost immunization rates were 80-89 percent (Table 6.3). 90 percent or more in Bali and DI Yogyakarta, With regard to specific vaccines, IDHS data show less than 50 percent in Maluku, Banten, and West that, although the coverage of the first doses of DPT Sulawesi, but only 35 percent in Papua (Figure 6.3). and polio vaccines are relatively high (88 percent Inequalities are large by economic status as well, 93 and 91 percent, respectively), only 72 percent and 76 and these have sustained over time (Figure 6.4). percent went on to receive the third dose of DPT and In 2013, only 39.5 percent of children from lowest polio respectively. The dropout rate between the first income quintile families received full immunization and third doses of vaccines is, therefore, 16 percent compared with 67.8 percent from the highest for DPT and 15 percent for polio. The main reason quintile. Of particular importance are the hard-to- for incomplete vaccination is failure to understand reach areas that require additional strategy, such as the need for return visits and for additional doses for Sustainable Outreach Service, to ensure adequate complete immunization. immunization coverage.   Table 6.3 Immunization Coverage Rates (Various Years) Source Vaccine IDHS Riskesdas SUSENAS WHO-UNICEF MoH 2012) (2013) (2012-2014) (2012-2014) (2012-2014) BCG 89% 88% 94% 97% 90% DPT3 72% 76% 73% 82% 77% Polio3 76% 77% 74% 83% 81% Measles 80% 82% 89% 82% 84% Fully immunized 66% 59% 68% – – 65 According to WHO guidelines, children are considered fully immunized when they have received one dose of BCG, three doses each of the DTP and polio vaccines, and one dose of the measles vaccine. HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better Data quality is a major issue, particularly in calculating between administrative data and surveys, and target populations for coverage estimates. There between surveys themselves. Coverage data at is a considerable variation in coverage estimates lower levels are unreliable due to poor reporting Figure 6.2 DPT3 and Measles Immunization Coverage vs Income (2014) percentage DPT percentage Measles 100 100 Sri Lanka Malaysia Sri Lanka Cambodia Thailand China Rusia Thailand China Brazil Vietnam Rusia Vietnam Cambodia Brazil Malaysia 90 90 Lao PDR Lao PDR Philippines India India 80 80 Philippines Indonesia Indonesia 70 70 South Africa South Africa LOWER UPPER LOWER UPPER 94 LOW INCOME MIDDLE INCOME MIDDLE INCOME HIGH INCOME LOW INCOME MIDDLE INCOME MIDDLE INCOME HIGH INCOME 60 60 250 500 1000 2500 10000 35000 100000 250 500 1000 2500 10000 35000 100000 GNI per capita, US$ GNI per capita, US$ Source World Development Indicators database Note both y- and x- axes logged Figure 6.3 DPT Immunization Rates by Province (2012) Percentage 100 (%) 80 Indonesia average 60 40 20 0 Riau Islands Bangka Belitung DI Yogyakarta Papua Maluku Central Kalimantan Riau West Java Central Java East Java Banten West Sulawesi Nanggroe Aceh West Papua South Sulawesi North Sumatra South Kalimantan North Maluku West Kalimantan West Sumatra Jambi South Sumatra West Nusa Tenggara Central Sulawesi Gorontalo Bengkulu Lampung Southeast Sulawesi DKI Jakarta North Sulawesi Bali East Nusa Tenggara East Kalimantan Source IDHS 2012 section six. A CASE STUDY ON IMMUNIZATION at village level (posyandu) mostly due to over- or urban areas, while all economic strata are more or underreporting by village midwives. Data Quality less equally likely to initiate vaccination, children from Self-assessment (DQS) reveals that the accuracy of the highest strata are far more likely to complete data for reporting from villages to districts level for vaccinations than those from lower economic three antigens (BCG, DPT/HB3, and measles) were quintiles. very low (below 20 percent). Household surveys lack standardization in survey methods. For administrative In addition to education, various factors such as birth data, there is a gap between real population data order, distance to health facility and ownership of reported by villages and target/projected population health insurance are among the factors that affect used by DHOs. Districts are obligated by MoH to immunization rate. Higher birth order also lowered use data from intercensal surveys, but this may not the odds of being vaccinated. Households who report account for recent migration patterns (rural-urban)/ that distance to the health facility is a serious issue in population mobility and differences in Crude Birth Rate getting medical help are both less likely to initiate and estimation between national and district level. Some less likely to complete vaccinations. Similarly, although areas may, therefore, be underestimating coverage vaccination is provided at little or no cost through the (if they have experienced outmigration), while others public system, children who have not received any are overestimating coverage (urban areas). UNICEF DPT vaccine were also more likely to be uninsured surveys in major urban areas highlight this effect by than children who received three doses of the vaccine. identifying significant numbers of unimmunized and underimmunized children in migrant communities and Recent analysis of data from rural areas suggests a among the urban poor. higher rate of immunization among those children whose mothers had antenatal care (Suparmi 2014). Women with more education were more likely to Level of development (as measured by the Human vaccinate their children, and unvaccinated children Development Index), health sector investment (as 95 are disproportionately concentrated in rural areas in measured by Public Health Development Index), and Indonesia. Although only one-half of the country’s health-worker density (as measured by numbers population is rural, approximately 63 percent of all of doctors) are also positively associated with unvaccinated children lived in these areas. Analyses immunization rates. Outbreaks of disease continue suggest that the determinants of vaccination are to occur, indicating significant subpopulations of sharply divided by urban-rural status. For example, unvaccinated and undervaccinated children among while fully immunized and fully unimmunized children geographically and socially isolated groups (for show the same wealth gradient in rural areas, the example, migrants, the urban poor, and people in same is not the case for children in urban areas. In rural/isolated areas). Figure 6.4 DPT3 Immunization by Economic Status DPT 100 coverage (%) 80 Top 20% Middle 40% 60 Bottom 40% 40 20 1995 2000 2005 2010 2015 Source IDHS (various years) HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better Service Delivery Most children are vaccinated at posyandu in costs for puskesmas and posyandu for immunization. Indonesia. Immunization services–as with several Unlike for most other health services, however, the other preventive health services–are offered at central government is responsible for procurement several points of contact with the health system of all vaccines and syringes. NIP also provides in Indonesia: these include integrated health technical assistance, guidelines, monitoring and service posts (posyandu), village maternity clinics/ evaluation, quality control, and training, as well as village health posts (polindes/poskesdes), health conducts supplementary immunization activities centers (puskesmas), schools, and at government such as campaigns. NIP uses a standardized tool for and/or private hospitals or clinics. Indonesia also assessing supply-side readiness for immunization sometimes conducts national or subnational at the subnational government level, annually immunization campaigns. A large majority (almost sampling across provinces and districts. BAPPENAS three-fourths) of all vaccinated children in coordinates the integration of immunization into Indonesia, however, receive their immunization at long-term (20 year), medium-term (five year), and 96 posyandu, followed by 10 percent at puskesmas, 10 annual plans. BAPPENAS also coordinates with MoH percent at private clinics and hospitals (although to set targets and to outline activities needed to be this can be as high as 50 percent in some financed to achieve them; MoH then prepares the provinces), and the remainder at polindes and other budget for submission to MoF. BAPPENAS has set a places (including midwives’ homes), emphasizing target of 95 percent of districts with greater than 80 the importance of posyandu kader and community percent vaccination rates as part of Indonesia’s next participation in immunization services.66 Routine five-year plan (RPJMN 2015-2019). immunization is also provided for school children via the School-based Immunization Month (Bulan Regulations require all government-procured Imunisasi Anak Sekolah, BIAS) program; the vaccines to be supplied by PT Bio Farma, a program targets children in the first through third state-owned enterprise.67 Regional health offices grades in order to boost immunity to measles and coordinate the provision and distribution of vaccines diphtheria, as well as to provide future maternal that are in the national immunization schedule. immunity against tetanus. The number of vaccines procured is based on the estimated number of entitled beneficiaries; thus, The central government is responsible for procuring in principle, Indonesia aims to provide universal vaccines and district governments are responsible immunization coverage. There are national for service delivery. MoH’s National Immunization guidelines governing immunization service delivery Program (NIP) that began in 1977 oversees the to ensure safe injection practices, counseling, and immunization program and performs forecasting waste disposal. and planning for vaccine procurement. As with other health services, district governments are responsible The regulatory authority that oversees PT Bio Farma for service delivery, including immunization is BPOM. BPOM has undergone assessments using equipment and supplies and providing operational the WHO NRA assessment tool and is considered 66 National Immunization Coverage Survey 2007. 67 PT Bio Farma is a WHO prequalified supplier that also exports to over 133 countries; 40 percent of Bio Farma products are used domestically, while 60 percent are exported. section six. A CASE STUDY ON IMMUNIZATION “fully functional” by WHO,68 all vaccine doses due to adequate measures taken to ensure service are therefore assumed to be of “assured quality”. continuation. Starting 2014, vaccine prices are listed This functional status must be sustained over on the e-catalog similar to all other procurement time through regular reassessments that require processes by the government. A fixed charge for significant resources and expertise on the regulatory transportation to provincial drug warehouses is agency’s part. While, to date, Bio Farma has been included in the procurement price. The transportation reliably supplying all vaccines used in the Expanded cost is fixed, which means there is no difference Program for Immunization (EPI) program, past between the transport cost for provinces with difficult experiences have shown that some delays in the access, such as Papua, and relatively well-developed introduction of vaccines were caused by delays in provinces with good road access, such as West Java. production by Bio Farma (Rubella, JE) (Hadisoemarto et al. 2016). Facility data indicate that the availability of vaccines at puskesmas was generally good, but immunization- The price of vaccines in Indonesia is generally related training of staff was not. According to the comparable to the prices paid by Gavi and UNICEF. 2011 NIHRD facility census, more than 90 percent of The exceptions are BCG and Hepatitis B single-dose all puskesmas reported availability of government- vaccines for which Indonesia pays twice (BCG) to mandated vaccines such as measles, DPT, polio, and four times (Hepatitis B single dose) the price paid BCG vaccines.69 Service readiness problems were by UNICEF. There was a switch in procurement notable in three provinces–Papua, West Papua, and roles in 2014, where the Directorate-General of Maluku–where less than 80 percent of puskesmas Pharmaceuticals and Medical Devices took over the reported availability of the measles, DPT, polio, responsibility of procuring vaccines, among other and BCG vaccines. These were also some of the program-specific medicines. Anecdotal evidence aforementioned three eastern provinces with relatively suggested some initial stock-out problems due to low immunization rates. A notable area of deficiency 97 the switch, although it did not affect service provision was with regard to staff training for immunization: Table 6.4 Vaccine Prices E-Catalog 2016 price per dose (US$) Gavi and UNICEF 2016 prices Vaccine Type Central prices Provincial prices per dose (US$) BCG 0.23 0.27 0.06-0.16 Diphtheria-tetanus 0.14 0.16 0.14 Tetanus 0.12 0.13 0.07-0.13 Tetanus-diphtheria 0.14 0.15 0.10-0.15 Measles 0.18-0.23 0.25 0.23-0.50 OPV 0.16 0.18 0.18-0.21* Hepatitis B 1.58 1.86 0.20-0.42 DPT HB-HiB 1.68 1.36 1.15-2.35** IPV 1.19 NA €0.75 - €2.4*** Source MoH Regulation No. 89/2014, UNICEF, and Gavi. Note *2015 prices. **No price for five doses available at UNICEF and prices are for different packaging size. Central prices include transportation to central warehouse; provincial prices include transportation to provincial warehouse. UNICEF prices are FCA nearest international airport Incoterms. ***Price for IPV in Euros. 68 BPOM (Indonesia National Agency of Drug and Food Control) was last assessed by WHO on vaccine regulation in June 2012. 69 Latest IFLS 2014 data findings show a slightly lower rate than what the NIHRD facility census found: the availability of mandated 6vaccines in the puskesmas is about 80 percent. HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better only 45 percent of puskesmas in the country had at cold-chain and waste disposal. Only 15 percent of least one staff member trained in EPI in the previous cold-chain equipment was reported to be equipped two years. Limitations exist in terms of the number with electronic continuous temperature monitoring and capacity of staff at national and provincial level to (2014 JRF). The MoH is currently piloting the use of conduct the planned intensive monitoring, technical the UNICEF Cold Chain Equipment Manager (CCEM) assistance, and follow-up action. High staff turnover tool in 22 provinces to identify cold--chain needs. at the district level (midwives, EPI managers, cold- chain technicians, and so forth)–often every three to Routine surveillance needs to be strengthened six months–inhibits continuity, training, development to improve its ability to timely detect and respond of expertise, and commitment to EPI activities. to outbreaks. Surveillance and reporting system Limited staffing standards, for example, requirements components consist of routine surveillance, coverage for health worker expertise in data management is monitoring, immunization safety, and adverse events another constraint. management. Routine surveillance is conducted through regular monthly reporting of VPDs and weekly In contrast to puskesmas, the availability of vaccines for Acute Flaccid Paralysis (AFP). AFP surveillance at private facilities was poor. The IFLS 2014 data identifies high-risk areas for wild polio virus and showed only about one-quarter of private facilities uses a mechanism to monitor polio eradication in reported availability of measles, DPT, polio, and BCG Indonesia. The indicator for coverage monitoring is vaccines (Table 6.5). the percentage gap between DPT3 survey coverage and officially reported figures. The central government Cold-chain management is functional but needs collects data from all provinces and districts. Declining further improvement to enhance quality. Cold-chain VPD surveillance performance, large discrepancies inventory assessments conducted in 2014 found between administrative coverage data, limited 98 that only 70 percent of cold-chain equipment were functioning of laboratory capacity and inadequate functional, 18 percent were working, and 12 percent AEFI surveillance are some of the issues faced by were working but needed attention. Main issues the surveillance system and requires adequate raised from EVM were temperature management, subnational financial support. stock management issues, and low levels of maintenance of equipment. The 2013 EPI review also As is the case with other health services, households highlighted these issues, reporting a large proportion face physical and time barriers to accessing of district-level cold-chain equipment due for immunization services. As noted earlier, children replacement, inadequate response to temperature living in rural areas and poor households are less deviations, vaccine wastage rates not monitored or likely to be fully immunized, despite the free used for planning, and inadequate budgeting for immunization services for all children provided by Table 6.5 Availability of Government-mandated Vaccines at Private Clinics70 Availability of vaccines (%) Facility survey Measles DPT Polio BCG IFLS 2007 23.4 24.8 25.5 22.6 IFLS East 2012 9.7 9.7 9.7 9.7 IFLS 2014 26.1 27.1 26.6 25.6 70 IFLS 2007 and 2014 were provincially representative surveys covering 13 provinces (representing 83 percent of Indonesia’s population): North Sumatra, West Sumatra, South Sumatra, Lampung, DKI Jakarta, West Java, Central Java, DI Yogyakarta, East Java, Bali, West Nusa Tenggara, South Kalimantan, and South Sulawesi. The survey included 952 puskesmas and 1,595 private clinics/practitioners; IFLS East 2012 is a provincially representative survey covering seven provinces (representing 7 percent of Indonesia’s population): East Nusa Tenggara, West Kalimantan, Southeast Sulawesi, Maluku, North Maluku, West Papua, and Papua. The survey included 98 puskesmas and 185 private clinics/practitioners. section six. A CASE STUDY ON IMMUNIZATION the government. The real costs of accessing health the responsibility of the Health Promotion Unit of the care are the transport and opportunity costs which MoH (PROMKES). Gavi and UNICEF have supported a are not covered by the government. A recent study variety of communication materials for use by health indicates that increasing the number of posyandu per workers. There remains, however, a reported lack of 1,000 population improves the probability of children communications strategy aimed at health workers, receiving complete immunization by 54 percent program managers, and policy makers. Most DHOs (Maharani and Tampubolon 2014). This signifies the periodically identify chronic low coverage or high importance of the policy to promote a more even dropout rate areas for “sweeping” activities that target distribution of posyandu to improve immunization children who did not show up when expected. This coverage. Better distribution of immunization providers is evidenced from the name-based infant registers would reduce the distance to health providers which, maintained by health workers/kaders. in turn, would shorten the time needed and lower the financial costs to access services. There is a small antivaccination movement in Indonesia, however, overall vaccine refusal is Poor knowledge regarding vaccination benefits, reported to be very low. Refusals stem from religious schedule frequency, and normal side effects lead objection (halal/haram) or pseudoscience beliefs to higher-than-expected DPT3 dropout rates. (false autism link). Evidence that some people refuse DPT dropouts may also be associated with DPT vaccines on religious grounds is limited. Additional reactogenicity and inadequate prevaccination study is required to evaluate this perspective further. counseling on expected adverse reactions by service Vaccine refusal is considered to be a sensitive providers. Kaders often do not have the training to subject, and people may not be willing to verbalize address concerns and objections to vaccination, their reason for refusal The MoH has responded nor do they routinely track defaulters in their to antivaccination concerns by engaging religious communities. Communication and demand creation is groups and utilizing social media. 99 HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better Financing and JKN Total expenditure on routine immunization from The national immunization program in Indonesia all sources of financing is estimated to be in the also receives technical support from UNICEF and range of US$155 million in 2014–less than 1 percent WHO. This is in the form of cold-chain evaluation of THE. Routine immunization expenditures have and upgrading; information, education, and increased significantly in recent years, up from communication material (IEC) development; ~US$57 million in 2010.71 Almost all expenditure on immunization advocacy; and management and routine immunization is generally assumed to be logistics training (EVM); as well as surveillance and public, amounting to about 2 percent of all public coverage survey. As noted earlier, even though expenditure on health in 2014. Roughly 60 percent of Indonesia’s dependence on external financing for this was central government and 40 percent was by health is generally low, the external financing share subnational governments. Some of the financing for of its immunization program is relatively significant immunization services comes from JKN.72 The exact and is crucial to the sustainability of immunization magnitude is not, however, easy to compute as there financing and quality service delivery. Programs 100 is no direct line to immunization services under JKN. traditionally supported by external donors, such Syringes and safety boxes are financed jointly by the as health-system strengthening activities, and central and subnational governments (80 percent civil society organization support are at risk of not central and 20 percent subnational cost-sharing). receiving government funding and oversight as Operational costs–including for the cold chain and Indonesia graduates from Gavi. others related to immunization delivery–are borne by subnational governments. The central government Gavi has disbursed ~US$121 million since 2002 is tasked with monitoring and evaluation of the when it became active in Indonesia in supporting immunization program, and provides additional on- the introduction of the Hepatitis B vaccine. Table demand financial support to districts. 6-6 shows type and amount of Gavi support under several windows: new and underused vaccine External financing plays a relatively important role support (NVS), vaccine introduction grants (VIG), and for immunization in Indonesia. In 2014, ~US$14.3 health system strengthening (HSS). In addition to million–9 percent of government expenditure financing the introduction of the Hepatitis B vaccines, on immunization–was financed by Gavi, up from the NVS window–which accounts for 51 percent ~US$0.2 million in 2011. The Gavi-financed share of of the total disbursements to date–is financing the government immunization expenditure increased introduction of the pentavalent vaccine over 2013- significantly in 2013 because Gavi is cofinancing the 16. The IPV vaccine is also being introduced in the introduction of the pentavalent vaccine until 2016, country via the same support mechanism; IPV is following which Indonesia is expected to graduate scheduled for a single-phase rollout in July 2016.73 from Gavi financing. Indonesia received support under the VIG window 71 Ministryof Health, Gavi Annual Progress Report (2010, 2014). 72 A specificexample is that part of the puskesmas capitation payment that can be used to improve immunization service delivery. 73 There are benefits to offering the oral polio vaccines (OPV) because administration does not need a trained medical staff member; and also because after vaccinating children, the virus continues to be shed and is picked up by other children. The downside, however, is that this form of the virus mutates. Before a country is fully vaccinated, OPV is generally recommended followed by transition to IPV (which is more expensive). section six. A CASE STUDY ON IMMUNIZATION in 2002, 2013, and 2015. Indonesia received HSS sources (for example, the WHO-UNICEF JRF). For grants from 2008-09 and 2012-15. HSS grants are evaluation purposes, Gavi also requires countries to aimed at strengthening health systems to improve demonstrate routine mechanisms to independently immunization outcomes and account for 21 percent assess and track changes in quality of administrative of total disbursements from Gavi to date. HSS in data. In addition, different types of support, such as Indonesia originally focused on maternal and child- HSS, have additional specific requirements. health activities and strengthening civil society organization involvement in immunization service There is no evidence of significant immunization- delivery. It faced significant delays in disbursement, related OOP expenditures. Systematic information however, until grant activities were refocused on on OOP expenditure on routine immunization is immunization outcomes in 31 districts with low not available for Indonesia. It is generally assumed coverage and high child mortality. to be negligible, however, since the majority of   vaccinations are delivered via the puskesmas/ Current requirements for Gavi financing require posyandu system and are, therefore, theoretically free compliance across several domains, such as Gavi’s of charge. Some puskesmas charge a nominal user or Transparency and Accountability Policy (TAP), aide registration fee for all services; this fee is variable and memoires, grant terms and conditions, cofinancing is waived in some districts for priority activities such policies, and submission of financial statements as immunization. In principle, both public and private and external audit reports. In addition, countries are health facilities are entitled to receive free vaccines also subjected to strict performance monitoring, from the government; for private facilities, this is and are required to submit further annual reports regardless of whether or not they are empaneled with where release of tranches is based on the receipt of JKN. Private health facilities may provide different satisfactory documents and availability of funds. In vaccines compared with those mandated by the addition, Gavi and partner countries will also need government, however, in such cases, JKN members 101 to agree on a performance framework based on as well as nonmembers must bear the cost for these existing monitoring and evaluation plans and other nonroutine vaccines out of pocket. Table 6.6 Type and Amount of Gavi Support for Indonesia Approvals Commitments Disbursements Type of support (US$) (US$) (US$) % Disbursed 2001 - 2020 2001 - 2020 2000 - 2016 Civil Society Organization support (CSO) 3,900,500 3,900,500 4,000,500 103 Health system strengthening (HSS 1) 24,827,500 24,827,500 24,827,500 100 Immunization services support (ISS) 12,636,000 12,636,000 12,636,000 100 Injection safety support (INS) 9,856,844 9,856,844 9,856,844 100 New and underused vaccine support (NVS) 72,191,000 88,658,500 61,832,000 70 Hb0 17,511,000 17,511,000 17,511,000 100 IPV 3,503,500 19,971,000 – – Penta 51,176,500 51,176,500 44,321,000 87 Vaccine introduction grant (VIG) 7,579,500 7,579,500 7,579,500 100 Total 130,991,344 147,458,844 120,732,344 82 Source Gavi. Note Information is as of April 30, 2016. HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better Availability of information on budgeting and by other MoH units beyond the EPI subdirectorate. expenditure on immunization activities varies by At the national level, much needed nonoperational levels of government, but is generally poor when immunization activities to comprehensively support the it comes to information from the subnational level. NIP, such as surveillance, health promotion, regulatory Central government budgeting on immunization systems strengthening and laboratory support, are programs is easily available, and national coverage the responsibility of directorates/subdirectorates of and vaccine administrative data are available as they MoH beyond the EPI unit. It is also important to note are reported by the JRF and on the MoH website. In that in Indonesia, through UNICEF and WHO, Gavi addition, regulations state that grants from NGOs has financed catalytic nonoperational immunization must be reported to relevant ministries. As previously activities, for example, technical assistance to conduct noted, however, provincial and district-level budgets assessment for immunization program scale up, and expenditure is not available, nor is it reported evaluation of postvaccine introduction, and data quality to central authorities. At the service level, it is not assessment. Systematic information on the cost of possible to track immunization resources received nonoperational immunization activities is currently beyond vaccine stock and administration monitoring. unavailable, although essential for further analysis of immunization financing in Indonesia. In 2016, a new DAK regulation enables more oversight to be given to subnational governments, Immunization expenditures are projected to increase potentially leading to better funding security for further as Indonesia experiences economic growth, each health program, including immunization. DAK indicating a need to secure sustainable funding for guidelines for 2016 provided detailed guidance and immunization. Elasticity analysis by Nader et al. (2014) indicators for activity cost components.74 The cost showed that countries spent about US$6.32 for every components include puskesmas staff transportation $100 in GNI increase from 2006 to 2012.75 Ensuring 102 cost for outreach activities to posyandu or home efficiency of spending is also a key issue. An analysis visit; BIAS implementation; introduction of new of government immunization program expenditures vaccines; capacity building (EVM, DQS); advocacy; in 51 lower- and lower-middle-income countries and outreach and coordination. Districts now have indicated that the mean cost per surviving infant on to submit a proposal, following which technical routine immunization was US$12 in 2006, and US$20 verification will be conducted by the EPI unit to in 2012. In Indonesia, the cost of immunization per ensure all components are included and costed. surviving infant in 2014 was ~US$33, a significant increase from US$13 in 2013, due to pentavalent Financing of immunization programs in Indonesia vaccine introduction (Table 6.7). Similar to Indonesia, must take into account the cost of activities conducted the cost per surviving infant in countries such as Table 6.7 Cost per Surviving Infant in Select Comparators Cost per surviving infant (US$) Country 2013 2014 Indonesia 13 33 Cambodia 40 33 Lao PDR 31 – India 25 – Sri Lanka 29 18 Vietnam 14 – Source Gavi Country Annual Progress Reports 2013 and 2014. Note Cost per surviving infant is calculated as total expenditures for immunization divided by number of surviving infants. 74 MoH Regulation No. 82/2015. 75 Expenditure data are analyzed from 68 of 73 GAVI Phase-II lower- and lower-middle-income countries over 2006–12. section six. A CASE STUDY ON IMMUNIZATION Cambodia and Sri Lanka varies by year, which could There are concerns about the usage of capitation also be due to new vaccines introduction. payments to improve immunization service delivery. Anecdotal evidence suggests that, due to confusion JKN’s benefit package currently covers routine among subnational governments and providers, JKN immunizations for children under five and tetanus capitation payments are often prioritized for financing immunization for pregnant women at primary health curative care only, even though it is supposed to facilities. Immunization services are free in public be used for individual promotive and preventive health facilities, regardless of JKN membership activities, including for immunization. Forty percent status. Unlike in public health facilities, JKN of the capitation payment is meant for operational members have to present their JKN card when support, however, it is not clear how the health facility visiting empaneled private facilities to obtain uses it to improve immunization services such as immunization services to avoid paying fee for upgrading and maintaining cold chain, availability of service (for example, immunizations provided by equipment (vaccine carrier, temperature monitoring the government). According to guidelines, however, device) and training for vaccinators. The capitation immunization services in hospitals are not covered payment is not currently linked to attainment of key by JKN–routine vaccines are provided free of health outcomes such as immunization. charge, but the cost to administer the vaccine is not. The MoH is advocating for adding immunization for The treatment and care of AEFI cases resulting from school-aged children, boosters for children aged 18 immunization, is unclear following the introduction months and 24 months in the JKN benefit package of JKN. Prior to JKN introduction, the government to ensure that the entire routine immunization covered the costs associated with the management package outlined in the national immunization of AEFI cases, including those not covered under schedule is covered under JKN. For new vaccines, any health insurance scheme. In contrast, after JKN there is no separate procedure to include the implementation, only JKN members are covered for 103 vaccines in JKN benefit package. Any new vaccines medication and treatment resulting from AEFI cases. included in the routine immunization package, This might have implications for the confidence of such as IPV, are automatically covered under JKN’s health care workers to administer the vaccines as benefit package. well as the willingness of the community to seek immunization services. HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better Introduction of New Vaccines Expanding access to nontraditional vaccines will capacity, because the present packing volume of considerably increase the funding requirements. rotavirus vaccines is approximately 7-18 times greater As previously mentioned, Indonesia is planning to than the packing volume of traditional vaccines introduce several new vaccines (Measles Rubella, (for example, DPT vaccines). Initiatives to minimize Japanese Encephalitis, Pneumococcal, Rotavirus, the packing volume may be required and have and Human Papilloma Virus) in the next few years. implications on additional budgeting for cold-chain These new vaccines are typically costlier. The and logistics systems (Suwantika et al. 2014). Human financial projection that includes only Pentavalent resources capacity needs to be strengthened in the and excludes other new vaccines, suggests that form of adequate training of health care workers the country cofinancing will likely increase from and adjustments of the number of vaccinators US$2.1 million in 2013 to US$32.6 million in 2017.76 as the number of injections per child per session The additional funding requirements will also have to increases. Improved incentive and support from the take into account the nonvaccine immunization costs government to support quality and cost-effective 104 (surveillance, quality assurance, training, and so forth). vaccines production by Bio Farma is needed as well as strengthened coordination and cooperation among Rigorous cost-effectiveness analysis, comprehensive different stakeholders including manufacture, national financial projections, fiscal space analysis and detailed regulatory authorities, and NIP. financial planning for the new vaccines would be useful as part of the transition plan to ensure adequate Learning from the successful introduction of financial capacity to meet the increased demand for pentavalent in Indonesia, several key factors that resources. Moreover, as Indonesia is graduating from include local scientific evidence, financing, production, Gavi, it may limit the access to grants that have been and political support must be taken into account used in the past to support the cost of preintroduction for the planned introduction of new vaccines. This activities, such as health worker training, IEC, social was evidenced by the successful introduction of mobilization, and technical assistance. Under the pentavalent vaccine in Indonesia where there was exceptional catalytic funding support, Gavi will support an adequate, evidence-based recommendation the introduction of JE, MR and HPV in 2017 and the from WHO and local experts, such as the Indonesian government is expected to fund them going forward. Technical Advisory Group on Immunization (ITAGI) and IPV introduction in 2016 receives full support from Gavi the Indonesian Pediatric Society, following local cost- until 2018, but will be fully funded by the government effectiveness studies. In addition, there was sufficient in the subsequent years. government financing for the new vaccine, and Bio Farma was able to produce the required amount of Introduction of new vaccines will require systematic pentavalent vaccine to meet the needs of the country. service delivery readiness to accommodate the More importantly, political support was successfully changes in service modalities and the scope of garnered from key stakeholders, including MoH, services provided. New delivery technologies and parliament, and religious societies (Hadisoemarto et al. readiness of cold-chain storage and logistic systems 2016). Ensuring that all these key factors are in place is may be needed. For example, rotavirus vaccines key when the government is planning the introduction may require additional cost to expand cold-chain of new vaccines such as MR, JE, and HPV. 76 Gavi Alliance, estimates as of September 2013 from Saxenian et al. 2014 section six. A CASE STUDY ON IMMUNIZATION 105 HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better 106 section seven. DISCUSSION AND POLICY OPTIONS section 7 . 107 DISCUSSION AND POLICY OPTIONS HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better Indonesia has made key strides towards attaining countries spend more than these targets and have yet UHC in terms of population coverage. The country to achieve UHC, while others spend less and manage has made a commitment to attain UHC by 2019 when to ensure access to a basic health package for their everyone in Indonesia is expected to have coverage entire population. 108 under JKN. In 2015, nearly 160 million individuals, or more than 60 percent of the population, have been Benchmarking–as opposed to measuring covered by JKN. Nevertheless, Indonesia faces key deviations from targets–can be a better challenges in order to meet its 2019 population way to assess the sufficiency of resources. coverage target as well as on other, arguably more Global and regional benchmarks indicate that important, dimensions of UHC, including service Indonesia’s health system remains significantly coverage and financial protection. underresourced. At 3.6 percent of GDP, Indonesia’s THE levels are among the lowest in the world, and From a health financing perspective, one of the key are particularly low when benchmarked against bottlenecks to attaining UHC in Indonesia is the other lower-middle-income countries (5.9 percent relatively low quantum of health spending. There is of GDP) and across the EAP region (6.6 percent of no clear normative answer to the question of how GDP). Although low levels of health spending can much a country should spend on health (Savedoff also be a sign of efficiency, as discussed below, 2003). A substantial body of literature focuses on the this is clearly not the case for Indonesia. The issue of sufficiency in terms of assessing deviations, average THE rate among developing countries for instance, from a health spending target of 5 that have already attained three of health SDGs77 percent of GDP or public spending of 15 percent of is 6.6 percent of GDP, significantly higher than the the government budget. Although spending targets amount spent by Indonesia in terms of share of can serve as inputs for estimation of the magnitude GDP. Actuarial estimates for expanding UHC also of global financing gaps, they are usually not very indicate underresourcing of Indonesia’s health helpful at the country level, given the wide variations system, to the extent of at least 1 percent of GDP in country and health-system contexts. Resource (Guerard et al 2011). Total health spending in needs will vary country-by-country, and even the most countries such as Malaysia and Thailand–both of ambitious expenditure targets will not ensure that which are further along the UHC spectrum than UHC will be achieved in all countries. Indeed, many Indonesia–is in the 4-5 percent of GDP range. 77 A neonatal mortality rate of less than 12 per 1,000 live births, an under-five mortality rate of less than 25 per 1,000 live births, and an MMR of less than 70 per 100,000 live births. section seven. DISCUSSION AND POLICY OPTIONS Although Indonesia is following a SHI model for services (constraining spending on other necessary attaining UHC by 2019 in principle, in reality the expenditures). Given the general unpredictability and health system is financed by a combination of undesirability of health shocks and expenditures, sources and separate flows. The four primary sources OOP spending should generally only be used as a of health financing in the country include OOP means for managing overutilization and reducing spending by households, government budgetary waste in more advanced health systems, and not supply-side health spending (both at the central as a primary mechanism for resource generation in and subnational levels), SHI, and external financing. developing health systems such as Indonesia’s. Despite increases in public financing in recent years, the fundamental structure of health financing has High levels of OOP spending are, in large part, a remained largely unchanged in Indonesia because of result of relatively low levels of public financing for concomitant increases in OOP spending for health. health in Indonesia. Despite recent increases in public financing–including via the expansion of SHI as well as OOP spending by households remains the largest increases in government budgetary health spending– source of financing for health in Indonesia at 45.3 Indonesia’s health financing transition is stalled percent of THE in 2014. OOP payments are an because OOP spending has risen at commensurate inefficient and inequitable means of financing rates. For OOP spending to decline significantly, public health systems. OOP payments connect utilization financing for health will have to increase at a rate faster of health services to an individual’s or household’s than the rise in OOP spending for health. ability to pay; deter utilization (especially for the poor), thus exacerbating or sustaining inequalities; No country has attained the SDGs and reduced OOP and expose individuals or households to the risk spending on health to less than 30 percent of total of impoverishment that results from high levels of health spending without public expenditures on health expenditures when they do utilize health health being at least 2.7 percent of GDP, much higher 109 Table 7.1 Indonesia Compared With Countries That Attained Key Health SDGs With an OOP Share of Total Health Spending <30% Total health Total health Public OOP spending GNI per Under- Maternal spending spending spending on Neonatal share of Country capital five mortality per capita, share of health share mortality* total health (US$) mortality* ratio* (US$) GDP (%) of GDP (%) spending (%) Indonesia 3,383 126 3.6 1.1 13.5 27.2 126 45.3 Argentina 12,241 605 4.8 2.7 6.3 12.5 52 30.7 Fiji 4,888 204 4.5 3.0 9.6 22.4 30 23.0 Brazil 11,491 947 8.3 3.8 8.9 16.4 44 25.5 Belize 4,376 279 5.8 3.9 8.3 16.5 28 23.0 Turkey 10,394 568 5.4 4.2 7.1 13.5 16 17.8 El Salvador 3,949 280 6.8 4.5 8.3 16.8 54 28.8 Romania 9,805 557 5.6 4.5 6.3 11.1 31 18.9 Hungary 13,406 1,037 7.4 4.9 3.5 5.9 17 26.6 Jordan 5,359 359 7.5 5.2 10.6 17.9 58 20.9 Thailand 5,648 360 6.5 5.6 6.7 12.3 20 7.9 Samoa 4,042 301 7.2 6.5 9.5 17.5 51 5.9 Costa Rica 10,071 970 9.3 6.8 6.2 9.7 25 24.9 Bosnia and 4,907 464 9.6 6.8 4.0 5.4 11 27.9 Herzegovina Source World Development Indicators database (Income level and health expenditure 2014, SDGs indicators 2015). Note *Mortality rates are per 1,000 live births for neonatal and U5 and per 100,000 live births for MMR. ** Indonesia data is based on the NHA country report, 2014 (Ministry of Health - Center for Health Economics and Policy Studies - AIPHSS, 2015) HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better than the rate for Indonesia (1.5 percent of GDP in The central government does not have mechanisms to 2014) (Table 7.1). Although this does not mean that if incentivize the generation of outputs/outcomes from Indonesia were to increase public financing for health use of resources, nor does it have clear policy levers to to 2.7 percent of GDP, it will then attain the SDGs and influence the allocation of resources at the subnational reduce OOP spending shares to less than 30 percent. level. Although some of the allocations of resources The implications are more so that public financing for are based on district characteristics, the capacity health will need to rise significantly beyond current of districts to plan for, absorb, and realize outputs/ levels in order for Indonesia to make progress on outcomes is not a key determining factor; the focus improving service coverage and financial protection. to date has largely been on ensuring compliance Nevertheless, despite OOP spending for health with rules/norms rather than on building the capacity remaining high, there is evidence that the incidence of and/or incentivizing districts to effectively utilize of catastrophic health expenditures has declined. resources in order to improve service delivery. The key This is possibly a result of the relative progressivity policy challenge will be ensuring that these additional of OOP spending in Indonesia’s health financing resources are absorbed and utilized effectively and system. In spite of this progressivity, high OOP that district governments and public facilities are spending deters utilization by the poor and reduces provided with the necessary technical assistance the redistributive capacity of health financing and is and incentives to do so. Some districts continue to undesirable even if this results in higher levels of OOP view health as a revenue-generating sector and have spending being incurred by the rich. revenue-raising targets from user charges that are then pooled at the district treasury level along with Supply-side government budgetary expenditures other revenue sources. are the second-largest source of health expenditures in Indonesia (41.4 percent of THE). Given Indonesia’s SHI is the third largest source of, and agent for, 110 decentralized governance arrangements, most health expenditure in the country (13 percent of THE). government health expenditures occur at the district Almost one-half of all social health expenditure is level. Most government revenues are, however, sourced from the central government in the form raised at the central level and are transferred to of premium payments for the poor and nearpoor. subnational governments using a complex system of Despite relatively large increases in SHI expenditures intergovernmental fiscal transfers. For the most part, in recent years, new concerns have emerged districts have discretion on how much gets spent regarding their equity and financial sustainability. on health at the local level. Although government Although Indonesia has successfully instituted a budgetary allocations to health are increasing from single-payer SHI system, contribution collection a relatively low base, they could be made more among nonpoor informal workers has been difficult effective and efficient in achieving policy objectives (under current regulations, this group must contribute and health outcomes. in order to enroll in JKN). Few nonpoor informal workers have enrolled to date and those that have The generally low levels of government budgetary are adversely selected, undermining equity, and expenditures for health reflect both a low threatening financial sustainability of JKN. government revenue mobilization effort as well as a relatively low prioritization given to health, especially Since the 2008 Health PER, there has been some at the central government level. Recent policy significant progress with the implementation of the efforts have led to a reprioritization for health at the JKN program. The implementation of JKN has raised central level, with health poised to receive the legally expectations for improved access to health care and mandated 5 percent of government budgetary reduced OOP spending. There is an encouraging allocations in 2016. There have also been increases trend from 2005 that the share of OOP spending to in the intergovernmental fiscal transfers earmarked THE is decreasing, and the government is increasing for health via DAK. Indonesia does not have an the budget allocation for health sector. However, explicit results-based orientation in its system of OOP expenditure continues to be the biggest source intergovernmental fiscal transfers. of Indonesia’s health spending and the effect of JKN expansion on the composition of health spending remains to be seen. section seven. DISCUSSION AND POLICY OPTIONS JKN provides comprehensive coverage without fragmentation of planning, financing flow, reporting 111 copayments or quantitative limitations, however, and monitoring requirements, and management of JKN reimbursements do not cover the full cost of services and human resources. Moreover, there is a care and significant cofinancing from government risk for continuity of funds which could be influenced budgetary expenditures remains in the public health by global economy. system. It will, therefore, be critical to: (i) integrate and leverage demand-side JKN financing with Decentralization poses a significant risk to the success government budgetary supply-side financing to and sustainability of the externally financed health attain improvements in health outputs and outcomes; programs. Although the central government procures and (ii) ensure that the partial reimbursement model and distributes drugs and vaccines, provincial and does not threaten quality of care rendered by the district governments manage the operations of public empaneled private facilities. Targeting of JKN needs health facilities and services. This poses numerous to be improved so that the poorest 40 percent of challenges to program evaluation and sustainability. For the population are covered as intended with central example, expenditure on the immunization program government financing. from subnational levels is not reported back to MoF or MoH. Management capacity and commitment to key Although external sources of financing are not health programs is extremely variable across different a dominant overall source of health financing, provinces/districts, leading to varying service coverage accounting for less than 1 percent of THE, it is the rates. There has been some anecdotal evidence of fourth largest source of financing for health and limited allocated operational budget for key health an important source of financing and technical programs at subnational government level that could assistance for specific programs. International potentially lead to suboptimal service delivery. development partners may bring in global experiences and introduce innovative interventions. Stronger and clearer links to JKN is key to the External resources may also fill in the gap where sustainability of externally financed health programs. government budget has less flexibility, as well as For example, how will the provider reimbursement improve accountability and good governance. The framework create incentives for the provision downside is that external financing comes with of services? In this regard, and as JKN expands HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better coverage, the key to financial and institutional opportunities exist for improving the efficiency sustainability will be for these programs to be better of Indonesia’s health system. In practical policy integrated within the context of UHC, both as an terms, efficiency improvements entail looking for indicator of progress towards UHC attainment (as opportunities to reduce costs without reducing recommended by a recent joint WHO-WB report on health services/outcomes and/or improving health this issue) as well as a program element to which a services/outcomes for the same costs. The WHO proportion of intergovernmental fiscal transfers and report indicates that, globally, the ten leading BPJS provider payments could be linked. Although sources of inefficiency include: (i) underuse of generic lessons from countries that are more advanced medicines and higher than necessary prices for along the UHC pathway suggests that it is not medicines; (ii) use of substandard and counterfeit necessary to subsume these programs and services medicines; (iii) inappropriate and ineffective use under JKN programs, there are benefits to ensuring of medicines; (iv) supplier-induced demand and closer coordination, integration, and to making overuse of some services; (v) inappropriate staff the UHC link more explicit. Better coordination mix and unmotivated workers; (vi) inappropriate of service delivery with empaneled nonpublic hospital admissions and length of stay; (vii) low use providers as well as joint monitoring and evaluation of infrastructure such as hospital beds; (viii) medical are some of the additional potential efficiency- errors and suboptimal quality of care; (ix) waste and enhancing benefits of closer UHC integration. fraud; and (x) inefficient mix and inappropriate level of interventions (WHO 2010b). Indonesia continues its commitment to meeting SDG targets and persisting challenges in the availability In Indonesia–as noted at various points in this policy and readiness of health service delivery systems. paper–prominent options for enhancing the efficiency This includes addressing chronic challenges in the of health expenditures include: 112 availability and readiness of health services, with increasing demand due to, and the promise of, JKN • reducing OOP payments by expanding and implementation. Even with the combined demand- deepening coverage; supply financing model, there is a disconnect between • improving primary care; public financing for and JKN benefits especially for • improving the distribution and quality of HRH; those living in relatively remote and rural locations • enhancing the effectiveness of of the country, and this limits the effective availability intergovernmental fiscal transfers; and of benefits for many JKN beneficiaries. Supply-side • Strengthening linkages between UHC and constraints reflect not only shortages in overall priority programs, for example immunization, numbers, but also in distribution given the geographic HIV/AIDS, and tuberculosis. context of Indonesia. They comprise all the factors that limit health care delivery at the point of service, Results-based incentives for service delivery including the number of physicians, nurses, and could be pursued to improve service coverage of midwives; the number of beds; medical equipment key priority programs and to enhance efficiency and technology; medicine supplies; and other basic in spending in the future. From the demand side, amenities. Although detailed data on the subnational Indonesia could build on extensive experiences in distribution of health expenditures including social implementing household and community conditional health insurance, are not available, there are cash transfers program that incentivize results at indications that inequities are prominent. household and local level, respectively, albeit at limited scale. From the supply-side, tying provider In order to resolve these persisting issues, financing payments to attainment of population-level service the health sector will require not only an increase coverage rates has been tried in other countries and in funding, but also well-targeted allocation could be piloted as a potentially effective mechanism and spending, and improved efficiency. Several to improve service readiness. section seven. DISCUSSION AND POLICY OPTIONS POLICY OPTIONS To make substantial progress towards service Ensure adequate public financing for UHC: While coverage and financial protection in order to achieve Indonesia has recently increased its government UHC by 2019, Indonesia would have to spend more, health spending, it remains one of the lowest in spend right, and spend better. The following are the world at 1.5 percent of GDP. No country has some policy options for Indonesia to consider: attained adequate SDGs and reduced OOP spending on health to less than 30 percent of total health Make the JKN benefits package explicit and adjust spending without public expenditures on health the package to be commensurate with financing being at least 2.7 percent of GDP, much higher and service delivery capacity: The current JKN’s than the current rate for Indonesia. It is, therefore, benefits package is comprehensive and is not crucial to increase government health spending as a explicit in that all medically necessary coverage necessary and critical but not sufficient, condition to is automatically deemed to be covered without progress towards achieving UHC. Acknowledging the any copayments, balanced billing, or expenditure challenges in increasing the fiscal space for public caps. All registered JKN members are entitled to financing for UHC, key options to address this include 113 a range of medical services, including a range a combination of: (i) increasing nonoil and gas tax of services that fall under promotive, preventive, revenues; (ii) central government’s reprioritization curative and rehabilitative services. While the list is of health (including from reduced fuel subsidies); comprehensive and specifies a negative list, it does (iii) efficiency gains; (iv) earmarked tobacco taxes; (v) not clearly spell out a positive list of what is covered complementary subnational financing; (vi) targeted under the program, the latter being inferred by incentives/penalties for enrolling the informal sector; providers from national clinical guidelines and from and (vii) incentives to formalize the informal sector. drugs that are included in the national formulary. To ensure that JKN’s covered services and benefits are Increase focus on primary health care, including available for all members and the resources (both prevention and promotion: There are concerns that financial and human) required to deliver the benefits the focus on UHC is for curative and rehabilitative are available, the JKN benefits package needs to be care and is distracting from the focus on improving more explicit and adjusted in line with current public primary health care and population/public health financing resources, economic growth and projected interventions. With the epidemiological transition macrofiscal trajectory, and service delivery capacity. towards NCDs already underway in Indonesia, this Indonesia may learn from other countries’ experience, will lead to an unsustainable fiscal burden on the JKN, such as Chile, in moving from a comprehensive greater OOP for those not covered or even increased benefit package to a basic set of explicit benefits numbers of patients foregoing treatment. Most cost- guaranteed with adequate financing from public effective interventions are usually delivered at the sources (via government budgetary supply-side population level (for example, increasing tobacco expenditures and/or social health expenditures). taxes to reduce smoking rates which are alarmingly Mechanisms can then be put in place such that high in Indonesia) as well as the primary-care level (for subsequent expansions to benefits are made in example, early diagnosis and treatment, community- tandem and commensurate with planned expansions level outreach, interpersonal communication for in public financing. behavior change and lifestyle modification). HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better Cover the nonpoor and eliminate mistargeting: Given results: In Indonesia’s decentralized context, challenges related to public financing, supply-side increased government health spending needs to readiness, equity in and financial sustainability of be complemented with system improvements to social health expenditure incidence, and implicit address the persistently low and unevenly distributed rationing, the following could be done related to quantity and quality of health services, with a focus JKN: (i) although improvements in socialization, on lagging and remote districts. First, improve local awareness, and availability of benefits may government’s (provincial and district) capacity to improve enrollment of the nonpoor in JKN, global prioritize, mobilize, plan, budget, and effectively experience indicates that this may not be sufficient utilize both supply- and demand-side financing, in and alternatives would need to be considered; order to improve availability and utilization of quality and (ii) mistargeting needs to be eliminated, and health services. Second, strengthen the monitoring the current system does not incentivize local and evaluation system to enable provision of an governments to enroll targeted beneficiaries. One independent assessment of results achievement to option may be for BPJS to transfer resources to local make local governments more accountable. Social governments based on verified local numbers rather accountability is another mechanism by which than on capitation as is currently the case. providers could be held accountable, building upon the experience from within and outside of Indonesia. Integrate supply-side and demand-side financing Third, introduce nonfinancial incentives (for example, to improve public and private provider supply benchmarking and public notification and rewards) side readiness: At the puskesmas level, where the and financial incentives tied to achievement of predominant provider payment method for health results by linking intergovernmental fiscal transfers facilities is capitation, this payment should be linked to achievement of results such as minimum service either directly or indirectly to the attainment of standards for health. 114 minimum standards. Facilities should be allowed discretion on how capitation funds are utilized, and Minimum Service Standards as an instrument that reimbursement from BPJS should not become a could potentially be used as Central government revenue source for district government for general- levers to influence sub national level: The most purpose use. More generally, as financing gradually recent amendment to the Decentralization Law in shifts from the supply-side to the demand-side UU 23/2014 states the Minimum Service Standards in Indonesia’s health system, an appropriate level (MSS) related to the distribution of governance of autonomy for health facilities–coupled with affairs and authority between the central and enhanced capacity to manage revenues–needs regional governments (province and district/ to be found for public health facilities. Inclusion of city), which are regulated based on the criteria of private providers should also focus on ensuring externality, accountability and efficiency. In particular, supply-side readiness at these facilities as well as it is underlined that the central government is providing adequate capitation amounts to level the responsible for setting the standards to be used by playing field with the public sector facilities that also local governments (provincial and district/city) as receive supply-side financing. At the hospital level, a reference in the implementation of basic health diagnosis-related group payments could be made services. Minimum Service Standard aims to ensure conditional on the adequacy of services provided in the delivery of essential services and to ensure the order to encourage investments in improving service accountability of different levels of government with readiness. As the health system evolves, additional the inclusion of a set of agreed indicators to measure measures to mitigate negative incentives inherent achieved results. As a planning and budgeting in capitation systems–such as overreferral and tool MSS is expected to serve as the reference for inappropriate referral to secondary care as well as prioritizing budget allocation for these basic health undertreatment–should be considered. services and is an instrument that could potentially be used as Central government levers to influence Increase effectiveness of intergovernmental sub national level. MSS is expected to be released fiscal transfers by improving local government as a presidential regulation. The mechanism to capacity, ensuring accountability, and incentivizing ensure sub national compliance by holding the section seven. DISCUSSION AND POLICY OPTIONS head of regions (Governor, Bupatis/Mayors) directly the quantum of financing required, but also on the accountable in achieving minimum services targets governance and service delivery mechanisms in is a potential game changer, but how the follow- place to deliver these services. As JKN expands up ministerial regulations and guidelines are coverage, the key to financial and institutional implemented will determine the impact of the MSS. sustainability will be for these externally-financed health programs to be better integrated within the Stronger and clearer links to JKN is key to the context of UHC. sustainability of externally financed health programs: While external financing constitutes a mere one Leveraging JKN provider payment mechanisms percent of THE, this finances several priority health to incentivize preventive/promotive services for programs, including HIV and AIDS, TB, malaria, results: Provider payment mechanisms under JKN and immunization. To continue progress made on are “passive” in that there are no explicit linkages these programs, there needs to be a transition plan with outputs/outcomes. Improved socialization to ensure that services continue to be available of guidelines on use of JKN capitation payments and scaled up, even after donors transition out of would help as would other mechanisms such as Indonesia. In countries that have not planned their introduction of “strategic” purchasing, e.g., to better transition, there has been disruption of services, integrate JKN provider payment mechanisms with which could have serious implications for health provision of preventive/promotive care so as to outcomes, such as control of MDR-TB. To transition improve the efficiency and financial sustainability of smoothly, Indonesia needs to focus not only on public financing for UHC in Indonesia. 115 HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better 116 APPENDIXES 117 APPENDIXES HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better Appendix A Interfiscal Transfer Diagram Profit Sharine Funds Tax Profit Sharing Balance General Allocation Funds Funds Property Tax Special Allocation Funds Income Tax PAPUA Special Autonomy Funds Tobacco Excise WEST PAPUA Special Autonomy Funds TRANSFER TO THE Special ACEH Special Autonomy Funds Natural Resources LOCAL LEVEL Autonomy Funds Profit Sharing 118 PAPUA Infrastructure Funds Forestry WEST PAPUA Infrastructure Funds General Mining Special Autonomy and Adjustment Funds Additional Teacher Income Gas and Oil Teacher Professional Allowance Geothermal School Operational Assistance Fisheries Adjustment Funds Local Incentive Funds Source Act No. 17, 2003 APPENDIXES Appendix B Flow of Funds fiscal balance transfer Ministry of Finance Other Ministries national deconcentration & budget co-administration grants mechanism and regional Ministry of office fund for hospital and/or loans Health social insurance contribution social insurance contribution Badan Penyelenggara Provincial Provincial Health Jaminan Sosial Government Office (PHO) Kesehatan BPJS Kesehatan social insurance grants contribution and/or loans District District Health External Donors Government Office (DHO) Hospitals 119 (Public, Private, Parallel Health System) local social scheme insurance Providers of Ambulatory funds; contribution co=financing Health Care national (GPs, specialist, Clinics) taxes grants Providers of Ancillary Services Autonomous Local Scheme Retailers and Other Providers of Medical Goods Providers of Preventive Care local scheme Providers of Health Care System contribution Administration and Financing local taxes Rest of Economy Rest of the World External Agencies Other Demand Side Schemes (VHI, NPISH) Population / Enterprises self insured in private companies out of poc et payment for direct access to health care Governmental financing scheme Social Insurance Financing scheme Providers Public financing Private financing Donor financing Private financing scheme Transfer between scheme HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better Appendix C Funds Flow of The Global Fund Grants FLOW OF FUND AND REPORTING FROM PR TO SR, SR CENTRAL / NGO AND SSR Government Special Account 1USD/IDR Account as approved by MoF 120 PRINCIPLE CCM/TWG RECIPIENTS 2USD/IDR Sub Recipients Sub Recipients Sub Recipients Account as Province Central NGO approved by MoF 3USD/IDR SSR SSR Account as SSR Central NGO approved by MoF Health Reporting Facilities Funding flows APPENDIXES Appendix D Tobacco Taxation Reform in Indonesia: Good for Public and Fiscal Health78 Tobacco use is a major but preventable public HEALTH AND SOCIAL IMPACT OF TOBACCO USE health risk in Indonesia and, contrary to trends in other middle-income countries, smoking prevalence Tobacco is harmful and kills prematurely almost and its intensity per adult is high and growing. one-half of its long-term consumers (Jha and Peto Two out of three adult men smoke, a rate that has 2014). In Indonesia, smoking was the third leading been increasing since the 1990s and is one of the risk factor of avoidable premature deaths (measured highest in the world. According to the 2011 Global in disability-adjusted life years or DALYs) among Adult Tobacco Survey (MoH and WHO 2012), 67.4 males in 2013, after dietary risks and high systolic percent of Indonesian men and 4.5 percent of women blood pressure. In terms of the number of years of (comprising 36.1 percent of the population or 61.4 life lost (YLLs) due to premature death, diseases million people) use tobacco in smoked or smokeless in which smoking is one of the risk factors, such as form. Tobacco use is more prevalent in rural areas cerebrovascular and ischemic heart diseases, and (39.1 percent) compared with urban areas (33.0 lower respiratory infections were the highest ranking percent). Smoking is the main form of tobacco use causes of mortality in 2013 (IHME 2013). 121 and more than one-third consume cigarettes. Kretek79 is the most popular kind of cigarette, followed by a Tobacco-related deaths as a proportion of total male wide distance by hand-rolled and white cigarettes. deaths have been increasing since 1990–from 13 The prevalence of kretek smoking is more common percent to 20 percent in 2013. Even among women, among men, middle-aged groups, and those with tobacco-related deaths have also increased, from lower levels of education. 6 percent in 1990 to 9 percent in 2013. Smoking- related diseases and premature mortality lead to Annual consumption of cigarettes per adult has made higher health care costs and lower labor productivity, Indonesia the fourth-highest cigarette consuming undermining the potential development of the country in the world. Rising incomes and greater country. For example, using recent health and affordability, and cigarette prices that are among the medical spending surveys in the United States, lowest in the world, are factors underlying growing researchers calculated that 8.7 percent of all health cigarette consumption in Indonesia. The ratio of the care spending, or US$170 billion a year, is for illness retail price of 100 packets of cigarettes to GDP per caused by tobacco smoke, and public programs like capita has decreased over time: from 5.8 percent in Medicare and Medicaid paid for most of these costs 2008 to 4.9 percent in 2012. Indonesia maintains one (Xu et al. 2015). In Indonesia, health care costs for of the most complex tobacco tax structures in the tobacco-related illnesses were estimated to amount world that promotes downward substitution to more to between US$319 million and US$1.2 billion per affordable products. There are no nationwide bans on year (Barber et al. 2008). tobacco advertising, promotion, and sponsorship. 78 This policy note was prepared by Patricio V. Marquez, Lead Public Health Specialist (HNP GP), The World Bank Group. 79 Kretek is a type of cigarette consumed in Indonesia. It contains a mix of tobacco leaves, cloves, and other additives. HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better Tobacco consumption in middle- and lower-income who are addicts to quit and preventing nonusers– households also competes with investment in especially young people, women, and the poor–from human capital (for example, health, education) ever starting. Higher tobacco taxes also provide which undermines longer-term labor productivity countries with additional revenue that can be used growth. Households with smokers dedicate 11.5 to fund vital health programs and other essential percent of monthly expenditures to tobacco, and public services–a highly relevant option for Indonesia such high spending has serious welfare implications. given the size of the current budget deficit that is The national nutritional surveillance system approaching 3 percent of GDP (Marquez 2016a).81 reported that paternal smoking is a predictor of an Indeed, tobacco taxation is an untapped source increased probability of short-term and chronic child of domestic financing that will be important for malnutrition (Xu et al. 2014). In addition, exposure the successful implementation of the Sustainable to second-hand smoke has a negative impact on Development Goals (SDGs) by 2030 as advocated in nonsmokers, particularly women and children at the the Financing for Development (Addis Ababa Action household level. More than 97 million nonsmokers Agenda) of (United Nations, 2015b). are regularly exposed to second-hand tobacco smoke known to be carcinogenic (CDC 2016). CONSUMERS RESPONSE AND OPTIMAL LEVEL OF TAXATION TOBACCO TAXATION: A WIN-WIN PUBLIC HEALTH AND FISCAL POLICY MEASURE Governments generally levy taxes on tobacco that are either excise (which is a selective consumption tax), As the part of the Framework Convention on Tobacco custom duties (on imports), and/or as a value-added Control (FCTC)(WHO 2003), a global treaty adopted tax (VAT) or general sales tax (GST). The excise tax 122 in 2005 and ratified by most countries in the world, levied on tobacco can be structured as a monetary WHO has identified six policies–encapsulated in the amount by quantity (for example, by package, piece, or acronym MPOWER–that can stamp out the tobacco weight) which is referred to as “specific”; or calculated epidemic. These six policies are: (i) monitor tobacco as a percentage of the price (“ad valorem”). Of these use and prevention policies; (ii) protect people from two types of excise tax, lower-income countries tend tobacco smoke; (iii) offer help to quit tobacco use; to use ad valorem excise taxes while higher-income (iv) warn people about the dangers of tobacco; (v) countries tend to use either specific or a mix of both enforce bans on tobacco advertising, promotion, and types of excise (WHO 2010b). sponsorship; and (vi) raise taxes on tobacco. Each letter of the acronym is important and necessary The extent to which consumers’ demand for a in the fight against the tobacco epidemic. Tobacco good changes in response to a price change due taxation is seen as the most cost-effective policy to adoption of higher taxes is known as the price measure to confront tobacco use and prevent its elasticity of demand. For example, if a price rise of negative health, social, and economic impact. 10 percent causes the quantity demanded to fall by 5 percent, the elasticity of demand is -0.5. The more Taxes on tobacco cost little to implement and lead price-responsive consumers are, the greater is the to a windfall of benefits. The primary motivations for elasticity of demand. Evidence from across the world imposing higher tobacco excise tax are to discourage shows that smokers’ demand for tobacco, while smoking and raise resources to compensate for inelastic, is nevertheless strongly affected by its price societal costs of smoking on nonsmokers (for (World Bank 1999). When the price of a good rises, example, due to higher health care costs and adverse people on low incomes are, in general, more likely to health effects from second-hand smoke)((Savedoff cut back their consumption of that good than people and Alwang 2015).80 Higher taxes on tobacco make on high incomes and, conversely, when the price falls, tobacco products less affordable, helping smokers they are more likely to increase their consumption. 80 See also WHO 2011 and Jha et al. 2012. 81 See also Marquez 2015b and 2016b. APPENDIXES Estimates of elasticity vary from study to study, but health risks, and save lives. In many countries, there is reasonable evidence that in low- and middle- where incomes and purchasing power are growing income countries, elasticity of demand is greater than rapidly, large price increases are required to offset in high-income countries. In the United States, for the impact of growth in real incomes on tobacco example, researchers have found that a price rise of 10 consumption habits. Strong tax administration is also percent for a pack of cigarettes decreases demand by critical to minimize tax avoidance and tax evasion, about 4 percent (an elasticity of -0.4). Studies in China to ensure that tobacco tax increases lead to higher have concluded that a price rise of 10 percent reduces tobacco product prices and tax revenues, as well as demand by between 6 and 10 percent (elasticity reductions in tobacco use and its negative health between -0.6 and -1.0). Studies in Brazil and South consequences. Africa have produced results in the same range. For low- and middle-income countries as a whole, then, Across the globe, several countries have a reasonable estimate of the average elasticity of implemented various types and size of tobacco demand would be -0.8, based on current data. A study taxes, and have earmarked the additional tax revenue by the Centers for Disease Control and Prevention collected to expand the fiscal space to fund priority (CDC)(World Bank 1999) found that demand elasticity investments and programs that benefit the entire among young adults aged between 18 and 24 in population, such as the expansion of Universal the United States was -0.6, higher than for smokers Health Coverage (WHO 2015). The Philippines is overall. Researchers conclude that when prices are often referred to as a benchmark for Indonesia due high, not only are existing young smokers more likely to its geographical proximity and similarities in many to quit, but that fewer potential young smokers will aspects. After the adoption of the ‘Sin Tax’ at the end take up the habit. of 2012, tobacco excise tax collection has more than doubled from the baseline 0.3 percent of GDP to The landmark World Bank report (1999) suggested a 0.8 percent of GDP in 2015. In the first three years of 123 pragmatic approach to define the “optimal tax level” implementation of the Sin Tax Law, US$3.9 billion or for cigarettes by observing the tax levels adopted by about 0.5 percent of GDP, in additional fiscal revenues countries with comprehensive and effective tobacco was collected. control policies. In such countries, the tax component of the price of a pack of cigarettes is between two- Following the tax revenue increase in the Philippines, thirds and four-fifths of the total retail cost. These 85 percent of the additional revenue has been levels are currently being used globally as a yardstick used for health programs, of which 80 percent is for proportionate increases in prices elsewhere, and to help finance the extension of fully subsidized imply, for example, that if tax is to account for four- health insurance for the poorest 40 percent of the fifths of the retail price, this requires prices to be population. As a result, PhilHealth nearly tripled the increased by four times the manufacturer’s (untaxed) number of families enrolled in the National Health price per pack.82 The impact on retail price would, of Insurance Program (NHIP) from 5.2 million families course, vary between countries, depending on retail in 2013 to 14.7 million families in 2014. Lessons factors such as the wholesale price, but broadly, an learned from the Philippines success story include increase of this order would raise the population- the clear focus on health, building strategic alliances weighted price by between 80 and 100 percent in and political support, and the use of strategic and low- and middle-income countries. effective communication (Kaiser et al. 2016). Following this commonly accepted approach, therefore, tax increases, using specific excise taxes TOBACCO TAX STRUCTURE IN INDONESIA or a combination of specific and ad valorem excise taxes, should aim to reduce the affordability of Indonesia applies multiple tax types (excise, customs tobacco products, decrease consumption, reduce duties, and VAT/GST) and its tobacco excise tax is 82 Forinstance, if a nontax price is equal to $0.50, then the tax rate would be 0.5 x 4 = $2. The retail price would be equal to $2.50 ($2 tax plus $0.50 manufacturing cost). HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better structured as a specific tax. By law, the maximum has been reduced. Indonesia’s tobacco market is still allowable cigarette excise tariff is 57 percent of the characterized, however, by a wide range of cigarette corresponding retail price (harga jual eceran–HJE). prices, specific excise tariffs between and within Other neighboring countries such as the Philippines market segments, and excise taxes as a percentage of and Singapore also use specific excise taxes while cigarette retail prices (Table D.1). Overall, the present Myanmar and Vietnam levy ad valorem taxes, and multitiered specific system favors smaller producers Malaysia and Thailand apply a mixture of both excise and those producing hand-rolled kretek–imposing taxes. Unlike most other countries such as Australia, upon them lower excise tariffs per stick and lower South Africa, and Norway which use a single-tiered excise taxation in relation to the HJE. system, Indonesia applies a multitiered approach in which cigarettes are classified into several categories The present simplification road map, if implemented, (World Bank 2015a). The specific rates per cigarette would lead in the coming years to a relatively higher stick vary by type of product–kretek and white increase in tariffs per cigarette stick of those lower- cigarettes by machine and kretek cigarette by hand- priced products and, more importantly, to a simple roll–and by production levels and the retail price. The tax structure that will substantially lower the cost of majority of tobacco users are smokers, and the vast tax administration. majority of smokers (88 percent) use kretek. ASSESSING THE POTENTIAL FISCAL IMPACT OF In order to reduce this complexity and existing TOBACCO TAX REFORM IN INDONESIA opportunities for tax avoidance, the government has been implementing a roadmap of tax structure As an input for the Ministry of Finance, a World Bank simplification and gradually increasing lower tariffs in Group (WBG) team assessed the impact (2015) of a each segment of the cigarette market. The number of two-phase approach for reforming the 2015 tobacco 124 excise tax tiers decreased from 19 to 13 between 2009 tax structure, in terms of consumption reduction and and 2013 and to 12 tiers by 2015 and the differential potential fiscal revenue mobilization, while managing between the highest and the lowest excise tax rate the potential negative impact on employment: Table D.1 Cigarette Consumption by Tier (2014) Volume (billions Banderole price Consumption in 2014 Type of sticks) (HJE) (IDR per stick) (billions of sticks) (percentage of total) >= 2 800 and above 212 61.5 Machine-rolled kretek 588 and above 17 4.9 (SKM) <2 511-587 21 6.1 >= 2 820 and above 16 4.7 White cigarettes (SPM) 520 and above 2 0.6 <2 425-519 2 0.6 825 and above 13 3.8 >= 2 606-824 40 11.6 417 and above 5 1.4 Hand-rolled kretek (SKT) 0.3 - 2 385-416 5 1.4 0.05 - 0.3 286 and above 5 1.4 < 0.05 286 and above 7 2.0 TOTAL 345 100 Source Ministry of Finance; World Bank staff calculations 2015. APPENDIXES OPTIONS FOR PHASE ONE OF REFORM (2016-17) to the permitted legal limit of 57 percent for machine- made cigarettes. The increase in the average excise tax The first part of tobacco excise tax reforms, which is assumed to raise revenue while lowering tobacco could be implemented in 2016-17, focuses on consumption due to inelastic demand. reducing the number of excise tax tiers for machine- made cigarettes (SKM and SPM), while maintaining Two options are outlined below: the tax treatment of hand-made cigarettes (SKT). • • Combine existing excise tiers, with the This policy change would be consistent with highest excise tariff on the tiers being combined the government’s plans outlined in the “tobacco to apply to the remaining tiers; and roadmap”. The streamlining of tax tiers is to simplify • • In addition to the first option, there is room to administration and contribute to improved tax raise excise tariffs for the remaining tier(s) and compliance (by reducing the incentive for producers still stay within the legal limit. To significantly to manage production facilities to fall under a lower mobilize additional revenues, this would entail excise tax tier) and, therefore, revenue collection. raising the excise tax rate on the category of cigarettes with the largest market share: Two options were considered: machine-rolled kretek (SKM) with production • Combine SKM and SPM tiers so that there are volume of more than two billion sticks, which only two tiers for machine-made cigarettes, one accounted for 61.5 percent of total cigarette for a production volume of less than two billion volumes in 2014. The excise tariff for this sticks and one for a production volume of more category was calculated to be IDR 415 per stick, than two billion sticks; and which is 52 percent of HJE (IDR 800 per stick). • Combine all SKM and SPM tiers into one tier for machine-made cigarettes, regardless of The 2016-17 reform options were grouped into production volume. three scenarios for estimation of impact purposes 125 (although many more are possible) in Table D-2. The second part of the reform assessed was to increase If the government wants to retain an excise tax the average excise tax (which in 2015 is estimated to differentiation by production volume, that is two be a weighted average of 48 percent for all cigarettes) tiers for machine-made cigarettes as per Scenario Table D.2 Reform Scenarios (2016-17) Scenario Number of Tiers Excise Tariff (for machine made) and resulting tariff as a percentage of HJE Baseline Current (as in 2015) number of Current excise tariffs for machine-made tiers as in the 2015 tiers: 12 in total (3 SKM, 3 SPM, 6 regulation–ranging from IDR 220 to IDR 425 per stick. SKT). -> 52 percent of HJE for machine-made and weighted average of 48 percent of HJE for all cigarettes. Scenario 1 Two tiers for machine-made (SKM IDR 305 for tier with production of <2 billion sticks and IDR 425 and SPM combined): (1) for <2 per stick for tier with production of >=2 billion sticks. billion sticks; and (2) for >=2 billion -> 52-58 percent of HJE for machine-made and weighted sticks. average of 49 percent of HJE for all cigarettes. Scenario 2 Combine all SKM and SPM Impose a single tariff of IDR 425 per stick for all machine-made tiers into one tier for machine- cigarettes. made cigarettes, regardless of -> 52-63 percent of HJE for machine-made and weighted production volume. average of 49 percent of HJE for all cigarettes. Scenario 3 Two tiers for machine-made (SKM IDR 305 for tier with production of <2 billion sticks and an and SPM combined): (1) for <2 increased IDR 550 per stick for tier with production of >=2 billion sticks; and (2) for >=2 billion billion sticks, which, under the assumption of full pass-through, sticks. represents 57 percent of HJE. -> 52-58 percent of HJE for machine-made and weighted average of 52 percent of HJE for all cigarettes. HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better 1 and 3, it was recommended that the government those in 2014 (Table D-1) and that overall 2015 consider lowering the threshold distinguishing larger consumption, in number of sticks, will increase and smaller segments of production (in 2015 around by 1 percent, relative to 2014. two billion sticks of cigarettes) as this will increase • The baseline scenario for 2016 is based on the cost for producers to break their production into 2015 excise tariffs and corresponding retail smaller legal entities in order to take advantage of prices (which are assumed to be equal to lower excise taxes. corresponding banderole prices, HJE)–as in the 2015 regulation. This implies that the difference To ensure that the estimated revenue and public between the total 2016 baseline and 2015 health impacts of the reform are not reduced in consumption is only due to an estimated real the following year(s) due to inflation, an automatic GDP growth in 2016 of 5.5 percent. adjustment mechanism could be incorporated into • Weighted average price elasticity of -0.5 (Ahsan the tariff design and stated in a regulation. While et al. 2009) is used with price elasticity of -0.45 the inflation feedback loop is impossible to avoid, for machine-made and -0.75 for hand-made no matter how the retail prices are determined, it cigarettes. might be minimized by applying a properly designed • Weighted average income elasticity of 0.6 formula based on which excise tariffs and regulated (Ahsan et al. 2009) is used with income elasticity prices would be automatically adjusted annually of 0.65 for machine-made and 0.35 for hand- for inflation. Formula design should be carefully made cigarettes. explored–one possibility may be to incorporate the • No materially significant substitution effects previous 12-month average inflation rate and the between machine-made cigarettes and hand- Bank Indonesia inflation target with certain weights. rolled kretek cigarettes so that if machine-made cigarettes become relatively more expensive 126 from the reforms, we do NOT expect demand SIMULATIONS AND SUMMARY OF ESTIMATED for relatively cheaper hand-made cigarettes to IMPACT ON REVENUE AND TOBACCO increase. CONSUMPTION • Full pass-through of excise tax increases to the consumers. This means that for the simulation, Simulations of the expected impact on tobacco the banderole price would increase in consumption (in billions of sticks) and excise revenue accordance with any increase in excise tax and it (IDR trillion, excluding regional cigarette tax of 10 is assumed that the banderole price is equal to percent of the excise tariff and VAT of 8.4 percent of the retail price. Industry discussions indicate that the retail price) in 2016 for the baseline and reforms the banderole price is not necessarily binding– scenarios were done. that is, the retail price is, in reality, higher than the banderole price. The following key assumptions (in addition to those summarized in Table D.2) were made: The expected outcomes from the various simulations • For 2015, it is assumed that the market share are presented in Table D.3. of total consumption by tier will be equal to Table D.3 Simulation Results Consumption Change over Revenue Collection Change over Scenario (Billions of Sticks) Baseline (%) (Trillions of IDR) Baseline (%) Baseline 359 n.a. 126.4 n.a. Scenario 1 356 -0.8 128.7 +1.9 Scenario 2 350 -2.3 131.6 +4.1 Scenario 3 336 -6.4 147.2 +16.5 APPENDIXES OPTIONS FOR PHASE 2 OF REFORM and thereby influencing the behavior of consumers (2017 ONWARDS) within a country. Indeed, while tax policy can help reduce negative externalities associated with The second phase of the reforms, which could tobacco consumption, the taxation model needs be implemented starting 2017, would continue to to avoid providing incentives to switch down to implement the tobacco roadmap by simplifying the cheaper cigarette brands in response to tax-related tax treatment of hand-made kretek cigarettes. This and other price increases. Furthermore, consumers’ could potentially be accompanied by compensatory price sensitivity and brand-switching behaviour, measures for tobacco factory workers and tobacco manufacturers’ pricing strategy including brand and clove farmers. repositioning, differential tax shifting, and cross-brand price subsidy, can have potential consequences on tax As recently proposed by WHO (2016), one option to revenue collection at an aggregate level. To this end, advance with the tobacco taxation reform agenda in as advised by the IMF, it is of critical importance that Indonesia would be to move from 12 to four tiers in Indonesia should adopt a simpler structure of taxation three years, therefore accelerating the simplification to have a greater influence on the relative prices of process while at the same time increasing rates and different tobacco products across the price bands. generating more revenues with no major shocks to the industry as a whole. IMPACT ON EMPLOYMENT IN THE TOBACCO To this end, the following simplification steps are INDUSTRY suggested: • 2017: merging SKM and SPM creating the group In Indonesia, it is reported that less than 1 percent of SKM/SPM with a reduction in number of tiers arable land is used for tobacco cultivation, and that from 12 to nine. These already have similar rates most tobacco-growing farmers do not depend solely 127 per tier (keeping tiers I, IIA and IIB). on tobacco cultivation to make a living (American • 2018: removing the price tiers for all groups and Cancer Society and World Lung Foundation 2012). merging IIIA and IIIB, keeping tiers by type of Farmers typically cultivate a combination of main cigarette and production volume. The number crops, including soybeans, corn, tobacco, cassava, of tiers falls from nine to five. peanuts, rice, fruits, and vegetables, to minimize • 2019: merging SKT I with SKM/SPM II – there is risks. Given this situation, tobacco farmers may not a fall in the number of tiers from five to four.. be significantly impacted by the increase in tobacco excise tariff. Moreover, excess production of tobacco Under the WHO proposal, an additional two years leaves may potentially be exported. On the other would be needed to finalize the simplification to two hand, it is not clear if a similar situation applies for tiers: clove-growing farmers. For hand-rolled kretek • 2020: merging SKT II and III, keeping a factory workers, it is not yet clear what the alternative substantially lower rate for this group compared livelihoods for these workers are, and if they will to SKM/SPM group–the number of tiers need support, such as training, to transition to other changes from four to three. livelihoods. • 2021: merging production tiers in SKM/SPM I with SKM/SPM II and SKT I, the number of tiers The World Bank is conducting, over the period from changes from three to two. 2016 to mid-2017, tobacco industry employment studies to inform the policy recommendations on The consolidation scenarios proposed for Indonesia reforms to the hand-rolled kretek cigarette segment take into account recent findings documented in an and options for compensatory measures to workers. IMF assessment done in Pakistan (Cevik 2016) that The importance of this analytical work followed the show that the structure of cigarette taxes is critical in President of Indonesia’s statement in June 2016 on determining the relative prices of different tobacco the country’s intention to ratify the FCTC. The tobacco products and brands across the price spectrum industry employment study is expected to be ready by mid-2017. HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better TOBACCO TAX INCREASES AND THE RISK OF ILLICIT TRADE The tobacco industry and other vested interests argue that tax increases on tobacco products fuel illicit trade. Accumulated international experience, however, exposes the flaws in this argument. While high taxes may create incentives for illicit trade, different country experiences show that illicit trade can be controlled by legal means and by increased law enforcement, controls over the distribution chain, improved technologies, and better use of data help to reduce illicit trade and complement tobacco tax reforms (Marquez 2015a).83 THE WAY FORWARD Tobacco taxation reform, including the drastic reduction in tax tiers for different categories of cigarettes, is a major potential policy tool for the Government of Indonesia to use to reduce the severe public health burden of smoking-related disease and 128 premature mortality in Indonesia. Tobacco taxation can also be a significant contributor to state revenue collection for expanding the fiscal space to support UHC and other essential investments that benefit all. Given fiscal pressures and of unmet health needs of the population in Indonesia, now is a particularly relevant time to focus on using tobacco taxation increases as a source to raise public revenue over the near and medium term. 83 See also WHO 2013 and van Walbeek et al. 2013. APPENDIXES 129 HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better References Abhimanyu, N., S.P Zodpey, S. Ughade, and S.I. BMI Research. 2015. “Indonesia Pharmaceuticals and Bangdiwala. 2011. “Neonatal morbidity and Healthcare Report (Q3).” London: BMI Research. mortality in tribal and rural communities in central Centers for Disease Control and Prevention (CDC). India.” Indian Journal of Community Medicine 36(2): 2016. “Secondhand Smoke (SHS) Facts.” http:// 150-158. www.cdc.gov/tobacco/data_statistics/fact_ sheets/secondhand_smoke/general_facts/ Ahsan, A., S. Barber, S.M. Adioetomo, and D. Setyonaluri. 2009. “Economics of Tobacco Taxation Cevik, S. 2016. “Smoke Screen: Estimating the Tax in Indonesia.” Demographic Institute, Faculty of Pass-Through to Cigarette Prices in Pakistan.” IMF Economics, University of Indonesia, Indonesia. Working Paper WP/16/179. Washington DC: IMF. American Cancer Society and World Lung Chisholm, D., and D.B. Evans. “Improving health system Foundation. 2012. “Tobacco Atlas.” Fourth Edition. efficiency as a means of moving towards universal 130 coverage.” World Health Report (2010), Background Asia Pacific Observatory and WHO. 2015. “Health Paper No. 28. Geneva: World Health Organization. System in Transition, Country Report: Indonesia.” Final draft, unpublished. Country Team, Asia Pacific Cotlear, D., S. Nagpal, O. Smith, A. Tandon, and R. Observatory and WHO. Cortez. 2015. “Going Universal: How 24 Developing Countries Are Implementing Universal Health Badan POM. 2016. “Badan POM Terus Perangi Coverage Reforms from the Bottom Up.” Peredaran Obat Ilegal di Indonesia”. http://www. Washington DC: World Bank. pom.go.id/new/index.php/view/pers/316/ BADAN-POM-TERUS-PERANGI-PEREDARAN- Fenochietto, R., and C. Pessino. 2013. “Understanding OBAT-ILEGAL-DI-INDONESIA.html Countries’ Tax Effort.” IMF Working Paper WP/13/244. Washington DC: International Barber, S., S.M. Adioetomo, A. Ahsan, and D. Monetary Fund. Setyonaluri. 2008. “Tobacco Economics in Indonesia.” Paris: International Union Against Gibbons, L., J.M. Belizán, J.A. Lauer, A.P. Betrán, Tuberculosis and Lung Disease. M. Merialdi and F. Althabe. 2010. “The Global Numbers and Costs of Additionally Needed and Bitran, R. 2013. “Explicit Health Guarantees for Unnecessary Caesarean Sections Performed per Chileans: The AUGE Benefits Package.” UNICO Year: Overuse as a Barrier to Universal Coverage.” Studies Series 21. Washington DC: World Bank. Background Paper. Geneva: WHO. Bitran, R. 2014. “Universal Health Coverage and the Guerard, Y., M. Wiener, C. Rokx, G. Schieber, P. Challenge of Informal Employment: Lessons from Harimurti, E. Pambudi, and A. Tandon. 2011. Developing Countries.” HNP Discussion Paper. “Actuarial Costing of Universal Health Insurance Washington DC: World Bank. Coverage in Indonesia.” HNP Discussion Paper. Washington DC: World Bank. Hadi, U., van den Broek, P., Kolopaking, E. P., Zairina, Jakarta Globe. 2016. “Counterfeit Drugs Worth Rp N., Gardjito, W., & Gyssens, I. C. 2010. Cross- 56 Billion Seized in Special Operation”. http:// sectional study of availability and pharmaceutical jakartaglobe.beritasatu.com/news/counterfeit- quality of antibiotics requested with or without drugs-worth-rp-65b-seized-special-operation prescription (Over The Counter) in Surabaya, Indonesia. BMC Infectious Diseases, 10, 203 Jha, P., R. Joseph, D. Li, C. Gauvreau C, I. Anderson, P. Moser, S. Bonu, I. Bhushan, and F.J. Chaloupka. Hadi U et al. 2008, Audit of antibiotic prescribing in 2012. “Tobacco Taxes: A Win-Win Measure for Fiscal two governmental teaching hospitals in Indonesia. Space and Health.” Mandaluyong City, Philippines: Clinical Microbiology and Infection, Volume 14 , Asian Development Bank. https:/ /www.adb.org/ Issue 7 , 698 - 707. sites/default/files/publication/30046/tobacco- taxes-health-matters.pdf Hadisoemarto, P.F., M.R. Reich, and M.C. Castro. 2016. “Introduction of pentavalent vaccine in Indonesia: a Jha, P., and R. Peto. 2014. “Global Effects of Smoking, policy analysis.” Health Policy and Planning. of Quitting, and of Taxing Tobacco.” New England Journal of Medicine 370: 60-68. http:/ /www.nejm. Harimurti, P., E. Pambudi, A. Pigazzini, and A. Tandon. org/doi/full/10.1056/NEJMra1308383. 2013. “The Nuts & Bolts of Jamkesmas: Indonesia’s Government-Financed Health Coverage Program.” USAID, JHPIEGO, and Indonesia OBGYN Association. UNICO Studies Series No. 8. Washington DC: 2015, “Retrospective Review of Factors Associated World Bank. with 112 Maternal Deaths in 12 Hospitals”, 2015, presented at OBGYN Congress, Bandung. Hidayat, B. 2016. “INA-CBG’s claim patterns: Is there any indication of fraudulent practices?” Kaiser, K., C. Bredenkamp, and R. Iglesias. 2016. 131 Presentation, August 25, 2016. Jakarta. “Sin Tax Reform in the Philippines: Transforming Public Finance, Health, and Governance for Hidayat, B., Mundiharno, J. Nemec, V. Rabovskaja, More Inclusive Development. Directions in C.S. Rozanna, and J. Spatz. 2015a. “Financial Development--Countries and Regions. Washington Sustainability of the National Health Insurance DC: World Bank. https:/ /openknowledge. in Indonesia: A First Year Review.” National Social worldbank.org/handle/10986/24617 Security Council, GIZ, Policy Brief. Lustig, N. 2015. “Inequality and Fiscal Redistribution in ———. 2015b. “Out-of-Pocket Payments in the Middle Income Countries: Brazil, Chile, Colombia, National Health Insurance of Indonesia: A First Year Indonesia, Mexico, Peru, and South Africa.” Review.” Policy Brief. Jakarta. Commitment to Equity Working Paper No. 31. New Orleans: Tulane University. Hollingsworth, B. and S.J. Peacock. 2008. “Efficiency Measurement in Health and Health Care.” Maharani, A. and G. Tampubolon. 2014. “Has Routledge International Studies in Health Decentralization Affected Child Immunization Economics. New York: Routledge. Status in Indonesia?” Global Health Action 7: 24913. Hsiao, W.C. 2007. “Why Is A Systematic View Of Health Marquez, P.V. 2015a. “World No Tobacco Day 2015: On Financing Necessary?” Health Affairs 26(4): 950-961. illicit trade and taxes.” http://blogs.worldbank.org/ health/world-no-tobacco-day-2015-illicit-trade- Institute for Health Metrics and Evaluation. and-taxes “Indonesia.” http://www.healthdata.org/indonesia. ———. 2015b. “Making the public health case for Institute for Health Metrics and Evaluation (IHME). tobacco taxation.” http://blogs.worldbank.org/ 2013. “Global Burden of Disease. Country Profile health/making-public-health-case-tobacco- for Indonesia.” Available at http://www.healthdata. taxation. org/indonesia HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better ———. 2016a. “Economic slowdown and financial Ministry of Health, 2015. “Profil Kesehatan Indonesia shocks: Can tobacco tax increases help?” 2014.” http://blogs.worldbank.org/voices/economic- slowdown-and-financial-shocks-can-tobacco- Ministry of Health and Social Welfare (Republic of tax-increases-help Liberia). 2008. “National Health and Social Welfare Policy.” ———. 2016b. “Plain packaging & tobacco taxes: an antidote for manipulation and deception.” http:// Missoni, E. and G. Solimano. 2010. “Towards Universal blogs.worldbank.org/health/plain-packaging- Health Coverage: the Chilean Experience.” World tobacco-taxes-antidote-manipulation-and- Health Report Background Paper 4. Geneva: WHO. deception Mulholland, E.K., L. Smith, I. Carneiro, H. Becher, Millennium Challenge Account - Indonesia. 2015. and D. Lehmann. 2008. “Equity and child- “Stunting and the Future of Indonesia.” Jakarta: survival strategies.” Bulletin of the World Health MCA Indonesia. Organization 86(5): 321-416. Ministry of Health (Government of Indonesia). Nader, A.A., C. de Quadros, C. Politi, and M. 2011. “Indonesia Human Resources for Health McQuestion. 2015. “An analysis of government Development Plan Year 2011-2025.” Jakarta: immunization program expenditures in lower and Ministry of Health. lower middle income countries 2006–12.” Health Policy and Planning. 30(3): 281-288. ———. 2013. “Annual Gavi Program Progress Report 2013.” National Institute for Health Research and Development 132 (MoH). 2013. Riset Kesehatan Dasar - Riskesdas. Ministry of Health (Center for Health Economic Policy Studies), AIPHSS. 2015. “Indonesia National Organisation for Economic Co-operation and Health Account 2014, Final Report.” Jakarta: Development (OECD). 2011. “Burden of out-of- Australia Indonesia Partnership Health System pocket health expenditure.” Health at a Glance Strengthening. 2011. OECD Indicators. Paris: OECD. Ministry of Health and World Bank. 2014a. “Supply- Pisani, E., M. Kok, and K. Nugroho. 2016. “Indonesia’s Side Readiness for Universal Health Coverage: road to universal health coverage: a political Assessing the Depth of Coverage for Non- journey.” Health Policy and Planning (forthcoming). Communicable Diseases in Indonesia.” Jakarta: World Bank, Ministry of Health. Pradiptyo, R. 2015. “Error, Fraud and Corruption Prevention to Strengthen Financial Governance ———. 2014b. “Universal Maternal Health Coverage? and Management in National Healthcare Security.” Assessing the Readiness of Public Health Presentation at Otoritas Jasa Keuangan Conference Facilities to Provide Maternal Health Care in in Bali. Department of Economics, Faculty of Indonesia.” Jakarta: World Bank. Economics and Business, Universitas Gadjah Mada. Ministry of Health and World Health Organization. Rokx, C., G. Schieber, P. Harimurti, A. Tandon, and A. 2012. “Global Adult Tobacco Survey: Indonesia Somanathan. 2009. “Health Financing in Indonesia: Report 2011.” Jakarta: National Institute of A Reform Road Map.” Washington DC: World Bank. Health Research and Development, Ministry of Health, Republic of Indonesia and New Delhi: Rokx, C., J. Giles, E. Satriawan, P. Marzoeki, P. World Health Organization, Regional Office for Harimurti, and E. Yavuz. 2010. “New Insights into the South East Asia. http://www.who.int/tobacco/ Provision of Health Services in Indonesia: A Health surveillance/survey/gats/indonesia_report.pdf Workforce Study.” Washington DC: World Bank. Puspitasari, H. P., Faturrohmah, A. and Hermansyah, A. Suparmi. 2014. “Antenatal and neonatal visits increase 2011, Do Indonesian community pharmacy workers complete immunization status among children respond to antibiotics requests appropriately?. aged 12-23 months in rural area of Indonesia.” Tropical Medicine & International Health, 16: Health Science Journal of Indonesia 5(2): 73-77. 840–846. Suwantika, A.A. and M.J. Postma. 2014. “Expanding Savedoff, W.D. 2003. “How Much Should Countries access to non-traditional vaccines: a perspective Spend on Health?” Discussion Paper No. 2, Geneva: from Indonesia.” Expert Review of Vaccines 13(12): World Health Organization. 1419-1421. Savedoff, W.D., R. Bitrán, D. de Ferranti, V.Y Fan, Ujoh, F., and F. Kwaghsende. “Analysis of the Spatial A. Holly, R. Moreno-Serra, P. Saksena, A.L. Distribution of Health Facilities in Benue State, Smith, P.C. Smith and K. Xu. 2012. “Transitions Nigeria.” Public Health Research 4(5): 210-218. in Health Financing and Policies: Final Report of the Transitions in Health Financing Project.” UNICEF/WHO/World Bank/UN. 2014. “Levels & Washington DC: Results for Development Institute. Trends in Child Mortality: Estimates Developed by the UN Inter-Agency Group for Child Mortality Savedoff, W.D., and A. Alwang. 2015. “The Single Estimation.” New York: UNICEF. Best Health Policy in the World: Tobacco Taxes.” CGD Policy Paper 062. Washington DC: Center United Nations. 2003. “Indicators for Monitoring for Global Development. http:/ /www.cgdev.org/ the Millennium Development Goals. Definitions, publication/single-best-health-policy-world- Rationale, Concepts and Sources.” New York: tobacco-taxes; United Nations. 133 Saxenian, H., R. Hecht, M. Kaddar, S. Schmitt, T. United Nations (Department of Economic and Ryckman, and S. Cornejo. 2014. “Overcoming Social Affairs, Population Division). 2015a. “World challenges to sustainable immunization financing: Population Prospects: The 2015 Revision, Volume I: early experiences from Gavi graduating countries.” Comprehensive Tables.” New York: United Nations. Health Policy and Planning 30(2): 197-205. United Nations. 2015b. “Addis Ababa Action Agenda Schoeps, A., S. Gabrysch, L. Niamba, A. Sié, and H. of the Third International Conference on Financing Becher. 2011. “The Effect of Distance to Health- for Development.” http:/ /www.un.org/esa/ffd/wp- Care Facilities on Childhood Mortality in Rural content/uploads/2015/08/AAAA_Outcome.pdf Burkina Faso.” American Journal of Epidemiology 173(5): 492-98. van Walbeek, C., E. Blecher, A. Gilmore, and H. Ross. 2013. “Price and Tax Measures and Illicit Trade in Siallagan, T. 2015. Challenges in Implementing the Framework Convention on Tobacco Control: Indonesia Case-based Groups (INA-CBGs). BPJS What We Know and What Research Is Required.” Presentation at Otoritas Jasa Keuangan (Financial Oxford Journals: Nicotine & Tobacco Research Service Authority) Conference, Bali, 7-8 September 15(4):767–76. 7-8, 2015. Wagstaff, A. 2010. “Social health insurance Soenarto, Y., A.T. Aman, A. Bakri, H. Waluya, A. reexamined.” Health Economics 19(5): 503-517. Firmansyah, M. Kadim, I. Martiza, D. Prasetyo, N.S. Mulyani, T. Widowati, Soetjiningsih, I.P Karyana, WHO and World Bank. 2015. “Tracking Universal W. Sukardi, J. Bresee, and M.A. Widdowson. 2009. Health Coverage: First Global Monitoring Report.” “Burden of severe rotavirus diarrhea in Indonesia.” Geneva: World Health Organization. Journal of Infectious Diseases Supplement 1, S188-S194. HEALTH FINANCING SYSTEM ASSESSMENT spend more . spend right . spend better WHO (World Health Organization). 2000. “The World ———. 2015. “WHO report on the global Health Report 2000: Health Systems: Improving tobacco epidemic, 2015: raising taxes Performance.” Geneva: World Health Organization. on tobacco.” http://apps.who.int/iris/ bitstream/10665/178574/1/9789240694606_eng. ———. 2003. “WHO Framework Convention pdf?ua=1 on Tobacco Control.” Geneva: World Health Organization. http://apps.who.int/iris/ ———. 2016. “Tobacco Excise Taxation in Indonesia.” bitstream/10665/42811/1/9241591013.pdf Technical brief note shared by The World Health Organization based on bilateral discussions held ———. 2010a. “Monitoring the Building Blocks of with the Ministry of Finance, Centre for Fiscal Health Systems: A Handbook of Indicators and Policy And Department of Customs and Excise, Their Measurement Strategies.” Geneva: World March 11, 2016. Health Organization. Widayati, A., Suryawati, S., de Crespigny, C., & Hiller, J. E. ———. 2010b. “WHO Technical Manual on Tobacco (2011). Self medication with antibiotics in Yogyakarta Tax Administration.” Geneva: World Health City Indonesia: a cross sectional population-based Organization. http://www.who.int/tobacco/ survey. BMC Research Notes, 4, 491. publications/tax_administration/en/index.html World Bank. 1999. “Curbing the Epidemic: ———. 2010c. World Health Report: Health Systems Governments and the Economics of Tobacco Financing–The Path to Universal Coverage. Control.” Washington DC: The World Bank. Geneva: World Health Organization ———. 2012a. “Indonesia Public Expenditure Review: 134 ———. 2011. “Effectiveness of Tax and Price Policies Optimizing Sub-National Performance for Better for Tobacco ControI.” IARC Handbooks of Cancer Services and Faster Growth.” Jakarta: World Bank. Prevention: Tobacco Control. Volume 14. Lyon, France: International Agency for Research on ———. 2012b. “Indonesia: Repeat Public Expenditure Cancer. http://www.iarc.fr/en/publications/pdfs- and Financial Accountability (PEFA) Report & online/prev/handbook14/handbook14.pdf Performance Indicators.” Jakarta: World Bank. ———. 2013. “Protocol to Eliminate Illicit Trade ———. 2012c. “Targeting Poor and Vulnerable in Tobacco Products.” Geneva: World Health Households in Indonesia.” Jakarta: World Bank. Organization. http://apps.who.int/iris/ bitstream/10665/80873/1/9789241505246_eng. ———. 2012d. “Indonesia Economic Quarterly – pdf December 2012: Policies in Focus.” Jakarta: World Bank. ———. 2013a. “The World Health Report 2013: Research for Universal Health Coverage.” Geneva: ———. 2013. “Spending more or spending better: WHO. Improving education financing in Indonesia.” Jakarta: World Bank. ———. 2013b. “Service Availability and Readiness Assessment (SARA). An annual monitoring system ———. 2014a. “Indonesia Economic Quarterly – for service delivery. Reference Manual.” Geneva: December 2014: Delivering Change.” Jakarta: World World Health Organization. Bank. ———. 2014. “Q&As: Health Systems.” http://www.who. ———. 2014b. “The Production, Distribution, and int/topics/health_systems/qa/en/. Performance of Physicians, Nurses, and Midwives in Indonesia: An Update.” Jakarta: World Bank. ———. 2015a. “Indonesia tobacco excise tax reform: Evaluating potential revenue and public health impact of a reform proposal to streamline the excise tax structure and increase the average excise tax rate for machine-made cigarettes: Policy Note.” Jakarta: World Bank. ———. 2015b. “Country Partnership Framework for the Republic of Indonesia for the period FY16 - FY20.” Washington DC: World Bank. ———. 2015c. “Indonesia Systematic Country Diagnostic: Connecting the Bottom 40 percent to the Prosperity Generation.” Washington DC: World Bank Group. ———. 2015d. “In Times of Volatility.” Indonesia Economic Quarterly October 2015. Jakarta: World Bank. ———. 2015e. “Reforming amid uncertainty.” Indonesia Economic Quarterly December 2015. Jakarta: World Bank. 135 ———. 2016a. “Financing: Where does funding for the Health System come from? How are the funds pooled? How are they allocated and spent?” http:// go.worldbank.org/I9NCO1V9N0. ———. 2016b. “Indonesia’s Rising Divide.” Jakarta: World Bank. Xu, X., E.E. Bishop, S.M. Kennedy, S.A. Simpson, and T.F. Pechacek. 2015. “Annual Healthcare Spending Attributable to Cigarette Smoking: An Update.” American Journal of Preventive Medicine, 48:3: 326–333.