Improving Service Levels and Impact on the Poor A Diagnostic of Water Supply, Sanitation, Hygiene, and Poverty in Indonesia INDONESIA This work was financed by the World Bank Water and Sanitation Program and the Swedish International Development Cooperation Agency and was a multi-Global Practice initiative led by Water and Poverty with significant support from Governance and Health, Nutrition, and Population. Improving Service Levels and Impact on the Poor A Diagnostic of Water Supply, Sanitation, Hygiene, and Poverty in Indonesia © 2017 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington, DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Please cite the work as follows: World Bank. 2017. Improving Service Levels and Impact on the Poor: A  Diagnostic of Water Supply, Sanitation, Hygiene, and Poverty in Indonesia. WASH Poverty Diagnostic. World Bank, Washington, DC. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522- 2625; e-mail: pubrights​@­worldbank.org. Cover design: Bill Pragluski, Critical Stages LLC. Contents Acknowledgments and Dedication vii Executive Summary ix Abbreviations xxi Chapter 1  Introduction and Background 1 A Global Mandate for Quality, Sustainability, and Equitable Distribution of Services 1 Overview of the Water Supply and Sanitation Sector in Indonesia 2 Approach and Methodology of the WASH Poverty Diagnostic 4 Roadmap of the Report 7 Notes 8 Chapter 2  Rising Inequality and Why It Matters 9 The Evolution of Poverty and Inequality in Indonesia 9 The Source of Inequality and Implications for Service Delivery 13 Notes 14 References 14 Chapter 3  The Role of WASH in Human Development 15 Pathways of Exposure and Impacts on Health and Nutrition 15 Growing Tall and Smart with WASH 17 The Stubborn Problem of Child Stunting in Indonesia and Linkages with WASH 19 Not by WASH Alone: How Multisectoral Interventions Help Improve Nutrition 23 Notes 25 References 25 Chapter 4  WASH Service Delivery Progress and Challenges in Indonesia 29 Stopping Open Defecation in Indonesia: A Global Success Story 30 Location Matters: Subnational Estimates of Sanitation Coverage and Associations with Poverty 32 Second Generation Challenges for Rural Indonesia: Moving up the “Sanitation Ladder” 37 Why High Levels of Sanitation in Urban Settings Have Not Fully Eliminated Fecal Contamination 38 The Evolution of Drinking Water Supply in Indonesia 39 Drinking Water Access and the Poor 41 Location Matters: Subnational Estimates of Access to Water Supply and Associations with Poverty 43 Water Quality: Little Information on a Potentially Widespread Problem 46 Notes 48 References 49 Chapter 5 Urban Water Service Delivery Constraints on and Opportunities for Reaching the Poor 51 Why Urban Water? 52 Urban Water Institutional and Service Delivery Context 53 Constraints On and Opportunities for Connecting the Poor to Urban Water Services 55 Notes 65 References 66 Improving Service Levels and Impact on the Poor iii Chapter 6 Priorities for Future Policy and Investments in Water Supply and Sanitation 67 Key Recommended Actions 69 Notes 77 Reference 77 Appendix A  Interactive Dashboard 79 Indonesia Water Supply, Sanitation, and Hygiene Poverty Diagnostics 79 Boxes Box 1.1: Defining Access to Improved Drinking Water 6 Box 3.1: Water Supply, Sanitation, and Hygiene and their Effect on Maternal Health 17 Box 4.1: Increased Bottled Water Usage Distorts Trends in Access to Piped Water Supply: Data and Definitions 40 Box 5.1: The Benefits of a Household Piped Water Connection 63 Box 6.1: Access to Water Supply and Sanitation under the 2019 Universal Health Coverage Target for Indonesia 72 Box 6.2: Strengthening Nutrition-Sensitive Actions in PAMSIMAS to Reduce Stunting in Children under 5 73 Figures Figure 2.1: Trend in Poverty Reduction 2002–15 10 Figure 2.2: Trends in Inequality, Households Per Capita Consumption, 1980–2015 11 Figure 2.3: Change in Inequality, Gini Coefficient 12 Figure 2.4: Children in Rural Areas Lack Access to Multiple Services 13 Figure 3.1: F-Diagram Showing How Inadequate WASH Affects Child Health and Nutrition 16 Figure 3.2: Pathways between Poor WASH and Early Child Development 18 Figure 3.3: Stunting Rates, by Wealth Quintile, 2007–13 20 Figure 3.4: Relationship between ODF and Height-for-Age and ODF and Cognitive Scores 22 Figure 3.5: Difference in Probability of Being Stunted, 2013 24 Figure 4.1: Access to Sanitation and Drinking Water, by Urban and Rural Wealth Quintile 31 Figure 4.2: Correlation between Poverty Rate and Improved Sanitation Access in Kabupatens/Districts, 2002–15 33 Figure 4.3: Share of T60 and B40 with Improved Sanitation and Poverty Rate, by District (Kabupaten) 34 Figure 4.4: Share of T60 and B40 with Improved Sanitation and Poverty Rate, by District (Kabupaten) in Java and Non-Java 35 Figure 4.5: Access to Improved Sanitation in Urban and Rural Areas, 2002–13 36 Figure 4.6: Percentage of B40 and T60 Population Practicing Open Defecation in Urban and Rural Areas 36 Figure 4.7: Trend in Improved Sanitation Access in Indonesia’s Largest Cities 37 Figure 4.8: Growing Inequality in Improved Sanitation Access in Rural Settings for B40 versus T60 38 Figure 4.9: Fecal Waste Management Flow Chart 39 Figure 4.10: Results from Study of On-Site Sanitation for Dense Urban Areas in Indonesia 40 iv Improving Service Levels and Impact on the Poor Figure 4.11: Access to Improved Drinking Water in Urban and Rural Areas, 2002–13 41 Figure 4.12: Inequalities in Access to Improved Drinking Water by Income Distribution 42 Figure 4.13: PAMSIMAS Water Supply Beneficiaries, T60–B40, Piped– Non-Piped, 2012–15 42 Figure 4.14: Change in Primary Source of Drinking Water by Consumption Quintile, 2002–15 43 Figure 4.15: Drinking Water Access in Rural and Urban Settings for B40 versus T60, 2002–15 44 Figure 4.16: Share of T60 and B40 with Piped Water Access (Primary and Secondary) and Poverty Rate in Urban Districts (Kota) 44 Figure 4.17: Share of T60 and B40 with Piped Water Access (Primary and Secondary) by Hibah and Non-Hibah Participating Urban Districts (Kota) 45 Figure 4.18: Piped Water Access (Primary and Secondary Source) and Population Growth Rate, by City Category, 2011–15 46 Figure 4.19: Piped Water Access (Primary and Secondary Source) for B40 versus T60, by City Category, 2011–15 47 Figure 4.20: Households Using Groundwater as Primary Drinking Water Source and Distance to Feces Containment B40 versus T60 in Urban Areas 48 Figure 5.1: Access to Piped Water by Income Quintile (Q1–Q5) in Urban Areas 52 Figure 5.2: Overview of Government of Indonesia Actors Involved in Urban Water Service Delivery 54 Figure 5.3: Central Government Patterns of Spending on Water Supply and Sanitation, 2001–13 56 Figure 5.4: Subnational Government Patterns of Spending on Water Supply and Sanitation, 2001–13 57 Figure 5.5: Capital/Investment Mapping 57 Figure 5.6: DAK Water Allocation per Capita in 2015 and 2017 versus Water Access 58 Figure 5.7: Top 10 Percent versus Bottom 10 Percent Performance Average (Unweighted) 62 Figure 6.1: Government of Indonesia 2019 Universal Access Targets and Service Delivery Platforms 68 Figure A.1: Panels for Selecting the Variables 80 Maps Map 2.1: Poverty Rate and Poverty Headcount, by Geographic Location, 2015 11 Map 3.1: Stunting Rates in Indonesia, 2013 19 Map 3.2: Changes in District Stunting Rates, 2007–13 20 Map 4.1: Poverty Rate and Access to Improved Sanitation, by Province, 2015 32 Map A.1: Double Map Province: Access to Improved Sanitation and Stunting, 2013 80 Map A.2: Double Map Districts: Access to Improved Water and Access to Piped Water, 2015 81 Map A.3: Single Map District: Poverty Rate and Access to Improved Water 82 Photo Photo 3.1: Malnourished and Nourished Gut Lining 18 Improving Service Levels and Impact on the Poor v Tables Table ES.1: Key Recommended Actions and Responsible Agencies xvi Table 1.1: Overview of Water Supply and Sanitation Sector in Indonesia 2 Table 5.1: Central Government Water Supply Spending as Percentage of Total Spending/Infrastructure Spending 55 Table 5.2: Summary of PDAM Performance Indicators, 2013–15 61 Table 6.1: Key Recommended Actions and Responsible Agencies 75 vi Improving Service Levels and Impact on the Poor Acknowledgments and Dedication The WASH Poverty Diagnostic for Indonesia was led by Claire Chase (Economist, Water Global Practice) and Maraita Listyasari (Water and Sanitation Specialist, Water Global Practice). The core team includes Bambang Suharnoko (Economist, Poverty Global Practice), Imam Setiawan (Research Analyst, Poverty Global Practice), Hendra Murtidjaja (Consultant), William Gunawan (Consultant), and Priyanka Verma (Consultant). Background papers were prepared by Emmanuel Skoufias (Lead Economist, Poverty Global Practice), Katja Vinha (Consultant), Rebekah Kathryn Pinto (Consultant), Lisa Cameron (Consultant), Daniel Harris (Research Associate, Politics and Governance Programme, ODI) and Nathaniel Mason (Senior Research Fellow, Water Policy Programme, Overseas Development Institute, ODI). Aroha Bahuguna (Operations Analyst) provided analysis on utility performance benchmarking. The team is grateful for feedback and discussion with Kathleen Anne Whimp (Lead Public Sector Management Specialist), Shomikho Raha (Senior Public Sector Specialist) and Cut Dian R. D. Agustina (Economist) on Governance and Fiscal issues, with Lilian Pena Pereira Weiss (Senior Water Supply and Sanitation Specialist) on Water sector issues, and with Vikram Sundara Rajan (Senior Health Specialist), Ali Subandoro (Nutrition Specialist), and Sitaramachandra Machiraju (Senior Water and Sanitation Specialist) on child health and nutrition issues. The team received guidance and advice from peer reviewers Matthew Grant Wai-Poi (Senior Economist), Fook Chuan Eng (Lead Water and Sanitation Specialist), Ajay Tandon (Lead Economist), and Jose Antonio Cuesta (Senior Economist). Additional guidance was received from the WASH Poverty Diagnostic global team members including Craig Kullmann (Senior Water and Sanitation Specialist), Luis Andres (Lead Economist), Vivek Srivastava (Lead Public Sector Development Specialist), and Elizabeth Clementine Loughnan (Consultant). We also thank Cristobal Ridao-Cano, Jyoti Shukla, Almud Weitz, Sudipto Sarkar, Taimur Samad, Salman Zaidi, and Ana Revenga for their feedback. Editorial support was provided by Molly McCloskey. The World Bank and the Task Team greatly appreciate the collaboration with the Government of Indonesia, especially the National Development Planning Agency (Bappenas) and Ministry of Public Works and Housing in the preparation of this report. An advanced draft of the WASH Poverty Diagnostic for Indonesia was presented and discussed with government, donor partners, and civil society representatives during a consultation workshop on June 7, 2017, in Jakarta. This report is dedicated to an inspiring Government official, Bapak Nugroho Tri Utomo (former Director of Urban, Housing and Settlement, National Development Planning Agency—Bappenas), who passed away on June 24, 2016, during the preparation of this analytical work. Pak Nugroho exhibited a strong passion and commitment to the water and sanitation sector in Indonesia, and was a tireless advocate in helping to ensure that every single household in Indonesia gains access to a safe drinking water supply and dignified sanitation. He is, and will be, sadly missed. May his soul rest in peace. Improving Service Levels and Impact on the Poor vii Executive Summary Inequality is on the rise in Indonesia. Although the country made significant progress in reducing poverty from 24 percent, at the time of the Asian financial crisis in 1997, to 11 percent in 2014, and maintained 6 percent annual growth for a decade up to 2015, consumption growth has not been evenly distributed across the population. The poorest 40 percent of Indonesians now account for just a fifth of total household consumption, while the richest 20 percent account for nearly half. This places Indonesia among the countries with the highest levels of inequality in East Asia—just below Malaysia, the Philippines, and China—and above the average of five Organization for Economic Co-operation and Development (OECD) countries with the highest levels of inequality. Indonesia is undergoing rapid urbanization, and although this can be accompanied by strong economic growth, it creates a number of challenges, including disparities in income and access to services. The urban population accounts for about half of the country’s total population, a figure estimated to rise to approximately 68 percent by 2025. Underinvestment in urban infrastructure and lack of adequate planning limits the potential economic growth and development benefits of growing cities and contributes to widening inequalities. Over the past decade, for every 1 percent increase in urbanization, Indonesia achieved only two percent gross domestic product (GDP) growth, below the return on urbanization in other Asian countries such as China, Vietnam, and Thailand, which have significantly benefited from economies of agglomeration. Unequal access to services at the beginning of life is a key driver of inequality. Children who Unequal access are born into poverty are more likely to be deprived of critical services such as health care, to services at the nutrition, education, water supply, and sanitation—placing them at an unfair disadvantage from beginning of life the outset. When accessible, these services help level the playing field for the next generation makes it more difficult to break by providing the basic conditions that allow children and adults to lead healthier, and more out of poverty educated and productive lives. Effective service delivery is essential to the future well-being of later in life. society, and is key to economic growth and prosperity. New evidence shows that owning a toilet, drinking clean water, and living in a community where most of one’s neighbors own a toilet are important drivers of child growth and cognitive development in Indonesia. Repeated exposure to fecal pathogens—especially common in areas where open defecation is practiced, fecal waste management is inadequate, and water quality is poor—can cause inadequate absorption and nutrient loss through diarrhea and poor gut function. These conditions stunt a child’s growth, causing irreversible impairment to development, learning, and earning—the effects of which extend over generations. Despite recent gains, many millions of Indonesians still go without improved water and Poverty is only sanitation. In Indonesia, 87 percent of the population has access to improved drinking water one determinant and 61 percent has access to improved sanitation, a 39 and 36 percentage point increase, of WASH access: respectively, since 1990. Although these gains are commendable, there are still close to poor urban dwellers often 100 million people without improved sanitation and 33 million without improved drinking water, have better according to the WHO-UNICEF Joint Monitoring Programme for Water Supply and Sanitation access than (WHO and UNICEF 2015). These summary figures hide the persistent divides between urban wealthier rural and rural populations and among different income levels in access to services, and they mask inhabitants. underlying gaps in quality faced by all households, regardless of income or geographic location. Growing incomes are helping to both reduce poverty and increase access to proper water and sanitation, but gaps in access between the poor and rich remain, and in some cases are widening. For drinking water (urban and rural) and urban sanitation, access increased in Improving Service Levels and Impact on the Poor ix parallel for the top three quintiles (top 60, or T60) and the bottom two quintiles (bottom 40, or B40); however, overall levels of access are lower among the B40 and the gaps remain large. For the most recent year of data (2015), there was a 14 percentage point gap between the T60 and B40 in access to improved water in urban areas (84 percent vs. 70 percent), and a 10 percentage point gap for rural water (64 percent vs. 54 percent). For rural sanitation, the gap between the T60 and B40 households has increased, and in 2015 stood at 20 percentage points (55 percent vs. 35 percent). Although rates of open defecation in rural areas decreased at the same rate in B40 and T60 households, B40 households were more likely to transition from open defecation to basic latrines, whereas their T60 counterparts transitioned to improved latrines. The gap in access to sanitation between B40 and T60 households in urban areas is also substantial (19 percent), but it has narrowed from 25 percent in 2002. Indonesia’s rapid Factors other than poverty also significantly affect access to drinking water and sanitation, urbanization particularly geographic location. Urban dwellers in the lower income quintiles are more likely to could either use improved toilets and drink clean water than rural dwellers in the upper income quintiles. exacerbate or There is also variation at the local government (LG) level. District poverty rates do not neatly reduce inequality. correspond with either levels of access or equity of access to improved sanitation; some poor districts are doing a better job than wealthier districts, and a far better job than some of their poorer peers. For example, despite there being no significant difference in poverty levels between Java and non-Java districts, Java districts have achieved higher levels of coverage overall for both B40 and T60 households. Failure to address However, it is the persistent gaps in service quality—rather than barriers to access—that are the sanitation the main challenge facing Indonesia at the outset of the Sustainable Development Goal (SDG) needs of period. Although most households are gaining access to drinking water and sanitation due to urban dwellers rapid urbanization and increasing living standards, not everyone is benefitting from the same increases inequality. quality of service. In 2015, 33 percent of T60 households had a piped water connection in urban areas, compared with only 20 percent of B40 households. Furthermore, it is estimated that more than a quarter (27 percent) of B40 households drink groundwater that is unsafe, due to inadequate protection from environmental contamination. The Government of Indonesia (GoI) has set an ambitious target for universal access to improved water by 2019, aiming for 60 percent coverage of piped and 40 percent coverage of non-piped water sources in urban areas. However, given these patterns of access between B40 and T60 households, it is likely that B40 households will remain on a non-piped service for longer than T60 households. Safe drinking Progressive approaches to urban sanitation have led to millions of Indonesians gaining access water and access to improved services over the past decade. Despite these gains, an estimated 95 percent of to sanitation not fecal waste still makes its way into the nearby environment due to poor quality on-site septic only support tanks, lack of adequate emptying and disposal, or dysfunctional wastewater treatment. These child health, but are drivers conditions elevate the cost of water treatment, and lead to environmental degradation, greater of cognitive risk of disease, and poor child health and stunting. The poor in urban Indonesia are not only development. less likely to have adequate sanitation, but are more likely to live in areas where their neighbors also lack these services. Failure to address the sanitation conditions of urban dwellers, especially those living in informal settlements, could exacerbate inequalities, and is among the greatest threats to the inclusive growth and sustainability of Indonesian cities. A poor-inclusive The water and sanitation sector in Indonesia is at a pivotal juncture in the post-2015 SDG era, approach to where success will be defined by service quality, sustainability, and equitable distribution of WASH access services. The GoI’s own ambitious target of achieving universal access to water supply and can help drive sanitation by 2019 is 11 years ahead of the SDG target. The challenge to achieving these a reduction in overall inequality. targets, and achieving them on schedule, is compounded by the trend of rising income inequality and rapid urbanization in Indonesia. In contrast to the SDGs, the 2019 universal access target has no clearly stated poor-inclusive mission guiding it, despite evidence that the poor are less likely to have access to higher quality water and sanitation services, and are more likely to suffer the negative consequences of this lack of access, such as poor health and nutrition. x Improving Service Levels and Impact on the Poor The objective of this report is to provide an empirical basis for more inclusive and equitable An estimated service delivery in the water and sanitation sector in Indonesia. Although the GoI has established 95% of fecal a program and strategy for achieving universal access to water supply and sanitation and zero waste in Indonesia slums (the 100-0-100 program, which aims for 100 percent access to water supply, zero urban still makes its slums, and 100 percent access to sanitation), these targets will be achieved through different way into the service level sub-targets. For water supply, the target is for 40 percent of the population to have environment. access to piped water and 60 percent to non-piped (in urban areas, 60 percent piped and 40 percent non-piped), whereas for sanitation, universal access is defined as 15 percent of the population having access to basic sanitation (a toilet that ensures hygienic separation of human excreta from human contact), 12.5 percent to centralized and decentralized sewerage systems, and 72.5 percent to on-site sanitation with improved fecal waste management. A poor-inclusive approach to universal access—one that improves the ability of and opportunity Sanitation levels for the poor and vulnerable to benefit from water and sanitation services—can help to ensure of a community, that Indonesia not only achieves its service delivery targets, but that water supply and sanitation once they reach a critical mass, are become key drivers of a reduction in inequality, enhanced health and well-being, and economic more important growth and prosperity. Policy recommendations are prioritized based on their expected impact than those of any on these development goals, and the strength of the evidence base for the solution proposed. one household. Table ES.1 summarizes the key recommended actions and the responsible agencies/ stakeholders. Key Facts and Recommended Actions Fact 1 The government of Indonesia’s 100-0-100 target is universal access to improved water supply, but current patterns in equity of access to piped water suggest that low-income households are likely to remain on a non-piped service for longer than non-poor households. Currently, of the 29.6 percent of urban households with access to piped water supply, the B40 make up just 7.5 percent, whereas the T60 make up 22.1 percent. A number of barriers, including (1) financial sustainability and performance of PDAMs; (2) government budget allocation and spending; (3)  perceptions and behavioral constraints; and (4) lack of legal frameworks for equitable service delivery prevent low- income households from accessing piped water connections. Recommended Action Expand piped water services to a larger share of the bottom 40 percent in urban areas. •• Improve the efficiency and performance of Perusahaan Daerah Air Minum (water utility; PDAMs) to generate a virtuous cycle of performance, tariff increases, cost recovery, and expansion of connections, especially to poor households. The National Urban Water Supply Program (NUWSP), the main delivery mechanism for the urban water supply platform, includes a robust emphasis on performance improvement of PDAMs. The program could be enhanced through capacity building for LGs and PDAMs on incorporating equity and social concerns into tariff structures, and guidance on structuring cross-subsidization between customers in order to protect the poor and vulnerable. Additional capacity building on project preparation and project proposal development should cover (1) how to assess affordability of water tariffs; (2) willingness among poor households to pay for piped water connections; and (3) incorporation of low-income households, including customers of Water Hibah (an output-based grant scheme for piped water), in the overall performance improvement and investment plan. Improving Service Levels and Impact on the Poor xi •• Expand financing options for low-income households to connect to piped water. Piped water connection fees are unaffordable for households living near or below the poverty line, and although the Hibah scheme has incentivized more poor- inclusive service delivery, not all PDAMs are eligible to participate, leaving a large share of poor households unable to connect. Subsidized credit and savings schemes, including microfinance, could be an alternative that allows households to spread the cost of the connection over time. Better coordination between Hibah and microfinance schemes can be achieved through the platform approach, taking advantage of a common policy framework regardless of the source of financing and greater flexibility at the LG level to partner with private sector actors. In addition, the existing targeting mechanism for Hibah beneficiaries that is based on electricity usage could be combined with income targeting to better identify eligible low-income households for financial subsidies. •• Raise awareness of the benefits of piped water—both among consumers and among local government actors—to shift consumer behavior and dependence on alternatives, and to build the political will for improvements in water supply to poor households. Awareness campaigns have been missing from most water supply programs. Although most households treat their water before drinking, either through boiling or filtration, they are unaware of the potential for recontamination during storage. Awareness campaigns, in accordance with a Water Safety Plan to achieve water quality standards, can be coordinated by the Ministry of Public Works and Housing (MoPWH) and Ministry of Health (MoH), and implemented in part by PDAMs. These campaigns can help to increase demand for clean water and put pressure on PDAMs and LGs to expand provision of piped water services to unserved communities and/or improve the quality of existing services. •• Adjust the current intergovernmental fiscal transfer system to better align transfers to needs. Although current levels of government budget allocation to water supply are insufficient to achieve the universal access targets for water supply, existing fiscal transfers could be allocated more efficiently to address needs. Basic information on water access is readily available; however, data on the Special Allocation Fund (Dana Alokasi Khusus, or DAK) transfers show a declining association between DAK allocations and water coverage at the district level. Additional considerations for aligning fiscal transfers to needs through the General Allocation Fund (Dana Alokasi Umum, or DAU) point to population growth in urban centers, and in suburban districts in particular. To better align fiscal transfers with population growth trends will require adoption of a per capita calculation, as opposed to the current per region calculation, to ensure equitable distribution of public resources according to population density of cities and districts. This alignment does not address the need for more financing to the sector overall. Commercial loans and private investment, including business-to-business collaboration, should be explored to better understand how these additional sources of financing can help bridge the gap. Fact 2 Groundwater quality is not consistently monitored, and representative data are not available. However, water quality surveys conducted in several cities show the potential risk for contamination is severe. A large share of the B40 uses groundwater sources for drinking. In 2015 data showed that over a quarter (27 percent) of the B40 drink unsafe groundwater, compared with 14 percent of the T60. Contamination stems from poor quality septic tanks and untreated domestic wastewater, as well as from landfill and industrial effluent. xii Improving Service Levels and Impact on the Poor Recommended Action Improve the quality of alternative water sources for those who will remain on non-piped water supply. •• Enhance monitoring for water quality risks for all source types, and make this information publicly available. Consumers are largely unaware of the variable quality of drinking water from different sources and the particular risks posed by poor household water storage practices and poor fecal waste management. Water sector strategy should account for the potential water quality risks of poor sanitation, and the respective investments of the water and sanitation sub-sectors should be aligned. This alignment is especially important in areas facing technical barriers to piped water. The Local Development Planning Agency (Bappeda) at city level can ensure that the needed alignment of water and sanitation is reflected in the respective strategy documents (the  Master Plan for Drinking Water and the City Sanitation Strategy). Bappeda could also oversee integration of data from the two sub-sectors into planning, implementation, and monitoring. •• Strengthen regulatory control for small-scale water providers to ensure that regulations on drinking water quality are met. For refilled bottled water, enhanced control could be achieved by linking water quality monitoring, under the responsibility of MoH, with the licensing process, under the Ministry of Industry (MoI). Fact 3 Between 2006 and 2015 access to improved sanitation grew at a rate of 6.5 percent annually. However, there were still close to 100 million people without improved sanitation in 2015 and the majority of these lived in rural areas. Just 48 percent of the population has improved sanitation in rural areas, compared with 76 percent of those in urban areas, a gap of 28 percentage points. Rates of open defecation have declined at similar rates between the B40 and T60 since 2002, but B40 households were more likely to move to basic latrines, contributing to the widening gap in access to improved sanitation between the B40 and T60. Recommended Action Support the bottom 40 percent in gaining access to improved sanitation. •• Strengthen the Sanitasi Total Berbasis Masyarakat (Community-Based Total Sanitation, or STBM) strategy by revisiting the zero-subsidy approach in order to move up the sanitation ladder. Although global practice suggests subsidies can harm sanitation behavior-change efforts, experience shows that when well-targeted, delivered through an efficient channel, and affordable, subsidies can be an effective mechanism to reach poor households which otherwise cannot afford the high lump- sum cost of a toilet. Targeting subsidized credit and savings schemes through existing targeting systems that are already working well to identify low-income households for social assistance—such as the Unified Database (UDB) operated by the National Team for the Acceleration of Poverty Reduction (TNP2K) and the Ministry of Social Affairs (MoSA)—can be an efficient and transparent way to reach households most in need of subsidies and achieve higher levels of service. The UDB contains socioeconomic and demographic information for the approximately 40 percent of the population with the lowest welfare status, the equivalent of 24 million households, or 96 million individuals. The MoH should take a leadership Improving Service Levels and Impact on the Poor xiii role in adapting the existing policy on sanitation subsidies to address the financial constraints of poor households, and MoPWH should work with TNP2K and MoSA to adopt the UDB for targeting assistance under the Community-Based Rural Water Supply and Sanitation Program (PAMSIMAS). Fact 4 The vast majority of households in urban areas use an improved toilet connected to a septic or sewerage system (78 percent), but less than 2 percent of those are connected to sewerage. However, a combination of high idle capacity for existing sewerage networks, poor performing septage treatment plants, limited improvements to fecal sludge management, and poor quality investments in on-site sanitation systems results in 95 percent of fecal waste making its way into the nearby environment through the process of containment, emptying, transport, treatment, and disposal. Conditions of high population density and inadequate fecal waste disposal interact to make poor sanitation particularly risky to the health of people population health in urban areas. Recommended Action Bring more households into the full sanitation and fecal waste service chain in urban areas. •• Adopt a more holistic and inclusive approach to planning for citywide sanitation to accommodate the range of solutions required to meet universal access targets in urban areas. Planning should cover the full fecal waste service chain and outline a progressive roadmap for bringing the entire population into this service chain. Local solutions are complex, requiring a combination of piped and non-piped technologies, such as septic tanks, sewerage, decentralized small-scale wastewater treatment plants, and fecal sludge management. District heads and mayors need to be given responsibility for ensuring consistency in planning, budgeting, and execution; flexible funding arrangements; and technical assistance and capacity building where needed. This approach requires a delicate balance between the national government’s fiscal leverage to incentivize investment in sanitation, and granting greater autonomy to LGs to decide where and how to invest those resources. •• Adapt sanitation behavior change to behavioral issues common in the urban sanitation space. The universal access targets will be met primarily through on-site sanitation systems with fecal sludge management (72.5 percent); smaller shares are planned for centralized and decentralized sewerage (12.5 percent) and basic sanitation (15 percent). Low consumer demand for fecal waste management services is a reflection of both the lack of integrated services and the lack of knowledge about safe management and disposal practices. Part of the solution will require generating the necessary demand and changing the behavior of individuals, communities, and providers. But behavior change cannot happen in a vacuum—it also requires a coherent policy framework, clarity on institutional arrangements, and adequate enforcement of LG ordinances for design, construction, and desludging (World Bank and Australian Aid 2013). Coordination between MoH and MoPWH will be needed for effective implementation of STBM in urban areas, along with the Ministry of Environment and Forestry (MoEF) to enforce new regulations on effluent standards. •• Elevate the profile of sanitation in political and fiscal discussions, as well as in intra- household decision-making. This change could require a shift in the narrative around xiv Improving Service Levels and Impact on the Poor urban sanitation to emphasize not only elements of modernity and competitiveness, but also the lifelong effects on intellectual and economic potential of early life stunting, caused in part by poor sanitation. AKKOPSI (Regency/City Alliance for Better Sanitation) could lead advocacy efforts with mayors and district heads. Fact 5 An estimated 9 million children (37 percent) under five in Indonesia are stunted. Children in rural areas are more likely to be stunted than children in urban areas, but a child from the lowest income quintile is just as likely to be stunted whether he or she lives in an urban area (48 percent likelihood) or a rural area (49 percent). Owning a toilet and having access to clean drinking water supply, as well as living in a community where most of one’s neighbors own a toilet, are important drivers of child growth and cognitive development in Indonesia. Access to WASH is just one key driver of nutrition, with food security, care, and access to health care being additional factors. The nutritional impact of WASH investments can be  enhanced through multisectoral convergence to ensure that children have simultaneous access to all drivers of nutrition. Recommended Action Champion multisectoral approaches to reduce child stunting. •• Capitalize on synergies of multisectoral approaches. Progress toward reducing stunting in Indonesia can be enhanced by coordinated multisectoral interventions that address effectively the four key underlying determinants of nutritional status— food security, access to health care, child care practices, and access to water and sanitation. Geographic targeting can be used to reach areas where undernutrition and underlying deprivations are high. In these areas, interventions should be co- located to achieve service improvements across multiple sectors that impact stunting. PAMSIMAS can serve as the main platform for multisectoral convergence between WASH and other programs addressing nutrition outcomes in young children, with oversight of implementation coordinated through the National Development Planning Agency (Bappenas). •• Crowd in resources until communities achieve near universal coverage of sanitation. There is now compelling evidence, both within Indonesia and globally, that sanitation levels of a community are more important than those of any one household. The evidence shows that health and nutritional benefits mainly accrue after a minimum threshold level of coverage is surpassed, and that full benefits may only be achieved as sanitation becomes universal. This evidence supports existing sector practices, defecation free (ODF) areas, and suggests that resources should which aim for open-​­ be spent on bringing as many communities as possible to universal or near-universal levels of coverage in order to realize the health benefits of sanitation. •• Adapt water and sanitation interventions to be more “child-centric.” The five pillars of STBM ([1] stop open defecation; [2] hand washing with soap; [3] household safe water treatment and storage, and safe food handling; [4] safe disposal and management of solid waste; and [5] safe disposal and management of wastewater) are comprehensive across WASH services, but may still miss some of the dominant fecal contamination pathways that affect small children. An emerging approach known as “baby WASH” or “child-centered WASH” focuses on interrupting exposure pathways that are most strongly associated with subsequent diarrheal disease. Improving Service Levels and Impact on the Poor xv The  MoH should adapt existing STBM behavior-change communication materials and LG capacity building to incorporate baby WASH, while implementation of the approach should be aligned with the current nutrition-sensitive pilot of PAMSIMAS. •• Target slum areas and informal settlements with multisectoral action. Conditions of poverty, overcrowding, and poor quality services interact to magnify the risks of poor water and sanitation in densely populated urban slums. The speed and scale of urbanization in Indonesia contributes to the urgency with which these challenges must be addressed. Multisectoral approaches have largely focused on rural areas, but the challenge in urban slums and informal settlements is complex, as an effective response involves a multitude of actors and is complicated by institutional constraints and tenure insecurity. Additional work is needed to understand the contamination pathways unique to these settings, and how to effectively engage different actors under the National Slum Upgrading Program (KOTAKU). Table ES.1: Key Recommended Actions and Responsible Agencies What Who Reduce Inequalities in Access and Quality Expand piped water services to a larger share of the B40 in urban areas Improve the efficiency and performance of PDAMs MoPWH, Bappenas, MoHA, private sector Enhance the capacity of LGs and PDAMs on tariff- MoHA, MoPWH, donor setting to support the establishment of cost- agencies, recovery tariffs Center of Excellence (CoE) • Conduct analytical work on the implementation of program, Association new regulations on tariffs and subsidies of PDAMs (Persatuan • Add specific tariff-setting content to existing Perusahaan Air Minum capacity building programs Seluruh Indonesia, or PERPAMSI), NUWSP Additional financing mechanisms to ease the Bappenas, MoPWH projects, financial and liquidity constraints faced by the poor NGOs, local financing • Continue and improve the Water Hibah scheme by institutions linking with investment on capacity improvement • Scale up microfinance and similar mechanisms • Encourage collaboration between Hibah and microfinance schemes • Combine existing targeting mechanism with income targeting to better identify eligible low- income households table continues next page xvi Improving Service Levels and Impact on the Poor Table ES.1: Continued What Who Increase demand and raise consumer awareness of MoPWH, MoH, PDAMs the benefits of piped water Adjustments to the current intergovernmental fiscal Bappenas, MoPWH, MoF, transfer system to better align transfers to needs donor agencies • Exercise alternative approach in the allocation of DAK and DAU • Diagnose private sector involvement in water sector, including commercial loans, private investment, and business-to-business collaboration in bridging the financing gap Improve the quality of alternative water sources for those who will remain on non-piped water supply Consistently monitor water quality risks to drinking MoH, District Health Office, water supplies, piped or non-piped, and make this Bappeda information publicly available • Strengthen the critical link across water and sanitation sub-sectors—e.g., ensure the alignment of the Master Plan for Drinking Water and City Sanitation Strategy Strengthen regulatory control for small water MoH, MoI providers to ensure that regulations on drinking water quality are met • Link water quality monitoring with licensing process for refilled bottled water providers Support the B40 in gaining access to improved sanitation • Strengthen STBM strategy by revisiting the “zero- Bappenas, MoH subsidy” for poor households • Identify various financial schemes to move up the Bappenas, MoH, MoPWH sanitation ladder, such as DAK, Sanitation Hibah, Village Grant, and community social responsibility (CSR) funds • Explore the possibility of targeted subsidy for Bappenas, MoH, MoPWH, the poorest segment of people to move up the PAMSIMAS sanitation ladder (from basic to improved latrines) • Adopt existing targeting systems that are already Bappenas, MoH, MoPWH, working well identifying low-income households Ministry of Social Protection (such as the UDB from TNP2K) to ongoing programs, including STBM, PAMSIMAS, etc. table continues next page Improving Service Levels and Impact on the Poor xvii Table ES.1: Continued What Who Bring more households into full sanitation and fecal waste service chain in urban areas Take holistic approach to planning in implementing Bappenas, MoPWH, Bappeda, citywide sanitation-inclusive approach PPSP • Apply the fecal waste diagram as a tool to assess citywide sanitation and identify priorities for city sanitation strategy • Ensure consistency in sanitation management at local level through PPSP (Acceleration of Urban Sanitation Development Program) and link it with decision-making on investment using central budget Adapt approach to behavior issues in urban MoH, MoPWH, MoEF sanitation, including enforcing the effluent standard • Effective implementation of STBM in urban areas • Enforce new regulations on effluent standards Elevate the profile of sanitation in political and MoHA, Bappenas, MoPWH, fiscal discussion MoH, AKKOPSI Improve Health, Nutrition, and Early Child Development Champion multisectoral approaches to reduce child stunting Capitalize on synergies of multisectoral Bappenas, MoPWH, MoH, approaches, including strengthening the existing Ministry of Social Protection scaling up nutrition (SUN) program and alignment with non-cash nutrition support Crowd in resources until communities achieve Bappenas, MoH, MoPWH, high coverage of sanitation Bappeda Adapt water and sanitation interventions to be MoH, PAMSIMAS more “child-centric” • Adapt existing STBM behavior-change communication materials and LGs capacity building programs to incorporate “baby WASH” • Ensure that the implementation of the “baby WASH” approach aligns with current nutrition- sensitive pilot of PAMSIMAS Enhance water supply and sanitation interventions MoH, Bappenas, STBM to be more impactful on nutrition outcomes Target slum areas and informal settlements Bappenas, MoPWH, Vice with multisectoral action President’s Office xviii Improving Service Levels and Impact on the Poor References WHO and UNICEF. 2015. Progress on Drinking Water and Sanitation: 2015 Update and MDG Assessment. WHO/UNICEF Joint Monitoring Programme for Water Supply and Sanitation. New York: UNICEF. World Bank and Australian Aid. 2013. Urban Sanitation Review: Indonesia Country Study. Washington, DC: World Bank. Improving Service Levels and Impact on the Poor xix Abbreviations AKKOPSI Regency/City Alliance for Better Sanitation APBD Local government budget APBN National government budget B40 Bottom two quintiles, or bottom 40% Bappeda Local Development Planning Agency Bappenas National Development Planning Agency BPPSPAM Support Agency for the Development of Drinking Water Supply System BPS Badan Pusat Statistik or Central Bureau of Statistics BPSPAMS Badan Pengelola Sistem Pelayanan Air Minum dan Sanitasi (village water boards) CHEF Care, Health, Environment (or WASH) and Food Security CLTS Community-Led Total Sanitation CoE Center of Excellence Program CSS City Sanitation Strategies DAK Special Allocation Fund (Dana Alokasi Khusus) DAU General Allocation Fund (Dana Alokasi Umum) DFAT Department of Foreign Affairs and Trade, formerly known as AusAid GDP Gross domestic product GoI Government of Indonesia IDR Indonesian Rupiah IFLS Indonesia Family Life Survey IMF International Monetary Fund IndII Indonesia Infrastructure Initiative (project funded by DFAT) JMP Joint Monitoring Programme for Water Supply and Sanitation LG Local government MDG Millennium Development Goal MoEF Ministry of Environment and Forestry MoEMR Ministry of Energy and Mineral Resources MoF Ministry of Finance MoH Ministry of Health MoHA Ministry of Home Affairs MoI Ministry of Industry MoPWH Ministry of Public Works and Housing MoSA Ministry of Social Affairs NTT East Nusa Tenggara Improving Service Levels and Impact on the Poor xxi NUWSP National Urban Water Supply ODF Open Defecation Free OECD Organisation for Economic Co-operation and Development PAMSIMAS Community-Based Rural Water Supply and Sanitation PDAM Perusahaan Daerah Air Minum (Water Utility) PDPAL Perusahaan Daerah Penanganan Air Limbah (Regional Wastewater Treatment Enterprise) PERPAMSI Persatuan Perusahaan Air Minum Seluruh Indonesia (Association of PDAMs) PPSP Acceleration of Sanitation Development Program RISKESDAS Indonesia Basic Health Research RPJMN Rencana Pembangunan Jangka Menengah-Nasional (National Medium Term Development Plan) SANIMAS Sanitasi Oleh Masyarakat (Sanitation by Communities) SDG Sustainable Development Goal STBM Sanitasi Total Berbasis Masyarakat or Community-Based Total Sanitation STH soil-transmitted helminths Susenas National Socio-Economic Survey T60 Top 3 quintiles, or top 60% TNP2K National Team for the Acceleration of Poverty Reduction UDB Unified Database UNICEF United Nations Children’s Fund UPTD District Technical Implementing Unit USAID United States Agency or International Development WASH Water Supply, Sanitation, and Hygiene WHO World Health Organization xxii Improving Service Levels and Impact on the Poor Chapter 1 Introduction and Background Indonesia is the fourth most populous country in the world, with 252 million people spread over a vast equatorial archipelago of 6,000 inhabited islands; the country has a total of more than 13,000 islands, which extend nearly 6,000 kilometers east to west and across three time zones. The population distribution and levels of development vary considerably across the islands. The largest population cluster is on Java, which hosts 60 percent of the country’s inhabitants, followed by Sumatra, which has a larger land area but hosts less than a third of the population. Approximately 118 million people (46 percent of the population) currently live in rural areas, where the majority of the poor are concentrated.1 A Global Mandate for Quality, Sustainability, and Equitable Distribution of Services The Millennium Development Goals (MDGs) came to a conclusion in 2015. Worldwide, 2.1  billion people gained access to improved sanitation, while 147 countries met the MDG drinking water target, 95 countries met the MDG sanitation target, and 77 countries met both. To continue these efforts and shift the world onto a sustainable path, world leaders gathered on 25 September 2015 at the United Nations in New York to adopt the 2030 Agenda for Sustainable Development. The 2030 Agenda comprises 17 new Sustainable Development Goals (SDGs), or Global Goals,2 which will guide policy and funding for the next 15 years. Universal access to clean water and sanitation is 1 of 17  Global Goals that make up the 2030 Agenda for Sustainable Development. Goal 6 aims to “Ensure availability and sustainable management of water and sanitation for all,” with two main targets: •• Target 6.1: By 2030, achieve universal and equitable access to safe and affordable drinking water for all •• Target 6.2: By 2030, achieve access to adequate and equitable sanitation and hygiene for all and end open defecation, paying special attention to the needs of women and girls and those in vulnerable situations In this context, the GoI, through the National Medium Term Development Plan (Rencana Pembangunan Jangka Menengah Nasional, or RPJMN) for the period 2015–19, has set the target of universal access to water supply and sanitation by the end of 2019.3 To achieve this universal access target, Ministry of Public Works and Housing (MoPWH) has launched the 100-0-100 program (100 percent access to water supply, zero urban slums, and 100 percent access to sanitation). The 100-0-100 program defines specific service levels to be met for universal access by the end of 2019: Water Supply •• 40 percent of the population with access to piped water and 60 percent to non-piped water (in urban areas 60 percent piped and 40 percent non-piped); •• 85 percent of urban areas receiving at least 100 liters per capita per day; Improving Service Levels and Impact on the Poor 1 •• 15 percent of urban areas receiving a basic level of 60 liters per capita per day; •• all supplies meeting the 4K standards (Kualitas, Kuantitas, Kontinuitas, Keterjangkauan) for quality, quantity, continuity and affordability Sanitation •• 15 percent of the population having access to basic sanitation (a toilet that ensures hygienic separation of human excreta from human contact); •• 12.5 percent to centralized and decentralized sewerage systems; •• 72.5 percent to on-site sanitation with improved fecal waste management. Overview of the Water Supply and Sanitation Sector in Indonesia Table 1.1 provides a high-level overview of the Water Supply and Sanitation Sector in Indonesia. Chapter 5 provides further detail on the institutional setup of Urban Water as a basis for Table 1.1: Overview of Water Supply and Sanitation Sector in Indonesia Sub- Service Service sector National authority authority provider Programs Urban •• Ministry of Public Works District PDAMs (district NUWSP (under preparation) Water and Housing (lead water utilities) The delivery mechanism for the Supply institution) urban water supply platform of the •• National Development urban water program, which includes Planning Agency a robust emphasis on performance (Bappenas, for national improvement and capacity building of planning and monitoring) PDAMs and Local governments (LGs) •• Ministry of Health (for to accelerate urban piped water water quality standards) service provision. •• Ministry of Home Water Hibah Affairs (for institutional improvements in An output-based grant (Hibah) subnational governments scheme for household piped water and setting up the connections, particularly for low- monitoring on minimum income households. The Hibah service standards) program provides reimbursement to the LG once connections have been independently verified. The program serves as an incentive to LGs to ensure water service delivery. Eligibility is restricted to “healthy” PDAMs that are performing well and have adequate spare production capacity and the ability to pre-finance capital investments. table continues next page 2 Improving Service Levels and Impact on the Poor Table 1.1: Continued Sub- Service Service sector National authority authority provider Programs Rural •• Ministry of Public Works District Village- PAMSIMAS Water and Housing (lead and village level CBOs The government’s main platform for Supply institution) governments predominantly; expanding Community-Based Rural •• National Development since the referred to as Water Supply and Sanitation, which Planning Agency Village Law Badan Pengelola targets underserved and low-income (Bappenas, for national (passed in Sistem communities. PAMSIMAS devolves planning and monitoring) 2014) Pelayanan planning and management of •• Ministry of Villages, Air Minum water systems to local government Underdeveloped Regions dan Sanitasi and communities through the and Transmigration (BPSPAMS) establishment of village BPSPAMS. (for institutional under the improvements in village national government) platform to •• Ministry of Home Affairs accelerate (for capacity building rural water for community-based supply access organization [CBO] that (PAMSIMAS) manage the rural water and sanitation systems) Urban •• Ministry of Public Works District Regional PPSP Sanitation and Housing (lead Wastewater Acceleration of Urban Sanitation institution) Treatment Development Program—a national •• National Development Enterprise program to assist LGs in conducting Planning Agency (Perusahaan comprehensive citywide sanitation (Bappenas, for planning Daerah planning through the preparation and monitoring) Penanganan of City Sanitation Strategies (CSS). •• Ministry of Health (for Air Limbah, Currently, PPSP is in its second advocacy and behavior or PDPALs), phase (PPSP-2) and focused on change) PDAMs, the transition from planning to •• Ministry of Environment technical unit implementation. and Forestry (for setting under specific up the effluent standard) district office SANIMAS or District Community-managed decentralized Solid Waste wastewater treatment system—a Management national program to expand Office decentralized community-managed wastewater systems, which may include public facilities or decentralized sewerage systems with a communal treatment facility. These systems were developed as a medium-term solution to serve high-density residential areas, with plans to eventually connect to municipal sewerage. table continues next page Improving Service Levels and Impact on the Poor 3 Table 1.1: Continued Sub- Service Service sector National authority authority provider Programs Sanitation Hibah An output-based grant (Hibah) scheme for sewerage connections, particularly for low-income households. The Hibah program provides reimbursement to the LG once connections have been independently verified. The program serves as an incentive to LGs to ensure sanitation service delivery. Improving Urban Fecal Sludge Management Activities initiated to help LGs improve existing fecal sludge management practices, which include (1) improving on-demand services, and (2) initiating regular desludging services. Rural •• Ministry of Health (lead District Sanitarian, STBM Sanitation institution, for behavior and village Community Community-Based Total Sanitation— change) governments Health Centre adopted in 2005 by combining •• National Development since the (Puskesmas) community-led total sanitation with a Planning Agency Village Law market-based approach and behavior- (Bappenas, for planning (passed in change communication. Sanitasi Total and monitoring) 2014) Berbasis Masyarakat (STBM) is seen •• Ministry of Public as the primary national approach to Works and Housing (for scaling up rural sanitation. technical advisory and support) diagnosis of the constraints to, and opportunities for, extending piped water access to the poor in urban settings provided in chapter 6. Given the complexity of the water and sanitation sector in Indonesia—with differing institutional, financial, and technical arrangements across sub-sectors—it was not possible to provide a similar level of detail for each of the sub-sectors within the present study, and it is recommended that these sub-sectors be the subject of further analytical work moving forward. Approach and Methodology of the WASH Poverty Diagnostic The Indonesia WASH Poverty Diagnostic seeks to strengthen the evidence base on inequality of access to, and quality of, water and sanitation services between the bottom 40 percent and the top 60 percent of the population in Indonesia. The World Bank Group’s new goal of boosting shared prosperity shines a spotlight on the welfare of the B40 population. Although growth itself 4 Improving Service Levels and Impact on the Poor will be important to achieve the goal of shared prosperity, it has become increasingly clear that promoting equality of opportunities has the dual advantages of fostering fairness and helping countries to achieve their aspirations of economic prosperity. Equalizing basic opportunities for children today will promote social mobility and reduce income inequality tomorrow. The diagnostic tailors a pilot framework developed by the Water Global Practice that focuses on a set of core questions. The overall objectives are to (1) deepen understanding of the extent and distribution of inequalities in access to water supply and sanitation services; (2) inform policy and investment strategy to achieve national and global targets; and (3) increase client capacity for identifying and diagnosing service delivery constraints to improving the quality of water supply and sanitation services for the poor. The following core questions guide the Diagnostic: 1. Who and where are the poor and bottom 40 percent of national distribution of income? This is a fundamental question for identifying the population of interest, and draws mainly on evidence and analysis produced under the Poverty Global Practice Indonesia country work program. 2. What are the linkages and synergies between WASH and other sectors? This question addresses how lack of access to water and sanitation affects early child health, nutrition, and development. Inequality in outcomes, such as income, is influenced by inequality of opportunity that includes access to basic services such as water supply and sanitation, and the inability of households to benefit from the synergies associated with access to more than one basic service. 3. What is the level of access and quality of WASH services experienced by the poor and bottom 40 percent compared to the non-poor and top 60 percent? Analysis of “access” for the diagnostic is based on the WHO/United Nations Children’s Fund (UNICEF) Joint Monitoring Programme (JMP) global categories of improved and unimproved service levels, but prioritizes Government of Indonesia (GoI) methods and definitions to facilitate policy dialogue. Where possible, additional dimensions of service quality are analyzed to begin to characterize levels of service as defined under the SDG framework. 4. What are the WASH service-delivery constraints and potential solutions to improving services to the poor and bottom 40 percent? Addressing the previous core questions helps to frame an institutional and problem-driven political economy analysis in one of the four sub-sectors. The diagnostic draws on a variety of existing data sources. Socioeconomic household survey data (primarily Susenas) is used for distributional and geospatial analysis to characterize how access to services varies at national and subnational levels. To reflect the nature of decentralized service delivery and responsibility in Indonesia, a key unit of analysis of the diagnostic is the district. The analysis also investigates differences across cities of different sizes (e.g., metropolitan, small, medium, and big cities). Econometric analysis using Susenas is used to understand the determinants of access, while the Indonesia Family Life Survey (IFLS) is used to model impacts of access to WASH in early childhood on later life nutrition and cognitive outcomes; the Indonesian Basic Health Research (RISKESDAS) survey is used for distributional and econometric analysis of access to the determinants of nutrition. Spatial mapping is used to visualize associations between access to WASH, poverty, and health outcomes. Stakeholder consultations were conducted throughout the preparation of the WASH Poverty Diagnostic. These consultations included Bappenas, MoPWH, Ministry of Health (MoH), Ministry of Environment and Forestry (MoEF), several development partners and relevant projects, such as UNICEF, United States Agency or International Development (USAID), Department of Foreign Affairs and Trade (DFAT), IUWASH Plus (Indonesia Urban Water Sanitation and Hygiene, or Penyehatan Lingkungan untuk Semua, a USAID-funded project) and IndII (Indonesia Infrastructure Initiatives, a DFAT-funded project), as well as the World Bank Improving Service Levels and Impact on the Poor 5 The GoI aims to Indonesia Water, Health, Poverty, and Governance teams, and other stakeholders. A final achieve universal consultation was held on June 7, 2017, after completion of the diagnostic to solicit feedback access to water on the policy recommendations and incorporate suggestions into the final report. and sanitation by 2019. It is increasingly recognized that indicators of access to “improved” services are insufficient for capturing all crucial preconditions for the desired poverty and prosperity outcomes over the long term. Monitoring of drinking water and sanitation services during the MDG period focused on indicators that identify adequate protection from outside contamination, particularly fecal matter (for drinking water), and hygienic separation of human excreta from human contact (for sanitation). The SDG targets, which are more in line with the current indicators used by the GoI, provide further granularity on service levels by including additional dimensions such as availability, accessibility, and quality. Existing socioeconomic household surveys in Indonesia do not yet capture these dimensions consistently, necessitating the use of multiple surveys, which are limited in population representativeness. To overcome data limitations, the WASH Poverty Diagnostic uses a variety of published evidence and secondary data sources. Importantly, the WASH Poverty Diagnostic uses GoI definitions, rather than those of the JMP , to categorize access to improved drinking water and sanitation (see box 1.1). Box 1.1: Defining Access to Improved Drinking Water There are differences in the definition of access indicators used by the Government of Indonesia (GoI) and the WHO/UNICEF Joint Monitoring Programme (JMP). The GoI defines “access to improved water” as the share of households whose primary source of drinking water is either: (1) piped water; (2) protected pump/well/­ spring water, at a minimum distance of 10 meters from a fecal disposal site; or (3) rain water. Excluded from the definition are: (1) pump/“protected” well/spring water, at less than 10 meters from a fecal disposal site; (2) “unprotected” well/spring; (3) bottled water; and (4) others. Bottled water is regarded as an “unsustainable” source in view of its relative price compared with other sources, but recent increases in use of bottled water as the primary source of drinking water prompted the GoI to modify the definition to include households using a safe and sustainable water source for both cleaning and cooking (even if their drinking water is not from a sustainable source, such as bottled water). The Central Bureau of Statistics (BPS) compiles this indicator through the National Socio-Economic Survey (Susenas), which is conducted annually. In addition to differences in how access is defined, different data sources and underlying population data contribute to different estimates. The JMP uses both the Demographic and Health Survey (DHS) and Susenas and applies a linear regression model to estimate coverage, while GoI refers only to Susenas. 6 Improving Service Levels and Impact on the Poor Roadmap of the Report The remainder of this report is structured in 6 parts. Chapter 2 gives an overview of recent trends in poverty and the increase in inequality in Indonesia—in particular, inequality of opportunity and implications for economic growth and stability. Chapter 3 reviews global and Indonesia-specific evidence on the links between water supply and sanitation, and health and nutrition, including econometric anlaysis of the impact of WASH on early child health conducted under the Diagnostic. In addition, analysis is presented on the role of multisectoral interventions to address stunting. Chapter 4 presents a core set of diagnostics of water supply and sanitation, including recent trends in access and quality, disparities across geography and income groups, and key service delivery challenges facing the sector. The chapter begins with a diagnostic of the sanitation sector, both urban and rural, and then turns to urban and rural water supply. Chapter 5 outlines the constraints on and opportunities for delivery of piped water to the poor in urban settings. A focus on the urban water sector is timely and opportunistic given the context of rapid urbanization in Indonesia, the longstanding challenges faced by urban utilities to keep up with the resulting rising demand, and the current focused efforts by government urban water counterparts to tackle these issues. Chapter 6 presents key recommended actions to orient future water supply and sanitation policy and investment toward a more inclusive approach. Recommendations are prioritized based on their expected impact on the development goals of (1) reducing inequality; (2) enhancing health and well-being; and (3) promoting economic growth and prosperity, as well as on the strength of the evidence base for the solution. The chapter also presents proposed future analytical work on topics that were not fully addressed in the report. The evidence summarized in this report has benefited from a set of topical background papers. The background papers include: 1. Determinants of Access to Improved Water Sources in Indonesia from a Household Demand Perspective, which estimates the influence of households’ socioeconomic, demographic, and geographic characteristics on the probability of choosing specific sources of drinking water. It then quantifies and synthesizes this understanding of determinants, and suggests how this knowledge could be utilized to provide scientific evidence in support of the GoI’s efforts to accelerate access to improved water sources in Indonesia. 2. Child Stunting and Cognitive Impacts of Water, Sanitation and Hygiene in Indonesia, which examines the relationship between poor households and community water and sanitation services and child stunting, underweight, and cognitive development in Indonesia, using a sample of 3,049 children from the IFLS. 3. Operationalizing a Multi-Sectoral Approach for the Reduction of Stunting in Indonesia: An Application Using the 2007 and 2013 RISKESDAS, which conducts analysis of the distribution, co-distribution, and synergies between the underlying determinants of nutrition and their effect on height-for-age outcomes in children under five. The analysis provides a practical diagnostic framework for identifying potential “binding constraints” in the Indonesian context to support efforts to reduce child stunting and malnutrition. 4. Identifying and Overcoming Binding Constraints to Piped Urban Water Services for the B40 in Indonesia, which uses problem-driven political economy and institutional analysis to examine the WASH service delivery constraints and potential solutions to improving urban piped water services to the poor in Indonesia. Additionally, a dynamic dashboard accompanies the report, which contains all underlying data. The dashboard is intended for use by the GoI, the World Bank, and development partners to monitor inequalities in WASH services, along with poverty levels, and prevalence of health outcomes among children under five, including stunting. The dashboard can help to inform Improving Service Levels and Impact on the Poor 7 geographic and sub-group targeting for more pro-poor and poor-inclusive WASH interventions, as well as identify where further inquiry is needed to understand why service delivery fails within certain groups or geographic areas. The dashboard is available at: http://witiestudio​ .­com/worldbank-map/ (see appendix A for examples). Notes 1. World Bank Development Policy Review 2014 Indonesia: Avoiding the Trap. 2. UNDP, http//www.undp.org. 3. National Medium Term Development Planning 2015–19. 8 Improving Service Levels and Impact on the Poor Chapter 2 Rising Inequality and Why It Matters Core Question: Who and where are the poor and bottom 40 percent (B40) of national distribution of income? Key facts • Between 2002 and 2015 Indonesia reduced poverty from 18.2 percent to 11.2 percent. • 28.6 million Indonesians live below the poverty line and 62 million are vulnerable to poverty. • Poverty rates are higher in Eastern Indonesia, compared with other regions, but the largest number of people living below the poverty line are in the islands of Java and Sumatra in Western Indonesia. • Consumption inequality, as measured by the Gini index, increased from 36 to 41 between 2002 and 2015, one of the largest increases in the world over this period. • The level of wealth inequality in Indonesia is one of the highest in countries for which there are data; by some estimates, the richest 1 percent own 50 percent of all financial and property wealth. • The poorest 40 percent now account for only a fifth of total household consumption, while the richest 20 percent account for nearly half. • Around a third of total inequality in Indonesia is due to inequality of opportunity— circumstances that give a child an unfairly disadvantaged start in life. • A key driver of better opportunities for the next generation begins with improvements in the delivery of basic services. The Evolution of Poverty and Inequality in Indonesia Understanding how poverty and inequality have evolved, and exploring the pathways to reducing poverty and boosting shared prosperity, are questions of central importance both globally and in Indonesia. While Indonesia has achieved solid economic growth and reduced poverty from 18.2 percent in 2002 to 11.2 percent in 2015, there are still 28.6 million Indonesians who live below the poverty line and 62 million people vulnerable to poverty.1 Alongside poverty reduction, the GoI recently declared rising inequality the administration’s top priority for 2017 (Gibson 2017). Improving Service Levels and Impact on the Poor 9 Figure 2.1: Trend in Poverty Reduction 2002–15 25 20 Poverty rate (%) 15 10 5 0 02 03 04 05 06 07 08 09 10 11 12 13 14 15 20 20 20 20 20 20 20 20 20 20 20 20 20 20 Year Rural National Urban Source: Susenas, World Bank calculations. Between 2002 Between 2002 and 2015 poverty decreased by more than a third in Indonesia. Urban areas and 2015, poverty experienced a slightly faster rate of poverty reduction, at 4 percent annually, compared to a decreased by reduction of 3 percent in rural areas (figure 2.1).2 more than a third in Indonesia, but consumption Poverty rates are higher in Eastern Indonesia, compared with other regions in the country. Eight inequality out of 16 provinces that exceeded the national poverty rate of 11.2 percent in 2015 are significantly located in Eastern Indonesia. Provinces in Papua Island, East Nusa Tenggara, and Maluku have increased. more than one-fifth of their population below the poverty line (map 2.1). However, the largest number of people living below the poverty line are on the islands of Java and Sumatra in Western Indonesia. This is primarily due to greater population density—in fact, half of the Indonesian population is living in Java. Vulnerability to poverty remains high in Indonesia, with a large share of the population living just above the official poverty line.3,4 This group accounts for 27 percent of the population, so that the poor and the vulnerable together make up 38 percent of the population (that is, the bottom 40 percent). Living standards for households classified as vulnerable remain low, and a small shock to income or expenditures for this group can easily send them into poverty (World Bank 2016a). Inequality in Indonesia has been steadily rising despite rapid growth over the past decade. Between 2002 and 2015 the country’s mean per capita consumption grew an average of 4.6 percent per year, and gross domestic product (GDP) per capita (measured at 2010 constant US$) rose from US$2,259 in 2002 to US$3,834 in 2015 (WDI). During this same period consumption inequality, as measured by the Gini index, increased from 36 to 41, as shown in figure 2.2 (Badan Pusat Statistik) (BPS). This places Indonesia among the countries with the highest levels of inequality in East Asia, just below Malaysia, the Philippines, and China. 10 Improving Service Levels and Impact on the Poor Map 2.1: Poverty Rate and Poverty Headcount, by Geographic Location, 2015 a. Poverty Rate b. Poverty Headcount 4.9–6.8 13.0–16.5 10,805–16,139 41,154–54,974 0 250 500 750 km 6.8–8.4 16.5–20.0 0 250 500 750 km 16,139–22,040 54,974–85,993 0.0–3.3 8.4–10.8 20.0–26.0 0-6377 22,040–29,320 85,993–1,45,796 3.3–4.9 10.8–13.0 26.0–46.0 6377-10805 29,320–41,154 1,45,796–6,26,529 Source: Susenas, World Bank calculations. Figure 2.2: Trends in Inequality, Households Per Capita Consumption, 1980–2015 45 Asian financial 40 crisis 35 30 Pre-crisis Post-crisis and Reformasi 25 Year 20 15 10 5 0 0 0 99 01 93 09 11 13 3 4 95 5 15 97 07 8 9 0 8 0 19 19 19 20 19 20 20 20 20 19 19 20 20 19 20 Gini Coefficients (points) Source: Susenas, World Bank staff calculations. Indonesia now has one of the highest levels of wealth inequality amongst countries for The richest which there are data. In 2014, according to Credit Suisse, the richest 10 percent of people 1 percent of owned 77 percent of all financial and property wealth, the fourth highest level of wealth Indonesians own half of the inequality in the dataset. The richest 1 percent owned half of all wealth. The World Bank country’s wealth. (2016a) identified high wealth inequality as a key driver of rising consumption inequality in Indonesia, as this high concentration of wealth in the hands of a few provides both higher incomes today for wealthy families—affording better health care and education for their children, greater opportunities, and better jobs for tomorrow—and wealth that is passed down between generations. The increase in inequality is the result of the richest 20 percent enjoying the most economic growth, while the B40 enjoyed relatively little. While the richest 10 percent enjoyed annual Improving Service Levels and Impact on the Poor 11 Figure 2.3: Change in Inequality, Gini Coefficient 60 50 40 Gini coefficient 30 20 10 0 a na s ia a ia os m nd ne si di es d na La hi ay la bo In pi on C et ai al ilip am Vi Th d M In Ph C 1990s 2000s Source: Zhuang et al. 2014. consumption growth of over 6 percent and the second richest 5.3 percent, consumption for the B40 grew at only 3–4 percent annually between 2002 and 2015. The B40 now account for only a fifth of total household consumption, while the richest 20 percent account for nearly half. Inequality has been rising rapidly compared to Indonesia’s neighbors, although it has not yet reached the levels seen in some countries. Indonesia started out in the 1990s with the lowest Gini coefficient in the region, but by the late 2000s, inequality had reached the levels seen in its peers (figure 2.3).5 The size of the increase was also larger than that seen in any other country in the region, except China, and most of this increase happened over a short time. Importantly, other fast-growing East Asian neighbors such as Malaysia, Thailand, and Vietnam experienced stable or declining inequality over the same period (World Bank 2016a). Generally speaking, higher income inequality reduces and destabilizes economic growth, according to an International Monetary Fund (IMF) study (Ostry and Berg 2011). When the share of total income held by the richest 20 percent of the population increases by 5 percentage points annually, economic growth falls by 0.4 percentage points. Conversely, when the share of total income held by the poorest 20 percent of the population increases by 5 percentage points, growth increases by 1.9 percentage points. Urbanization Indonesia is also undergoing rapid urbanization, and while this can be accompanied by strong can catalyze economic growth, it creates a number of challenges, including disparities in income and access strong economic to services. For example, in just the past few years poverty has increased by 2 percentage growth, but it can points in Jakarta, despite the downward trend nationwide, and this could be related to high also exacerbate disparities in rates of migration from rural areas. In the absence of adequate planning, rapid urbanization income and also creates challenges for effective delivery of services, the absence of which contributes to access to widening inequalities. These inequalities can cause social friction, leading to a rise in crime services. and violence, which threaten the sustainability of the growth process and undermine the benefits of urbanization (UN Habitat 2008). 12 Improving Service Levels and Impact on the Poor The Source of Inequality and Implications for Service Delivery Sometimes inequality incentivizes people to work hard and take risks, leading to innovation Equitable and economic growth. But when not everybody begins life with the same opportunities, due to distribution of factors beyond their control, inequality is considered unfair. This sort of inequality (inequality of services can opportunity) makes people uncomfortable and can lead to social conflict (Atinc et al. 2005). level the playing field and make Thus, a society that ensures equality of opportunity for its citizens helps engender greater equitable trust, more efficient growth, and better institutions. Although it is difficult to guarantee equality outcomes in outcomes such as income—and many would argue against this as a policy objective— more likely. ensuring equitable distribution of basic services and interventions can “level the playing field,” which can make equitable outcomes more likely. Around a third of total inequality in Indonesia is due to inequality of opportunity—that is, when About one third a child is born into circumstances that give him or her a disadvantaged start in life (World Bank of inequality in 2016a). Inequality of opportunity can stem from birthplace, race, gender, or other inherited Indonesia is characteristics. Basic opportunities, such as access to water and sanitation, education, and due to inequality of opportunity health care, are other circumstances that are out of a child’s control but which compelling that children evidence shows dictate future opportunities. Often these deprivations overlap. In Indonesia, experience early for instance, evidence indicates that one-fifth of rural children lack simultaneous access to on in life. critical services such as health care, education, and transportation services (figure 2.4). A key driver of better opportunities for the next generation is improved delivery of basic services. Improved delivery In Indonesia, responsibility for delivery of basic services, including water supply and sanitation, of services is key was devolved to local governments (LGs) after the democratization and decentralization to creating better reforms in the late 1990s. However, LG entities often lack capacity, a supportive regulatory opportunities for the next environment, coherent institutional and fiscal arrangements, and political incentives to generation. effectively deliver these services. As a result, service delivery breaks down—despite supportive policies and high-level political commitment. Addressing the critical gap between policy and implementation that impedes service delivery requires doing business differently: understanding the functioning of the public sector and the politics of reform, and taking action in the context of those constraints. The following chapter reviews global and Indonesia-specific evidence on the links between water supply and sanitation and human development outcomes, illustrating the ways in which water and sanitation services, along with other key determinants of nutrition, can safeguard a child’s future health, and social and economic well-being. Figure 2.4: Children in Rural Areas Lack Access to Multiple Services a. Urban access to services, 2011 b. Rural access to services, 2011 Poor Poor Poor health 9% education 2% Poor health 3% 3% access 5% access education 6% access 40% 41% access 8% 1% 2% 2% 7% 20% 5% 6% 22% Poor 18% Poor transportation transportation 26% 50% Source: World Bank 2015 An Unfair Start: How Unequal Opportunities Affect Indonesia’s Children. Improving Service Levels and Impact on the Poor 13 Notes 1. The poverty line used throughout the report is Indonesia’s national poverty line, currently approximately 330,000 IDR per person per month, set by BPS unless stated otherwise. 2. The poverty rate increased by nearly 2 percentage points between 2005 and 2006 due to sharply higher rice prices, a result of a ban on rice imports (World Bank 2006a Making the New Indonesia Work for the Poor). 3. The official method used by the Central Bureau of Statistics (BPS) to set the poverty line is the basic needs approach, begun in 1984. Since poverty is defined as the inability to meet basic food and non-food needs, the “basic needs approach” is based on the consumption module of Susenas that is collected annually from around 65,000 households. Currently, the consumption module includes 216 food items and 94 non- food items. Based on the “basic needs approach,” the indicator used to measure poverty is the Head Count Index, defined as the number or percentage of poor people living under the poverty line. 4. Vulnerable is defined as households with at least a 10 percent chance of being below the poverty line in the following year. Using this definition, the World Bank calculated the vulnerability line based on Susenas panel data 2008–10. The calculated vulnerability line is approximately 1.5 times the poverty line (World Bank 2012). 5. Note on Figure 4: Consumption Ginis for all countries except Malaysia, which uses income. The periods for each country are: Indonesia 1990–2011; Malaysia 1992–2009; the Lao People’s Democratic Republic 1992–2008; China 1990–2008; Vietnam 1992–2008; Thailand 1990–2009; the Philippines 1991–2009; and Cambodia 1994–2008. References ., Menendez, M., Ozler, B., Prennushi, G., Atinc, T. M., Banerjee, A., Ferreira, F. H. G., Lanjouw, P Rao, V., Robinson, J., Walton, M., and Woolcock, M. 2005. World Development Report 2006: Equity and Development. World Development Report. Washington, DC: World Bank. Gibson, Luke. 2017. “Towards a More Equal Indonesia.” Oxfam. Ostry, J. D., and A. Berg. (2011). Inequality and Unsustainable Growth; Two Sides of the Same Coin? (11/08). International Monetary Fund. UN Habitat. 2008. State of the World’s Cities 2008–2009: Harmonious Cities. Earthscan. World Bank. 2015. An Unfair Start: How Unequal Opportunities Affect Indonesia’s Children. Jakarta: World Bank. World Bank. 2016a. Indonesia’s Rising Divide. Washington, DC: World Bank Group. Zhuang, J., R. Kanbur, and C. Rhee. 2014. “Asia’s income inequalities.” In Inequality in Asia and the Pacific: Trends, Drivers, and Policy Implications, edited by J. Zhuang, R. Kanbur, and C. Rhee. New York: Asia Development Bank and Routledge. 14 Improving Service Levels and Impact on the Poor Chapter 3 The Role of WASH in Human Development Core Question: What are the linkages and synergies between WASH and other sectors? Key facts • Children living in an environment contaminated by feces are more likely to be infected by disease-causing pathogens, leading to diarrhea, poor gut function, and stunting. • An estimated 9 million children (37 percent) under five are stunted in Indonesia, and children in the lowest quintile (49 percent) are more likely to be stunted than children in the highest (29 percent). • Stunting is higher in rural areas (42 percent) than in urban (33 percent), but children in the lowest quintile are just as likely to be stunted whether they live in urban areas (48 percent) or in rural (49 percent). • Owning a toilet and having access to clean drinking water supply, as well as living in a community where most of one’s neighbors own a toilet, are important drivers of child growth and cognitive development. • Poor access to WASH is just one key driver of stunting and malnutrition, with food security, child care practices, and access to health care being additional factors. • Children who have simultaneous access to multiple determinants of nutrition are taller on average, but very few children meet this criteria, suggesting that the nutritional impact of sector-specific interventions may be limited by poor access to the other drivers of nutrition. Pathways of Exposure and Impacts on Health and Nutrition Poor quality latrines, inadequate fecal waste management, and open defecation are the main sources of environmental fecal contamination that cause disease. Pathogens in the environment are spread through drinking and washing water, hands, vectors (such as flies), and the soles of the feet or shoes. The “F-diagram” (figure 3.1) illustrates how these fecal pathogens can enter into the open environment, and illustrates the role that sanitation, clean water supply, and hygiene each play in blocking these contamination routes. The evidence underlying the F-diagram has recently expanded to show how children, in particular, face high risk of exposure (Mahmud and Mbuya 2015). For instance, a recent Improving Service Levels and Impact on the Poor 15 Figure 3.1: F-Diagram Showing How Inadequate WASH Affects Child Health and Nutrition Sanitation Clean water supply Hygiene Fluids Feces Fingers Future Food victim Flies Fields/ floors Source: Wagner and Lanoix 1958 adapted by the World Bank. study in Bangladesh found that 98 percent of soil samples from areas where small children play tested positive for E. coli at extraordinarily high levels. Structured observations of children in the study showed them frequently mouthing hands and objects that had touched the soil, and 18 percent of children were observed putting soil directly into their mouths (Ercumen et al. forthcoming). Infants and children also risk being fed water and food that has become contaminated by poor hand hygiene, and by dirty utensils and containers used for preparation, storage and serving. Food and water can be contaminated at a more macro level due to inadequate management and disposal of feces, disposal of untreated wastewater into water supply sources, and storm runoff. Due to a pregnant woman’s immune response, and the permeability of the placenta to disease-causing pathogens, researchers hypothesize that these risks can affect an unborn fetus (see box 3.1) (Campbell et al. 2015). Stunting Access to a safe water supply improves the overall disease environment and enables behaviors is caused that influence child health and nutrition. Access to piped water in urban areas has been shown by multiple to decrease infant mortality. For example, privatization of water services in Buenos Aires led to determinants, improved access and service quality, resulting in an 8 percent reduction in child mortality from and sector- specific infectious disease. In the poorest areas that benefited the most from the service expansion, interventions child mortality declined by 26 percent (Galiani, Gertler, and Schardrogsky 2005). Lack of may have access to a continuous source of safe water can have negative health consequences if limited impact. households revert to using unimproved sources of water for even short periods of time (Hunter, Zmirou-Navier, and Hartemann 2009). Adequate quantities of water are also needed to practice hygiene behaviors such as handwashing, washing utensils and containers, and cleaning objects and surfaces used by children (Howard and Bartram 2003). A child who lives in an environment contaminated by feces is more likely to be infected by disease-causing pathogens, including bacteria, viruses, and other microorganisms. An estimated 58 percent of diarrheal disease is due to poor water supply, sanitation, and hygiene (Prüss-Ustün et al. 2014), causing the loss and malabsorption of nutrients. Protozoa and helminths (worms) live in feces and are transmitted in water (schistosomiasis) and soil-transmitted helminths (STH), causing infections that lead to anemia, malnutrition, stunted growth, and impaired physical and cognitive development. These outcomes are associated with lower school attendance and educational attainment—factors that limit future economic productivity (Victora et al. 2008). 16 Improving Service Levels and Impact on the Poor Box 3.1: Water Supply, Sanitation, and Hygiene and their Effect on Maternal Health Poor water supply and sanitation can contribute to maternal outcomes through factors such as hygiene and quality of piped water and sewerage systems in the home or at a health facility where antenatal care or delivery takes place (Campbell et al. 2015). Direct evidence about the effect of household-level water supply, sanitation, and hygiene (WASH) availability on maternal outcomes in Indonesia is limited to a 2003 study in Surabaya (Taguchi et al.), which found that lack of a toilet facility in the home more than doubled the probability of maternal mortality. Using panel data from the Indonesia Family Life Survey (IFLS), a sample of 456 women who were pregnant at the time of the survey were tracked to a subsequent wave, and their pregnancy outcomes (live birth or miscarriage) observed. After controlling for other confounding factors, preliminary results suggest that having access to improved water at the time of pregnancy is associated with a 7 percentage point increase in the probability of a live birth. The increase is particularly associated with women living in rural areas where access to clean water is lower. Access to improved water also reduces the probability of miscarriage by 2 percentage points. Having an improved toilet in the home, or living in a community with high coverage of toilets at the time of pregnancy, was not associated with pregnancy outcomes. There is robust evidence that access to WASH decreases the incidence of diarrhea in young children. Synthetic review and meta-analysis of health impact assessments of water and sanitation interventions (Wolf et al. 2014) show water interventions reduce diarrhea morbidity by 34 percent, sanitation interventions reduce it by 28 percent, and promotion of handwashing with soap results in a 40 percent reduction (Freeman et al. 2014). The largest health effects for improved water are for piped water supply, with a greater benefit associated with higher quality piped water—water that is safe and continuously available (Wolf et al. 2014). Open defecation and poor sanitation are harmful not only to those who lack access to toilets, Stunting reflects but—due to externalities—to other households in the community as well. Researchers have the cumulative demonstrated across countries and data sets that the health benefits of sanitation mostly effects of accrue as sanitation becomes universal (Andres et al. 2014; Hunter and Prüss-Ustün 2016; infection and undernutrition, Larsen et al. 2017). A number of randomized controlled trials of household sanitation and is regarded interventions have failed to demonstrate health impacts (Clasen et al. 2014; Patil et al. 2014), as the non- and researchers hypothesize that this is due to insufficient community-wide coverage and income face behavior change. The implication is that sanitation interventions that fail to adequately reduce of poverty. the pathogen load in the environment will not achieve substantial nutritional impacts. Growing Tall and Smart with WASH Researchers have recently begun to look at the impacts of WASH on stunting. This work points to environmental enteric dysfunction (EED), or poor gut function, caused by repeated ingestion of fecal bacteria (Humphrey 2009) as the primary pathway linking unsafe WASH to poor health and nutrition outcomes, especially stunting. A child who suffers from EED is less able to Improving Service Levels and Impact on the Poor 17 Photo 3.1: Malnourished and Nourished Gut Lining Malnourished Nourished Source: Garcia 1968. Figure 3.2: Pathways between Poor WASH and Early Child Development Exploratory play; Chronic Hand-to- inflammation mouth Stunting and behaviors Anemia Early child Microbial Environmental development ingestion enteropathy Altered brain Low Low development Unhygenic Psycho-social nutrient nutrient diet stimulation diet absorption Source: Ngure et al. 2014. absorb nutrients due to a damaged gut lining (photo 3.1). At the same time the small intestine becomes more porous, and disease-causing pathogens enter the bloodstream more easily, activating an immune response and diverting energy from human growth. These conditions do not produce overt symptoms like diarrhea, making them harder to track and quantify, although estimates suggest that up to 43 percent of stunting may be due to these silent infections (Guerrant et al. 2012). Stunting has Few studies have extended this work to understand the potential impact of WASH on early lifelong effects childhood development. Early childhood experiences—encompassing care, stimulation and on cognitive learning, nutrition, and stress, especially in the first 1,000 days of life—have a profound development, impact on brain development. The chronic gut inflammation characteristic of EED, caused in earnings, and part by poor WASH, makes it more difficult to absorb nutrients and leads to poor cognitive intergenerational poverty. development (Ngure et al. 2014). At the same time, children who are sick often or who aren’t growing as well as their peers may be treated differently or sheltered from their social environment, depriving them of early psycho-social stimulation opportunities critical for brain development (Ngure et al. 2014) (figure 3.2). 18 Improving Service Levels and Impact on the Poor The Stubborn Problem of Child Stunting in Indonesia and Linkages with WASH Stunting is a severe and persistent problem in Indonesia, where an estimated 9 million children In a six-year under five (37 percent) are stunted (RISKESDAS 2013). This situates Indonesia just behind period (2007–13), India in terms of the share of children who are stunted.1 Stunting is chronic undernutrition— rates of stunting defined as a child whose length/height is below minus 2 standard deviations of the median in the poorest households rose, height for a child of the same age from the reference population. Since it reflects the cumulative whereas those effects of infection and undernutrition from the time a child is in the womb, stunting is often in the richest referred to as the non-income face of poverty and is regarded as one of the best indicators of declined. overall human development. Stunting has lifelong adverse consequences for cognitive development, human capital, productivity, earnings, and intergenerational transmission of poverty (Victora et al. 2008). The prevalence of stunting is widespread and distributed across the Indonesian archipelago (map 3.1). In rural areas the prevalence of stunting is 42 percent, while in urban areas it is 33 percent (Skoufias 2016). Although stunting rates are higher in rural areas, research has shown that slum populations are particularly at risk—more so than rural and urban (total) populations (FAO 2017). Map 3.1: Stunting Rates in Indonesia, 2013 Stunting rate 11.1–37.2 37.2–70.4 Source: Calculations based on the 2007 and 2013 RISKESDAS. The prevalence of stunting varies by wealth quintile (figure 3.3). A striking 29 percent of children in the topmost quintile are stunted, while 49 percent are stunted in the bottommost (Skoufias 2016). Furthermore, more children in the poorest households were stunted in 2013 than in 2007, rising from 43 percent in 2007 to 49 percent in 2013; the same is not true for wealthier households, where rates of stunting are declining.2 Children in wealthier households were 5 percentage points less likely to be stunted in 2013 than in 2007. Therefore, whereas in 2007 the difference in stunting rates between children from the poorest households and those from the wealthiest households was 10 points, by 2013 it had increased to 20 points. There is no clear pattern in trends of stunting prevalence across districts between 2007 and 2013. In the maps below (map 3.2), the upper panel shows changes in stunting rates for those districts where the stunting rate in 2007 was above the national stunting rate of 36.8 percent; the lower panel shows changes in stunting rates for those districts where the stunting rate in 2007 was below 36.8 percent. In both panels, districts in green saw a more than four percentage point decrease in stunting between 2007 and 2013, while those in red saw a more than four percentage point increase in stunting. Many of the districts with stunting rates above the national average in 2007 had lower stunting rates in 2013, and many of the districts with stunting rates below the national average in 2007 had a higher prevalence of Improving Service Levels and Impact on the Poor 19 Figure 3.3: Stunting Rates, by Wealth Quintile, 2007–13 100 Percentage of children with 80 HAZ < –2 SD 60 48 49 43 42 44 43 40 39 39 40 37 35 33 33 28 29 20 0 1 2 3 4 5 2007 2010 2013 Source: Calculation based on the 2007, 2010, and 2013 RISKESDAS. Map 3.2: Changes in District Stunting Rates, 2007–13 a. Below national average (36.8%), 2007 Percentage point change in stunting Stunting decreased Stunting remained constant Stunting increased Stunting in 2007 below 36.8 percent figure continues next page 20 Improving Service Levels and Impact on the Poor Map 3.2: Continued b. Above national average (36.8%), 2007 Percentage point change in stunting Stunting decreased Stunting remained constant Stunting increased Stunting in 2007 above 36.8 percent Source: Calculations based on the 2007 and 2013 RISKESDAS. stunting in 2013. Only a few districts that had stunting rates lower than the national average were able to decrease stunting by more than 4 percentage points, whereas a number of districts with stunting rates above the national average saw an increase in stunting. Child Stunting and Cognitive Impacts of Water and Sanitation in Indonesia The WASH Poverty Diagnostic used panel data from the IFLS to analyze associations between Stunting has access to WASH in utero and in the first 2 years of life, and nutrition and cognitive outcomes for lifelong effects the same children later in life (see Cameron 2017). First, local polynomial regression analysis on cognitive was used to explore the relationship between community open defecation and child nutrition and development, earnings, and cognitive outcomes. The analysis shows that children living in villages (urban and rural combined) intergenerational where a higher proportion of households have a toilet are closer to meeting growth standards and poverty. score higher on cognitive tests than children in villages with a lower proportion of toilets. Two- thirds of the gains in average height-for-age z-scores (HAZ)3 accrue after a threshold of around 60 percent coverage of improved sanitation is surpassed (figure 3.4). Data from the 2013 round of RISKESDAS show that just 27 percent of children in rural settings and 62 percent in urban settings are living in communities where coverage of sanitation is at adequate levels of 75 percent or above (Skoufias 2016), putting the majority of children, especially in rural areas, at risk. Regression analysis was used to examine the effect of three independent variables of interest on the nutrition and cognitive outcomes of Indonesian children. The three variables are: household-level sanitation, community-level sanitation, and access to improved household-level water sources during the window of opportunity (from the time a child is in utero until 2 years of age). Outcomes include a child’s cognitive test scores from the 2007 round of IFLS when a child is between the ages of 7 and 16, and anthropometric outcomes, including stunting, underweight, HAZ, and weight-for-age z-scores (WAZ), measured when a Improving Service Levels and Impact on the Poor 21 Figure 3.4: Relationship between ODF and Height-for-Age and ODF and Cognitive Scores a. Village ODF and HAZ z-score (Pooled) b. Village ODF and cognitive scores (Pooled) –1.2 12.0 Mean village height-for-age z–score Mean village cognitive score –1.4 11.5 –1.6 11.0 –1.8 10.5 –2.0 10.0 –2.2 9.5 0 0.2 0.4 0.6 0.8 1.0 0 0.2 0.4 0.6 0.8 1.0 Mean village ODF rate (%) Mean village ODF rate (%) 95% CI Ipoly smooth Kernel = epanechnikov, degree = 0, bandwidth = 0.14, pwidth = 0.21 Kernel = epanechnikov, degree = 0, bandwidth = 0.23, pwidth = 0.34 Source: Cameron et al. 2017. Note: ODF = open-defecation free. child is under 5 years of age.4 The analysis controls for parental/caregiver education, mother’s age at the time of birth, age at the time of measurement (either anthropometric or cognitive testing), and decile of per capita consumption. Sub-district and IFLS wave fixed effects were also included. Results indicate that Indonesian children who were exposed to a poor water and sanitation environment early in life were more likely to be stunted or underweight and to score lower on cognitive tests than those with adequate levels of water and sanitation. Children living in communities with lower levels of open defecation during the critical window of the first 1,000 days of life are 11 percentage points less likely to be stunted and 5 percentage points less likely to be underweight. These children also score 1.3 points higher on cognitive tests—equivalent to about a 0.33 standard deviation increase in cognitive score. Both water and sanitation access at the household level are associated with better early childhood outcomes in Indonesia. Children who have access to an improved water supply gain 0.41 standard deviation in WAZ,5 while an improved toilet in the home reduces the likelihood of being underweight by 3 percentage points and improves cognitive test scores by 0.37 percentage points. Forty-four percent of households in the study were located in urban areas. There was no significant difference in health or cognitive outcomes for children living in urban as compared with rural areas. Owning a toilet These findings suggest that owning a toilet, as well as living in a community where most of and living in a one’s neighbors own a toilet, are important drivers of a child’s cognitive development. The only community where other published evidence of a sanitation/cognitive development link comes from an econometric most people own study in India (Spears and Lamba 2016), which uses variation in latrine construction at the a toilet helps drive cognitive district level in a child’s first year of life and matches it with child-level data on cognitive development. achievement to estimate the impact of India’s Total Sanitation Campaign (TSC) on cognitive outcomes. The authors found that children exposed to TSC in their first year of life were better able to recognize letters and simple numbers by the age of six. The cognitive effects from early life exposure to safe water and sanitation are comparable to a range of early childhood education and health interventions, suggesting that access to water and sanitation is among a core set of critical early childhood development interventions. 22 Improving Service Levels and Impact on the Poor Not by WASH Alone: How Multisectoral Interventions Help Improve Nutrition Operationalizing a Multisectoral Approach for the Reduction of Stunting in Indonesia Stunting is widely understood to be caused by multiple underlying factors, including food insecurity, poor child care practices, and lack of access to health services, water, and sanitation. The United Nations Children’s Fund (UNICEF) conceptual framework of child undernutrition, initially proposed in 1990 (UNICEF 1990), was one of the first attempts to emphasize household food security, WASH services, and maternal and child care practices as the main underlying determinants of child nutrition in developing countries. A fundamental premise of this conceptual framework is that increases in access to adequate services in any  one of the drivers of nutrition—for example, food security alone—cannot substitute for  inadequate levels of access to the other determinants. While there is widespread acknowledgment of the key underlying determinants of nutrition, there is limited quantitative information on the interdependence of adequate (or inadequate) access to the determinants for child nutrition. The WASH Poverty Diagnostic undertook an econometric analysis to quantify the relationships between and interdependence of the determinants of child nutrition outcomes using the 2007 and 2013 RISKESDAS surveys. Indicators were constructed for four groups of child nutrition determinants proposed in the UNICEF framework—child care practices, health, environment (water and sanitation), and food security (CHEF).6 Each indicator is comprised of various components, depending on the data available in the survey, with the definition of “adequacy” based on national and/or accepted international standards.7 In consideration of the complexity of the linkages between the underlying determinants of nutrition and the economic situation of the family, the analysis is also carried out separately for urban and rural households, and for resource-rich T60 and resource-poor B40 households, as well as for districts with high stunting rates and those with low stunting rates. A more holistic view is provided regarding the extent to which adequate levels of the four determinants—food security; adequate caregiving resources at the maternal, household, and community levels; access to health services; and a safe and hygienic environment—are, both on their own and in combination, associated with better nutrition as measured by HAZ and stunting rates. In Indonesia, survey results indicate overall low levels of access to the key drivers of nutritional health. Nationally, in 2013, just 7 percent of children under 5 had access to adequate child care practices—defined by factors such as early and appropriate breastfeeding, handwashing, a smoke-free home environment, and complementary feeding. A similarly low percentage of children (14 percent) had access to adequate food, such as protein, calories, exclusive breastfeeding, and nutrition of the mother. About 56 percent of children had access to basic drinking water and improved sanitation, including adequate levels of community coverage of sanitation, and 46 percent had access to adequate health services such as prenatal care, vitamin A, and immunizations. Substantial inequalities in access to adequate levels of CHEF determinants remain between The health rural and urban areas, between districts with high and low stunting rates, and between poorer benefits of and wealthier households. In 2013 fewer children in rural areas had access to all four drivers sanitation of nutrition than children in urban areas. The largest discrepancy was in access to an adequate mostly accrue as sanitation environment, with only around 40 percent of those in rural areas having access and around becomes 70 percent of those in urban areas having access. Access to an adequate environment is also universal and the nutrition driver showing the largest differences between wealth quintiles. Only 1 percent pathogens in of children in the lowest wealth quintile had access to an adequate environment, whereas in the environment the highest quintile 92 percent of the children had access to an adequate environment are adequately reduced. (Skoufias 2016). Improving Service Levels and Impact on the Poor 23 Access to all four CHEF determinants is exceedingly uncommon, suggesting that nutritional impacts of sector-specific interventions may be limited by inadequate access to the underlying drivers of nutrition. In 2013, 23 percent of children between 0 and 5 years of age did not have adequate access to any of the four determinants of nutrition, while less than 1 percent of children had simultaneous access to all four key underlying determinants of nutrition. For the poorest 20 percent of children, 68 percent lack adequate access to all four determinants. Children under 5 in Indonesia with simultaneous access to adequate levels of two of the four drivers of nutrition have higher mean height for age z-scores. Stunting rates are even lower among children with simultaneous access to adequate levels of three of the four drivers of nutrition (figure 3.5). These children had a 13.4 percent lower likelihood of stunting, compared to the reference group without access to any drivers, whose rate of stunting is 41.4 percent. This pattern is consistent across rural and urban areas, and for children living in households in the top 60 percent of the wealth distribution. These results illustrate the importance of coordinated multisectoral policies and suggest that the success of “sector-specific nutrition- sensitive” initiatives could be enhanced by better coordination and integration across sectors to effectively address multiple underlying determinants of nutrition. In the next chapter, the report turns to the core diagnostics of the water supply and sanitation sector in Indonesia, including recent trends in access and quality, disparities across geography and income groups, and key service delivery challenges facing the sector. The chapter begins with a diagnostic of the sanitation sector, both urban and rural, and then turns to urban and rural water supply. The trends and patterns described provide overall guidance on sector policy priorities, particularly as they relate to the B40. However, the list of challenges described is not necessarily exhaustive, and is limited in the sense that it uses existing data and evidence (primary data collection was not part of the WASH Poverty Diagnostic in Indonesia). These trends and patterns, along with recent sector studies and reports and global experience and evidence, form the basis of the policy recommendations in chapter 6. For the urban water sub- sector, the policy recommendations benefit from an in-depth institutional and political economy analysis described in chapter 5. Additional analytical work could shed further light on some of the challenges discussed for urban sanitation, rural sanitation, and rural water, and lead to more targeted policy recommendations (see chapter 6). Figure 3.5: Difference in Probability of Being Stunted, 2013 National, relative to the reference group –0.052 1 –0.089 2 Nutrition drivers –0.134 3 –0.109 4 –0.200 –0.150 –0.100 –0.050 0 Percentage point difference in stunting rate Source: Calculation based on RISKESDAS 2013. 24 Improving Service Levels and Impact on the Poor Notes 1. Indonesia stunting rate is 37 percent and is ranked 25; India is 24. Global Nutrition report. 2. These analyses, and all that follow, are based on the official HAZ scores calculated by the Indonesian Ministry of Health. 3. Height-for-age z-scores (HAZ) measure the deviation of a child’s height from the median of children of the same age in a reference population. A HAZ less than 2 standard deviations below the median for the reference population is classified as stunted; a HAZ less than 3 standard deviations is classified as severely stunted. 4. For children who were in utero during the window of opportunity, their anthropometric measures from the subsequent wave, when they were under-5 years, were used. 5. Weight-for-age z-scores (WAZ) measure the deviation of a child’s weight from the median of children of the same age in a reference population. A WAZ less than 2 standard deviations below the median for the reference population is classified as underweight; a WAZ less than 3 standard deviations is classified as severely underweight. 6. CHEF indicators consolidate similar groupings of determinants as proposed in the UNICEF framework. 7. Further details on the definitions of adequacy can be found in Skoufias 2016. References Andres, L. A., B. Briceño, C. Chase, and J. A. Echenique. 2014. “Sanitation and Externalities: Evidence from Early Childhood Health in Rural India.” Policy Research Working Papers, World Bank, Washington, DC. Cameron, L., C. Chase, G. Joseph, and R. Pinto. 2017. “Child Stunting and Cognitive Impacts of Water, Sanitation and Hygiene in Indonesia.” Policy Research Working Paper. World Bank, Washington, DC. Campbell, O. M., L. Benova, G. Gon, K. Afsana, and O. Cumming. 2015. Getting the Basic Rights—The Role of Water, Sanitation and Hygiene in Maternal and Reproductive Health: A Conceptual Framework. Tropical Medicine and International Health 20 (3): 252–67. Clasen, T., S. Boisson, P. Routray, B. Torondel, M. Bell, O. Cumming, J. Ensink, M. Freeman, M. Jenkins, M. Odagiri, S. Ray, A. Sinha, M. Suar, W. Schmidt. 2014. Effectiveness of a Rural Sanitation Programme on Diarrhoea, Soil-Transmitted Helminth Infection, and Child Malnutrition in Odisha, India: A Cluster-Randomised Trial. The Lancet Global Health 2 (11). FAO (2017). Assessment of Nutritional Status in Urban Areas. http://www.fao.org/ag/agn​ /­nutrition/urban_assessment_en.stm. Freeman, Matthew C., Meredith E. Stocks, Oliver Cumming, Aurelie Jeandron, Julian Higgins, Jennyfer Wolf, Annette Prüss-Ustün, S. Bonjour, P. R. Hunter, L. Fewtrell, V. Curti. 2014. “Systematic Review: Hygiene and Health: Systematic Review of Handwashing Practices Worldwide and Update of Health Effects.” Tropical Medicine & International Health 19 (8): 906–16. . Gertler, and E. Schargrodsky. 2005. “Water for Life: The Impact of the Privatization Galiani, S., P of Water Services on Child Mortality.” Journal of Political Economy 113 (1): 83–120. Garcia S. 1968. “Malabsorption and Malnutrition in Mexico.” Am J Clin Nutr. Sep; 21 (9): 1066–76. PMID: 5675845. Improving Service Levels and Impact on the Poor 25 Guerrant, R. L., M. D. Deboer, S. R. Moore, R. J. Scharf, and A. M. Lima. 2012. The Impoverished Gut a Triple Burden of Diarrhoea, Stunting and Chronic Disease. Nature Reviews Gastroenterology & Hepatology 10 (4). Howard, G., and J. Bartram. 2003. Domestic Water Quantity, Service Level and Health. Geneva: WHO. Humphrey, J. H. 2009. Child Undernutrition, Tropical Enteropathy, Toilets, and Handwashing. The Lancet 374: 9694. . R., and A. Prüss-Ustün. 2016. “Have We Substantially Underestimated the Impact of Hunter, P Improved Sanitation Coverage on Child Health? A Generalized Additive Model Panel Analysis of Global Data on Child Mortality and Malnutrition.” PLoS One 11 (10). . R., D. Zmirou-Navier, and P Hunter, P . Hartemann. 2009. Estimating the Impact on Health of Poor Reliability of Drinking Water Interventions in Developing Countries. Science of the Total Environment 407 (8): 2621–24. Larsen, D. A., T. Grisham, E. Slawsky, and L. Narine. 2017. “An Individual-Level Meta-Analysis Assessing the Impact of Community-Level Sanitation Access on Child Stunting, Anemia, and Diarrhea: Evidence from DHS and MICS Surveys.” PLoS Neglected Tropical Disease 11 (6): e0005591. https://doi.org/10.1371/journal.pntd.0005591 Mahmud, I., and N. Mbuya. 2015. Water, Sanitation, Hygiene, and Nutrition in Bangladesh: Can Building Toilets Affect Children’s Growth? Washington, DC: World Bank. Ngure, F. M., B. M. Reid, J. H. Humphrey, M. N. Mbuya, G. Pelto, and R. J. Stoltzfus. 2014. “Water, Sanitation, and Hygiene (WASH), Environmental Enteropathy, Nutrition, and Early Child Development: Making the Links.” Annals of the New York Academy of Sciences 1308 (1): 118–28. Patil, Sumeet R., Benjamin F. Arnold, Alicia L. Salvatore, Bertha Briceno, Sandipan Ganguly, John M. Colford Jr., and Paul J. Gertler. 2014. “The Effect of India’s Total Sanitation Campaign on Defecation Behaviors and Child Health in Rural Madhya Pradesh: A Cluster Randomized Controlled Trial.” PLoS Medicine 11 (8): e1001709. Prüss-Ustün, A., J. Bartram, T. Clasen, J. M. Colford, O. Cumming, V. Curtis, and S. Cairncross. 2014. “Burden of Disease from Inadequate Water, Sanitation and Hygiene in Low- and Middle-Income Settings: A Retrospective Analysis of Data from 145 Countries.” Tropical Medicine & International Health 19 (8): 894–905. Skoufias, E. 2016. Operationalizing a Multi-Sectoral Approach for the Reduction of Malnutrition in Indonesia: An Application Using the 2007 and 2013 RISKESDAS. Washington DC: World Bank. Spears, D., and S. Lamba. 2016. “Effects of Early-Life Exposure to Sanitation on Childhood Cognitive Skills: Evidence from India’s Total Sanitation Campaign.” Journal of Human Resources 51 (2). UNICEF. 1990. “Strategy for Improved Nutrition of Children and Women in Developing Countries.” . C. Hallal, R. Martorell, L. Richter, and H. S. Sachdev. 2008. Victora, C. G., L. Adair, C. Fall, P “Maternal and Child Undernutrition: Consequences for Adult Health and Human Capital.” The Lancet 371 (9609): 340–57. 26 Improving Service Levels and Impact on the Poor Wagner, E.G., Lanoix, J.N., Excreta Disposal for Rural Areas and Small Communities, Geneva: World Health Organization. . Higgins. Wolf, J., A. Prüss-Ustün, O. Cumming, J. Bartram, S. Bonjour, S. Cairncross, and J. P 2014. “Systematic Review: Assessing the Impact of Drinking Water and Sanitation on Diarrhoeal Disease in Low- and Middle-Income Settings: Systematic Review and Meta- Regression.” Tropical Medicine & International Health 8 (19): 928–42. Improving Service Levels and Impact on the Poor 27 Chapter 4 WASH Service Delivery Progress and Challenges in Indonesia Core Question: What is the level of access and quality of WASH services experienced by the poor and bottom 40 percent as compared to the non-poor and top 60 percent? Key facts •• Since the launch of the STBM, access to improved sanitation has grown at 6.5 percent annually, but there are still 47 million people defecating in the open and another 52 million using sanitation that is considered unsafe. •• 76 percent of the population in urban areas has improved sanitation compared with just 48 percent in rural areas, a gap of 28 percentage points. •• District poverty rates do not neatly correspond with either levels of access or equity of access to improved sanitation at the district level; some poor districts are doing a better job than wealthier districts, and a far better job than some of their poorer peers. •• Despite there being no significant difference in poverty levels between Java and non-Java districts, Java districts have achieved higher levels of coverage overall for both B40 and T60 households. •• B40 households are not achieving higher levels of service in rural areas, which is contributing to the widening gap in access to improved sanitation between the B40 and T60. •• In urban areas, 95 percent of fecal waste makes its way into the nearby environment through the process of containment, emptying, transport, and disposal, despite high coverage (78 percent) of on-site septic tanks. •• While trends in access have increased in parallel since 2002, 80 percent of the population in urban areas has access to improved drinking water supply, compared to just 60 percent in rural areas. •• Gaps in access to improved water between B40 and T60 households remain, most starkly in urban areas. •• Hibah-participating kotas demonstrate higher, but not necessarily more equitable, access to piped water supply. •• Data on water quality suggest that the potential risk of contamination is severe, especially in more dense urban settings. Improving Service Levels and Impact on the Poor 29 In 2015 the WHO/UNICEF JMP estimated that access to improved drinking water was 87  percent in Indonesia, a 39 percentage point increase since 1990, and access to improved sanitation was 61 percent, a 36 percentage point increase. Indonesia achieved the MDG target for water, and while good progress was made towards the sanitation goal, it was not achieved. There are still close to 100 million people without improved sanitation and 33 million without improved drinking water, according to the JMP for global monitoring purposes WHO and UNICEF (2015). When considering inequalities in access, Joint Monitoring Programme (JMP) estimates show that overall access levels to sanitation are lower among the poor and lower in rural areas. The B40 households—especially those in rural areas—have not gained access to improved sanitation at the same rate as their T60 counterparts. Trends for drinking water appear more equitable, although a relatively low share of the rural population drinks piped water on premises (figure 4.1). The sharp increase in unimproved drinking water is partly due to households—particularly those in the upper wealth quintiles in urban areas—drinking bottled water as their primary source, which JMP does not categorize as improved. As discussed in this report, the vast majority of bottled water users have access to another source of improved water, reflecting a change in preferences rather than access. Stopping Open Defecation in Indonesia: A Global Success Story In 2015 the WHO/UNICEF JMP estimated that access to improved sanitation was 61 percent, a 36 percentage point increase since 1990 (WHO and UNICEF 2015).1 Although this increase was substantial, it fell short of achieving the MDG sanitation target of reducing by half the proportion of the population without access to sanitation. While there are still close to 100 million people without improved sanitation, Indonesia has become a global success story due to the rapid and sustained reduction in the practice of open defecation that was achieved in less than a decade. Community-Led Total Sanitation (CLTS) was adopted nationwide in 2005, leading to the launch of the STBM by MoH in 2008. These combined approaches and programs heralded a major shift in the trajectory of sanitation in the country, and between 2006 and 2015 access to improved sanitation grew at 6.5 percent annually, up from an annual growth rate of 3.4 percent in the years between 2002 and 2004. STBM is a community-based total sanitation strategy focused on behavior change at the household and community levels to cut off contamination pathways for E. coli, a primary pathogen causing diarrheal disease. STBM includes 5 pillars: (1) stop open defecation (including using a community “triggering” approach to motivate people to stop open defecation); (2) handwashing with soap; (3) household safe water treatment and storage, and safe food handling; (4) safe disposal and management of solid waste; and (5) safe disposal and management of wastewater. The open defecation pillar of STBM was a major departure from previous publicly funded toilet construction programs in Indonesia as it explicitly shunned household subsidies and embraced a community mobilization approach to stopping open defecation. STBM The STBM program approached the open defecation problem from three directions, with the prioritized an aim of achieving “total sanitation.” It aimed to improve the enabling policy environment to improved policy make sanitation a priority for LGs, to create demand for safe sanitation and hygiene through environment, community empowerment and behavior change, and to improve the supply of sanitation demand creation, and products and services so that households had access to the materials they needed at a price sanitation service they could afford. The program only provided subsidies to build communal sanitation facilities improvement and implemented a community award to incentivize achievement and sustainability of total to address the sanitation. STBM has mainly been a rural-focused program, but was recently expanded to open defecation urban settings. problem. 30 Improving Service Levels and Impact on the Poor Figure 4.1: Access to Sanitation and Drinking Water, by Urban and Rural Wealth Quintile Sanitation trends by Sanitation trends by rural wealth quintile urban wealth quintile 100 100 3 1 5 20 11 7 1 48 21 8 5 1 31 13 10 7 8 1 96 24 17 3 24 12 80 5 28 19 41 14 18 7 89 80 2 21 19 22 10 51 49 24 5 60 78 21 14 3 83 56 60 30 15 7 67 20 19 27 40 27 26 10 21 6 40 47 26 41 59 8 19 10 20 20 27 11 71 88 13 12 51 32 45 7 17 26 57 0 0 Poorest Second Middle Fourth Richest Poorest Second Middle Fourth Richest Trends in sanitation coverage (%) by rural Trends in sanitation coverage (%) by urban wealth quintile from 1995 to 2012 wealth quintile from 1995 to 2012 Open defecation Other unimproved Shared Improved Sanitation trends by Sanitation trends by rural wealth quintile urban wealth quintile 100 100 36 33 29 29 17 10 5 0 0 49 15 26 30 21 80 80 27 34 33 68 47 51 61 67 75 35 60 44 58 20 60 60 56 56 31 46 40 40 37 59 40 48 43 43 20 20 75 27 33 22 10 20 23 54 63 70 20 32 53 1 3 2 8 2 9 3 11 20 0 0 Poorest Second Middle Fourth Richest Poorest Second Middle Fourth Richest Trends in sanitation coverage (%) by rural Trends in sanitation coverage (%) by urban wealth quintile from 1995 to 2012 wealth quintile from 1995 to 2012 Unimproved Other improved Piped on premises Source: WHO/UNICEF JMP, 2015. Improving Service Levels and Impact on the Poor 31 Location Matters: Subnational Estimates of Sanitation Coverage and Associations with Poverty Where people At provincial level, there is no clear correlation between poverty and access to improved live—not just sanitation, pointing to the presence of other factors affecting sanitation coverage. The following how much map (map 4.1) shows the geographic dispersion of poverty and access rates at provincial level they earn— for the most recent year data are available (2015). Some provinces with above average poverty determines access to rates have above average sanitation coverage, while others with below average poverty rates improved water have below average sanitation. For example, the provinces that make up the island of Java and sanitation. mostly have above average poverty levels, but also enjoy above average access, while the Western Indonesian island of Sumatra generally has lower levels of access irrespective of poverty levels. Eastern Indonesia—especially Papua, Maluku, East Nusa Tenggara, and Central and West Sulawesi—has high poverty levels and low access, while Southeast Sulawesi and West Papua have above average levels of access despite having above average levels of poverty. These findings point to the presence of factors other than poverty that influence sanitation coverage. These are explored in map 4.1. At district level there is a trend toward improvement in sanitation coverage over time, but there is wide variance across districts and not only according to poverty levels (figure 4.2). There are some poor districts that are doing a better job than wealthier districts with regard to sanitation coverage, and a far better job than some of their poorer peers. The implication of these discrepancies is that strategies that improve coverage in low-poverty districts may be vastly different from those that would work in high-poverty districts. To date, the national STBM program has relied on community empowerment and behavior Map 4.1: Poverty Rate and Access to Improved Sanitation, by Province, 2015 Categorization based on proportion Below average proportion of poverty, above average proportion of access Above average proportion of poverty, above average proportion of access Below average proportion of poverty, below average proportion of access Above average proportion of poverty, below average proportion of access No data Source: Susenas. 32 Improving Service Levels and Impact on the Poor Figure 4.2: Correlation between Poverty Rate and Improved Sanitation Access in Kabupatens/Districts, 2002–15 a. 2002 b. 2007 c.2010 d. 2015 100 100 100 100 90 90 90 90 80 80 80 80 Improved Sanitation(%) Improved Sanitation(%) Improved Sanitation(%) Improved Sanitation(%) 70 70 70 70 60 60 60 60 50 50 50 50 40 40 40 40 30 30 30 30 20 20 20 20 10 10 10 10 0 10 20 30 40 50 0 10 20 30 40 50 0 10 20 30 40 50 0 10 20 30 40 50 Poverty rate(%) Poverty rate(%) Poverty rate(%) Poverty rate(%) Improved Sanitation Smoothing Source: Susenas. change—an approach that may work better in some areas than others due to many underlying factors, including the quality of program implementation (Cameron et al. 2017), levels of social capital in the community (Cameron et al. 2015), and participation of local champions. Another contextual factor could be population density. Lower population density may make it more difficult to mobilize households and bring the community together to “trigger” behavior change. For rural sanitation, district investment priorities and fiscal capacity likely play a smaller role, since government expenditures on infrastructure and implementation of STBM are minor. Poverty rates do not neatly correspond with either levels of access or equity of access to improved sanitation at the district level. Each pair of bars in figure 4.3 illustrates the share of the T60 and B40 with access to improved sanitation in the district (n=412) and the district poverty rate (the national average poverty rate in 2015 of 11.2 percent is represented by the red line). Almost all districts show higher coverage among the T60. While a larger concentration of high-poverty districts are clustered in the left side of the figure, where overall access levels are lower, both above average and below average poverty districts are found across the distribution of coverage levels. Distinguishing between Java and non-Java districts reveals a noticeable difference in coverage levels between B40 and T60 households (figure 4.4). Despite there being no significant difference in poverty levels between Java and non-Java districts (12 and 13 percent, respectively), Java districts have achieved higher levels of coverage overall for both B40 and T60 households. The average level of improved sanitation coverage among B40 in non-Java districts is 35 percent, compared with 51 percent in Java districts, while it is 67 percent among T60 in Java districts, compared with 51 percent in non-Java. While the average gap between B40 and T60 households is similar across Java and non-Java districts, there is more variation in the gap in non-Java districts. In addition, there appears to be a slight downward trend in the gap in Java districts as higher levels of coverage are achieved. Improving Service Levels and Impact on the Poor 33 Figure 4.3: Share of T60 and B40 with Improved Sanitation and Poverty Rate, by District (Kabupaten) 100 100 90 90 80 80 70 70 Percent 60 60 Percent 50 50 40 40 30 30 20 20 10 10 0 0 T60 B40 % poor Source: Susenas. Place seems to be a stronger driver of access to sanitation than poverty. Both urban and rural areas have achieved substantial progress in sanitation since 2002, with the steepest gains seen in rural areas—a difference likely due to those areas having had lower levels of access to begin with. In 2002 just 18 percent of the rural population had improved sanitation and 41 percent were defecating in the open. In 2015 these figures were 48 percent and 20 percent respectively. While these gains were large, rural areas have not yet reached the levels of access seen in urban areas. The most recent data for 2015 show that 76 percent of the population in urban areas have improved sanitation compared with just 48 percent in rural areas, a gap of 28 percentage points (figure 4.5). Moreover, households in the T60 of the income distribution are more likely to have access to sanitation facilities within urban or rural areas, but the same is not true across these areas (figure 4.6). The B40 in rural areas are most likely to defecate in the open—29 percent do not have a toilet, compared with 15 percent of the T60; in urban areas, 12 percent of the B40 and 3 percent of the T60 do not have a toilet. In other words, households belonging to the top 60 percent of the income distribution in rural areas are worse off than those in the bottom 40 in urban areas. A possible explanation could be the fundamental assumptions of the STBM approach, requiring LGs to conduct community empowerment and behavior change activities, and households to seek out and procure their own materials to improve their sanitation. Low-density rural settings can make these requirements particularly challenging. While coverage of sanitation is higher in urban areas, the location within a city seems to have an effect on access. Cities in Indonesia face challenges to service delivery driven by rapid population increase, inadequate planning, and underinvestment in infrastructure. In four of Indonesia’s largest cities—Jakarta, Medan, Surabaya, and Makassar—the suburban or peripheries of these cities have both higher levels of poverty and lower levels of access than core urban centers (figure 4.7). In Makassar coverage is 25 percentage 34 Improving Service Levels and Impact on the Poor Figure 4.4: Share of T60 and B40 with Improved Sanitation and Poverty Rate, by District (Kabupaten) in Java and Non-Java a. Java districts 100 55 90 45 80 35 70 60 25 Percent Percent 50 15 40 5 30 –5 20 10 –15 0 –25 b. Non-Java districts 100 55 90 45 80 35 70 25 60 Percent Percent 50 15 40 5 30 –5 20 –15 10 0 –25 T60 B40 Gap Source: Susenas. points lower in suburban areas than in the city’s core. In Jakarta the difference is 15  percentage points. Suburban areas of Makassar also have higher levels of poverty (8 percent) than does Makassar’s core (1 percent). In contrast, there is a 15 percentage point gap in access between the Jakarta core and Jakarta suburban areas, yet rates of poverty are the same (3 percent). In all cities, progress is notable in the periphery areas— doubling or near doubling access since 2002. These areas now enjoy higher levels of access than the national average, although they still lag behind the urban cores. Urban slums remain a major challenge in Indonesian cities and are a visible marker of urban poverty and the gaps in access to basic infrastucture. An estimated 22 percent of Indonesia’s urban population (approximately 29 million people) is living in slums with low levels of access to basic services. In 2014, it was estimated that 30  percent of slum dwellers (9 million people) lacked safe drinking water, and 37 percent (11 million people) lacked sanitation.2 Improving Service Levels and Impact on the Poor 35 Figure 4.5: Access to Improved Sanitation in Urban and Rural Areas, 2002–13 a. Urban 100 90 80 70 60 Percent 50 40 30 20 10 0 2002 2004 2007 2010 2012 2015 b. Rural 100 90 80 70 60 Percent 50 40 30 20 10 0 2002 2004 2007 2010 2012 2015 Open defecation Unimproved Improved Source: Susenas. Figure 4.6: Percentage of B40 and T60 Population Practicing Open Defecation in Urban and Rural Areas 50 40 30 Percent 20 10 0 2000 2002 2004 2006 2008 2010 2012 2014 2016 B40-Urban T60-Urban B40-Rural T60-Rural Source: Susenas. 36 Improving Service Levels and Impact on the Poor Figure 4.7: Trend in Improved Sanitation Access in Indonesia’s Largest Cities 100 100 90 90 80 80 70 70 60 60 Percent Percent 50 50 40 40 30 30 20 20 10 10 0 0 2002 2015 2002 2015 2002 2015 2002 2015 2002 2015 Jakarta Medan Surabaya Makassar Core Sub-urban / Periphery Urban National Source: Susenas. Second Generation Challenges for Rural Indonesia: Moving up the “Sanitation Ladder” There are still 47 million people defecating in the open in Indonesia and another 52 million using Despite improved sanitation that is considered unsafe, most of these living in rural areas. When access to improved access to sanitation began to accelerate in 2007 on the heels of the new national strategy and total sanitation in sanitation program, the acceleration for rural households was higher in the top three consumption urban areas, high levels of fecal quintiles, and that gap has persisted to the present (figure 4.8). The most recent evidence contamination points to a gap in access to improved sanitation in rural areas of 20 percentage points (2015). persist. Open defecation has declined at similar rates in the B40 and T60 since 2002, but B40 households were more likely to move to basic latrines,3 contributing to the widening gap in access to improved sanitation between the B40 and T60. This finding is consistent with a 2011 Impact Evaluation of STBM in East Java, which found that poor households— defined in the study as those in the bottom 20 percent of the distribution of non-land assets—did not improve their sanitation as a result of the program, while those in the top 80 percent were more likely to construct toilets, stop defecating in the open, and correctly dispose of child feces (Cameron and Shah 2011). In 2015 more than half (55 percent) of the households in rural areas either had no toilet or were using unimproved latrines. Poor quality pit latrines fail to adequately isolate human waste from the environment and create breeding grounds for flies and other vectors that spread disease. The largest share of unimproved sanitation for the most recent year of data (2015) are facilities that dispose of waste into fields, water bodies, or open land, essentially equivalent to open defecation. Analysis using RISKESDAS data for 2013 shows that in rural settings only 27 percent of children live in communities with sanitation coverage greater than 75 percent, but evidence from Indonesia shows that two-thirds of the gains in average height-for-age z-scores (HAZ) accrue after a threshold of around 60 percent percent coverage is surpassed, and that full benefits may only be achieved when coverage becomes universal. As a result, the majority of children in rural areas are exposed to a contaminated environment even if they themselves live in households with sanitation (Skoufias 2016). The primary barrier to toilet construction cited by households is high cost. In the 2011 Impact Evaluation of STBM, the average reported expenditure necessary to construct a toilet was Improving Service Levels and Impact on the Poor 37 Figure 4.8: Growing Inequality in Improved Sanitation Access in Rural Settings for B40 versus T60 70 60 50 40 30 20 10 0 2002 2015 Improved sanitation B40 T60 Source: Susenas. US$119, equivalent to 94 percent of one month’s average (reported) household income (Chase and Briceno unpublished report). Such a large lump sum outlay can be impossible for poor households to afford, but credit and formal savings products that would allow households to spread payments over time are not widespread, limiting the ability of poor households to invest in improved sanitation. Why High Levels of Sanitation in Urban Settings Have Not Fully Eliminated Fecal Contamination Despite high levels of access to improved sanitation in urban settings, low levels of service quality and a heavy reliance on households to manage their fecal waste result in massive levels of environmental contamination. The vast majority of households (78 percent) in urban areas use an improved toilet connected to a septic or sewerage system (“septic tanks” or, more often, pit latrines or soak pits); of this 78 percent, less than 2 percent are estimated to be sewerage connections. Lateral sewers and household connections are often excluded from central government and development partner investments, resulting in 50 percent idle capacity on average for existing networks in 13 cities (World Bank 2015). At the same time, while there has been an increase in construction of septage treatment plants, these investments have not been followed by improvements to fecal sludge management or upgrades to on-site sanitation. An assessment conducted by MoPWH in 2012 found that over 90 percent of existing sludge treatment plants built since 1990 (n=150) are no longer operational or are poorly performing. As a result of these conditions, most fecal waste (95 percent) makes its way into the nearby environment through the process of containment, emptying, transport, and disposal (figure 4.9). The remaining 5 percent of fecal waste is delivered to a treatment plant, but even these plants are in desperate need of improvement. Conditions of high population density and inadequate fecal waste disposal interact to make poor sanitation particularly risky for the health of populations in urban areas. Thirty-eight percent of children in urban settings live in communities with coverage of sanitation below 75 percent (Skoufias 2016). However, the measure of “community” used in this study is less meaningful in urban settings, where biological contaminants can rapidly spread among urban populations through groundwater supplies and piped water systems, and during heavy rains and flooding. 38 Improving Service Levels and Impact on the Poor Figure 4.9: Fecal Waste Management Flow Chart Direct sewerage Total <1% Wastewater wastewater safely collected treated Communal 1% sewerage Urban <0.5% Septage safely Septage safely population collected treated/disposed 110 million Septic tanks 4% no sewerage 62% Septage and wastewater Other on-site unsafely disposed <23% 95% Open defecation 14% Source: EAP Urban Sanitation Review Indonesia Country Study, 2013. For example, in Jakarta, where an estimated 63 percent of waste is being discharged into the ground untreated, four-fifths of the economic losses associated with existing sanitation conditions—estimated at US$1.4 billion per year (or US$139 per person)—can be attributed to health-related impacts (World Bank 2016c). A  recent study shows that children living in households with poor fecal waste disposal practices, such as toilets that discharge directly into drains, had 3.78 times higher prevalence of enteric infection than children in other households, including those without toilets (Berendes et al. 2017). On-site sanitation systems in Indonesian cities do not achieve effluent quality standards, especially in challenging environments.4 Recent research has shown that not one of the seven types of on-site sanitation systems sampled in five cities was in compliance with effluent standards of less than 3,000/100 mL fecal coliform (figure 4.10). Most on-site sanitation in Indonesia is based on the anaerobic system, which does not reduce microbiological content, allowing untreated or partially treated wastewater to seep into the ground. The absence of practical national standards for on-site sanitation systems in challenging areas, inadequate quality of construction, and lack of knowledge on operations and maintenance are additional factors leading to high fecal coliform counts (World Bank 2017 Forthcoming). The Evolution of Drinking Water Supply in Indonesia In 2015 the WHO/UNICEF JMP concluded that Indonesia had achieved the MDG target on drinking water supply. An estimated 87 percent of the population had access to improved drinking water in 2015—a 39 percentage point increase since 1990. Due to different calculation methods, GoI estimated 70 percent coverage of improved drinking water (see box 1.1 for an explanation of the calculation method),5 which translates into annual growth of 2.7 percent between 2002 and 2015. Still, there are notable disparities in access to improved water supply and type of water source between urban and rural settings. While access to improved water supply has increased in parallel for urban and rural areas since 2002, in urban areas 80 percent of the population has access, compared with only 60 percent in rural areas (figure 4.11). A small share of the population (3 percent) continues to use surface water in rural areas, down from 5 percent in Improving Service Levels and Impact on the Poor 39 Figure 4.10: Results from Study of On-Site Sanitation for Dense Urban Areas in Indonesia E. coli content found within the effluent 15,000 12,000 9,000 Mg/L 6,000 3,000 0 Modular Tank with Tank with Tank with Horizontal Tank with Upflow septic tank 3 chamber 3 chamber 2 chamber septic 1 chamber anaerobic 3 filter 2 filter 1 filter tank and drainfield filter tank E. coli Source: World Bank 2017. Box 4.1: Increased Bottled Water Usage Distorts Trends in Access to Piped Water Supply: Data and Definitions The rapid expansion of bottled water usage and household preferences for drinking bottled water can distort access trends if analyses fail to account for other household sources of water. Until 2011, bottled water for drinking was not considered an improved or sustainable source of drinking water in official statistics. However, as bottled water use began to increase—from less than 2 percent in 2002 to 29.5 percent in 2015—the rise led to distorted access figures for other water sources, most notably piped water. In response, Susenas began collecting data in 2011 on the source of water used for bathing and washing for households whose primary drinking water source is bottled water. These surveys show that the vast majority of bottled water drinkers use another source of improved water in the home for bathing and washing (surveys available for the years 2011 to 2015). Since 2011, bottled water has been classified as an improved source if the household uses a safe and sustainable water source for bathing and washing. Thus, 92 percent of bottled water users are classified as having improved water. 40 Improving Service Levels and Impact on the Poor Figure 4.11: Access to Improved Drinking Water in Urban and Rural Areas, 2002–13 a. Urban b. Rural 100 100 90 90 80 80 70 70 60 60 Percent Percent 50 50 40 40 30 30 20 20 10 10 0 0 2002 2004 2007 2010 2012 2015 2002 2004 2007 2010 2012 2015 Surface Water Unimproved Improved Source: Susenas. 2002; a larger share of rural households uses protected wells (28 percent) and springs (15 percent) than use them in urban areas (16 and 3 percent, respectively). Just 6 percent of rural households use tap water as their primary source of drinking water, compared with 16 percent of urban households. Notably, 44 and 14 percent of urban and rural households respectively drink bottled water as their primary source (box 4.1). Drinking Water Access and the Poor As of 2015 there are still an estimated 20 million households in Indonesia using unimproved Disparities persist in drinking water sources. Of these 20 million unserved households, the majority—13.5 million— access to improved are located in rural areas, and 43 percent of these (5.8 million households) were classified as drinking water poor and vulnerable. Nevertheless, inequalities in access to improved drinking water by income between poor and wealthy households, distribution are modest and have lessened over time. Figure 4.12 shows the extent of income- and between related inequality in access to improved drinking water between households at different points urban and rural along the income distribution. The diagonal line in the graph represents perfect equality—in populations. other words, when the share of the population in the first 20 percent of the income distribution accounts for a 20 percent share of access to drinking water. Between 2002 and 2007 there was no change in inequality in access, but a trend towards greater equality in access can be seen between 2007 and 2015. In 2015 the bottom 20 percent accounted for approximately 17 percent of those with access to drinking water. Despite these improvements, there are still gaps in access to improved water between B40 and T60 households. While the gaps are greater in urban areas, they are growing in rural areas. In 2015 the gap in access to improved water in urban areas dropped to 14 percentage points— returning to its 2002 level—although the gap had been trending at around 17 percentage points since 2010. In rural areas, there was a 10 percentage point gap between B40 and T60 in access to improved water in 2015. The gap has steadily increased from 6 percentage points in 2002. The driver of the growing gap in access to improved water sources in rural areas is not clear from existing data. The government’s main platform for expanding access to rural water supply, Improving Service Levels and Impact on the Poor 41 Figure 4.12: Inequalities in Access to Improved Drinking Water by Income Distribution 1.0 1.0 Cum. papulation share with improrved water 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0 0 0.2 0.4 0.6 0.8 1.0 Cum. proportion of income 2002 2007 2015 Equality Source: Susenas. Figure 4.13: PAMSIMAS Water Supply Beneficiaries, T60–B40, Piped–Non-Piped, 2012–15 70 60 50 Beneficiaries (millions) 40 30 20 10 0 2012 2013 2014 2015 T60 piped T60 non-piped B40 piped B40 non-piped Source: Management Information System PAMSIMAS. PAMSIMAS, targets underserved and low-income communities. The program began in 2006, and by the end of its second phase in 2015 it had reportedly provided access to clean drinking water for an estimated 9 million beneficiaries, or 3.5 percent of the population in Indonesia (PAMSIMAS 2015). Program data for the years 2012 to 2015 indicate the vast majority of water services provided were piped, with nearly equal numbers of beneficiaries identified as poor and non-poor6 (figure 4.13). 42 Improving Service Levels and Impact on the Poor Figure 4.14: Change in Primary Source of Drinking Water by Consumption Quintile, 2002–15 60 50 Percentage of increased use 40 30 20 10 0 –10 –20 –30 p w ace er w ed rin d ed w ed rin d m / Ta pu ole sp cte sp cte at er ct l g ttl t l g p el rf el ec w h at te te Bo Su e re ot ot in ro ro Bo Ra Pr Pr np np U U Q1 Q2 Q3 Q4 Q5 Source: Susenas. Whether improved or unimproved, the source of drinking water reported by households differs by consumption quintile. Household data by consumption quintile show that the use of piped (tap) water for drinking has declined across all quintiles, but most dramatically in the top-most quintile. There were also large declines in the use of well water, both protected and unprotected, across all quintiles. Using bottled water for drinking has increased by 50 percentage points in  the top quintile and 12 percentage points in the poorest 20 percent of households by consumption (figure 4.14). Location Matters: Subnational Estimates of Access to Water Supply and Associations with Poverty As with sanitation, access to improved drinking water is strongly determined by geographic location. Coverage of improved drinking water is higher in T60 households than in B40 households. However, being in an urban setting is a stronger determinant of access: B40 households in urban areas are better off than T60 households in rural settings (figure 4.15). In 2015, 37 percent of rural households drank unimproved water. Households in high-poverty rural districts were less likely to drink piped water as their primary source, and the overall proportion of piped water (including public taps from small rural piped systems) is low (7 percent) compared with urban areas. There is not a clear association between district poverty rates and access to piped water in urban districts (kotas). Each pair of bars in figure 4.16 illustrates the share of T60 and B40 households with access to piped water (including primary and secondary sources) in that district, with the district poverty rate represented by a diamond. In the majority of kotas there is little to no gap; 29 percent of kotas have higher levels of access among the T60, and 34 Improving Service Levels and Impact on the Poor 43 Figure 4.15: Drinking Water Access in Rural and Urban Settings for B40 versus T60, 2002–15 100 80 60 Percent 40 20 0 2000 2002 2004 2006 2008 2010 2012 2014 2016 B40-Urban T60-Urban B40-Rural T60-Rural Source: Susenas. Figure 4.16: Share of T60 and B40 with Piped Water Access (Primary and Secondary) and Poverty Rate in Urban Districts (Kota) 100 45 90 40 80 Piped water coverage (percent) 35 70 Poverty rate (percent) 30 60 25 50 20 40 15 30 20 10 10 5 0 0 T60 B40 % poor Source: Susenas 2015. percent of kotas have higher levels of access among the B40. Districts with above average poverty rates are mostly clustered in the lower half of the distribution, but most of the kotas have below average poverty. Some districts are doing much better than others, irrespective of poverty. Hibah-participating kotas show higher, but not necessarily more equitable, access to piped water supply. The performance-based Water Hibah program has contributed to an estimated 265,000 piped water supply connections for poor urban households in 151 LGs since 2012 (Indii 2014a). For the 26 Hibah-participating kotas, access to piped water supply is significantly higher (5 percent) than in non-Hibah participating kotas for both T60 and B40 households. 44 Improving Service Levels and Impact on the Poor However, there is no statistically significant difference in the average gap in access between T60 and B40 households (figure 4.17) across the two groups of districts.7 Hibah-participating districts had a larger gap in access between T60 and B40 than non-Hibah participating districts did in 2012 (13.5 percent vs. 10.3 percent p=0.13). The gap declined for both groups of  districts in 2015, but remained larger in Hibah-participating districts (11.0 percent vs. 8.4 percent p=0.20). However, this does not necessarily imply that the Hibah program is not achieving its pro-poor objectives, since the evidence suggests that the program indeed targets kotas with larger gaps in access to begin with. Small and big cities increased the share of households with piped water connections between 2011 and 2015, despite rapid population growth averaging around 2 percent per year (figure 4.18). Piped water connections in metropolitan areas declined by 2.8 percentage points over the period between 2011 and 2015, and these cities also experienced growth of 2.5 percentage points per year. While Jakarta experienced one of the lowest population growth rates (1.2 percent) among cities categories,8 the share of households in Jakarta with piped connections declined by 1 percentage point. Medium-sized cities of between 100,000 and Figure 4.17: Share of T60 and B40 with Piped Water Access (Primary and Secondary) by Hibah and Non-Hibah Participating Urban Districts (Kota) a. Hibah participating Kotas 100 15 90 10 80 5 70 0 Percent 60 –5 50 –10 40 –15 30 20 –20 10 –25 0 –30 b. non-Hibah participating Kotas 100 15 90 10 80 5 70 0 60 Percent 50 –5 40 –10 30 –15 20 –20 10 0 –25 T60 Bottom 40% Gap Source: Susenas 2015 and CPMU Hibah, MOPWH. Improving Service Levels and Impact on the Poor 45 Figure 4.18: Piped Water Access (Primary and Secondary Source) and Population Growth Rate, by City Category, 2011–15 50 Piped water access (%) 45 40 Rate of change Average population City category 2011–15 (%) growth rate (%) 35 Jakarta –0.6% 1.2% Metropolitan –1.5% 2.5% 30 2011 2012 2013 2015 Big city 2.3% 2.1% Medium city –0.8% 2.3% Jakarta Metropolitan Big city Medium city Small city Small city 1.2% 2.0% Source: Susenas. Note: Results are not representative at levels below district level. 500,000 inhabitants have, at 35 percent, the lowest share of piped water connections, while small cities have the highest at 45 percent. Small, medium, and big cities—all experiencing annual growth of over 2 percent per year since 2011— maintain the widest gaps in access to piped water between the B40 and T60 households (figure 4.19). In small cities 37 percent of B40 households have piped water connections compared with 49 percent of T60. Meanwhile, inequality gaps are closing in metropolitan cities and have reversed in Jakarta, where 42 percent of B40 households have piped water connections compared with 39 percent of T60. Water Quality: Little Information on a Potentially Widespread Problem A recent water quality survey was conducted in the city of Yogyakarta, finding that nearly all sources of improved water were contaminated with E. coli.9 Wells, boreholes, and protected springs had the highest levels of contamination (90 percent), while piped water was lower at 77 percent. Ready-to-drink samples that were reportedly treated by boiling, filtration, or another method, showed similarly high levels of fecal contamination: 68.9 percent for water coming from wells, boreholes and protected springs; 73 percent for piped water; and 52 percent for bottled water. The fact that 73 percent of samples from piped water sources were contaminated even after reported treatment suggests that the water is not being properly treated, is not treated to the same degree as water from other sources, or is becoming re-contaminated due to unhygienic storage (not covered and/or exposed to direct contact with flies, dust, and dirt). Available evidence suggests that PDAM water quality is likely to be a significant problem, especially with dilapidated piped networks, groundwater contamination, intermittent service, and low pressure. There is limited public transparency regarding PDAM water quality results, and MoH reports that only 23 percent of drinking water is currently tested for quality, with a target of just 50 percent by 2019 (MoH 2015). Indeed, District water utility (PDAMs) are not 46 Improving Service Levels and Impact on the Poor Figure 4.19: Piped Water Access (Primary and Secondary Source) for B40 versus T60, by City Category, 2011–15 a. Metropolitan b. Jakarta 60 60 50 50 40 40 Percent Percent 30 30 20 20 10 10 0 0 2011 2012 2013 2014 2015 2011 2012 2013 2014 2015 T60 Metropolitan B40 Metropolitan T60 Jakarta B40 Jakarta c. Small city d. Medium city 60 60 50 50 40 40 Percent Percent 30 30 20 20 10 10 0 0 2011 2012 2013 2014 2015 2011 2012 2013 2014 2015 T60 Small city B40 Small city T60 Medium city B40 Medium city e. Big city 60 50 40 Percent 30 20 10 0 2011 2012 2013 2014 2015 T60 Big city B40 Big city Source: Susenas. legally required to provide clean water: the current Regulation on Drinking Water Supply Systems (Government Regulation No. 122/2015) allows PDAMs to provide water that requires one further stage of treatment before drinking, placing the burden on households to treat piped water before drinking. Groundwater quality is not consistently monitored, but available evidence indicates that the potential risk for contamination is severe, especially in dense urban settings. A study of groundwater quality in large cities in Java shows high levels of contamination from septic Improving Service Levels and Impact on the Poor 47 Figure 4.20: Households Using Groundwater as Primary Drinking Water Source and Distance to Feces Containment B40 versus T60 in Urban Areas 100 90 80 70 60 Percent 50 40 30 20 10 0 '02-B40 '15-B40 '02-T60 '15-T60 Improved groundwater >10m Unimproved groundwater or improved groundwater <10m Source: Susenas. tanks and untreated domestic wastewater, as well as landfill and industrial effluent contamination.10 In 2015, 57 percent of the B40 in urban areas used groundwater sources, a decline from 73  percent in 2002. A smaller proportion of the T60, 32 percent, used groundwater sources (figure 4.20). Although use of groundwater for drinking is declining, estimates show an increasing share of households are using a protected groundwater source for purposes other than drinking. Leaving aside industrial and other forms of contamination, nearly half (43 percent) of “protected” groundwater sources used by the B40 are less than 10 meters from an excreta disposal site such as a cubluk (wet pit latrine) or septic tank (often poorly constructed and not properly sealed). Adding these to the unprotected sources implies that in 2015 over a quarter (27 percent) of the B40 drink unsafe groundwater, though this has fallen from over half (52 percent) in 2002. For the T60, meanwhile, the share has fallen from 35 to 14 percent. Contaminated groundwater is particularly an issue in urban settings where there is overcrowding, poor quality septic tanks, and lack of fecal waste management. Building on the challenges identified in this chapter, the following chapter will outline the constraints on, and opportunities for, delivery of piped water services to the poor in urban settings. Although piped water service delivery in urban settings is not the only challenge facing the water sector and the GoI, it was prioritized for deeper investigation on the basis of current patterns in equity of access, rapid urbanization in Indonesian cities, and the underlying financial and performance hurdles in the urban water sector, which together present a unique challenge to extending access to low-income households. Notes 1. The GoI applies stricter criteria to classify improved sanitation and only considers private or shared pour flush latrines that dispose of feces in a septic tank as improved, estimating that 62 percent of the population had access to improved sanitation in 2015. 48 Improving Service Levels and Impact on the Poor 2. World Bank, Project Appraisal Document of national Slum Upgrading Project, June 2016. 3. A basic latrine is one that does not hygienically separate feces from the environment as opposed to an improved latrine. 4. environments such as dense urban areas, high groundwater, frequently flooded areas, and housing built along the coastline or beside rivers and lakes. 5. GoI exclusively uses Susenas to calculate estimates of access to water supply and requires that groundwater sources (borehole, well, or spring) are located at least 10 meters distance from the feces containment structure (see box 1.1) in order to be considered improved. 6. PAMSIMAS management information system identifies “poor” and “non-poor” beneficiaries. 7. Eligibility criteria for Hibah is based on a household’s power voltage and household assets, rather than income poverty. 8. Categories are based on government regulation No.26 Year 2008 and enforced through Policy and Strategy for National Urban Development (KSPPN) issued by Bappenas as a basis for urban development for 2015–45. 9. Survey was conducted alongside March 2016 Susenas data collection by BPS, UNICEF, Bappenas and MoH. 10. World Bank, Java Water Resources Strategies Study Report, 2012. References Berendes, D., A. Kirby, J. A. Clennon, S. Raj, H. Yakubu, J. Leon, K. Robb, A. Kartikeyan, P.  Hemavathy, A. Gunasekaran, and B. Ghale. 2017. “The Influence of Household-and Community-Level Sanitation and Fecal Sludge Management on Urban Fecal Contamination in Households and Drains and Enteric Infection in Children.” The American Journal of Tropical Medicine and Hygiene 96 (6): 1404–14. Cameron, L., and M. Shah. 2011. “Risk-Taking Behavior in the Wake of Natural Disasters.” Discussion paper. University of California, Irvine. Cameron, Lisa A., Susan Olivia, and Manisha Shah. 2015. “Initial Conditions Matter: Social Capital and Participatory Development.” IZA Discussion Paper 9563. Available at SSRN: https://ssrn.com/abstract=2708376 Cameron, Lisa A., and Manisha Shah. 2017. “Scaling Up Sanitation: Evidence from an RCT in  Indonesia.” IZA Discussion Paper 10619. Available at SSRN: https://ssrn.com​ /­abstract=2940609 IndII (Indonesia Infrastructure Initiative). 2014a. Output-based Financing for Urban Water Supply, Towards a Nationwide Water and Sanitation Hibah in 2015-2019. ———. 2014b. Journal of the Indonesia Infrastructure Initiative—PRAKARSA. Jakarta: Australia Indonesia Partnership. MoH. 2015. Mewujudkan Aksesibilitas Air Minum dan Sanitasi yang Aman dan Berkelanjutan Bagi Semua. BPS, Bappenas, MoH and UNICEF. PAMSIMAS. 2015. Management Information System PAMSIMAS. http://mis.pamsimas.org​ report/kpi_new/ (accessed October 25, 2016). /­ Skoufias, E. 2016. Operationalizing a Multi-Sectoral Approach for the Reduction of Malnutrition in Indonesia: An Application using the 2007 and 2013 RISKESDAS. Washington DC: World Bank. Susenas. 2015. Survei Ekonomi Nasional. Badan Pusat Statistik. Improving Service Levels and Impact on the Poor 49 WHO and UNICEF. 2015. Progress on Drinking Water and Sanitation: 2015 Update and MDG Assessment. WHO/UNICEF Joint Monitoring Programme for Water Supply and Sanitation. Geneva: WHO and New York: UNICEF. World Bank. 2015. An Unfair Start: How Unequal Opportunities Affect Indonesia’s Children. Jakarta: World Bank. ———. 2016c. Poor Sanitation Costs Jakarta about IDR 16.2 Trillion (USD 1.4 Billion) per Year— The Economic Impacts of Sanitation. Jakarta: World Bank. ———. 2017. “Study on Small Onsite Sanitation in High Dense and Challenging Areas in Indonesia.” Jakarta: World Bank. 50 Improving Service Levels and Impact on the Poor Chapter 5 Urban Water Service Delivery Constraints on and Opportunities for Reaching the Poor Core Question: What are the WASH service-delivery constraints and potential solutions to improving services to the poor and bottom 40 percent? Key facts • The GoI’s 100-0-100 target is for universal access to improved water supply, but current patterns in equity of access to piped water suggest that low-income households are likely to remain on a non-piped service for longer than T60 households. • The barriers preventing low-income households from accessing piped water connections include (a) government budget allocation and spending; (b) financial sustainability and performance of PDAMs; (c) perceptions and behavioral constraints; and (d) legal frameworks for equitable service delivery. • At current levels, government budget allocation to water supply is insufficient to achieve the universal access target by 2019; it is also dominated by central government financing and is limited in the extent to which it can incentivize subnational spending on poor-inclusive investments. • A large share of PDAMs perform poorly on core sector diagnostics and face a number of challenges to turning around performance, including political and financial interdependencies between local government and PDAMs, lack of incentives to improve performance, and limited mechanisms to sustain service delivery for the poor. • Low-income consumers are viewed as costly and risky to serve. They are perceived as unwilling to pay for piped water services, despite the fact that they pay a higher price for water in the informal market. The Hibah scheme is designed to counteract some of these perceptions, but analysis suggests it is not necessarily leading to more equitable coverage. • There is a lack of overall legal frameworks for equitable service delivery, complicated by the fact that poor residents in urban areas often live in informal settlements, lacking the formal registration and legal documents required for utility service. Improving Service Levels and Impact on the Poor 51 Why Urban Water? The focus on urban water for in-depth institutional and political economy analysis is timely and opportunistic. There is a clear historical trend of utilities facing challenges in trying to keep up with rising demand resulting from population growth and rapid urbanization. Half the population of Indonesia resided in urban areas in 2010, and this figure is projected to rise to 68 percent by 2025 (UN-DESA 2014). In the absence of adequate planning, the growth of cities can introduce a number of challenges, including disparities in income and access to services, which exacerbate inequalities. Despite these trends, focused efforts are being made by urban water government counterparts to tackle the issue of utility underperformance in order to meet universal access targets for improved water by 2019. Reaching the target of 60 percent coverage of piped and 40 percent coverage of non-piped improved water sources in urban areas will require 16.5 million new household piped water connections by 2019. But non-piped sources, such as groundwater, will remain a key means for extending improved access in order to reach the target. Currently, of the 29.6 percent of urban households with access to piped water supply, the B40 make up just 7.5 percent, while the T60 make up 22.1 percent. Given current patterns in equity of access to piped water (figure 5.1), B40 customers may be more likely to remain on a non-piped service for longer than T60 customers. This disparity indicates an urgent need to support the water sector and the GoI to develop viable strategies to help overcome challenges to increasing piped water access among the poor, as well as to ensure that the groundwater sources that make up the other 40 percent of improved coverage meet the 4K standards of quality, quantity, continuity, and affordability (kualitas, kuantitas, kontinuitas dan keterjangkauan). The current challenges facing the urban piped water sector have negative impacts on poor people, whether they have a connection or not. Piped water, especially that provided by PDAMs, represents the most economically efficient route to safe, affordable, and environmentally sustainable water services in urban areas. At present, alternative sources such as wells and boreholes are unlikely to be consistently safe in urban areas (Foster, Lawrence, and Morris 1998), especially given the large share of urban households (78 percent) using on-site pit latrines and soak pits that are a source of groundwater contamination. Moreover, the poor already pay several times more than the official tariff for vendor-supplied water; a piped water connection could thus result in substantial cost savings for poor households, if they are able to overcome the initial connection fee barrier through subsidies (World Bank 2006b). Figure 5.1: Access to Piped Water by Income Quintile (Q1–Q5) in Urban Areas Q1 3.4% Q5 Q2 10.1% 4.1% Q3 5.2% Q4 6.8% Source: Susenas 2015. 52 Improving Service Levels and Impact on the Poor The aim of this chapter is to outline the current insitutional and service delivery context for urban water supply and the constraints on extending piped water access to the poor in urban settings. The chapter will provide key recommended actions for improving the delivery of this service. Urban Water Institutional and Service Delivery Context Responsibility for basic service provision, which includes water supply and sanitation, has been decentralized to the district level through laws such as 23/2014 on Regional Government and 33/2004 on Fiscal Balance between Central Government and Regional Governments. In the case of water supply, it is now well established in legal terms that local (District) governments must provide a minimum standard of water service to citizens. Provincial governments are required to mediate on issues spanning district government boundaries, and central government retains a mandate for overall coordination, strategy, and policy formulation, as well as for safeguarding the availability of services to all. Within this broad framework, roles and responsibilities in the sub-sector are distributed across a wide variety of actors at different levels of government (figure 5.2). National level: The Ministry of Public Works and Housing (MoPWH) and the Ministry of Energy and Mineral Resources (MoEMR) have, respectively, responsibilities for policy and technical standards of surface and ground waters. MoPWH, and particularly its Directorate General of Human Settlements (DGHS) and Support Agency for the Development of Drinking Water Supply System (BPPSPAM), is the key ministry for technical and implementation issues on water supply and sanitation. DGHS has responsibility at the national level for facilitating the provision of water throughout the country—key units within it include the Directorate of Water Supply Systems (DITPAM). The BPPSPAM focuses more on monitoring the performance of the PDAMs. Standards for drinking water quality are set by Ministry of Health (MoH), whereas the Ministry of Environment and Forestry (MoEF) monitors the quality of water bodies, especially river water. A set of high level “4K” standards has also been issued by the National Development and Planning Agency (Bappenas) and enforced by Presidential Regulation No. 185/2014, and covers quality, quantity, continuity, and affordability (kualitas, kuantitas, kontinuitas dan keterjangkauan). The Ministry of Finance (MoF) has a key role in determining national budget envelopes, and is therefore central to water supply (and sanitation) planning. Preparation of National Plans (Rencana Pembangunan Jangka Menengah Nasional, RPJMN), including the current RPJMN 2015-2019, is led by Bappenas and reflects the vision of the elected president. Finally, the Ministry of Home Affairs (MoHA) is responsible for promoting institutional improvements in subnational governments, setting up the monitoring on minimum service standards, and supporting the financial management for local government (LG) enterprises, including PDAMs (ADB 2012; key informants). Provincial level: A Technical Implementation Unit (Satker), appointed by MoPWH, is required to receive and consolidate requests from district governments for water and sanitation infrastructure investment projects. These form the basis for most of MoPWH’s investment programs (strategic projects can also be determined by central government). The Satker coordinates with the national level to select projects to be supported (World Bank 2016b). District governments: Governments at district level are assigned powers over their own budget and planning processes in a manner that broadly reflects the national process, involving LG working units (Satuan Kerja Perangkat Daerah, SKPD), which include district-level planning and finance departments (equivalent to MoF and Bappenas at national level) and relevant dinas or local service offices (e.g., for public works, dinas pekerjaan umum, or for health, dinas kesehatan). The district head wields significant influence over investment priorities within the sector at the local level, whereas the local parliament (DPRD) works with the district head on budgeting, and also monitors implementation. Improving Service Levels and Impact on the Poor 53 Figure 5.2: Overview of Government of Indonesia Actors Involved in Urban Water Service Delivery Bappenas Ministry of Ministry of Ministry of (pokja lead) finance home affairs health National pokja AMPL Ministry of Ministry of Bureau of Other public works environment statistics ministries Target setting and DG of cipta policy karya Finance and DG for DG for development development regional public controller finances health Support Agency bureau (BPKP) for the development of Effluent standard and drinking water water resource quality supply services (BPPSPAM) Annual audit of Tariff Sector monitoring PDAM guidance DG of water performance Resources Drinking Water • Finacial assistance quality standard • Capacity building Monitor PDAM for LGs and PDAMs • Develop performance relationship based on BPKP’s within National water audit report Water resources management members utilities association • Support (PERPAMSI) performance Province Regional water improvement government supply system District government Annual budget process Tariff payment, Local legistlative District head/ PDAM Client/Citizen opting out, etc (DPRD) mayor Leadership appointment Market choices Political parties Alternative Note: providers Guidance/support Oversight Coordination Source: Harries et al. 2016. With decentralization, governance responsibility for basic services falls under the category of “concurrent government affairs” (Urusan pemerintahan konkuren), implying a division of responsibility between central and district governments (Law 23/2014). Ultimately, the district head/mayor has significant oversight authority as a result of his/her influence over strategic human and financial resource decisions, if not day-to-day operations and management of PDAMs. District heads also have regulatory responsibility for water tariffs, which involves signing off on proposals submitted by the PDAM directors and approved by the supervisory board. Some mayors also draw up performance contracts with PDAMs as a basis for more clearly defined oversight responsibilities. 54 Improving Service Levels and Impact on the Poor In urban areas, the provision of clean drinking water is undertaken predominately by PDAMs,1 each providing clean water connections to households within a single district or city. Out of a total of 512 districts across Indonesia, 423 have water utilities providing services in their area. Although the majority (386) are PDAMs, owned and managed by district/city governments, an additional 37 water companies are owned and managed by MoPWH or by the private sector. Taken together, these 423 water utilities have reported serving more than 9 million households as of 2015—equivalent to about 42 million individuals, or 16 percent of the total population of Indonesia (MoPWH, 2015). Constraints On and Opportunities for Connecting the Poor to Urban Water Services Why Urban B40 Households Have Lower Access to Piped Water than T60 Households There are a number of barriers preventing low-income households from accessing piped water connections. These include (a) government budget allocation and spending; (b) financial sustainability and performance of PDAMs; (c) perceptions and behavioral constraints; and (d) legal frameworks for equitable service delivery. Government Budget Allocation and Spending Despite decentralization of water and sanitation service delivery to the district level Government beginning in 2001, financing continues to be dominated by central government investments. budget allocation Finance from both national and local governments to urban water services has been for water supply isn’t enough to increasing in recent years, but it remains a small share of total and infrastructure spending. meet universal Coordinating between investments by different parts of government—with central access targets, government mainly investing “upstream” in urban water networks, and LGs funding and local “downstream” components—is a continuing challenge. In 2013 government expenditure government for water supply totaled US$476 million (IDR7.0 trillion), and was largely sourced from the spending is insufficient central government budget, with only 0.3 percent of sector expenditure coming from LGs. to bridge the Central government spending (adjusted for inflation/real and nominal) has increased both financing gap. in absolute terms (figure 5.3) and as a proportion of total allocations—though spending on water supply is still a modest share of total infrastructure spending, at less than 10 percent (table 5.1).2 The available data on sector financing suggests that current levels of government spending, although they have significantly increased, are inadequate to achieve universal access. The cost of universal access to water supply is estimated by the Directorate General of Human Settlements, MoPWH (Cipta Karya) to be approximately US$26 billion (IDR254 trillion), of which central government contributes US$1.4 billion (IDR13.5 trillion) annually up to 2019 through national budget allocation (APBN), for a total expenditure of Table 5.1: Central Government Water Supply Spending as Percentage of Total Spending/Infrastructure Spending 2005 2006 2007 2008 2009 2010 2011 2012 2013 Central government spending 0.13 0.2 0.28 0.31 0.5 0.43 0.34 0.37 0.62 Infrastructure spending 3.1 3.5 5.5 5.3 6.1 6.6 4.8 4.9 9.3 Source: World Bank 2016b. Improving Service Levels and Impact on the Poor 55 Figure 5.3: Central Government Patterns of Spending on Water Supply and Sanitation, 2001–13 8,000 7,000 6,000 5,000 IDR billion 4,000 3,000 2,000 1,000 0 2005 2006 2007 2008 2009 2010 2011 2012 2013 Water supply (nominal) Sanitation (nominal) Water supply (real) Sanitation (real) Source: World Bank, 2016b. US$7 billion (IDR67.5 trillion) (MoPWH 2015).3 Bridging the financing gap will require accessing provincial and LG budgets (APBD), PDAM internal resources, public multilateral and bilateral development financing, commercial loans, and private investments—as well as household out-of-pocket spending, which is currently estimated to be approximately one-third of total expenditure on water supply from all sources (World Bank 2016b). In particular, the government aims to attract US$1.5 billion in private sector financing through PPPs and business-to-business schemes, and US$860 million in commercial financing. The efforts to mobilize diverse funding sources in partnership toward achieving the shared goal of universal access has led the GoI to adopt common sectoral policies through national platforms of delivery that apply regardless of the source of funds. Current local government spending for water and sanitation is not sufficient to meet the financing gap for universal access. Provincial and district (subnational) government is expected to finance nearly half (47 percent) of expenditures required to meet universal access targets for water supply—a total of US$12.4 billion (IDR119 trillion), or US$2.5 billion (IDR24 trillion) annually (MoPWH 2015). Although subnational expenditures for the water and sanitation sector have doubled in real terms since 2001, by 2013 they were US$0.9 billion (just over IDR9 trillion) (figure 5.4). Taking these past expenditures for water supply and sanitation as a guide, a substantial financing gap remains. National Government has limited leverage to incentivize subnational spending on urban water supply. The vertical imbalance characteristic of the Reformasi-era4 fiscal-federal system has helped to promote state cohesion (Harris & Foresti 2010), but has also had important implications for investments in urban water supply—including the extent to which national government can push for prioritization of pro-poor investments by district governments. Most transfers to LGs (General Purpose Grants or Dana Alokasi Umum, DAU) go to support general expenditure, and are not earmarked (figure 5.5). The Special Allocation Fund (Specific Purpose Grants or Dana Alokasi Khusus, DAK), on the other hand, is allocated by region and earmarked for specific projects according to national priorities. It also includes some pro-poor criteria 56 Improving Service Levels and Impact on the Poor Figure 5.4: Subnational Government Patterns of Spending on Water Supply and Sanitation, 2001–13 10,000 9,000 8,000 7,000 6,000 IDR billion 5,000 4,000 3,000 2,000 1,000 0 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 Sub national (nominal) Sub national (real) Province (nominal) Province (real) District (nominal) District (real) Source: World Bank 2016b. Figure 5.5: Capital/Investment Mapping Ministry Ministry of Other of finance public works ministries Donor DAU DAK Satker at Asset transfer province Province Grant government Captial Injection through additional equity Service Local government PDAM Client Asset There’s no direct transfer to LG- transfer Service all on budget and/or on treasury Annual remittance treated as local revenue follows criterias based on current regulation Other providers (Non-PDAM) Note: Cash transfer Technical assistance Asset transfer Service provision Source: Harries et al. 2016. Improving Service Levels and Impact on the Poor 57 regarding how it is distributed among districts. The DAK, however, is relatively modest in relation to total LG financing, and the portion that is earmarked to water supply and sanitation is even smaller (most recently 0.15 percent) (World Bank 2016b). The national government’s ability to incentivize subnational governments to target poor households is, in particular, even more limited; it has increased with the national roll-out of the Water Hibah program from 2015, but eligibility criteria for LGs and PDAMs (such as sufficient raw water supplies and adequate treatment) has, along with other issues, limited the expansion of the program. Despite an increase in DAK transfers from central government, transfers are still not aligned to needs and there is limited accountability for fund allocation. DAK transfers have increased significantly over the past several years. In 2013, DAK for water supply was approximately US$62 million (IDR600 billion), and increased to US$104 million (IDR1 trillion) in 2015 and Figure 5.6: DAK Water Allocation per Capita in 2015 and 2017 versus Water Access a. 2015 200,000 180,000 Corr = –0.3769 DAK water/capita allocation 160,000 140,000 120,000 100,000 80,000 60,000 40,000 20,000 0 0 10 20 30 40 50 60 70 80 90 100 Water access (% 2015) b. 2017 200,000 180,000 Corr = –0.1769 DAK water/capita allocation 160,000 140,000 120,000 100,000 80,000 60,000 40,000 20,000 0 0 10 20 30 40 50 60 70 80 90 100 Water access (% 2015) Source: World Bank calculation, Ministry of Finance. 58 Improving Service Levels and Impact on the Poor US$118 million (IDR1.14 trillion) in 20175. The number of districts receiving DAK also declined, from 430 cities/districts in 2015 to 269 in 20176. Two challenges persist with DAK transfers. The first is that DAK water supply allocations are not aligned to needs. Figure 5.6 shows the correlation between the DAK allocation for water in 2015 and 2017, and the coverage of improved water in districts in 2013 and 2015. As DAK allocations have increased over time, there is a weaker association between allocations and improved water access figures. The second challenge is that DAK allocations are not based on performance— though this appears to be changing as of 2017. Funds are released upon submission of reports on absorption, rather than on the basis of how funds have been used and what has been achieved. Financial Sustainability and Performance of PDAMs Efforts to incentivize poor-inclusive service delivery need to factor in the overall financial health, PDAMs play efficiency, and performance of PDAMs. A major challenge to extending connections to low- a big part in income households is the perception that doing so would put the financial and technical health the provision of the PDAM in jeopardy, especially if not accompanied by targeted technical assistance and of drinking water, but they capacity building. Utilities require economies of scale for sustainable operation and cost lack incentives recovery from across the customer base. However, as illustrated in this section, a large share to improve of PDAMs perform poorly on core sector diagnostics and face numerous challenges to turning performance and around performance—including the political and financial interdependencies between LGs and are not obligated PDAMs, and the lack of broader incentives to improve performance. to serve the poor. Financial Sustainability Current PDAM water tariffs are not economically efficient in managing demand and resources, do not raise enough revenues for the PDAMs to become financially independent, tend to benefit high-income rather than low-income households, and are complex and difficult to administer effectively, creating opportunities for fraud and corruption. District heads have regulatory responsibility for water tariffs. For water utilities this is guided by the recent MoHA Regulation 71/2016, which requires tariffs to achieve full cost recovery and includes specific measures to protect poor customers to ensure that tariffs do not exceed 4 percent of household income. The new regulation lacks detail, however, on how compliance will be enforced across districts. Increasing Block Tariffs (IBTs) have been criticized for failing to achieve cross-subsidy, because high-volume consumers are not necessarily wealthier than lower volume consumers. The Regulation acknowledges this and allows for charges to vary according to a household’s classification. Households classified as MBR (Masyarakat Berpenghasilan Rendah or low-income household) would pay lower rates across the consumption blocks. Government Regulation 122/2015 includes more general provisions on tariff-setting, including affordability and the provision that national government should retain an overarching responsibility for ensuring access for low-income groups. The rationale behind the tariff calculation is generally not well understood by local governing bodies, creating an incentive to push tariffs down rather than allow for cost recovery. In some cities water tariffs for the poor (“social tariffs”) are set so artificially low (US$0.03/m³) that even with cross-subsidies from higher paying customers, PDAMs are forced to sell water to the poor at a loss. This ends up disadvantaging the poor because PDAMs are discouraged from serving predominantly poor areas. PDAMs can propose tariff levels and structures based on guidelines issued by MoHA, with the head of local government given the authority to approve them. Although no DPRD approval is needed, LGs, and even the PDAMs’ management, still prefer to seek DPRD agreement on tariff increase, so as not to be responsible for the decision. Guiding legislation and processes for tariff approval may be insufficient to mitigate the strong political incentive of elected officials to suppress tariffs. However, a new MoHA regulation, No 70/2016, will require LGs to subsidize the difference if tariffs are set below cost-recovery levels. Improving Service Levels and Impact on the Poor 59 Whether or not tariff increases are approved may still depend on the nature of the relationship between an individual mayor and the respective PDAM, and the level of understanding of both PDAMs and LGs on the need for cost-recovery tariffs. Connection fees required to fund capital investments are a barrier for low-income households, but have so far only been addressed through the Hibah. The government’s output-based grant (Hibah) scheme for household water connections is meant to circumvent this barrier, by reimbursing district government for connection fees for low-income households. The program was rolled out nationwide in 2015, but still only covers a fraction of the eligible population. In part, this is due to strict eligibility criteria that LGs and PDAMs are required to meet in order to participate. The criteria includes PDAMs achieving “healthy” performance status, and having adequate spare production capacity and the ability to pre-finance capital investments. Other barriers to participation include a lack of tools and information to identify eligible low-income households, concern that targeting such households under the water Hibah will damage the financial position of the PDAM, and a generally limited understanding of the purpose of the Hibah and how to incorporate the program into an overall business plan. Technical Aspects Available performance data7 suggest that a large share of PDAMs lag on key sector performance indicators, but that some progress has been made. For the three years shown in table 5.2, a little over 50 percent of PDAMs are classified as “healthy” based on the multi-indicator rating system of BPPSPAM, whereas around 20 percent are classified as “sick.” Data available for 2015 indicate water consumption of 18.9m3 per household (connection) per month, and an average number of operating hours per day of 19.2. Water losses hovered at around a third of the distributed total in 2015,8 and only around 26 percent of PDAMs evaluated in 2014 and 2015 were applying full cost-recovery tariffs. The financial position of most PDAMs remains precarious, with the majority loss-making.9 There is wide variation in performance measures between “top performers” and the rest of PDAMs. Using BPPSPAM data for the period 2011–14, utilities were categorized as top, bottom, or middle performers on each performance indicator, using the following composition: i. The Top 10 percent performers—utilities whose average performance on an indicator for the period 2011–14 (or 2010–15, depending on availability of data) was greater than the 90th percentile (or less than the 10th percentile in the case of non-revenue water); ii. The Bottom 10 percent performers—utilities whose average performance on an indicator was less than the 10th percentile (or greater than the 90th percentile in the case of non-revenue water) for the same period; and iii. The Middle 80 percent performers—utilities whose average performance on an indicator was between the 10th and the 90th percentile. As shown in figure 5.7, the top 10 percent performers on each indicator are doing much better than the middle 80 percent and bottom 10 percent—achieving full performance on quality and duration of supply hours. Service coverage, average revenue, and number of connections for the top 10 percent performers increased between 2011 and 2014, whereas other indicators remained stable. On the other hand, PDAMs in the bottom 10 percent of performers for duration of supply did worse over time, and none passed quality standards under MoH regulations. The middle 80 percent of performers on water quality also showed a sharp decline in the percentage of samples meeting the required standards over the period. Despite the poor performance of many PDAMs, most PDAMs and their owners lack incentives to improve performance. Although performance targets are included in some GoI programs, these are only for the duration of the program and there is no incentive to maintain performance 60 Improving Service Levels and Impact on the Poor Table 5.2: Summary of PDAM Performance Indicators, 2013–15 2013 2014 2015 Total number of PDAMs 383 383 386 Number of PDAMs evaluated by BPPSPAM 350 359 368 Category       Healthy 176 182 196 Unhealthy 104 103 100 Sick 70 74 72 Total residents in administrative areas served by PDAMs 229,564,729 232,402,248 232,670,949 Total residents in service areas served by PDAMs 141,378,957 146,958,850 149,380,197 Number of subscribing customers 8,816,286 9,260,268 9,828,054 Population served 57,588,990 61,489,535 64,155,423 Domestic water consumption (m3/customer 17.4 19.23 18.87 household/month) Non-revenue water 33.00% 32.79% 32.47% Billing collection efficiency 86.8% 94.47% 94.05% Total PDAMs applying full cost recovery tariffs a 105 93 96 Service operating hours per day 18.4 18.6 19.22 Number of PDAMs by category: number of subscribing       customers <10,000 155 157 154 10,000–50,000 160 165 174 50,000–100,000 19 22 25 >100,0000 16 15 15 Source: BPPSPAM 2013–15. a. Tariffs are assessed to provide full cost recovery on the basis of an assumed 100% billing collection efficiency. As can be seen from the table, collection efficiency has fallen below 100% in recent years. The calculations also assume a generic level of non revenue water (NRW) of 20%, which is lower than recorded levels. As such, the proportion of PDAMs achieving full cost recovery is likely to be lower than the figures reported here suggest. levels. In addition, although BPPSPAM conducts performance audits of PDAMs, there is no systematic monitoring or evaluation of performance in relation to program participation, resulting in broken feedback loops. With regard to LGs, the power of the district head to approve budget allocation and tariffs, and to appoint PDAM leadership, appears to be a key source of dysfunction in the former’s oversight role—resulting in a reluctance to raise tariffs to full cost-recovery levels or to disconnect illegal connections, and leading to losses in revenue. Consequently, utilities are often caught in a negative downward spiral of poor cost recovery, low investment in existing infrastructure (let alone network expansion), deteriorating service levels, and falling customer satisfaction and willingness to pay. These factors constrain the ability of PDAMs to sustainably serve low-income customers. Overcoming Perceptions and Behavioral Constraints the reluctance of piped water Low-income consumers are less likely to be served by piped providers, and have less of a providers to serve voice, politically, compared to high-income groups. There are few, if any, intermediary institutions the poor means addressing bringing together the poor, PDAMs, and LG officers. This situation is complicated by the fact perceived that the poor are generally ill informed about the advantages of piped water connections or and actual about the disadvantages, in terms of cost and quality, of alternatives (see box 5.1). They are disincentives. Improving Service Levels and Impact on the Poor 61 also distrustful of the service provided by PDAMs, and are reluctant to get locked into a contract for water services that are unreliable. Piped providers reportedly perceive low-income customers as costly and risky to serve. There are a number of issues that create a disincentive for piped providers to actively extend access to B40 households, in particular: •• limited expectations that low-income households will behave as “good customers”; •• perception that poor households are reluctant or are unable to commit funds for a water connection since they are vulnerable to shocks, such as inflation, job layoff, or sudden illness; Figure 5.7: Top 10 Percent versus Bottom 10 Percent Performance Average (Unweighted) Service coverage Non-revenue water 100 70 90 60 80 70 50 60 Percent Percent 40 50 40 30 30 20 20 10 10 0 0 2011 2012 2013 2014 2011 2012 2013 2014 Staff productivity Consumption of water 300 30 Number of customers per employee 250 25 m3/month/connection 200 20 150 15 100 10 50 5 0 0 2013 2014 2015 2011 2012 2013 Duration of supply Quality 24 100 90 20 80 16 70 Hours/day 60 Percent 12 50 40 8 30 4 20 10 0 0 2011 2012 2013 2014 2012 2013 2014 Bottom 10% Middle 80% Top 10% figure continues next page 62 Improving Service Levels and Impact on the Poor Figure 5.7: Continued Cost coverage Average revenue 1.4 7,000 IDR per m3 of water sold 1.2 6,000 1.0 5,000 Percent 0.8 4,000 0.6 3,000 0.4 2,000 0.2 1,000 0 0 2012 2013 2014 2011 2012 2013 2014 Unit costs Number of connections 1,600 70,000 IDR per m3 of water produced 1,400 60,000 1,200 50,000 1,000 Number 40,000 800 30,000 600 20,000 400 200 10,000 0 0 2013 2014 2015 2011 2012 2013 2014 Bottom 10% Middle 80% Top 10% Source: Calculations based on BPPSPAM Performance Audits (2011–14). Note: a) based on unweighted observations; b) changes are computed based on the same composition of utilities in two consecutive years; c) the trends are based on data for at least 20 utilities. Box 5.1: The Benefits of a Household Piped Water Connection • Health effect: convenient water supply linked to better hygiene, improved health and development outcomes, and a cleaner environment; • Small enterprise effect: opportunities to open small businesses, such as catering services, drink packaging, seed farms, soybean cake production, etc.; • Wealth effect: increase in property values due to availability of piped water (for owner-beneficiaries); • Prestige effect: increase in sense of self-worth through better and more convenient facilities; and • Income effect: savings in unit price of water, savings in storage, and time savings in purchasing vended water Improving Service Levels and Impact on the Poor 63 •• perception that low-income households are unwilling to pay cost-recovery tariffs, and that “social” tariffs will damage the financial position of the PDAM; •• uncertainty around tenure and ownership rights (largely left to LGs to resolve); •• perception that low-income communities are more territorially dispersed within the administrative area; •• perceived lack of sophistication in filling out forms; •• perception that poor households do not understand PDAM constraints and are quick to protest when service delivery fails to meet expectations. It is not known to what extent these perceptions reflect unconscious bias or the actual situation.10 The Hibah scheme is designed to counteract some of the tendencies on the part of district government to view poor households as costly and risky to serve. The Hibah incentivizes PDAMs and district government to overcome governance and financial barriers (real and perceived) to serving low-income households, since connection fees are reimbursed to PDAMs once the connection has been independently verified. Analysis of Hibah- versus non-Hibah-participating kotas does not indicate more equitable coverage between T60 and B40 in Hibah-participating kotas, although this could be due to different baseline coverage among the two groups, as previously discussed. Moreover, because the Hibah only reimburses for connection fees, LGs and PDAMs participating in the program tend to target expansion to areas where additional capital investments in piped networks would not be needed11—a tendency that could result in some remote low-income settlements being excluded. It appears that the purpose of the Hibah program is still not well understood by some LGs and PDAMs, which also limits expansion. Poor households are likewise at a disadvantage as a result of their own perceptions and behaviors. They may lack awareness of the risks of drinking contaminated water (whether groundwater or  vendor-supplied) and so do not demand improvements in public services. Although poor households perceive water tariffs as costly, they fail to calculate the full cost of buying water from informal vendors, both in terms of direct expenditure and time spent. Analysis of cost-recovery tariffs suggests that low-income households paying tariffs would save between 5 and 12 percent of their income compared to expenditure on vendor-supplied water (World Bank 2006b). Legal and Regulatory Frameworks for Equitable Service Delivery Currently, there is no clearly stated poor-inclusive mission underlying the 2019 universal access target for water supply. Although the 100-0-100 program target is universal access, given current trends, and in the absence of an explicit approach for reducing inequality in access, it is likely that B40 households will remain on non-piped sources for longer than T60 households. Moreover, although affordability targets for water supply aim for water tariffs that do not exceed 4 percent of household income, and regulations are in place for achieving this (Regulation 71/2016 and Regulation 122/2015), an explicit strategy to enforce this at LG level has not yet been outlined. There are no laws, regulations, or contractual requirements for PDAMs to consistently serve the poor. The lack of legal guidance on equitable service delivery is further complicated by the fact that poor residents in urban areas are more likely to live in informal settlements, lacking formal registration and legal documents required for utility service. If these settlements occupy government-owned land, utilities do not have legal authority to build infrastructure or extend services. In such cases, PDAMs rely on LGs to pave the way to service delivery. However, in most cases LGs do not articulate any poor-inclusive or pro-poor strategy as part of the planning process required by MoPWH in its Master Plan for Water Supply System (RISPAMS). 64 Improving Service Levels and Impact on the Poor Regulatory control of alternative water service providers is fragmented, with local government responsible for enforcing the regulations. Currently B40 households are more likely to use alternative sources for drinking (with the exception of bottled water, now the preferred drinking water source for T60 households in urban areas), which are harder to regulate—most notably groundwater sources such as private wells and boreholes, and branded or non-branded (refill) bottled water. There is little clarity on who holds ultimate responsibility for regulation of abstraction and quality of water resources, on which all forms of urban access depend. Law 7/2004 on Water Resources and a new government regulation (PP No.122/2015 on Cultivation for Water Resources) left ground/surface water management coordination unresolved, with responsibility for surface water assigned to MoPWH, and for groundwater to LGs, in coordination with MoEMR. Though the annulment of Law 7/2004 in 2015 and resinstatement of the previous Law on Water Resources (11/1974) provides an opportunity for a more integrated approach, signs are not promising that the annulment is leading to greater integration, particularly with regard to the relationship of water resources to urban water supply. Regulations do exist for bottled water, including under MoI Decree 705/2003 for branded bottled water, and MoH Decree No. 492/2010 for unbranded/offered by refill kiosks. There is, however, no regulatory control of private tanker operators, smaller mobile vendors, or piped customers who resell water informally to neighbors. The full extent of this informal water market is not clear—Susenas data imply it is a minor component of service provision, but it may be significant in some cities, such as Jakarta (Kooy 2014). In the following chapter the report will conclude with a presentation of key recommended actions to orient future water supply and sanitation policy and investment toward a more inclusive approach. Recommendations are prioritized based on their expected impact on the development goals of (a) reducing inequality; (b) enhancing health and well-being; and (c) promoting economic growth and prosperity, as well as on the strength of the evidence base for the solution proposed. Notes 1. There are also small urban water utilities that are not constituted as PDAMs (that is, as LG-owned but officially independent enterprises), including community-based organisations, and technical units housed within District Government with varying degrees of autonomy: “A UPTD (Unit Pelaksana Teknis Daerah) is a Regional Technical Implementation Unit attached to an LG agency that is responsible for service provision but has no autonomy to retain income for expenses. A BLUD (Badan Layanan Umum Daerah) is also a technical operation unit of an LG agency that is allowed to operate with flexibilities or exemptions from rules applicable to the Dinas. It has some additional degree of autonomy to a UPTD, but is more complicated to establish.” (World Bank and AusAid 2013, 21). We nonetheless focus on PDAMs in this study, as the most prominent form of piped water provider. 2. All estimates exclude transfers to sub-national government. 3. Historical (2013) exchange rate was used; if current (2016) exchange rate is used, the US$ figures will be lower. 4. The reform era following the fall of Suharto in 1998 and encompassing a process of democratic decentralization as the centerpiece of a broad but fundamental set of reforms to the formal institutions of the Indonesian state, the most significant of these being the model of decentralization for service delivery. 5. Historical (2013) exchange rate was used; if current (2016) exchange rate is used, the US$ figures will be lower. 6. The smaller number of districts receiving DAK funding in 2017 may be due to changes in the procedures on submission of requests and evaluation of proposals by Bappenas, MoPWH, and MoF. Proposals are now required to go through a stringent review process, covering the number of beneficiaries, alignment with national priorities, readiness criteria, local fiscal capacity, and prior performance. 7. Based on audited reports collated the by the Support Agency for PDAMs (Badan Pendukung Pengembangan Sistem Penyediaan Air Minum, BPPSPAM) and analyzed by the Financial Improving Service Levels and Impact on the Poor 65 and Development Supervisory Agency (Badan Pengawasan Keuangan dan Pembangunan, BPKP), with additional analysis by the World Bank. The data are not fully reliable due to measurement and sampling issues and missing values in some years but remain the only available source of PDAM performance indicators. 8. The rating system evaluates performance on the basis of audited reports from BPKP . Indicators, which are incorporated into a total score, include: financial measurements such as return on equity; service aspects such as coverage and customer growth; metrics of operational performance such as non-revenue water; and human resource aspects such as staff to customer ratio. 9. Analysis of data from 386 PDAMs (BPPSPAM). 10. The evidence of perceptions and behavioral constraints in this section is sourced from (World Bank 2006b) and World Bank, Project Appraisal Document of National Urban Water Supply Project, April 2017. Further qualitative research is recommended to fully substantiate the claims. 11. Central Project Management Unit (CPMU) Hibah. (2016). Program Hibah Air Minum dan Sanitasi—Progress Report. Jakarta: Ministry of Public Works and Housing. References ADB. 2012. Water Supply and Sanitation Assessment, Strategy, and Road Map. Manila: ADB. /33808​ Available online at: https://www.adb.org/sites/default/files/institutional-document​ /files/indonesia-water-supply-sector-assessment.pdf. Foster, S.S., A. Lawrence, and B. Morris, B. 1998. Groundwater in Urban Development: Assessing Management Needs and Formulating Policy Strategies. Vol. 390. Washington, DC: World Bank. Harris, D., and M. Foresti. 2010. Indonesia’s Progress on Governance: State Cohesion and Strategic Institutional Reform. London: ODI. Harris, D., Mason, N., Rimbatmaja, R. 2016. Identifying and Overcoming Binding Constraints to Piped Urban Water Services for the B40 in Indonesia. London: ODI. Kooy, M. 2014. “Developing Informality: The Production of Jakarta’s Urban Waterscape.” Water Alternatives 7 (1): 35–53. MoPWH (Ministry of Public Works and Public Housing). 2015. Rencana Pencapaian Air Minum dan Sanitasi 2015-2019. Jakarta: MoPWH. Susenas. 2015. Survei Ekonomi Nasional. Badan Pusat Statistik. UN-DESA 2014 (2014). Revision of World Urbanization Prospects. New York: United Nations Department of Economic and Social Affairs. World Bank. 2006b. Enabling Water Utilities to Serve the Urban Poor. Washington, DC: World Bank Group. ———. 2016b. More and Better Spending: Connecting People to Improved Water Supply and Sanitation in Indonesia, Water Supply and Sanitation Public Expenditure Review (WSS-PER). Jakarta: World Bank. 66 Improving Service Levels and Impact on the Poor Chapter 6 Priorities for Future Policy and Investments in Water Supply and Sanitation Key Recommended Actions •• Expand piped water services to a larger share of the B40 in urban areas. •• Improve the quality of alternative water sources for those who will remain on non- piped water supply. •• Support the B40 in gaining access to improved sanitation. •• Bring more households into the full sanitation and fecal waste service chain in urban areas. •• Champion multisectoral approaches to reduce child stunting. •• Enhance water supply and sanitation interventions to have greater impact on nutrition outcomes. The water and sanitation sector in Indonesia is at a pivotal juncture in the post-2015 SDG era, where success will be defined by service quality, sustainability, and equitable distribution of services. The GoI has established its own ambitious target to achieve universal access to water supply and sanitation by 2019–11 years ahead of the SDG target. The existing challenges to achieving these targets, let alone achieving them on schedule, are compounded by rising income inequality and rapid urbanization in Indonesia. The recommendations outlined in this chapter aim to support the GoI in achieving its universal access target, with a particular focus on the needs and constraints of the B40. The goal is to orient future water and sanitation investments around three sector priorities: (a) reduce inequalities in access and quality of water and sanitation services; (b) enhance the health and nutritional impact of water and sanitation investments; and (c) promote economic growth and vibrant cities through more sustainable service delivery in urban areas. Going forward, the Government of Indonesia will establish several national platforms of service delivery to bridge the gap between policy and implementation. These platforms include (a) rural water supply—community-based and institutionally driven; (b) urban water supply through LGs and PDAMs; (c) regional water supply systems1; and (d) urban sanitation2 (figure 6.1). The platforms allow for (a) one national policy applicable to all sources of financing; (b) common planning documents at LG level; and (c) one monitoring system. In addition, the platforms include flexible funding arrangements and management that will no longer rely solely on national government funding (APBN); support for the development of regulations and guidelines; Improving Service Levels and Impact on the Poor 67 Figure 6.1: Government of Indonesia 2019 Universal Access Targets and Service Delivery Platforms ‘Indonesia’s 100-0-100 Program’ Policy target to achieve 100% safe water access, 0% urban slums, 100% sanitation access Delivery platform Stategies and delivery mechanism to achieve targets Rural and Urban water Regional Urban waste community supply water supply water based PAMSIMAS NUWSP Piped Non-Piped National investment roadmap Investment needs and potential sources of finance to achieve targets Source: Development of Urban Water Supply Investment and Service Improvement Framework, Final Report, 2016. and capacity building at all levels of government. This common policy framework will provide a consistent basis to inform all infrastructure investments, regardless of the source of financing (APBN, APBD, bilateral and multi-lateral development banks, and local banks). Additionally, local government will be in the driver’s seat with the flexibility to involve communities and other donors, NGOs, universities, and private sector actors. The 100-0-100 program and strategy to achieve universal access should adopt a poor-inclusive approach to ensure that Indonesia not only achieves its service delivery targets, but that water supply and sanitation become key drivers of reduced inequality, enhanced health and well- being, and economic growth and prosperity. A poor-inclusive approach is one that improves the ability and opportunity of the poor and vulnerable to benefit from water supply and sanitation services, using the tools of financing, targeting, legal frameworks, and institutions. For water supply, this implies the need for a fifth standard of keadilan (equity) alongside the 4K standards of quality, quantity, continuity, and affordability (kualitas, kuantitas, kontinuitas dan keterjangkauan), as well as an explicit strategy for achieving this standard. The evidence in this report aims to provide an empirical basis for shifting the approach toward more inclusive and equitable service delivery in the water and sanitation sector. A key input for policy and recommendations is a dynamic dashboard that provides visual perspective of the overlaps between poverty, lack of access, and child health and nutrition (see appendix A).3 Disaggregation to sub-district level, and mapping capabilities for informal settlements and urban slums, could further advance poor-inclusive service delivery. Recommendations are grounded in the findings of the WASH Poverty Diagnostic and, for urban water, in the institutional and political economy analysis described in chapter 5. Even though a similar in-depth analysis was not conducted for urban sanitation, rural sanitation, or rural water as part of the WASH Poverty Diagnostic, the recommendations that are specific to these 68 Improving Service Levels and Impact on the Poor subsectors benefit from recent sector studies and reports, as well as global experience and evidence. Future analytical work could address these subsectors in depth, especially regarding targeted financial subsidies and urban sanitation service delivery challenges. Key Recommended Actions Expand Piped Water Services to A Larger Share of the B40 in Urban Areas Improve the efficiency and performance of PDAMs to generate a virtuous cycle of performance, tariff increases, cost recovery, and expansion of connections, especially to poor households. NUWSP is the delivery mechanism for the urban water supply platform and already includes a robust emphasis on performance improvement of PDAMs through a number of program components, including: •• Incentives and performance-based approaches, including a comprehensive framework of performance incentives that tie sources of additional financing to achievement of performance targets; •• Joint performance monitoring assessment of LGs and PDAMs using a universal and integrated monitoring and information system accessible to financing partners; •• Capacity building and technical assistance to LGs and PDAMs, covering technical, human resources management, utility reform, financing, good water governance, citizen engagement, and the identification, development, and preparation of projects and project proposals; •• Performance-based contracts/agreements between LGs and PDAMs to formalize the relationship and provide reassurance to central government that performance of PDAMs is being monitored by asset owners. These components of the NUWSP framework will be tested by the GoI, especially MoPWH, under the Directorate General of Human Settlements, and Bappenas, in select provinces that are home to a wide range of PDAMs with various capabilities. A critical input into the performance-improvement cycle of PDAMs will be the establishment of cost-recovery tariffs, but tariff schedules need to adopt realistic affordability benchmarks for low-income households. The affordability target under the 100-0-100 program aims for water tariffs that do not exceed 4 percent of household income. More evidence is needed to evaluate whether 4 percent is a reasonable expectation for households living below the poverty line. Capacity building for LGs and PDAMs on tariff-setting must include specific provisions to incorporate equity and social concerns into tariff structures, as well as guidance on how tariffs can be structured to allow for cross-subsidization between customers in order to protect the poor and vulnerable. Additional capacity building on project preparation and project proposal development, undertaken through the CoE program managed by the MoPWH, should cover (a) how to assess affordability of water tariffs; (b) willingness among poor households to pay for piped water connections; and (c) incorporation of low-income households, including Water Hibah customers, in the overall performance improvement and investment plan. Additional financing mechanisms are needed to ease the financial and liquidity constraints faced by the poor. Piped water connection fees are unaffordable for households living near or below the poverty line. These costs, and the ability of poor households to pay them, need to be factored into investment decisions. The output-based grant Hibah scheme for piped water has incentivized more poor-inclusive service delivery, but not all PDAMs are eligible to Improving Service Levels and Impact on the Poor 69 participate, leaving a large share of poor households unable to connect to piped water services. Subsidized credit and savings schemes are alternatives that allow households to spread the cost of the connection over time. Over the past several years, microfinance for household water facilities has become more common. An initiative of USAID’s IUWASH program has financed connection fees for approximately 15,000 households in low-income communities of North Sumatra, DKI/West Java/Banten, Central Java, East Java, and South Sulawesi.4 Despite government support for microfinance, the scale-up of the approach has been constrained by chronic raw water shortages (particularly in Java and Sumatra), the protracted development of distribution networks, and the availability of alternative mechanisms (such as Hibah) covering the connection fee. Firstly, better coordination between Hibah and other microfinance schemes can be achieved through the platform approach, taking advantage of a common policy framework regardless of the source of financing and more flexibility at the local government level to partner with private sector actors. Secondly, the existing targeting mechanism for Hibah beneficiaries that is based on electricity usage could be combined with income targeting to better identify eligible low-income households for financial subsidies. Improving the Increasing demand and raising awareness of the benefits of piped water—among both performance consumers and local government actors—is needed to shift consumer behavior and dependence of PDAMs and on alternatives, and to build the political will for improvements in water supply to poor households. setting cost- Awareness-raising could increase demand for efficient, sustainable water supply services, but recovery tariffs the poor can has been missing from most water supply programs. Most households treat their water before afford will help drinking through boiling or filtration, but are unaware of the potential for recontamination during increase piped storage. Awareness and behavior change campaigns, coordinated by MoPWH and MoH, and water access for implemented in part by PDAMs, can help increase demand for clean water and put pressure on the poor. PDAMs and LGs to expand provision of piped water services to unserved communities and/or improve the quality of existing services. Importantly, technical assistance provided through NUWSP should emphasize the health and economic benefits of supplying clean water to poor households to help raise the profile of WASH investments in district decision-making. The current Adjustments to the current intergovernmental fiscal transfer system are needed to better align intergovernmental transfers to needs. While current levels of government budget allocation to water supply are transfer system insufficient to achieve the universal access targets for water supply, existing fiscal transfers must better could be allocated more efficiently to address needs. Basic information on water access is align transfers readily available; however, data on DAK transfers show a declining association between DAK to needs. allocations and water coverage at the district level. Additional considerations for aligning fiscal transfers to needs through DAU point to population growth in urban centers, and suburban districts in particular. To better align fiscal transfers with population growth trends will require adoption of a per capita calculation—rather than the current per region calculation—to ensure equitable distribution of public resources according to population density of cities and districts. This alignment does not address the need for more financing for the sector overall. Commercial loans, private investment, and business-to-business collaboration should be explored to better understand how these additional sources of financing can help bridge the gap. Improve the Quality of Alternative Water Sources for those who Will Remain on Non-Piped Water Supply Drinking water supplies, whether piped or non-piped, need to be consistently monitored for water quality risks, and this information made publicly available. Consumers are largely unaware of the quality of drinking water from different sources and the particular risks posed by poor household water storage practices and poor fecal waste management. Water sector strategy should account for the potential water quality risks of poor sanitation, and the respective investments of the water and sanitation sub-sectors should be aligned. This alignment is especially important in areas facing technical barriers to piped water. At city level, Bappeda can ensure that the needed alignment on water and sanitation is reflected in the respective strategy documents (Master Plan for Drinking Water and City Sanitation Strategy). Bappeda could also oversee integration of data from the two sub-sectors into planning, implementation, and monitoring. 70 Improving Service Levels and Impact on the Poor Regulatory control for small scale water providers should be strengthened to ensure that Regulatory regulations on drinking water quality are met. For refilled bottled water, this could be control must be accomplished by linking the water quality monitoring (under the responsibility of MoH) with the strengthened and water quality licensing process (under MoI). monitoring improved. Support the B40 to Gain Access to Improved Sanitation Indonesia achieved substantial progress in reducing open defecation and now faces the second generation challenge of moving households up the sanitation ladder. Achieving universal access will require Indonesia to strengthen the STBM program by revisiting the zero-subsidy approach to sanitation and linking it with subsidized credit and savings mechanisms to reach the poorest households. Global practice suggests that subsidies can harm sanitation behavior- change efforts; however, experience shows that when subsidies are well targeted, delivered through an efficient channel, and affordable, they can be an effective mechanism for reaching poor households who otherwise cannot afford the high lump sum cost of a toilet. Septic tanks are a desirable level of service, but the cost is often prohibitive for poor households. Targeting subsidized credit and savings schemes through existing targeting systems that are already working well to identify low-income households for social assistance (e.g., the UDB operated by TNP2K and MoSA) can be an efficient and transparent way to reach households most in need of subsidies in order to achieve higher levels of service. The UDB contains socioeconomic and demographic information for approximately 40 percent of the population with lowest welfare status, the equivalent of 24 million households, or 96 million individuals. Moreover, this approach would accommodate a multisectoral strategy for targeting poor households, since the UDB is used for targeting other social assistance, including Subsidized Rice for the Poor (Raskin), Public Health Insurance (Jamkesmas), Cash Assistance to Poor Students (Bantuan Siswa Miskin), and the Family Hope Programme (Program Keluarga Harapan, or PKH), which has a nutrition component. The MoH should take a leadership role in adapting the existing policy on sanitation subsidies to address the financial constraints of poor households, whereas MoPWH should work with TNP2K and MoSA to adopt the UDB for targeting assistance under PAMSIMAS. Bring More Households into the Full Sanitation and Fecal Waste Service Chain in Urban Areas The range of solutions required to meet universal access targets in urban areas in a cost- A holistic effective manner demands that cities and towns take a more holistic and inclusive approach to and inclusive planning for citywide sanitation. Planning needs to cover the full fecal waste service chain and approach outline a progressive roadmap for bringing the entire population into this service chain. Local to citywide sanitation will solutions are complex, requiring a combination of piped and non-piped technologies, such as bring more septic tanks, sewerage, decentralized small-scale wastewater treatment plants, and fecal households into sludge management. District heads and mayors need to be given responsibility for ensuring the service chain. consistency in planning, budgeting, and execution; flexible funding arrangements; and technical assistance and capacity building where needed. This approach requires a delicate balance between the national government’s fiscal leverage to incentivize investment in sanitation, and granting greater autonomy to LGs to decide where and how to invest those resources. Behavior change has been the cornerstone of a successful effort to stop open defecation in Indonesia, but needs to be adapted to behavioral issues common in the urban sanitation space. The universal access targets will be met primarily through on-site sanitation systems with fecal sludge management (72.5 percent), whereas smaller shares will be met by centralized and decentralized sewerage (12.5 percent), and basic sanitation (15 percent). Currently, although 78 percent of the population in urban areas uses a septic tank, 95 percent of fecal waste ends up in the nearby environment. Low consumer demand for fecal waste management services is a reflection of both the lack of integrated services and the lack of knowledge about safe management and disposal practices. Part of the solution will require generating the Improving Service Levels and Impact on the Poor 71 necessary demand and changing the behavior of individuals, communities, and providers. But behavior change cannot happen in a vacuum—it requires a coherent policy framework, clarity regarding institutional arrangements, and adequate enforcement of local government ordinances for design, construction, and desludging (World Bank and Australian Aid 2013). Coordination between MoH and MoPWH will be needed for effective implementation of STBM in urban areas, along with enforcement by MoEF of new regulations on effluent standards. Elevating the profile of sanitation in political and fiscal discussions, as well as in intra-household decision-making, will be crucial to achieving universal access targets. This change may require a shift in the narrative around urban sanitation to emphasize not only elements of modernity and competitiveness, but also the lifelong effects on intellectual and economic potential of early-life stunting, caused in part by poor sanitation. AKKOPSI could lead advocacy efforts with mayors and district heads. Champion Multisectoral Approaches to Reduce Child Stunting A multisectoral Efforts to improve early-life outcomes for children, especially reducing stunting in Indonesia, approach is should capitalize on the synergies of multisectoral approaches. Progress toward reducing critical if early life stunting in Indonesia can be enhanced by coordinated multisectoral interventions that address outcomes are to effectively the four key underlying determinants of nutrition.5 Evidence shows the effects will be improved and stunting reduced. vary by the wealth status of the household, and across rural and urban areas. Thus, a one-size- fits-all approach to multisectoral programming is not likely to be as effective as multisectoral programs that are tailored and targeted to specific geographic locations and poverty levels. Geographic targeting can be used to reach areas where undernutrition and underlying deprivations are prevalent. In these areas, interventions should be co-located to achieve service improvements across multiple sectors that impact stunting (see box 6.2). PAMSIMAS could serve as the main platform for multisectoral convergence between WASH and other programs addressing nutrition outcomes in young children, while implementation across relevant sectors could be coordinated through Bappenas. In addition, the strengthening of social and behavior-change communication, including development and execution of country plans and communication strategies for improving nutrition as part of the Scaling Up Nutrition (SUN) Movement, is needed at national level (see box 6.1). Enhance Water Supply and Sanitation Interventions to Have Greater Impact on Nutrition Outcomes An overarching message of the WASH Poverty Diagnostic in Indonesia is that existing WASH interventions are failing to produce outcomes of sufficient quality to impact child nutrition. Box 6.1: Access to Water Supply and Sanitation under the 2019 Universal Health Coverage Target for Indonesia Indonesia plans to attain Universal Health Coverage (UHC) by 2019, whereby all people receive the quality, essential health services they need, without being exposed to financial hardship. The World Bank and WHO have proposed a dashboard (known as the UHC Dashboard) of common and comparable indicators across countries to track coverage of prevention and treatment interventions related to the health SDGs. Improved water and improved sanitation are included in the dashboard of eight core tracer indicators covering health promotion, illness prevention, treatment, rehabilitation, and palliative care. 72 Improving Service Levels and Impact on the Poor Box 6.2: Strengthening Nutrition-Sensitive Actions in PAMSIMAS to Reduce Stunting in Children under 5 PAMSIMAS, which targets 27,000 villages, is the GoI’s platform approach for bringing sustainable clean drinking water and sanitation services to rural communities. The program adopts the STBM approach, focused on behavior change at the household and community level, including 5 pillars of (a) stopping open defecation; (b) handwashing with soap; (c) household safe water treatment, storage, and food handling; (d) safe disposal and management of solid waste; and (e) safe disposal and management of wastewater. Nutrition-sensitive actions will be piloted through PAMSIMAS across four districts in two provinces. The objective of the pilot is to gain implementation knowledge, strengthen the evidence base, and derive lessons for scaling up multisectoral interventions impacting child nutrition outcomes. The key performance indicators are: a. Increased coverage of WASH interventions for households with pregnant women, lactating mothers, and children under five b. Increased number of ODF villages c. Increased number of villages practicing all five pillars of STBM d. Increased demand and uptake of health and nutrition services in the pilot villages e. Increased number of villages leveraging village funds (Dana Desa) for WASH and nutrition activities The proposed pilot will leverage the operational mechanisms instituted under PAMSIMAS, combining the strengths of STBM with Infant and Young Child Nutrition (IYCN) practices implemented at the village level by: a. Prioritizing villages with high malnutrition rates b. Enhancing the community-action planning process of PAMSIMAS to prioritize households with pregnant women, lactating mothers and children under five c. Implementing Community-wide Behavior-Change Communication campaigns for ODF, embedded with nutrition messages; reinforcing WASH messages through local health/nutrition centers (Puskesmas, Posyandu, etc.) d. Monitoring progress of WASH access and usage by target groups and nutrition programs using citizen engagement tools such as social audits, community score cards, feedback loops, etc. Evidence from the Diagnostic shows that one way to enhance sanitation interventions for greater nutritional impact is to crowd in resources until communities exceed a high threshold of coverage of improved sanitation. There is now compelling evidence, both within Indonesia and globally, that sanitation levels of a community are more important than those of any one household. The evidence shows that health and nutritional benefits mainly accrue after a threshold level of coverage is surpassed, and that full benefits may only be achieved as sanitation becomes universal. This evidence supports existing practice, which aims for ODF areas, and it suggests that resources should be spent on bringing as many communities as possible to universal or near universal levels of coverage in order to realize the health benefits of sanitation. Improving Service Levels and Impact on the Poor 73 Water and sanitation interventions could have a greater impact on nutrition by adopting a “child-centric” approach. The five pillars of STBM are comprehensive across WASH services, but may still bypass some of the dominant fecal contamination pathways that affect small children. An emerging approach known as “baby WASH” or “child-centered WASH” focuses on interrupting exposure pathways that are most strongly associated with subsequent diarrheal disease. Sanitation interventions need to include measures that ensure cleanliness of a child’s play environment (such as safe disposal of child and animal feces), and separation of livestock and domestic animals from the main housing compound. Equally important are washing hands with soap before preparing food and feeding/breast feeding, and after handling child feces, and using only treated drinking water for preparation of liquid and solid food for infants and young children. Importantly, these recommendations go deeper and are more targeted to child- related exposures than the existing pillars of the STBM program. MoH should adapt existing STBM behavior-change communication materials and local government capacity building to incorporate baby WASH, while implementation of the approach should be aligned with the current nutrition-sensitive pilot of PAMSIMAS. Multisectoral approaches need to be adapted to work in densely populated urban slums, Resources need where conditions of poverty, overcrowding, and poor quality services interact to magnify the to be crowded in risks of poor water and sanitation. Representative data are not available for urban slums in until communities Indonesia, but RISKESDAS data from 2013 show that stunting rates among children in the reach a coverage bottom income quintile in urban areas are nearly 1.5 times those for urban children as a whole level high enough to impact health (48 vs. 33 percent) and higher than those for children in rural areas (42 percent). Multisectoral and nutrition approaches have largely focused on rural areas, but the challenge in urban slums and informal outcomes. settlements is complex, as an effective response involves a multitude of actors and is complicated by institutional constraints and tenure insecurity. Additional analytical work is needed to investigate the WASH characteristics and other nutritional determinants in urban slums and informal settlements and to determine how to effectively engage different actors under the National Slum Upgrading Program (Kotaku). In particular, a better understanding is needed of the contamination pathways unique to these settings, where the typical play environment of children includes solid waste disposal sites and contaminated water bodies, and the disease vectors include not just flies, but also cockroaches and rats. Table 6.1 summarizes the key recommended actions that will support the GoI to achieve the universal access target, with a particular focus on the needs and constraints of the B40. Suggestions are made as to the appropriate responsible agency/stakeholder, based on consultation with government actors and stakeholders. The WASH Poverty Diagnostic in Indonesia covered a broad range of challenges facing the Water Supply and Sanitation Sector in meeting the Universal Access Targets. However, the diagnostic is not exhaustive. Specific challenges recommended for further analytical work include: •• Water tariff affordability and impact of changes in tariffs on households living below the poverty line; •• Economic benefits of WASH provision for reducing income inequality, to include benefit incidence of net public spending on WASH; •• Relationship between poor WASH, disease environment, and child nutrition in dense urban informal settlements, and implementation of multisectoral approaches in these settings; •• Water Resources and Water Security Diagnostics, and the relationship with poverty; •• Institutional and political economy analysis of service delivery of urban sanitation, and of rural water and sanitation. 74 Improving Service Levels and Impact on the Poor Table 6.1: Key Recommended Actions and Responsible Agencies What Who Reduce Inequalities in Access and Quality Expand piped water services to a larger share of the B40 in urban areas Improve the efficiency and performance of PDAMs MoPWH, Bappenas, MoHA, private sector Enhance the capacity of LGs and PDAMs on tariff- MoHA, MoPWH, donor setting to support the establishment of cost- agencies, recovery tariffs Center of Excellence (CoE) • Conduct analytical work on the implementation of program, Association new regulations on tariffs and subsidies of PDAMs (Persatuan • Add specific tariff-setting content to existing Perusahaan Air Minum capacity building programs Seluruh Indonesia, or PERPAMSI), NUWSP Additional financing mechanisms to ease the Bappenas, MoPWH projects, financial and liquidity constraints faced by the poor NGOs, local financing • Continue and improve the Water Hibah scheme by institutions linking with investment on capacity improvement • Scale up microfinance and similar mechanisms • Encourage collaboration between Hibah and microfinance schemes • Combine existing targeting mechanism with income targeting to better identify eligible low- income households Increase demand and raise consumer awareness of MoPWH, MoH, PDAMs the benefits of piped water Adjustments to the current intergovernmental fiscal Bappenas, MoPWH, MoF, transfer system to better align transfers to needs donor agencies • Exercise alternative approach in the allocation of DAK and DAU • Diagnose private sector involvement in water sector, including commercial loans, private investment, and business-to-business collaboration in bridging the financing gap Improve the quality of alternative water sources for those who will remain on non-piped water supply Consistently monitor water quality risks to drinking MoH, District Health Office, water supplies, piped or non-piped, and make this Bappeda information publicly available • Strengthen the critical link across water and sanitation sub-sectors—e.g., ensure the alignment of the Master Plan for Drinking Water and City Sanitation Strategy table continues next page Improving Service Levels and Impact on the Poor 75 Table 6.1: Continued What Who Strengthen regulatory control for small water MoH, MoI providers to ensure that regulations on drinking water quality are met • Link water quality monitoring with licensing process for refilled bottled water providers Support the B40 in gaining access to improved sanitation • Strengthen STBM strategy by revisiting the “zero- Bappenas, MoH subsidy” for poor households • Identify various financial schemes to move up the Bappenas, MoH, MoPWH sanitation ladder, such as DAK, Sanitation Hibah, Village Grant, and community social responsibility (CSR) funds • Explore the possibility of targeted subsidy for Bappenas, MoH, MoPWH, the poorest segment of people to move up the PAMSIMAS sanitation ladder (from basic to improved latrines) • Adopt existing targeting systems that are already Bappenas, MoH, MoPWH, working well identifying low-income households Ministry of Social Protection (such as the UDB from TNP2K) to ongoing programs, including STBM, PAMSIMAS, etc. Bring more households into full sanitation and fecal waste service chain in urban areas Take holistic approach to planning in implementing Bappenas, MoPWH, Bappeda, citywide sanitation-inclusive approach PPSP • Apply the fecal waste diagram as a tool to assess citywide sanitation and identify priorities for city sanitation strategy • Ensure consistency in sanitation management at local level through PPSP (Acceleration of Urban Sanitation Development Program) and link it with decision-making on investment using central budget Adapt approach to behavior issues in urban MoH, MoPWH, MoEF sanitation, including enforcing the effluent standard • Effective implementation of STBM in urban areas • Enforce new regulations on effluent standards Elevate the profile of sanitation in political and MoHA, Bappenas, MoPWH, fiscal discussion MoH, AKKOPSI Improve Health, Nutrition, and Early Child Development Champion multisectoral approaches to reduce child stunting Capitalize on synergies of multisectoral Bappenas, MoPWH, MoH, approaches, including strengthening the existing Ministry of Social Protection scaling up nutrition (SUN) program and alignment with non-cash nutrition support table continues next page 76 Improving Service Levels and Impact on the Poor Table 6.1: Continued What Who Crowd in resources until communities achieve Bappenas, MoH, MoPWH, high coverage of sanitation Bappeda Adapt water and sanitation interventions to be MoH, PAMSIMAS more “child-centric” • Adapt existing STBM behavior-change communication materials and LGs capacity building programs to incorporate “baby WASH” • Ensure that the implementation of the “baby WASH” approach aligns with current nutrition- sensitive pilot of PAMSIMAS Enhance water supply and sanitation interventions MoH, Bappenas, STBM to be more impactful on nutrition outcomes Target slum areas and informal settlements Bappenas, MoPWH, Vice with multisectoral action President’s Office Notes 1. The Regional water systems platform is still at early development stage. These systems are planned for areas facing water scarcity. Despite Indonesia having relatively high rainfall, water shortages may result from insufficient storage capacity, poor water quality, and competing water demands. If a PDAM has insufficient water resources within its jurisdiction, it may seek water from neighboring LGs. Regional water systems would fall under the mandate of the Province, which may develop the system to supply multiple PDAMs. The facilities are constructed by DG Cipta Karya and then managed by provincial water institutions. 2. The urban sanitation platform is centered around three major areas of activities: (a) national community-based sanitation program (SANIMAS); (b) national urban sewerage; (c) septage management, including upgrading on-site sanitation and developing new septage treatment plants (IPLTs). 3. http://witiestudio.com/worldbank-map/. 4. IUWASH Annual Progress Report 2013, 2014, 2015. 5. A Multisectoral Nutrition Framework and Action Plan (MNFAP) has already been developed, which serves to guide internal World Bank multisectoral engagement for nutrition in Indonesia. The MNFAP identifies specific opportunities for multisectoral action, incorporates nutrition activities into multisectoral programs and analytical work, strengthens integration, and develops concrete implementation plans. The MNFAP is aligned with the recommendations of the WASH Poverty Diagnostic. Reference World Bank and Australian Aid. 2013. Urban Sanitation Review: Indonesia Country Study. Washington, DC: World Bank. Improving Service Levels and Impact on the Poor 77 Appendix A Interactive Dashboard Indonesia Water Supply, Sanitation, and Hygiene Poverty Diagnostics Spatial Analysis to Guide Pro-Poor and Poor-Inclusive Water and Sanitation Interventions in Indonesia Two of the main challenges in implementing pro-poor and poor-inclusive water supply, sanitation, and hygiene (WASH) interventions and adopting a multisectoral approach to reducing stunting is the coordination of multiple stakeholders across many sectors and the use of many different data sets. Ensuring that decision-makers have the tools to identify locations with multiple deprivations—high poverty, low access to improved water, lack of access of improved sanitation—is essential for the future well-being of these disadvantaged communities. Data and analysis can inform and facilitate actions that optimize efforts to reduce poverty and stunting along with efforts to increase access to WASH and maximize the use of available resources. The WASH-Poverty dashboard offers a new tool that provides information through maps—at both province and district levels—that visualize access to improved water and sanitation, poverty rates, and health outcomes (diarrhea and stunting) to help monitor inequalities in WASH services; this information can be used by the government, the World Bank, and other development partners. The dashboard illustrates how such data can inform geographic targeting to extend coverage and improve service quality for more pro-poor and poor-inclusive interventions (and the associated sanitation and hygiene efforts) to have the most impact on Indonesia’s health. Furthermore, the dashboard can be used to identify areas where further inquiry is needed to understand why service delivery fails within geographic areas. The dashboard can also model how changes in variables might influence the districts in which poverty, access to WASH, and health outcomes would overlap. For each variable presented, the dashboard provides different choices to allow the user to select the scope (overall, urban, or rural), category (all population, T60, B40), data year (from 2102 to 2015), and data unit (households or individuals). For easier selection, each column has a drop-down button with choices (see figure A.1). There are two different types of spatial analysis: a. Double Maps The double map screen (maps A.1–A.3) allows the user to make a comparison of two  different variables in a particular year, such as “access to improved water—all population—2015—urban—households” with “poverty rate 2015,” or “open defecation 2013” with “stunting 2013,” or “access to piped water 2014” with “access to improved sanitation.” Once both variables have been selected, the dashboard will automatically display the variables next to one another. Map keys are provided in the bottom left-hand corner of each map as a guide to the color coding. In addition, the map will show the title of each province or district and its numbers for easier identification. Improving Service Levels and Impact on the Poor 79 Figure A.1: Panels for Selecting the Variables Map A.1: Double Map Province: Access to Improved Sanitation and Stunting, 2013 Access to improved sanitation – all population – 2013 – overall household 0–10 10–20 20–30 30–40 40–50 50–60 60–70 70–80 80–90 >90 No Data Stunting - all population - 2013 - overall - household 0–10 10–20 20–30 30–40 40–50 50–60 60–70 70–80 80–90 >90 No Data b. Single Maps The single map screen allows the user to see the display of the overlay from two different variables, with the desirable range of each variable based on the latest data (Year 2015 for poverty and WASH access; and Year 2013 for health outcomes). Once both variables and the range have been selected, the tool will automatically display the variables geographically on the map with the map guide provided in the bottom left-hand corner. Provinces or districts that meet the selected first variable will be highlighted in yellow, 80 Improving Service Levels and Impact on the Poor Map A.2: Double Map Districts: Access to Improved Water and Access to Piped Water, 2015 Access to improved water – all population – 2015 – overall-household 0–10 10–20 20–30 30–40 40–50 50–60 60–70 70–80 80–90 >90 No Data Access to piped water – all population – 2015 – overall-household 0–10 10–20 20–30 30–40 40–50 50–60 60–70 70–80 80–90 >90 No Data and provinces or districts that meet the second parameter are highlighted in green. Provinces or districts that meet both variables are highlighted in blue and will be displayed in tabular form. Changing the variables or ranges allows the user to explore different dimensions of poverty, WASH access, and health outcomes, which may impact geographic targeting of the poor with water and sanitation interventions. The datasets were taken from Susenas (for poverty rate and access to water and sanitation) from 2012 to 2015, and from RISKESDAS (for diarrhea and stunting rate) for 2013, and data on open defecation is taken from STBM monitoring data. The dashboard was designed to be open to future improvement, such as (a) adding more time series for existing data sets; (b)  adding new data sets, such as nutrition and health data; (c) lowering the level of data analysis below district level (e.g., village); (d) translation into Bahasa Indonesia; or (e) fully customized index/rating for legend. To ensure its sustainable operation and maintenance, the dashboard will be linked to, and integrated with, the Government-led National Water and Sanitation Information Services (NAWASIS) under Bappenas leadership. Bappenas has agreed to host the dashboard. Improving Service Levels and Impact on the Poor 81 Map A.3: Single Map District: Poverty Rate and Access to Improved Water Districts with poverty rate >40% and access to improved water <50% Overall all population poverty rate 2015 household Overall all population access to improved water 2015 household Meet both criteria 82 Improving Service Levels and Impact on the Poor W17018