1 All Drops in the Bucket for Universalization Public Expenditure Review of Water and Sanitation in Brazil March 2022 © 2022 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. 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 Design of the report and photos by: Jose Ramon Galvez Perez/ Adriana Carolina Rocha Albuquerque. About the Water Global Practice Launched in 2014, the World Bank Group’s Water Global Practice brings together financing, knowledge, and implementation in one platform. By combining the Bank’s global knowledge with country investments, this model generates more firepower for transformational solutions to help countries grow sustainably. Please visit us at www.worldbank.org/water or follow us on Twitter: @WorldBankWater. About GWSP This publication received the support of the Global Water Security & Sanitation Partnership (GWSP). GWSP is a multidonor trust fund administered by the World Bank’s Water Global Practice and supported by Australia’s Department of Foreign Affairs and Trade, Austria’s Federal Ministry of Finance, the Bill & Melinda Gates Foundation, Denmark’s Ministry of Foreign Affairs, the Netherlands’ Ministry of Foreign Affairs, the Swedish International Development Cooperation Agency, Switzerland’s State Secretariat for Economic Affairs, the Swiss Agency for Development and Cooperation, and the U.S. Agency for International Development. Please visit us at www.worldbank.org/gwsp or follow us on Twitter: @TheGwsp. Acknowledgements The team that conducted this study consisted of Midori Makino (task team coleader) and Christian Borja-Vega (task team coleader), Julio Gonzalez (World Bank), Nicolas Drossos (World Bank), Claudia Tufani (World Bank), John Burgess (World Bank), Rui Cunha Marques (University of Lisbon), Evan Kresch (Oberlin College), Joaquim Bento de Souza Ferreira Filho (São Paulo University), and Emmanuel Garcia (Johns Hopkins University). Paula Carvalho Costa (World Bank), Maye Rueda Gomez (World Bank), and Alejandra Hernandez (World Bank) provided administrative support. The study was supported by the following World Bank’s leadership team: Paloma Anos Casero (Country Director for Brazil), Sophie Naudeau (Operations Manager), Rita Cestti (Water Practice Manager), Rafael Munoz (Program Leader Macroeconomics), Luis Andres (Program Leader for Infrastructure), Renato Nardello (Program Leader for Sustainable Development). In addition, the team received technical guidance and support from the following World Bank staff: Gustavo Saltiel, George Joseph, Dambudzo Josephine Muzenda, Norhan Mohamed Sadik, Qiao Wang, Anne Shrestha, and Alex Giron Gordillo. The excellent assistance provided by the Country Management Unit for Brazil is also acknowledged. The team is grateful for valuable comments received at various stages of the report from the following World Bank’s staff: Juliana Menezes Garrido, Paula Pedreira de Freitas de Oliveira, Viviane Virgolim Zamian, Stela Goldenstein, Victor Vazquez, Smita Misra, Dominick Revell de Waal, Vincenzo di Maro, Gabriel Lara, Gabriel Zaourak, and Kjetil Hansen. The team would like to extend thanks to the following Brazilian institutions that contributed with insights and views during consultation processes and during Brazil’s Water Week (2020): Ministry of Economy, Subsecretary of Infrastructure Development, National Sanitation System, National Water Agency, Ministry of Regional Development, National Health Foundation, Brazilian Association of Regulatory Agencies, Brazil TRATA Institute, National Development Bank of Brazil (BNDES), Federal Public Bank (CAIXA), Municipal Governments, and State Owned Operators. Contents Executive Summary ............................................................................................................................ 11 1. Introduction: What is this Water and Sanitation Public Expenditure Review about? ........................................................................................................................ 19 2. Macroeconomic Trends and Infrastructure Investments....................................................... 25 3. Public Financial Management, Governance and Institutional Arrangements of WSS ............................................................................................................. 39 4. WSS Access and Performance Benchmarks............................................................................ 49 5. Allocation of Public Expenditures in the WSS Sector ............................................................. 63 6. WSS Subnational Planning, Budgeting, and Spending............................................................ 71 7. Investments and Expenditures: A Shaken WSS Sector ......................................................... 105 8. Provisions of the New WSS Law, WSS Policies, and their Impacts ....................................... 119 9. COVID-19, Public Expenditures, and Health Burdens from Limited WSS Access .............................................................................................................. 129 10. Conclusions and Policy Recommendations .......................................................................... 139 References........................................................................................................................................ 143 Annex ............................................................................................................................................... 153 4 List of Figures and Tables Figure 1: Analytical framework of the WSS PER Global Initiative ...................................................... 23 Figure 2: Recent WSS PERs Supported by the Global Initiative ......................................................... 24 Figure 3: Primary Fiscal Deficit of Brazil, 2011-2018 (left) and Effect of Public Investment on GDP 2016-2019 (right) .................................................................................... 27 Figure 4: Model of Investment Flows for Main Infrastructure Sectors in Brazil ................................ 28 Figure 5: Infrastructure Investments by Sector (% GDP), 2019 ......................................................... 30 Figure 6: Infrastructure Services, Quality and Competitiveness in Brazil .......................................... 31 Figure 7: Tax Collection by Municipality (2010, R$ 000) .................................................................... 32 Figure 8: GDP variation according to scenarios. Percent change deviation from baseline, accumulated in 2030 ....................................................................................... 35 Figure 9: Modern Institutional Arrangement of WSS Provision in Brazil ............................................41 Figure 10: Benchmarks of Public Investment and Financial Management Institutions, 2018 .................................................................................................................... 43 Figure 11: Composition of the Municipal Public Budget .................................................................... 47 Figure 12: Access to Different Types of WSS by State and Metropolitan Regions ............................ 52 Figure 13: Percentage of Water Losses to Total Water Production in Brazil, 2011-2019 (left) and Average Cost (US$billion) of 1 percent Change in Coverage and Water Losses in Brazil 2011-2020 (right).......................................................................... 53 Figure 14: Types of Non-Revenue Water ........................................................................................... 54 Figure 15: Benchmarks of WSS Service Quality by States, 2019 ........................................................ 56 Figure 16: Water Service Quality and Access Index 2018 (left) and Wastewater Treatment Index (m3 treated/consumed) 2018 (right) ........................................................... 57 Figure 17: Operational Revenues/Net Investments and Cash Flow Risks of State WSS SOEs ................................................................................................................................ 59 Figure 18: Non-Revenue Water % per Connection by State WSS Companies, 2012-2017 (left) and Spending per Worker of State WSS companies, 2017 (right) ................ 61 Figure 19: Tariff Changes among Selected WSS State Companies (US$/m3) 2012-2017 (left) and Productivity of Workers per Connection, 2017 (right) .......................... 62 Figure 20: Federal Government Budget Allocation as Percentage of GDP, Selected Sectors (2004-2017) ................................................................................................. 66 Figure 21: Budget Under-Execution in Brazil, Real Values Adjusted with IPCA.................................. 66 Figure 22: Ratios of Executed and Committed Budgets .................................................................... 67 5 Figure 23: Consolidated National Public Expenditures (2002-2017) as Percent of GDP ……………68 Figure 24: Percentage of GDP of Programmed Expenditure to Functional Sectors (2002-2017) ................................................................................................................ 69 Figure 25: Distribution of Federal Budget for Urban and Rural Areas ............................................... 70 Figure 26: Federal Expenditure on Basic Sanitation (Capital and Operational Expenditure) ........................................................................................................................... 70 Figure 27: Population and WSS Budget Allocation at the State Level................................................ 73 Figure 28: Budget Execution of WSS at the State Level (2013-2019) ................................................. 73 Figure 29: Total Expenditure in WSS, by State (2019) (In Million R$)................................................. 75 Figure 30: Per Capita Expenditure in WSS, by State in R$ (2019) ...................................................... 76 Figure 31: WSS Expenditure by State for 2013 and 2019 .................................................................. 77 Figure 32: Ratio of Transfers to Municipalities Executed Payments against Commitments for Functional Classification No. 17 (Saneamento Básico) .............................. 78 Figure 33: Ratio of Transfers to Municipalities Executed Payments against Commitments for Functional Classification No. 17 (Saneamento Básico) in Rural Areas** ................................ 79 Figure 34: Ratio of Transfers to Municipalities Executed Payments against Commitments for Functional Classification No. 17 (Saneamento Básico) in Urban Areas** ............................... 80 Figure 35: Efficiency of Spending by Broad Functional Categories: Ratio of Executed/ Appropriated Budgets to States and Municipalities, 2005-2018................................................................... 81 Figure 36: Percent of Municipalities according to the Preparation of Water and Sanitation Municipal Plans according to Municipal Population Size and Region .................... 83 Figure 37: Registry of UVUC in Municipalities as proportion to their revenues 2013-2018 84 Figure 38: Types of WSS Providers (left) and Type/Status of Municipal Plans (right)......................... 85 Figure 39: Efficiency of WSS and Financial Liquidity of Municipalities ............................................... 86 Figure 40: Operational Revenues (median, R$) for Different Types of Entities of Coverage for Water (left) and Sanitation (right)................................................................. 87 Figure 41: Brazil is Lagging in Terms of Coverage, with Limited Spending, SNIS ................................ 88 Figure 42: Index of Municipal Development for Health and WSS in Brazil......................................... 89 Figure 43: Distribution of Municipal per Capita Spending on WSS 2015 (left) and Stunting Rate (2017) (right).............................................................................................. 90 Figure 44: Share of water expenditures and subsidies per deciles .................................................... 96 Figure 45: Water Omega – Targeting subsidies by region, by state ................................................... 99 list of figures and tables 6 Figure 46: Sanitation/Wastewater Omega – Targeting subsidies by region, by state…………………….99 Figure 47: Water Omega- Access cross map, by state ..................................................................... 100 Figure 48: Sanitation Omega- Access cross map, by state .............................................................. 100 Figure 49: Water Errors of Exclusion, by region (left) and state (right)........................................... 101 Figure 50: Sanitation/wastewater Errors of Exclusion, by region, by state..................................... 103 Figure 51: Water and Sanitation/Wastewater Omegas - Errors of exclusion map, by state ................................................................................................................................. 104 Figure 52: Annual Capital Costs of Reaching the Entire Population to Achieve Universal Access of Different Service Levels by 2030, Brazil................................................ 107 Figure 53: Water and Sanitation Public Investments as % of GDP in Brazil, 1971-2019 .................. 108 Figure 54: Brazil pension spending is high: Share of population and spending as % of GDP, 2017 (left) and Debt and public spending during the COVID-19 pandemic by country, 2020 (right) ........................................................................................................ 111 Figure 55: Investment in Infrastructure as a Percentage of GDP .................................................... 112 Figure 56: Number of interruptions in WSS public works and infrastructure upgrades, 2005-2015 ............................................................................................................ 113 Figure 57: Balance of Expenditures and Revenues in the WSS Sector in Brazil, 2010- 2023………………………………………………………………………………………………...………… 113 Figure 58: Public expenditure in LAC (left) and in Brazil (right) (average 2007-17) ......................... 115 Figure 59: Wastewater treatment spending per capita compared to country’s GDPper capita, 2018 (left) and Distribution of coverage of sanitation by type of service in Brazil, 2019 (right) ............................................................................................ 116 Figure 60: WSS budget allocations, commitments and distribution ............................................... 117 Figure 61: Implementation and legality of municipal plans ............................................................ 121 Figure 62: Years in which WSS policy was created in municipality ................................................. 122 Figure 63: Year of BSP creation by municipal GDP quintile, 2000-2017 .......................................... 122 Figure 64: B2B competition model of private sector participation and stewardship to WSS SOEs .......................................................................................................................... 124 Figure 65: Child mortality over time ................................................................................................ 126 Figure 66: Share of avoidable deaths 0-5 years old Brazil ............................................................... 126 Figure 67: Treated and Non-treated before Treatment – Child Mortality ...................................... 127 Figure 68: Treated and Non-treated before Treatment – Share of Avoidable Deaths 0-5 Years ................................................................................................................... 127 7 Figure 69: Relationship between support provided to COVID, fiscal space and debt* (left) and measures adopted by 26 WSS state companies (SCs) for maintaining service during the pandemic (right).............................................................. 131 Figure 70: Fiscal support measures during the first 9 months of the COVID-19 pandemic (% GDP)................................................................................................................. 133 Figure 71: Brazilian states that enacted financial support measures across payments and tariffs of WSS services, 2020 ........................................................................ 133 Figure 72: Number DALY’s Lost and Deaths due to Diarrheal Diseases Attributable to Unimproved WASH by Federation Unit ............................................................................ 136 Figure 73: WSS Investments in Billion US$ (2018 prices) .................................................................. 137 Table 1: Annual increase (%) in investment in WASH to achieve universalization by 2033, by region .................................................................................................................. 34 Table 2: Macro regions GDP increase. % variations, accumulated in 2030 ........................................ 36 Table 3: Real consumption by household group. Percent deviation from baseline in 2030 ............. 37 Table 4: Coverage Rates of Piped water/Treated Water, Sewerage and Wastewater Treatment in Brazil ...................................................................................................................................................51 Table 5: Share of Population with Water and Sewer Access, with Connection and Paid Water and WasteWater Bill in Brazil, by Rural/Urban, Expenditure Deciles ................... 92 Table 6: Share of Population with Water and Sewer Access, and Share of Population with Connection and Paid Water and WasteWater Bill in Brazil, by Region, State ................ 93 Table 7: Amount and Share of expenditure on water, share of exp on sewer (monthly)................. 95 Table 8: Omega Decomposition for Water Subsidies ......................................................................... 97 Table 9: Omega Decomposition for Wastewater Subsidies ............................................................... 98 Table 10: Errors of inclusion and exclusion of water consumption subsidy, national, urban, rural ........................................................................................................................... 102 Table 11: Errors of inclusion and exclusion of wastewater consumption subsidy, national, urban, rural ............................................................................................................ 102 Table 12: Responsibilities of different Levels of Government for basic public goods .......................110 Table 13: Cases, deaths during the COVID-19 pandemic.................................................................. 130 list of figures and tables 8 Abbreviations ABAR: Associação Brasileira de Agências de Regulação ABES: Associação Brasileira de Engenharia Sanitária e Ambiental ANA: Agencia Nacional de Águas (National Water Agency) AREs: Agências de Regulação Estaduais BC: Banco Central do Brasil BNDES: Banco Nacional de Desenvolvimento Econômico e Social BOOST: BOOST Public Expenditure Database BRDE: Banco Regional de Desenvolvimento do Extremo Sul, BSP: WSS Plan CAPEX: Capital Expenditure CEDAE: Companhia Estadual de Águas e Esgotos do Rio de Janeiro CEF: Caixa Econômica Federal CERI: Centro de Estudos em Regulação e Infraestrutura CESAN: Companhia Espírito-Santense de Saneamento CESB: Companhias Estaduais de Saneamento Básico CGU: Controladoria-Geral da União (Office of the Comptroller General) Cisb: Comitê Interministerial de Saneamento Básico CNI: National Federation of Industry CODEVASF: São Francisco and Parnaíba Valleys Development Company DALY: Disability Adjusted Life Years DCA: Declaração das Contas Anuais DRU: Desvinculação de Receitas da União (Delinking of Federal Revenues) EBAPE: Escola Brasileira de Administração Pública e de Empresas EESB: Empresas Estaduais do Saneamento Básico EMEs: Emerging Market Economies FGV: Fundação Getulio Vargas FGTS: Fundo de Garantia do Tempo de Serviço (Severance Pay Fund) FUNASA: Fundação Nacional de Saúde (National Health Foundation) GBD: Global Burden of Disease GDP: Gross Domestic Product GRP: Gross Regional Product IBGE: Instituto Brasileiro de Geografia e Estatística IBNET: International Benchmarking Network for Water and Sanitation 9 IDB: Inter-American Development Bank IFDM: Índice Firjan de Desenvolvimento Municipal (Index of Municipal Development) IFGF: Índice Firjan de Gestão Fiscal (FIRJAN Fiscal Management Index) IHME: Institute for Health Metrics and Evaluation IMF: International Monetary Fund INFRALATAM: Infrastructure Investments in Latin America Database IPCA: Índice Nacional de Preços ao Consumidor Amplo (Extended National ConsumerPrice Index) ITB: Instituto Trata Brasil IPTU: Imposto sobre a Propriedade Predial e Territorial Urbana (Property Tax) ISS: Imposto Sobre Serviços (Taxes on Services of any Nature) JMP: Joint Monitoring Program LAC: Latin America and the Caribbean Region LDO: Lei de Diretrizes Orçamentárias (Budget Guidelines Law) LOA: Lei Orçamentária Annual (Annual Budget Law) LRF: Lei de Responsabilidade Fiscal (Fiscal Responsibility Law) MBSP: Municipality WSS Plan MDR: Ministério do Desenvolvimento Regional (Ministry of Regional Development) MDS: Ministério da Saúde (Ministry of Health) ME: Ministério da Economia MMA: Ministério do Meio Ambiente MPO: Ministério do Planejamento e Orçamento e Gestão (Ministry of Planning, Budgeting and Management) NDB: New Development Bank NRDBs: National and Regional Development Banks NRW: Non-Revenue Water NTS: National Treasury Secretariat OECD: Organization for Economic Co-operation and Development OFSS: Orçamento Fiscal e da Seguridade Social (Fiscal and Social Security Budget) PAC: Programa de Aceleração de Crescimento PB: Performance Budgeting PEF: Plano de Equilíbrio Fiscal PEFA: Public Expenditure and Financial Accountability PER: Public Expenditure Review abbreviations 10 PFM: Public Financial Management PPOs: Private Providers and Operators PPP: Public-Private Partnership PLANSAB: National WSS Plan PMSB: Plano Municipal de Saneamento Básico (Municipal WSS Plan) PNAD: Pesquisa Nacional por Amostra de Domicílios PNDS: Pesquisa Nacional sobre Demografia e Saúde PPA: Plano Plurianual (Pluriannual Plan) REM-F: Ranking de Eficiência dos Municípios - Folha SABESP: Companhia de Saneamento Básico do Estado de São Paulo SDG: Sustainable Development Goal SEF: State Finance Secretariat SEFAZ: State Finance Secretariat SEPLAG: State Planning and Management Secretariat SIAFI: Integrated Financial Administration System SIGA: Sistema de Informações sobre Orçamento Público Federal SIOP: Sistema Integrado de Planejamento e Orçamento do Governo Federal SNIS: Sistema Nacional de Informações sobre Saneamento SNS: Secretaria Nacional de Saneamento (National Secretariat of Sanitation) SOE: State Owned Enterprise SWA: Sanitation and Water for All TCEs: Tribunais da Contas Estaduais TCU: Tribunal de Contas da União TFP: Total Factor Productivity UNICEF: United Nations Children’s Fund UVUC: Unverified Unpaid Commitments VUC: Verified Unpaid Commitments WASH: Water, Sanitation and Hygiene WEF: World Economic Forum WHO: World Health Organization WSS: Water Supply and Sanitation WWTP: Wastewater Treatment Plant YLL: Years of Life Lost Indicator Executive Summary 12 Brazil has a pressing need for better Water Supply and Sanitation Services(WSS), but in recent years public investment in the sector has been declining both in real terms and as a proportion of GDP and is often spent behind schedule and inefficiently. Volatile macroeconomics, fiscal strain, and bureaucratic tangles have created a climate of general uncertainty for a sector that is key to the country’s future economic growth and productivity. Low investment to maintain and expand the network goes a long way toward explaining why only 27 percent of Brazilians live in municipal areas where WSS services have been rated as satisfactory. To meet its future goals of prosperity and human capital development, the country needs to invest more in WSS infrastructure and improve the execution of investments. Access for all to WSS services is key to preventing disease and reducing high levels of pollution discharged into rivers and soil across the country. Good WSS helps make citizens healthy, productive members of society. It heads off economic burdens to federal, state, and municipal governments that come from high health care spending and environmental damage. Access to piped water connection differs significantly between rural and urban areas. It is estimated that 93 percent of the country’s urban population have access to piped water connection but only 32 percent of its rural counterparts do likewise headlining a wide urban-rural gap in access to piped water connection in the country. The urban-rural gap for sewer connection is found to be even wider – only 5 percent of the population living inrural areas are connected to the sewers while close to 71 percent of the urban population have access to sewer connection. The urban-rural gap in access to piped water and sewer connection are further reflectedacross expenditure per capita deciles which could be a proxy for household wealth. Here, it is found that only 60 percent of individuals from the bottommost expendituredecile have access to piped water connection whereas as much as 93 percent of individuals from the topmost decile have access to 13 piped water connection. In fact, the share of individuals with access to piped water steadily increases across each decile. The same is observed for sewer connection. Only 35 percent of individuals from the bottommost decile are connected to the sewers while as much as 85 percent of individuals from the topmost expenditure decile have access to sewer connection as their sanitation facility. A new legal framework for water and sanitation was approved in 2020. The new legal framework provides national guidelines and rules for the allocation of federal public resources and financing to the WSS sector. The main objective of the law is to enable theuniversalization of services, by providing the regulatory uniformity of the sector and allowing the regionalized provision of the service as an instrument to guarantee the economic and financial sustainability of providers. Nevertheless, this important steppingstone in regulatory reform of the sector, the financial fragility is being exacerbated by the COVID-19 pandemic. A horizon of general economic recovery that was in sight in 2019 vanished in 2020 as the COVID-19 lockdown exposed the country’s fragility, leading to GDP contraction of 9.7 percent in the second quarter of 2020. Multiple WSS SOEs, already out of compliance with regulations established by law, were pushed toward insolvency. A critical initiative of the new law is to reduce the monopoly power of WSS SOEs and foster competition in service provision by ramping up public-private partnerships in the sector. The goal is that states and municipalities will “reset” ongoing concessions contracts and create a bigger role for private capital, but only with explicit roles, stewardship, and comprehensive oversight from the National Water Agency. Furthermore,the new WSS law of 2019 establishes equal grounds of competition between state-ownedand private companies. Issues of economic impact get insufficient consideration in the budgetary process of the WSS sector. There are four basic elements to consider: regulation, type of costs, forms and sources of financing, and tax and fiscal policies. In the long bureaucratic process of arriving at WSS budgets, little thought goes toward how to craft the best combinations of these elements and reduce the specific economic burdens that they create. The sector has suffered from high federal budget under-execution, a slow tariff reform for increasing revenues, and weak targeting of cross-subsidization policy in the past decade. When funds are allocated, excess bureaucracy and outdated rules often prevent the money from being spent on schedule. Brazil could realize a potential gain of 4.4 percent of GDP by addressing these inefficiencies in public spending in the infrastructure sectors, and up to 1 percent of GDP from WSS infrastructure alone (Ministério da Economia 2020; World Bank 2017a; IPEA-ENAP 2017), studies have estimated. Despite these potential benefits, Brazil’s WSS budget cycle still operates executive summary 14 with scant connection between policy making and funding capacity. Many municipalities are trapped in those pitfalls because local governments depend excessively on the central government for funding for investments. WSS expenditures remain vulnerable not only to macroeconomic shocks, but to limitedfinancial maneuverability caused by high personnel costs and other expenses. The absence of harmonized public accounting in the Fiscal Responsibility Law (LRF) has permitted a rapidly growing payroll at the state level. In addition, the sector struggles to meet high fixed and asset maintenance costs, which have long lifetimes. Complex procurement processes and limited operational management restrict implementation of expenditures. As a result, the WSS sector is left with scant ability to use public budgets, whether federal or local, to invest. Moreover, high spending on pensions makes it harder for public WSS SOEs to thrive financially and generate multi-year investment plans. Other pressure comes from spending on electricity, which is the main cost of water utility companies after labor. SOEs in metropolitan areas and state capitals face additional pressure because they cross- subsidize their counterparts in smaller cities. At most SOEs, costs are pushed up by widespread loss of water passing through outdated infrastructure, while revenues are curtailed by chronic non-payment of customer bills. Despite important regulatory advancements in the sector, Brazil still lacks comprehensive penalties or rewards for water companies based on the number and duration of supply interruptions to consumers. Improving efficiency of operational spending could amplify budgets by transferring costs to water tariffs. Moreover, a bigger role for private companies could foster a climate of competition that would help improve service quality. Brazil’s financial rules and regulations favor reducing expenditures rather than raisingrevenues when attempting to address budget problems. This reality results in dwarfedinvestments to replace assets, improve the quality of infrastructure, and expand coverage. Scant attention goes to devising ways to raise SOEs’ revenues. Brazil is the seventh largest economy in the world, it ranks 112 in the sanitation ranking among 200 countries1. Sanitation expansion between 2011 and 2019 showed an average rate in the decade of 4.1percent per year, which was smaller than the previous decade (2000–2009): 4.6 percent per year. The Basic Sanitation Index showed that Brazil performs worse in the WSS sector than North America and Europe, but also to those of some countries in North Africa and the Middle East, with much lower average incomes than Brazil. Even countries like Ecuador, Chile, Honduras, and Argentina rank better in terms of WSS. 1 The index considers aspects of coverage, quality and service performance of water and sanitation services (WSS). See Instituto Trata do Brasil (2019) Ranking do Saneamento Básico. Also see, Instituto Trata do Brasil (2014) ‘Benefícios Econômicos da Expansão do Saneamento Brasileiro. 15 Public budget allocations to WSS account for a smaller share of GDP than those of comparable sectors. In the last decade, the GDP share of WSS budget has declined while in certain other sectors the figure has proportionately increased (basic education) or remained reasonably stable (primary healthcare). Only four out of the 27 states of Brazil had greater WSS expenditures in 2019 than in 2013 (Sergipe, Rio Grande do Sul, Tocantins, and Distrito Federal). This is large part of why Brazil’s water network has remained practically unchanged over the last decade. Brazil has made laudable gains in sanitation coverage in that period, however. The cost of adding one more household to a water supply network (marginal cost) is high compared to adding a household to the sewage network or treating one extra cubic meter of wastewater. To spend resources that they do get in a timely, efficient way, municipalities need to enactMunicipal WSS Plans (PMSBs)—but many have not done so. Without those plans in place and strong technical capabilities, it is difficult to honor concession contracts efficiently and spend public resources on schedule. Federal funds for WSS go predominantly to states with higher populations. Efficiency of spending these transfers showed lower fluctuations between 2005 and 2018 but had no real improvements in overall efficiency. executive summary 16 One of the key findings of these estimates is that investment in the WSS sector is their high concentration on raw inputs, machinery, and construction. 50 percent of the urban water investments is concentrated in only five states which include São Paulo, Rio de Janeiro, Minas Gerais, Rio Grande do Sul, and Santa Catarina. In these five states, annual investments in urban WSS systems of around 0.7 percent of their GDP, on average, will need to be made to reach universalization objectives. COVID-19 is causing new pressure by raising contingent expenditures and limiting capital investments. In the time of a pandemic, strong governance and financially sustainability ofstate- owned WSS facilities could help combat economic contraction, by spending public resources efficiently, and promoting equity and development goals. In its current state, Brazil’s WSS system cannot deliver this support. Brazil hopes to achieve universal WSS service by 2033, but if current investment rates and practices continue, that date may be pushed to 2050—about 20 years late. The costs to society and the economy will be enormous if the country does not step up to the WSS challenge. It is not too late. With well-planned shifts in policy and resources, and faithful implementation of new laws, Brazil can meet the 2033 target. Subsidy reform is essential to ramp up access of water and sanitation to the poorest. An analysis of water and sanitation tariff and consumption subsidies offer insights of how regressive these subsidies are. A higher proportion of poorer individuals are being excluded from receiving the water and wastewater subsidies compared to that of richer individuals. Water and wastewater consumption subsidies inadequately target to the most vulnerable or poor populations. To improve the targeting performance of the water and wastewater consumption subsidies, the country could expand their service area for piped water and subsequently increase access to piped water among its population especially in rural or difficult-to-access communities. Nationally, it is found that the error of inclusion to be around 22 percent while the error of exclusion is around 57 percent for water subsidies. This means that as much as 22 percent of top 60 percent households of the income distribution have received the water subsidies when they should not have and that 57 percent of bottom 40 percent households of the income distribution did not receive water subsidies when they should have. The new water and sanitation law in Brazil promotes conditions to reduce risks andovercome challenges faced by the sector. Nonetheless, the quality of its implementation is essential to attain sector’s objectives and deliver results. The law promotes the standardization of WSS contracts for all types of service providers, which will reduce both regulatory perils and transaction costs of compliance to meet performance standards. The new stewardship role that ANA will play in the implementation of the law, also calls for institutional capacity support and development. ANA will have the legal authority bestowed to (i) design standardized contracts which must consider, explicitly, key performance parameters of efficiency and universalization; (ii) identify complementary sources of revenue that can improve overall financial performance of WSS SOEs and providers; and (iii) develop a technical procedure for calculating insurance or compensation schemes for reversible assets not amortized at the time of contract termination. 17 The WSS public expenditure review of Brazil has important findings and policy recommendations. FINDING Fragmentation across programs, levels of government, and budget 1 and non-budget sources results in efficiency loss, overlaps, and lack of coordination. Support technical capacity to help local WSS strategic plans improve federal-to-local transfers. This would improve (1) sector outcomes, (2) RECOMMENDATION budget execution efficiency, and (3) equity in the allocation of 1 investments. The new WSS Law will give theregulatory body ANA an explicit role in supervising and coordinating WSS investment and expenditure programs, so strengthened administrative capacity for ANA would bring better institutional coordination. FINDING WSS financial and investment gaps are increasing over time, at a higher 2 rate than budget and revenue growth, with modest rates of return against capital (investments) and limited technological innovations (metering, supervision, service delivery) in the sector. RECOMMENDATION Diversify and expand WSS SOEs’ revenues and ramp up private-sector 2 investments in the sector under the new WSS law, so as to marshal extra resources to enhance service quality and expand coverage in vulnerable areas. executive summary 18 FINDING Budgets allocated, and expenses incurred in the sector are required 3 to balance out at zero economic profit, but WSS SOEs are far from that goal because expenses add to their financial profit, raising the economic cost of service delivery. RECOMMENDATION Systematically apply Performance Budgeting (PB) to make allocations 3 and expenses converge to zero economic profit and increase fiscal transparency. Specify guidelines and institutional support to expand the use of PBs. FINDING With WSS investments almost cut in half (from US$6.8 to US$3.6 billion), 4 the country is doing more with fewer resources and finding ways to optimize its budgetary allocation on a yearly basis. Non- revenue water and water loss are major drags on performance. Employ programs to reduce non-revenue water and water loss,helping RECOMMENDATION improve financial performance and increase returns ofinvestments in 4 the sector. The Instituto Trata Brasil and the World Bank have estimated that gains from these steps could total R$21-37 billion (US$4-$7 billion). Comprehensive implementation of the 2019 WSS law could greatly aid FINDING the goal of universal WSS services by 2033. Universalization in turn 5 would bring extra benefits to Brazil in the form of avoidance of extra costs to the economy from disease that is caused by lack of access and unimproved WSS. Develop a robust framework of coordination between the Ministries RECOMMENDATION of Rural Development and Health to implement the law. Advocate for 5 multi-year planning and incremental budgets to attain universalization objectives and reduce costs to taxpayers and the health care system. 1. Introduction: What is this Water and Sanitation Public Expenditure Review about? 20 Water Supply and Sanitation services (WSS) are paid for in Brazil by a combination of tariffs from water users, public expenditures for programs, subsidies (from taxes) or grants (transfers), and private sector investments. The current report focuses on the public expenditure component of the sector. It looks at trends and state of spending allocations in the sector, reviewing their efficiency, equity, and main institutional bottlenecks. The report includes a benefit incidence analysis to demonstrate the importance of ramping up investments and achieving the socioeconomic benefits of universal service. Also, the report highlights the vulnerabilities that public expenditures face in relation to the macroeconomic and fiscal trends of the country, the current COVID-19 reallocation of funds, and how the recently approved water and sanitation law can provide a regulatory framework to enhance investments from the private sector. The delivery of water and sanitation services is a complex process involvingmultiple government ministries and agencies, as well as private sector and non-government providers. Public Financial Management (PFM) is an analysis method that maps complex funding and accounting issues and is used in this report. Analyzing WSS public funding this way is particularly fruitful for understanding its high infrastructure costs, prevalence of natural monopolies, and benefits for public health, socioeconomic well-being, and the environment. I. Objective The objective of this WSS Public Expenditure Review (WSS PER) is to analyze the trends, barriers, and opportunities of public expenditures to improve efficiency and equity of the sector. The WSS PER makes use of available information to identify ways in which public expenditures function and contribute to development outcomes. The analysis is then used to identify reforms that would make public spending more effective, efficient, and equitable, as well as detailing the sector’s information and financial gaps. WSS services constitute an integrated set of activities that need regulation, monitoring, and planning. This is particularly relevant for Brazil, because its WSS sector is undergoing shifts in regulation and policy with the aim of (1) reaching the ambitious goals of the 2033 universalization agenda, (2) increase general service quality and performance, and (3) leverage public finances to make the best possible use of scarce resources and meet development outcomes. 21 II. Audience This study provides information useful for institutions in charge of the budgetary process in Brazil, as well as public sector agencies that carry out the execution, oversight, and planning of the sector. Institutions related to the budgetary process in Brazil include the Ministério da Economia (ME), Tribunal de Contas da União (TCU), Controladoria-Geral da União (CGU), Tribunais da Contas Estaduais (TCEs), Comitê Interministerial de Saneamento Básico (Cisb). Public institutions in charge of WSS include Ministério do Desenvolvimento Regional (MDR), Agencia Nacional de Águas (ANA), Ministério do Meio Ambiente (MMA), Associação Brasileira de Engenharia Sanitária e Ambiental (ABES), Banco Nacional de Desenvolvimento Econômico e Social (BNDES) and other national and regional development banks (NRDBs), Empresas Estaduais do Saneamento Básico (EESB), Associação Brasileira de Agências de Regulação (ABAR), Agências de Regulação Estaduais (AREs), Departamento Nacional de Obras de Saneamento, Ministério da Saúde (MDS), Fundação Nacional de Saúde (FUNASA), municipal governments and private providers and operators (PPOs) of water supply and sanitation. 22 III. Scope and Definitions In Brazil WSS is referred to as the Saneamento Básico (Water and Sanitation in English) sector. It includes water supply, sanitation, sewerage, wastewater treatment, and pluvial infrastructure2. In the past few decades, definition of WSS in Brazil has evolved,with the notion of ‘‘water supply and sanitation’’ being progressively replaced by the concept of ‘‘basic sanitation,’’ which encompasses a more comprehensive approach to policy interventions on human habitats, aimed at protection of public health and the environment. This WSS PER functions as a standalone study given the complex institutional, regulatory, and budgetary processes that contribute to the delivery of services, resulting in large geographical variations. Also, since Brazil is a federal country, there are nuances and differentiated policies at the local level that affect the performance and efficiency of public expenditures of the sector. Water loss and energy efficiency for Brazil’s water utility sector varies by geographical area and has achieved little progress over the last five years. The average water loss in water utility companies in Brazil is about 40 percent (including both physical and apparent losses). In some companies, loss exceeds 60 percent. The high level of water loss reduces companies’ revenues, and consequently, their ability to obtain financing and invest in improvements. Expenditures on electricity are the water utility companies’ biggest cost after labor. As such, the companies would do well to enhance their energy efficiency. The expenditure categories in this study do not include the entire public investments of water supply and sanitation. This is because of the complexity of budget classification practices. Water infrastructure provision, for instance, involves a value chain of elements (collection, treatment of raw water, supply, reservation, storage, collection, and treatment of effluents, etc.) that requires investment resources all alongthe way. Some elements are classified in Brazil as basic sanitation, others as health management or environmental sanitation. To give another example, basin management involves elements such as protection against extreme climate events and ecosystem protection. Both “basin management” and “infrastructure” have different costs (budgetary functions and financing sources) that are managed by different institutions, such as a state management body, a supply concessionaire, or a hydroelectric power generation concessionaire. Hence, this analysis is not able to consider categories that may be important for the quality of WSS but are not included as a functional classification in the budget process for “Basic Sanitation.” 2 Also, it may include solid waste management. 1. introduction: what is this water and sanitation public expenditure review 23 IV. About the Report This WSS Public Expenditure Review aims to provide an update and detailed analysis ofthe trends of allocations, WSS services, and targets to achieve universalization in Brazil. The report follows the guidelines of the WSS PER initiative of the Global Water Practice and the Guidance Note of Public Expenditure Reviews for Water (World Bank 2012). The Brazil WSS PER has a scope of a comprehensive analytical framework (Figure 1) and discusses reforms (regulatory and institutional) that could help fill in the budgetary and investment gaps of the sector. The overall Water, Sanitation and Hygiene Public Expenditure Review (WASH PER) global initiative provided support in defining the scope and methodology used to standardize the analytical framework with WASH PERs conducted in seven countries (Figure 2). This report contains the following sections: Section 1 is the introduction. Section 2 presents the macroeconomic picture of the country and the levels of infrastructure investments. Section 3 reviews the institutional and public financial management system to illustrate the budget cycles and allocation system in Brazil. Section 4 shows basic WSS information in terms of coverage, service performance, quality, and financial and operational benchmarks across WSS state-owned enterprises (SOEs) and utilities. Section 5 shows the allocation of public expenditures, and Section 6 details issues related to subnational planning, budgeting, and spending that explain the status FIGURE 1 Analytical framework of the WSS PER Global Initiative 1 LIGHT Levels, Trends, & Composition Financial Gaps Literature Review MEDIUM Targeting Efficiency Analysis Benefit Incidence Analysis Institutional Review 3 COMPREHENSIVE Source: Water GWSP 2019. 24 FIGURE 2 Recent WSS PERs Supported by the Global Initiative WSS Spending as share of Gov Expenditures Bottom Quintile 3nd Quintile Top Quintile 2nd Quintile 4th Quintile Source: Water GWSP 2019. of inefficiencies and inequities in the use of public resources in the sector. Section 7 presents the funding gaps and gridlocks of investments and expenditures in the sector. Section 8 presents provisions and WSS policies impacted by the new WSS law. Section 9 contains a brief discussion of COVID-19 implications for the budget endowment of the sector, and health burdens from inadequate WASH. Finally, Section 10 concludes and presents policy recommendations. 1. introduction: what is this water and sanitation public expenditure review 2.Macroeconomic Trends and Infrastructure Investments 26 Public budgets allocated to the water and sanitation sector (WSS) do not rise at a steady pace but fluctuate considerably amidst fragile and volatile macroeconomic conditions. Brazil’s GDP contracted by 3.5 percent in 2015 and 3.3 percent 2016, marking the deepest recession in the country’s modern history. The economic crisis brought a rapid rise in unemployment—2.6 million jobs were lost in 2015 and 2016. Recovery was weak, with growth of just 1.1 percent in both 2017 and 2018. The unemployment rate peaked at 13.6 percent in March 2017 and only slowly declined to 12.7 percent in March 2019, with most of the regained jobs being low-skill and informal. Fiscal multipliers of public investments had little effect in containing drastic declines in GDP (Figure 3). But those economic downturns were dwarfed by shocks that hit the country in 2020 with the onset of COVID-19. Unemployment reached almost 20 percent in the first months of the pandemic. This prolonged crisis has threatened a decade of social progress from 2005 up to 2015, when 24.8 million Brazilians escaped poverty and the Gini coefficient fell from 0.59 in 1999 to 0.51. Sustained growth and fiscal consolidation in Brazil are vital to the stability of public funding for WSS. However, the literature has shown that when countries undertake fiscal adjustments, public investment usually takes a disproportionate share of the cuts. This pattern is particularly strong among emerging economies (World Bank 2003; Serven 2007). The apparent reason is that capital expenditure cuts may be more politically palatable than cuts in current expenditures (Ardanaz and Izquierdo 2017; Arezki and Ismail 2013). Ramping up WSS public investments depends on favorable macroeconomic conditions and sectoral clarity (regulatory, operational, and institutional) to bring complementary investments from the private sector. For its part, the recovery in private investment in Brazil’s WSS is expected to gain momentum only gradually, given low-capacity utilization and policy uncertainly about the pace of structural reforms. Swift progress on reform implementation, including accelerating fiscal consolidation, removing regulatory and policy distortions in product and factor markets, and improving the business environment, would help GDP growth. Fiscal constraints, on the other hand, will hold backgovernment consumption and public investment in the coming years, making infrastructure development highly dependent on private sector participation (Figure 4). Macroeconomic and tax policies affect the federal budget’s levels and allocations, particularly for the WSS sector, by reducing states’ financial strains and containing the increasing severance and staff costs of state-owned enterprises (SOEs). More than half of the states face liquidity constraints. Eight of them have already declared a state of financial misstep. This is due to rapid growth in payroll and pensions, and suspension of 27 FIGURE 3 Primary Fiscal Deficit of Brazil, 2011 – 2018 (left) and Effect of Public Investment on GDP 2016-2019 (right) (% GDP; 12-m flows) (% GDP; 12-m flows) 4.0 4.0 3 3.0 3.0 Cumulative spending multiplier 2.0 2.0 2 1.0 1.0 0.0 0.0 1 -1.0 -1.0 -2.0 -2.0 -3.0 -3.0 0 -4.0 -4.0 Nov -11 Nov -12 Nov -13 Nov -14 Nov -15 Nov -16 Nov -17 Nov -18 0 1 2 3 4 5 6 7 8 Quarter Source: Haver analytics, Goldman Sachs Global Note: Evaluated at high, 95th percentile, and low, 5th Investment Research Report 2019. percentile, initial stock of public capital over GDP ratios. Brazil thick red line and black dotted lines, 95% confidence intervals. Source: Izquierdo et al. 2019 and IMF 2016b, “Brazil’s Fiscal Multipliers.” WSS service payments due to COVID-193, resulting in revenue stagnation. Subnational bankruptcy presents a serious contingent liability to the federal government and reduces its budgetary levels, because most subnational debt (about 12 percent of GDP) is directly owed or guaranteed by the federal government, which may also be forced to take over subnational public services. Recent efforts to restructure state debt owed to the federal government had limited impact on resolving the states’ liquidity crisis. The efforts failed to address its underlying causes and created defaults and federal bailout. To overcome this, the federal government is moving towards a multi- year program (Plano de Equilíbrio Fiscal, PEF) to provide US$2.5 billion in federal guarantees for new loans to states in exchange for adjustment measures that will restore fiscal balance.4 3 This refers to payment of WSS services that consumers have not made or have “suspended” as result of COVID- 19, and that therefore reduce or stagnate the enterprise’s revenue. 4 The program would force a combined projected fiscal adjustment of about US$10.2 billion (0.5 percent of GDP) for the 13 states that potentially can join. 2 FIGURE 4 Model of Investment Flows for Main Infrastructure Sectors in Brazil • Sector Investors /EPC* • Instututional Investors Private Sector • Equity Funds & Asset Managers Sources of Funds • Commercial banks • Corporate bonds (incl. SOE issuances) Ports, Airports, Toll Some Supply roads, Railroads and Water User (esp. rolling stock), Treatment, Securitizable Fees Some Mass Transit Some Revenue Mobile, Fixed Generation Distribution (especially BRT & distribution lines, internet Light rail) ICT Energy Transport Water and Sanitation Flow of 80-90% 40-60% 30-70% 20-40% PPPs funds 10-20% 40-60% 30-70% 60-80% Transmission, Urban & rural Rural Some retail Telephony, roads, Shipping consumption channels, Tax Some fiber for poor Waterways, Wastewater Public Payers optic Railbeds, Most Treatment & Transfers backbone Mass Transit Sewerage Infrastructure Consumption (especially (tariff) support metros), Air for poor, irriga- Traffic Control tion, Flood Mgt. Federal, Local Budgets • DFls and MDBs Public Sector • Public bond financing Sources of Funds • State Dev Banks (BNDES) • * Engineering, Procurement, and Construction (EPC). Source: World Bank 2020, Cascade Paper. 29 The PEF law also includes changes to the Fiscal Responsibility Law (LRF) to better prevent fiscal imbalances. The absence of harmonized public accounting in the LRF has permitted rapid growth of payrolls at the state level, the main engine of states’ public expenditure. Changes to the LRF establish a stricter definition of personnel expenditures, limiting the interpretation from states’ courts of accounts. This would force all states to cut personnel expenditures by about US$6.3 billion (0.3 percent of GDP) in five years.5 A Fiscal Management Council, already foreseen in Art. 67 of the LRF, will improve coordination among the federal government, states, and municipalities. This could bring more certainty to fiscal and budgetary cycles. Reducing the federal government’s contingent liabilities should be accomplished through the appropriate sharing of fiscal risks among federal, state, and municipal governments (Dos Santos and De Castro 2018). This requires the creation of a framework that may reduce moral hazards in fiscal relationships for WSS between MDR/ANA and MinSaude. It will also need the acceleration of fiscal reforms in subnational governments to limit their structural expenditure growth. Given the strong importance of consumption taxes for state and municipal governments,6 the less decentralized states (with higher dependence on consumption in their GDP) have been the hardest hit by COVID-19’s economic impacts, including many poorer states in the North and Northeast of Brazil. In view of these policies, public WSS expenditures remain vulnerable to macroeconomic risks, with limited maneuverability to meet the sector’s targets. Private sector participation, combined with an improvement in efficiency of public expenditures and the simplification of the tax system, could make investments thrive and have greater impact. Higher investment in infrastructure is also necessary to ensure that WSS infrastructure is replenished and expanded to meet current and increasing demand. Increasing public investment at times of economic slack and low interest rates is an effective short-term boost of demand and adds to long-term benefits from increasing productive capacity. Nonetheless, Brazil has posted one of the lowest infrastructure investment levels (2.0 percent of GDP on average between 2000 and 2017) compared to its peers (Figure 5), resulting in a deterioration of the infrastructure stock, and creating acute bottlenecks for production.7 5 Also, branches of state governments will no longer be allowed to authorize salary increases beyond the terms of their mandates. 6 The municipal tax burden is determined primarily by two taxes: Tax on Services of any Nature (ISS) and the Property Tax (IPTU). While the first is an indirect tax, levied on the service sector’s production, the second is a direct tax on urban real estate. Between 1980 and 2016, these two taxes accounted for, on average, more than 60 percent of the municipal tax burden. In 2016, this rate was 57.7 percent. 7 According to ME and IBGE, low total factor productivity (TFP) growth between 1996 and 2018 has limitedthe increase in per capita income to just 0.7 percent per year since the mid-1990s. This figure is one tenth the rate in China and only half the average in OECD countries. 30 FIGURE 5 Infrastructure Investments by Sector (% GDP), 2019. 4.2% GDP 0.4 0.7 2.2% GDP 1.1 0.2 0.5 0.7 2.0 0.9 Annual investments average (2011 – 2 018) Necessary investments (modernization) Transport Energy Telecommunications Water and Sanitation Source: Lins and Ferreira 2019. Insufficient investment in infrastructure, particularly in WSS, hinders the potential of Brazil to enhance its competitiveness and human capital. The limited investment is already exacting a toll on infrastructure quality, holding back the country’s competitiveness (Figure 6). The gaps of investment vis-à-vis other countries have been increasing over the last 10 years. Because Brazil has limited fiscal space and a declining demographic dividend,8 accelerating productivity growth remains key to the country’s long-term growth. To this end, reforms should focus on boosting Brazil’s global competitiveness through infrastructure investments. The COVID-19 pandemic continues to increase contingent expenditures and aggravate limited capital investment by lowering tax revenues. The northern and northeastern states, for instance, lost close to R$113.7 billion in revenues (13.5 percent of total 2019 revenues). This has resulted in annual fiscal deficits of above 5 percent of GDP.9 Figure 7shows the increase of tax collection in many municipalities between 2005 and 2017. However, rising tax collection does not necessarily mean that the new resources would be available for ready spending10: there are regulatory restrictions on how municipalities use taxes. 8 The demographic dividend is the accelerated economic growth that can result from a rapid decline in acountry’s fertility and the subsequent change in the population’s age structure. 9 See (World Bank, 2020). COVID-19 in Brazil Impacts and Policy Responses. Washington D.C. http://documents1.worldbank.org/curated/en/152381594359001244/pdf/Main-Report.pdf 2. macroeconomic trends and infrastructure 31 FIGURE 6 Infrastructure Services, Quality and Competitiveness in Brazil 2017 Access to Select Infrastructure Services,circa Correlation between Competitivenessand 2008-2018 Infrastructure Quality 7 100 6 90 80 Percent of population Infrastructure Quality Index 70 5 60 50 40 4 30 20 10 0 3 2008 2018 2008 2018 2008 2018 2010 2015 Water supply Sewerage Electricity Internet/ICT and drainage 2 1 Rural Urban 2.5 3.5 4.5 5.5 Global Competitiveness Index Source: IADB, 2019. Setor de infraestruturas, em Source: WEF 2017. Perception-based infrastructure pesquisa para o biênio 2018-19. index. mean that the new resources would be available for ready spending10: there are regulatory restrictions on how municipalities use taxes. Municipal tax revenues are insufficient to address expenditure inefficiency (World Bank, 2017). Richer municipalities, with high fiscal revenues per WSS connection, tend not to be more efficient in the provision of WSS than poorer municipalities (World Bank 2018). Thus, it is likely that to comply with the regulatory framework of tax policy, many wealthy municipalities are spending resources that do not necessarily improve WSS outcomes. Because of the drastic increase in the demand for WSS and a demographic transition, the country requires a shift in efficiency of spending that is critical to achieving universalization. The macroeconomic and fiscal factors are the main determinants of the sustainability of public funding in key development sectors of Brazil, including WSS (World Bank, 2017). Undoubtedly, the country spends a lot more than it can afford, with current spending consistently rising and putting fiscal sustainability at risk. On top of that, the turmoil ofCOVID-19 has badly harmed the country’s SOEs (World Bank 2020a). 10 Municipalities must share their accounts regarding actions of the previous year to the NTS (STN). The NTS then has 60 days to make this information available to the population. Despite this rule, data of the 1,000 municipalities are not readily available or presented inconsistencies which prevented a deep analysis of public expenditures locally. 32 FIGURE 7 Tax Collection by Municipality (2010, R$) Collection of Tax Revenue by Municipality (2005) Collection of Tax Revenue by Municipality (2017) 2,000,010 2,000,010 4,000,010 7,000,000 21,000,000 4,000,010 7,000,000 21,000,000 Source: National Treasury Secretariat - NTS (Secretaria do Tesouro Nacional - STN) 2018 and FIRJAN Fiscal Management Index – IFGF 2017. I. WSS investment multipliers and impacts Using a computable general equilibrium model of Brazil, the TERM-BR11, the impacts of increasing investments in the Water and Sanitation (WSS) sectors in Brazil were estimated. The TERM-BR model calibrated for the year 2015 using the Brazilian input-output matrix of that year and other official databases that allow a detailed characterization of the Brazilian economy, such as the National Household Sample Survey (PNAD), the Consumer Expenditure Survey (POF), the Municipal Agricultural Production (PAM), all from the Brazilian Institute of Geography and Statistics (IBGE)12. The analysis was based on comparing a baseline scenario with simulation results. The model was calibrated to year 2015 compared to the period 2015 – 2 018 (last year available of the Brazilian National Accounts) when the baseline is set to represent the observed past values of the macroeconomic aggregates. The baseline is compared to future scenarios 11 See the link for a description of the model https://www.copsmodels.com/term.htm 12 A background paper on the TERM-BR analysis for this public expenditure review describes the methodology, steps, assumptions, and details of the estimates. 2. macroeconomic trends and infrastructure 33 based on projections of the economy with macroeconomic information from the IPCC SSPS2 scenario13, mostly for the GDP increases forecast, as well as with population growth estimates by IBGE, both nationally and by state. Once the baseline is estimated, the policy simulations14 can be compared to it, and the differences are reported as the policy results. The simulations were based on four scenarios, according to different assumptions related to a) sources of financing to the WSS sector (external and internal)15, b) changes in WSS investments, and c) productivity shocks based on the burden of disease from limited WSS access. The scenarios were carried as follows: • SCEN1: increase in WSS investment, all the investment is funded from external financial flows. No labor productivity shock. • SCEN2: increase in WSS investment. Part of the investment is funded through a reduction in domestic consumption, part is funded from external financial flows. No labor productivity shock. • SCEN3: same as SCEN1 + labor productivity increase. • SCEN4: same as SCEN2 + labor productivity increase. The simulations comprise investment shocks in the WS sector. The average annual investment needs (and the percent change in investments) between 2019 and 2033 to reach universalization is shown below (Table 1). Investment in the WSS sector uses mostly products from the metal and mechanical sectors, machinery, as well as construction. Hence, the geographical location of the industries producing those goods is important for determining the results. The investment in WSS results in labor productivity gains due to improvements in general health conditions. This is particularly true for the poorest populations. The laborproductivity16 gains are lower in the Southeast states because of high WSS coverage and less incidence of WSS diseases, compared to states in the North or the Northeast regions. The number of DALYs lost in a year due to unimproved WASH, then, is a proxy of the time loss that could be prevented by providing improved WASH. 13 For a description of the SSPS2 scenario see the following reference of the Inter-governmental Panel for Climate Change (IPCC). https://www.ipcc.ch/site/assets/uploads/2018/02/WG1AR5_Chapter12_FINAL.pdf 14 The scenarios elaboration for the simulations involves two main sources of data: on investment byregion in WSS, and on the productivity gains due to the investments on WSS. 15 These are important assumptions because with external funding for WSS investments, the macroeconomic variables affected are related to the external sector’s balance of payments, where are for internal sources of WSS investments, there are implications of reduced internal consumption to compensate the increase in internal WSS investments. 16 The labor productivity effect is based on the Disability-adjusted Life Years (DALY) concept. Roughly, theDALY is a metric that evaluates the burden of disease attributable to unimproved WASH (Water, Sanitation and Hygiene) services. 34 TABLE 1 Annual increase (%) in investment in WASH required to achieve universalization by 2033, by selected states. Region % variation Region % variation 1 Rondônia 15.85 9 Bahia 8.36 2 Acre 8.36 10 Region Southeast 8.14 3 Amazonas 15.97 11 Sao Paulo 0 4 Roraima 4.61 12 Paraná 0 5 Pará 16.98 13 Region South 12.10 6 Amapá 23.43 14 Mato Grosso do Sul 6.01 7 Matopiba 15.15 15 Mato Grosso 12.37 8 Region Northeast 11.07 16 Goiás and Federal District 6.05 Source: Original data from Instituto Trata Brasil, 2019. This is the measure of year productivity improvement we have used in the simulations17. The higher productivity gains from investments in WSS appear in states in the North region. States in the Southeast and South regions in Brazil present relatively lower productivity gains, since those states are already in a much better situation with respect to access to WSS services, hence they have lower burden of disease from WSS. As it can be seen, GDP increases in all aggregated regions for all four simulations (Figure 8). The exception is São Paulo (SP) state, which shows a slight fall in GDP (-0.025% accumulated in 2030) in SCEN3. This happens because SP is the largest and richest state in Brazil, and the fall in consumption required in that scenario is particularly important in that state. The largest increase in GDP due to WSS investments is observed in the Northeast (NE) region (the poorest of Brazil). This region has a larger share of the investment goods industries than the Center- west (CO) or Southern (S) regions, for example. At the same time, the NE region has the third largest investment inthe WSS sector (2015), after the Rest of the Southeast region (RSE), North (N) and São Paulo (SP). Although the RSE region and São Paulo have a larger value of investment in WS than the NE in the base year, the impact of the simulation on those region´s GDP is relatively smaller, due to the greater diversification of their economies when compared to the NE. The gains in GDP under the four scenarios show that impacts in São Paulo are smaller compared to other regions, when productivity gains are added to the simulations. 17 These estimates were based on the DALY Burden of Disease described in Section 9.II 2. macroeconomic trends and infrastructure 35 FIGURE 8 GDP variation according to scenarios. Percent change deviation from baseline, accumulated in 2030. 100 90 80 % GDP variation 70 40 0 North Northeast Sao Paulo Southeast South Central-West Scenario 1 Scenario 2 Scenario 3 Scenario 4 Source: Estimates from the TERM-BR model, 2021. 36 TABLE 2 Macro regions GDP increase. % variations, accumulated in 2030. SCEN2 SCEN4 SCEN1 SCEN3 Macro Region External funding + Domestic funding + External funding Domestic funding productivity productivity 1N 0.37 0.85 0.25 0.74 2 NE 0.43 0.95 0.29 0.83 3 SP 0.09 0.15 -0.03 0.04 4 RSE 0.20 0.30 0.11 0.22 5S 0.18 0.38 0.08 0.28 6 C0 0.09 0.27 0.04 0.22 Source: TERM-BR model results. N: North; NE: Northeast; SP: São Paulo; RSE: Rest of Southeast; S: South;CO: Center-west region. São Paulo is a state where WSS investments requirements to reach universalization have lowest impacts, yet the state receives substantial investments (see section 5 and 6). Likewise, this state has a relatively smaller share of the least skilled workers compared to the regional low skill labor rate average. Based on those conditions, the labor productivity shocks assumed for the region are negligible (Table 2), causing the other regions, where productivity growth is higher, to attract both labor and capital from SP. Even though the productivity gains in SCEN4 are enough to compensate for the GDP loss in SCEN3, it´s not as big as in the other regions. The simulated policy, then, would lead to a decentralization of the economic activity in the country, toward the least developed regions which would benefit the most. The greater GDP increase is observed in the Northeast (NE) region. This region has a larger share of the investment goods industries than the Center-west (CO) region, for example. At the same time, the region has the third largest investment in the WS sector in the database (2015), after the Rest of the Southeast region (RSE) and São Paulo. Although the RSE region and São Paulo have a larger value of investment in WASH than the NE in the base year, investment shocks are smaller, as seen before. Likewise, the impact of the simulation on those region´s GDP is relatively smaller, due to the greater diversification of their economies when compared to the NE. Model results also shed light to the distributional effects of the simulated policies (Table 3). In the table, real household consumption is used as a proxy for welfare. Households are classified according to household income, from POF1 (poorest) to POF10 (richest). As it can be seen in the table, SCEN3 and SCEN4 cause a general fall in households consumption compared to the baseline, due to the assumed hypothesis of domestically funding the investment in WSS. The increase in productivity included in SCEN4, however, partially compensates the fall in consumption, reducing the burden of the policy on consumption. 2. macroeconomic trends and infrastructure 37 TABLE 3 Real consumption variations by household deciles. Percent deviation from baseline, accumulated in 2030. SCEN2 SCEN4 SCEN1 SCEN3 Household group External funding + Domestic funding + External funding Domestic funding productivity productivity 1 POF1 (poorest) 0.28 0.26 -0.37 -0.35 2 POF2 0.47 0.68 -0.19 0.06 3 POF3 0.41 0.60 -0.26 -0.03 4 POF4 0.38 0.55 -0.30 -0.09 5 POF5 0.35 0.50 -0.34 -0.15 6 POF6 0.32 0.46 -0.39 -0.21 7 POF7 0.31 0.44 -0.42 -0.25 8 POF8 0.30 0.42 -0.45 -0.28 9 POF9 0.28 0.40 -0.51 -0.34 10 POF10 (richest) 0.27 0.38 -0.60 -0.44 Source: TERM-BR model results. GDP increases are expected even under the hypothesis of domestically funding part of the required investments. The positive effects on labor productivity generated by investments in WASH increase GDP gains and point to a decentralization of the effect toward the poorest regions in Brazil. The results show that unlocking budgetary inefficiencies and increasing investments of the WSS sector bring tangible benefits to the poorest. Improving access to WSS contributes to improved health, generates savings for households and national health budgets, and contributes to poor households’ economies by through reduced costs, and losses of time. These savings are translated into higher consumption which can be instrumental for moving their situation out of poverty or at least reduce poverty risks. These figures are important to identify the potential impacts of WSS investments, considering improvements of operational management and expansion of expenditures. A World Bank study18 in two states (São Paulo and Minas Gerais) highlighted the importance of financial and operational management to expand expenditures and raise their impact on infrastructure sectors, including WSS. The study showed that the execution of public works—the backbone of infrastructure investment— 18 Blancas, L., Chioda, L., Cordella, T., Oliveira, A., & Várdy, F. (2011). Do procurement rules impact infrastructure investment efficiency? An empirical analysis of Inversão das Fases in São Paulo State(Policy Research Working Paper No. 5528). Washington, D.C. 38 is a protracted and delay-prone endeavor in Brazil.19 The analysis found that the recent reform meant a 24-day reduction in procurement processes for large projects and a 7 percentage points drop in the likelihood of court challenges irrespective of project size. In general, delays in WSS procurement are not due to the complexity of design and scope. Instead, they grow from an unwieldy procurement process that undermines the efficiency of spending. Over the last three decades, low and declining WSS public investment in Brazil have been eroding the capital stock of WSS infrastructure, thereby increasing future financial and fiscal costs of the sector. Public investment has been chronically low and private investment in WSS infrastructure kept its pace to meet similar levels of Latin American countries. Nevertheless, the recession of 2015 and 2016 and the recent fiscal and financial crises plummeted private investments. The macroeconomic vulnerabilities enhanced budgetary instability of the sector of the Federal Government, and, in addition, the stiffnesses between the federal, state, and municipal regulatory frameworks, which define their roles and responsibilities of the sector, contributed to forming public expenditure bottlenecks. Even though the constitution grants that the regulation and provision of water and sanitation services rests within municipalities, state companies are de facto in charge of water and sanitation services. Compared to investment requirements for universalizing water services (0.1 percent of GDP), sanitation investments would require four times total investments (0.4 percent of GDP). This is because first more people have no access to safely managed sanitation compared to safely managed water services. Also, the target population for sanitation services is concentrating in mid-size and large cities, and population growth, considering projected internal migration from rural to urban areas, contribute to the increasing costs of sanitation provision. Also, the technologies for wastewater treatment and sewerage collection are more expensive than conveyance systems that transport water to the end user. These conditions would require that committed financial resources reach the local level and respond efficiently to the needs of municipalities and municipal blocs that are being established by the new basic sanitation law of 2020 (Law 14.026/2020). Tariff policies will need to safeguard the long-term financial viability of water and sanitation utilities, while ensuring predominantly that sanitation services remain accessibleand affordable for the poor. To meet those objectives, macroeconomic and budgetary stability are necessary conditions to ensure that the sector has lower financial uncertainty. 19 The study compared procurement performance of the largest water and sewage utilities along three efficiency dimensions: procurement process duration, likelihood of complaint resolution litigation, andprices paid to contractors. 2. macroeconomic trends and infrastructure 3 3.Public Financial Management and Institutional Arrangements of WSS The Brazilian Public Financial Management (PFM) is robust and generally follows 40 international budget cycle standards. Brazil has a well-established legal framework for the formulation, execution, and monitoring of the budget, including medium-term perspectives. The budget cycle starts with the strategic goals of the Government to be implemented with the following successive steps, starting with the: pluriannual and annual planning and budgeting, budget execution, financial reporting, and ending with the external oversight of the Supreme Audit Institution (SAI) and the legislative bodies. The planning function is placed under the responsibility of the Subsecretary of Governmental Planning (SEPLA), within the Special Secretary of Finance of the Ministry of Economy. SEPLA uses the SIGPLAN system for the preparation and management of Multi-annual Plan (Plano Plurianual, PPA). The PPA is the main instrument of the Federal Government for the implementation of its public policies. From a 4-year medium-term perspective, the PPA has the function to promote governmental action, along with other important tools, such as the Budget Guidelines Act (LDO) and the Annual Budget Law (LOA). One important element of PFM for WSS is to better align the PPA according to WSS medium term objectives, mostly related to improving efficiency of public expenditures and ramp up investments. Policy-based budgeting and productive spending of WSS have been hindered over the late decades, by the mandatory spending and budget rigidity, weaknesses in the subnational fiscal framework, and inefficiencies in the tax system (PEFA 2009 and IMF art. IV 2020). Beyond low investments in WSS infrastructure, the challenge remains in coordinating the multilayer programs and in finding management models that guarantee the economic and financial sustainability of the water supply systems in cities and mainly in rural areas. Although promising models such as the private, not-for-profit Union of Community Associations for the Maintenance of Water Supply and Sanitation Services (CENTRAL) have been developed, their coverage and operational capacity remains limited(as reported in the PAD-Bahia Produtiva). Thus, key WSS policy objectives towards the Universal access to Water and Sanitation, meant to be implemented through the Federal “Agua para todos program”; the State “Agua para todos program”; and various Municipal programs, as is the case of various programs in Fortaleza: Programa de Infraestrutura em Educação e Saneamento; Regulação Fiscalização e Controle dos Serviços Públicos pela ACFOR; Convivência Urbana20; and in Campo Grande (Mato Grosso do Sul): Urbanização e Saneamento de Fundos da Vale21. 20 See for further reference: https://www.opovo.com.br/noticias/fortaleza/2019/12/30/projetos-de-intervencao-urbana-serao- implantados-em-fortaleza.html 21 For further reference see: https://schettini.eng.br/web/servicos/ . FIGURE 9 Modern Institutional Arrangement of WSS Provision in Brazil 41 Budget allocation Capital investment programs for Financial support and co-fi nancing of water and sanitation water and sanitation services Direct public Indirect administratio concession Shared management Autonomous Public state Mixed company WSS Public Private entity company of (50+1% state State owned consortia concession (municipality) WSS ownership company Management and budget Indirect provision to SOE execution (state and municipality) PPP Concession or private concession Municipal stewardship and planning Regulatory Entity Service/ Tariff Consumer (residential / commercial/industrial Source: Own elaboration. 4 42 Most WSS services in Brazil are provided by public administration, through municipalities, mixed- capital companies, and public companies (Figure 9). Another option is to delegate, by bidding, the partial provision through concession contracts to private companies. In addition, municipalities may enter a public consortium for the purpose of directly providing WSS services. In some cases, however, the public consortium does not operate the WSS service, instead transferring the function to WSS State-Owned Companies (CESBs, Companhias Estaduais de Saneamento) or to private operators, through bidding. These leave the sector with limited room to harmonize and coordinate expenses and investments between public and private entities.22 Overall, PFM reforms in recent years have been designed aiming at supporting greater operational efficiency in spending. The Government has been strengthening its capacity and procedures for managing spending more efficiently, both at the Ministry of Economy (former MPO), and the executing agencies levels, although it was still in the early phases of these reforms, and it would be too early to assess their impact. The Ministry of Economy (ME) was continuing to refine the systems for the performance reporting and evaluation of federalprograms and has developed systems for monitoring the execution of voluntary transfers to subnational levels, through agreements (convenios) with states and municipalities. The automated financial management system (SIAFI) is considered as a critical support tool for managers to support further operational efficiency gains. However, the main priority has been to control aggregate spending, with relatively modest efforts towards shifting away from compliance to autonomy of the line ministries and spending units, and this orientation has not sufficiently encouraged managers to focus more on productivity and outcomes. A World Bank report (2018) recommends practical ways to improve governance of spending in the WSS sector. It cites ways to make financing more agile within the structure of the federal system, namely: (1) evaluate and organize different financing channels according to the highest funding gaps, (2) promote long-term vision for financing, in view of the horizons and amounts required for larger investments, and (3) identify cash flow generated by the “3Ts”— taxes (which may include subsidies), tariffs (which may include collection), and transfers (not incremental cost support). Brazil performs relatively low on specific public financial management indicators compared to other countries with similar levels of development and income. The IMF (2018) has published a benchmark study in which Brazil ranks low in terms of project selection, availability of funding, transparency in execution, project management, management of PPPs, central-local coordination, and multi-year budgeting (Figure 10). These are all areas in which the WSS sector is struggling, and where reforms could greatly aid the overall execution and quality of WSS infrastructure. A deeper dive into sector trends in the following section will give context for the financial and budgetary challenges. 22 See for a description of CESBs in Brazil the following link: https://aesbe.org.br/novo/companhias-estaduais-de- saneamento-sao-certificadas-por-orgaos-reguladores-com-a-comprovacao-de-capacidade-economico-financeira/ . 43 FIGURE 10 Benchmarks of Public Investment and Financial Management Institutions, 2018 1. Fiscal Rules 15. Monitoring of Assets 2. National & Sectoral Planning 14. Project 3. Central-Local Coordination 13. Transparency of Execution 4. Management of PPPs 12. Availability of Funding 5. Company Regulation 11. Protection of 6. Multiyear 10. Project Selection 7. Budget Comprehensiveness 9. Project Appraisal Brazil EME LAC Other BRICS Note: EME=Emerging Market Economies; BRICS= Brazil, Russia, India, China, South Africa; LAC=Latin America and the Caribbean Region. Source: IMF 2018 and Brazil: Technical Assistance Report of Public Investment Management Assessment.IMF Country Report No. 18/249. There is a high volatility of the budget allocation for the WSS in Brazil. In the case of the States of Ceará and Minas Gerais, budget allocation for WSS is volatile in various of its programs, like in infrastructure, Water Resources, Funds for management of hydric resources, and other multisectoral projects. In the financial reports, the information is summarized by function and subfunctions in several budget lines. The WSS is not a high priority sector in Brazil, as observed in budget allocation that remains at very low levels, when compared with the total budgets. The analysis of the expenditures presented in the latest Pluriannual and Annual Plans and budgets, by government functions and subfunctions, is demonstrating the primacy of resource allocation in the sectors declared as of high priority, such as Health, Education, Social Security and Urbanism, mounting a percentage of more than 70 percent of the predicted total, on average. WSS is laying among the bottom sector budget lines. At the Federal level, the WSS budget allocation is lower than 0.02 percent and 0.1 percent of the total annual budgets, in 2019 and 2020, respectively. The same trend has been observed at the State and Municipal levels, with a remarkable exception in the State of Ceará, with a budget allocation reaching 1.98 percent and 1 percent of the total annual budgets in 2019 and 2020. 44 Budget Execution in WSS is also demonstrating a very weak budget credibility at the Federal, State and Municipal levels. At the Federal level, WSS is fully executed at the main WSS function, while WSS budget channeled through the Social Assistance, Health, and Environment government functions, is presenting and very weak execution reaching -70 to -100 percent of the allocated amounts. Two lessons surface from recent PFM analysis of the WSS sector. First, the degree to which performance information is used by public-sector organizations is key to the success of PB reforms.23 Second, the use of such information is highly dependent upon institutional considerations such as power relationships, leadership, and culture. Also influencing are macro- factors such as the economic and political environment, specific institutional factors including procedures, capacity (both of people and systems), incentives and safeguards for performance, and political economy of the federal government and subnational governments and other stakeholders. In Brazil, many of these barriers may stand in the way of embracing PB’s underlying assumptions of technical rationality, openness, and transparency. Furthermore, strategic objectives within the WSS subsector are often misaligned, creating room for uncertainties of budgetary allocations and executions. A state government and its municipalities may have conflicting objectives because the former is committed to water environmental protection while the latter may give little thought to how dumped raw sewage will affect municipalities downstream. Hence the 2019 WSS legislation in Brazil aims to align and cluster municipalities so that they can allocate more resources to sewage treatment. Estache et al. (2016) found that sewage treatment is between 18 and 46 percent higher in municipalities that are clustered in a single metropolitan area. The clusters created better conditions for municipalities to access more federal transfers. When it comes to allocative decisions, spending on direct social transfers has been increasing, while longer-term investment, such as WSS infrastructure, has been losing ground. The latter is usually the first to suffer cuts when governments need to reduce spending. According to an IADB study conducted in 2019, there is a potential gain of 4.4 percent of GDP through addressing inefficiencies in public spending in the infrastructure sectors, and up to 1 percent of GDP through the WSS infrastructure sector alone. In the decentralizing of WSS services over the last two decades, the devolving of responsibilities came without an accompanying granting of fiscal powers to cover thesector’s increasing financing gaps. This dissociation between policy decisions and funding capacity means local 23 See https://blog-pfm.imf.org/pfmblog/2017/01/sequencing-of-performance-based-budget-reforms.html 3. public financial management, and institutional arrangements of 45 governments depend excessively on the central government for financial resources for investments. This disconnect may contribute to budgets that vary over time, susceptible to political cycles, and with multi-year uncertainty about funding’s frequency and size. Together, these elements hinder the ability to plan and carry out long-term investments that are critical to the sustainability and good functioning of the sector (IDB 2018). Despite the political economy and decentralization of WSS, official governance indicators for resource management and accountability for Brazil’s SOEs (Figure 10) show good performance compared to other countries with similar levels of income. But these positive numbers consider only flows and expenditures from federal accounting and allocations, and not the efficiency of budget spending in the different layers of subnational government. The resources allocated to the WSS sector at the federal level are relatively low with high inefficiency. For the decentralized funds, it is difficult to assess their efficiency (FGV-CERI 2016). In the WSS and other sectors the quality of internal and external audits provide information to improve management, and generate budgetary surpluses, which can be accounted by using unpaid commitments (FGV EBAPE 2017). Moreover, as for external control, the Courts of Accounts do not demand disclosure of unpaid commitments, and the Office of the Comptroller General (CGU) ignores the issue of unpaid commitments when ranking transparency based on online information provided by state and municipalgovernments to the public. Besides the growing debt at all levels of government, the weak regulation of unpaid commitments is reducing the credibility and transparency of the budget at all levels of government (FGV-EBAPE 2017). Brazil’s PFM institutions are stronger than the averages for EMEs and LAC in some areas and weaker in others. The stronger areas include national planning, budget comprehensiveness, company regulation, and asset monitoring. The weaker areas are theallocation and implementation phases, especially project appraisal and selection, protection of investments, funding availability, and project management. Most institutionsare assessed as medium or low in terms of practical implementation and effectiveness (IMF 2018). State and Municipal governments have a too narrow fiscal space, to be able to raise external funds for WSS financing. States and municipalities have been struggling with high debt and severe liquidity constrains. Some of the largest states have already defaulted in their debt repayments and are running payment arrears (wages and suppliers). The Federal government has provided substantial support through debt service relief over the years and, in the context of the 2020 War Budget, the support helped offset revenue shortfalls and covered extraordinary spending during the pandemic. So, reforming the subnational fiscal framework is a key priority towards sustaining the provision of core public services. (IMF Art IV 2020) 46 WSS budget allocation is at a very low level at the Federal, State and Municipal levels, jeopardizing WSS policy implementation. The projected investment needs are huge over the next ten years, exceeding R$600 billion in CAPEX, by 2033 (as this year is the expected target for WSS universalization in Brazil), to extend the networks, on the top of the capital needs (of more than R$145 billion) for the replacement and enhancement of the current WSS assets, related to the existing network. At the same time, budget allocations for WSS are below 0.1 percent of the total annual budgets at the 3 government levels. When considering the very low budgets allocated to WSS and the even lower budget execution rate in the sector the financing of the projected investments is simply not achievable by 2033. Federal, State and Municipal PFM have been facing very weak budget credibility, seriously impacting WSS coverage. The comparison between budgeted and budget execution amounts reveals a low capacity in implementing the extremely low budgeted WSS expenditure, due to (a) generally small deviations in aggregate expenditure; but (b) high deviations in expenditure composition; (c) deviation in revenue collection; and (d) the balance of expenditure payment arrears. Weak budget credibility, combined with very low Federal budget allocated to WSS, may delay the results of the regional approach in the sector. According to the new WSS framework, the Federal Government is “pushing” for a regional approach in WSS. States are required to structure city blocks aiming at scaling up and cross-subsidizing cities willing to use concessions, while the federal funds will be conditioned to cities that opt to join blocks. However, the extremely low Federal Government budget allocation, may not serve as a catalyzer for the implementation of the new framework. Current regulations at the State and Municipal levels would need to be updated to meet the legal requirements and may delay the implementation of the new WSS framework. The internalization of the new legal WSS framework would need a regulatory change at the State and Municipal levels and the regulatory agencies trained in the new legal frameworks, before the Federal, State, and Municipal Governments been able to update their WSS policies, Pluriannual Plans and Budgets and start implementing their WSS budgets. Technical operational and financial capacity looks very weak at the Municipal level for the elaboration, evaluation, prioritization, and implementation of WSS projects, including the more challenging PPPs. The technical capacities of Municipal and State WSS institutions are weak for the full implementation and monitoring of the new legal WSS framework, while public Federal and State banks are lacking the technical capacity needed for couching the public administration and financing PPPs. Figure 11 shows the composition of a municipal public budget in Brazil. On the left-handside are the different sources of inputs (yellow arrows) such as municipal revenues, government transfers, and government funds. On the right-hand side are the different budget outputs (yellow arrows) all drops in the bucket for universalization: public expenditure review of water and sanitation in Brazil 3. public financial management, and institutional arrangements of 47 FIGURE 11 Composition of the Municipal Public Budget Society Commerce Industry Culture Services Administration Municipal Revenue Social Assistance State Gov. Health Education Municipal Gov. Transfers Municipal School Transport Other Transfers Public Housing Budget Basic Sanitation FUNDEB/FUNDEF Plans Security Labor Management Sports and Leisure Environmental Management LDO* Industry, Commerce, Services Policies *LDO: Budget Guidelines Law (Lei de Directrizes Source: Authors’ own elaboration. such as expenditure in health, education, and basic sanitation. The municipal public budget is affected by society (teal rectangle) and also by economic activities (green circle) such as commerce, industry, and services. The municipal government is in charge of managing the budget and implements it through plans (among them the MBSP), policies, and the budget guidelines law. There are 19 budget lines in the Federal Government budgets for WSS, attributed to 4 different ministries. Although, there are not clear indications on how these budgeted amounts would be executed at the Federal, State and Municipal levels, while the main WSS program is financed by the Conditioned Priority Investments (PPA 2020 – 2 023) and distributed in 4 budget lines, including one state, one regional, and two administrative activities. The National Treasury oversees the Federal transfers to the Municipalities, but on its website only global amounts to the municipalities are reported, without indicating the funds allocated to WSS (RREO 2020). At the State and Municipal levels, the budget lines are clearly presenting the WSS budgets, while the use of internationally accounting and reporting standards allow the comparison between budgeted and executed funds. 48 Despite a relevant legal framework in place, institutional weaknesses can hinder proper linkages between medium term planning and budget execution in some States and/or Municipalities. Weaknesses have been observed in the PEFA 2009 for Ceará state, while stronger linkages have been reported in the PEFA 2009 for Minas Gerais. SEPLAGs at the State and Municipal levels are the main entities coordinating the planning and budgeting activities. A formal annual budget calendar is in place, while all budget units, including the ones involved in WSS are informed through a budget circular about their allocated ceilings for their Annual Budget Law (LOA) preparation. Existing instructions and financial IT systems allow the budget units to prepare their budget proposals on time. An adequate level of stakeholder participation and discussions exists, with inter-Secretariat committees supervising the overall budgeting process. Final decision on budget ceilings appropriation and distribution is subjected to the Executive’s final review and approval by the State and Municipal assemblies. Budget proposals are usually prepared and submitted within the statutory deadlines while the Assemblies use to review and approve the budget proposals on a timely manner, before the beginning of the next fiscal year. Associated planning and budgeting documents include: the Multiyear Action Plan (PPA); the Budget Guidelines Law (LDO); and the Annual Budget Law (LOA). Performance-based budgeting in the public sector could have a positive impact on WSS, but only if they are carefully planned and implemented, with the full engagement from all Government levels (Federal, State, and Municipal). In practice, the development of budgets based on the relationship between program funding levels and expected results from each program can be used as a tool to manage more cost-efficient outlays and effective budget outcomes. However, this approach need to be carefully designed, with fully engaged PFM instances in Brazil at the Federal, State, and Municipal levels, using the theory-driven analysis, focusing on “common themes related to international policy diffusion, organizational change, stakeholder politics and gaming, communication and information management, principal–agent dynamics, and institutional constraints” (Ho et al 2019), with extensive support from well-experienced practitioners, having long professional experience in implementing performance budgeting reforms. In sum, the recent economic challenges in Brazil, aggravated by the global pandemic, do not favor an increase of public sector investments in WSS over the upcoming years. In 2021, the Public Financial Management (PFM) in Brazil was operating in a narrowed economic environment that impacted the budgetary outcomes, such as: (i) aggregate fiscal discipline; (ii) strategic allocation of resources; and (iii) efficient service delivery. As State and Municipal governments have too narrow available fiscal space, they may not be able to raise internal or external funds for financing the WSS sector. Better coordination and capacity enhancement of all levels of government to improve budget execution and their financial management are fundamental for the implementation of the WSS law. The Federal Government can facilitate efficiency in public WSS budgets and create secondary regulations for private sector infrastructure investments. However, State and Municipal implementation capacities must be enhanced to make the best use and allocation of WSS financing. 3. public financial management, and institutional arrangements of 4. WSS Access and Performance Benchmarks 50 I. Access and Service Performance According to the Instituto Trata Brasil (2019a), the country still has almost 35million people without access to treated water, 100 million without sewage collection, and only 47 percent of sewage produced in the country is treated. This complicates the tasks of preventing diseases and reducing the current high levels of pollution discharge into rivers and soils across the country. Municipalities and state governments have limited financial resources to tackle the increasing challenges of providing safely managed WSS (ITB 2019a). The last decade (2011–2019)24 has brought only slow progress in WSS coverage, as data from PNAD (IBGE) and Instituto Trata do Brasil show (Table 4). In Brazil, the Continuous National Household Sample Survey (PNAD) of 2019 demonstrates current unmet needs. The general water distribution network, which served 82.4 percent of households in 2011, remained practically unchanged at 83.7 percent in 2019 and 82.8 percent in 2020.25 The same held for coverage of piped water in the 100 largest cities of Brazil. For sanitation, the story is slightly better in terms of expansion of coverage from 2011 to 2020, yet the overall coverage rate remains low. The national rate of safely managed (meaning sewerage is collected) increased 8.9 percent between 2011 and 2020, from 48.1 to 52.4 percent. The rise in WSS access was driven mostly by smaller urban centers and rural areas, since sewerage collection coverage for the 100 largest cities in Brazil increased by only 1.6 percent, from 69.1 percent in 2011 to 70.2 percent in 2020. One important coverage rate that increase substantially was wastewater treatment as a percentage of total water consumed. This percentage is important to avoid wastewater being spilled in the environment, creating hotspots of health risks for the population. In 2011, close to 37 percent of water consumed was treated in wastewater treatment plants (WWTP). By 2019, this percentage had increased to 47 percent, though it dropped to 43.2 percent in 2020 due to increasing water consumption in the COVID-19 pandemic. In the 100 most-populated cities of Brazil, the percentage of treated sewage with respect to water consumed increased by 5.8 percentage points between 2011 and 2020. 24 For 2020, data is not strictly comparable because of sample changes for the PNAD COVID-19 survey. 25 The decline could have been explained by the default payments since many households lost their main source of income during the first COVID-19 lockdown. Even though WSS SOEs implanted policies to waive and defer WSS payments, many families could not afford WSS payments even with the flexibility of waivers and payment deferrals. 51 TABLE 4 Coverage Rates of Piped water/Treated Water, Sewerage and Wastewater Treatment in Brazil Coverage of piped Coverage of Treated or treated water (% sewerage collection sewerage/water population) (% population) consumed (%) 100 Largest 100 Largest 100 Largest Year National National National cities cities cities 2011 82.4 93.5 48.1 69.1 37.5 46.7 2012 82.7 93.5 48.3 69.4 38.7 48.8 2013 82.5 92.9 48.6 69.1 39.0 48.1 2014 83.0 93.3 49.8 70.4 40.8 50.3 2015 83.3 93.4 50.3 71.1 42.7 51.7 2016 83.3 93.6 51.9 72.2 44.9 54.3 2017 83.5 94.6 52.4 72.8 46.0 55.6 2018 83.6 93.3 53.2 73.3 46.3 56.1 2019 83.7 93.6 53.5 73.9 46.9 56.7 2020* 82.8 90.7 52.4 70.2 43.2 52.5 Percent change 0.5 -3.0 8.9 1.6 15.2 12.4 (2011-2020**) Note: Largest cities in terms of population. * Estimate based on JMP 2019 and PNAD 2020. ** For investments the percent change is for 2011-2018 Source: PNAD 2011-2020 and Instituto Trata de Brasil 2019. Coverage of the general water network is uneven by region, varying from 58.8 percent inthe North region to 92.3 percent in the Southeast. Regional inequalities also show up in the proportion of households with access to the general sewage system: in 2019, the North and Northeast regions had the lowest coverage, with 27.4 percent and 47.2 percent, respectively, while coverage in the Southeast region reached an estimated 88.9 percent. The South and Midwest regions had the same coverage of 68.7 percent (Gomes et al. 2020). Regional disparities in access to budget allocations for cities with the largest populations (Figure 12) are making the expansion of services more expensive. Insufficient budgets for capital investments and service expansion are leaving fast-growing areas with lower rates of coverage. In areas with high population densities, access to WSS has not changed much over the last decade, even though expenditures in those areas can provide higher economic returns. In addition, WSS spending in urban areas has stagnated with higher rates of spending inefficiency. 52 FIGURE 12 Access to Different Types of WSS by State and Metropolitan Regions (a) Household access to water by type (b) Household access to water by type- Metropolitan Regions (% coverage) 100 2016 Central west 2017 2018 40 2019 2016 Northeast 60 2017 2018 2019 40 2016 North 2017 20 2018 2019 0 Southeast 2016 Goiânia (GO) Aracaju (SE) Manaus (AM) Maceió (AL) Macapá (AP) Natal (RN) São Paulo (SP) Fortaleza (CE) João Pessoa (PB) Salvador (BA) Porto Alegre (RS) Belém (PA) 2017 Belo Hrizonte (MG) Recife (PE) Grande São Luís (MA) Curitiba (PR) Grande Vitoria (ES) Rio de Janeiro (RJ) Florianópolis (SC) Vale do Rio Cuiaba Cuiaba (MT) 2018 2019 2016 South 2017 2018 2019 0 20 40 60 80 100 Access to water Deep well with Access to water Deep well with Springwater Springwater network pump network pump Low scale Other Low scale Other well source well source Source: PNAD, 2019 – 2020 Source: PNAD, 2019-2020 Water losses are increasing economic costs at such rates as to ultimately diminish the liquidity of WSS SOEs (ITB 2019a, 2019b). Brazil has water losses of 39 percent nationally (Figure 13), which is higher than losses in least-developed countries such as Bangladesh (22 percent), Uganda (33 percent), and South Africa (34 percent). In the largest 100 cities of Brazil, however, water losses have showed a sharp decline in the last three years, falling to 34 percent in 2019. Increasing coverage of WSS services and declining WSS system losses are necessary conditions to reach universalization in 2033. Over the last decade (2011-2020), the percent change in the rate of coverage for water supply was only 0.5 percent compared to 8.9 percent for sanitation coverage and 15.2 percent for wastewater treated (Figure 13). The rate of water loss, in contrast, increased by 1 percent between 2011-2020. Rates of WSS coverage and water losses have held largely steady during the COVID-19 pandemic, during which many households have been unable to pay for services (World Bank 2020a). 4. wss access and performance benchmarks all drops in the bucket for universalization: public expenditure review of water and sanitation in Brazil 53 FIGURE 13 Percentage of Water Losses to Total Water Production in Brazil, 2011-2019 (left) and Average Cost (US$ billion) for a 1 percent Change in Coverage and Water Losses in Brazil 2011-2020 (right) 41 WSS subsector Percent of Delta (percent Average cost 40 change in total WSS change in rate (2011-2020; coverage cost investment 2011-2020) USD billion) 39 38 How much does it 37 cost 1 percent 25 0.5 2.54 increase in water 36 supply? 35 How much doesit 34 cost 1 percent 40 8.9 0.22 33 increase in 32 sanitation 31 How much does it 2011 2012 2013 2014 2015 2016 2017 2018 2019 cost a 1 percent 20 15.2 0.6 increase in waste- National 100 largest cities water treatment? How much does it cost a 1 percent 15 -1.0 0.72 decline in water Source: PNAD 2011-2020 and ITB losses? The WSS subsector with the highest change in coverage between 2011 and 2020 was wastewater treatment, followed by sanitation and water supply. Given the proportion of WSS investment in these subsectors against total WSS investment, and the percentage change in the rate of water losses, the unit cost of a one percent change in these coverage rates ranges from US$0.6 to US $2.5 billion. Wastewater treatment investments represent the lowest proportion of WSS investments, compared to water supply, water loss, and sanitation investments. However, this subsector showed the lowest cost for a one percent change in the rate between 2011 and 2020. The costliest subsector was water supply (US$2.5 billion between2011 and 2020 to increase the rate of coverage by just 0.5 percentage points). This is because rates of water supply coverage in Brazil are already relatively high, and the cost of adding one more of the remaining households to the network is higher compared to the other subsectors (ITB 2019 and MDR 2018). Despite these varying costs of increasing coverage rates of WSS subsectors and reducing water loss, Brazil is still focusing a relatively large proportion of its budgets on water supply expansion. These expenditures are going to a handful of states with high population density and high number of urbanized areas, where the costs of an extra connection are comparatively low. States such São Paulo, Rio de Janeiro, Minas Gerais, and Bahia have allocated more than 40 percent of the WSS federal budget to this work. This creates a population-driven incremental budget effect because increasing coverage comes at increasing unit costs. 54 BOX 1 Non-Revenue Water and Water Losses in WSS systems The World Bank’s database on International Benchmarking Network for Water and Sanitation Utilities (IBNET) assesses the performance of water utility operators globally. Among the operators covered by the database, the average water loss was estimated at 38 percent. In addressing this problem, operators generally cannot eliminate all physical and commercial water losses, as this is neither economically nor financially feasible. A reasonable target for loss reduction is to cut it by half, which would make WSS production and distribution systems economically viable and water resources sustainable. Water loss reduction programs should always consider the trade-offs between the value of the water saved and the value of the investment needed to achieve that saving, not only in infrastructure but in the variable costs of commercial management and operations. At a certain point, when water losses are extremely low, the cost of additional loss reduction becomes prohibitively high. Types of water losses in a typical WSS system are outlined below. The principles of water balance and economic efficiency lie behind the need to target and reduce water losses. Also, efforts to curtail losses and non-revenue water improve the effectiveness of public expenditures by increasing the value of investments in expanding or rehabilitating systems. Non-revenue water (NRW) is an indicator that signals quality of investment oriented towards improving operational performance of WSS utilities and SOEs. NRW can occur through physical losses from leaking and broken pipes, caused by poor operations and maintenance, lack of active leakage control, and poor quality of assets. Malfunctioning meters and theft of water also contribute to the totals. FIGURE 14 Types of Non-Revenue Water Water Exported Billed Water Exported Revenue Water (corrected for known errors) Billed Metered Consumption Billed Authorized Revenue Water Consumption Billed Unmetered Consumption Authorized Unbilled Metered Volume Consumption Consumption Unbilled From Own Authorized Sources Consumption Unbilled Unmetered (corrected Consumption for known errors) System Input Customer Metering Volume Inaccuracies Water Supplied Apparent Unauthorized Losses Consumption Non-revenue Water Systematic Data Handling Errors Water Leakage on Transmission Losses and Distribution Mains Water Real Leakage and Overflows at Imported Losses Utility´s Storage Tanks (corrected for known Leakage on Service errors) Connections up to the Pointof Customer Metering Note: All data in volume for the period of reference, typically one year. Source: IBNET 2019 and International Water Association 2017. 4. wss access and performance benchmarks 55 II. Service Quality, and Financial and Operational Performance The budget allocated to the WSS sector provides only limited support and incentives toenhance service quality in a sustainable manner. To ensure water supply continuity, water regulators can use direct regulation, which requires that the continuity of the drinking water service be considered when water tariffs are set. The lack of earmarked funding for water and sanitation quality leads to limited progress in service delivery and heightens the financial imbalances of WSS SOEs. Customers in almost half of Brazilian states suffer frequent service interruptions (Figure 15). Reducing the frequency and duration of unplanned supply interruptions requires improving management and operational costs efficiency at water companies. It is essential to estimate costs and levels of efficiency, since, depending on the regulatory model applied, these costs will be transferred to the water tariffs and budget allocations. In monopoly structures of water and sewerage, customers have no option to switch to another supplier when the performance of their provider is poor. To avoid unplanned water supply interruptions, water companies need to increase their maintenance budgets or improve their investment programs’ quality. Brazil still lacks comprehensive penalties or rewards for water companies when the number, duration and/or frequency of their unplanned supply interruptions increases or decreases. Brazil’s financial rules and regulations usually focus on lowering expenditures more thanraising revenues for the goal of addressing budget imbalances. This has dramatically reduced the budget allocations to the WSS sector. In a ranking of 136 countries on public expenditure efficiency,26 the World Economic Forum (2018) placed Brazil in 133rd place, just above Zimbabwe, El Salvador, and Venezuela. The combination of low public spending efficiency and fiscal vulnerabilities exacted a toll on the federal WSS budget. Against the background of budget rigidities, rising mandatory spending, and falling revenues, discretionary public investment expenditure has proven to be the only remaining item where cuts can be made. Attempts at fiscal consolidation have almost always resulted in cuts to capital outlays, significantly reducing spending on WSS infrastructure. In Brazil, expenditures on electricity represent the main cost that water utility companies bear, after expenditures on labor (World Bank 2013). As such, water utility companies have a major incentive to enhance their energy efficiency. Moreover, the high volumes of water losses reduce companies’ revenues, and consequently, their ability to obtain financing and invest in improvements (World Bank 2013). 26 See for instance: WEF, 2018. http://reports.weforum.org/pdf/gci-2017-2018-scorecard/WEF_ GCI_2017_2018_Scorecard_EOSQ043.pdf 56 FIGURE 15 Benchmarks of WSS Service Quality by States, 2019 Between 4 and 6 days 100 Porto Alegre (RS) Northeast SP Belo Horizonte (MG) DF Florianópolis (SC) (b) For those with access to the network, how often is water available? Grande Vitória (ES) MG PR RJ RN Midwest Curitiba (PR) PR SE MA PIRA ES Goiânia (GO) SC (d) Access vs Supply by States Maceió (AL) MT AL TR 80 Fortaleza (CE) OP PR access to water network Metropolitan areas PE São Paulo (SP) South AM Between 1 and 3 days Aracaju (SE) MA Manaus (AM) João Pessoa (PB) North Belém (PA) How often is the supply available for those who have access to the network 60 Río de Janeiro (RJ) AC Natal (RN) AP Southeast Macapá (AP) PA Salvador (BA) Daily Grande São Luis (MA) RO Vale do Río Cuiabá(MT) Recife (PE) 40 100 20 40 0 60 80 100 60 40 80 Between 4 and 6 days Palmas (TO) Between 4 and 6 days Rio Grande do Sul Belo Horizonte (MG) Espírito Santo Porto Alegre (RS) Paraná (c) For those with access to the network, how often is water available? Vitória (ES) (a) For those with access to the network, how often is water available? Roraima Campo Grande (MS) Mato Grosso do Sul Fortaleza (CE) Santa Catarina Florianópolis (FC) Tocantins Río de Janeiro (RJ) Goiás Boa Vista (RR) São Paulo São Paulo (SP) Minas Gerais Goiânia (GO) Piauí Between 1 and 3 days Teresina (PI) Amazonas Between 1 and 3 days State Capitals Maceió (AL) Rondônia Aracaju (SE) Río de Janeiro States Curitiba (PR) Pará Macapá (AP) Amapá Mato Manaus (AM) GrossoCeará João Pessoa (PB) Alagoas Salvador (BA) Sergipe Belém (PT) Distrito Federal Natal (RN) Maranhão Bahia Brasilia (DF) Paraíba Daily Porto Velho (RO) Daily Río Grande do Norte Recife (PE) Acre Cuiabá (MT) Pernambuco São Luis (MA) Río Branco (AC) 20 0 0 60 40 80 20 40 100 60 80 Source: SNIS 2019 and PNAD 2019. 4. wss access and performance benchmarks 57 If Brazil were to undertake a nationwide effort to reduce water losses and enhance energy efficiency by water utilities, significant gains would result, which could facilitate more expenditures and investments in the sector. According to the World Bank (2013), the estimated potential gains from water loss reduction and energy efficiency enhancements under three scenarios, assuming a time horizon to 2025, could range between R$21 billion and R$37 billion (US$4 billion and US$7 billion). The WSS sector has high inefficiencies in using public expenditures as well as insufficient investments compared to other sectors. This results in large geographical variations in the quality of water and wastewater treatment services (Figure 16). For both water and sanitation, rates of lack of coverage have not changed much in recent years. That is, even though there are improvements in the indicators that attempt to measure access to these services, the number of households without access to water supply and sanitation has remained relatively stable. This shows that the public policies on basic sanitation have failed to keep up with the pace of the country’s growth, urbanization, and the building of precariously located settlements. In many municipalities the true state of WSS access and service quality is unknown. For instance, a substantial number of municipalities in the north and northeast regions have no information on wastewater treatment. This makes it difficult to identify solutions and allocate budgets. In addition, in many rural areas, there is no information about individual sanitation solutions, such as septic tanks and pit latrines. This lack of complete information on wastewater treatment coverage also risks environmental degradation: without knowledge of hotspots of spilled wastewater, it is impossible to assign appropriate yearly budgets to invest in wastewater treatment plants and infrastructure. FIGURE 16 Water Service Quality and Access Index 2018 (left) and Wastewater Treatment Index (m3 treated/consumed)2018 (right) (92, 100] (73 ,92] No data [51, 73] (90,100) [0,51] [50,90) No data [0,50) 0 Source: SNIS 2019. 58 Although Brazil has multiple operators, each of them has exclusive rights to a geographic area, which limits competition and investment within it. Easing entry, based on regulatedaccess fees, and allowing different franchisees to compete in the same area would encourage economies of scale and reduce costs to ensure better quality of services. Currently, drinking water supply services, sanitary sewerage, and drainage and rainwater management are responsibilities of multiple institutions. The institutions that share responsibilities are the Ministry of Regional Development’s National Secretariat of Sanitation (MDR-SNS), the Ministry of Health (MinSaude-FUNASA), and the regulatory bodies of the states and municipalities, with occasional service outsourcing to private companies in certain municipalities. With so many organizations having a role, the result is inefficient budgetary coordination. Fostering that coordination is key to solving complex issues of improving overall quality of services and reducing harm to the environment caused by limited access to these services. In Brazil, 70 percent of municipalities have no sewerage treatment plants, causing water pollution and health problems. Efficiency of STPs is low, and distribution system losses reach 37 percent on average in states that have CESBs. The sector also faces increasing challenges of water security due to climate change.27 WSS companies owned by states, operating under financial strain and with increasing demand for their product, are not obliged to reduce expenses and improve operating revenues to be able to be included in the public budget. State-owned companies linked to the WSS SOEs (defined as public companies and mixed-capital companies, subsidiaries,and controlled companies) operate with their budgets organized and monitored with participation of the Ministry of Economy (formerly MPO) but they are not assessed by the legislative process of budget preparation. A situation of poor financial performance with limited cash flows besets most WSS SOEs due to low and volatile operational revenues (Figure 17). This condition is holding back the potential for public expenditures and investments in the sector. Data from SNIS (2019) show that between 2005 and 2018 cash insufficiency was relatively high in practically all states, particularly in the North and Northeast. Also, the ratio of operational revenue to investments has consistently declined because of a marginal increase in operational revenues amid a sharp decline in investments due to high staff and current expenditures. The fluctuations of WSS investments across SOEs result from complex budgetary process,with limited incentives to improve financial performance. There are also regional disparities in WSS SOEs investments due to larger allocations to WSS budgets in states with higher SOE capitalization, larger populations, and better financial performance. The high percentage of investments in sanitary sewage—40 percent—is justified by this subsector’s generally low coverage, especially for sewage treatment infrastructure. 27 Climate change is increasing water variability. Severe droughts and floods alike are becoming more intense and frequent. From 2003 to 2016, about 47.5 percent of municipalities in the country faced floods. In the same period, 50 percent of municipalities faced droughts. In the Northeast, the drought problem was particularly high, with 78.5 percent of municipalities experiencing one. 4. wss access and performance benchmarks 59 FIGURE 17 Operational Revenues/Net Investments and Cash Flow Risks of State WSS SOEs Note: The ratio is reported x 100. Cash flow insufficiency index refers to the insufficiency of cashflows to cover operational costs of utilities, as defined by indicator IN101 of the SNIS. Source: SNIS 2019. The Southeast region had the largest investments in absolute value and higher operating revenues and investments, with total revenue of R$6.5 billion (US$1.2 billion) in 2018, equal to 55.5 percent of total national WSS investments per year. Of the value invested in the Southeast, SABESP accounted for R$3.88 billion (60.7 percent of the region’s total) and 33.7 percent of the total for the country. The lowest regional investment occurred in the Northern region, which captured just 3.5 percent of total investments over the last five years (SNIS 2019 and Banco de Nordeste 2018). Service performance of WSS companies and SOEs varies markedly and lacks a cohesivebudgetary policy that sets common rules to disincentivize poor performance. Some state regulatory agencies28 oversee services in some cities without proper municipal authorization. 28 The country has 50 regulatory agencies for water and sanitation services, consisting of 24 states agencies, one district agency, 22 municipal agencies, and three from municipal consortiums (micro- regional). They regulate in total 48.8 percent of all concessions of the country. Some are multi-sectorial (electricity, gas, transport) while others are dedicated exclusively to water and sanitation. Of the total, 27 regulate water and sanitation business corporations studied in this paper, of which seven are municipal/district and 21 are regional/micro-regional bodies regulating 26 utilities. 60 Many other cities served by WSS companies and SOEs have no regulatory agencies, primarily because they lack the financial viability required for creating and maintaining them. As a result, investments to reduce inefficiencies and increase performance are made at the discretion of state and municipal authorities. Variation of non-revenue water levels between states ranges from 25 percent to 60 percent for the WSS SOEs (Figure 18). Total water losses of 38 percent in Brazil translates into 6,530 million cubic meters of water per year, with an economic value of R$8.1 billion (US$1.5 billion). A 15 percent reduction in water losses would bring savings of R$3.8 billion (US$ 750 million).29 The economic value of water losses (US$1.5 billion) represents about 9 percent of Brazil’s government primary budget deficit (US$17 billion). To put these losses in international context, water losses globally add up to 126,000 million cubic meters per year, with a value of US$39 billion. The Latin America and Caribbean Region (LAC) loses about 25,400 million cubic meters per year, with a total economic value of US$8 billion. The regions of sub-Saharan Africa and Southeast Asia has less water losses than Brazil: 5,200 versus 6,530 million cubic meters per year30, respectively (Liemberger and Wyatt 2019). Non-revenue water and water loss reduction programs are effective interventions to improve financial performance and increase the returns of investments in the sector. Utilities no longer need to miss out on meeting revenue potential or incurring unnecessary operational expenses due to NRW. NRW and water loss reduction programs help identify the sources of NRW and curtail them. When implemented strategically, these solutions can be combined to create a system that will pay for itself, improve budget expenditures effectiveness, and produce cost savings into the future. Recurrent tariff increases are partly a result of inefficiencies in WSS SOE service delivery, and they amplify uncertainty among consumers, diminish affordability of services, and drive high dropout rates of household connections. The WSS sector has traditionally operated as a monopoly, with discretion for tariff revision exercised at the state levels. The lack of standardization and basic rules, including tariff revision criteria, indemnity, inspection, goals, and contractual obligations, heighten transaction costs and discourage the entry of new private players into the sector. Changes in tariffs are typically in response to insolvency of WSS SOEs limiting targeted use of tariff subsidies (Figure 19). 29 Instituto Trata Brasil 2013. 30 There are slight variations between sources: an approximation (using the NRW model at 1,000 m3 per day) results in an annual volume of 6,888 million cubic meters; while the Wyatt/IDB methodology results in 6,526 million cubic meters per year. 4. wss access and performance benchmarks 61 FIGURE 18 Non-Revenue Water % per Connection by State WSS Companies, 2012 – 2 017 (above) and Spending per Workerof State WSS companies, 2017 (below) 2012 2015 2017 100 100 100 80 80 80 60 60 60 40 40 40 20 20 20 NRW (%) NRW (%) NRW (%) 0 0 0 600 000 Average annual expense per employee (R$/employee). 500 000 400 000 300 000 200 000 100 000 0 utilities Source: SNIS 2019. 62 Because conditions in each state are quite different, the variation in tariff changes is large. The new WSS law provides for more standardization of basic rules for tariff changes and revisions. In sum, investments in infrastructure can bring wide benefits to other sectors. For households, especially those with low income, investments to upgrade transport, electricity, safe water and sanitation, and other basic facilities has a direct impact of creating better quality of life. But on overall infrastructure quality, Brazil ranks 116 out of138 countries in the latest World Economic Forum survey (WEF 2018). Brazilian businesses are suffering competitive disadvantages because low investment translates into high costs of basic public services, transport, and logistics. Making infrastructure more efficient and resilient could save the country economic costs in the future. Brazil remains seriously behind in WSS investment, a vital kind. According to the National WSS Plan (PLANSAB) 2014, to reach government targets of universalization for WSS service by 2033,31 the country will need to invest about R$26 billion (US$ 5 billion) per year (0.4 percent of GDP) for the next 14 years. But during the past two decades it has invested only R$12 billion (US$2.2 billion) per year. 31 The National Basic Sanitation Plan set targets to reach 99 percent water supply and 92 percentsewerage by the year 2033. 4. wss access and performance benchmarks 6 5. Allocationof Public Expenditures in the WSS Sector 64 This section analyzes recent budget and expenditure developments in Brazil in the basic sanitation32 sector. The section also looks at different levels of government (federal, state, and municipal) within the context of a broader analysis of growth and poverty reduction. The overarching task is to offer insights into how to best allocate resources, given the government’s current levels of budget execution and expenditure. The aim is to help the government map out plans to attain universalization of basic sanitation services. The federal government’s budget proposal for 2021, sent to the National Congress, provides for spending of just over R$694 billion (US$128 billion), of only 10 to 12 percent is for investment in basic sanitation works and services across the country.33 If confirmed, this will be the lowest investment value for the sector in the past five years. The budget text was sent by the federal government to the National Congress so that the parliamentarians could analyze the numbers and act on the values for the coming year. According to the government’s proposal, R$233 billion (US$43 billion) would go to the Ministry of Regional Development, which, among other things, is tasked with fulfilling investments related to development in the country’s regions and applying resources in such fields as sanitation as provided for in the Constitution. Another R$40 billion (US$7.4 billion) would go to the São Francisco and Parnaíba Valleys Development Company (Codevasf), a public company that focuses on the progress of the riverside regions of theSão Francisco and Parnaíba rivers and their tributaries, in the states of Minas Gerais, Bahia, Pernambuco, Alagoas, Federal District, Goiás, Sergipe, Piauí, and Maranhão. More than R$331 billion (US$61.3 billion) would be for the National Health Foundation (FUNASA), an agency of the Ministry of Health, which is currently the government institution responsible for promoting WSS solutions in rural areas. The main objective is to bolster WSSfor disease prevention and control, as well as to formulate and implement promotion actions,sanitation, and health protection related to actions established by the National EnvironmentalHealth Surveillance Subsystem. Finally, R$89 billion (US$16 billion) would be invested in risk and environmental disaster management (SIOP 2021 and SIGA34 2021). The Box 2 below provides an idea about the difficulties of working with Brazil’s Budget 32 In Brazil, basic sanitation includes water supply, sewerage, wastewater treatment, and solid waste andflood management. 33 An expenditure is not necessarily an investment. For example, payment of interest on government debtis not an investment, just a cost. The same is true for salaries for government employees who keep an office running. An investment is normally an expenditure that is expected to generate a return. Investments in WSS may also include private sector investments. 34 SIGA Brazil is a federal public budget information system that allows broad and easy access to data from the Integrated Financial Administration System-SIAFI and other databases on public plans and budgets. This access can be made by SIGA Brazil Panels and SIGA Brazil Reports. 65 BOX 2 WSS Budget Data Challenges Getting reliable data on WSS budget and expenditure on public accounts in Brazil faces multiple challenges. These are mostly due to structure—the administrative decentralization of the basic sanitation sector and the ensuing multiplicity of actors at the federal, state and municipality levels. Municipalities hold the ultimate responsibility to provide the service, but the sector historically comes with a service provision structure in which the states provided such services. In addition, the law allows for the service to be provided directly by the municipality, or through a concession which could be in practice be carried out by a state utility. Therefore, basic sanitation service schemes are diverse and heterogeneous across the country. This, in turn affects the collection, integration, standardization, consistency, and ultimately the quality of budget and expenditure data at the different levels of government. For example, expenditure data at the state and municipal level are self-reported. This means that data are not collected in a standardized manner and there are no ways to effectively verify them. Regulatory agencies for the sector can be state, municipal, or intermunicipal agencies, with each one having its own procedures for reporting, monitoring, and evaluation. Service providers report their finances and indicators based on specific state or local requirements and not on standardized national ones. Expenses are classified into two economic categories: current and capital. Current expenses guarantee the maintenance of the functioning of public services and are not directly linked to the formation or acquisition of a capital asset. Capital expenditures contribute to increasing the entity’s equity. Current expenses do not have limits under the regulatory framework, and they have been increasing over time. Some states and municipalities report different disaggregation levels of expenditures, while some others do not. Some others report how much they spend on the rural or urban sector, while others do not. Despite laudable efforts by national agencies and departments in charge of national accounting, budgeting and statistics, this framework continues to undermine public oversight and budget analysis of the basic sanitation sector. and Expenditure data, particularly at sub national level. The sources of information vary substantially in terms of quality and consistency depending on the agency that report them. Also, municipalities need obtaining support to systematize the reporting of budgetary data within the multiple information systems that share information with the public about the total amounts and changes in budgetary allocations, implementation, and efficiency. I. Budget Allocation and Public Expenditurein Basic Sanitation at the Federal Level The WSS budget allocation in Brazil accounts for a smaller share of GDP than comparable sectors such as basic education and primary health care. In fact, the share of basic sanitation budget has been decreasing over time while other sectors have proportionately increased (basic education) or remained reasonably stable (primary healthcare) (Figure 20). Brazil’s WSS sector has had high budget under-execution in the past decade. Budget execution rates reveal high levels of inefficiency in spending in WSS, further constraining thesector (Figure 21). Capacity limitations at different administrative levels (but mainly at the municipal) may be among the critical factors slowing progress of physical implementation and consequently budget utilization. Some factors would enhance budget execution if improved include (1) operative and management capacity of service providers, (2) transparency in the allocation of resources, and (3) participation of the central/state government in assisting less-developed service providers. 66 FIGURE 20 Federal Government Budget Allocation as Percentage of GDP, Selected Sectors (2004-2017) 2.00 1.50 % GDP 1.00 0.50 0.00 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Basic sanitation Basic education Primary health care Note: Basic Education includes obligatory transfers. Note: As comparators we have used basic sanitation and primary health care. Not all education falls under the responsibility of the municipalities – just basic education (primary level). High school level is the responsibility of the state. For health, primary care is the responsibility of municipality and secondary care of the state. Thus, since high school education and secondary care fall under the state’s responsibilities, expenditures on these are excluded from the comparison. Source: BOOST, World Bank, IBGE 2004 – 2 017. FIGURE 21 Budget Under-Execution in Brazil, Real Values Adjusted with IPCA Basic Sanitation Budgets 2000-2020 Federal Spending Program of Basic Sanitation, 2015-2019 Real Terms (in R$ million) Real Terms (in R$ million) 5000 2000 4500 Federal Government Spending-Program 4000 1,10 1500 Millions (BRL) 3500 3000 959 743 2500 1000 1,16 706 2000 1500 500 964 1000 668 681 557 500 350 0 0 2011 2017 2012 2013 2015 2014 2019 2016 2018 2001 2010 2007 2002 2003 2005 2020 2004 2006 2009 2008 2000 2015 2016 2017 2018 2019 Committed Budget real terms (million) Disbursed Committed but not disbursed Executed Budget real terms (million) Note: Real values adjusted with IPCA Source: SNIS 2000-2020. Source: SIOPS 2015-2019. 5. allocation of public expenditures in the wss 67 FIGURE 22 Ratios of Executed and Committed Budgets 4 3.5 3 2.5 2 1.5 1 0.5 0 2010 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 Ratio of Executed vs Committed Budget (inflation adjusted values)Ratio of Executed vs Committed Budget in constant terms (2015=100) Note: The deflator used to calculate inflation adjusted values is BC/IBGE’s GDP Implicit Price deflator. Source: SIOPS 2020 and TCU 2020. At the federal level, the budget process is uncertain and budget efficiency is very volatile. Budgets are uncertain given the fiscal and macro vulnerabilities during several periods between 2000 and 2020 (Figure 22). Moreover, budgets may be executed at different levels of efficiency because there are federal projects that increase the budget execution. The WSS sector has low efficiency of about 33 percent of executed federal budgets compared to appropriated budgets (SIOP, 2019).35 These vulnerabilities of budget execution efficiency create illiquidity, since taxes are subject to revenue sharing and tax breaks awarded by the federal government reduce the budget transfers that subnational governments receive. On the other hand, solvency risks in subnational government, such as the fiscal crisis in Rio de Janeiro in 2016, can present contingent liabilities to the federal government, because states might default on their debts to the federal government or receive bailouts. There is no consistency of budget allocations and efficiency parameters required in the sector, and budget cycles are highly sensitive to political discretion and fiscal constraints.As Figure 22 shows, even when adjusting for inflation, only after 2015 is there an increment in budget efficiency (larger proportion of executed versus committed budget). 35 The data considered are those of the approved budgets established annually in the Annual Budget Law (LOA) and those of financial execution over the fiscal years 2010 to 2019, with the analysis focusing on the period between 2012 and 2019, to guarantee the comparability of information. The source of the information is the Federal Budget Panel, of the Integrated Planning and Budget System (SIOP), whose data refer to the base of the Federal Government’s Integrated Financial Administration System (SIAFI) (SIOP, 2019). 68 FIGURE 23 Consolidated National Public Expenditures (2002-2017) as Percent of GDP 04- Administration 1.7 2.05 08- Social assistance 0.87 1.6 06- Public security 1.11 1.35 11- Job 1.1 1. 02- Judiciary 2 1.2 05- National defense 1.09 3 15- Urbanism 0.83 1.04 26- Transport 0.85 1.9 01- Legislative 0.53 0.64 20- Agriculture 0.49 0.55 03- Essential to Justice 0.21 0.38 17-Sanitation 0.25 0.39 14- Citizenship Rights 0.09 0.21 18-Environmental management 0.2 0.23 19-Science and technology 0.15 0.17 13- Culture 0.13 0.16 23- Trade and Services 0.1 0.24 27- Sport and Leisure 0.08 0.08 16- Housing 0.07 0.1 25- Energy 0.05 0.56 22- Industry 0.05 0.14 21- Agrarian organization 0.05 0.1 24- Communications 0.04 0.25 07- Foreign Relations 0.04 0.12 0.0 0.5 1.0 1.5 2.0 Consolidated national public expenditures (2002-2017) % GDP Note: Sanitation includes water supply, wastewater and sanitation, and some pluvial infrastructure. Source: SIOP and BOOST 2002-2017. Budgetary allocations to the sector have declined by a higher proportion than the decline in executions. For that reason, the ratio of executed against committed budgets has increased since 2007, in constant terms, but in real terms the ratio stayed relatively steady. Macro and fiscal challenges have reduced overall public funding in many development sectors of the economy. Budget uncertainty and fluctuations complicate planning of thesector. Moreover, many provisions do not allow a programmatic approach to steadily increase budgets to meet universalization goals. Figure 23 shows that the basic sanitation sector has suffered a declining share of public expenditure as part of national GDP over time (2002 to 2017), going from 0.39 percent of GDP in 2002 to 0.25 percent in 2017. Moreover, of the more than twenty functional sectors examined, only six had increases in their budgets as a share of GDP during the 2002-17 period. These were Social Assistance, Public Security, Urbanism, Expenses Essential to Justice, Citizenship Rights, and Science and Technology. Figure 24 is a different visualization of Figure 23 to make clearer the functional sectors’ trends over the period studied. Social Assistance showed the steepest increase, going from 0.87 percent of GDP in 2002 to 1.61 percent in 2017. The remaining fourteen functional sectors, including WSS, registered decreased shares of GDP over the last two decades. Basic sanitation went from 0.39 percent of GDP in 2002 to 0.25 percent in 2017. 5. allocation of public expenditures in the wss 69 FIGURE 24 Percentage of GDP of Programmed Expenditure to Functional Sectors (2002-2017) 2.0% Administration (1.7%) Social Assistance (1.61%) 1.5% Public Safety (1.35%) Labor (1.12%) Judiciary (1.09%) 1.0% Defense (1.04%) Urbanism (0.87%) Transport (0.85%) Legislative (0.53%) 0.5% Agriculture (0.49%) Justice (0.38%) Basic Sanitation (0.25%) Energy (0.05%) 0% 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 Source: SIOPS 2019 and TCU 2019. A point of comparison is the Energy sector (not including oil), which shows a sharp decline in 2004 (from 0.56 percent of GDP in 2002 to 0.05 percent in 2017). This was due mainly to reforms that the government carried out in the sector. These introduced a model that aimed to attract long- term private investment to the sector and relied heavilyon competition, while also addressing certain existing market imperfections. Urban areas receive a larger share of federal budget for WSS than rural areas. This divide was particularly large between 2007-2013, mostly because of the Growth Acceleration Program (PAC– Programa de Aceleração de Crescimento) and the Water Law of 2007. In recent years, federal budget flows to urban areas have dramatically decreased, while the share for rural areas has remained stable but very low (Figure 25). Federal expenditure on WSS is mainly driven by capital expenditure, but it is insufficient and with critically low levels over the last 8 years. Capital expenditure has fluctuated over time as a result of different government programs, policies, and legislation. On the other hand, recurrent expenditure (OPEX) on basic sanitation from the federal government has been low but stable over time (Figure 26). Federal and subnational fiscal policy interact in important ways. Federal fiscal policy affects subnational governments’ finances, which in turn constitute an important fiscal risk for the federal government. Federal rules on pensions, health, WSS, energy, and education dictate much of subnational spending. 70 FIGURE 25 Distribution of Federal Budget for Urban and Rural Areas 8,000 6,000 Verified (millions) 4,000 2,000 0 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 rural urban total Note: All values are in 2018 reais. Despesas líquidas has been translated as verified budget, which is the variable used here to account for expenditure. Source: BOOST and World Bank 2004-2018. FIGURE 26 Federal Expenditure on Basic Sanitation (Capital and Operational Expenditure) 8,000 6,000 Verified (millions) 4,000 2,000 0 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 recurrent expenditure (OPEX) capital expenditure (CAPEX) Note: Note: All values in 2018 Reais. Despesas liquidas has been translated as verified budget, which is the variable used here to account for expenditure. Source: BOOST and World Bank 2004-18. 5. allocation of public expenditures in the wss 7 6.WSS Subnational Planning, Budgeting, and Spending 72 Understanding the dynamics of local (state and municipal) budget allocationand public expenditure offers insights into how to allocate resources in the basic sanitation sector. It is important to underscore, as mentioned on Box 2, that local- level data should be taken with caution, given that state and municipal expenditures are self-reported, and the form of reporting—far from being standardized nationally—is determined by state or municipal authorities. This makes the data difficult to analyze.36 However, given that the framework in the basic sanitation sector gives pre-eminence to the local level—particularly the municipal level—there is no choice but to try to make the most of this data despite its restrictions and challenges.37 I. State Level WSS Spending38 The states’ institutional organization is generally aligned with the federal one in WSS, though some responsibilities are entirely with the federal government. Some topics are regulated bythe Union (Brazil’s civil, commercial, penal, electoral, agricultural, maritime, aeronautical, and labor codes of law), as well as some economic activities, including WSS, energy, postal services, and all types of procurement at the federal, state, and municipal levels. Budget typically reflects the size of a state39 (in population and Gross Regional Product—GRP), but does not necessarily reward states that are the most efficient in terms of budget execution. Thus, the states of São Paulo and Minas Gerais, as the two most populated states in Brazil, get the largest shares of budget. Likewise, Roraima, Acre,and Amapá states, as some of the least populated states, receive smaller shares of the national budget (Figure 27). São Paulo, Rio de Janeiro, and Minas Gerais are the states with the biggest GRPs in the country, while Roraima, Acre, and Amapá have the smallest GRPs, status that influences their budget allocations. 36 One of the results of these limitations with data at state and municipal level is that we cannot analyze rural and urban expenditure (the data are misleading), and capital and recurrent expenditure (the datado not exist) at these levels of government. 37 This document uses despesas liquidadas to account for expenditure following a recommendation from the Public Accounting Subsecretary, Under the Federal Treasury. The group cites the so-called princípio de competência/regime de competência (accrual basis/accrual regime) by which expenses are considered at the (closest) time thatthey occur and not when they are paid. If despesas pagas is used, there is a risk of underestimating thegovernment’s expenditure commitments. 38 Budget and expenditure data for basic sanitation at the state level must be taken with prudence giventhat the structure of the service provision (whose ultimate responsibility rests with the municipalities and not with the states) as well as the reporting of data, which is not standardized, may create distortions that can affect the analysis of the sector. 39 The states receiving the larger proportions of WSS budgets also have higher WSS coverage. 73 FIGURE 27 Population and WSS Budget Allocation at the State Level Metropolitan areas of Brazil, 2020 Water supply and sanitation budget allocation by state (US$, 000, 2017) RS GO RJ SC ES CE MG +20m PA PB BA SE Metropolitan Area Core +3m DF PI TO +1m MT MA RR R SP PR PE MS RN AM A... A... A... AC AL AM AP BA CE DF ES GO MA MG MS MT PA PB PE PI PR RJ RN Note: Green areas in the map signify areas of more than 1 million inhabitants. Source: IBGE 2020 and TCU, SNIS 2018 FIGURE 28 Budget Execution of WSS at the State Level (2013-2019) 6,000 4,000 Millions (BRL) 2,000 0 2013 2014 2015 2016 201 2018 2019 Committed Paid Note: Constant 2013 prices. Source: DCA/Finbra 2013- 74 Budget execution of WSS across states has been more efficient over the last few years, but total budget/spending in the sector at the state level has fallen drastically. Figure 28 shows that budget for WSS at the state level is spent more efficiently than at the federal level. At the state level, executed or paid budget was well over 75 percent of committed budget between 2013 and 2019. However, budget size for WSS at the state level decreased drastically over this period. Between 2013 and 2019, budget for WSS (committed budget) went from R$6,000 million to just over R$2,000 million. Analysis of state-level data shows that WSS expenditure varies widely among states. Figure 29 shows total expenditure in WSS by state for 2019, in total or aggregate/absolute amounts and disaggregated by expenditure components.40 Goiás and Santa Cataria states spend the smallest amounts in the sector, while São Paulo and Bahia states spend the largest.41 However, expenditure data may be unreliable because they are not standardized among municipalities and states. Hence, the graph that shows disaggregation will show a mixture of urban, rural, administrative, and other expenses. Another factor that makes it difficult toaccount for WSS expenses at the state level is the framework in the provision of service. The current framework gives responsibility for provision of WSS service to the municipality but allows for different arrangements in which state-level providers take part as well as private bodies. There are around 5,570 municipalities in Brazil and by law each one of them must determine the best or most feasible way to provide WSS services to its population. Therefore, Brazil has a large range of options and possibilities in the provision of service. Per capita WSS spending for 2019 also shows wide differences between states, but the order of states changes in relation to total or absolute expenditure amounts. Figure 30 presents per capita expenditure in WSS by state for 2019, in total or aggregate amounts (left) and disaggregated by expenditure components (right). As in the total or absolute values for WSS expenditure for 2019 (Figure 29), the per capita values for 2019 show wide differences between states. However, the order of states in terms of per capita expenditure is different than when ranked by total or absolute expenditure. In other words, states that spend more on WSS in absolute terms are not necessarily the same states that spend more in per capita terms.42 40 While existing data show rural and urban classifications, it is uncertain whether these data are correct,as there may be deficiencies in the declaration of data, which at this level of government is self- reported. Therefore, analysis and conclusions about state levels of expenditure for rural and urban areas will not be done here. 41 The second graph that shows state-level expenditure disaggregated by components should be taken with caution, given that expenditure data at the state and municipality levels are self-reported and the form of reporting is determined by state or municipal authorities. 42 There seem to be other factors which may explain differences in states spending. These may concern not only population size, but structure of the service provision (SOE vs private, state vs municipal, etc.),efficiency in spending, managerial/administrative capacity, and even factors such as geography. 6. wss subnational planning, budgeting, and spending 75 FIGURE 29 Total Expenditure in Basic Sanitation, by State (2019) (In Million R$) Expenditure in basic sanitation at state level, Expenditure in basic sanitation at state level, aggregated (2019) disaggregated in components (2019) Goiás 0.11 Goiás Santa Catarina 0.39 Santa Catarina Rondônia 3.58 Spending in Basic Rondônia Mato Grosso 4.13 Sanitation - Mato Grosso Alagoas 5.76 2019 Alagoas Maranhão 8.48 Maranhão Distrito Federal 10.30 Distrito Federal Amapá 14.44 Amapá Minas Gerais 14.89 Minas Gerais Roraima 23.11 Roraima Piauí 23.86 Piauí Mato Grosso do Sul 23.97 Mato Grosso do Sul Tocantins 28.13 Tocantins Pará 28.64 Pará Paraíba 35.19 Paraíba Espírito Santo 42.27 Espírito Santo Rio Grande do Sul 53.33 Rio Grande do Sul Amazonas 62.18 Amazonas Sergipe 94.32 Sergipe Río de Janeiro 101.63 Río de Janeiro Acre 114.25 Acre Pernambuco 193.88 Pernambuco Ceará 231.08 Ceará Bahia 359.35 Bahia São Paulo 494.17 São Paulo 0 50 100 150 200 250 300 350 400 450 500 0 50 100 150 200 250 300 350 400 450 500 Millions (BRL) Millions (BRL) rural urban administrative other Note: There are no data for Paraná and Rio Grande do Norte for 2019. Source: DCA/Finbra 2019. For example, São Paulo is the state that spent the most in 2019 in total or absolute expenditure on WSS, R$493 million (Figure 29), but in per capita expenditure, it ranked only tenth, at R$11 per capita (Figure 30). Acre is the fifth state in terms of absolute/ total expenditure in the sector (R$114 million in 2019), but in per capita terms, it ranked number one (R$131 per capita). Sergipe state is seventh in terms of total WSS expenditure (R$94 million in 2019) but was second (R$41 per capita) in per capita expenditure. As is true in total WSS expenditure, per capita expenditures also show enormous differences between states. An extreme example is Santa Catarina, which spent R$0.06 per capita on WSS in 2019, while Sergipe spent R$41. In other words, Sergipe spent 683 times what Santa Catarina spent. The enormous differences come because of political will, municipal capacity, and technical designs of projects. Goiás, Paraná, Rio Grande do Norte, and Santa Catarina, meanwhile, spent R$1 or less per capita on averageduring the 2013-19 period. These differences are due to limited planning and political priorities in the budget allocations to the sectors, without a standard norm. 76 FIGURE 30 Per Capita Expenditure in Basic Sanitation, by State in R$ (2019) Per Capita Expenditure in Basic Sanitation Per Capita Expenditure in Basic Sanitation at State Level, Aggregated (2019) at State Level, Disaggregated in Components (2019) Goiás 0.0 Goiás Santa Catarina 0.1 Santa Catarina Minas Gerais 0.7 Minas Gerais Mato Grosso 1.2 Mato Grosso Maranhão 1.2 Maranhão Alagoas 1.7 Alagoas Rondônia 2.0 Rondônia Pará 3.4 Pará Distrito Federal 3.5 Distrito Federal Rio Rio Grande do Sul 4.7 Grande do SulRío Río de Janeiro 5.9 de Janeiro Piauí 7.3 Piauí Mato Grosso do Sul 8.7 Mato Grosso do Sul Paraíba 8.8 Paraíba Espírito Santo 10.6 Espírito Santo São Paulo 10.8 São Paulo Amazonas 15.2 Amazonas Amapá 17.4 Amapá Tocantins 18.1 Tocantins Pernambuco 20.4 Pernambuco Bahia 24.3 Bahia Ceará 25.5 Ceará Roraima 40.1 Roraima Sergipe 41.4 Sergipe Acre 131.4 Acre 0 25 50 75 100 125 0 25 50 75 100 125 BRL BRL rural urban administrative other Note: There are no data for Paraná and Rio Grande do Norte for 2019. Source: DCA/Finbra 2013 – 2 019. Across time, there have been significant drops in WSS expenditure in individual states in recent years. Figure 31 shows WSS expenditure for all states in 2013 and in 2019. Only four out of the 27 states of Brazil spent more on WSS in 2019 than in 2013 (Sergipe, Rio Grande do Sul, Tocantins, and Distrito Federal).43 The remaining 23 states all recorded a decline in WSS expenditure in the 2013-2019 period (Figure 31).44 The most significant drops occurred in Maranhão, Alagoas, and Espírito Santo states. There the 2019 figures had fallen more than 80 percent over 2013’s levels. 43 This study refers to the federative units of Brazil as states. Strictly speaking, the Federal District is nota state, but we refer to it as one to allow practical comparisons. 44 There are no data for Paraná and Rio Grande do Norte states for 2019, but it is likely that their patternfollowed that of most of the states, that is, a decrease in basic sanitation expenditure for 2019. 6. wss subnational planning, budgeting, and spending 77 FIGURE 31 WSS Expenditure by State for 2013 and 2019 800 600 Millions (BRL) 400 200 0 Paraná Rio Grande do Norte Mato Grosso Alagoas Amazonas Pernambuco Santa Catarina Minas Gerais Amapá Maranhão Distrito Federal Piauí Mato Grosso do Sul Sergipe Acre São Paulo Espírito Santo Goiás Pará Paraíba Rondônia Roraima Ceará Bahia Tocantins Rio Grande do Sul Río de Janeiro 2013 2019 Source: DCA/Finbra 2013-2019. II. Regional and Local Spending Constitutionally required transfers from the federal government to states and municipalities, as well as earmark budgets for special funds, greatly reduce the federal government’s options for adjustments of its resources. Ordinary taxes (impostos) collected by the federal government are subject to revenue sharing with states and municipalities, and these transfers involve the WSS sector. To raise revenues for exclusive federal use, the federal government has introduced several “contributions” (contribuições,de-facto taxes). These are devoted to specific purposes, such as social security, WSS, and transport infrastructure. To gain more budgetary flexibility, the government in 1994 introduced a delinking of federal revenues (Desvinculação de Receitas da União, DRU) which frees up 20 percent of revenue from contributions (as well as some smaller revenues) for general budget use. However, states and municipalities have not fully implemented the DRU in revenue contributions for WSS SOEs. Introducing a limit on spending growth and reorienting WSS budget allocations to stateswith lowest coverage would serve to delink budget cycles from revenue performance or economic activity. It could also reduce macroeconomic uncertainty since the maximum level of federal expenditures (adjusted for inflation) will be predetermined, increasing the predictability 78 of fiscal policy as well. The implementation of those expenditure rules would benefit multi-year budgeting and reduce regional inequities and inefficiencies inspending (Figure 32). The current system of revenue earmarking and indexation is a relic from a hyperinflationary past. It was an effort by the legislative branch to establish spending priorities beyond the annual budget cycle and protect them from executive discretion inbudgetary execution. The target of WSS universalization would benefit from spending priorities that extend beyond the annual budget cycle. In rural areas, transfers to municipalities have declined overall in terms of efficiency, with marginal variations in the ratios of executed against committed transfers. In the northern regions in particular, variations of efficiency are high, and levels of efficiency arelow. Reporting in rural areas is generally inadequate. The North, South, Southeast, and Northeast regions have relatively higher efficiency in urban areas compared to rural ones, whereas the Central-West region has higher efficiency in rural areas. FIGURE 32 Ratio of Transfers to Municipalities Executed Payments against Commitments for Functional Classification No. 17 (Saneamento Básico) 2005-2018 2015-2018 1.0 1.0 Ratio of transfers to municipalities-executed Ratio of transfers to municipalities-executed 0.8 0.8 against committed against committed 0.6 0.6 0.4 0.4 0.2 0.2 0 0 CO DF NE NO NI* SD SL DF CO Ni* NE NO SD SL Note: The ratio measures the level of efficiency of expenditures of transfers to municipalities. The higher theratio the better the efficiency. NI refers to non-identified expenditures under the transfers. The regions of Brazil are CO (Central-West), NE (Northeast), NO (North), SD (Southeast), SL (South), and DF (Federal District). Source: Based on data from OFSS Gasto Social e Despesas Públicas, Brazil 2019. 6. wss subnational planning, budgeting, and spending 79 Nevertheless, comparing efficiency within urban and rural areasfor the period of 2005 to 2017/2018 shows that overall efficiency declined in heterogeneity but there were only marginal changes in the Southeast and Northeast regions (Figure 33 and Figure 34). The North region experienced a decline of efficiency between 2005 and 2017 in rural areas and a marginal increase in efficiency for urban areas between 2005 and 2018. Making WSS expenditures more efficient is vital, given the importance of these basicservices in improving human capital and enhancing incomes. Brazil has an average of per capita income of US$12,000, yet Brazilian regions have startling inequalities. The North region, for example, has average per capita income of only US$7,800 which is comparable to Guyana. The Northeast region’s figure is US$5,900, like the level in Lesotho. At the other end of the scale, the Southeast and South regions have average per capita incomes of US$13,500 and $15,400, respectively, comparable to countries such as Bulgaria and Malaysia. Improving the efficiency of spending will contribute to reduction of regional and local inequality of income and socioeconomic status. FIGURE 33 Ratio of Transfers to Municipalities Executed Payments against Commitments for Functional Classification No. 17 (Saneamento Básico) in Rural Areas** 2005-2018 2015-2017 1.0 1.0 Ratio of transfers to municipalities-executed Ratio of transfers to municipalities-executed 0.8 0.8 against commitments against commitments 0.6 0.6 0.4 0.4 0.2 0.2 0 0 CO DF NE NO SD SL CO* NO NE SD SL *CO region includes Distrito Federal (DF). **Includes direct allocations to rural areas under Saneamiento Básico functional classification, plus transfers to municipalities for that same classification. Data for 2018 for rural areas are incomplete and thereforenot included in the graph. The regions of Brazil are CO (Central-West), NE (Northeast), NO (North), SD (Southeast), SL (South), and DF (Federal District). Source: Based on data from OFSS Gasto Social e Despesas Públicas, Brazil 2019. 80 FIGURE 34 Ratio of Transfers to Municipalities Executed Payments against Commitments for Functional Classification No. 17 (Saneamento Básico) in Urban Areas* 2005-2018 2015-2018 1.0 1.0 Ratio of transfers to municipalities executed Ratio of transfers to municipalities executed 0.8 0.8 against commitments against commitments 0.6 0.6 0.4 0.4 0.2 0.2 0 0 CO DF NE NO NI* SD SL CO* DF NE NO SD SL Note: * The NI identifications are partial, missing values from 2015 onwards, and are therefore not includedin the graph because averages will be biased against totals and rural areas. The regions of Brazil are CO (Central-West), NE (Northeast), NO (North), SD (Southeast), SL (South), and DF (Federal District). Source: Based on data from OFSS Gasto Social e Despesas Públicas, Brazil 2019. Data show slow change and high variation of efficiency scores, yet some areas showed increasing efficiency over the analyzed period (2005-2018). Regions that recorded a decline in efficiency were the Northeast, Midwest, and Distrito Federal. Companies in the Southeast of the country present, on average, the highest efficiency scores, while those inthe North region show the lowest. Rigidities in the budget allocation process for WSS and other functional classifications are jeopardizing the achievement of targets in the COVID-19 era. Many areas of the country remain far from the goals. For instance, ANA (2019) showed that 31 percent of thecountry’s people live in areas that have low water assurance (this means they are facing rationing, network deficiencies, or alert in periods of drought, making it necessary to seek new water sources). Forty-one percent live in areas whose production system requires expansion. Only 27 percent of Brazil’s population lives in municipal areas whose WSS services have been rated satisfactory. Seventy-eight percent of people are predominantly supplied by surface water, while for 22 percent it is mostly groundwater. 6. wss subnational planning, budgeting, and spending 8 FIGURE 35 Efficiency of Spending by Broad Functional Categories: Ratio of Executed/Appropriated Budgets to States and Municipalities, 2005-2018 Note: The dotted line represents the average ratio for all categories, 2005-2018. Source: Based on data from OFSS Gasto Social e Despesas Públicas, Brazil 2019. Spending efficiency of federal transfers to municipalities became less volatile across local governments between 2005 and 2018, but there was no progress on overall efficiency improvements. This happened in part because (1) the private sector took overmany of the functions and investments in the sector, putting municipalities in charge of complex projects that were difficult to implement with the current budgetary arrangements, and (2) planning and financial technical capacities of municipalities have not improved over the years to allow them to execute spending of transfers more efficiently. Expenditures for the WSS have low federal allocations and efficiency. Budgets are a frequent object of analysis in Brazil sixteen years after the Fiscal Responsibility Law (FRL) came into force. The nation is facing a serious threat to the fiscal sustainability of its governments. Efficiency of the WSS sector compared to other sectors in the budgetary functional classification is one of the lowest (Figure 35). However, an accounting mechanism of “carry-over” or “deficit carry-over” improves budget balances and obtains incremental fiscal appropriation of government funds. Increasing the efficiency of the execution of public WSS expenditures is a necessary condition effective implementation of the WSS law. 82 III. WSS Plans and Spending Efficiency Municipalities are the stewards of sanitation services and can provide them directly orthrough concession. These governments are tasked to organize, regulate, and superviseprovision of these services, as well as to elaborate Municipal WSS Plans (PMSBs)45. Municipalities are holder of the four functions essential for the coordination of sanitation services: provision of service, regulation, inspection, and planning. However, in metropolitan regions and microregions, states share these regulatory and institutional functions with municipalities. In addition to regulatory and financing restrictions, the evolution of the sector depends on the efficiency of the supplying companies. In terms of financing restrictions and limits on WSS investments in Brazil, Kresch (2020) found that ambiguity in intragovernmental relations reduces investment and, hence, the quality of WSS service.46 The study found that a legal reform in 2005 that reduced the risk of appropriation between the various levels of government strongly influenced investment in this sector post- reform: municipalities with self-run water and sanitation companies nearly doubled their investment. Furthermore, this increase in investment led to an increase in access to the WSS system that resulted in a 7 percent decrease in under-one-year-old child mortality compared to before the reform. Institutional and regulatory reforms like these can have large impacts on welfare. Taxpayer resources can be more effectively used in public investment if focused on areas that the private sector cannot finance, such as expansion and distribution of services. Reducing the government’s role in financing of the WSS sector would imply supporting municipal development plans that consolidate budgets and transfers, and reforming tariffs to link them to performance. Consolidating budgets of the WSS sector in a single investment facility would increase the efficiency of expenses, especially in three areas: maintenance and recovery, strategic bottlenecks in infrastructure segments, and expansion of basic services. This could most benefit the low- income population and expand capacity-building programs to prioritize infrastructure projects in zones that hada single metropolitan area and several municipalities. Furthermore, state policies and municipal plans would help mobilize more private financing for infrastructure, tapping national and international sources alike. 45 This basically means that of the 4 functions, Planning is responsibility of the Municipality and cannot be delegated. The other 3 (provision of service, regulation, and inspection) can be somehow delegated through concessions or other arrangements. Given that the PMSB implies the planning of the WSS themes, we say that planning is embedded in the PMSB. 46 The study used an administrative panel dataset on the Brazilian water and sanitation sector, the Sistema Nacional de Informações sobre Saneamento (SNIS) dataset. 6. wss subnational planning, budgeting, and spending 83 FIGURE 36 Percent of Municipalities according to the Preparation of Water and Sanitation Municipal Plans according toMunicipal Population Size and Region501 to 10,000 10,001 to 20,000 20,001 to 50,000 50,001 to 100,000 100,001 to 500,000 More than 500,000 501 to 10,000 10,001 to 20,000 20,001 to 50,000 50,001 to 100,000 100,001 to 500,000 Less than 500 Less than 500 North North South South Brazil Brazil Northeast Northeast Central-West Central-West Southeast Southeast More than 500,000 2011 2017 Regulated Plan Plan not regulated Plan to be prepared Not plan at all Note: Each horizontal gray line represents 20 percent, cumulative up to 100 percent. Source: IBGE 2018, SIOP 2019, and SNIS 2018. More than half of the municipalities in Brazil do not have or are in the process of elaborating their Municipal WSS Plans (PMSBs).47 Figure 36 shows the percentage of municipalities that do not have a plan (sem plano), with a plan in preparation (com plano em elaboração), with a plan but without regulation (com plano sem regulamentação), and with a regulated plan (com plano regulamentado). The classification is done by the population size of municipalities and by geographical regions. The graph shows a comparative between 2011 and 2017. Figure 36 shows that in 2017, only 2,313 municipalities (41.5 percent of the total) had a PMSB (either regulated or not). This percentage was 10.9 percent in 2011. The South Region registered the most significant increase between 2011 (13.5 percent) and 2017 (72.9 percent). Moreover, in 2017, the proportion of municipalities with plans (either regulated or not) was quite unequal in regional terms, varying from 15.7 percent in the Northeast to 72.9 percent in the South. In 2017, the states with the highest percentages of municipalities with PMSB were Santa Catarina (87.1 percent) and Rio Grande do Sul (75.5 percent), whereas the states with the lowest percentages were Paraíba (13.0 percent), Pernambuco (14.1 percent) and Bahia (14.6 percent). 47 The national guidelines for sanitation in Brazil require all municipalities to develop and implement a Municipal Basic Sanitation Plan (PMSB) and a Municipal Basic Sanitation Policy. PMSBs are important because they create the municipal framework to manage and develop the basic sanitation services, as well as determining how budgets will be executed. In addition, PMSBs present an opportunity to join technical decisions and economical and social viability on sanitation services. However, for many municipalities, these plans are not fully developed or are non-existent. 84 FIGURE 37 Registry of UVUC in Municipalities as proportion to their revenues 2013-2018 Registration of unverified unpaid commitments in relation to the revenues (average of municipalities) 201 2014 2015 2018 3 7% (n=4005) (n=4270) (n=3964) (n=4180) 6% 5% 4% 3% 2% 1% 0 <20 <50 <100<500 >500 <5 <10 <20 <50 <100<500 >500 <5 <10 <20 <50 <100<500 >500 <5 <10 <20 <50 <100 <500 >500 Population (per 1000 inhabitants) Other exp. Personnel exp. (including taxes and contributions) Investments Note: Amounts are reported with inflation adjustments using the IPCA. Revenues are composed by taxes, tariffs, and transfers (3t). Municipalities listed in each year are those that sent data to STN in that year andregistered at the Uvuc. Data from 2015 was collected from Siconfi and SDI/ME. Source: Based on FGV (2017) and own estimates. Of the 139 municipalities of Tocantins, only five reported to have a plan in 2011, and 62 in 2017 (IBGE 2018). Municipal WSS Policies are less common in the least populous municipalities. Among the 42 municipalities with more than 500,000 inhabitants in 2017, 69.0 percent (29 municipalities) declared in 2017 that they had such a policy. In 2011 just 22 municipalities did (Figure 37). Another six stated in 2017 that they were drafting such a policy. In municipalities with up to 5,000 inhabitants, the number reporting they had such a policy increased by 72 percent (from 277 to 477), while in those with 5,001 to 10,000 inhabitants, the rise was 62 percent (from 269 to 435). Brazil received a favorable position in the Public Expenditure and Financial Accountability (PEFA) assessment. Brazil’s federal government got the highest grade in PEFA’s evaluation when it comes to disclosing Verified Unpaid Commitments (VUC). This term refers to unpaid commitments that are verified by the government, which accepts fully or partially the service rendered or goods delivered, recognizing the bill as a liability.These expenses accomplish the budgetary phase of expenditure verification, designated as settled (liquidadas), as the liability is registered in the accrual-accounting regime. These debts, therefore, compromise revenues from future fiscal years. 6. wss subnational planning, budgeting, and spending 85 FIGURE 38 Types of WSS Providers (left) and Type/Status of Municipal Plans (right) WS service provider type Regulated plan Municipal-run company Plan not regulated State-run company Plan to be prepared No observations No plan at all Note: The state of Mato Grosso (in central-west Brazil) eliminated its state WSS company in 1998 and thus is white on the map. Source: SNIS 2019. Data on WSS company type from the Ministry of Cities. Administrative boundary map fromIBGE. Funasa and Instituto Trata Brasil 2016. However, it is important to remember that while the PEFA evaluation takes into consideration the VUCs, it says nothing about Unverified Unpaid Commitments (UVUCs), which have skyrocketed in the last decade, perhaps because they have not been subjectto any regulation (Figure 38). This end-year-flexibility is known in Brazil as restos a pagar não processados (UVUC). Those unpaid commitments refer to committed expenses that are not verified (not settled) by the government as they did not go through the verification stage. In Brazil, “unpaid commitment” is, in general, an issue treated in nationwide research asa measure of indebtedness. It is also associated with accounting or fiscal maneuvers (Augustinho et al. 2013) and classified as a form of “creative accounting” (Almeida 2011 and Melo et al. 2014) to bypass the regime of fiscal discipline in place. The issue has not been associated with the issue of regulating the mechanism of budget deficit or credit carry-over. Transparency and accountability of local budgetary information, if coupled with multi-year flexibility measures, would help avoid discretionary and inefficient use of public funds in the WSS sector. The CGU and the TCU have pointed in recent years to the 86 FIGURE 39 Efficiency of WSS and Financial Liquidity of Municipalities Budget efficiency performance of municipalities Financial liquidity by municipality, 2019 for basic sanitation of REM-F index, 2018 Low High Low High 0.34 0.49 0.49 0.81 0.24 0.41 0.6 0.84 Note: The performance index REMF at the municipal level measures the spending efficiency by which core social sectors (education, health, and basic sanitation) deliver the best outcomes of access and quality of service, with the least budget possible. Basic sanitation lags the education and health sectors in terms of performance. See more at https://www1.folha.uol.com.br/remf/ . The right-hand map displays a subindex of the IFGF fiscal management index. The IFGF is composed of five components: own revenue, staff costs, investments, liquidity, and debt costs. The measure presented on the map corresponds to the liquidity subindex of the IFGF 2019. It verifies whether the municipalities leave sufficient resources to honor their outstanding liabilities year-to-date, calculating the municipality's liquidity as a proportion of current net revenues. An index below 0.5 signals limited liquidity and unsustainable finances. See more information on the methodology here https://www.firjan.com.br/publicacoes/ . Source: REMF, FIRJAN, 2019. uncontrolled increase of unpaid commitments and their use as a mechanism to manage the primary result in the federal government. Nonetheless, a reduction in their general use would not solve the problem. Regional and local governments have engaged in the same practice, with external control bodies having similar low success in containing the growth of UVUC use. Municipal plans could be a tool to ramp up investments and make better use of public expenditures in the WSS sector--but many municipalities have developed no plans (Figure 38). Improving municipal plans to make more efficient use of public expendituresin WSS must be accompanied by budgetary flexibilities to promote multiannual budgets. Multiannual budgets can rely on budget carry-over, which in Brazil can be done through multiple mechanisms (FGV-EBAPE 2017). One involves budgetary appropriations (créditos especiais) expenditures authorized in the current fiscal year that can be reopened the following year, without the need for a new authorizing law. Another channel is to identify the financial surplus registered on the balance sheet, which can then be used to open budgetary credits in the following year. 6. wss subnational planning, budgeting, and spending 87 FIGURE 40 Operational Revenues (median, R$) for Different Types of Entities of Coverage for Water (left)and Sanitation (right) 1,000,000 Median of operational revenues (R$, 2018) Median of operational revenues (R$, 2018) 1,500,000 2,000,000 600,000 1,000,000 400,000 200,000 500,000 0 0 Local Microrregional Regional Local Microrregional Regional Local Microrregional Regional Local Microrregional Regional 2005 2018 2005 2018 Note: Local refers to municipal revenues generated under a single municipality. The difference between microregional and regional is the number of municipalities clustered to provide services. Regions have autonomy in determining their models of service provision and management. Annual operating revenue is the amount resulting from the provision of the water supply and sanitation services, exclusively collected from tariffs and/or fees, excluding the amounts resulting from the sale of water exported wholesale (raw or treated). Source: SNIS 2019. Numerous studies in the last two decades have tried to explain the evolution of the efficiency of sanitation operators in Brazil (Da Motta and Moreira 2006, Ferro et al. 2014,and Cavalcanti et al. 2020). The studies analyzed the performance of these companies, inmany cases measuring the effect of the nature of the property (private versus public) and the jurisdiction of the operation (regional versus local). However, these studies did not consider that spending efficiency and financial liquidity of municipalities (Figure 39) vary broadly and lack the means to improve on those indicators. In face of a budget that is increasingly impeded by compulsory expenditures, municipalities are using accounting tools to postpone expenditures that are not executed via accounts payable. But the institutionalization of these budget adjustments limits their liquidity and pushes the municipalities into a vicious circle of financial inefficiency and insolvency. Technical operational and financial capacity is weak at the municipal level for the elaboration, evaluation, prioritization, and implementation of WSS projects, including the more challenging Public- Private Partnerships. Moreover, public federal and state banks lack the technical capacity for these functions. Improving these skills could improve the financial space and give regional and municipal clusters better operating margins and solvency. 88 FIGURE 41 Brazil is Lagging in Terms of Coverage, with Limited Spending, SNIS 350 1.4 300 1.2 250 1.0 200 0.80 15 0.60 100 0.40 50 0.20 0 0.00 AM MA MG MT AC MS GO ES TO PA PI RN RO SC PR BA PB AP SP RS SE DF PE CE RR AL Total spending in WSS per capita R$ 2017 (left) RJ Total minimum water tariff (circa 2015/2016)USD/m3 Source: SNIS, 2019 Regulations at the state and municipal levels have not been updated to meet the new legal requirements, which may delay implementation of the new WSS framework. Staff training on the new legal frameworks is also incomplete. Completing this would allow federal, state, and municipal governments to update their WSS policies, pluriannual plans,and budgets and start implementing their WSS budgets along the new regulatory lines. Public operators serve 94 percent of the market but remain unable to effectively increase investments. Public utilities invest on average 18 percent of revenues in capital expenditure (CAPEX), most of it (57 percent) financed by equity.48 The best public and private sector performers, in contrast, invest 35 percent of revenues in CAPEX, most of it financed by debt. To increase private sector financing in WSS, the federal government is proposing draft bills that aim to (1) strengthen the National Water Agency (ANA) to be the national reference regulator, (2) help municipalities and states to follow federal reference regulation to access federal financing and technical assistance, and (3) enforce concession contracts, signed after a competitive process that includes the state utilities as well as private operators (Figure 40). 48 Only 28 percent of the utilities’ CAPEX is financed by debt. Budgetary transfers finance only 15 percentof CAPEX and may be further reduced due to budget constraints. 6. wss subnational planning, budgeting, and spending 89 FIGURE 42 Index of Municipal Development for Health and Basic Sanitation in Brazil 2005 Lowest municipal Highest municipal 2016 development development Note: The Index of Municipal Development (IFDM) is a study of the Brazilian system of PFM that annually monitors the socioeconomic development of all of the more than 5,000 Brazilian municipalities, in three areas: employment and income, education, and health (including “Basic Rural Sanitation). It compiles PFM and municipal performance indicators based on official public statistics provided by the Ministries of Labor,Education, and Health (including “Basic Sanitation Rural”). Source: Research team’s own estimates based on data from IFDM. https://www.firjan.com.br/ifdm/ There are limited budgetary standards to promote budget efficiency when revenues of WSS SOEs are covered more than 50 percent through tariffs, so that per capita WSS spending, through a combination of federal and state resources, can adjust based on financial sustainability (Figure 41). Brazil’s fiscal problem is structural and common to the three levels of government; it is related to the high budget share that goes to compulsory expenditures, notably personnel expenses. Thus, in times of revenue decline, such as the current period, there is little room for maneuver to adjust expenditure, leaving the public accounts extremely vulnerable to the economic situation. Although municipal programs for heath and WSS have improved (Figure 42), budget planning and execution are still at the mercy of fiscal vulnerabilities. As a result, in most Brazilian municipalities significant imbalances result as revenues decline and amounts that must be covered by the municipality’s own tax levy grow, or federal transfers. This severely limits ability to finance WSS projects (IFGF 2017). Municipal development has improved, at least for rural areas, with the adoption of theNational WSS Plan (PLANSAB). However, this has not matched the efficiency in public expenditures in the most remote and least populated municipalities, which tend to have increasing population growth rates. 90 FIGURE 43 Distribution of Municipal per Capita Spending on WSS 2018 (left) and Stunting Rate (2017) (right) Per capital basic spending in water and sanitation (R$), 2018 Stunting rate by municipality (newborns), 2017 (300, 400) (107, 300) Very high (42, 107) High (6, 42) Avg. (0, 6) Low Very no data low Source: Estimated with data from SNIS 2019 (left) and Atlas Brasil 2018 (right). For that reason, targets for coverage, access, and quality of WSS services are still lagging in many regions and municipalities. The evolutionof this municipal development index determines, in a disaggregated way, whether improvement that occurred in each municipality results from the adoption of specific policies under the stewardship of local governments. Better WSS can bring dramatic human benefits. Affordable, reliable, easily accessible water and sanitation services prevent thousands of children dying every year from preventable diseases such as diarrhea and cholera. Healthier children absorb nutrients properly, develop stronger brains and bodies, do better in school, and end up making a fuller contribution to society. The COVID-19 pandemic has shown how quickly a pathogencan spread when people are not able to wash their hands with soap and water. But lagging WSS coverage is the reality, exacting major human costs. Without further investment, girls and women are forced to continue the time-consuming, back-breaking work of fetching water. They are left exposed to the indignity and dangers of going to the toilet in insecure facilities or in the fields and streets. Despite the evolution of coverage inrecent decades, 18.3 percent of children aged 0 to 14 years are still living in private households that have no access to the general water supply network. Only 55.5 percent of households have sewage from the general network or a septic tank, and only 39 percentof the country’s sewage is treated.49 In areas that have high rates of stunting, no data onsanitation spending are available, but a connection seems highly likely (Figure 43). 49 See, for instance, http://www.tratabrasil.org.br/images/estudos/itb/beneficios/Press_Release_-_ Benef%C3%ADcios_do_saneamento_no_Brasil.pdf 6. wss subnational planning, budgeting, and spending 91 IV. Benefit Incidence: Distributionand targeting of WSS subsidies Properly designed subsidies for piped water and sewerage are important instrument to increase access to water and sanitation. In the context of poverty especially, where market forces alone do not result in adequate levels of service provision and consumption, subsidies can be important to help address the gaps in affordability withoutnecessarily jeopardizing the objectives of cost recovery or economic efficiency. Data from the Brazil Consumer Expenditure Survey-POF 2017 shows that access to pipedwater and sewer connections differed significantly between rural and urban areas. It is estimated that 93 percent of the country’s urban population have access to piped water connection but only 32 percent of its rural counterparts do likewise headlining a wide urban-rural gap in access to piped water connection in the country. The urban-rural gap for sewer connection is found to be even wider – only 5 percent of the population living inrural areas are connected to the sewers while close to 71 percent of the urban population have access to sewer connection (Table 3). The urban-rural gap in access to piped water and sewer connection are further reflectedacross expenditure per capita deciles which could be a proxy for household wealth. Here, it is found that only 60 percent of individuals from the bottommost expenditure decile have access to piped water connection whereas as much as 93 percent of individuals from the topmost decile have access to piped water connection. In fact, the share of individuals with access to piped water steadily increases across each decile. Thesame is observed for sewer connection. Only 35 percent of individuals from the bottommost decile are connected to the sewers while as much as 85 percent of individuals from the topmost expenditure decile have access to sewer connection as theirsanitation facility (Table 3). However, a considerable number of individuals who have either access to piped water or sewer connection were reported to have not paid water bill in the previous month at the time of the data collection. This could reflect several issues in the water sector such as non-revenue services, payment disruptions, households who are connected to piped water or sewer facilities in place but did not receive any service or even private piped water or sewer connection. 92 TABLE 5 Share of Population with Water and Sewer Access, with Connection and Paid Water and WasteWater Billing Brazil, by Rural/Urban, Expenditure Deciles Share of Population Share of Population Share of Population Share of Population with Access to Piped with Access to Sewer with Access to Piped with Access to Sewer Water Connection and Connection and Paid Water Connection Connection Paid Water Bill Wastewater Bill Brazil 83.8% 67.5% 61.0% 49.1% Rural 31.7% 20.1% 5.0% 3.4% Urban 92.8% 75.7% 70.7% 57.0% Expenditure per capita deciles 1 68.5% 46.9% 35.5% 24.6% 2 76.2% 57.9% 46.1% 34.7% 3 79.9% 63.3% 51.2% 41.1% 4 81.3% 67.2% 54.5% 45.6% 5 84.3% 70.4% 61.2% 51.8% 6 85.5% 72.4% 62.8% 53.7% 7 87.9% 75.3% 67.5% 57.2% 8 89.8% 77.2% 71.4% 61.1% 9 91.4% 76.3% 75.6% 62.2% 10 93.3% 67.8% 84.3% 59.2% Source: Brazil Consumer Expenditure Survey - POF, 2017/18 By examining the share of population with connection to piped water and sewer services across the different regions, similar findings regarding gap in access are uncovered. As shown in Table 4, access to piped water and sewer connection is lowest in the Northern region in Brazil at 60 percent and 19 percent, respectively. It is also estimated that the state of Rondônia, which is surrounded by the Amazon rainforest and the fifth least populated state in the country, has the lowest access to both piped water and sewer connections at 37 percent and 5 percent, respectively. Access to both piped water and sewer connection in two other neighboring states, the Amazonas and Acre, are also among the lowest in the country. On the other hand, access to piped water and sewer connection are highest in the Southeast region at 92 percent and 87 percent, respectively. The state of São Paulo, a predominantly urban state which is also the most populous state in the country, has the highest share of population with access to piped water and sewer connection at 97 percent and 92 percent, respectively. This is followed by the country’s Federal District, located in the Central-West region at 95 percent and 84 percent, respectively (Table 4). 6. wss subnational planning, budgeting, and spending 93 TABLE 6 Share of Population with Water and Sewer Access, and Share of Population with Connection and Paid Waterand WasteWater Bill in Brazil, by Region, State Share of Population Share of Population Share of Population Share of Population with Access to Piped with Access to Sewer with Access to Piped with Access to Sewer Water Connection and Connection and Paid Water Connection Connection Paid Water Bill Wastewater Bill Nation 83.8% 67.5% 61.0% 49.1% North 59.5% 37.6 18.8% 12.3% % Rondônia 36.6% 27.8% 5.0% 4.3% Acre 50.2% 32.9% 29.4% 16.6% Amazonas 70.0% 50.2% 27.5% 19.1% Roraima 87.3% 66.0% 30.2% 25.8% Pará 54.4% 27.1% 14.8% 7.2% Amapá 50.3% 11.6% 5.8% 3.7% Tocantins 87.5% 80.8% 31.8% 29.8% Northeast 76.7% 59.1 37.4% 29.4% % Maranhão 63.8% 33.2% 13.3% 8.5% Piauí 84.4% 65.4% 7.2% 6.6% Ceará 79.4% 69.9% 38.6% 33.2% Rio Grande do Norte 81.2% 64.9% 22.7% 19.2% Paraíba 70.7% 58.3% 44.8% 36.0% Pernambuco 72.2% 51.7% 45.7% 32.9% Alagoas 67.2% 54.8% 24.4% 20.3% Sergipe 87.3% 73.6% 48.7% 43.0% Bahia 83.4% 65.6% 52.0% 40.4% Southeast 91.8% 73.2 86.5% 68.4% % Minas Gerais 88.7% 77.2% 81.0% 70.8% Espírito Santo 86.3% 70.5% 74.3% 60.4% Rio de Janeiro 84.3% 41.4% 82.2% 37.2% São Paulo 96.5% 83.6% 91.8% 79.7% South 86.8% 78.5% 62.1% 54.4% Paraná 89.3% 81.3% 65.1% 57.9% Santa Catarina 82.4% 74.1% 55.7% 48.8% Rio Grande do Sul 87.0% 78.3% 63.1% 54.4% Central-West 86.7% 78.7 49.8% 44.7% % Mato Grosso do Sul 88.7% 80.0% 44.3% 38.6% Mato Grosso 80.0% 68.3% 30.1% 26.1% Goiás 85.7% 81.6% 46.8% 43.8% Distrito Federal 94.9% 82.8% 84.2% 73.4% Source: Brazil Consumer Expenditure Survey - POF, 2017/18 94 Regarding the share of expenditure on water and wastewater, data shows that households in Brazil spend on average about 0.21 percent of their total monthly consumption expenditures on water and about 0.07 percent of their total monthly consumption expenditures on wastewater. These estimates roughly translate to R$37.6 (US$11.8)50 andR$17.5 (US$5.5), respectively, on average. These results are presented in Table 5 below. On average, the urban population in Brazil spends more on both water and wastewatercompared to their rural counterparts. Across urban areas the values are estimated to be 0.23 percent (R$38.8 or US$12.1 per month) and 0.08 percent (R$16.8 or US$5.3) respectively; while in rural areas, the values are estimated to be around 0.06 percent(R$16.8 or US$5.3) and 0.01 percent (R$11.2 or US$3.5), respectively. The amount of household water and wastewater expenditure also increase steadily across the expenditure deciles. The population in the bottommost decile on average spend only R$20.4 (US$6.4) and R$9.9 (US$3.1) for water and wastewater expenditures respectively but the population in the topmost expenditure per capita decile spend roughly R$50.6 (US$15.8) and R$21.8 (US$6.8), respectively, for water and wastewater on average. However, through a closer examination of the share of the water and wastewater expenditure respectively out of the total consumption expenditure of the households in each decile, it can be observed that individuals in the lower deciles would actually allocate a higher share of their total expenditures on both the water and wastewater expenditures compared to individuals in the higher expenditure deciles. In other words, although perhaps not substantive, the burden of water and wastewater expenditure is higher for poorer individuals than that of richer individuals (Table 5). The proportion of beneficiaries who receive tariff subsidies within each decile is higher in the upper deciles compared to that in the lower deciles, for both water and wastewater subsidies. In other words, a higher proportion of poorer individuals are being excluded from receiving the water and wastewater subsidies compared to that of richer individuals. For example, as much as 29 percent of individuals in the bottommost decile are excluded from receiving water subsidies while as much as 60 percent of individuals from the same decile are excluded from receiving wastewater subsidies. Meanwhile, the corresponding figures for individuals who were excluded fromreceiving the subsidies at the topmost decile are 7 percent and 14 percent only. While 10 percent of the water subsidies received nationally are accrued to the bottommost decile, as much as 12 percent of the water subsidies are accrued to the topmost decile (Figure 44). The distribution of water expenditures made by the households across the deciles on the other hand is more pro-rich. The share of water expenditure is more concentrated at the higher deciles compared to the lower deciles. 50 Using the official 2017 LCU exchange rate of US$1 = R$3.2. 6. wss subnational planning, budgeting, and spending 95 TABLE 7 Summary; Amount and Share of expenditure on water, share of exp on sewer (monthly) Share of Household Water Share of Household Household Water Household Wastewater Expenditure out ofTotal Wastewater Expenditure Expenditure in R$ Expenditure in R$ Expenditure out of Total Expenditure Nation 37.6 0.21% 17.5 0.07% Rural 16.8 0.06% 11.2 0.01% Urban 38.8 0.23% 17.5 0.08% Expenditure per capita deciles 1 20.4 0.67% 9.9 0.18% 2 27.5 0.53% 12.8 0.16% 3 31.8 0.45% 14.6 0.14% 4 34.7 0.38% 16.1 0.12% 5 38.6 0.34% 17.7 0.12% 6 40.2 0.29% 18.2 0.10% 7 42.4 0.25% 18.9 0.09% 8 45.1 0.21% 19.9 0.08% 9 47.5 0.17% 20.9 0.06% 10 50.6 0.08% 21.8 0.03% Source: Original calculations using –data from Consumer Expenditure Survey – POF 2017/18, and IBNET.See the data and methodology section, and the appendix, for more detail. Note: Subsidy beneficiaries are households. All figures are calculated using sample weights. Total expenditure is total household expenditure in all categories. For example, 14 percent of the total water expenditure comes from the topmost decile while only 6 percent of the total expenditure is accrued to the bottommost decile. On the other hand, and although it is not shown in the graph, both the distributions of wastewater expenditures and wastewater subsidies turned out to be regressive or pro-rich. Subsidies are more important for the poor compared to the rich (materiality of subsidies). This analysis found that, in the absence of water and wastewater subsidies, the share of both expenditures as a share of total expenditures at the household level will increase significantly and in fact, more dramatically for the poorer as compared to the rich. For example, it is estimated that the share of water expenditure will increase by roughly 4.7 times for individuals from the bottommost decile from 1.4. to roughly 6.6 percent if the subsidies were not available to them. Similarly, it is estimated that the share of wastewater expenditure out of the total expenditure at the household level could increase from 0.7 percent to 8.1 percent translating to 11.5 times increase. Among the rich,the removal of subsidies is estimated to increase the share of both water and wastewater expenditures as a share of total household expenditures from 0.1 percent to 0.4 percent, respectively. This shows that the subsidies are especially important to alleviate the burden of water and wastewater consumption among the poorest in the country. 96 FIGURE 44 Share of water expenditures and subsidies per deciles 16% 14% 12% 10% 8% 6% 4% 2% 0% 1 poorest 2 3 4 5 6 7 8 9 10 richest Share of Water Expenditure in the Decile out of Total Water Expenditure in All Deciles Share of Water Subsidy Received in Each Decile out of Total Water Subsidy Received in All Deciles Note: The graph compares Share of Water Expenditure in the Decile out of Total Water Expenditure in All Deciles vs. Share of Water Subsidy Received in Each Decile out of Total Water Subsidy Received in All Deciles (Decile Share Distribution of Water Expenditure vs. Decile Share). Subsidy beneficiaries are households. All figures are calculated using sample weights. Total expenditure is total household expenditure in all categories. Source: Own calculations using data from Consumer Expenditure Survey – POF 2017/18, and IBNET. See the data and methodology section, and the appendix, for more detail. Results of the Omega Decomposition analysis performed shown that water and wastewater subsidies in Brazil inadequately target the poor51. Table 6 and Table 7 demonstrate how access factors (SA*C) and subsidy design factors (T*R) drive the omega (Ω) value, nationally and across urban and rural population for both water and wastewater subsidies. A value below 1 indicates that the factor lowers performance (a regressive subsidy); whereas a value above 1 indicates that the factor boosts performance in targeting the poor (a progressive subsidy). For Brazil, the omega value is estimated to be at 0.88 nationally for water subsidies and 0.73 for wastewater subsidies. This indicates that the water and wastewater consumption subsidies are targeted inadequately to the most vulnerable or poor population. An advantage of the omega decomposition analysis is that it allows examining the contributors to the omega values; that is, the factors that affect subsidy performance and that render them regressive or progressive. It was found that nationally, both the quantities of consumption (Q) and access to facilities (SA) were contributing negatively to the performance of the water and wastewater consumption subsidies in targeting the 51 Omega decomposition shows the profile of efficiency in targeting subsidies to the lowest deciles of theincome distribution. 6. wss subnational planning, budgeting, and spending 97 TABLE 8 Omega Decomposition for Water Subsidies Omega (Ω) SA C T R Q Potential ac- Receipt of Connection Quantities omega>1 cess to water subsidy (for Rate of rate (for those (for those pro-poor connection those subsidization* with access) connected) (service area) connected) Nation Poor 0.8752 0.7870 1.0000 0.9981 0.7124 11.8272 All 0.8490 1.0000 0.9976 0.6805 13.1215 Poor/All 0.9270 1.0000 1.0005 1.0470 0.9014 Rural Poor 1.0104 0.3316 1.0000 0.9993 0.8229 11.3990 All 0.3224 1.0000 0.9989 0.8048 11.8708 Poor/All 1.0286 1.0000 1.0004 1.0226 0.9603 Urban Poor 0.9126 0.9080 1.0000 0.9979 0.7017 11.8685 All 0.9331 1.0000 0.9975 0.6737 13.1899 Poor/All 0.9731 1.0000 1.0005 1.0417 0.8998 Note: Poor households are defined as belonging to the first four deciles of the expenditure per capita distribution. *Includes only group of households receiving subsidies. Source: Original calculations using data from Consumer Expenditure Survey – POF 2017/18, and IBNET. See the data and methodology section, and the appendix, for more detail. poor. Thus, to improve the targeting performance of the water and wastewater consumption subsidies, the country could expand their service area for piped water and subsequently increase access to piped water among its population especially in rural ordifficult-to-access communities. The water consumption subsidies performed slightly better in targeting the poor in the rural areas (Ω =1.01) compared to in urban areas (Ω =0.91). However, the same could not be said for wastewater consumption subsidies (Ω =0.77 and Ω =0.79 respectively). Again, the main contributors to the bad performance of the water subsidies in targeting the por among urban areas were quantities of water consumed and access to piped water connection. On the other hand, the targeting performance of wastewater consumption subsidies are similarly inadequate across urban and rural areas. Similar results are reflected spatially across the different geographical divisions in thecountry. Water and wastewater consumption subsidies performed relatively better in targeting the poor population in the northern regions (Ω =0.94 and Ω =0.81 respectively) relative to the rest of the country (Figure 45). This is perhaps because the northern regions are less populated and have more rural population compared to the rest of thecountry. Similarly, it is estimated that the water consumption subsidies performed the best in targeting the most vulnerable population in the Amazonas and Sergipe states both of which are predominantly rural states. 98 TABLE 9 Omega Decomposition for Wastewater Subsidies Omega (Ω) SA C T R Q Potential Receipt of Rate of access to Connection Quantities omega>1 subsidy (for subsidization wastewater rate (for those (for those pro-poor those (for those witha connection with access) connected) connected) subsidy) (service area) Nation Poor 0.7285 0.5076 1.0000 1.0000 0.8781 12.1009 All 0.6325 1.0000 1.0000 0.8637 13.5546 Poor/All 0.8026 1.0000 1.0000 1.0167 0.8927 Rural Poor 0.7722 0.0432 1.0000 1.0000 0.9112 11.2127 All 0.0516 1.0000 1.0000 0.8949 12.3846 Poor/All 0.8376 1.0000 1.0000 1.0182 0.9054 Urban Poor 0.7899 0.6311 1.0000 1.0000 0.8775 12.115 All 0.7252 1.0000 1.0000 0.8634 13.5659 Poor/All 0.8702 1.0000 1.0000 1.0164 0.8931 Note: Poor households are defined as belonging to the first four deciles of the expenditure per capita distribution. Source: Original calculations using data from Consumer Expenditure Survey – POF 2017/18, and IBNET. See the data and methodology section, and the appendix, for more detail. On the other hand, the targeting performance of wastewater consumption subsidies are found to be most adequate in the state of Rondônia (Figure 46). Several states have both relatively better targeting performance for water or wastewater consumption subsidies in targeting the poor and high level of access to piped water connections or sewer connections. By combining the two indicators of omega values and access levels, it is possible to identify states which have high level of access to either piped water connection or sewer connection and high performing wateror wastewater subsidies, respectively (Figures 47 and 48). Thus, states like Amazonas, Acre, Roraima in the North region as well as Mato Grosso do Sul and Rio Grande do Sul are found to have both relatively higher levels of access to piped water and relatively better targeting performance of water consumption subsidies as compared to the rest of the country. On the other hand, states like Pará, Bahia and Rondônia have both low access to piped water connections and inadequate performing water consumption subsidies to target the poor. These are the states where improving access and coverage of piped water connection should be prioritized. Finally, São Paulo, the country’s most populous state has high access to piped water connection, but the water consumption subsidies poorly perform in targeting the poor in this state. 6. wss subnational planning, budgeting, and spending 99 FIGURE 45 Water Omega – Targeting subsidies by region, by state Good targeting Average targeting Poor targeting Very poor targeting Poor Water Omega targeting <=0.90 0.91-1 Source. Original calculations using data from Consumer Expenditure Survey – POF 2017/18, and IBNET. See the data and methodology section, and the appendix, for more detail. Similar findings, combining omega values and access levels, can be observed for wastewater consumption subsidies (Figure 48). States like Rondônia, Amazonas and Tocantins are considered to have both high level of access to sewer connection and relatively better performing wastewater consumption subsidies in targeting the poor; whereas states like Pará and Bahia again have both low levels of access to sewer connections and wastewater subsidies that inadequately target the poor. Again, expansion of service areas for sewer connections should be prioritized in these states. FIGURA 46 Sanitation/Wastewater Omega – Targeting subsidies by region, by state Poor targeting Average targeting Average targeting Good targeting Wastewater Omega Very poor <=0.90 targeting 0.91-1 Source. : Original calculations using data from Consumer Expenditure Survey – POF 2017/18, and IBNET. See the data and methodology section, and the appendix, for more detail. 100 FIGURA 47 FIGURA 48 Water Omega- Access cross map, by state Sanitation Omega- Access cross map, by state Water Omega vs. Access Sanitation Omega vs. Access High, high High, high High, low High, low Low, high Low, high Low, low Low, low Source: Own calculations using data from Consumer Expenditure Survey – POF 2017/18, and IBNET. See the data and methodology section, and the appendix, for more detail. Note: “If the value above or equal to the median of the indicator, the state is defined as “High”; otherwise as “Low”. States with both “Low” omega value and “Low” access are highlighted in yellow in the map. “If the value above or equal to the median of the indicator, the state is defined as “High”; otherwise as “Low”. States with both “Low” omega value and “Low” access are highlighted in yellow in the map. Inefficient distribution of subsidies could further contribute to the poor targeting performance of consumption subsidies. These inefficiencies could arise from what is called error of inclusion and error of exclusion. The former is defined as relatively richer individuals or households (T60) who have received the subsidies when they should not have; while the latter is defined as relatively poorer individuals who did not receive the subsidies when they should have. The results are presented in Table 10 and Table 11. Nationally, the error of inclusion is found to be around 22 percent while the error of exclusion is around 59 percent for water subsidies. This means that as much as 22 percent of T60 households have received the water subsidies when they should not have and that 59 percent of B40 households did not receive water subsidies when they shouldhave. The error of inclusion is much higher in rural areas, estimated to be around 67 percent. This would indicate that a lot of the water subsidies were reaching to the better- off households in rural areas and would suggest the need to improve the level of targeting precision to the poor in these areas. On the other hand, the error of exclusion is higher in urban areas, around 60 percent, which suggest more outreach to improve piped water access among the urban poor can still be improved. Similarly, Table 11 shows the errors of inclusion and the error of exclusion for the wastewater consumption subsidy. Both errors of inclusion and exclusion are high as a nation, and across rural and urban areas. The extremely high level of the inclusion error for rural areas is consistent with extremely low sewer access. Only the households in the richest deciles in rural Brazil have the access to a network sewer. 6. wss subnational planning, budgeting, and spending 101 FIGURA 49 Water Errors of Exclusion, by region (left) and state (right) Very low error Low error High error Average error Water Error of Exclusion <=0.40 Very high 0.41-0.60 error 0.61-0.80 Source. Original calculations using data from Consumer Expenditure Survey – POF 2017/18, and IBNET. See the data and methodology section, and the appendix, for more detail. 102 TABLE 10 Errors of inclusion and exclusion of water consumption subsidy, national, urban, rural Errors of inclusion Errors of exclusion Brazil 0.2150 0.5864 Rural 0.6707 0.2978 Urban 0.0939 0.6022 Note: Poor households are defined as belonging to the first four deciles of the expenditure per capita distribution. Error of inclusion is measured by the percentage of all beneficiary households that are rich; error of exclusion is measured by the percentage of poor households that do not get a subsidy. All figures are calculated using sample weights. Source: Original calculations using data from Consumer Expenditure Survey – POF 2017/18, and IBNET. See the data and methodology section, and the appendix, for more detail. Consistent with the results that water consumption subsidies performed relatively better in targeting the poor population in the northern regions, the two regions in the north show much lower level of errors of exclusion (North 39 percent and Northeast 42 percent respectively), relative to the other three regions that are all above 60 percent. On the other hand, states in the South and South-eastern regions have higher levels of error of exclusion. In particular, Federal District has the highest error of exclusion at 80 percent, suggesting poor targeting precision in reaching the poorest, probably due to the high population density of slums. The same can also be said for São Paulo (Figure 49). TABLE 11 Errors of inclusion and exclusion of wastewater consumption subsidy, national, urban, rural Errors of inclusion Errors of exclusion Brazil 0.4939 0.6426 Rural 0.9597 0.4158 Urban 0.3701 0.6449 Note: Poor households are defined as belonging to the first four deciles of the expenditure per capita distribution. Error of inclusion is measured by the percentage of all beneficiary households that are rich; error of exclusion is measured by the percentage of poor households that do not get a subsidy. All figuresare calculated using sample weights. Source: Original calculations using data from Consumer Expenditure Survey – POF 2017/18, and IBNET. See the data and methodology section, and the appendix, for more detail. 6. wss subnational planning, budgeting, and spending 103 FIGURA 50 Sanitation/wastewater Errors of Exclusion, by region, by state Very low error Low error High error Average error Sanitation Error of Exclusion Very high <=0.40 error 0.41-0.65 0.66-0.90 Source. Original calculations using data from Consumer Expenditure Survey – POF 2017/18, and IBNET. See the data and methodology section, and the appendix, for more detail. Similarly, for sanitation/wastewater consumption subsidies, the errors of exclusion are notably higher in the southern regions. The South, Southeast, and Central-West regions all have higher levels of error of exclusion (around 70 percent), suggesting a wide margin for the improvement of targeting. Again, predominantly urban, and populous states like São Paulo are identified to have among the highest error of exclusion (Figure 50). By combining the two indicators of the errors of exclusion and omega values, it is possible to identify states which have high level of exclusion errors of water or wastewater consumption subsidy and low targeting performance of water or wastewater subsidies, respectively. These are shown in Figure 51. Several states in both North and Northeast regions have better performing water or wastewater consumption subsidies in targeting the poor (i.e., a high omega value) and lower levels of error of exclusion. For example, states like Amazonas, Acre, Roraima in theNorth region as well as Tocantins, Maranhão, Piauí, Ceará, Paraíba, and Pernambuco in theNortheast are found to have both relatively better targeting performance of water consumption subsidies and have relatively less error in excluding the poor as beneficiaries compared to the rest of the country. On the other hand, most states in Central-West and Southeast regions (except Mato Grosso do Sul and Minas Gerais), as wellas Amapá in North, Rio Grande do Norte in Northeast, and Paraná in South have both relatively lower targeting performing of water subsidy and higher level of errors in excluding the poor from benefiting from the subsidies. 104 FIGURA 51 Water and Wastewater Omegas - Errors of exclusion map, by state Water Omega vs. Error of Wastewater vs. Exclusion High, high High, high High, low High, low Low, high Low, high Low, low Low, low Source. Original calculations using data from Consumer Expenditure Survey – POF 2017/18, and IBNET. See the data and methodology section, and the appendix, for more detail. Similar findings can be observed for wastewater consumption subsidies, also combining the omega values and the error of exclusion indicators. States like Rondônia, Amazonas, Tocantins, Pernambuco in the north, and Rio de Janeiro in the south are considered to haveboth relatively better performing wastewater consumption subsidies in targeting the poor and lower level of exclusion error; whereas states like Amapá, Rio Grande do Norte, Paraná and most of the states in Central-West again have both wastewater subsidies thatinadequately target the poor as well as high level of exclusion error (Figure 51). Reforming subsidies has the difficulty of the trade-off between rapidly cutting budget- financed subsidies and avoiding an adverse impact on the poor. Also, removing of any type of subsidy is politically inviable. A gradual reform of subsidies could, first, identify the sources of the errors of inclusion and exclusion in terms of monetary magnitude, to then adjust budgets and reduce gradually subsidized tariffs. Then with those adjustments any remaining financial resources freed could be used to improve targeting tothe poorest, particularly for sanitation services, where the urban poor are mostly facing those types of tariff affordability barriers. 6. wss subnational planning, budgeting, and spending 10 7.Investments and Expenditures: A Shaken WSS Sector 106 Investments in WSS in recent years have ranged between R$11.4 billion (US$2.2 billion) in 2011 and R$13.2 billion (US$3.6 billion) in 2018 (ITB 2018). Many of the construction materials for WSS infrastructure are imported into the country, so due to large depreciation in the Brazil Real the WSS investments in US dollars have declined over the last decade. Current investment rates and changes in coverage over the last decade suggest the country could reach universalization only after 2050, a delay of about 20 years from the 2033 target. The target will be met only if the sector accelerates coverage growth with a robust investment framework to promote innovation. It would also need a simplified budget framework that includes explicit criteria to invest in subsectors that bring the most cost-effective solutions and maximize coverage. Investment would bolster subsectors that bring the highest co-benefits in health, environmental preservation, and human capital development. I. Funding Gaps for Water, Sanitation,and Hygiene 2020 – 2030 Expanding water and sanitation services by strengthening the systems that deliver them is the bedrock of economic growth and sustainable and human development. The water and sanitation funding gaps in Brazil are increasing rapidly, with unmet demand for services leading to rapidly increasing social and economic costs. Brazil executed federal WSS funds of about R$1 billion (US$192 million) in urban areas and R$210 million in rural areas (US$40 million) in 2020. For the first time that year, the budget execution for urban WSS was higher that the planned budget. Over the previous 20 years, the percentage of federal WSS budget executed against the total budget allocation varied between 50 and 64 percent. The gains in efficiency materialized due to a reduction in the budget allocation, and a concentration of projects that executed at higher-than-average rates. Brazil will need over the next decade to increase its budget execution by tenfold to ensure safely managed services are provided universally, according to recent data fromthe Joint Monitoring Program and Sanitation and the Water for All (SWA) WSS costingtool (2020). This extra capital investment will both build new infrastructure and maintain and upgrade the current services, and reach the unserved population by 2033. In aggregate Brazil will require between US$4.9 and US$9.8 billion annually for the WSS sector to reach at least universal basic coverage or even safely managed services (Figure 52). 107 FIGURE 52 Annual Capital Costs of Reaching the Entire Population to Achieve Universal Access of Different ServiceLevels by 2030, Brazil 16,000 14,000 12,000 Million USD 10,000 8,000 6,000 4,000 2,000 0 Ending Water Sanitation Hygiene WASH Water Sanitation WatSan Targets Open 6.1 + 6.2 Defecation Maintaining services for served in 2020 Reaching unserved 2020 to 2030 Universal Basic Service Universal Safely Managed Service Population group Output Targets served Ending OD Water Sanitation Hygiene WASH Water Sanitation WatSan 6.1 + 6.2 Reaching unserved 19 322 1,246 15 1,583 1,442 5,770 7,213 8,634 2020 to 2030 Total cost Maintaining services 339 1,077 1,997 233 3,307 2,339 255 2,594 5,363 for served in 2020 Reaching unserved 0.00% 0.02% 0.06% 0.00% 0.08% 0.08% 0.30% 0.38% 0.04% 2020 to 2030 Percent GDP Maintaining services for served in 2020 0.02% 0.06% 0.10% 0.01% 0.17% 0.12% 0.01% 0.13% 0.03% Source: WASH costing tool, SWA 2020. Although these figures are considerably higher than the budget execution of 2020, theyrepresent only a small fraction of Gross Domestic Product (GDP). For instance, the totalcost of reaching universalization of basic services for both water and sanitation,52 would amount to only 0.25 percent of Brazil’s GDP. Reaching SDG Targets 6.1 and 6.2 by 2033, including preserving current customers and adding new ones would, would cost just 0.4 percent of GDP. And reaching safely managed universalization would amount to just 0.70 percent of GDP. 52 Making adequate investments could see disease burden and epidemic risk decline, including fast- moving killers such as cholera. For example, improved hygiene—through water and soap—is critical inthe fight against COVID- 19. Yet one in four—24 percent—of health care facilities lack basic water services, one in 10—10 percent—have no sanitation service, and one in three—32 percent—lack hand hygiene facilities at points of care. 108 FIGURE 53 Water and Sanitation Public Investments as % of GDP in Brazil, 1971 – 2 020 0.50 Debt crisis, Hyperinflation, PAC and Economic hyperinflation, Debt renegotiation Currency closure of PANASA fiscal recession 0.45 closure of and national crisis and flexibility expenditure- and tariff and prices Brazilian recovery plan fiscal containment program, focused fiscal 0.40 Housing Bank adjustment Bankruptcy of FGTS LRF adjustment (Fundo du Garantiado 0.35 Tempo de Serviço) 0.30 Law 4.162 0.25 (2020) 0.20 0.15 0.10 Planasa Macroeconomic turmoil Regulatory Regulatory moderation consolidation 1971 1977 1972 1973 1975 1991 2011 1974 1976 1979 1981 1997 2017 1978 1987 1992 2012 1993 1995 2013 2015 1982 1983 1985 1994 1996 1999 2014 2016 2019 2020 1984 1986 1989 1998 2001 2018 1988 1990 2010 1980 2007 2002 2003 2005 2004 2006 2009 2008 2000 Note: LRF = Fiscal Responsibility Law. Source: Research team’s own elaboration based on World Bank and IMF data 2020. The long-term trend of WSS public investments as a percentage of GDP shows a declining pattern with periods of high volatility, primarily induced by macroeconomic and fiscal instability. Figure 53 shows a steady increase in these investments from 1971 until 1981. After that period, however, a series of macro-economic and fiscal vulnerabilitiescreated conditions for high budgetary planning and allocation uncertainties. This resulted in a gradual reduction of investments in the sector as a proportion of GDP between 1987 and 2007. In 2007 an updated regulatory framework brought hope for better conditions and certainty to expand the budgetary allocation. However, the regulatory changes proved insufficient. The fiscal and currency vulnerabilities facing the country continued to drive down WSS investments as a percent of GDP. With a new WSS law now in effect, the viability of financial resources to support its implementation are as important as the law itself. The regulatory changes of 2007 provide a precedent for targeted investment to facilitate the implementation of the law. Under those changes, the federal government launched the Growth Acceleration Program—PAC—with the objective of promoting economic growth by investing in various sectors of infrastructure, including WSS. In 2011, calls were made for infrastructure project in a second phase, PAC II. WSS received strong financial stimulus, especially for the expansion of water supply systems and sanitary sewage. The National WSS Plan (PLANSAB), approved in 2013, was the strategic foundation forreforms inserted in the WSS Law. Under the new law, PAC’s investments are linked to short- and medium- 7. investments and expenditures: a shaken wss 109 term planning objectives. PLANSAB is the result of a broad debate between the federal government, the National Sanitation Secretariat Ministry of Cities (now the Ministry of Regional Development), and the national councils, such as the Council of Cities (ConCidades). Citizens participate through public consultations. The new law requires more targeted and scaled-up program to ensure (1) a fiscal stabilization program for CESBs that face financial strains due to COVID-19, (2) incentives to promote efficiency improvements and equity of the CESBs, and (3) investments to enhance municipal capacities and overcome barriers of contracting arrangements and planning for service delivery. The development bank CAIXA Econômica Federal (CEF) monitored financing operations of 2,250 WSS works between 1990 and 2016 and found that 56 percent of the contracts were related to the expansion of the water supply. The contracted investments amount to R$42.2 billion (US$8.0 billion), 75 percent of which were financed through the Sanitation for All program. The remaining 25 percent consisted of programs monitored by the CAIXA partnership program. Public ownership has held generally constant in the WSS sector, mainly in the form of state- owned companies. However, the private sector has gained significant participation in investments due to the scaling up of the Sanitation for All program. Private companies account for almost a third of the loans made by CEF in the last nine years. Though the number of contracts by private entities is modest, the contracts are generally large, carrying out projects that demand high investment. However, these operations are also riskier than normal and have shown mixed results in overall performance (FGV 2017). Other development banks have structured other WSS projects with similar results. II. Gridlocks of Efficient WSS Spending Even though only 12 percent of the federal government’s primary spending goes to payroll, the aggregate wage bill of all levels of government in Brazil is high by international standards (World Bank 2017). The country’s public sector wage bill increased from 11.6 percent of GDP in 2006 to 13.1 percent in 2015, leaving behind even Portugal and France, which had larger wage bills than Brazil a decade ago (World Bank2017). The WSS sector faces this same high wage bill- but it is driven by high salaries of public officials, including those employed in the WSS sector, rather than an excessive number of employees. WSS SOEs are facing financial and fiscal vulnerabilities due to increasing wage bills. Thishas limited the sector’s potential to steadily increase investments. In addition, regulatory uncertainty undermines WSS public service when different levels of government share a mandate (Table 12) on public service provision (Kresch 2020). This creates conditions wherebudgets cannot adapt to emerging challenges, and the additional public resources for specific WSS infrastructures become 110 TABLE 12 Responsibilities of different Levels of Government for basic public services Water and Sanitation Electricity Airports Country Local State Federal Local State Federal Local State Federal Bangladesh X X X Ghana X X X X Honduras X X X X X X Uganda X X X X X Brazil X X X X X X X X Indonesia X X X X X X Iran X X X X Vietnam X X X X X Germany X X X X X Japan X X X X X X Netherlands X X X X X USA X X X X X X Source: Adapted from Kresch, 2020. The Buck Stops Where? Federalism, Uncertainty, and Investment in the Brazilian Water and Sanitation Sector. American Economic Journal: Economic Policy, 12(3): 374–401 redundant because of unclear rules of the types ofinvestments supported by three different layers of government. Financial stress has alsoled some SOEs to be taken over by states. Brazil’s high public spending on pensions (Figure 54) hinders the ability of public WSS enterprises to invest in building much-needed new facilities and in maintaining existing ones. The main driver of this fiscal pressure is growth in mandatory primary expenditures, which consist largely of severance payments and pensions. Primary expenditures grew on average by 6.1 percent per year in real terms between 2006 and 2018 (World Bank 2017, TCU 2019). This rise originates from the indexing of large parts of federal primary expenditure to GDP, revenues, or the minimum wage, as well as widespread earmarking of revenues and mandated minimum spending levels. Certain budgetary policies aimed at improving efficiency of social assistance programsand pensions could have positive repercussions in the WSS sector. A case in point, according to the Public Expenditure Review of Brazil (World Bank 2017), is a reform to consolidate social pensions with other social assistance programs by aligning their targeting and levels of benefits. 7. investments and expenditures: a shaken wss 111 FIGURE 54 Brazil pension spending is high: Share of population and spending as % of GDP, 2017 (left) and Debtand public spending during the COVID-19 pandemic by country, 2020 (right) Brazil Mexico Peru Paraguay Chile Colombia Fiscal measures against COVID-19 as % of GDP Uruguay Argentina 1 2 14 2 Perú Brazil 1 Chile 12 20 Tailandia 10 9 Polónia 8 Africa do Sul 8 7 Rep. Checa 6 Arabia Saudita 6 Austria 5 Indonesia Paquistão Argentina 4 4 Bulgaria Filipinas Rússia Colombia Índia 3 Turquía México 2 2 % Population 65+ Pensions % GDP 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 Public Debt as % of GDP Source: OECD, 2018 and IMF, 2020. This could generate significant savings and free up resources to spend on urgent social needs such as access to water and sanitation, earlychildhood education, and old age care. Despite fiscal risks and macroeconomic vulnerabilities, investment needs to grow dramatically if Brazil is to meet the SDG6 challenge.53 The global gaps in infrastructure inLatin America and the Caribbean call for expanding investments considerably, particularlyin Brazil, where the combination of public and private investments in infrastructure (including WSS) reached a small proportion of GDP in 2018 (Figure 55). The sector’s budget cycles are determined and influenced not only by national regulations and strategies, but by external and macroeconomic conditions. The planned investments, that is, the amounts included in the federal budget, and the investments executed, show a large gap of efficiency for the WSS sector in Brazil. The fiscal outlook is also vulnerable to significant contingent liabilities, mainly related to subnational fiscal distress.54 53 LICS and MICS need an estimated global annual investment of US$114 billion to meet SDG6 access challenges, according to the World Bank. 54 For further detail, see World Bank, 2017. http://documents1.worldbank.org/curated/en/643471520429223428/pdf/Volume-1-Overview.pdf 112 FIGURE 55 Investment in Infrastructure as a Percentage of GDP 6 5 4 As percentage of GDP 3 2 1 0 Trinidad and Tobago Haiti Argentina Paraguay Mexico Guatemala El Salvador Brazil Dominican Republic Uruguay Chile Costa Rica Caribbean average Colombia Guyana Belize Ecuador Latina America and the Panama Honduras Nicaragua Peru Bolivia Private Public Note: Public investment data for all countries are averages for the years 2008-17, with the exception of the Dominican Republic (2009-17), Ecuador (2008-16), El Salvador (2008-15), and Haiti (2012-16). Source: Data for public investment from INFRALATAM. Data for private Investment from PPI. Figure 2.1 in Cavallo et al. 2020. Just to expand coverage, the World Bank (2018) estimates that universalization of WSS in Brazil would require investments of R$317 billion (US$60 billion) in 20 years or R$16 billion (US$3 billion) per year.55 These are large sums, but they would bring economic and social benefits to the country in the order of R$537 billion (US$102 billion). Most importantly, universalization would unlock strong future gains from externalities in health, productivity, and environmental enhancement. For now, however, public budget for WSS is proportionally the lowest among infrastructure subsectors (as a percent of GDP) and is spent with relatively high inefficiency. WSS spending inefficiency is even higher when compared to social sectors. International benchmarking of Brazil’s wage bill as a percent of GDP, of public expenditures, and of revenues, highlights that Brazil exceeds the average of countries of all income groups. As a share of GDP, Brazil’s overall wage bill is also higher than that of any regional average of countries. For infrastructure, the proportion of public expenditure as a percentage of GDP is lower than the LAC average with high costs of compensations, pensions, and severance payments (PER 2017). 55 World Bank (2018). Volume V–Tema 4: Sustentabilidade Financeira Diálogos para o aperfeiçoamento daPolítica e do Sistema de Recursos Hídricos no Brasil. World Bank-Brasília, DF/Brasil. 7. investments and expenditures: a shaken wss 113 FIGURE 56 Number of interruptions in WSS public works and infrastructure upgrades, 2005-2015. Municipal Public Company 1 1 3 Private 5 6 10 CESBS 6 3 42 State 6 7 27 Municipalities 6 2 19 Concluded Normal Inadequate Source: FGV-CERI, 2017. FIGURE 57 Balance of Expenditures and Revenues in the WSS Sector in Brazil, 2010- 2023* 27.6 20.2 19.9 19.9 19.8 19.4 19.4 19.6 19.1 18.9 18.5 18.7 18.7 18.1 18.2 18.6 17.3 17.7 17.4 17.4 17.5 17.8 16.8 16.9 16.7 16.7 16.9 16.9 2.0 2. 1. 1.4 -0.4 -1.9 -1.8 -1.3 -1.8 -2.0 -2.6 -2.3 -3.0 - 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 Primary balance Net revenue Total expenditure * Projections for 2020 onwards. 2010 R$ billion. Source: TCU, 2020 and SIOP, 2019. 114 Private investment accounts for just a small portion of total investment in WSS. Brazil would do well to seek to expand the private role, in part by increasing the size of each private project: currently the investment range of WSS projects are R$27 (US$ 5.1 million) to R$121 million (US$ 23 million) per project (FGV-CERI 2017). Interruptions of WSS public works at the municipal level (Figure 56) pose a major obstacle to budget implementation.The delays in public works increase economic costs to municipalities, which have limited fiscal resources to cover these types of liabilities. The private sector is more agile in the execution of contracts compared to the public sector. The public sector reports only 16 percent completion of contracts signed six years ago. In contrast, private entities have reached 100 percent completion of contracts signed six or more years ago. Public contracts’ delays have worsened the imbalance of expenditures and revenues in Brazil’s WSS (Figure 57). Transparency of budget execution, openness of procurement, and efficiency of cash management are critical to the stability and predictability of investment and to reducing opportunities for rent seeking. The Ministry of Economy and the TCU should focus on protecting investment expenditures within ministries (MDR, MinSaude) and agencies (ANA). This could be accomplished by appropriating total project costs at the commencement of a project, preventing those budgets from being “raided” to meet unrealistic execution targets, and allowing some carry-over of unspent appropriations to future years. Greater transparency and accountability regarding budget management, monitoring, and evaluation would help as well. These conditions would strengthen incentives to deliver WSS projects on time and on budget and ensure value for money andintegrity in the use of public resources. Spending on public sector wages in Brazil is high (Figure 58) and divided roughly equally between federal, state, and municipal governments. States and municipalities oversee most expenditures in public safety, education, and health, and WSS, areas which generate most of their wage bills. High spending on labor is impeding local governments from increasing capital investments in critical sectors such as WSS. Reformsconducted in 2017 and consolidated in 2018 and 2019 sought to restore fiscal sustainability, by increasing employees’ social security contributions and private sector participation in selected state-owned enterprises and reducing tax exemptions. But over the last year, WSS SOEs have not yet seen financial returns on their balance sheets from those policies. Moreover, the COVID-19 pandemic heightened their financialstress due to contingent measures undertaken to ease the impact of the economic downturn. 7. investments and expenditures: a shaken wss 115 FIGURE 58 Public expenditure in LAC (left) and in Brazil (right) (average 2007-17) 14% 12.4% 9.3% 12% 10% 8% 7.4% 10% 8.4% 6% 5.3% 8% 5.7% 6% 4% 4% 2% 1.2% 2% 0% 0.7% 0% Current Health expenditure Compensation of employees Current education expenditure Current Health expenditure Public investment in Compensation of employees Current education expenditure Public investment in infrastructure infrastructure Note: Data for LAC includes weighted GDP for Argentina, Brazil, Chile, Colombia, Mexico, and Peru. Source: Research team’s own elaboration based on data from IMF Government Finance Statistics (Compensation of employees as % of GDP) and World Development Indicators (WDI). Average 2007-17 (circa). For infrastructure, INFRALATAM for public sector data 2007-2017. Includes telecommunications, water, energy, and transport. In addition to the wage policy reforms, higher quality of public spending is needed. This requires the institutionalization of a system of regular and rigorous policy monitoring andevaluation. Monitoring should provide focus on results and connect budgetary inputs to outcomes. Evaluation can provide evidence that points to the most efficient policy solutions that will achieve objectives. The evaluation of spending efficiency cannot be a one-time exercise. It must be become a routine, institutionalized process that guides policy making for the sector. In 2016, the execution rate of the WSS federal budget was 64 percent, or R$409 million(US$181 million) out of a budget of R$640 million (US$284 million).56 In 2020, the sectorhad actual over- execution of funds due to extraordinary measures taken against COVID-19—the sector’s overall budget was R$889 million (US$395 million) and execution reached R$1,000 million (US$444 million), a spending rate of 112.5 percent. 56 See data from (CGU, 2020) Controladoria Geral da Uniao at http://www.portaltransparencia.gov.br/ funcoes/17-saneamento?ano=2016 11 FIGURE 59 Wastewater treatment spending per capita compared to country’s GDP per capita, 2018 (left) and Distribution of coverage of sanitation by type of service in Brazil, 2019 (right) 100 1 0 7 7 10 0 0 Water treatment spending per capita USD 200 80 32 Switzerland 41 Population (%) 39 Japan 160 1 France U. E. 60 United Kingdom Netherlands 120 Singapore Saudi Arabia Germany Finland 40 South Korea 80 60 Poland Portugal Sweden 49 52 Taiwan 20 40 China Brazil Spain Italy India Indonesia Mexico Russia 0 0 1,000 4,000 16,000 64,000 National Rural* Urban Nominal GDP per capita USD (logarithmic scale) Limited service Safely managed Unimproved Basic service Open defecation Source: OECD 2018. Source: JMP 2020. Functional classifications of WSS into urban (functional classification 17) and rural (functional classification 511) can complicate data collection. These classifications often mix different types of programs within MDR (which handles urban WSS) and MinSaude (rural WSS). Hence it is difficult to distinguish programs that are earmarked from those that are planned within the budget cycle. Limited data are making it difficult to plan and allocate budgets to WSS subsectors, such as sanitation. Safely managed WSS is the highest standard for connectivity to waterand sewerage, according to the Joint Monitoring Program by UNICEF and WHO.57 But Brazil lacks comprehensive data on the quality of water that is supplied by the public network and consumed by the population (ANA 2019). Also, Brazil has limited information on safely managed categories of sanitation, particularly in rural areas (Figure 59). It is difficult to know exactly the coverage gaps and where corrective spending should go, particularly in rural areas. As a result, Brazil does not allocate federal and subnational budgets to the sanitation subsector, and per capita spending on wastewater treatment is relatively low with respect to Brazil’s per capita GDP (Figure 59). 57 See Joint Monitoring Program, UNICEF-WHO 2017. The full datasets for Brazil are available at https://washdata.org/data/downloads#BRA 7. investments and expenditures: a shaken wss 11 FIGURE 60 WSS budget allocations, commitments, and distribution Federal Spending on Basic Sanitation (2068), 2015-2019 Per Capita Spending on Basic Sanitation, 2019 2,000 60 1,10 1,500 Per capita spending 40 Millions (BRL) 959 1,000 743 1,16 706 20 500 964 681 0 668 Espirito Santo Alagoas Minas Gerais Amazonas São Paulo Pernambuco Maranhão Distrito Feral Rio Piauí Sergipe Santa Catarina Gran do SulRio Goiás Pará Paraiba Rondônia Matto Grosso Janeiro Tocantins Mato Grosso do Sul Ceará Amapá 557 Bahia 350 0 2015 2016 201 2018 2019 Disbursed Committed but not disbursed Basic Sanitation-Rural Basic Sanitation-Urban Note: real values adjusted with IPCA Source: DCA, 2019-Secretaria do Tesouro Nacional. Source: SIOPS. Data from Secretaria do Tesouro Nacional (DCA) and Sistema Integrado de Planejamento e Orçamento (SIOP) (Figure 60) bear out that limited information on sanitation leads to poor budget execution of the WSS sector. This results in large variations in per capita expenditures on sanitation, with zero sanitation spending in many rural areas. States including Bahia, Ceará, Piaui, Tocantins, Espírito Santo, and Minas Gerais, are among the few that allocate more than 40 percent of per capita WSS spending to rural WSS. Overall, the process of formulating and executing WSS federal budgets lacks consonance, compatibility, and integration between PPA, LDO, and LOA. Among the main challenges: (1) There is lack of consistency between what is planned in the PPA and the physical goals established in the LDOs. (2) Budgetary execution of WSS programs is distributed to many actors with different objectives that are not aligned to maximize the value of investments. (3) Many budgets of state companies above the allocation are authorized in the LOA andthis became higher during COVID-19. (4) There are divergences between the priority goals stipulated by the Government in theLDO and those effectively registered in the LOA. (5) Priority programs have not been implemented. 118 Contrary to common perceptions, the vast majority of WSS projects come in close to their estimated costs. It is just a few large projects that have large over-runs. These costblowouts usually relate to projects that are announced without enough planning and evaluation being done in advance. The bigger the project, the more common and larger the cost overruns. Common reasons cited are specification changes and problems associated with land acquisition and permits (ANA 2019). Finally, there are large variations in per capita expenditures on sanitation, with negligible sanitation spending in many rural areas. States like Bahia, Ceará, Piaui, Tocantins, Espírito Santo and Minas Gerais, are the few states with more than 40 percent ofper capita WSS spending allocated to the rural WSS subsector. In most states with federalbudget WSS programs cost overruns of capital investments projects are about one quarter above the estimated cost. Around the World, the affordability of WSS is feasible based thresholds on what households spend on these services with respect to their total consumption or expenditures. An “affordable” water service must be in the range of 2 to 3 percent of total household consumption or expenditures, and between 1 and 2 percent for sanitation services. Overall, a range of “affordability” for WSS is, hence, on average between 3 to 5 percent of total household expenditure. Both water and wastewater services in Brazil have affordability thresholds above the global thresholds, which means that these services are unaffordablefor the poorest. The poorest households that are in the lowest 10 percent of the income distribution spend between 5 and 7 percent of total expenditures on water and sanitation services. The subsidies help to make these services more affordable. However, the subsidiesare regressive and benefiting more the richest households, and water and sanitation tariffs are increasing over time. Subsidy reform and efficiency improvements of WSS services could then be pro-poor policies that would make these services affordable over time. 7. investments and expenditures: a shaken wss 11 8. Provisions of the New WSS Law, WSS Policies,and their Impacts 120 The new 2019 WSS law creates a long-needed opportunity to integrate macro-fiscal frameworks, bringing greater private capital into WSS, and embedding WSS investment planning in medium-term budgeting. This will help coordinate WSS public investment across levels of government and improve the standards for private sector participation in building WSS infrastructure. The legal framework will bring help to states at thelower end of the income scale, by setting quality standards and practices inrelation to fiscal discipline (including fiscal sustainability), improving management of PPPs contracts, and emphasizing project appraisal and selection and better monitoring of public assets. An important provision of the new law is development of a fiscal risk tool for public- private partnerships (PPPs). This will be standard to all types of contracts and layers of government and smooth the way for greater private capital. The tool’s aim is to (1) gather relevant PPP project data, (2) quantify the short and medium-term impact of a WSS PPP on the government’s deficitand debt under both cash- and accrual- based reporting standards, and (3) perform sensitivity analysis of the potential fiscal impact of a PPP.58 Another key provision is to encourage the drafting and implementation of PMSB in each municipality. These plans are elaborated studies of the municipality’s WSS Policy, bringing diagnosis, objectives, and goals for the universalization of WSS in the municipality. Overall, the new law can bring more consistency and certainty to development of WSS plans in areas with the largest coverage gaps. After December 31, 2019, existence of a plan became a necessary condition for accessing federal budgetary resources or financial resources managed or administered by an organ or entity of the federal public administration, when destined to basic sanitation services (§ 2 of Article 26 of Decree No. 7,217). Existence of a plan is also a condition for the validity of contracts for delegating the provision of water and sanitation services, according to item I of Article 39 of Decree no. 7,217. In Brazil, it is common that public policies are implemented in a two-step process. First, abroader instrument (such as the policy–Política) is created, to lay out the guidelines and understanding 58 Once project-specific and macroeconomic data are introduced, this tool generates standardized outcomes. These include (1) project cash flows over the whole life cycle. (2) fiscal tables and charts, on a cash and accrual basis— i.e., a government cash statement, income statement, and balance sheet, (3) debt sustainability analyses with and without the PPP project, and (4) sensitivity analyses of main fiscalaggregates to changes made within macroeconomic and project-specific parameters. See IMF 2015. 121 FIGURE 61 Implementation and legality of municipal plans Implementation Legal instrument used Does the municipality have a Basic Sanitation Policy? Was the Policy implemented via legal instrument? 5.60% 11.25 35.26% 53.49% 94.40% In progress Yes No Source: IBGE 2019. of the service that the population will receive. Secondly, a more technical instrument or plan is prepared. It lays out the targets and objectives for the service delivery. Federal Law 11445 of 2007 provides that a WSS Policy should be elaborated by a PMSB.Issues to settle in the plan include designation of a monitoring and oversight body to regulate the sector, definition of parameters for service delivery, definition of rights and obligations of users, establishment of mechanisms that allow civil society oversight, and establishment of information systems. Finally, the federal Law requires that a WSS Plan be reviewed every four years. As of 2017, less than 54 percent of municipalities in the country had implemented a WSSPolicy, while around 11 percent said that a policy was in progress, an IBGE survey found. Of the municipalities that had implemented a policy, almost 95 percent had done so via a legal instrument (decree, ordinance, or law) (Figure 61). The majority of PMSB were implemented after 2007 (Figure 62). While some municipalities had plans in place before the federal law, they did not explicitly conform to the components outlined in the legislation, and for all intents and purposes we exclude them from analysis of the efficacy of BSPs. The implementation of BSPs varied by municipality characteristics as well as by year. As seen in Figure 63, the year of implementation varied by the income of the municipality. 122 FIGURE 62 Years in which basic sanitation policy was created in municipality 300 Number of municipalities 200 100 0 1973 1991 2011 2017 1979 1997 1978 1970 2012 2013 1962 1993 1995 2015 1983 2014 1994 1999 2016 1966 1989 1998 2001 1968 1986 1990 2010 2007 1980 2002 2003 2005 2004 2006 2009 2008 2000 Source: Research team’s own elaboration based on SNIS 2019. FIGURE 63 Year of BSP creation by municipal GDP quintile, 2000 – 2017 300 Number of municipalities 200 100 0 2011 2012 2017 2013 2015 2014 2016 2001 2010 2007 2002 2003 2005 2004 2006 2009 2008 2000 1st quintile 2nd quintile 3rd quintile 4th quintile 5th quintile Source: Research team’s own elaboration based on SNIS 2019. 8. provisions of the new wss law, wss policies, and their impacts 123 I. Main features of the New Law and Recent Private Sector Projects in WSS The new regulatory framework for WSS recognizes that the public sector alone lacks themeans to fill in the sector’s increasing and large financial gaps. Its goal is to give the sector more clarity on the rules and open the way to greater investment from all sources and facilitate expansion of service scope and quality. But a key objective is to make the sector more attractive to private investment, which could mean big progress in reducing the financial burden of state-owned companies. A critical element of the new law is to eliminate the monopoly power of WSS SOEs. The law prohibits new “Program Contracts” (Contratos Programa) for the provision of water supply and sewage services.59 These contracts have long been used to let state-owned companies provide services without a bidding process. The new approach is a “business- to-business—B2B” model. Private companies will be allowed to provide water and sewage services on an equal footing with state-owned ones. The goal is that with competition quality standards will improve and WSS SOEs will be able to devote the limited public resources to improving their financial positions, efficiency, and equity (Figure 64). The new WSS law gives the national regulator (ANA) authority to standardize basic rules,60 which could make the use of public resources more efficient. ANA will determine regulatory guidelines to be followed by state and city regulatory agencies. Allocation of federal resources will be conditioned on compliance with the ANA guidelines,61 which has the potential to improve overall efficiency in spending and direct funding to prioritized areas. Certain mandatory provisions of the new WSS law could bolster equity in budget allocations to the sector. The concession regime explicitly promotes service expansion, reduction of non- revenue water, quality of service, efficiency, and rational use of water, electricity, and other natural resources. Those targets could help make the distribution of public resources more equitable in high-potential areas that have not necessarily benefited from federal and state transfers and subsidies. 59 The Sanitation Law allows existing Program Contracts to remain in force until their expiration if the state-owned companies are able to prove their economic and financial capability as service providersthrough either their own resources or third-party financing. The main criteria for remaining in force is the capability to bring universal access to water and sewage services in the relevant areas under the contract by December 31, 2033. 60 They include tariff revision criteria, indemnity, inspection, goals and contractual obligations, burdened transaction costs, and other factors that have discouraged the entry of private players into the sector. The rules are also intended to address citizen and labor union concerns that private WSS investors willcare only about profits and not service, and that to control that, private WSS will be subject to strong regulatory oversight. 61 The law states that operators that do not meet defined goals may be subject to sanctioning measures, including cancellation of the concession. Giving ANA authority to edit reference standardsalso strengthens the sector’s governance and regulation, providing important support to local and regional regulatory agencies in the improvement of operating contracts. 124 FIGURE 64 Business model for to private sector participation and stewardship of WSS SOEs Example of a B2B model PPP Public Private operator WSS SEO Fee Operational Spending Performance Efficiency Investments Bank (fiduciary agent) and operation Other private of the hired investors Cash management Remaining fee service scope Access and Bill payment Equity/Fairness Affordability Society Source: World Bank-IFC 2020. Furthermore, the allocation of resources will be subject to circular economy principles since the law intends to prioritize those projects that result in steady revenues (Werneck, Vieira, and Youseff 2020). The new law’s provisions concerning regionalization could help the equity of service delivery. The regionalized provisions provide for three types of groupings: (1) metropolitan region, urban cluster, or micro region: a unit created by each state under a law, by grouping adjacent municipalities; (2) regional unit of WSS: a unit created by each state under a law, by the grouping of municipalities that do not need to be contiguous, toadequately meet the requirements of hygiene and public health or to ensure economic and technical feasibility to less favored municipalities; and (3) reference block: grouping of municipalities that do not need to be contiguous, established by the federal government, and if the government does not, it can be formally created by means of associated management of service holders. Under these new provisions, a handful of projects are being structured as a fresh startfor better WSS service. In Rio de Janeiro, for instance, BNDES62 helped structure a concession contract under the new law with investments totaling R$31 billion (US$5.7 billion). These will ease the financial burden of the state by making public resources available for current and recurrent spending on WSS operations. 62 Brazilian development bank BNDES had already started to structure concession and privatization projects before the bill passed. The two most advanced projects are in the Metropolitan Region of Maceió, in the state of Alagoas, and in Rio de Janeiro. In addition, the bank has been structuring projects in the states of Acre, Amapá, Ceará, and Rio Grande do Sul. Together with the Alagoas and Rioprojects, these investments could total R$50 billion (US$9 billion). 8. provisions of the new wss law, wss policies, and their impacts 125 The beneficiary of these investments will be the Companhia Estadual de Águas e Esgotos do Rio de Janeiro (Cedae),63 which will remain responsible for the collection, treatment, and supply of drinking water for the private operator that will be chosen by auction. In Espírito Santo, a 30-year concession will require investments estimated at R$580 million ($104 million), which will benefit more than 400,000 people. The project was structured by national development bank BNDES and the state-owned sanitation company Companhia Espírito- Santense de Saneamento (CESAN). Alagoas has, meanwhile, started a public consultation to launch the first phase of a sanitation concession for 13 municipalities in the metropolitan region of Maceió. The project will require investments totaling about R$2.6 billion (US$624 million) with support from BNDES.64 In addition, the New Development Bank (NDB)65 approved in December 2020 a $164 million loan to an urban, rural, and social infrastructure program. The loan to the Far South Regional Development Bank (Banco Regional de Desenvolvimento do Extremo Sul, BRDE) is for on-lending to sustainable infrastructure in the states of Rio Grande do Sul, Paraná, and Santa Catarina. The goal is to help sub-national governments that have limited access to infrastructure financing. It is important to highlight, however, that WSS Policies of municipalities will be fundamental to guaranteeing the success of initiatives like these. II. Impact of the Implementation of WSS Policy on Health Indicators Brazil is a large middle-income country that has achieve relatively modest results with the provision of WSS services to its population. By law, WSS is a responsibility of municipalities, and in a country with 5,570 municipalities, the oversight of WSS service delivery at the municipal level has been a challenge. In 2007, the federal government implemented Law 11445, with the goal of standardizing guidelines for WSS service delivery across the subnational entities. 63 Cedae’s reach will include 13 municipalities in the Metropolitan Region of Rio de Janeiro, maintaining the operation of the major water systems: Guandu, Imunana-Laranjal, Lajes and Acari. The arrangement also includes four blocks of service delivery clustering multiple municipalities. 64 Sanitation has become a major focus for BNDES. In Brazil, approximately 74 million people (36 percent of the population) do not have access to sewage collection, and only 58 percent have water services with adequate frequency and quality. Brazil plans to announce in the next few months water and sewage PPPs for many other municipalities. 65 The New Development Bank is an international development bank established in 2014 by the BRICS countries (Brazil, Russia, India, China and South Africa) which aims to mobilize resources for infrastructure and sustainable development projects in BRICS and other emerging economies, as well as in developing countries. 12 FIGURE 65 FIGURE 66 Child mortality over time Share of avoidable deaths 0-5 years old Brazil .45 .5 Avoidable deaths – 0 to .4 .45 Child mortality .35 .4 .3 .35 .3 .25 2011 2017 2012 2013 2015 2014 2019 2016 2018 2001 2010 2007 2002 2003 2005 2004 2006 2009 2008 2011 2000 2017 2019 2012 2013 2015 2014 2016 2018 2001 2010 2007 2002 2003 2005 2004 2006 2009 2008 2000 Source: DataSus 2019. The Law set out clear principles and guidelines that had to be observed for elaboratingWSS policies at the municipal level. After Law 11445 was approved, there was a major change in the institutional framework in which municipalities were responsible for the provision of WSS services, which could impact in service delivery and quality and subsequently health indicators. Figure 65 and Figure 66 show time trends for two health indicators that are directly associated with WSS services: child mortality and share of avoidable deaths for children zero to five years old. For the country, the figures do not suggest a clear break in the trend of these two variables after Law 11445 was implemented. Thus, it is an unanswered question whether policy implementation had an impact on health outcomes. It is also possible to compare the trends for treated (municipalities adopting the law and Basic Sanitation Municipal Plans (BSMP)) and non- treated municipalities without those policies and plans in place. Treated and non-treated municipalities both experienced a decline in these two indicators since 2007. Figure 67 and Figure 68 show these time trends for the two groups of municipalities. It may be too soon to expect strong causation between WSS Plans and indicators: more than half of the municipalities in Brazil do not have or are only in the process of elaborating their plans. These plans can affect health indicators because they create a framework to manage and develop WSS services, as well as determine how budgets willbe executed. The national guidelines for sanitation in Brazil require all municipalities todevelop and implement a Municipal WSS Plan (MBSP) and a Municipal WSS Policy. MBSPs are important because they create the municipal framework to manage anddevelop the WSS services, as well as determining how budgets will be executed. In addition, MBSPs 8. provisions of the new wss law, wss policies, and their impacts 127 FIGURE 67 FIGURE 68 Treated and Non-treated before Treatment – Child Treated and Non-treated before Treatment – Share of Mortality Avoidable Deaths 0-5 Years .45 .55 (mean) sh_avoided_mort (mean) sh_child_mortality .5 .4 .45 .35 .4 .3 .35 .25 .3 -14 -9 -4 0 -14 -9 -4 0 Treated Non-treated Treated Non-treated Source: DataSus 2019. 128 present an opportunity to join technical decisions and economical and social viability on sanitation services. Overall, the regulation of 2007 conveyed a less-than-optimistic view of the efficacy of WSS policies, but the new law approved last year has more promising impacts for efficacy. After policy implementation, municipalities typically take on more in-house staff.But this increase in labor does not bring an increase in WSS services. The result is a decrease in service productivity and financial performance. One potential explanation is that WSS policies and plans do not have to include provisions for good financial governance. The implementation of the water and sanitation law will be challenging for the country. The standard of universalizing these services call for ramping up investments, improving efficiency of public budget execution, and making the sector financially sustainable and resilient. However, special attention must be placed on bolstering competition of service delivery contracts. The monopoly power of municipalities for signing contracts with State’s SOEs without a competitive bidding process generates inefficient resource allocation and inequity in the distribution of WSS services. Private sector participation itself would not guarantee better WSS outcomes. Nevertheless, it can promote better practices for higher quality of service delivery, subject to adequate monitoring and supervision from municipalities. In sum, the study makes the following conclusions about how to leverage the new WSS law to make public WSS expenditures more efficient and effective: • The role of the regulator under the new law will be paramount to reducing inequitiesin investments and performance and improving budget efficiency. • Emerging threats to municipal liquidity are crimping funding for WSS projects in alltypes of municipalities, whether good or bad performers. • Limited reporting of SNIS data by service providers in many municipalities is limitingdecision making in the sector. MDR, FUNASA, and ANA hold key stewardship roles toprovide stronger incentives to increase data reporting to SNIS. 8. provisions of the new wss law, wss policies, and their impacts 12 9. COVID-19, Public Expenditures, and Health Burdens from Limited WSS Access 130 The COVID-19 pandemic highlights the urgency of establishing public policies toreduce social inequalities and provide access to basic services that guarantee quality of life and health. During the pandemic, WSS service providers faced the challenge of maintaining the drinking water supply and sanitation services despite heightened financial and operational pressure on them. At the start of the pandemic, the sector’s players (municipalities, service providers, and regulators) immediately took steps aimed at protecting both utilities’ employees and customers, especially the most financially vulnerable (Capodeferro and Smiderle 2020). These steps included (1) changes in the period of billing, (2) restrictions on service requested by customers, (3) postponement of tariff rises, (4) payment waivers of water and sewage tariffs; (5) tariff exemptions for vulnerable users, and (6) suspension of legal action against non-payment (Table 3 and Figure 69). Brazil showed high debt and limited fiscal space to further support WSS utilities during the COVID-19, with a large share of fiscal support during 2020 as a percentage of GDP (Figure 70). TABLE 13 Cases, deaths during the COVID-19 pandemic COVID-19 cases, deaths, and mortality rates in Latin America and the Caribbean Country Cases (thousands) Deaths Deaths per 100,000 Brazil 6,927 181,835 86.81 Mexico 1,256 114,298 90.58 Argentina 1,503 41,041 92.24 Colombia 1,435 39,195 78.94 Peru 985 36,677 114.65 Chile 574 15,931 85.06 Ecuador 202 13,875 81.21 Bolivia 147 9,024 79.48 Guatemala 129 4,445 25.77 Panama 195 3,382 80.97 Honduras 115 2,989 31.18 Rep. Dom. 155 2,364 22.24 Total LAC 14,103 472,801 - Source: Johns Hopkins August. 2020. 13 FIGURE 69 Relationship between support provided to COVID, fiscal space and debt* (left) and measures adopted by 26WSS state companies (SCs) for maintaining service during the pandemic (right) GDP % 0 2 4 6 8 00 88% 88% Argentina 6 62% % SC Pakistan 2 Brazil 14.6 Egypt 1.7 75 India 7 19% 15% 12% South Africa 9.9 Colombia 5.3 Mexico 1.1 Debt as percentage of GDP 50 China 5.9 Tailandia 12.5 Poland 11.7 Romania 5.4 Turkey 13.8 Philippines 3.4 Restriction of presential service Peru 15.7 and availability of virtual channels Tariff exemption for vulnerable users Postponement of Easing payment of United Arab tariff adjustments Change in the method of billing against non payment* water and sewage bills Suspension of procedures 25 Emirates 2.1 Saudi Chile 10.7 Indonesia 3.8 Arabia 3.1 Kazakhstan 4.9 Russia 3.4 0 -4 -2 0 2 Primary fiscal balance (2019) Measures for Measures for employee protection users’ protection * Size of bubbles indicate the total support for COVID-19 fiscal (direct) and credit as percentage of GDP. Source: Our World in Data 2020, and Capodeferro and Smiderle 2020. It is no secret that the COVID-19 pandemic and measures against it badly hurt the WSS sector (World Bank 2020). Revenue shortfalls from a decline in commercial and industrialwater use and non-payment of bills by struggling residential customers hobbled WSS utilities across the country. Revenue concerns are spurring utilities to find ways to offsetthe shortfalls. Dealing with the persistent problem of non-revenue water (NRW) is a good place to start. During the first six months of the pandemic, the Ministério do Desenvolvimento Regional (MDR) announced suspension of payments by heavily indebted companies and according to the Fundo de Garantia do Tempo de Serviço (FGTS)66. In addition, the National Water and Sanitation Agency (ANA) Resolution 21/2020 provided an extension to use grants for the national water public works until the end of 2020. These measures sought to minimize economic impact in the productive sectors due to the pandemic. But they created new financial pressure on the WSS SOEs. 66 The Fundo de Garantia do Tempo de Serviço (FGTS) or Severance Pay Fund was created with the objective of protecting employees dismissed without good cause by the employer, by opening an account linked to the employment contract. At the beginning of each month, employers deposit in accounts opened at CAIXA commercial bank, in the name of employees, the amount corresponding to 8 percent of each employee’s salary. Thus, the FGTS is constituted by the total of these monthly deposits and the amounts belong to employees who, in some situations, may get the total amount deposited in their names. 9. covid-19, public expenditures, and health burdens from limited wss access 132 I. Emergency Fiscal Measures and Budgetary Allocations The government enacted COVID-related measures for budgetary reallocation to the WSSsector but these were insufficient because many were zeroed out by exemptions. Underthe federal Legislative Decree of “public calamity,” No. 06/2020,67 financial measures related to combating the coronavirus included specific exemptions. The main exemptionswere for spending that (1) will not affect the goals of fiscal results provided for in the LDO, (2) did not have to be offset through reduction of other expenses or the creation or increase of taxes or a source of collection, and (3) included estimates of the budgetary-financial impact and a statement by the originator of the expenditure that it has budgetary and financial adequacy with the annual budget law and compatibility with themultiannual plan and with the law of budget guidelines. Together, these measures allowed Brazil to deliver high levels of fiscal and liquidity support (Figure 70) but without the flexibility in budgetary reallocations necessary to make use of these financial resources more effective in achieving short-run outcomes. Aproblem for WSS in the special budgetary provisions is the rules that prohibit borrowing for the payment of current expenses – even during the financial year of the officially declared public calamity. The condition of “public calamity” does allow credit operations carried out for the refinancing of securities debt to be used for the payment of interest and charges. It waives compliance with constitutional and legal restrictions on the creation, expansion, or improvement of government action that leads to increased spending and the granting or expansion of a tax incentive or benefit that results in a waiver of revenue—as long as itis not an expense that has the sole purpose of facing the calamity and its social and economic effects. The law says that it must not imply a permanent expense, but it must have the sole purpose of addressing the calamity and its social and economic consequences. It is intended, therefore, to allow the creation of a more flexible “parallel budget” that facilitates hiring, an increase in public expenses and indebtedness,all in a more transparent and legally secure way to allow expenditure control. Under the new rules, it will no longer be necessary to indicate the source of financing for an expense, which may be covered by the issuance of public debt. 67 The purpose of the law was to obtain the benefits of Article 65 of the Fiscal Responsibility Law, namely, the exemption from achieving the tax results provided for in the LDO and the suspension of the commitment limitation mechanism. Also, Art. 3 of the Law clarifies that it must not imply a permanent expense, but it must have the sole purpose of addressing the calamity and its social and economic consequences. 40 30 20 45 1 5 2 5 3 Source: Monitoring Responses in Latin America and the Caribbean, World Bank Global Water Practice, 2020. 5 1 Fiscal support measures during the first 9 months of the COVID-19 pandemic (% GDP) FIGURE 70 Brazilian states that enacted financial support measures across payments and tariffs of WSS services, 2020 FIGURE 71 Source: IMF fiscal monitor database, 2020. Uganda Mexico States Egypt Vietnam 18 Mexico Niger Gabon Ethiopia Jordan Waive Payments by Special Groups Tunisia Zambia Namibia Moldova Albania Armenia Saudi Arabia Ghana Morocco Philippines Russia Senegal Indonesia Tajikistan Georgia Belize Fiscal support Honduras Kazakhstan Barbados Mali 9. covid-19, public expenditures, and health burdens from limited wss Colombia Romania China Argentina Uzbekistan Bahrain Argentina Bulgaria Azerbaijan India Serbia Ireland Estonia Malaysia Lithuania Greece Liquidity support (excl. Central Bank) Hungary Iceland The Netherlands Cyprus States Finland South Africa 8 Israel Chile Mongolia Portugal Sweden Chile Austria Freeze Tariff Adjustments Switzerland Luxembourg Latvia Malta Norway Hong Kong SAR Poland Thailand Australia Turkey Korea United States Slovenia Brazil Denmark Peru Brazil Bolivia Belgium Canada Spain Czech Republic Singapore France New Zealand United Kingdom Mauritius Japan Italy Germany 13 134 Concerning coordination across sectors, the national health contingency plan does not mention water and sanitation as key elements of the response. During the pandemic, states, their health institutions, and their water utilities seem to have been acting independently from each other, with limited general guidance from the national government.Gains in efficiency and equity of public spending are best achieved with integrated policies between health and other sectors, including WSS. At the municipal level, contingency plans did not include actions related to water and sanitation services (Abraham 2020). The key measures to support WSS SOEs and consumers during the COVID-19 pandemic such as deferring water bills and freezing tariff rises, took place in only a handful of states (Figure 71) and without clear federal guidelines on adjusting financial imbalances through contingent federal and state budgets. Most northern states did not fully implement these measures. There were no amendments to the Annual Budget Law to provide flexibility for progressively reducing financial stress of the WSS SOEs. It appears there was no provision in the Annual Budget Law for the commitment of public financial resources to the granting of these subsidies (Capodeferro and Smiderle 2020). 135 In practice, the lack of financial support for WSS SOEs’ balance sheets have resulted in higher burden to taxpayers during the pandemic. Taxes are financing WSS tariff subsidies (Narzetti and Cunha 2020), but many households are not benefiting from the subsidies because they do not use WSS services. The effect has been to preserve and, in some cases,increase financial deficits in the WSS sector. II. Burden of Disease: Economic Costsof Lack of WSS Coverage and Gains from Expanding It Better effectiveness and efficiency of public expenditures in WSS in Brazil are critical toavoiding extra costs to the economy. Based on the Global Burden of Disease, the estimated days of productivity lost per year due to lack of WSS services reached 125,00068 due to resulting diarrheal disease (Figure 72). The DALY (Disability Adjusted Life-Years) indicator offers another insight into the economic burden of disease in Brazil. By calculating number of years of productive life lost due to unimproved WASH services andGDP per capita for the year 2019, it is possible to assign an economic value to those years.In Brazil, the economic burden of disease of unimproved WSS or lack of access to it altogether can be estimated by state, allowing comparisons. The jurisdiction with the lowest DALY69 lost is Distrito Federal, with 360 days per year.The state with the highest DALY loss per year is Santa Catarina, at 18,300 days. These figures underline the massive inequities in WSS access in Brazil. In 2019, the economy of Brazil lost close to US$ 1.3 billion due to unimproved WASH, based on estimates using the Global Burden of Disease data70. States including Pará, Pernambuco, Ceará, Maranhão, Santa Catarina, and Bahia each accrued economic burdensper year of more than US$100 million. Other states, including São Paulo, Tocantins, Amapá, Roraima, and Distrito Federal, had economic burdens of US$15 million or less during the year. Another way of measuring the burden comes from the Instituto Trata Brasil, which has reported that in 2018 Brazil registered 233,880 hospitalizations for diseases caused by precarious WSS and 2,180 deaths. People who live with a lack of WSS are vulnerable to multiple diseases (ITB 2019a). 68 For more details about the cost to the health system in Brazil, see BRK Ambiental 2021: https://blog.brkambiental.com.br/saneamento-basico-e-saude-publica/ 69 For methodological insights of the burden of disease, DALY, YLL and other related indicators, seeWagner et al. (2015) https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4689490/ 70 A methodological note was developed for those estimations based on the GBD for Brazil. 136 FIGURE 72 Number DALY’s Lost and Deaths due to Diarrheal Diseases Attributable to Unimproved WASH by Federation Unit Deaths due to Diarrheal Diseases DALY’s Lost due to Diarrheal Diseases Attributable to Unimproved WASH Attributable to Unimproved WASH DALY’s Deaths 10,080.4 - 18,333.3 111.6 - 278.7 5,038.9 - 10,080.4 60.3 - 111.6 2,527.8 - 5,038.9 28.6 - 60.3 1,662.8 - 2,527.8 15.8 - 28.6 359.4 - 1,662.8 3.6 - 15.8 Source: Estimated with data from GBD IHME 2019, PNAD 2015, and PNDS 2006. States’ and municipalities’ influence over health care, as well as the many social and economic, behavioral, and environmental factors that shape health, is most apparent in their budget decisions. Preventing increasing health costs is an intrinsic objective of public expenditures. With the COVID-19 pandemic, health spending expanded to make up more than 15 percent of state and local expenditures. However, the limited nature of WSSservices produced extra costs for health care systems locally and nationally. Hence improving budgetary efficiency could become an effective instrument for improving Brazilians’ health and quality of life. Two critical ways to reduce the economic burden of WSS diseases are to reduce WSS service interruptions and expand coverage to areas that have no service. Concerning the first, interruptions, the PLANSAB review of 2019 found that the North, Northeast, and Midwest regions were on trends for meeting the 2033 targets in terms of lowering service interruptions. However, it is possible that data fragility, as discussed early, has contributed to painting a more promising picture in those macro-regions than is fact. The South, on the other hand, remained clearly further from the target, with 6.4 million peopleand businesses affected by service stoppages or interruptions (MDR 2018). 9. covid-19, public expenditures, and health burdens from limited wss 137 FIGURE 73 WSS Investments in Billion US$ (2018 prices) 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 2011 2012 2013 2014 2015 2016 201 2018 National 100 largest cities Source: Instituto Trata Brazil 2019. Concerning the second way, expansion of service, investments from public resources declined between 2011 and 2018 due to a higher proportion of funds going to the pension and severance costs of WSS SOEs (Figure 73). Most major cities in the country have a low level of reinvestment in the WSS sector. Of the amount they collect, only a small portion goes toward making improvements in the services and expanding them to new areas. Among the 100 largest cities in the country, 70 invest less than 30 percent of what they collect in the sector itself. In this environment, Brazil is increasingly turning to private money to fill the gap. The new legal framework for WSS is providing extra incentives for private investment to expand coverage and improve service delivery. With the new legal framework, the objective of universalization will encompass integrated management and targets for service delivery improvements. Health and environmental co- benefits are to be considered when crafting budget allocations and investments in the sector. The expansion of sewage treatment services, for instance, will reduce natural waste in basins, rivers, streams, and seas, aiding in targets that all Brazilian regions are attempting to reach by 2033. The key challenge would be to provide the WSS sector more tools and capacitiesnot only to implement the new WSS law but also to bolster coordination and engagement with the health sector. The new WSS legal framework gives ANA attributions to standardized guidelines at the subnational level and homogenize regulatory agencies’ capacities. This new stewardship roles could also open doors to closer engagements with local health institutions, strategies, and programs. A close coordination and engagement between WSS and health, at various levels of government, could improve efficiency and effectiveness to reach public health outcomes. 138 Universalization of water and sewage services will reduce Brazil’s annual health costs by up to R$1.45 billion (US$270 million), data from the National Confederation of Industry (CNI, 2017) suggest. In the economic field, for each R$1 allocated to the sector, R$2.50 is generated in the associated chains, and for each R$1 billion (US$180 million), 60,000 jobs are generated (MDR 2020). The gains go beyond lower health costs, however, to include such measures as lower environmental clean-up costs. Hutton (2012) estimates a total return of US$8.93 in Brazil for every US$1.00 invested in expanding WSS coverage towards universalization. That is more than double the World Health Organization’s estimate for the global ratio. Reducing health costs through WSS investments is a smart strategy to make the best use of public resources. Considering all public health benefits (beyond the burden of disease avoided through WSS investments) and productivity gains, and environmental externalities, shows that the WSS sector is essential for more resilient and inclusive growth and poverty reduction. As WSS investments decline over time, become less effective, and are not used to ramp up coverage rates, the health care costs and productivity losses due to WSS disease will increase over time. More efficient investments in WSS will help build health sector’s resilience against future diseases: Safely managed WSS services are essential to support municipalities with lowest WSS implementation capacity with important outcomes of reducing the incidence of diseases that afflict the poor, predominantly in urban areas of Brazil. 9. covid-19, public expenditures, and health burdens from limited wss 10. Conclusions and Policy Recommendations 140 Brazil has still to adopt measures to improve budget execution performance. The WSS sector still faces multiple institutional issues that constrain effectiveness and efficiency of budget execution. All over the world, budgets are never big enough, and Brazil is no different. But the country’s three tiers of government, bringing on with rigidity, fiduciarycontrol, poor governance, and excess bureaucracy, lead to weak disbursements and investment quality in the water sector. Procurement is another process that delays project implementation. In addition, leadership and attracting skilled human resources to the sector can hobble performance of public financial management systems. Achieving universal access to water and sanitation services will require huge investments. To meet the targets, the volume of investment needs to increase substantially in the coming years. Efficient public financial management will be fundamental to opening bottlenecks. These include overlapping roles of institutions, competing funding, conducting of projects without appropriate planning and design, and lack of integration with other sectors. Poor planning and alignment of the public policies, and lack of a monitoring and evaluation framework to analyze each step of budget execution, from preparation to completion of a project represent opportunities for reform. Fragmentation across programs, levels of government, and budget and non-budget sources results in efficiency losses due to overlaps and lack of coordination. Spending inprograms often is co- financed by various on-and-off funding sources, without much being coordinated. This fragmentation makes effective monitoring and policy evaluation difficultand, in many cases, results in inefficiency and overlap. Projects should be framed with a multi-year horizon, sequenced around agreed priorities, and built upon a coordinated approach both between the multiple levels of government and within them. Furthermore, governance practices, including transparency and accountability, need attention. WSS institutions need strengthening to deal with discretion volatility, as more checks and balances make it harder for governments to change fiscal policy for reasons unrelated to the current state of the economy. 141 Here are some core findings of this study: • Brazil is a great paradox in the field of water access: it has 12 percent of the fresh surface water reserves of the planet, yet millions of people in its cities can be unable to obtain clean water. • Coverage of the piped water network has barely changed in the past decade. Sewage network coverage has achieved laudable but far from adequate gains. • Major inequities persist in the WSS sector, with city people having far higher access to service and investment budgets than do rural people. There are similar inequities between states and regions. • PFM environment in Brazil is not favorable for public sector investments in WSS, over the upcoming years. In 2021, the Public Financial Management (PFM) in Brazil, will be operating in a narrowed economic environment that could impact the budgetary outcomes, such as the (i) aggregate fiscal discipline; (ii) strategic allocation of resources; and (iii) efficient service delivery. • Despite these considerable challenges, capital spending on WSS has been declining inabsolute terms and as a portion of GDP. • WSS capital programs often face delays in spending funds that have been authorized.These and other investment inefficiencies are declining, though not fast enough. 10. conclusions and policy recommendations 142 • Brazil is struggling to do more with less, but its declining budget allocations and high severance and payroll spending for WSS employees create obstacles to higher investment. • Private capital can bring new efficiencies and service quality to WSS, offsetting declines in public funding, but parties that might want to invest face numerous barriers in the regulatory framework. • Local strategic plans for WSS have shown limited capacity to improve local outcomes, budget execution efficiency, and equity in the allocation of investments. • Data on municipal WSS expenditures are scarce, often unreliable, and available only through complex information systems. • COVID-19 is causing new pressure on WSS finances by raising contingent expenditures and further limiting capital investments. • If current policies and investment rates continue, Brazil stands to miss by 20 years its target to achieve universal WSS by 2033. • To put WSS on track, the sector’s budget needs a new and robust investment framework to promote project innovation, and a simplified budget framework with explicit criteria to invest in subsectors that bring the most cost-effective solutions. References 144 Abraham, M. 2020. Orçamento de guerra contra a Covid-19. Revista Gen Juridico. http://genjuridico.com.br/2020/04/16/orcamento-de-guerra-covid-19/ Albuquerque, B. 2011. “Fiscal institutions and public spending volatility in Europe.” Economic Modelling Volume 28, Issue 6, Pages 2544-2559. Elsevier. https://www.sciencedirect.com/science/article/pii/S0264999311001891 Almeida, M., 2011. Nota técnica: restos a pagar e artifícios contábeis. Brasília, IPEA. Disponível em: https://www.mansueto.files.wordpress.com/2011/02/nota-rap-20111.pdf ANA. 2019. “SDG 6 in Brazil. ANA’s vision of the Indicators.” National Water Agency, Ministryof Regional Development. https://www.ana.gov.br/noticias/ods-6-ana-lanca-publicacao-inedita-para- acompanhar-metas-para-agua-e-saneamento-ate-2030-1/ods6_en_digital.pdf Anillo, A., F. Boehm, and J. Polo-Otero. 2014. “Governance Matters: Universal Access to Water.” https://ssrn.com/abstract=2393718 Ardanaz, M. and Izquierdo, A. 2017. Current Expenditure Upswings in Good Times and Capital Expenditure Downswings in Bad Times?: New Evidence from DevelopingCountries. IDB Working Paper No. IDB-WP-838. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3103799 Arezki, R. and Ismail, K. 2013. Boom–bust cycle, asymmetrical fiscal response and theDutch disease. Journal of Development Economics, 101, pp.256-267. Armelin, P. and Rabioglio, R. 2019. “Improving Living Standards in Brazil ThroughWater and Sanitation Infrastructure Development. The Role of the World Bank.” Harvard Kennedy School. Cambridge, MA. Atlas Brasil. 2018. Atlas do Desenvolvimento Humano no Brasil. PNUD Brasil, Ipea e FJP, 2020. http://www.atlasbrasil.org.br/consulta Augustinho, S.M., de Oliveira, A.G. and de Lima, I.A. 2013. A “contabilidade criativa” e a inscrição de restos a pagar como “expediente” para alcançar o superávit primário.REUNIR Revista de Administração Contabilidade e Sustentabilidade, 3(4), pp.127-147. https://reunir.revistas.ufcg.edu.br/index.php/uacc/article/view/167 Banco de Nordeste. 2018. Saneamento Básico. Situacao Atual e Perspectivas. Caderno Setorial ETENE. Number 3. Issue 37th. Escritório Técnico de Estudos Econômicos doNordeste. Blancas, L., Chioda, L., Cordella, T., Borges de Oliveira, A. and Várdy, F. 2011. “Do Procurement Rules Impact Infrastructure Investment Efficiency? An Empirical Analysis of Inversão das Fases in São Paulo State.” Policy Research Working Paper5528. Washington, DC: World Bank. BNDES. 2016. Saneamento básico no Brasil: desafios e perspectivas. BNDES. https://www.bndes.gov.br/wps/portal/site/home/conhecimento/noticias/noticia/ saneamento-basico BOOST. 2002-2018. Public Expenditure Database-Brazil, World Bank Group. https://datacatalog.worldbank.org/dataset/brazil-boost-public-expenditure-database Bose, N., M.E. Haque, and D.R. Osborn. 2007. “Public Expenditure and Economic Growth: A Disaggregated Analysis for Developing Countries,” The Manchester School, 75(5), DOI <10.1111/j.1467-9957.2007.01028.x> 145 BRK Ambiental 2021. “Os Impactos da Relação entre Saneamento Básico e Saúde Pública”. Saneamento Sociedade e Meio Ambiente. Saneamento em Pauta. BKR ambiental, Brasil. https://blog.brkambiental.com.br/saneamento-basico-e-saude-publica/ Capodeferro, M.W., and J.J. Smiderle. 2020. “The Brazilian sanitation sector’s response to COVID- 19.” Revista de Administração Pública, 54(4), 1022-1036. Epub August 28, 2020. https://dx.doi.org/10.1590/0034-761220200324 Cavalcanti, A., A. Teixeira, and K. Pontes. 2020. “Evaluation of the Efficiency of WSS Integrated Management in Brazilian Municipalities.” International Journal of Environmental Research and Public Health, 17(24), 9244. https://doi.org/10.3390/ijerph17249244 Cavallo, E.A., Powell, A. and Serebrisky, T. 2020. From Structures to Services: The Path to Better Infrastructure in Latin America and the Caribbean. IDB Publications (Books). https://ideas.repec.org/b/idb/idbbks/10460.html Congressional Budget Office. 2011. “CBO’s 2011 Long-Term Budget Outlook.” Washington DC. https://www.cbo.gov/sites/default/files/112th-congress-2011-2012/reports/06-21- Long- Term_Budget_Outlook.pdf Congressional Research Service. 2020. “Latin America and the Caribbean: Impact ofCOVID- 19.” Washington DC. https://fas.org/sgp/crs/row/IF11581.pdf CGU (Controladoria-Geral da União). 2016. Portal da Transparência - Saneamento. http://www.portaltransparencia.gov.br/funcoes/17-saneamento?ano=2016 CNI (Confederação Nacional do Indústria). 2017. ‘O financiamento do investimento em infraestrutura no Brasil’. As of 3 March 2017: http://www.portaldaindustria.com.br/publicacoes/2016/7/ofinanciamento-do- investimento-em-infraestrutura-no-brasil-uma-agenda-para-sua-expansaosustentada/ Da Motta, R. and Moreira, A. (2006) Efficiency and regulation in the sanitation sector inBrazil. Utilities Policy. Volume 14, Issue 3, September 2006, Pages 185-195. https://www.sciencedirect.com/science/article/pii/S0957178706000154 DATASUS. 20148. Trata Brasil e CEBDS destacam benefícios com expansão do saneamento. Disponível em http://datasus.saude.gov.br/noticias/atualizacoes/391-trata-brasil-e-cebds- destacam-beneficios-comexpansao-do-saneamento DATASUS. 2019. Informações de Saúde (TABNET) – Estatísticas Vitais. Disponível em http://www2.datasus.gov.br/DATASUS/index.php?area=0205&id=6937 DCA/Finbra. 2013-2019. Declaração das Contas Anuais. Siconfi – Secretaria do Tesouro Nacional (STN). https: siconfi.tesouro.gov.br/siconfi/index.jsf;jsessionid=faXkYCagxqbkD7Z96b2fFACV.node1 Deloitte. 2017. Water Tight 2.0: The Top Trends in the Global Water Sector. https: www2.deloitte.com/content/dam/Deloitte/pl/Documents/Reports/pl_Water-Tight-2- 0-The-top-trends-in-the-global-water-sector.pdf Dos Santos, A.P., and K.P. De Castro. 2018. Local Governments’ Tax Burden in Brazil: Evolution and Characteristics. Taxes and Taxation Trends. Edited by Jolanta Iwin-Garzyska. DOI: 10.5772/intechopen.74808 Estache, A., G. Garsous, and R. Seroa da Motta. 2016. “Shared Mandates, Moral Hazard, and Political (Mis)alignment in a Decentralized Economy,” World Development, Elsevier, vol. 83(C), pages 98-110. 146 Ferraz and, C., and F. Finan. 2008. “Exposing Corrupt Politicians: The Effects of Brazil’s Publicly Released Audits on Electoral Outcomes.” Quarterly Journal of Economics123 (2): 703–45. Ferro, G., E.J. Lentini, A.C. Mercadier, and C.A. Romero. 2014. “Efficiency in Brazil’s water and sanitation sector and its relationship with regional provision, property and the independence of operators.” Utilities Policy, Volume 28, Pages 42-51, ISSN 0957-1787, https://doi.org/10.1016/j.jup.2013.12.001. FGV-EBAPE 2017. “Unpaid commitments and loss of budgetary credibility.” Brazilian Journalof Public Administration. Revista de Administracao Publica Rio de Janeiro 51(4):580-595, July - Aug. FGV-CERI. 2016. Efetividade dos Investimentos em Saneamento no Brasil: da disponibilidade dos recursos financeiros, até a implementação do sistemas de abastecimento de água e esgotamento sanitário. Centro de Estudos de Regulaçãode Infraestrutura--Fundação Getulio Vargas. Rio de Janeiro. Brasil. Global Burden of Disease (GBD). 2019. Institute for Health, Metrics and Evaluation. Seattle, WA. http://www.healthdata.org/gbd/2019 Gomes, F., et al. 2020. “Access to drinking water and sewage treatment in Brazil: A challenge for the control of waterborne infectious diseases.” Revista do Instituto deMedicina Tropical de São Paulo, 62, e71. https://doi.org/10.1590/S1678-9946202062071 Gontijo, V. 2012. Instrumentos de planejamento e orÇamento. Orçamento da União -Câmara dos Deputados. https://www2.camara.leg.br/orcamento-da-uniao/cidadao/entenda/cursopo/planejamento GWSP. 2019. Connecting the Drops : Global Water Security and Sanitation Partnership Annual Report 2019 (English). Washington, D.C. : World Bank Group. http://documents.worldbank.org/curated/en/997021571169834156/Connecting-the- Drops-Global-Water-Security-and-Sanitation-Partnership-Annual-Report-2019 Ho, A.T.-K., de Jong, M., & Zhao, Z. (Eds.). (2019). Performance Budgeting Reform: Theories and International Practices (1st ed.). Routledge. https://doi.org/10.4324/9781351055307 Hutchings, P., et al. 2015. “A Systematic Review of Success Factors in the Community Management of Rural Water Supplies over the Past 30 Years,” Water Policy, 17(5), DOI <10.2166/wp.2015.128> Hutton, G. 2012. “Global Costs and Benefits of Drinking-Water Supply and Sanitation Interventions to Reach the MDG Target and Universal Coverage.” World Health Organization, Geneva, Switzerland https://www.who.int/water_sanitation_health/publications/2012/globalcosts.pdf IBGE. 2020. “Metropolitan Regions, Urban Agglomerations, and Integrated regions,” byWilliam Ferreira. IBGE. 2017-2020. Estatísticas. Disponível em https://www.ibge.gov.br/estatisticas/todos-os-produtos-estatisticas.html IBGE. 2018. “Munic: More than half of Brazilian municipalities did not have any basic sanitation plan in 2017.” Agência IBGE Notícias - Social Statistics. https://agenciadenoticias.ibge.gov.br/en/agencia-press-room/2185-news-agency/ releases- en/22616-munic-more-than-half-of-brazilian-municipalities-did-not-have- any-basic- sanitation-plan-in-2017 reference 147 IB-NET. 2020. “The International Benchmarking Network of Water and Sanitation Utilities.” Washington, DC: World Bank. https://www.ib-net.org/about-us/publications/ IFGF (Índice Firjan de Gestão Fiscal). 2017. Recorte Municipal Abragencia Nacional. Pesquisas de Estudos Socioeconomicos. FIRJAN. Brasilia. https://www.firjan.com.br/data/files/3F/22/D1/8B/7AF3E5107210A3E5A8A809C2/ IFGF%20-%20English%20Version.pdf IFDM (Indice FIRJAN de Desenvolvimento Municipal). 2005-2016. Consulta ao Indice- Pesquisa-Downloads. https://www.firjan.com.br/ifdm/downloads/ IFGF (Índice Firjan de Gestão Fiscal). 2017. Recorte Municipal Abragencia Nacional. Pesquisas de Estudos Socioeconomicos. FIRJAN. Brasilia. https://www.firjan.com.br/data/files/3F/22/D1/8B/7AF3E5107210A3E5A8A809C2/ IFGF%20-%20English%20Version.pdf IFGF (Índice Firjan de Gestão Fiscal). 2019. Um Raio-X Na Gestão Pública dos Municipios Brasileiros. FIRJAN. Brasilia. https://www.firjan.com.br/data/files/8F/50/19/81/B2E1E610B71B21E6A8A809C2/ IFGF- 2019_estudo-completo.pdf IMF. 2000. Equity and Efficiency in the Reform of Price Subsidies. Washington D.C. https://www.imf.org/external/pubs/ft/equity/index.htm Inter-American Development Bank. 2018. Development in the Americas: Better Spending forBetter Lives. Washington D.C.: Inter-American Development Bank. IMF. 2000. Equity and Efficiency in the Reform of Price Subsidies. Washington D.C. https://www.imf.org/external/pubs/ft/equity/index.htm IMF. 2015. Making Public Investment More Efficient. PP4959. Washington D.C. https://www.imf.org/en/Publications/Policy-Papers/Issues/2016/12/31/Making-Public- Investment-More-Efficient-PP4959 IMF. 2016a. Public Investment – The Perils of Premature Election Commitments. By J. Harris. PFM-IMF. https://blog-pfm.imf.org/pfmblog/2016/11/public-investment-the-perils-of- premature- election-commitments.html IMF. 2016b. “Fiscal Multipliers for Brazil.” T. Matheson and J. Pereira. IMF: Washington, DC. https://www.imf.org/en/Publications/WP/Issues/2016/12/31/Fiscal-Multipliers-for- Brazil- 43831 IMF. 2020. Fiscal Monitor. https://www.imf.org/en/Publications/FM/Issues/2020/04/06/ fiscal- monitor-april-2020 Instituto Trata Brasil (ITB). 2019. Paneil de Saneamento do Brasil. https://www.painelsaneamento.org.br/explore/ano?SE%5Ba%5D=2018 Instituto Trata Brasil (ITB). 2019ª. Perdas de água. Desafios para disponibilidade hidrica eavanco da eficiencia do saneamento básico. http://www.tratabrasil.org.br/images/Release_Perdas_de_%C3%81gua.pdf Instituto Trata Brasil (ITB). 2018. Paneil de Saneamento Brazil. https://www.painelsaneamento.org.br/explore/ano?SE%5Ba%5D=2017&SE%5Bo%5D=a Instituto Trata Brasil (ITB). 2013. Ranking do Saneameto. http://www.tratabrasil.org.br/component/estudos/itb/ranking-do-saneamento-2013 148 Instituto Trata Brasil (ITB). 2016. Ranking do Saneameto. http://www.tratabrasil.org.br/datafiles/estudos/ranking/2016/relatorio-completo.pdf Inter-American Development Bank (IADB). 2018. Development in the Americas: Better Spending for Better Lives. Washington D.C.: Inter-American Development Bank. International Water Association. 2017. “Non-revenue water management, the challengefacing water and sewage companies.” https://www.openintl.com/non-revenue-water-management-the-challenge-facing-water- and-sewage-companies/ IPEA-ENAP (2017). Burocracia Federal de Infraestrutura Econômica: reflexões sobre capacidades. Jean Marlo Pepino de Paula et al. – Brasília. Izquierdo Alejandro; Ruy Lama; Juan Pablo Medina; Jorge Puig ; Daniel Riera-Crichton; Carlos Vegh; Guillermo Javier Vuleti (2019). “Is the Public Investment Multiplier Higher in Developing Countries? An Empirical Investigation.” National Bureau ofEconomic Research Working Papers Series. Working Paper 26478. Cambridge, MA. Janson, N. 2017 “Current governance and performance of the water supply and sanitation sectorin the Caribbean.” In IDB/GOJ Conference: Financing Water Utilities of the Caribbean – Towards Efficient and Financially Sustainable Services. Kingston, Jamaica, 2017. Johns Hopkins University School of Medicine, Coronavirus Resource Center. 2020.“Mortality Analyses.” Kingdom B., R. Liemberger, and P. Marin. 2006. “The Challenge of Reducing Non-Revenue Water.” http://documents.worldbank.org/curated/en/385761468330326484/The-challenge- of- reducing-non-revenue-water-NRW-in-developing-countries-how-the-private- sector-can- help-a-look-at-performance-based-service-contracting Kresch, E.P. 2020. “The Buck Stops Where? Federalism, Uncertainty, and Investment in the Brazilian Water and Sanitation Sector.” American Economic Journal: Economic Policy,12(3): 374–401 Legatum Prosperity Index. 2019. “Brazil Prosperity Score”. 2019 Legatum Prosperity Index. https://docs.prosperity.com/8215/7407/3636/Brazil_2019_PIcountryprofile.pdf Lentini, E., and G. Ferro. 2014. Políticas tarifarias y regulatorias en el marco de los Objetivos de Desarrollo del Milenio y el derecho humano al agua y al saneamiento.Santiago de Chile: CEPAL. Lentini, E. 2015. El Futuro de Los Servicios de Agua y Saneamiento en América Latina – Desafíos de los Operadores de Áreas Urbanas de más de 300,000 habitantes. Inter-American Development Bank, CAF and ADERASA, Washington, DC.. Liemberger, R., and A.S. Wyatt. 2019. “ Quantifying the global non-revenue water problem.” Journal of Water Supply (2019) 19 (3): 831–837. https://doi.org/10.2166/ws.2018.129 Lins, C., and G. Ferreira. 2019. “Logistics infrastructure in the Southeast region: An analysisof possible pathways for a partnership between Brazil and China.” CEBRI. Rio de Janeiro. Luh, J., and J. Bartram. 2016. “Drinking Water and Sanitation: Progress in 73 Countries inRelation to Socioeconomic Indicators,” Bulletin of the World Health Organization,94(2), DOI http://dx.doi.org/10.2471/BLT.15.162974 reference 149 Manghee, S., and C. Van den Berg. 2012. “Public Expenditure Review from the Perspective of the Water and Sanitation Sector: Guidance Note.” Water Papers, Washington, D.C.,The World Bank. http://documents.worldbank.org/curated/en/667911468340140917/Public-expenditure- review-from-the-perspective-of-the-water-and-sanitation-sector-guidance-note Matheson, T., and J. Pereira. 2016. Fiscal Multipliers in Brazil. IMF Working Paper, No. 79. Maziotis, A., A. Villegas, and M. Molinos-Senante. 2020. “The cost of reducing unplannedwater supply interruptions: A parametric shadow price approach.” Science of the Total Environment, Volume 719, 137487, ISSN 0048-9697, https://doi.org/10.1016/j.scitotenv.2020.137487 Mckinsey & Company. 2020. Brazil 2020 Opportunity Tree. https://www.mckinsey.com/br/~/media/McKinsey/Locations/South%20America/ Brazil/Our%20Insights/Brazil%202020%20Opportunity%20Tree/ McKinsey2020OpportunityTree.pdf Mckinsey & Company. 2020. “Spending Reviews: A More Powerful Approach to Ensuring Value in Public Finance.” Research Article. Insights No. 3. https://www.mckinsey.com/industries/public-and-social-sector/our-insights/ spending- reviews-a-more-powerful-approach-to-ensuring-value-in-public-finances# Mckinsey & Company. 2013. “Infrastructure productivity: How to save $1 trillion a year.” The McKinsey Global Institute. New York. Melo, M.A., Pereira, C. and Souza, S. 2014. Why do some governments resort to ‘creative accounting’ but not others? Fiscal governance in the Brazilian federation. International political science review, 35(5), pp.595-612. https://journals.sagepub.com/doi/abs/10.1177/0192512114543160 Ministério do Desenvolvimento Regional (MDR). 2020. Novo Marco do Saneamento entra em vigor e trará avanços econômicos, na saúde e no meio ambiente em todo o País.MDR- SNISA, https://drive.google.com/file/d/1m8c6a9H7Vu_8IhyiQGQihDMcslKF-SVI/view Ministério do Desenvolvimento Regional (MDR). 2020. Plano Nacional de Saneamento Básico. https://www.gov.br/mdr/pt-br/assuntos/saneamento/plansab Ministério do Desenvolvimento Regional (MDR). 2018. PLANSAB. Relatório de AvaliaçãoAnual. Secretaria Nacional de Saneamento. MDR. https://antigo.mdr.gov.br/images/relatorio_plansab_2018.pdf Ministério da Economia (2020). Guia Prática de Análise Custo-Benefício de Projetos de Investimento em Infraestrutura. SEPEC, ME. Brasilia. Narzetti, D.A., and R. Cunha. 2020. “Models of Subsidies for Water and Sanitation Servicesfor Vulnerable People in South American Countries: Lessons for Brazil.” Water 2020,12, 1976; doi:10.3390/w12071976 OECD. 2018. OECD Economic Surveys: Brazil 2018, OECD Publishing, Paris. http://dx.doi.org/10.1787/eco_surveys-bra-2018-en Orçamento Fiscal e da Seguridade Social (OFSS). 2019. Gasto Social e Despesas Públicas, Brazil. https://dados.gov.br/dataset?tags=Gastos+Sociais https://www.tesourotransparente.gov.br/ckan/dataset/gastos-sociais-do-governo- central- despesas-ofss-exceto-previdencia-social 150 Pesquisa Nacional por Amostra de Domicílios (PNAD). 2011-2020. Instituto Brasileiro deGeografia e Estatística. https://www.ibge.gov.br/estatisticas/sociais/populacao/9171-pesquisa-nacional-por- amostra- de-domicilios-continua-mensal.html Pesquisa Nacional de Demografia e Saúde (PNDS). 2006. Pesquisa Nacional de Demografiae Saúde da Criança e da Mulher (PNDS-2006). Ministério da Saúde, Brasil. http://bvsms.saude.gov.br/bvs/pnds/banco_dados.php Pesquisa de Orçamentos Familiares (POF). 2017. Instituto Brasileiro de Geografia e Estatistica – IBGE. https://www.ibge.gov.br/estatisticas/sociais/educacao/9050- pesquisa-de- orcamentos-familiares.html?=&t=o-que-e Pinilla-Rodriguez, D., and Y. Torres-Sanchez. 2019. Gasto Social en el Acceso a Agua Potabley Saneamiento de las Poblaciones Rurales en América Latina. Revista Problemas del Desarrollo, 196 (50), https://www.probdes.iiec.unam.mx/index.php/pde/article/view/63499/60268 Public Expenditure and Financial Accountability (PEFA). 2009. Brazil: Federal PublicFinancial Management Performance. Based on the PEFA Methodology. https://www.pefa.org/node/2511 / https://www.pefa.org/country/brazil Raiser, M., Clarke, R., Procee, P., Briceno-Garmendia, C., Kikoni, E., Kizito, J. and Viñuela, L. 2017. Back to Planning: How to Close Brazil’s Infrastructure Gap in Times ofAusterity. World Bank, Washington, DC. https://openknowledge.worldbank.org/handle/10986/28286 REM-F. 2018. Ranking de Eficiência dos Municípios – Folha. Disponível emhttps://www1.folha.uol.com.br/remf/ Resource Governance Index. 2018. Natural Resource Governance Institute. https://resourcegovernanceindex.org/ Rozenberg, J., and M. Fay. 2019. “Beyond the Gap : How Countries Can Afford the Infrastructure They Need while Protecting the Planet.” Washington DC: World Bank. Sanitation and Water for All (SWA). 2020. WASH SDG costing tool. UNICEF - United Nations International Children’s Emergency Fund. https://www.sanitationandwaterforall.org/tools-portal/tool/sdg-costing-tool Serven, L. 2007. “Fiscal rules, public investment, and growth.” World Bank Policy ResearchWorking Paper No. 4382. Washington DC: World Bank. SIGA. 2021. SIGA Brasil: Painel Especialista. http://www9.senado.leg.br/QvAJAXZfc/opendoc. htm?document=Senado%2FsigaBrasilPainelEspecialista. qvw&host=QVS%40www9&anonymous=true&select=LB137,2019 Sistema Integrado de Planejamento e Orçamento do Governo Federal (SIOP). 2002-2021. Dados Abertos. https://www1.siop.planejamento.gov.br/siopdoc/doku.php/acesso_publico:dados_abertos SNIS. 2000-2020. Paneil de Informacoes sobre Saneamento – Brasil. http://www.snis.gov.br/painel-informacoes-saneamento-brasil/web/ reference 151 Tesouro Nacional Transparente (TNT). 2017. Balanço do Setor Público Nacional (BSPN) -2017. Ministério da Economia. https://www.tesourotransparente.gov.br/publicacoes/balanco- do-setor-publico-nacional-bspn/2017/114 Tesouro Nacional Transparente (TNT). 2019. Operações de crédito de Estados e Municípios. Ministério da Economia. https://www.tesourotransparente.gov.br/temas/estados-e-municipios/operacoes-de- credito-de-estados-e-municipios Tesouro Nacional Transparente (TNT). 2020. Uma história sobre o Balanço do Setor Público Nacional (BSPN). Ministério da Economia. https://www.tesourotransparente.gov.br/historias/bspn Tribunal de Contas da União (TCU). 2019. Orçamento e Finanças 2019. Brasilia. https://portal.tcu.gov.br/transparencia-e-prestacao-de-contas/servico/?#!consulta- a- execucao-orcamentaria-e-financeira-do-tcu&cod=241 Tribunal de Contas da União (TCU). 2020. Orçamento e Finanças 2020. Brasilia. https://portal.tcu.gov.br/transparencia-e-prestacao-de-contas/servico/?#!consulta- a- execucao-orcamentaria-e-financeira-do-tcu&cod=241 Tesouro Nacional Transparente (TNT). 2020. Uma história sobre o Balanço do Setor Público Nacional (BSPN). https://www.tesourotransparente.gov.br/historias/bspn Tesouro Nacional Transparente (TNT). 2019. Operações de crédito de Estados e Municípios. https://www.tesourotransparente.gov.br/temas/estados-e-municipios/operacoes- de- credito-de-estados-e-municipios UNICEF. 2017. “Choosing Public Expenditure Analytical Tools for Use in the WASH Sector.”WASH Guidelines G01/17. New York. https://www.unicef.org/sites/default/files/2019- 12/WASH-Guidelines-Choosing- Public-Expenditure-Analytical-Tools-Use-in-WASH- Sector-2017.pdf Wagner, R. G., et al. 2015. “Differing Methods and Definitions Influence DALY estimates: Using Population-Based Data to Calculate the Burden of Convulsive Epilepsy inRural South Africa.” PloS one, 10(12), e0145300. https://doi.org/10.1371/journal.pone.0145300 Werneck, B., G. Vieira, and L. Youssef. 2020. Brazilian New Basic Sanitation Law: The Reform and Its Implication for New Investments. Mayer Brown LLP. Illinois, USA. https://www.mayerbrown.com/-/media/files/perspectives-events/ publications/2020/09/the-legal-reform-of-brazilian-basic-sanitation-act.pdf WHO/UNICEF JMP (Joint Monitoring Program): https://washdata.org/data/household#!/ World Bank. 2003. The limits of stabilization: infrastructure, public deficits and growth in Latin America. The World Bank. https://elibrary.worldbank.org/doi/abs/10.1596/978-0-8213-5489-6 World Bank. 2012. “Guidance Note: Public Expenditure Review from the Perspective of the Water and Sanitation Sector.” S. Manghee and C. Van Den Berg. Water Papers 74742.World Bank: Washington, DC. http://documents.worldbank.org/curated/en/667911468340140917/Public- expenditure- review-from-the-perspective-of-the-water-and-sanitation-sector-guidance- note 152 World Bank. 2012. “Guidance Note: Public Expenditure Review from the Perspective of theWater and Sanitation Sector.” S. Manghee and C. Van Den Berg. Water Papers 74742.World Bank: Washington, DC. World Bank. 2013. Manual for Performance-Based Contracting by Water Utility Companies in Brazil. International Finance Corporation. World Bank: Washington, DC. World Bank. 2017. “A Fair Adjustment: Efficiency and Equity of Public Spending in Brazil.”Public Expenditure Review. World Bank: Washington, DC. https://www.worldbank.org/en/country/brazil/publication/brazil-expenditure- review- report World Bank, 2017a De Volta ao Planejamento: Como Preencher a Lacuna de Infraestruturano Brasil em Tempos de Austeridade. Relatório N.º 117392-BR. World Bank. World Bank. 2018. Volume V –Tema 4: Sustentabilidade Financeira Diálogos para o aperfeiçoamento da Política e do Sistema de Recursos Hídricos no Brasil. WorldBank- Brasília, DF/Brasil. World Bank. 2020. “COVID-19 in Brazil. Impacts and Policy Responses.” World Bank: Washington, DC. http://documents1.worldbank.org/curated/en/152381594359001244/pdf/Main- Report.pdf World Bank. 2020. WASH and COVID-19. https://www.worldbank.org/en/topic/water/brief/wash-water-sanitation-hygiene-and- covid-19 World Bank. 2020c. Monitoring Responses in Latin America and the Caribbean. SecuringWater and Sanitation Services in Times of COVID-19. World Bank Global Water Practice. World Bank: Washington, DC. https://app.powerbi.com/ view?r=eyJrIjoiNGI5NGY5ZTktZGVkMC00NDdiLTkwNGMtYzcxYzAwZjQxMWRlIiwid CI6IjMxYTJmZWMwLTI2NmItNGM2Ny1iNTZlLTI3OTZkOGY1OWMzNiIsImMiOjF9 WDI 2021. World Development Indicators, 2000-2020. The World Bank. World Economic Forum (WEF). 2017. The Global Competitiveness Report 2017-2018. World Economic Forum, Geneva. http://www3.weforum.org/docs/GCR2017-2018/05FullReport/ TheGlobalCompetitivenessReport2017%E2%80%932018.pdf World Economic Forum (WEF). 2018. Efficiency of Government Spending Scorecard. WorldEconomic Forum, Geneva. https://reports.weforum.org/pdf/gci-2017-2018-scorecard/WEF_GCI_2017_2018_ Scorecard_EOSQ043.pdf World Health Organization. 2020. “Water, sanitation, hygiene, and waste management forthe COVID-19 virus: Interim guidance.” https://www.who.int/publications- /i/item/water-sanitation-hygiene-and-waste- management-for-covid-19. Worldwide Governance Indicators (WGI). 2018. http://info.worldbank.org/governance/wgi/ reference Annex 154 The Instruments and Organizationof WSS Budgeting The Brazilian budgetary model, defined in the 1988 Federal Constitution, consists of three instruments: the Pluriannual Plan (PPA), the Budget Guidelines Law (LDO), and the Annual Budget Law (LOA). The PPA is valid for four years. Its function is to establish the guidelines, objectives, and medium-term goals of the public administration and its programs and budget. LDO states public policies and funding priorities for the next financial year on an annual basis. LOA estimates revenue and sets a schedule of expenses for the financial year. Thus, based on a detailed execution of the work plan for a given fiscal year, defined in the LOA, priorities are defined in the LDO according to the current policy, and these priorities are transmitted to the PPA. In this way, the LDO acts as a link between the LOA’s expenditure schedule and the medium-term planning of the PPA (Gontijo 2012). These three instruments interact in terms of planning and execution with state and local governments. Public service holders have authority to delegate WSS supervision, inspection, and service provision. Service provision planning, however, cannot be delegated, meaning that municipalities are responsible for preparing a Municipal Sanitation Plan (PMSB) and ensuring its feasibility. Without those plans it is difficult to honor concession contracts efficiently and spend public resources in a timely way. The WSS Law requires regulation and inspection to be performed by independent entities, that is, service providers do not have the legal right to self-regulate. It is the responsibility of the municipality to carry out these functions, or the regulatory entity. There are limited earmarked provisions to enforce concerning these important functions. Brazil has 27 CESBs including 25 companies with mixed ownership but public management. In geographic terms, seven operate in the North region, nine in the Northeast region, four in the Mid- West, four in the Southeast region, and three in the South region. Economy of scale is an important aspect of the WSS sector as the scope forcompetition is very limited. High fixed costs and asset maintenance costs (with long lifetimes) characterize the sector. Another noteworthy characteristic is its generally low technological content and slow adoption of technological advances. A system of cross-subsidies between different classes of consumers makes coverage of low-income population tenable (Narzetti and Cunha 2020). By the same logic, the cross-subsidies scheme has been extended to municipalities that have lower revenue potential. In that case, the adoption of a single state-level tariff (based on global costs) establishes a cross-subsidy mechanism by which municipalities of higher revenues support to those of smaller revenues or higher costs. Asa general pattern, larger metropolitan areas and state capitals subsidize smaller and non-urban cities. These arrangements create disincentives for small-revenue municipalities to improve their financial performance and capacities. 155 TABLE A1 State Profiles of Investment Needs Average Total Average Annual Total How Many Times Population Population Annual Investment Investment Investment Should Investment Be With With Sewage Investment Needs Needs State 2014-2018 Increased for Drinking Collection 2014-2018 2019-2033 2019-2033 (US$ million Universalization? Water 2018 2018 (US$ million (US$ million (US$ million of 2018) of 2018) of 2018) of 2018) Distrito Maintain Federal 99% 89% 319.3 63.9 783.5 52.3 current level Maintain current São Paulo 96% 90% 6,606.2 1,321.3 20,044.6 1,336.2 level Maintain Paraná 94% 71% 1,393.0 278.5 3,839.6 255.9 current level Pernambuco 81% 28% 1,001.2 200.3 4,387.6 292.6 1.38 Roraima 81% 52% 80.0 16.0 472.8 31.4 1.49 Minas Gerais 82% 72% 1,572.6 314.6 13,407.5 893.7 2.71 Espírito Santo 81% 55% 396.4 79.3 2,765.7 184.3 2.11 Mato Grosso do Sul 86% 50% 316.2 63.4 2,280.3 152.1 2.18 3.83 Rio de Janeiro 90% 65% 1,194.7 238.9 14,429.3 962.0 2.08 Sergipe 87% 25% 181.1 36.4 1,219.4 81.3 3.06 Goiás 86% 46% 667.6 133.6 6,909.0 460.6 Bahia 82% 40% 830.1 166.1 8,314.3 554.3 3.24 Tocantins 79% 26% 198.5 39.7 1,740.1 116.0 2.65 Rio Grande do Norte 87% 24% 248.0 49.6 2,686.7 179.1 3.37 Santa Catarina 89% 24% 704.9 141.1 8,759.0 584.1 3.72 Mato Grosso 89% 36% 283.9 56.8 4,897.6 326.5 5.17 Rio Grande do Sul 86% 32% 680.6 136.1 12,911.3 860.7 6.12 Paraíba 74% 36% 173.8 34.7 3,901.9 260.1 7.16 Alagoas 75% 21% 90.4 18.2 2,163.5 144.4 7.69 Acre 47% 10% 49.0 9.9 494.2 33.1 2.96 Ceará 59% 25% 307.7 61.4 6,746.2 449.6 6.94 Piauí 76% 14% 83.7 16.8 4,185.4 279.1 16.36 Amapá 81% 10% 8.9 1.7 582.1 38.8 18.43 Maranhão 56% 14% 200.9 40.2 5,480.8 365.3 8.66 Rondônia 49% 5% 62.4 12.4 1,691.0 112.7 8.22 Pará 46% 5% 163.1 32.5 5,127.6 341.9 9.57 Amazonas 35% 7% 109.2 21.8 3,013.4 200.8 8.03 Source: Instituto Trata de Brasil, and Paneil do Saneamento Básico 2019. 156 There are two main PFM entities at the State level: the State Planning and ManagementSecretariat (SEPLAG) and the State Finance Secretariat (SEF or SEFAZ, depending on thestate). The planning and budgeting function relies on SEPLAG (Secretaria de Estado de Planejamento e Gestão) as the central unit responsible for planning and coordinating government action to (1) provide technical assistance to sectors on the development and execution of their plans, programs, and budgets and (2) coordinate, consolidate, and present to the state assembly the budget documentation (PPA, LDO, andLOA). SEPLAG’s functions include public policy development; program/project planning, budgeting; monitoring and evaluation; financial information systems; human resources; and procurement of goods and services. SEPLAG is also responsible for managing externally financed credit operations. Financial planning, budget preparation and execution, including accounting and financial reporting, are the responsibilities of the State Finance Secretariat, following federal and state rules. Overall, PFM reforms in recent years have aimed at supporting greater operational efficiency in spending. The federal government has been strengthening its capacity andprocedures for managing spending more efficiently, both at the Ministry of Economy and the executing agencies. These reforms are still in their preliminary stages, making it too early to assess impact. MoP has continued to refine systems for performance reporting and evaluation of federal programs and has developed systems for monitoring the execution of voluntarytransfers to subnational levels, through agreements (convenios) with states and municipalities. The automated financial management system known as SIAFI is a critical tool for managers to make further gains in operational efficiency. However, the main priority has been to control aggregate spending. This orientation has not sufficiently encouraged budget priorities to focus more on productivity and outcomes. annex 157 Expenditures in WSS WSS (Saneamento básico) depends highly on municipal budget execution. Because Brazilian law makes municipalities ultimately responsible for the provision of water andsanitation services, execution of budget for the sector relies highly on this level of government (Figure A1). In 2017, WSS was second only to urbanism in terms of budget execution at the municipal level. Spending of this volume makes it vital to understand how municipal expenditure occurs, what are its patterns and characteristics, and what can be improved. Municipalities get the largest share of their WSS budgets from the federal government.States also receive significant federal resources. Figure A3 shows transfers made by the federal government to states (orange line) and municipalities (blue line). Between 2007 and 2014, municipalities and states saw an increase in transfers from the federal government. But as policies and laws changed in later years, a noticeable reduction occurred in the proportion of transfers to municipalities and states alike. Nevertheless, municipalities continued to get the largest share of federal resources for WSS. FIGURE A1 Distribution of Expenses by Level of Government (2017) By decreasing order of participation of Federal Gov. in total expenditure 05-National defense 07-External relationships 11-Job 21-Agrarian organization 28-Special orders 08-Social care 09-Social providence 22-Industry 20-Farming 19-Science and Technology 25-Energy 24-Communications 23-Business and services 02-Judiciary 10-Health 18-Environmental management 12-Education 26-Transport 03-Essential for justice 27-Sport and leisure 04-Management 13-Culture 01-Legislative 14-Citizenship rights 15-Town planning 06-Public security 17-Sanitation 16-Habitation 0% 25% 50% 75% 100% Municipalities State Federal Source: BSPN 2017. 158 FIGURE A2 Per Capita Spending on WSS in Brazilian States Total spending basic sanitation - 2019 250 200 Per capita spending on basics sanitation - 2019 Total Spending (millions) 60 150 Per capita spending 40 100 20 50 0 0 Espírito Santo Alagoas Minas Gerais Amazonas São Paulo Pernambuco Maranhão Distrito Feral Rio Gran do SulRio Piauí Sergipe Santa Catarina Goiás Pará Paraíba Rondônia Mato Grosso Janeiro Tocantins Mato Grosso do Sul Alagoas Amazonas Amapá Pernambuco São Paulo Maranhão Distrito Feral Sergipe Acre Santa Catarina Espírito Santo Piauí Goiás Tocantins Pará Paraíba Minas Gerais Ceará Rondônia Mato Grosso Rio Gran do Sul Mato Grosso do Sul Rio Janeiro Ceará Amapá B hi Bahia A Basic Sanitation - Rural Basic Sanitation - Rural Basic Basic Sanitation - Urban Sanitation - Urban Source: DCA, 2019-Secretaria do Tesouro Nacional Water distribution and sewage collection coverage by city size Water distribution coverage by city size Sewage collection coverage by city size % of population (2015) % of population (2015) 100 100 80 80 60 60 40 40 20 83.3% 87.3% 89.4% 20 50.3% 55.1% 55.2% 0 0 National Medium cities Large cities National Medium cities Large cities average (100-500k inh.) (>500k inh.) average (100-500k inh.) (>500k inh.) Source: SNIS 2019 annex 159 FIGURE A3 Allocation of Federal Government Basic Sanitation Budget by Level of Government (2004-2018) Allocation of Federal Government Basic Sanitation Budget (R$ millions) 4,000 3,000 2,000 1,000 0 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Municipalities States Other Note: All values are in 2018 million reais. Real estimates based on 2020 IMF forecast for real GDP growth. Source: BOOST, World Bank 2004-2018. 160 Burden of Disease Tables TABLE A2 Number of DALYs Lost and Deaths due to Diarrheal Diseases Attributable to Unimproved WASHby Federation Unit (2019) DALYs Deaths Federation Unit Total Females Males Federation Unit Total Females Males Distrito Federal 359.4 186.6 172.8 Distrito Federal 3.6 2.0 1.6 Roraima 439.4 220.8 218.6 Roraima 4.5 2.3 2.3 Amapá 854.7 401.3 453.4 Amapá 8.0 3.8 4.2 Tocantins 1,294.1 619.1 675.0 Tocantins 13.9 6.2 7.7 Espírito Santo 1,345.6 673.5 672.1 Espírito Santo 14.3 7.6 6.7 Acre 1,662.8 829.0 833.8 Acre 15.8 7.8 8.0 Rondônia 1,821.7 861.7 960.0 Rondônia 19.2 9.0 10.2 Piauí 2,164.2 1,130.2 1,034.1 São Paulo 22.4 13.0 9.4 São Paulo 2,248.3 1,187.3 1,061.0 Goiás 26.1 13.5 12.6 Sergipe 2,253.3 1,105.6 1,147.7 Mato Grosso 27.0 12.4 14.6 Mato Grosso do Sul 2,527.8 1,224.8 1,303.0 Mato Grosso do Sul 28.6 14.5 14.0 Rio de Janeiro 2,792.5 1,441.8 1,350.7 Sergipe 33.7 17.6 16.1 Goiás 2,875.9 1,465.6 1,410.3 Piauí 35.1 19.6 15.6 Mato Grosso 2,947.0 1,356.3 1,590.7 Rio de Janeiro 47.1 26.9 20.2 Paraíba 4,150.2 2,106.3 2,044.0 Amazonas 57.4 30.7 26.7 Rio Grande do Norte 4,493.4 2,031.1 2,462.3 Paraíba 58.1 32.1 25.9 Amazonas 5,038.9 2,646.3 2,392.6 Rio Grande do Norte 60.3 30.2 30.1 Rio Grande do Sul 5,775.4 3,097.4 2,678.0 Rio Grande do Sul 66.0 42.5 23.6 Alagoas 5,944.1 3,018.7 2,925.5 Alagoas 81.5 43.9 37.6 Minas Gerais 6,331.2 3,023.3 3,307.9 Paraná 83.1 47.9 35.2 Paraná 7,010.2 3,734.0 3,276.2 Minas Gerais 88.0 48.5 39.5 Maranhão 10,080.4 4,672.9 5,407.5 Pará 111.6 57.9 53.7 Pará 10,243.7 4,653.4 5,590.4 Maranhão 130.8 64.0 66.9 Pernambuco 11,737.7 5,246.2 6,491.4 Pernambuco 177.9 89.1 88.7 Ceará 12,216.3 6,163.8 6,052.5 Ceará 183.2 99.9 83.2 Bahia 15,555.3 6,977.4 8,577.9 Bahia 203.4 93.8 109.5 Santa Catarina 18,333.3 8,971.1 9,362.3 Santa Catarina 278.7 162.1 116.6 Brazil 124,920.0 60,779.9 64,140.1 Brazil 1,659.9 888.7 771.2 Source: Estimated with data from GBD IHME 2019, PNAD 2015, and PNDS 2006. annex 161 TABLE A3 Economic Burden of Disease Attributable to Unimproved WASH, by Federation Unit from Low to High Burden(Million US$ 2019) Federation Unit Total Females Males Distrito Federal 7.2 3.3 3.9 Roraima 8.6 4.4 4.2 Amapá 12 5.6 6.4 Tocantins 13.7 6.8 6.9 São Paulo 14.1 6.8 7.3 Rondônia 17 7.6 9.4 Espírito Santo 17.6 8.1 9.5 Mato Grosso do Sul 17.6 7.3 10.3 Rio Grande do Sul 23.1 11.6 11.5 Sergipe 24.4 11.3 13.1 Piauí 25.7 12.3 13.4 Mato Grosso 26 10.5 15.5 Goiás 34.7 15.4 19.4 Paraíba 34.7 17 17.7 Paraná 35.9 17.3 18.7 Acre 38.3 18.9 19.4 Rio Grande do Norte 42.8 17 25.9 Rio de Janeiro 51 24.5 26.5 Minas Gerais 60.1 25.1 35.1 Alagoas 64.3 31.4 32.9 Amazonas 71 38.6 32.3 Pará 108.4 56.8 51.6 Pernambuco 109.7 42.4 67.3 Ceará 110 54.8 55.2 Maranhão 129.9 61.1 68.8 Santa Catarina 141.1 60.3 80.8 Bahia 174.6 66.7 107.8 Brazil 1,298.1 588.8 709.3 Source: Estimated with data from GBD IHME 2019, PNAD 2015, and PNDS 2006. 162 COVID19 and WSS FIGURE A4 COVID-19 Deaths and GDP Growth by Country 4 2 China 2020 IMF forecast for real GDP growth 0 Indonesia -2 Korea -4 Australia Russia US Turkey Japan -6 Saudi Arabia Germany Brazil Canada -8 South Africa Euro area Mexico -10 India France UK Italy -12 Argentina -14 0 200 400 600 800 1000 1200 1400 Covid-19 deaths per 1m population Source: IMF, Oxford Economica, WHO, Refinitiv and EFGAM calculations. Data as at 1 January 2021. annex 163 TABLE A4 Real GDP Growth, Percent Change from Previous Year 2018 2019 2020e 2021f 2022f World 3.0 2.3 -4.3 4.0 3.8 Advanced economies 2.2 1.6 -5.4 3.3 3.5 United States 3.0 2.2 -3.6 3.5 3.3 Euro area 1.9 1.3 -7.4 3.6 4.0 Japan 0.6 0.3 -5.3 2.5 2.3 Emerging market and developing economies 4.3 3.6 -2.6 5.0 4.2 EMDEs excluding China 3.2 2.3 -5.0 3.4 3.6 Commodity-exporting EMDEs 2.0 1.6 -4.8 3.0 3.2 Other EMDEs 5.7 4.8 -1.3 6.1 4.8 Other EMDEs excluding China 4.8 3.2 -5.3 3.9 4.1 East, Asia and Pacific 6.3 5.8 0.9 7.4 5.2 China 6.6 6.1 2.0 7.9 5.2 Indonesia 5.2 5.0 -2.2 4.4 4.8 Thailand 4.1 2.4 -6.5 4.0 4.7 Europe and Central Asia 3.4 2.3 -2.9 3.3 3.9 Russian Federation 2.5 1.3 -4.0 2.6 3.0 Turkey 3.0 0.9 0.5 4.5 5.0 Poland 5.4 4.5 -3.4 3.5 4.3 Latin America and the Caribbean 1.9 1.0 -6.9 3.7 2.8 Brazil 1.8 1.4 -4.5 3.0 2.5 Mexico 2.2 -0.1 -9.0 3.7 2.6 Argentina -2.6 -2.1 -10.6 4.9 1.9 Source: 2019 Legatum Prosperity Index (https://docs.prosperity.com/8215/7407/3636/Brazil_2019_PIcountryprofile.pdf) and SIGA Brasil (http://www9.senado.leg.br/QvAJAXZfc/opendoc.htm?document=Senado%2FSigaBrasilPainelEspecialista. qvw&host=QVS%40www9&anonymous=true&select=LB137,2019) 164 Inequality in WSS Spending Inequality in spending among Brazilian states is high. Improving equity is fundamentalfor advancing the universalization agenda and promoting territorial development. Budgetary prioritization to upgrade rural and peri-urban services of water, sanitation, andwastewater treatment is essential. Research (Estache et al. 2016) shows that informal arrangements between state and local authorities play a key role in addressing this principal–agent problem or moral hazard issues related to municipal-state conflicts between providing local sanitation services and protecting the quality of water bodies. This has created high inequalities in spending on that important subsector in the last 20 years: the more aligned a mayor of a municipality is with the state’s governor, the morelikely the municipality will receive budgetary support for sanitation. Politically and economically weaker municipalities are left behind in the process of setting budget allocations. Hence inequality is exacerbated by these principal-agent problems. Furthermore, the relationship between the effectiveness of budget implementation and speed of access is strongly positive, which leaves many poorly performing WSS SOEs withlimited capacity to attain universalization of services. The figures below show how FIGURE A5 FIGURE A6 Distribution of WSS expenditures (Brazil) and Effectiveness of Water and Sanitation Spendingand Sources of Income (USA), 2018 Impact of Increased Access, 2019 100% 90 100 80 More 80 COSANPA CAGECE Cumulative Share of Income Index of Impact for access EMBASA CEDAE 70 AGESPISA COMPESA 60 60 CORSAN CAERD COSAMA CAEMA CAERN COPASA 50 CESAN CAESA 40 DEPASA SANEAGO CASAL 40 Labor CASAN WSS Less CAGEPA SABESP Spending DESO 30 20 SANESULA CAER CAESB SANEPAR Capital 20 Business 0 10 20 30 40 50 60 70 80 90 100 10 Less More 0 Capital Gains Index of Feasibility of Implementation 0 10 20 30 40 50 60 70 80 90 100% Cumulative Share of Population Note: Note: Comparing sources of income in USA Source: Armelin and Rabioglio 2019. and WSS spending in Brazil. Source: Congressional Budget Office 2011, SIOP 2018and POF 2017. annex 165 FIGURE A7 Water Public Expenditures Reviews by Type (left) and Distribution of water lossesby municipality, 2016 (right) 12% 21% 67% National PERs with chapter/volume on water Rural Water Supply & Sanitation PERs Standalone Water PERS <20.0% 20.1 a 30.0% 30.1 a 40.0% >40.1% No data Source: World Bank 2012. Source: SNIS, 2018 unequal is Brazil in terms of the distribution of WSS expenditures comparing with variables related to sources of income in United States, that also present high inequality. Also, Figure A8 below shows how the states have a wide variation in terms ofeffectiveness in WSS expenditures correlated to the increased access impacts, which shows the potential of ramping up investments given its multiplicative effects for socialand economic outcomes. Furthermore, the expenditure reviews of WSS are typically not developed as stand-alone sectoral reports, so many of the specific issues related to WSS efficiency andequity in expenditures are not addressed. For instance, in many countries it is unclear where WSS expenditures are allocated, and how these expenditures are funneled into areas with better efficiency in the use and distribution of water and sanitation services. 166 Effect of WSS Policies on Service Provision,Quality, and Finances To estimate the impact of implementation of WSS Policies on various WSS outcomes, the study uses a method of staggered timing of the implementation of the policies. Anevent framework measures the efficacy of these plans on levels of service provision, quality, and financial performance. Specifically, we run the following estimation: Where is the outcome variable of interest in municipality and year , and is an indicator for years from the implementation of the WSS policy in the municipality. We control for a variety of time-varying municipality characteristics ( ) including population, GDP, yearly budget revenue, expenditures, and taxes. We also include municipality and year fixed effects and cluster standard errors at the municipality level. The intuition of the event study is that while the implementation of BSPs was not random across the country, the variation in the timing of their implementation can be used to identify any subsequent variation in WSS outcomes. We omit the indicator for the year prior to implementation ( , so the coefficients ( ) represent the change inthe outcome relative to the year before implementation. If the change was caused by theBSP, all coefficients before the policy should not be statistically different from 0, whereasthe coefficients associated with the post-implementation period should display a significant trend. We also include those municipalities that did not implement BSPs, so that the coefficients are interpreted relative to the year before the implementation, controlling for the decision of a municipality to implement a BSP in the first place. Measures for access to the WSS system are shown in Figures A8 and A9. Figure A8 displays the effect of BSPs on the volume of water consumed system-wide in a given year. There is no notable effect of the policies on increasing the amount of water provided to consumersin the system, with the coefficients remaining relatively flat and not statistically different from “no change” compared to the year prior to policy implementation. The degree of access to the WSS systems—as measured by the number of active water connections in a municipality—displays a similar small effect after BSP implementation (Figure A9). The coefficients after the implementation of the policies are very close to zero, indicating no significant increase in system connections after the policy. This may bedue to long construction times needed to increase service capacity. However, the lack of effect more than five years after the initial policy start makes it less likely that physical fixed costs are driving the lack of significant change in WSS companies. annex 167 FIGURE A8 FIGURE A8 Impact of BSP Implementation on Volume of Water Impact of BSP Implementation on Volume of Water Connections Consumed Volume of Water Consumed (,000 m3/year) Active Water Connections (,000s) 200 1000 Coefficient on Effect Size of Basic Sanitation Policy Coefficient on Effect Size of Basic Sanitation Policy 500 100 0 0 -500 100 -1000 -8+ -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8+ -8+ -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8+ Years Since Implementation of Basic Sanitation Policy Years Since Implementation of Basic Sanitation Policy Source: SNIS 2019. We were not able to look at the effect of changes to the quality of service due to data quality issues. While the benchmarking database provided by Sistema Nacional de Informações sobre Saneamento (SNIS) includes sections on aspects of service quality, the voluntary nature of the surveys used to build the database makes it difficult to acquire complete or comparable numbers. For example, SNIS lists two potential types ofquality issues, service outages and interruptions. The most recent SNIS diagnostic lists several potential issues with these data, including companies combining the two issues, regional companies listing outages for the entire company instead of in a given municipality, and companies not providing answers to these questions, among others (SNIS 2020). Any future analysis of the WSS system will require reliable benchmarking data, with extra emphasis on creating a comprehensive database of service quality issues. Despite the lack of significant change in quantity and quality of service provision, other aspects of the WSS system changed because of the implementation of the BSP. Figure A10 displays the event study on a measure of productivity—the number of in-house employees per connection. There is a sharp increase in the number of employees per connection after the implementation of a BSP, implying that WSS companies became less efficient (Figure A11) as a result of the policy. As the number of active connections did not significantlyincrease post-implementation, this change in the productivity measure appears to be driven by an increase in in-house staff, rather than a decrease in service connections. 168 FIGURE A10 FIGURE A11 Impact of BSP on Staff per Water Connection Impact of BSPs on Productivity for Bottom 40 andTop 60 Percentiles of Municipal Income Productivity Index: In-House Staff per Water Connection Productivity Index: In-House Staff per Water Connection (Employee/,000 Connections) (Employee/,000 Connections) .6 Coefficient on Effect Size of Basic Sanitation Policy Coefficient on Effect Size of Basic Sanitation Policy 1 .4 .5 .2 0 0 -.05 -.2 -8+ -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8+ -8+ -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8+ Years Since Implementation of Basic Sanitation Policy Years Since Implementation of Basic Sanitation Policy Bottom 40th Percentile Top 60th Percentile Source: SNIS 2019. FIGURE A12 FIGURE A13 Impact of BSP Implementation on Severance Impact of BSP on Financial Performance of WSSProviders Payments of WSS Providers Expenditures on Payroll Tax and Social Security (Reals 000) Financial Performance Indicator (Percent) 5 Coefficient on Effect Size of Basic Sanitation Policy Coefficient on Effect Size of Basic Sanitation Policy 200000 0 100000 -5 0 -10 7 8+ -8+ -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8+ -8+ -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 Years Since Implementation of Basic Sanitation Policy Years Since Implementation of Basic Sanitation Policy Source: SNIS 2019. annex 169 While the effects presented in Figure A10 are for the entire sample, there is significant variation across types of municipalities. Figure A11 displays the results of the event studysplit by level of income. While the effect is present for both municipalities in the bottom 40 and top 60 percentiles of income distribution, it is primarily driven by the poorer municipalities. The increase in in-house staff after the implementation of BSPs placed a significant newfinancial burden on WSS companies. The staffing rise made budgets less effective in reaching outcomes. In particular, the increase in staff led to large increases in payroll taxes, social security, pensions, and other expenditures related to providing benefits to the larger staff (Figure A12). This increased burden led to a decrease in financial performance. Figure A13 displays theevent study measuring the ratio of direct operating revenue to total service expenditures.Increases in payroll expenditures over time after the policy implementation are tracked by a concurrent decrease in the financial health of the utility. 170 Household Spending on WSS Annual per capita expenditure on water, sanitation, and hygiene (WASH) totaled US$89 in Brazil, in comparison to US$3 in Indonesia. Low and lower middle-income countries tend to have lower average per capita WASH expenditures than higher income countries. Thirty-one percent of Brazilians live in places of low water security, that is, they face rationing, according to analysis of water sources and infrastructure that supply Brazilian municipalities. Forty-one percent of people live in regions where production systems require expansion. Only 27 percent of the population lives in municipalities where the supply is considered satisfactory. Water distribution across income brackets is very unequal, as 40 percent of the unserved population are in the minimum wage or less income bracket (Figure A15). Unserved population within the minimum wage threshold represents 38 percent of the total who have no access to services (Narzetti and Cunha 2020). FIGURE A14 FIGURE A15 Annual Per Capita Expenditure on Water, Sanitation,and Correlation between Human Development and Hygiene in 2019, by Select Country (US$) Water Service Indices for Brazilian States, 2018 Netherlands 307 .84 Trinidad and Tobago 301 .82 DF Lebanon 15 South Africa 2 .80 Hungary 141 .78 SP Jamaica SC 92 RJ Brazil .76 RS 89 ES PR Argentina 85 .74 GO Human Development Index Costa Rica 82 MG .7 Peru 77 TO 2 AP Colombia 72 .70 RO CE RR RN Serbia 54 HDI Azerbaijan 48 PE AM .68 AC Bosnia and Herzegovina 46 SE BA Solomon Islands 46 .66 PA PI PB El Salvador 43 .64 MA AL Albania 41 Senegal 27 .62 0 50 100 150 200 250 300 .30 .35 .40 .45 .50 .55 .60 .65 .70 .75 .80 .85 .90 .95 1.00 Spending in U.S. dollars Water service index Source: Sanitation and Water for All (SWA), WASH Source: SNIS and Human Development Atlas 2018 costing tool 2020. annex 171 FIGURE A15 Households Without Access to Water (left) and Sanitation (right) by Minimum Wage 45% 45% 40% 40% 35% 35% 30% 30% 25% 25% 20% 20% 15% 15% 10% 10% 5% 5% 0% 5 to 10 MW 5 to 10 MW 2 to 3 MW 3 to 5 MW 2 to 3 MW 3 to 5 MW Up to 1 MW Up to 1 MW No income No income More 10 More 10 1 to 2 1 to 2 URBAN RURAL TOTAL URBAN RURAL TOTAL Source: Narzetti and Cunha 2020