Water and Sanitation Program: REPORT Delivering Water Supply and Sanitation (WSS) Services in Fragile States Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms Chris Heymans, Rolfe Eberhard, David Ehrhardt, and Shannon Riley August 2016 The Water and Sanitation Program is a multi-donor partnership, part of the World Bank Group’s Water Global Practice, supporting poor people in obtaining affordable, safe, and sustainable access to water and sanitation services. Authors The Water and Sanitation Program (WSP) is a multidonor partnership, Chris Heymans, Rolfe Eberhard, David Ehrhardt, and part of the World Bank Group’s Water Global Practice, supporting Shannon Riley poor people in obtaining affordable, safe, and sustainable access to water and sanitation services. WSP’s donors include Australia, Austria, Denmark, Finland, France, the Bill & Melinda Gates Foundation, Photo Credits: Kathy Eales, eThekwini Water, ONEA, Luxembourg, the Netherlands, Norway, Sweden, Switzerland, the Ouagadougou, Burkina Faso; and WSP/The World Bank. United Kingdom, the United States, and the World Bank. Standard Disclaimer: This volume is a product of the staff of the International Bank for Reconstruction and Development/The World Bank. The findings, interpretations, and conclusions expressed herein are entirely those of the author and should not be attributed to the World Bank or its affiliated organizations, or to members of the Board of Executive Directors of the World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Copyright Statement: The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development/The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978-750-8400, fax 978-750-4470, http://www.copyright.com/. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA, Fax 202-522-2422, E-mail pubrights@worldbank.org. © 2016 Water and Sanitation Program Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms Chris Heymans, Rolfe Eberhard, David Ehrhardt, and Shannon Riley August 2016 Table of Contents Acronyms and Abbreviations................................................................................................................................. vii Acknowledgments...................................................................................................................................................ix Definitions.................................................................................................................................................................x Executive Summary................................................................................................................................................xii I. Introduction....................................................................................................................................................... 1 II. Methodology..................................................................................................................................................... 3 III. Water Service for Poor Households in African Cities...................................................................................... 7 3.1 Current situation......................................................................................................................................... 7 3.1.1 Percentage of poor people who can access improved water services................................................ 7 3.1.2 Percentage of poor people who can access water piped to the premises.......................................... 8 3.1.3 Percentage of poor people who can access water from a public tap.................................................. 9 3.1.4 Time poor people must spend to get water......................................................................................10 3.1.5 Continuity of water service...............................................................................................................10 3.1.6 Affordability of water for poor people................................................................................................11 3.2 How some cities serve the poor.................................................................................................................11 3.2.1 Factors distinguishing utilities that serve the poor well from others....................................................13 3.3 Key points.................................................................................................................................................14 IV. The Political Economy of Improving Water Service for the Urban Poor........................................................15 4.1 Starting reform...........................................................................................................................................18 4.2 Building momentum and sustaining reform.................................................................................................22 4.2.1 Building and preserving strong internal capabilities and culture.........................................................23 4.2.2 Forging and embedding alliances with external stakeholders............................................................26 4.2.3 Creating and strengthening formal rules and structures (institutions).................................................27 4.3 Key points.................................................................................................................................................28 V. Financing the Extension of Good Service to the Poor....................................................................................31 5.1 How successful utilities financed improved service to the poor....................................................................31 5.2 How typical utilities could finance improved service to the poor..................................................................38 5.3 Key points.................................................................................................................................................41 VI. Practical Techniques to Achieve Widespread and Affordable Access...........................................................42 6.1 Techniques to overcome financial barriers to access...................................................................................42 6.1.1 Affordability of access connections...................................................................................................42 6.1.2 Ensuring affordability of on-going service..........................................................................................44 6.2 Techniques to overcome nonfinancial barriers to good service....................................................................48 6.2.1 Prepay water dispensers that cut the monopoly power of water vendors in Nairobi and Kampala......49 6.2.2 Small providers as utility agents serve informal areas in Ouagadougou..............................................49 6.2.3 Ensuring access in hard-to-reach areas in Nyeri and eThekwini (Durban)..........................................52 6.2.4 Do Pro-Poor Units help in serving the poor?.....................................................................................52 6.3 Key points.................................................................................................................................................52 VII. Conclusion.......................................................................................................................................................56 iv Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Table of Contents Appendixes Appendix A: Case studies: How Cities Turned Around Water Service for the Poor...............................................61 Appendix B: Suggestions for Further Research..................................................................................................90 Appendix C: Perspectives on Alternative or Supplementary Service Providers....................................................91 Appendix D: Utility Management Effectiveness Index..........................................................................................92 Appendix E: Longitudinal Utility Data..................................................................................................................94 Appendix F: Data Sources...............................................................................................................................129 Appendix G: References..................................................................................................................................132 Tables Table 2.1: Cities analyzed in study................................................................................................................... 4 Table 3.1: Proportion of monthly income spent on water by poorest 40 percent, Kenya..................................11 Table 3.2: Access to piped water, compared to access to utility-provided water.............................................13 Table 4.1: Key elements in starting reform, building momentum, and sustaining reform...................................16 Table 5.1: Improved water supply service to the poor.....................................................................................31 Table 5.2: Sources and amounts of capital investment financing.....................................................................33 Table 5.3: Summary of efficiency indicators, NYEWASCO...............................................................................34 Table 5.4: Summary of efficiency indicators, NWSC........................................................................................35 Table 5.5: Strategies for improving operating cash flow..................................................................................38 Table 5.6: Performance of successful utilities..................................................................................................40 Table 6.1: Summary of techniques used to achieve affordable access............................................................54 Table C.1: Perspectives on alternative or supplementary service providers......................................................91 Table D.1: Utility management effectiveness index..........................................................................................92 Table D.2: Inputs for utility management effectiveness index...........................................................................93 Table F.1: Household survey data sources...................................................................................................129 Table F.2: List of data sources for utility-reported data..................................................................................130 Figures Figure 3.1: Access to different levels of service, poorest 40 percent.................................................................. 7 Figure 3.2: Access to water piped to the premises, poorest 40 percent............................................................ 9 Figure 3.3: Access to public tap, poorest 40 percent........................................................................................ 9 Figure 3.4: Average roundtrip time to get water from public tap, poorest 40 percent........................................10 Figure 3.5: Continuity (hours of water supply service per day)..........................................................................11 Figure 3.6: Dakar provides water piped to the premises to (almost) all..............................................................12 Figure 3.7: Relationship between service to the poor and utility management effectiveness..............................14 Figure 4.1: Starting the reform, building momentum, and sustaining the reform................................................15 Figure 5.1: Growth in water connections in case study cities............................................................................32 Figure 5.2: Growth in connection density (connections per 100 people in city)..................................................32 Figure 5.3: Sources and amounts of capital expenditure financing (KShs billion)...............................................33 Figure 5.4: Capital expenditure financing and debt service payments...............................................................34 Figure 5.5: Sources and amounts of capital expenditure financing (UShs billion)...............................................35 Figure 5.6: Sources and amounts of investment, ONEA...................................................................................36 Figure 5.7: Sources and amounts of capital expenditure financing, SDE/SONES..............................................37 Figure 5.8: Comparison of utility operating efficiency........................................................................................39 Figure 5.9: Comparison of utility connectedness (connections per 100 people in city)......................................39 Figure 5.10: A financing strategy for universal access........................................................................................40 v Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Table of Contents Figure 6.1: Comparison of progressive tariff structures (eThekwini, SDE/SONES).............................................45 Figure 6.2: SDE/SONES tariff structure, 2014..................................................................................................46 Figure 6.3: NWSC tariff structure, 2015...........................................................................................................46 Figure 6.4: Delegation of water service in informal settlements in Ouagadougou...............................................49 Figure 6.5: Access to piped water to premise in Ouagadougou’s informal settlements......................................50 Figure A.1: Dakar provides water piped to the premises to (almost) all..............................................................62 Figure A.2: Operating expenses and tariffs, operating cost recovery ratio.........................................................64 Figure A.3: The number of connections in eThekwini increased by 35 percent over 11 years with some experimentation in alternatives.......................................................................................................67 Figure A.4: Operating cost coverage, eThekwini Water and Sanitation Unit.......................................................68 Figure A.5: Access to piped water by poor, Kampala.......................................................................................73 Figure A.6: Achieving financial self-sufficiency at NWSC...................................................................................75 Figure A.7: Sources and amounts of capital expenditure financing....................................................................81 Figure A.8: Cash from operations compared to debt service payments............................................................81 Figure A.9: Nonrevenue water, staff productivity (NYEWASCO).........................................................................82 Figure A.10: Cost recovery and real average water tariff over time......................................................................82 Figure A.11: Formal and informal settlements in Ouagadougou..........................................................................85 Figure A.12: Delegation of water supply to private operators in informal areas....................................................87 Boxes Box 3.1: SDGs ‘safe and affordable’ versus ‘improved’....................................................................................... 8 Box 4.1: Institutions and individuals in water sector improvements.....................................................................18 Box 4.2: Role of national and local political support in reforms............................................................................22 Box 4.3: Applicability of the findings in fragile environments................................................................................29 Box 6.1: The role of private providers in Greater Maputo....................................................................................51 Box 6.2: Pro-Poor Units: Drivers for serving the poor?.......................................................................................53 vi Acronyms and Abbreviations AAWSA Addis Ababa Water and Sewerage Authority AFD Agence Française de Développement APC Area Performance Contract CEO Chief Executive Officer CRA Conselho de Regulação do Abastecimento de Água DAWASCO Dar es Salaam Water and Sewerage Corporation DHS Demographic and Health Survey DRC Democratic Republic of the Congo EWURA Energy and Water Utilities Regulatory Authority FCFA West African CFA Franc GDP Gross Domestic Product GIS Geographic Information System GIZ German Corporation for International Cooperation GWCL Ghana Water Company Ltd GTZ Germany Agency for Technical Cooperation HDPE High-density polyethylene HUWSUP Hargeisa Water Supply Upgrade Project HWA Hargeisa Water Agency IBNET International Benchmarking Network for Water and Sanitation Utilities IBRD International Bank for Reconstruction and Development IDA International Development Association IDAMCs Internally Delegated Management Contracts IFI International financial institution ISO International Organization for Standardization JMP UNICEF Joint Monitoring Programme KANU Kenya African National Union KfW German Development Bank KShs Kenyan Shillings Lpcd Liters per capita per day LWSC Lusaka Water and Sewerage Company MD Managing Director MDG Millennium Development Goal MLD Million liters per day MOWASCO Mombasa Water Supply & Sanitation Company MPESA M for mobile, pesa is Swahili for money (mobile money transfer service) NCWSC Nairobi City Water and Sewerage Company NGO Nongovernmental organization NRM National Resistance Movement NRW Nonrevenue water NTU Nephelometric Turbidity Unit NWASCO National Water Supply and Sanitation Council (Zambia) vii Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Acronyms and Abbreviations NWSC National Water and Sewerage Corporation (Uganda) NYEWASCO Nyeri Water and Sewerage Company O&M Operations & Maintenance ONAS l’Office National de l’Assainissement du Sénégal ONEA l’Office national de l’eau et de l’assainissement PEMU Projet d’alimentation en eau potable en milieu urbain PPP Public-private partnership PS Parti Socialiste SDE Sénégalaise des Eaux SDG Sustainable Development Goal SEEN Société d’Exploitation des Eaux du Niger SONEES Société Nationale d’Exploitation des Eaux du Sénégal SONES Société Nationale des Eaux du Sénégal SPEN Société de Patrimoine des Eaux du Niger Tanga UWASA Tanga Urban Water Supply and Sanitation Authority UN-HABITAT United Nations Human Settlement Program UShs Ugandan Shillings USD United States Dollars WASH Water, sanitation, and hygiene WASREB Water Services Regulatory Board (Kenya) WHO World Health Organization WSP World Bank’s Water and Sanitation Program viii Acknowledgments The authors wish to gratefully acknowledge the inputs World Bank staff and consultants–Luis Andrés, Matar from the many people involved in developing this report. Fall, Yitbarek Tessema, Martin Gambrill, Mike Webster, Scott Taylor conducted the political economy analysis for Caroline Van Den Berg, Alexander Danilenko, Sophie four cities, with substantial contributions from Brian Levy. Trémelot, Oliver Mark Jones, Gustavo Saltiel, Mukami Jacqueline Homann produced the Geographic Information Kariuki, Gerard Soppe, Michel Duret, Samuel Mutono, System (GIS) maps and analysis with assistance from Nisha Ngoni Mudege, Seydou Traore, Taibou Adamou Maiga, Krishnan and Justin Baker. Dieudonné Sawadogo and Neil Denis Dakoure, Oumar Diallo, Adama Mbaye, Ajith Macleod provided valuable inputs to the report, including Kumar, Emmanuel Nkrumah, Maximilian Hirn, Gulilat lessons from their own professional experience managing Birhane Eshetu, Tesfay Bekalu Wondem, Kaposo Mwam- two utilities in Sub-Saharan Africa. buli, Chantal Richey, Dominick Revell de Waal, Odete Duarte Muximpua, Kevin Bender, Jean Doyen, Lilian A large group of senior utility managers–Eng. Joshua Mgeye- Otiego, and Charlotte Ndakorerwa–helped refine the kwa (Tanga UWASA), Eng. George Ndongwe (Lusaka main messages and assisted with data collection. Glenn WSC), Eng. Silver Mugisha (NWSC), Hamado Ouedraogo Pearce-Oroz, Bill Kingdom, Jyoti Shukla, Alex Bakalian, (ONEA), Moumouni Sawadogo (ONEA), Abdoul Bal and Jonathan Kamkwalala and many others provided guid- Cheikh Sall (SDE), Ednick Msweli (eThekwini), and Eng. ance throughout the assignment, and helpful feedback Peter Gichaaga (NYEWASCO) –shared their successes and during peer review sessions held in December 2015 and challenges in a video-conference in November 2015. January 2016. ix Definitions Alternative service providers. There are no standard definitions to classify alternative or small-scale providers, Operating cost coverage ratio. The operating cost coverage due mainly to their informal nature. From the literature, ratio is defined as revenue divided by operating expenses. captured most recently by Kariuki and Schwarz (2005), For the purposes of this report, operating cost coverage the independence of such providers is an important part data were collected from utility accounts, regulator reports, of understanding of them, recommending that the concept IBNET, and directly from the utilities. should essentially describe providers who are physically delinked from formal utilities. Kariuki and Schwarz Poor people. Poor people are defined as the bottom 40 suggest that this term should refer to those providers percent by income (or a proxy for income, such as asset producing and selling their own sources of water, but in ownership) of people living in cities. Poor people are the this report this is somewhat qualified. For example, the case focus of this study, in line with the World Bank goal of of Ouagadougou in this report shows that it is conceivable promoting shared prosperity by fostering income growth of that such providers may purchase water from a formal the bottom 40 percent of the population in every country. utility to manage distribution in areas where the utility is However, because urban incomes are typically higher not able to serve customers. than rural incomes, income of some of the urban poor (taken as the bottom 40 percent of the urban area income Good service (with respect to a city). A city in which at distribution) will be above the 40th percentile of income least 90 percent of the bottom 40 percent of households by nationally. income (or a proxy for income, such as asset ownership) have access to an improved water source, and water is Piped water service. Piped water service is classified as any available for at least 18 hours per day, on average. The word form of piped water to premise (piped water to dwelling or ‘good’ is used in the sense of ‘good relative to the typical piped water to the yard or plot), public tap or standpipe, situation’. and piped water from the neighbor. Improved water. Improved water is classified according to Piped water to premise. Piped water to premise is classified the World Health Organization/United Nations Children’s as piped water into dwelling or piped water to the yard Fund Joint Monitoring Programme (WHO/UNICEF or plot. The definition does not interrogate the extent to JMP) definition. It includes piped water into dwelling, which such a connection may in fact be shared, for example piped water into yard or plot, public tap or standpipe, between different tenants renting on the premises. tubewell or borehole, protected dug well, protected spring, and rainwater. Public tap. According to the JMP definition (used in this report), a public tap is a public water point from which Nonrevenue water. In this report, nonrevenue water people can collect water. Standpipes and water kiosks are (NRW) is primarily expressed as the percentage of water also classified as public taps. supplied into the distribution system that is not billed to consumers. It consists of physical losses through leaks in the Professional. For the purposes of this report, professional distribution system and commercial losses from inaccurate is defined to mean a set of practices and behaviors that metering and illegal or unbilled connections. strive for continued improvement towards excellence as defined by sector or other widely agreed standards in utility Other improved water. Other improved water includes management and governance (in other words, strategy, tubewell or borehole, protected dug well, protected spring, technical/engineering, customer services, financial, human and rainwater. In other words, it includes all sources of resources, procurement, ethics, and so on). See, for example, improved water other than piped water into dwelling, Effective Utility Management–A primer for water and piped water into yard or plot, and public tap or standpipe wastewater utilities. The discussion in the report emphasizes (collectively known as piped water in this report). though that the application of such standards should be x Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Definitions relevant to the context, rather than assuming that universal Utility. A utility is defined, for the purposes of this report, “best practice” is possible. as an entity engaged in the provision of potable water to customers irrespective of its particular institutional form and Shared tap. A shared tap is defined as any piped water ownership or management structure. This study found good service that is not piped to premise. In other words, a shared service to the poor among utilities that were both publicly tap is a public tap or standpipe, or piped water accessed and privately owned and/or managed and which had various from a neighbor’s tap. corporate identities, including being a department in a municipality. Institutional form and ownership are touched upon, but were not a focus of this study. xi Executive Summary Africa’s urban population will triple by 2050. People in However, in this challenging context, some cities are making these rapidly growing cities need safe, convenient, and significant strides in providing sustainable service to poor reliable water supplies. However, the proportion of Africa’s households. The table here shows seven cities (marked by urban population with improved water supply has barely asterisks) that provide relatively good service to the poor– grown since 1990. Research shows that water piped to the with good service defined as at least 90 percent access to premises is the standard to ensure adequate health (families improved water, and supply reliability of at least 18 hours who rely on water carried from shared taps often do not per day on average. This report aims to share lessons from get enough water for basic needs). Yet the share of the these cities’ achievements. urban population with water piped to their premises has declined, from 43 percent in 1990 to 33 percent in 2015. Five of the seven cities were analyzed in detail for this Poor families are the least likely to have water piped to their study. Ten cities that provide typical levels of service were premises, and the fact is that income levels remain low for also studied, to understand the range of service levels and many city-dwellers. The most vulnerable, therefore, will contexts across cities in Africa. bear the brunt of the inadequacy of water supplies. Access and service Cities analyzed in study The Sustainable Development Goals (SDGs) mark City Reliability Access to a notable shift from the access-focused Millennium (hours/day) improved water Development Goals (MDGs) to a new international (poorest 40%) monitoring approach that places greater emphasis on the eThekwini (Durban)* 24 100% quality of that access. With quality of access in mind, two Lusaka* 20 99% key measures stand out in the analysis. First, an average Ouagadougou* 23 98% of 97 percent of the poor have access to improved water Dakar* 24 97% service in the seven cities. The average access to piped water Nyeri* 24 96% by poor households is 90 percent. In four of these cities Niamey* 24 94% shared pipes are the main way of serving the poor, and in Kampala* 18 93% Dakar and Nyeri, 75 percent or more of poor customers Nairobi 20 86% have water piped to their premises. Second, in most of Hargeisa 20 76% the cities, residents received piped water for more than 18 Tanga 24 75% hours a day. Both accessibility (including the quantity of water) and reliability of water supply are important. Accra 18 65% Addis Ababa 16 100% Remarkably, good water service for the poor is found in Maputo 12 97% large, poor, and rapidly growing cities with arid climates, Kinshasa 11 93% such as Ouagadougou and Niamey, and in countries with Mombasa 4 92% low governance effectiveness according to the World Bank’s Dar es Salaam 8 89% Worldwide Governance Indicators. Clearly, serving poor Kaduna 14 29% people in fast growing and poor African cities is possible, * City with relatively good service for the poor. even in countries that are poor, arid, and suffer from Notes: governance problems. ‘Poorest 40%’ refers to households in the bottom 40 percent of income distribution in the city, or households in the bottom 40 percent of an asset ownership index if income data was not available. How have these cities managed to provide widespread, ‘Access to improved water’ is as defined by the Joint Monitoring Program reliable service to poor people in such challenging (see Definitions section). environments? The case studies of five cities show that Sources: Household surveys (see Table D.1 for list of sources); Utility improvements in service to the poor started with important research (see Table D.2 for full list of sources). changes in the general or specific political economy in xii Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Executive Summary Access to different levels of service, poorest 40 percent * Relatively good service to the poor. Source: Household survey data. Full list of sources for each city is included in D.1. the city or country. These changes allowed or enabled subject of this report.1 It is common cause that it is poor the turnaround of underperforming utilities. The utilities people that are most affected by an obstructive political improved their financial performance, generating surpluses economy and that this is experienced directly in the lack that supported investment in infrastructure to serve the of access to reliable, safe, piped water sources.2 This study poor. At the same time, the utilities adopted a range of pro- sought to understand the opposite side of this equation– poor strategies to overcome the financial and nonfinancial where and how water services are provided effectively to barriers to serving the poor. poor households living in cities in Africa. It is notable that, in all five of the case study cities, improvements Political economy of turning around water in service to the poor started with improvements in the service for the poor political economy of the sector and utility serving the The ways in which political economy can obstruct the city. Common features present in all cases were analyzed effective and efficient provision of services–through, for to understand how cities start reform, build momentum, example, patronage and corruption or rent-seeking– and sustain reform (see figure below). have been discussed at length elsewhere and are not the 1 Castalia Strategic Advisors. 2014. Barriers to Infrastructure Service Delivery in Sub- 2 Bratton, Michael, and Carolyn Logan. 2013. “Voters but not yet Citizens.” In Voting and Saharan Africa and South Asia. Democratic Citizenship in Africa, edited by Michael Bratton. Boulder: Lynne Rienner. 2 Bratton, Michael, and Carolyn Logan. 2013. “Voters but not yet Citizens.” In Voting and Democratic Citizenship in Africa, edited by Michael Bratton. Boulder: Lynne Rienner. Driving and sustaining reform Starting the reform Building the momentum Sustaining the reform Build Preserve Professional technical Internal capability and culture leader Catalytic event Forge Embed or space for reform Alliances external to the utility Secure political Create Strengthen leader Formal rules and structures xiii Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Executive Summary Summary: CITY and service improvements data City Dakar Ouagadougou Nyeri Kampala eThekwini Country Senegal Burkina Fasa Kenya Uganda South Africa Utility SDE ONEA NYEWASCO NWSC eThekwini municipality Country and city context Country GDP/capita (US$ 1,071 720 1,338 677 6,478 at PPP, 2014) Fresh water resources (m3 1,758 711 461 1,032 829 per capita, 2014) City population (million) 3.1 2.5 0.150 1.6 3.6 Worldwide Governance 0.02 (1996) -1.02 (1996) -0.34 (1996) -0.73 (1996) 0.88 (1996) Indicators – Government -0.39 (2014) -0.56 (2014) -0.30 (2014) -0.40 (2014) 0.33 (2104 Effectiveness 1996 and 2014 Access to water services (Household surveys data) ) Access to improved water 97% 98% 96% 93% 100% service (% of bottom 40%) (2014) (2010) (2013) (2011) (2015) Access to piped water 97% 89% 90% 78% 92% service (% of bottom 40%) (2014) (2010) (2013) (2011) (2015) Access to water on 75% 12% 88% 7% 45% premises (% of bottom (2014) (2010) (2013) (2011) (2015) 40%) Water reliability 23 22 24 18 24 (2014) (2010) (2013) (2011) (2015) Network expansion Increase in connections 189,000 to 368,000 60,000 to 207,000 9,800 to 23,400 46,000 to 214,000 369,000 to 498,000 (in city) (2002 to 2013) (2005 to 2014) (2006 to 2015) (2002 to 2015) (2004 to 2015) Connectedness index 8.7 to 12.0 4.5 to 8.8 8.9 to 14.9 4.4 to 9.0 11.6 to 13.9 (connections per 100 (2002 to 2013) (2005 to 2014) (2006 to 2015) (2003 to 2013) (2004 to 2015) people in city) Utility management effectiveness (Utility data) Cash collection efficiency Averaged 97% 78% to 97% 98% to 100% 85% to 95% 81% to 98% (%) (1996 to 2013) (2002 to 2013) (2006 to 2014) (2001 to 2011) (2004 to 2014) Nonrevenue water % 30% to 20% Averaged 17% 42% to 19% 43% to 35% Averaged 38% (1996 to 2013) (2002 to 2013) (2006 to 2014) (2001 to 2011) (2004 to 2014) Staff productivity (n / 1000 6.9 to 2.6 8.3 to 3.0 8.9 to 3.5 16.5 to 5.9 Averaged 3.6 (2005 connections) (1996 to 2013) (2002 to 2013) (2006 to 2014) (2001 to 2011) to 2015) Operating cost recovery 1.01 to 1.33 Averaged 1.33 1.17 to 1.26 1.18 to 1.31 Averaged 1.33 ratio (1998 to 2013) (2002 to 2013) (2006 to 2014) (2001 to 2011) (2010 to 2015) Management effectiveness 77 to 99 83 to 94 70 to 100 58 to 85 Averaged 90 index (1998 to 2013) (2002 to 2013) (2006 to 2014) (2001 to 2011) (2010 to 2014) xiv Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Executive Summary The start of successful reforms in the case studies displayed (IDAMCs). The IDAMCs successfully applied the three mutually reinforcing conditions (1) a catalytic event principles of incentivized management to the utility’s or space for reform; (2) a skilled technical leader motivated area managers. The utility leader in eThekwini innovated to improve service; and (3) a relatively stable political leader constantly, saying that at some points he felt like he was who supported and protected the reform. In Ouagadougou, ‘just trying to run faster than everyone else’ in coming up the catalytic event was a water resources crisis. In eThekwini with institutional and management solutions to an ever- (Durban), the end of apartheid in South Africa created the changing set of challenges. political impetus to redress water service inequalities. In both cases, a capable utility manager was able to chart the Making reforms sustainable is as much a challenge as way to addressing an urgent need, combining technical and getting them started. As utilities succeed in providing financial strategies with institutional reform, and winning better service, they gain more resources and become more the support of a political leader (Burkina Faso’s Minister of tempting targets for predation, whether for personal or Water and eThekwini’s mayor, respectively) who felt secure political gain. Yet, while success might be a double-edged enough politically to be able to risk change in pursuit of sword, it could also be the first line of defense against such service improvement. predation. Proven competence in service delivery can win support from external stakeholders and raise the political The political economy conditions that set these cities on the costs of predation. Uganda’s NWSC makes the point that path to reform cannot be created by outsiders. However, “In Kampala, the poor vote.” Nyeri’s water utility describes outsiders were crucial to the success of the reforms in how, when allegations of political meddling surfaced, some cases, providing financing and technical assistance to citizens called to ask how they could prevent it. the utility. In the case of Dakar, for instance, the World Bank and other development partners provided the In the cases studied, utilities consciously embed themselves finance needed for major increases in production capacity, in networks of relationships with external stakeholders and for expansion of the distribution network into poor who benefit from their success–especially their customers. communities. Equally important was technical assistance The NWSC and Nyeri’s utility each hold regular barazas for the design and implementation of new institutional (consultation with communities) to find out what service arrangements. In Nyeri (Kenya) technical assistance from the the community needs, and also to educate the community German development agency helped the municipal utility on its responsibilities. eThekwini Water established a to corporatize and commercialize. A loan from KfW (the consultative committee with low income communities. German development bank) then financed a new treatment The committee became more engaged and supportive when works and network expansion. These investments provided the utility addressed requests to increase the free basic the water and distribution needed to increase service in water allowance. The utility leader who drove eThekwini’s poor areas, and reduced operating costs. turnaround credits the continued renewal of his contract by the elected city council to eThekwini’s success in serving Institutional reforms in the five cities studied differed poor communities. widely from case to case, but in all cases they were motivated by solving pressing problems first. Institutional Formal structures, such as independent boards and reforms did not follow a ‘best practice blueprint’. While regulators, are not sufficient to ensure effective service in Dakar the institutional reforms were planned out in provision because they are not immune to predation or detail and implemented early, the other cities took a capture. However, these structures can be useful for bolstering more adaptive, evolutionary approach. Uganda’s National professional corporate cultures, and coordinating supportive Water and Sewerage Corporation (NWSC) started with relationships with external stakeholders. The affermage and foreign management contractors for service in Kampala. related contracts in Senegal provide clear rules that are costly Drawing on the experience from those contracts, it then to change. Provided the utility keeps doing a good job, developed Internally Delegated Management Contracts the contracts support success. In Burkina Faso, ONEA is publicly owned and operated. Its performance contract with xv Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Executive Summary government is supervised by a multistakeholder committee More utilities in Africa could emulate their peers in comprising representatives of customers, nongovernmental raising finance on the strength of improving cash flows. organizations, and the development partners who finance Consider what could be achieved if this happened in the sector. The committee monitors performance of both the Mombasa, a city where hardly any households in the utility and the government under the contract, on the basis poorest 40 percent have water piped to the premise, and of independently audited financial and technical reports. the 77 percent who can get water from standpipes have Such designs embed accountability to external stakeholders an average round-trip time of 22 minutes. If Mombasa’s in formal structures that can also help mobilize support utility improved operating efficiency on key indicators against predation. to levels achieved by well performing utilities studied, it would generate enough cash to service concessionary loans Financing the extension of good service to of US$1.0 billion–enough to provide universal access to the poor piped water, including water to the premise to 42 percent In the cases analyzed, grant finance formed 5 percent of poor households, while holding real increases in the (NYEWASCO), 28 percent (NWSC), 29 percent (SDE), average tariff to just 1.5 percent per annum. and 52 percent (ONEA) of the total investments, with the rest financed through loans and internally generated Practical techniques to achieve widespread cash. In Dakar, Kampala, Nyeri, and Ouagadougou, the and affordable access utilities borrowed money (mostly from development Using similar financing strategies, typical African cities may partners via national governments) to finance their be able to expand networks and increase access by the poor. investment programs. They serviced these loans with cash More finance and new infrastructure alone, however, will they generated, and also invested cash surpluses directly in not ensure that access is widespread and affordable. Utilities infrastructure expansion. in the cities that serve the poor relatively well also help poor customers to overcome financial and nonfinancial barriers To generate the operating surpluses needed to finance to access. They implement rising block tariff structures investments, utilities in the five cities improved their well, to ensure they truly benefit the poor, and they cross- management and cost effectiveness. They reduced subsidize residential consumption by charging higher tariffs nonrevenue water, increased collection efficiency and to commercial customers. They have successfully curbed boosted labor productivity. Adequate tariffs were also water mafia and other on-sellers who charge excessive mark- important. Average tariffs in these cities range from US$0.82 ups. Where informal land tenure does not allow service (NWSC) to US$1.08 (SDE/SONES). Real tariff increases provision, or the pattern of settlement or topography makes have been moderate, ranging from –8 percent to +3 percent conventional network designs infeasible, they improvise per annum, on average. new technical and institutional arrangements. This includes Growth in water connections in case study cities xvi Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Executive Summary Summary: REform features Catalyst for Mid-1990s. Solve Early- to mid-1990s. Early- to mid- Late 1990s. Donor 1992, Need to reform water shortage in Water shortage and 1990s. Water support for PPP serve all citizens in Dakar low water coverage rationing and coupled with the metro area in a in Ouagadougou poor water quality Ugandan opposition way that undid the and Bobo-Dioulasso problem to PPP. Conflict legacy of service created space for a discrimination under credible alternative, Apartheid in a situation where Uganda needed the development partners Nature of Reform plan Decided against Independent, President and utility Municipal unit not reforms developed with private operator to accountable drove strategy to interfered with; local stakeholders form independent structure; use of improve water supply. strategic planning and development national utility on loan financing; Early (external and focus on outcomes; partners; then short- a performance transparent internal) management incentivized term improvements, contract; involve audited systems contracts (1998– contracts with followed by consumers, and strategic 2004), and then senior management structural reform: government, NGOs, planning beyond internally delegated team; training and a public-private and development the immediate; management staff performance; partnership partners in robust revenue contracts since 2004. impressive customer (PPP), making a Supervision practice, no Past 3 years fresh call center; adjusting private operator Committee; Sub- preferential focus on equity. services to contexts; responsible for contracting of small treatment; active Robust corporate use free monthly providing services providers to reach engagement of training facility put in water allowances and managing peri-urban areas. customers, with place from national infrastructure “brand loyalty”. government to leverage access. working with small providers to deliver services where the Other utilities are also adopting new technologies to utility cannot, such as ONEA does in Ouagadougou. This better serve the poor. Some use ‘water dispensers’–prepaid is proving a handy option for a wider range of utilities, for standpipes–that can be placed at frequent intervals in example, in Maputo. How utilities used these techniques is poor communities. Utilities in Kenya use a mobile phone summarized in Table 6.1 in the main document. technology in which the customers can report their own meter reading, and then pay the bill using mobile eThekwini delivers water to poor households to meet basic money. This system allows customers to make smaller, needs for free, using technologies that can restrict the more frequent, payments that are in line with their own amount of water delivered to exactly the free allowance. household cash flows. In Kampala, the utility relies on standpipes to serve the poor, and ensures that standpipe tariffs are kept lower than Implications for the World Bank and other the tariff for water piped to the premise. If more than two development partners or three households are accessing water from the same The research shows that effective reforms have been initiated yard tap, the standpipe tariff is also applied in this case. in challenging circumstances. Would-be reformers, in In Ouagadougou, the utility enlists small entrepreneurs to politics or at the utility-level, if given an opening through resell water at controlled prices to residents of the rapidly an exploitable catalytic moment, use that opening to expanding informal settlements. build momentum. Development partners, spotting such xvii Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Executive Summary champions, should back them. Finance could be phased in, In summary with emergency needs addressed first. Technical assistance The cases studied show that it is possible for rapidly growing should help with governance and management reforms that African cities to offer the poorest 40 percent of their residents address immediate needs while building for the future. near-universal access to reliable and affordable water. The technical, financial, and managerial techniques used by Development partners can play a supportive role in all utilities in these cities are widely applicable. Whether they phases of reform–starting, building, and sustaining. are actually applied in any particular case depends largely However, they cannot expect to create from the outside the on whether political economy factors enable, support, conditions in which better service to the poor becomes a and protect the professional management and provision political economy priority, nor to succeed by transplanting of services by the utility. Where these conditions can be ‘best practice models’ regardless of context. locally created, and combined with the necessary financial and technical support, utilities in many cities will be able to extend and sustain water service to poor households. xviii I. Introduction Sustainable Development Goal (SDG) 6 targets “universal in 1990 to 33 percent in 2015.3 While data on access for and equitable access to safe and affordable drinking water the poor are not available continent-wide, the analysis and for all”. However, in Africa’s fast-growing cities, just discussion in this document supports the view that poor accessing water at all is a daily struggle for many poor people are almost certainly worse off than the average access families. To redress these problems, new approaches are statistics imply. needed in managing water supply in these cities. While it is often assumed that many cities on the continent are Rapid urbanization in Africa adds to the challenge. By 2050, structurally unable to develop, implement, and sustain the urban population will more than triple, with an average such new approaches, this report provides evidence that annual growth rate of 3.4 percent. Demand for water will transformation and achieving services to the poor are rise at similar rates. Water resources will get progressively indeed possible. more difficult and costly to obtain, as cities use the closest sources of water first, and then have to tap sources that are Need for new approaches to urban water further away. These sources require greater investment in supply in Africa pipes and pumping. As costs rise, reuse will become more In urban Sub-Saharan Africa, access to “improved” water viable and more important. has virtually stagnated at about 85 percent, whereas access 3 World Health Organization (WHO). Progress on Sanitation and Drinking Water: to water piped to the premises has fallen–from 43 percent 2015 Update and MDG Assessment. http://apps.who.int/iris/bitstream/10665/17775 2/1/9789241509145_eng.pdf. Improved water service did grow at a rate of one-third of a percent per annum, which may be considered virtually stagnant. Urban Africa: Rapidly growing and densifying. (Photo Credit: Kathy Eales). 1 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Introduction Exacerbating the situation is the relatively low incomes still mainly self-reported with limited external verification. of the African countries in which rapid urbanization is To develop a more complete perspective on the services that taking place. Many countries in Africa are expected to have poor families receive in the 17 cities, the analysis here also populations that are 50 percent urban while per capita used household survey data, presented in Section 3. incomes are at or below US$3,500. Case studies of five cities that turned around water service Such rapidly growing demand and resource scarcity will for the poor–Dakar, Senegal; eThekwini (Durban), South strain already stretched institutional capacity. The finance Africa; Kampala, Uganda; Nyeri, Kenya; and Ouagadougou, needed to bring water to cities and distribute it to poor Burkina Faso–are included in Appendix A. areas will increase significantly. Drawing on these cases and other research, Section 4 But more finance alone is not the answer–it would need to examines the political economy of service reform. It explores be accompanied by better approaches for providing water how changes in the political economy enabled five cities to services to the poor. If not, African cities run a high risk improve water services to the poor. of being marked by sprawling low income communities in which struggling families pay excessive prices for water Improved service typically requires significant capital of dubious quality, spend hours out of their day queuing expenditure; Section 5 explains how these expenditures for and carrying water, and cannot get enough water for were financed, and what other cities can learn from their daily hygiene needs. Rich households may be able to finance strategies. arrange adequate self-provision–for instance, through trucked delivery of water or private boreholes and in- Cities which served the poor well used a set of practical home treatment systems (as happens from Lagos and Port techniques for service delivery, billing, and cost recovery. Harcourt in the West to Mombasa and Dar es Salaam in the Section 6 shares these techniques, explaining how they East). Poor households will not be able to rely on and afford helped transcend the traditional trade-off between such alternatives. affordability and cost recovery, and overcame challenges such as water mafias and other barriers to service delivery in Some cities do serve the poor well–we unplanned settlements which lack clear rights of way and studied them to understand how legal tenure to land. To aid the search for better options, a case study approach is used to look for ‘what works’ in African cities in which Conclusions concerning the wider application of the the poor have access to relatively good water services, and approaches and appropriate roles for development partners to understand the reasons for their success. Data from a are drawn in Section 7. sample of 17 cities were analyzed–seven of which provide relatively good service to the poor, and 10 of which provide Appendix B offers suggestions for future research. “typical” lower levels of service. The five case studies then Perspectives on alternative or supplementary service analyze what some cities have been doing differently to providers are shared in Appendix C. Appendix D includes serve the poor better. a description of the utility management effectiveness index. Longitudinal graphs with data from the utilities Overview of the report serving all 17 cities is provided in Appendix E. A list of the The approach to the study is set out in Section 2. Although data sources used is provided in Appendix F. References utility data in Africa are improving, they are often confined are included in Appendix G. to formal “service areas” rather than cities at large, and are 2 II. Methodology The primary objective of this study is to find out how water services were getting water from pipes (figures some cities have been able to provide water services to ranged from 84 percent to 100 percent). In cities poor people relatively well and to learn from these relative where poor people did not receive good service, the successes. Given the lack of a comprehensive dataset on percentage of poor people getting water from pipes service metrics for poor people across African cities, a case was more variable (from 0 percent to 100 percent). study approach was adopted. The aim is to identify the key A comparison of household survey data on reported factors that contributed to good service provision for the access to piped water with utility-reported access to poor in some cities. piped water indicated that most people with piped water are served by the utility. The case study approach • This finding turned attention to utility management The methodology centers on in-depth case studies on five effectiveness as a possible driver of good service to the cities, focusing on political economy and governance in the poor, so the team constructed an index to measure water sector and its wider context, financing of access, and this. Analysis showed that of the seven cities with specific service and tariff offerings that enabled affordable good performance, six had effectively managed access for more poor people. Comparing the five cases utilities. None of the cities with more typical service revealed certain commonalities–for example, in how the had effectively managed utilities. The research then political economy aligned with serving poor people, or focused on the cities with good service to the poor how they financed expansions in access. There also were and effectively managed utilities, to find out what differences–for example, in the strategies used to provide had happened in these cities to enable poor people to affordable access, or in what prompted the prioritization of get good access to water, and the role of the utility in working towards universal access. The report summarizes improving service provision. and analyzes the commonalities and the diversity of the experiences across the five cities. All 17 cities are listed in Table 2.1. Cities with average water • To put these achievements in perspective, data were supply reliability of at least 18 hours per day are listed first, collected initially for a total of 17 cities, making it in descending order of access to improved water by the poor. possible to identify suitable case studies of improvement The seven cities identified as having relatively good service of services generally and to the poor especially, and to to the poor are marked with asterisks (‘*’). compare and contrast them with more “typical” cities. Defining ‘serving the poor’ in terms of access levels Ten typical cities and service quality, seven cities were then identified as The 10 typical cities–Addis Ababa, Kinshasa, Maputo, offering ‘relatively good service’ to the poor, taken to Mombasa, Dar es Salaam, Nairobi, Hargeisa, Tanga, Accra, mean at least 90 percent access to improved water by and Kaduna–were selected to represent the range of service the poorest 40 percent, and availability of water for at levels and contexts across cities in Africa. Data on these least 18 hours per day. The typical cities fell below this cities provide information on the range of access and service level of service to the poor–some by only a little, some to the poor across African cities, and allow for a contrast very substantially. between what was done in cities that have served the poor • The team then examined how poor people in the well and other more typical cities. They also provide some seven cities with good service to the poor get their narratives on possible techniques that can help to bring water. The data showed that the overwhelming service to the poor. majority of poor people in these cities with improved 3 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Methodology Table 2.1: Cities analyzed in study In Niamey, the data were not available in a timely manner City Reliability Access to due to the broader security situation at the time when field (hours/day) improved water visits were undertaken. In Lusaka, the level of services to (poorest 40%) the poor seemed substantially influenced by water services eThekwini (Durban)* 24 100% trusts and not the utility alone, which the team did not Lusaka* 20 99% investigate particularly during the field visit as the focus Ouagadougou* 23 98% as that stage was more on the utility. The Lusaka case is Dakar* 24 97% noted as an area for another study, as it may well juxtapose Nyeri* 24 96% the utility-led approach while the approach here was Niamey* 24 94% more multiinstitutional. In this study, the comparative Kampala* 18 93% data (especially in Section 3) include these two utilities Nairobi 20 86% nonetheless, and the discussion draws selectively on some of their experiences. Hargeisa 20 76% Tanga 24 75% Data sources Accra 18 65% The access and service delivery data come from household Addis Ababa 16 100% surveys, not utilities only, which have three advantages Maputo 12 97% over utility-reported data: households answer the questions Kinshasa 11 93% directly, questions are typically standardized across cities; Mombasa 4 92% and results can be disaggregated by income (or a proxy for Dar es Salaam 8 89% income, such as asset ownership). Kaduna 14 29% * City with relatively good service for the poor. By contrast, utilities estimate coverage. These estimates are often unreliable, especially estimates of the number of Notes: ‘Poorest 40%’ refers to households in the bottom 40 percent of income people getting water from standpipes. distribution in the city, or households in the bottom 40 percent of an asset ownership index if income data was not available. The data on the reliability of water supply come from the ‘Access to improved water’ is as defined by the Joint Monitoring Program utility serving the city, because these data are not included (see Definitions section). in household surveys. Sources: Household surveys (see Table D.1 for list of sources); Utility research (see Table D.2 for full list of sources). In this study, longitudinal data sets (for 10 years or more), reviews of existing literature, participant interviews, and Geographic Information Systems (GIS) images where available, were used to interrogate the dynamics that led to Case study cities and marked relatively good water services for poor people Five of the cities with relatively good service to the poor in some cities. The data and perspectives were shared during were chosen for case studies, partly on the basis of their a video-conference with key utilities, and preliminary service delivery achievements, but also considering the findings shared with individual utilities as well as at several assignment’s specific interest in whether and how utilities international and regional conferences. could secure such services. These cities are: eThekwini (Durban), Ouagadougou, Dakar, Nyeri, and Kampala.4 Limitations of the methodology Two cities with relatively good service to the poor were not A case study approach limits the claims that can be made: selected as case studies, for different reasons. it cannot establish universal relationships with statistical validity or prove that a particular technique is ‘the best’. 4 Each of the case studies is discussed cogently in Appendix A. 4 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Methodology However, causation between a particular factor and good Sanitation to the Urban Poor: A Scoping Study’7–and the service for poor people can be inferred from participants’ recently completed Fecal Sludge Management: Diagnostics own narratives or expert judgment. and Decision Support Tools for Service Delivery in Poor Urban Areas.8 Another important limitation is that cause and effect are always context-dependent. There are limits to what can be Finding perspective on the relevance of the utility concluded about which techniques are likely to work well model was a consideration. The widely reported weak elsewhere, since not all contexts are studied. performance of so many utilities in African cities has led to a questioning of the appropriateness of the utility model. The research is also limited by data quality. Most household This is NOT primarily a study of utility performance but surveys do not contain questions about reliability, of how some cities have been able to provide a relatively adequacy, affordability or provider type. While utilities good service to poor people. For a more general analysis of can supply some of this information, data from utilities are often not available for the exact geographic area covered utility performance, the reader is referred to a related World by a household survey. This is especially true for national Bank study prepared in parallel.9 It is interesting, therefore, utilities, which typically do not disaggregate data by city. that in the cities studied where the poor are served relatively Technical data–such as the layout of a piped network well, traditional utilities are the main service providers to the and how the network has grown over time–are often not poor (see Section 3.2), sometimes–but not always–working available and thus cannot be analyzed. with others to deliver services. Thus the question to be answered became, ‘how did political economy, financing, Three critical aspects of the focus of the managerial, and technical strategies allow the utilities serving study these cities to provide relatively inclusive, reliable service To avoid any misunderstanding on what the study sets out to poor households?’ A distinctive feature of this study is to do and what not, three points need to be made. the elaboration of a political economy framework to help to explain how significant improvements were achieved in The focus is on water, not on sanitation. Improved providing services to poor people. sanitation is essential for public health. By 2030, the SDGs aim for “access to adequate and equitable sanitation A political economy framework is core to the analysis. and hygiene for all” and to “end open defecation”.5 Rather than the typical, normative instruction to countries Increasingly, the World Bank, development partners, and about the way policy and governance should work, the tasks many governments acknowledge that this SDG requires of political economy analysis are to “illuminate constraints multifaceted links between improving water, sanitation, on decision making” and query “why do things work the way and hygiene (WASH), and also with urban planning, they do, and what are the implications for policy?”10 Thus, management and financing. While this study focuses on “…political economy analysis takes the existing situation as improving water supply service and does not address the its starting point, and then focuses on identifying feasible issues of wastewater or sanitation services at any length, it solutions to improving sector performance within existing fully acknowledges the importance of these more integrated incentives structures and relations of power to achieve ‘good approaches. The study focuses primarily on water in the enough governance’”.11 knowledge that there is a growing body of analytical work on improving urban sanitation for the poor in Africa Hawkins, Peter, Isabel Blackett, and Chris Heymans. 2013. Delivering Sanitation to the 7 Urban Poor: A Scoping Study. The World Bank. and globally–such as Faecal Sludge Management: Systems 8 World Bank. Forthcoming. Fecal Sludge Management: Diagnostics and Decision Support Approach for Implementation and Operation6 and ‘Delivering Tools for Service Delivery in Poor Urban Areas. 9 Van den Berg, Caroline, and Alexander Danilenko. Forthcoming. Performance of Water and Wastewater Utilities in Africa. The World Bank. 5 Sustainable Development Goal 6.2. 10 Levy, Brian. 2014. Working with the Grain: Integrating Governance and Growth in 6 Strande, Linda, Mariska Ronteltap, and Damir Brdjanovic. 2014. Faecal Sludge Development Strategies, 25. Oxford University Press. Management: Systems Approach for Implementation and Operation. London: IWA 11 Kooy, Michelle, and Daniel Harris. 2012. Political Economy Analysis for Operations Publishing. in Water and Sanitation: A Guidance Note. ODI. 5 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Methodology For water utilities, problem-driven political economy starts cities that allowed them to have utilities that serve the poor from the understanding that they, like other parastatals, do relatively well. not necessarily live up to their formal principal objective of the provision of services as expressed in policies and Key points stakeholder frameworks. Decisions may be made not on A case study approach allows for the development of some the basis of collective goods provision,12 but may be more insights into the reality of service delivery to the poor in affected by how and why competing interests engage in the selected African cities. The cases study cities were selected utilities or their environments.13 on the basis of their relative good performance in serving poor people. The proposition is that insights gained from As a result, the poor–who are often the least powerful cases of relative success can be useful in providing pointers constituency and the most likely to be marginalized by in how services to poor people might be improved in elite-level machinations–bear the brunt of these costs in other cities. This is particularly relevant in a context where lacking access to reliable, safe, piped water sources.14 Elites relatively good performance is the exception rather than the or other internal or external interest groups often are better norm. If it is accepted that the political-economy dynamics positioned to benefit from rent-seeking, extracting public prevailing in many African countries and cities typically resources from the utility through corrupt contracts or constrain good performance in serving poor people, then it patronage, such as using the utility as a “jobs bank”. These is particularly important to examine how these constraints activities have more immediate pay-offs for incumbent have been transcended in particular cases to enable good office-bearers that may be more attractive than any long run performance. In other words, the methodology is premised benefits that may eventually derive from popular support on the proposition that the answers (to service improvements generated by providing good services to more people. for poor people) are more likely to lie in the exceptions As a result, the political economy ‘problem’ that the case rather than a study of the norm, and this why the study studies seek to understand is what was distinctive about has looked at a few relatively successful case studies with a the situation and political economy in the five case study particular focus on political economy. 12 Manghee, Seema, and Alice Poole. 2012. Approaches to Conducting Political Economy Analysis in the Urban Water Sector. The World Bank. 13 Olson, Mancur. 1982. The Rise and Decline of Nations: Economic Growth, Stagflation, and Social Rigidities. 14 Bratton, Michael, and Carolyn Logan. 2013. “Voters but not yet Citizens.” In Voting and Democratic Citizenship in Africa, edited by Michael Bratton. Boulder: Lynne Rienner. 6 iii. Water Service for Poor Households in African Cities This chapter examines access to water services provision 3.1.1 Access to improved water services (access and quality) in 17 cities. The selection of cities is Using household survey data, access to improved water explained in Section 2 (Methodology). Household survey was calculated for each of the 17 cities, for the poorest data at a city level are used, together with utility-reported 40 percent of households. The data are shown in Figure service reliability data and utility management effectiveness. 3.1. Access to improved water is disaggregated into three Reliable data of this kind are not widely available and this categories: piped to premise (dark blue), shared tap (light is the reason the analysis is done for only 17 cities. The blue), and other improved (pink). Access to improved source of water for poor people (piped or not, and utility- water was analyzed because this is what household surveys provided or not) is examined and compared with the data measure. Note that improved water does not necessarily on access and quality of water services for poor people, and mean safe water (see Box 3.1). Note also that reliability of utility management effectiveness. The analysis shows that supply is not incorporated in the measure. where cities have relatively good access to water (access and quality) for poor people, this is predominantly provided by Cities with good access provide access to improved water to effectively managed utilities. On the basis of this analysis, 97 percent of the poor, on average. As a group they illustrate five cities were chosen to undertake more detailed case what is possible. On the other hand, in seven of the 10 studies which are the basis of the sections that follow. typical cities, less than 90 percent of poor households have access to improved water. 3.1 Current situation Data on service to the poor in the seven cities with Among the typical performers, Hargeisa is doing very well relatively good service are presented together with data for a city in a fragile state (see Box 4.3 for more on Hargeisa). from the 10 typical cities. This gives a reasonable sense of However, water supply for the poor remains far from ideal. the level and range of water service experienced by poor Just 76 percent have access to improved water, the others people in African cities, as well as what is possible. The primarily accessing water from tanker trucks that charge a data cover access for the poor, convenience, continuity of premium. About 55 percent have access to public taps. supply, and affordability. Figure 3.1: Access to different levels of service, poorest 40 percent * City with relatively good service to the poor. Source: Household surveys (see Table D.1 for list of sources). 7 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Water Service for Poor Households in African Cities In Kinshasa, another conflict-affected state, access to improved water by the poor is good at 93 percent, but Box 3.1: SDGs ‘safe and affordable’ service is inconsistent and the utility has low indicators versus ‘improved’ on management effectiveness. Kaduna–a city of almost 1 The Sustainable Development Goals (SDGs) target million people in Nigeria, an oil-rich country–surprisingly safe and affordable water for all. Unfortunately, data provides worse service than either Hargeisa or Kinshasa. consistent with this definition are not yet being widely There, the poor primarily get water from wells, less than a collected and reported. In the meantime, access to third of which are protected. improved water (the target of the MDGs) can be analyzed, but lacks the qualitative perspective that 3.1.2 Access to piped water to the premises is the major additionality in the SDGs. The limitations Recent research indicates that households benefit more to the improved water definition under the SDGs from water piped to the premises than they do from shared should, however, also be noted: taps from which the water must be carried back. Howard • Many of the ‘improved’ sources (and often, even and Bartram, for example, showed that the volume of water piped sources) are not delivering safe water used by households depends on accessibility as determined in urban areas (even if they may do so in rural primarily by distance and time, and that the quantities used areas). when water must be carried may not be sufficient to reduce • No measure of adequacy. health risks.15 • No measure of convenience. • No measure of affordability. Evans et al. found a clear difference between the volumes of water consumed when water is available on-site rather Most African utilities simply do not publish reliable than at a distance. The conclusion was that “at-home water data on the quality and safety of the drinking water supply has significant, measurable benefits when compared they supply. However, water sector professionals with shared water supply outside the home, (provided that know that wherever supply is intermittent, there is a the service provided is reliable enough to ensure access strong chance of piped water being contaminated to adequate quantities of water when required). Reliable as polluted groundwater infiltrates the piped at-home water supply results in higher volumes of water network during periods of depressurization. consumed, greater practice of key hygiene behaviors, a Most African water utilities report that they have reduction in musculo-skeletal impacts associated with intermittent supply. carrying water from outside the home, and improved water quality”.16 Access by the poor to water piped to the premises is shown in Figure 3.2. Water which does not meet drinking water standards may still be safe for other purposes such In Nyeri and Dakar, most poor people have water piped to as washing clothes–and can be purified for drinking their premises, at 88 percent and 75 percent, respectively. purposes using in-home filters, purification tabs or These cities show what is possible. However, in 12 of the 17 boiling. These techniques are widely used in better- cities studied, less than one-third of the poor have access to off African households, but may be more difficult water piped to the premises. In the 10 typical cities, access for poor families with few facilities and little extra to water piped to the premises for the poor ranges from 0 income. percent to 48 percent. There is thus no doubt that the proportion of poor people accessing safe water is lower than the access to ‘improved’ water would suggest, and in 15 Howard, G., and J. Bartram. 2003. Domestic Water Quantity, Service Level and Health. Geneva: World Health Organization. many cities also lower than the percentage able to 16 Evans, Barbara., Jamie Bartram, Paul Hunter, Ashley Rhoderick Williams, Jo-Anne access piped water. Geere, Batsi Majuru, Laura Bates, Michael Fisher, Alycia Overbo, and Wolf-Peter Schmidt. 2013. Public Health and Social Benefits of at-house Water Supplies. 8 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Water Service for Poor Households in African Cities Figure 3.2: Access to water piped to the premises, poorest 40 percent * City with relatively good service to the poor. Source: Household surveys (see Table D.1 for list of sources). 3.1.3 Access to water from a public tap opted for standposts as an interim option see this as realistic While water piped to the premises may seem a good pragmatism that positions them for improving services standard to aim for, this has not been achieved fully in most affordably. Also, the definitions of both “premise” and African cities, even among those that are doing relatively “public tap” are not always clear. In countries with high well in serving the poor. Often the next best thing is access levels of tenancy the landowner often gets a connection to to a public or shared tap. Some cities do especially well in the “premise” and then provides a tap for all tenants in the ensuring that poor households receive water in this way, yard to share. In other cities one tenant in the premise (or and some argue that public standpipes assist them in yard) that can afford a connection may sell water to his or securing payment from users and is often close enough to her “neighbors” who would otherwise have to walk to the the properties served that the inconvenience of walking standpipe. Nonetheless, Figure 3.3 provides a perspective to fetch water is largely mediated. This is an unresolved on the spread among the cities studied. argument though, the more so because utilities that have Figure 3.3: Access to public tap, poorest 40 percent * City with relatively good service to the poor. Source: Household surveys (see Table D.1 for list of sources). 9 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Water Service for Poor Households in African Cities In four of the cities that provide relatively good service to the poor (Niamey, Ouagadougou, Lusaka, and Kampala), more than 70 percent access water through a public tap. 3.1.4 Time poor people must spend to get water A public tap is only a good source if it is conveniently located. Convenience can be measured by the average roundtrip time spent collecting water from a public tap. Figure 3.4 shows the average roundtrip time for the poor to get water from a public tap. (The percentage of the poorest 40 percent accessing water from a public tap is included in parentheses next to the city name.) In four of the cities that provide good service to the poor (Niamey, Ouagadougou, Lusaka, and Kampala), more than 70 percent access water through a public tap. This water is fairly convenient, with roundtrip times averaging from 12 minutes (Niamey) to 21 minutes (Kampala), on average. By contrast, four of the typical cities have average collection times of 22 minutes or more. In Hargeisa, this figure is 42 minutes. All the utilities that have improved services to the poor have focused on piped water supply. (Photo Credit: ONEA, Ouagadougou, Burkina Faso). Figure 3.4: Average roundtrip time to get water from public tap, poorest 40 percent * City with relatively good service to the poor. Notes: The value in parentheses represents the percentage of people in the poorest 40 percent with access to a public tap. Kaduna is not included in these graphs because 0 percent of the poor have access to a public tap. eThekwini is not included because data on roundtrip times were not available. Source: Household surveys (see Table D.1 for list of sources). 10 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Water Service for Poor Households in African Cities 3.1.5 Continuity of water service 3.1.6 Affordability of water for poor people Continuity of service is the number of hours per day that The best way to measure the affordability of water in a city water is available, on average. Continuity matters because is to use survey data in which households report the actual poor households may not consume adequate quantities amount spent on water in a given period and their income of water if it is only available a few days per week. In or total expenditure in that period. Such data are available addition, wherever supply is intermittent, there is a strong for the three Kenyan cities in the sample from the Kenya chance of piped water being contaminated as polluted Baseline “State of the City” survey, as shown in Table 3.1. groundwater infiltrates the piped network during periods of depressurization. Nairobi’s poor households spend the highest proportion of their monthly income on water–10 percent. This figure is Unfortunately, standard household surveys do not ask lower in Mombasa and Nyeri, at 7 percent and 6 percent, people about how many hours a day or days a week they are respectively. able to access water. Still, in cities were most poor people rely on piped water, it is possible to estimate the hours of For the remainder of the sample, data on the actual service with data from the utility.17 These estimates are shown in Figure 3.5. amount spent on water were not available. This is an area that household surveys could usefully include questions Clearly, it is possible to provide water to poor people in on in the future. African cities for 24 hours a day, as Dakar, eThekwini (Durban), and Nyeri show. However, this is far from typical. Table 3.1: Proportion of monthly income spent on In Maputo, Addis Ababa, Kaduna, Kinshasa, Dar es Salaam, water by poorest 40 percent, Kenya and Mombasa, the utilities serving the city provide supply City Income/ Amount Proportion of month (KShs) spent on monthly 17 hours a day or less. Among the most difficult places for water/month income spent poor people to get water reliably is Mombasa, where service (KShs) on water is available every second day, for around six to eight hours. Mombasa 8,918 616 7% Nairobi 6,271 596 10% Nyeri* 8,543 509 6% 17 The utilities serving Ouagadougou, Dakar, Niamey, Kampala, and Accra are national utilities. Reliability data are for the entire utility service area. These cities * City with relatively good service to the poor. typically account for more than half of all utility customers. Source: Baseline: “State of the City” survey (Kenya), 2012–2013. Figure 3.5: Continuity (hours of water supply service per day) * City with relatively good service to the poor. Continuity data are for the entire utility service area. The utilities serving Ouagadougou, Dakar, Niamey, Kampala, and Accra are national utilities. Source: Utility research (see Table D.2 for full list of sources). 11 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Water Service for Poor Households in African Cities 3.2 How some cities serve the poor Figure 3.6 shows the extent of Dakar’s network throughout Understanding the current situation of water service for most of the city, which has greatly enhanced the city’s ability poor households in African cities, we need to explore how to reach more people in close proximity to where they live. poor people are served in the seven cities in which they receive relatively good water services, and how this differs Where access is good, utilities are the main service from other cities. providers for poor people Having established that piped technologies are relied on In cities with good service, water is provided to poor in the cities with relatively good service, the next question people through pipes is about the broad institutional arrangements. Are poor In the seven cities with relatively good water service for people primarily supplied by small scale entrepreneurs poor people, an average of 90 percent of the poor access running piped networks? Or by community organizations? water from pipes. In Dakar and Niamey, 94 percent and Or by conventional utilities? 97 percent, respectively, of poor people have improved water. All of those poor people are supplied by pipe. This Household surveys do not identify the type of organization is a valuable finding, in that it shows these cities are not which is supplying water. There is no other good data source relying on boreholes, or rainwater harvesting, or other less with this information, either. One way to gauge the type of conventional technologies, for providing water services to entity providing service is to compare access estimates from poor people. They seem to say that while these alternative household surveys with utility-estimated coverage rates. modes of supplying water may have benefits, such as This comparison is shown in Table 3.2 (because utilities conservation, some cost saving and others, they are difficult do not report service to poor people separately, it is only to move to scale. Cities that have reached the poor on possible to make this comparison using access levels for the scale have invariably developed their networks, whether total population). to connect individual properties and households on the premises, or through public standposts. Figure 3.6: Dakar provides water piped to the premises to (almost) all Malika Legend Piped Network Keur Massar Poorest 40% Improved Water Sources Percent Access 81% - 90% Rufisque 91% - 100% Malika Keur Massar Ngor ± Rufisque Yoff Daliford Ngor Yoff 0 2 4 8 Daliford Mbao Kilometers Mbao Rufisque Est Access to Improved Rufisque Est Water Sources Hann Bel-Air Water Access (2014) Hann Bel-Air Dakar, Senegal Data Sources: (1) ANSD (2) SONES Plateau Map Authors: Baker, Homann & Krishnan Plateau Date: 12/18/2015 Figure 3 Esri, HERE, DeLorme, MapmyIndia, © OpenStreetMap contributors, and the GIS user community 12 Esri, HERE, DeLorme, MapmyIndia, © OpenStreetMap contributors, and the GIS user commun Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Water Service for Poor Households in African Cities Table 3.2: Access to piped water, compared In Kampala and Nyeri, the utilities’ estimates of coverage to access to utility-provided water are lower than the access to piped services reported by City Access to piped Utility estimated household surveys. The utilities may be underestimating water, household coverage their effective coverage because they do not include illegal survey data connections among households served, or because they Durban* 97% 97% underestimate numbers served by each shared tap. In any Lusaka* 92% 87% case, the statistics suggest that all those with piped supply Ouagadougou* 94% are served by utilities. This is consistent with interviews and Dakar* 96% 100% field visits suggesting few, if any, alternative piped providers Nyeri* 94% 85% in these cities. Niamey* 95% Kampala* 83% 78% Utility coverage data were not available for the cities of Nairobi 83% 75% Niamey and Ouagadougou. We do know, however, that Hargeisa 60% 50% in Ouagadougou a growing number of poor people in informal settlements are served by small providers. These Tanga 71% 98% small providers on-sell utility water under an agency Addis Ababa 99% 89% arranged with the utility, as described in Section 6. Maputo 94% 44% Kinshasa 88% In conclusion, where the poor are relatively well served, service Mombasa 82% 57% is provided by utilities. Service is usually provided directly to Dar es Salaam 62% 46% customers, though this may be supplemented with on-selling Accra 57% 56% through small scale providers, as in Ouagadougou. Kaduna 18% 47% * City with relatively good service to the poor. In the typical cities, alternative providers have a Notes: larger role City-level coverage data was not available for three of the national By contrast, the typical cities have a wider range of water utilities. service providers. In Maputo, for instance, 94 percent of Source: Household surveys (see Table D.1 for list of sources); utility households (all income levels) report access to piped water, research (see Table D.2 for full list of sources). whereas the utility serving Maputo (AdeM) reports just 44 percent coverage. Small scale providers account for the other 50 percent. About 500 entrepreneurs manage over 800 piped A weakness in this method is that utilities’ methods of systems, providing mostly yard taps. In these cities, small estimating coverage vary, and may be unreliable, especially scale providers are important in filling the gap left by the as regards the number of people getting water from utility utility, and contribute to service to poor households being standposts. The results therefore need to be cross-checked better than it otherwise would be. Additional information against local knowledge of the prevalence of small scale piped water providers in the city. about this arrangement in Maputo is included in Box 6.1. In Nairobi, by contrast, most of the poor do obtain water The data in Table 3.2 suggest that, in Durban and Dakar, from the utility. However, because the utility has not everyone with piped water gets it from the utility. This extended service in a widespread manner, access to improved matches local reports that there are few, if any, small-scale water remains below 90 percent for poor people. piped water operators in these cities. Lusaka, by contrast, has 5 percent more of the population reporting access to piped water than the utility reports that it serves. This is consistent with field research showing significant involvement of community trusts in supplying water in some areas. 13 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Water Service for Poor Households in African Cities 3.2.1 Factors distinguishing utilities that serve the Strikingly, of the seven cities with relatively good service poor well from others to the poor, six have utilities that are good performers in We have seen that, in cities where poor people get good terms of management effectiveness. Just one, Lusaka, has a water services, the service is almost always provided by a composite management effectiveness score below 80. The utility. At the same time, cities where the poor are less well 10 cities with more typical levels of service for poor people served also have utilities. Just having a utility is clearly not all have utilities with more typical levels of performance. enough to get good water service to the poor. We need to understand what makes some utilities serve the poor well, To be sure that this striking result is not an artefact of while most do not. how the measures were calculated, we tested other cut-off points, and other indicators of utility performance. Among One factor which stood out as making a difference is how the indicators tested was one that measures general service effectively the utility is managed. A utility management performance in addition to management effectiveness, effectiveness indicator was developed. This indicator and one measure of utility performance developed by is a composite of the most important indicators for Performance of Water and Wastewater Utilities in Africa.18 We cost-effectiveness and financial sustainability for a also tested other measures of service to the poor. While one utility: nonrevenue water (NRW), collection ratio, staff or two cities moved from one quadrant to another in these productivity, and the operating cost recovery ratio (for tests, the overall conclusion remained the same. details, see Appendix E). Utilities that scored above 80 out of 100 on this measure were classified as good performers, the Van den Berg, Caroline, and Alexander Danilenko. Forthcoming. Performance of 18 rest as typical performers. Figure 3.7 shows the relationship Water and Wastewater Utilities in Africa. World Bank. between service for the poor and management effectiveness of the utility. Figure 3.7: Relationship between service to the poor and utility management effectiveness Aggregate utility management effectiveness score Service Typical (less than 80) Good (greater than or equal to 80) Total High Dakar 7 (greater than or equal to 90% access to Lusaka Durban improved water for poor, and reliability 18 Kampala hours per day or more) Niamey Nyeri Ouagadougou Low Accra 10 (Less than 90% access to improved water for Addis Ababa poor, or reliability less than 18 hours per day) Dar es Salaam Hargeisa Kaduna Kinshasa Maputo Mombasa Nairobi Tanga Total 11 6 17 Source: Household surveys (see Table D.1 for list of sources); utility research (see Table D.2 for full list of sources). 14 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Water Service for Poor Households in African Cities A broader sample of cities would probably show some cities 3.3 Key points that serve the poor well without having effectively managed The analysis of 17 cities shows that some cities have achieved utilities, and some effectively managed utilities do not serve over 90 percent access to improved water for the poor, in the poor well. The evidence does not prove that a well- contrast to a range of more typical performance. In the seven performing water utility is necessary for serving the poor. cities that provided both good access to the poor (more than The evidence here shows, however, that well-performing 90 percent) and reliable access (more than 18 hours per day), water utilities can and do serve poor people effectively. For the water was provided predominantly by well-managed six of the seven cities that serve the poor relatively well, a utilities in six of these seven cities. Five of these six cities (and well-performing utility was the route to service provision. their utilities) were studied in more detail as case studies, To put it another way, the overwhelming majority of poor namely Dakar, Durban, Kampala, Nyeri, and Ouagadougou. people with good service in the cities we studied got that (Timely data for the sixth city, Niamey, and its utility was service from a well-managed utility. not obtained.) See Appendix A. The sections that follow are based on the findings of these five case studies. 15 IV. The Political Economy of Improving Water Service for the Urban Poor Many critics say that African parastatals, including water networks and alliances for change and start to embed the utilities, are run to benefit vested political and economic reform legacy, even if not immediately comprehensively. interest groups, with little regard for the poor. The political In the cases looked at here, a political leader involved in economy question of interest, therefore, is ‘why and how do starting the reform was typically secure, and wished to some utilities provide good service to the poor’, despite this improve service sustainably. A technical leader, often from inauspicious environment? Five cases–Dakar, eThekwini, the utility, had a vision and entrepreneurial spirit to seize Kampala, Nyeri, and Ouagadougou–were analyzed to opportunities and influence the environment, forging a understand why and how such “islands of excellence” exist mutually beneficial partnership with the political leader. (see Appendix A for the full case studies). Building momentum. Public entrepreneurs often build In each case, a distinct turnaround was identified–the reform momentum by carving out sufficient bureaucratic utility serving the city once provided inferior or no service space to operate, even against the odds. Daniel Carpenter’s at all to the poor, and now provides relatively good service. (2001) classic study of how the United States of America The findings from the cases flow from three distinct but built agency capability in the late 19th and early 20th interrelated avenues of inquiry: how reforms are started, how centuries highlights the role of such “public entrepreneurs” reform momentum is built, and how successful outcomes that are able to build the internal capacity of the agency are sustained. Figure 4.1 shows these phases. and forge external alliances, thereby constructing a new coalition that will insulate the ongoing reforms from self- Starting reform. A catalytic event–such as a cholera interested “predatory” rivals.19 outbreak, a water shortage or an upheaval in the broader political environment–can create a space for reform. 19 Carpenter, Daniel. 2001. The Forging of Bureaucratic Autonomy: Reputations, Net- Exploiting this space for reform requires leadership, but works, and Policy Innovation in Executive Agencies, 1862-1928. Princeton: Princeton University Press; Levy. Working with the Grain: Integrating Governance and Growth in whatever the leaders do needs to build up to developing Development Strategies, 138–58. Figure 4.1: Starting the reform, building momentum, and sustaining the reform Starting the reform Building the momentum Sustaining the reform Build Preserve Professional technical Internal capability and culture leader Catalytic event or space for Forge Embed reform Secure Alliances external to the utility political leader Create Strengthen Formal rules and structures 16 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | The Political Economy of Improving Water Service for the Urban Poor Sustaining reform. Success begets success, but also brings utility more attractive to elites and other interest groups. risks and countervailing factors, for three key reasons. When utilities gain access to finance to implement capital First, as the leaders who created the success depart, their expenditure projects worth tens or hundreds of millions replacements may not share or understand the original of dollars, the stakes are even higher. vision, or may lack the skills and charisma to carry the organization forward. Second, success may actually reduce In these cases, sustaining reform was made possible by the perceived political value of high quality water services. continuing to do the two things initiated in the ‘building When memories of a water crisis have receded, widespread momentum’ phase–namely, by strengthening this internal reliable water supply may come to be taken for granted capacity, and embedding external alliances. Formal rules by both the public and politicians, and no longer seen as or structures (institutions) also need to be strengthened something to fight for. Third, the benefits of predation over time. A brief description of the six key elements that on the utility will increase once it becomes successful. helped start the reform, build momentum, and sustain As utilities expand service to the poor and others, they reform in each of the five cases are listed in Table 4.1, with generate operating cash surplus, making controlling the additional detail in Sections 4.1 (starting reform) and 4.2 Box 4.1: Institutions and individuals in water sector improvements In many developing countries, the prevailing lack of checks and balances (as would be provided by strong formal institutions), means that governments do not always make credible commitments in those polities (Keefer, 2004).20 However, the basis for credibility may emerge from personalized deals, often coupled with third party enforcement mechanisms. Such personalization alone may have limited impact on reform if the leaders disregard the institutional constraints and incentives, as well as alliances that make change happen. But it also highlights the importance of attention to informal institutions in political economy research on developing countries (see Levy, 2014: 153).21 Seminal analyses in political economy demonstrate the importance of institutions in structuring politics and relations between actors (North, 1990; Steinmo et al., 1992). Development practitioners have incorporated institutionalist perspectives into their policies and engagement inasmuch as they typically emphasize the development of formal rules, norms and practices, and these structural dimensions can be shaped, to some extent, by specific donor interventions. Indeed, there is a reluctance on the part of development partners to focus on individuals and personalities, not least because these factors lack generalizability and, more importantly, replicability. In the case of successful reforms in the urban water sector, the role of individual utility leaders, although not discounted altogether, is seen as unique. The common responses to NWSC’s William Muhairwe and eThekwini’s Neil MacLeod exemplify this tendency: Those averse to focusing on personalities note that ‘of course if we had 5 Williams and 5 Neils we could reform 10 water utilities–but we don’t–we can’t simply create these people and insert them. That’s why we need to focus on institutional and structural factors.’ The successful reform cases of utilities in Ouagadougou, Nyeri, Dakar, Kampala, and Durban suggest that both institutions and personalities are important. It has been the personalities that have actually driven institutional reform to improve service delivery and prioritize the poor–but to become sustainable, institutional systems had to be transformed so that they would work differently even after the reformer has set in motion and consolidated change. Those policy makers, development agencies or others who want to promote turnaround, need to be skilled at identifying and backing the right type of people, but also need to understand that they need support in their respective political economy environments–none of which would ever be exactly the same and replicable elsewhere. 20 Keefer, Philip. 2004. What does Political Economy Tell Us About Economic Development– and Vice Versa? The World Bank. 21 In general, our analysis is consistent with this view that institutions that are stronger, more inclusive, and more effective in terms of governance, produce better outcomes. This relationship does not appear in all contexts, however; Uganda is a notable outlier. There, good governance institutions are lacking, but the government can make credible commitments. 17 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | The Political Economy of Improving Water Service for the Urban Poor Table 4.1: Key elements in starting reform, building momentum, and sustaining reform Starting reform Building momentum and sustaining reform City Catalytic event Political Technical Internal External Formal rules Current utility, country or space for leader leader capacity and alliances and structures utility reform culture leadership Ouagadougou Early- to mid- Salif Diallo, Mamadou ISO certified Supervision Performance Hamado ONEA, 1990s. Water Minister Lamine (2009) Committee Contract Ouedraogo Burkina Faso shortage and Environment Kouate, Strategic (representatives Supervision and low water and Water Managing planning of consumers, Committee Moumouni coverage in (1995–1999) Director (1995– Service contract government, (representatives Sawadogo Ouagadougou 2005) for commercial NGOs, and of consumers, and Bobo- Dieudonne functions (2001– development government, Dioulasso Sawadogo, 2006) partners) NGOs, and Manager and development Secretary partners) General Nyeri Early- to mid- Mayor Eng. Joshua Transparent and Customers: WASREB Peter NYEWASCO 1990s. Water Jackson Nguiguti audited systems barazas, annual (regulator) Gichaaga Kenya rationing and Wanjage (1995–2014) Strategic open days, and poor water planning good service quality problem fosters “brand loyalty” Kampala Late 1990s. President William Early (external Political: Parliamentary Silver NWSC, Donor support Museveni Muhairwe and internal) continued approval of the Mugisha Uganda for PPP coupled (1986– (1998–2011) management support of tariff indexation with Ugandan present) contracts President (2004) and opposition to (1998–2004), Museveni political support PPP. Conflict and internally Customers: Pro- for “Water for created space delegated Poor Unit All” program for a credible management more recently alternative, contracts in a situation (2004–present) where Uganda Corporate needed the training facility development partners eThekwini 1992, Need Obed Mlaba Neil Macleod, Incentivized Customers: Incentivized Ednick (Durban) to serve all Executive Managing contracts agreed to contracts Msweli eThekwini, citizens in the Mayor (1996– Director (1994– with senior raise free with senior South Africa metro area in a 2011) 2014) management monthly water management way that undid team allowances from teams the legacy Strategic 6 m3 to 9 m3 of service planning discrimination Training under Apartheid Dakar Mid-1990s. Minister Madio Fall, Combination Development Affermage Abdoul Bal SDE/ Solve water of Water, Water Director of continued partners: World contract for SDE and Cheikh SONES, shortage in Mamadou at Ministry of leadership in the Bank, AFD Concession Sall Senegal Dakar Faye (1992– Water (1992– public sector contract for 2000) 2003) (the MD of SONES SONES was the previous MD of SONEES) and international, private sector expertise Source: Interviews with participants, Professor Scott Taylor. 18 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | The Political Economy of Improving Water Service for the Urban Poor (building momentum and sustaining reform). Box 4.3 With the support of development partners, a reform plan explores the applicability of these findings in fragile and was developed.25 In the short-term, additional boreholes conflict-affected states. were constructed and an existing treatment plant was expanded. A public-private partnership (PPP) contract was 4.1 Starting reform signed, giving a private operator (Sénégalaise des Eaux, SDE) In each turnaround case, three factors were important in the responsibility for providing service to customers and for starting reforms: a catalytic event, a political leader, and a operating and maintaining the infrastructure. utility leader. Box 4.1 elaborates the reasoning that shows the importance of individual actors, but also underscores Importantly, political elites in the ruling Parti Socialiste that their roles need to be viewed, developed, and supported (PS) as well as the opposition were able to reach consensus within their institutional settings. about both the importance and direction of reforms. To secure World Bank support for reforming the sector, the For each city, how political and technical leaders supported government had to commit to making the water sector water sector reform, in the context of a catalytic event, is financially sustainable–which it did. The President of described below. Senegal at that time, Abdou Diouf, was in a relatively Dakar secure position, and the PS had been in power since 1960. In 1994, Dakar suffered a water shortage. The utility serving Potential opponents were included in government to build the city was only able to supply 60 percent of demand. ownership of the reforms across the political spectrum. Water was rationed to 16 hours a day, with low pressure and Development partners supported this approach because, contamination of the piped supply in some areas. Rapid once the PS agreed to water sector reform, the ready-made population growth (more than 5 percent per annum) and consensus made proposed reforms easier to enact.26 This rapidly depleting water resources spelled imminent crisis. also enabled opposition leaders to maintain momentum around water sector reforms once they were elected to A small group of leading figures–from the Water Directorate, government in 2000. the private operator, and the World Bank–was “behind the reforms in Senegal and can collectively be credited for their Ouagadougou success”.22 Their personal desire to improve the quality of In the 1990s, Ouagadougou also faced a severe water public services for all residents, making these services more shortage. The city population had doubled from 1985 to inclusive, was one reason they were so successful. Among 2000, but the development of water service infrastructure these leaders, Madio Fall (Water Director at the Ministry had not kept pace with increasing demand. By 2000, just of Water from 1992 to 2003) was the key bureaucratic half the population had access to piped water (through entrepreneur involved in Dakar’s reform. Described as ‘a individual taps or communal standpipes).27 l’Office national towering and respected figure within the sector, Madio Fall de l’eau et de l’assainissement (ONEA), the utility responsible had the full support of the Minister of Water, Mr. Mamadou for providing water to Ouagadougou, was performing poorly Faye, who himself remained in place from 1992 until the and thus lacked the cash needed for service improvement. presidential elections in 2000’.23 Fall himself says he was Salif Diallo (Burkina Faso’s Minister of Environment and motivated ‘by the urge to improve the public side of water Water from 1995 to 1999) and Mamadou Lamine Kouate administration and delivery’, and “to ‘stop seeing women (Managing Director (MD) of ONEA from 1995 to 2005) on the streets of Dakar spending hours fetching water’.”24 were the political and technical leaders, respectively, that were critical to starting the water sector reform. It 22 Trémolet. 2006. Case Study on Senegal’s Water and Sanitation Sector Economic Regula- tion: Final Report. A Report to the World Bank on behalf of Castalia Advisors. May. 25 The World Bank’s contribution was considerable, but the success hinged largely on By Trémolet’s account, the list includes Fall and his two director generals in the Water the dedicated and creative responses by Senegalese role-players. Directorate; Water Minister Mamadou Faye; World Bank officials Jan Janssens, Matar 26 Beck, Linda. 1999. “Senegal’s Enlarged Presidential Majority: Deepening Democracy Fall and Fadel Ndaw; Mamadou Dia at SDE; and independent conciliator Jan Dirickx. or Detour?” In State, Conflict & Democracy in Africa, edited by Richard Joseph, 214. 23 Trémolet, 55. Boulder: Lynne Rienner. 24 Trémolet, 81. 27 World Bank. 2001. “Project Appraisal Document, Ouagadougou Water Supply Project.” 19 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | The Political Economy of Improving Water Service for the Urban Poor can be inferred that Burkina Faso’s then-president, the would expect that the utility would have been subject to strongman Blaise Compaoré, gave his political blessing. penetration by national and local political elites. However, Diallo and Kouate resisted suggestions from the World this was not the case in Nyeri in 1995. Mayor Wanjage Bank to introduce a private operator to manage the service. was a civic-minded reformer, with an uncommon level Instead, ONEA remained a government-owned, limited of autonomy. The utility’s small size and weak financial liability company. From 2001 to 2006, technical assistance standing minimized vested interests in continuation of from a private operator (Veolia) was provided through the status quo (low coverage and intermittent service) a performance-based service contract.28 In addition, for patronage, corruption or other forms of predation. In a Supervision Committee–comprising representatives other words, it seems the political benefits of improved of consumers, government, NGOs, and development service outweighed the political costs from loss of rent- partners–was established. The Supervision Committee seeking opportunities. It may also have helped that Nyeri monitors both ONEA and the government’s performance is a relatively ethnically and culturally homogenous town, against the Contract Plan (three-year performance contract), minimizing inter-group rivalries that could otherwise have but it also is an example of a formal structure that developed interfered with a drive for universal service and the use of an important informal role in connecting the key actors in cross-subsidization to achieve it. a common forum, from where they have been able to guide and support reform and progress in alliance mode. eThekwini (Durban) In Durban, South Africa, in the 1990s, the entire Nyeri sociopolitical context in which utility services were In the early 1990s, municipal water was rationed, unreliable, provided shifted dramatically. Under 46 years of apartheid and unsafe to drink. Although reliable data on the situation rule, black people mostly lived in separate areas where water before 1995 are not available, those who lived through the services were poor or nonexistent. The catalytic event, period recall that infrastructure investment had not kept therefore, was democratic transition in South Africa in up with population growth. Water was rationed by limiting 1994, from an exclusive race-based franchise to universal supply to a few hours per day. The poor families living in franchise. Reform of local government amalgamated the growing informal settlements lacked access to the piped previously racially separate local governments. Shortly water network. thereafter, the 1996 Constitution included a bill of rights that required the state to progressively realize the right The reforms at Nyeri Water and Sewerage to water. The Durban water department, previously Company (NYEWASCO) were initiated through the providing a high level of service to the city center and combined efforts of a political and a technical leader. The affluent white suburbs, now had the responsibility to political impetus for reform came from the mayor, Jackson manage and extend services over a much wider and poorer Wanjage, who had support from the municipal council. area. The service boundary was extended by 68 percent to Technical and managerial input came from the Municipal include more rural areas that added a further 9 percent to Engineer, Joseph Nguiguti, who at the time was responsible the population of 3.6 million people. for water supply because of his role in running all of the municipality’s engineering services. Together, they persuaded eThekwini provides a quintessential example of “public GTZ to let Nyeri join an existing program supporting entrepreneurship” under the leadership of Managing the creation of municipal-owned public companies with Director Neil Macleod, an engineer with a Master’s in autonomous management. Business Administration. In the early 1990s, Macleod acted early to reorient the water service towards a pro-poor focus, At the time of reform, predation and neo-patrimonial even before the first democratic elections in 1994. The leadership were reportedly widespread in Kenya, and one turnaround began with institutional reforms to incorporate multiple municipal water departments into a single 28 The contract with Veolia is described in more detail in Section 4.2. A description of metropolitan water department within the metropolitan the loan secured for ONEA is provided in Section 5.1. government. The utility moved rapidly to incorporate the 20 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | The Political Economy of Improving Water Service for the Urban Poor To get the water flowing reliably, safely and for all people in the city, creating conducive political conditions and developing technical, managarial and financial strategies and options mostly take time and tactical acumen. (Photo Credit: The World Bank). various township utilities into its management structure. sufficiently secure politically to establish and maintain the Macleod focused on expanding and improving service, and Water and Sanitation unit as a ‘semi-autonomous’ entity constantly innovated institutionally. He reminisces that he within the municipal administration. The roots of this felt like he was “just trying to run faster than everyone else” arrangement have been so firm that it has been in existence in putting together institutional, financial, and management and successful for over 20 years. models. Macleod’s success at starting reforms illustrates how speed can help set the agenda. Kampala In 1996, access to piped water (from private connections Ultimately, whatever significant changes Macleod proposed and public taps) was a mere 48 percent. Poor families or implemented as MD needed the consent of the City predominantly lived in the low-lying and swampy valleys Council. Creating proposals quickly let him set the agenda. between Kampala’s hills, relying on wells, springs, and Implementing change rapidly let him demonstrate success, water bodies as alternative sources of water for reasons overcoming theoretical objections and turning “new” into of convenience and cost.29 Much of this water was “status quo” before opposition developed. New politicians contaminated, often as a result of infiltration from latrines,30 were eager to be seen to be doing something quickly and which contributed to cholera outbreaks in December 1997 it was thus convenient to back a manager who had already to March 1998.31 demonstrated he was being effective in line with the new political agenda. 29 The World Bank. 2014. Do Pro-Poor Policies Increase Water Coverage: An analysis of service delivery in Kampala’s Informal Settlements. The World Bank, 14. 30 Kayaga, S., J. Fisher, and R. Franceys. 2009. “Improved access to urban water The Executive Mayor, Obed Mlaba (1996 to 2011), and services in Uganda.” Proceedings of the ICE: Municipal Engineer 162:3, 165–170. later City Manager, Mike Sutcliffe (2002 to 2012), helped Legros, D., M. McCormick, C. Mugero, M. Skinnider, D.D. Bek’Obita, and S.I. 31 Okware. 2000. “Epidemiology of cholera outbreak in Kampala, Uganda.” East Africa foster a stable and enabling political context. They were Medical Journal 77 (7): 347–349. 21 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | The Political Economy of Improving Water Service for the Urban Poor This poor service had been present for decades, however, championed private sector participation. A management and previous reform attempts had failed. The true catalytic contract with a private firm (Gauff) was signed in 1998. event that sparked the National Water and Sewerage In 1998, William Muhairwe was appointed as MD of Corporation’s (NWSC) successful reform in the late 1990s the NWSC. He quickly gained a reputation as an ardent was a call for privatization of parastatals nationwide, reformer. In the initial phase of reforms, the Muhairwe- and creative tension between development partners and led NWSC implemented a series of “Change Management Ugandan stakeholders over how this might apply in the water Programmes” geared toward the rapid improvement of sector. President Yoweri Museveni supported privatizing operations.33 Meanwhile, President Museveni, operating 85 percent of Uganda’s public enterprise sector by the end from a position of political strength under the prevailing of 1997 (Tangri, 1999: 56) as part of an internationally ‘no party’ system of politics, backed this reform, giving accepted debt forgiveness deal worth US$2 billion.32 Inside Muhairwe the autonomy to run the NWSC, first alongside Uganda, however, tensions remained between embracing private management contractors, then on his own. market-oriented economic governance on the one hand and the desire for national and social control of key entities Box 4.2 discusses the role of national and local political on the other. In the water sector, development partners leaders in the reforms. 32 International Monetary Fund. 2000. “HIPC Debt Relief for Uganda Increased 33 These included the 100 Days program, Service and Revenue enhancement project to a Total of US$2 Billion: Additional Relief vital for Uganda’s Poverty Reduction (SEREP), Area Performance Contracts (APC), Stretch out programme, Internally Programs.” International Monetary Fund. Accessed September 18, 2015. https://www. Delegated Area Management Contracts (IDAMCS), One Minute Management imf.org/external/np/sec/pr/2000/pr0006.htm Concept (OMM), and a Checkers system that deploys staff, often undercover, to expose bribery and other malpractices at branch offices (Muhairwe, 2010; see also WIN, 2009: 48). Box 4.2: Role of national and local political support in reforms Given the high degree of centralization that characterized African polities in the 1990s and the persistence of patron-clientelism, the scope for proactive, reform-oriented leaders, whether in politics or at the utilities, ordinarily would be circumscribed. Uganda’s utility circumvented this constraint because, in the case of water, at least, the president himself was the reformist leader. Among our cases, Uganda offers the clearest case of water sector reform fitting within the president’s political objectives at the time: he imbued the Managing Director, William Muhairwe, with a level of authority that may have even exceeded that of the water minister. The signs are that his successor, Silver Mugisha, has similarly paid considerable attention to working with the president and senior political leadership in developing new directions and focus points for the NWSC. Though comparative evidence is more limited, we can infer that Burkina Faso’s then-president, the strongman Blaise Compaore, similarly gave his political blessing to ONEA, and that the utility leadership has worked hard at making sure that this has been developed and grown. Kenya followed a different course in the 1990s, as opposition to the tactics of President Daniel Arap Moi caused growing alienation of the ruling Kenya African National Union (KANU) from the Kikuyu people, many of whom were embracing the nascent opposition in Central Province and Nyeri, its heartland. Thus, Nyeri’s water sector politics were largely divorced from national level politics. Senegal and South Africa, for their part, had more democratic, decentralized processes, thus utility reform in those countries would bear less of a presidential imprimatur than Uganda. The eThekwini case mirrors Nyeri in that the role of national political leaders was less visible and less germane to understanding the case. The national and local political transformation in 1994 undoing the apartheid system established a broad mandate for reform, with a focus on extending services to the poor. 22 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | The Political Economy of Improving Water Service for the Urban Poor 4.2 Building momentum and sustaining 4.2.1 Building and preserving strong internal reform capabilities and culture Political and technical leaders, working together in the Where conditions are favorable for reform, what should be context of a catalytic event, can spark reform. However, done first? Successful reforms are mostly problem-focused success is only possible if the balance of political economy (that is, responsive to the crises that drove the need for pay-offs remains in favor of reform and–once achieved–in reforms), well-timed, and adaptive. The utility may need favor of sustained good service, even as the attractions of a new water supply scheme, repairs to the distribution predation on the utility increase. The cases suggest that network, or help implementing systems to collect cash from building reform momentum and sustaining service required customers, or to send bills automatically. The case studies action on three interrelated fronts: are replete with evidence of utility leaders contributing to • Build and strengthen internal capacity and culture. virtuous circles by building internal capabilities and cultures The technocratic and managerial skills that helped that make the utility successful. launch the reforms were then used to build strong staff and managers. Other managerial techniques–such as Later on, this same culture helps to sustain reforms.34 A performance-based pay, inclusive corporate strategic professional culture provides a barrier against predation on planning, and general transparency–also helped build the utility since, being anathema to professional culture, and strengthen performance-based cultures. the organization itself will fight against it. Grooming • Forge and embed external alliances. Alliances future leaders and promoting training and development with customers, the government, development were particularly important in the cases studied. This is an partners, and other stakeholders were used to build indication that leadership is a critical variable, not solely at momentum for reform, and help sustain it. Utility the outset of reforms, but throughout the life of a successful leaders demonstrated a high degree of political utility, particularly in a precarious macro political economy savvy, understanding of the political and wider context. By creating a deep management bench, a utility sociocultural context, and the ability to navigate reduces the risk that losing a leader will undo its success. competing interests in an unsettled governance and institutional environment. Alliances constructed A performance-based corporate culture, with an emphasis in the momentum phase were instrumental in on transparency, also promoted reform momentum and maintaining reforms later. reform resilience. A performance-based pay system–based • Adopt formal rules and structures. Contracts, on annual reviews against performance indicators derived regulatory arrangements, and other formal frameworks from the utility’s overall mission, and well-defined job and institutions were used to support service delivery descriptions for staff–is important in many of the case improvements. Credible commitments through utilities. Many utilities also involve all staff in strategic multistakeholder frameworks, as well as legally- planning. This sets the scene for establishing ownership binding contracts between public and private parties, of the institution, with staff at all levels understanding, worked to reduce the risk of predation, and sustain operationalizing, and able to articulate the approach taken successful governance models. by the management. Importantly, these factors can be self-reinforcing and Kampala promote virtuous circles. Effective leaders build alliances William Muhairwe (MD of the NWSC, 1998 to 2011) first and create meritorious business cultures, which in turn addressed the culture of slackness pervading the NWSC strengthen institutions, and so on. How the case study when he took over. Garbage was left lying around offices. utilities built momentum and sustained reform is described To shock the organization out of its lethargy, he instituted a in the following subsections. 100-day turnaround program, in which managers and their 34 Carpenter. The Forging of Bureaucratic Autonomy, 14. 23 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | The Political Economy of Improving Water Service for the Urban Poor teams committed to extraordinary goals which they would towns than was the case under the previous MD. He has achieve within the first 100 days. This signaled that things also intensified efforts to reduce NRW, improve revenue, were changing, and made subsequent changes easier. and modernize the customer feedback systems. He then created a management system based on autonomy, Nyeri accountability, and incentives. He built on a foundation Under Joshua Nguiguti (MD from 1995 to 2014), established by international operators under management NYEWASCO adopted modern management techniques, contracts with clear targets, including with Gauff (1998 to serving to build internal capacity and culture. All staff are 2001), and subsequently with ONDEO (2002 to 2004). involved in strategic planning. Computerized systems are In parallel with these external management contracts, used. The systems are transparent and audited, reducing Muhairwe developed a series of internal management risk of malfeasance. Staff are provided with technical and contracts. The first of these were ‘area performance managerial training. All of this has built a strong corporate contracts’ which applied to the areas outside Kampala not culture which now resists predation. managed by the external management contractors. The NWSC managers for these areas were given performance This internal capacity has been preserved after Nguiguti’s targets and bonuses for meeting the targets. In 2004, a departure. The Board brought in a new MD from outside the similar concept called ‘Internally Delegated Management organization, who immediately confirmed his commitment Contracts’ was applied to the management of the NWSC’s to working with the established senior leadership. The Kampala service area also. current MD of NYEWASCO also recounted that if senior management were to try to cook the books to their personal The NWSC developed its staff. Leaders were hard-working, benefit, their actions would be quickly visible to the committed, and dynamic. Staff were granted autonomy accounting staff, who would report it. He explained that and were motivated by clear vision, mission, and objectives. this is the reinforcing power of a strong professional culture The utility invested in a corporate training facility and and transparent information systems. emphasized monitoring and evaluation, increased customer focus, outsourcing noncore activities, and information Dakar sharing through benchmarking.35 In Dakar, much of the PPP’s success was due to management capability in both the private and public institutions. From The appointment of quality technocratic leadership in the 1996, then new state-run asset holder, Société Nationale senior management of Uganda’s NWSC was instrumental des Eaux du Sénégal (SONES), maintained a degree in enabling reforms to stick, in securing political autonomy, of continuity by retaining the MD of the forerunner and in instituting rigorous performance metrics. When institution, Société Nationale d’Exploitation des Eaux du Muhairwe retired, there was an intermission amid some Sénégal (SONEES), while the affermage contract brought controversy over succession, but in the end his place in private sector expertise from a reputable international was taken by Dr. Eng. Silver Mugisha, a member of the company, SAUR. Senegal’s long experience with private management team that Muhairwe had developed. Good sector management–an affermage contract prior to the leadership, in turn, promoted and sustained virtuous 1972 nationalization and a consulting contract with SAUR circles within the organization, including the cultivation of between 1972 and 1995–facilitated trust, familiarity ,and a professional staff and an institutional culture of efficiency cooperation between public and private actors. and accountability. This has not meant sticking exactly to his predecessor’s agenda–in fact, in the three years or so eThekwini (Durban) since his appointment, Mugisha has been quite assertive eThekwini’s Neil Macleod built momentum through in driving a firmer agenda on the NWSC’s role in smaller technical innovation, changes in the institutional culture, and customer engagement, all of which led to improved service for the poor. This in turn increased political backing 35 Muhairwe, William. 2010. Making Public Enterprises Work. IWA Publishing. 24 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | The Political Economy of Improving Water Service for the Urban Poor Building alliances is not only about high-level politics – in fact, continously improving services and maintaining infrastructure is a concrete way to win and build customer support. (Photo Credit: ONEA, Ouagadougou, Burkina Faso). for the management approach adopted. Under Macleod, The MD negotiated a performance contract with the City eThekwini established a customer management unit with Manager linked to the water unit’s overall performance. This equal status to other major management units such as contract made 25 percent of his remuneration dependent finance and the engineering functions. Macleod was willing on meeting performance targets. These performance to innovate, using technological solutions–such as flow agreements were cascaded to the senior management team. limiters on customer connections and HDPE (high-density Critically, although not operating a classic corporatized polyethylene) pipes–which allowed the utility to provide utility, Macleod was able to establish and sustain the Water free basic water in poor communities.36 At the same time, and Sanitation unit as managerially ‘semi-autonomous’ management efficiency allowed eThekwini to cover all of its within the municipal administration, which protected operating and maintenance costs from the tariff. management and staff from outside interference in operations, and enabled accountability. eThekwini developed a people-focused and trust-based organizational culture in which staff contributions were Ouagadougou recognized and valued. The eThekwini management has In Burkina Faso, ONEA’s MD Mamadou Lamine Kouate consistently emphasized people as the most important asset. was in place throughout a key phase of the reform period The MD adopted a personal and engaging management (1995 to 2005), providing stability in the ‘building style, being quick to give visible recognition to good momentum’ phase. He developed the first strategic plan for staff performance through awards. He built trust in the the period 2004 to 2008. From 2001 to 2006, ONEA also management team and communicated effectively with staff, benefited from technical assistance from a private operator frequently visiting staff in the field. He fostered a positive (Veolia), which ran commercial operations. Veolia provided attitude by encouraging people to propose solutions to the two deputy managers, plus other short-term advisors, for problems they brought to management. ONEA’s commercial and finance departments. They set up new accounting and customer management systems, and helped ONEA identify illegal connections, improve 36 Additional information on this technological innovation is provided in Section 6.2.3. meter reading and meter repairs, and improve customer 25 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | The Political Economy of Improving Water Service for the Urban Poor service. These factors combined to improve management open day, operates an effective customer service line, makes effectiveness. From 2001 (start of reform) to 2006, a point of fixing burst pipes quickly, and reaches out to collection efficiency rose from 85 percent to 95 percent; serve poor communities as well as the middle class. staff productivity improved from eight staff per 1,000 connections to five staff per 1,000 connections; and NRW At the NWSC as well, managers have also made water remained low at about 17 percent. customers into advocates and allies. The NWSC has a well- developed customer service apparatus, which includes a In building a well-trained, motivated management team, dedicated Pro-Poor Unit focusing on informal settlements Kouate helped ensure that succession was smooth and in Kampala, and a range of feedback mechanisms, such as the reform sustained. In 2009, ONEA became ISO 9000 annual surveys that reach 3 to 5 percent of the now 214,000 certified. The company reports that this promotes internal customers in Kampala. The management team reported discipline. Management effectiveness indicators have that providing service to the poor is a key factor in ensuring continued to improve even in the course of a transition on continued political support of the utility since “in Kampala, leadership–by 2013, collection efficiency was 97 percent, the poor vote”. The NWSC has now also galvanized support staff productivity was three staff per 1,000 connections, and in communities and government for the “Water for All” NRW was maintained at about 17 percent. program. Again, service is embedded in a relationship with communication. Like NYEWASCO, the NWSC holds 4.2.2 Forging and embedding alliances with external barazas in communities. There, the utility listens to the stakeholders community’s needs, acknowledges its own problems, and Internal strength needs to be supplemented with external talks about how it plans to improve. Community leaders alliances. Alliances with customers, development partners, and politicians are also involved. In this way, the politicians government, and other stakeholders protect utility see what the utility is doing for the community and the fact performance, since stakeholders allied or networked with that the community supports the utility. This encourages the utility resist predation and other influences that may the politicians and community leaders to cooperate with undermine its ability to serve successfully. the utility and support it, rather than try to score political points that undermine sustainable service. Alliances with customers NYEWASCO’s customers exhibit “brand loyalty”. It is said eThekwini in South Africa established alliances by doing that, when travelling, many carry jerrycans of water from its work effectively, serving its primary (new) constituents the utility’s taps because they consider Nyeri to provide in the black majority and, in the process, aligning to the the best water in Kenya. Such loyalty helps preserve the transformative objectives of the post-apartheid government. utility’s success. Management recounted that, whenever The importance of listening to poor communities and there was a news report or rumor suggesting that political taking what they say seriously is recognized. After interference in the utility might be in the offing, customers establishing community consultation committees, the would call the utility office to ask what they, as customers, initial engagement was tepid. Community members were needed to do to head off such interference. In other words, not sure if the dialogue was meaningful and were wary of citizens of this town in Kenya seemed well aware of the the utility. The first major issue the consultation brought up risks of “predatory behavior”, and the damage it could do. was that the free water allowance was not enough to allow They have shown readiness to protect an entity that had families to meet their basic needs. In 2008, eThekwini demonstrated success. Water got Council approval to increase the nationally set allowance from 6 m3/month to a locally-set volume of 9 m3/ Other factors embedding NYEWASCO in a network month. After that, the communities knew the consultation of alliances includes being locally owned, and strong was meaningful and engaged more enthusiastically. The communication between the utility and the stakeholders then MD says his contract was renewed largely because the it serves. NYEWASCO fosters a relationship with citizens utility successfully served poor communities. through barazas (a deliberative gathering) and its annual 26 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | The Political Economy of Improving Water Service for the Urban Poor Alliances with development partners management and institutional models was valued. Moreover, In Dakar, the World Bank and Agence Française de the KfW provided a loan and helped NYEWASCO develop Développement (AfD) were the leading development a financial strategy that included lower operating costs and partners in Senegal’s water reforms. As in the Burkina Faso a grace period in which free operating cash flows could case, both were advocates of more extensive private sector be directed toward network improvements before debt participation. Yet the development partners worked with service obligations kicked-in.38 The payment discipline that the government as it opted to preserve some elements of the KfW instilled helped NYEWASCO maintain its own SONEES, its former (owner-operator) water parastatal, in commercial and financial discipline. the new institutional structure of SONES, which retained the water assets. Alliances with political leaders In Kampala, the NWSC has formed a powerful alliance The World Bank also helped to develop the affermage with President Museveni. He was instrumental in contract for the private operator. The experience in Burkina bestowing autonomy on the reformed utility (he allowed Faso, where the government resisted World Bank proposals MD Muhairwe to have a substantially free hand), and he for a more private sector-driven model, also highlights the insisted that government offices pay any arrears to the water value of a development partner being open to learn. This company. Essentially, President Museveni has been the flexibility and openness on the part of the Bank eventually “constant” among NWSC allies since the utility embarked was conducive to the emergency of a strong, professional upon reforms in 1996. With his support, the NWSC has ONEA, owned by the public sector, but engaged tactically been largely insulated from predation by other elites or with civil society, organized business, and–at the operational other internal or external interest groups. level –contracting of small providers to help extend services in low income settlements where the utility has been In eThekwini, the Executive Mayor and City Manager prevented by mandate from operating. supported the retention of Macleod as the head of water (a position he held for over 20 years).39 Macleod’s early success In Kampala, international engagement and support aligned and effectiveness were likely key factors in convincing them positively with the objectives of the Museveni government of the merits of retaining Macleod, and thereby avoiding at the time of reforms. Subsequently, the substantial support the turbulence that afflicted other local governments and of international development partners and lending agencies urban water systems. was instrumental to the NWSC’s success from 1998.37 German Corporation for International Cooperation (GIZ)– 4.2.3 Creating and strengthening formal rules and then Germany Agency for Technical Cooperation (GTZ)– structures (institutions) was a longstanding partner, providing technical assistance In isolation, formal rules and structures are an inadequate prior to reforms, and the Austrian government supported guarantee of sustained success. ‘Independent’ boards, for network infrastructure development. Among the largest example, are routinely replaced by politicians, sometimes in and most consistent supporters of the sector have been the breach of company law, or sometimes bending the rules to World Bank and the German Development Bank (KfW), fill boards with their own people. ‘Independent’ regulators which have provided funding, especially for the extension may be reluctant to approve tariff increases. of access to the poor. When coupled with the professionally capable organizations In Nyeri, development partners were quite marginal embedded in a web of stakeholder alliances, however, when reform started. At the momentum phase, however, formal regulatory and governance structures can contribute the GTZ’s help in developing commercial, business-like 38 How this loan was secured for NYEWASCO, and how the utility managed to 37 Schwartz, Klaas. 2008. “The New Public Management: The future for reforms in service its debt, is covered in Section 5.1. the African water supply and sanitation sector?” Utilities Policy Volume 16, Issue 1, 39 In other cities, senior managers have held their positions for much shorter periods. 49–58. The City of Tshwane, for example, has had five city managers in 12 years. 27 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | The Political Economy of Improving Water Service for the Urban Poor to the sustainability of reforms. Adherence to formal rules may in part be due to the fact that it regulates numerous and practices strengthens the utility, the regulator, and utilities, making it harder for political opportunists to government. Lessons from the World Bank’s work in fragile bend the rules for any individual utility. environments like Hargeisa in Somaliland has reiterated the critical value of developing such core capacity formally, even In Kampala, parliamentary approval of the tariff indexation though there may be periods in the immediate aftermath of took the politics out of tariff increases needed to keep pace conflict or other causes of fragility where the only way to with rising input prices. get water to people would be informally. In Ouagadougou, a Contract Plan with the government Contracts serves to regulate ONEA. Its effectiveness is increased Contracts with private parties can embed successful through a multistakeholder Supervision Committee and models. In Senegal the affermage contract defined the independent auditors, as explained below. service standards and remuneration for the operator, and provided mechanisms to monitor and enforce The takeaway from the case studies is that various regulatory systems can help utilities sustain success by protecting performance. While a contract between a public utility against predation. This report does not make a judgment and its government owner is generally easily amended or call about which form of regulation is best. ignored if the government so wishes, a contract with a private party cannot be changed without the consent of Formal supervision involving multiple stakeholders both parties. Thus, an affermage contract helped increase An innovative model used in Ouagadougou for formalizing the durability of the service delivery improvements in external relationships is ONEA’s multistakeholder Senegal. This system also evolved over time. While the Supervision Committee in Ouagadougou. The Supervision contracts and institutional structures were generally Committee comprises representatives of consumers, well specified from the beginning, the contract was later government, NGOs, and the development partners who amended and extended. finance ONEA. Its role is to monitor both ONEA’s and the government’s performance against the Contract Plan. The Incentivized contracts with senior management teams Committee does this through an annual meeting. Prior to may also help to embed good governance. If managers’ pay the meeting, Committee members receive not just a report depends on the results they deliver–as it did in Kampala from each party on its performance against the contract, but and eThekwini–those managers will have incentives to also the report of a financial auditor and a technical auditor preserve autonomy, ensure financial ring-fencing of the whose job it is to assure the quality of the information. The utility, and to avoid predation. auditors’ reports indicate the degree of confidence they have in the information presented. The auditors appear in person Regulation before the Committee and explain their reports. ONEA’s In Kenya, WASREB, the national regulator is credited management credits this committee with an important role by NYEWASCO with allowing adequate tariffs while in making sure that both the utility and the government promoting cost effectiveness.40 National tariff regulation play their agreed roles under the Contract Plan. mitigates the impact of local politics on tariff setting. WASREB also sets and enforces corporate governance rules This arrangement has institutionalized a web of relationships that utilities such as NYEWASCO must comply with. with external stakeholders. Those stakeholders have a These strengthen the utility against attempted political venue, an official role, and reliable information with which predation. WASREB’s own ability to remain independent to demand performance from the utility. They also have the opportunity to protect the utility against predation from government. Bringing customers, development partners, 40 Even with independent regulators, utilities feel that rules ensuring tariff adequacy need to be used with political savvy. This includes, for example, avoiding applying for and other stakeholders together–and giving them reliable a tariff increase during an election season. Afterwards, politicians will be less likely to information–raises the costs of predation, reducing the oppose it–indeed, they will be more interested in seeing how the utility can improve service during their term in office. risk of back-sliding. 28 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | The Political Economy of Improving Water Service for the Urban Poor Box 4.3: Applicability of the findings in fragile environments One may question if these findings are applicable in in fragile and conflict-affected states. The experience of the Hargeisa Water Agency (HWA) illustrate there relevance in such cities. The HWA provides water for over 500,000 people in or around Hargeisa, the capital of Somaliland. Currently, 76 percent of poor households have access to an improved water supply, and about 55 percent have access to piped water for 22 hours per day. Another 13 percent of the poor obtains water from private vendors. On management effectiveness measures, the HWA is covering operating costs from revenues, maintaining collections efficiency at 92 percent, and with nonrevenue water at 33 percent. How has Hargeisa put itself in a position to improve water services, while many others utilities in fragile states are financially unsustainable without government support? The answers may be in the political economy dynamics of the respective systems. In conflict-afflicted, poor governance environments, failure of government institutions is the norm, such as in Kinshasa, Democratize Republic of the Congo (DRC) where the utility, REGIDESO, focused on new investments and awarding construction contracts, rather than instilling management autonomy and sustained building of institutional systems.41 In contrast, Hargeisa’s surprising improvement (albeit from a low base) demands explanation. It seems possible that, with Somaliland having asserted a form of independence from the rest of Somalia (albeit contested) and external assistance being limited, there was no source of surplus in the water sector on which to prey. Equally, the local initiative that drove the creation of Somaliland and the HWA may evidence a polity that is relatively cohesive and focused on improving things for the community. In this environment, the politics appear to have been relatively favorable to the creation of a service-oriented utility with autonomous management. The things that Hargeisa credits for its success–good customer relations, transparency, accountability to stakeholders, and a focus on fair resolution of conflicts–are exactly in line with the governance and political economy approaches in the case studies. The HWA faces challenges of limited water resources and damaged infrastructure. Somaliland is a semi-arid country with low surface water potential, and the country suffers from persistent drought. The civil war that broke out in 1988 destroyed water infrastructure, as state institutions collapsed. When, after the war, some development partners provided investments in infrastructure, an emphasis and institutional strengthening helped put in place the organizational structure, policies, and procedures required for the HWA to better manage its finances, procurement, and human resources. With some external support, HWA’s management has improved its performance in key areas: water connections almost doubled from 11,900 (2006) to 21,000 (2014); Collection efficiency was 92 percent or higher in 2012 to 2014. Access to piped water by the poor is still only 55 percent, but reliably available 22 hours per day, on average. The HWA still has a long way to go to achieve its vision for clean, affordable, and safe water for every resident of Hargeisa city, but illustrates that, despite a challenging starting point–with the right external support aimed at building core systems and capable management–services can be improved and new approaches introduced. 41 Doyen, Jean H. 2015. “REGIDESO 2006-2014: Restoring & Expanding Capacity in a Post-Conflict Context.” WSP, November; SDE-ERANOVE. 20015. “Rapport Annuel 2014; Contrat de Service.” March. 29 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | The Political Economy of Improving Water Service for the Urban Poor 4.3 Key points Though reform is often sparked by a catalytic event such as a water supply crisis, and driven by utility and political leadership, the key reforms ultimately cannot be driven by outsiders. Development partners could, however, provide initial finance for the turnaround, but their longer term impact is highest when they carefully direct financial and technical support towards bolstering reform and enhancing core systems. Especially in fragile environments, the latter is often an essential aspect of sustainable impact. A combination of internal competence and motivation, external alliances, and institution-building can shape the institutional changes required to put service providers on track. There are no golden rules or magic bullets–context is the key determinant, demanding that external support and local reform is problem-focused and adaptive. The very notion of ‘best practices’ seems out of place in such processes–what is needed from local role-players and development partners is creative, adaptive engagements to allow reform progress at a pace that locals are comfortable with, and that ultimately transcends the individual leaders, but getting embedded in locally grown institutions. While the search for context-independent best practices seems out of place, the good news is that alignment of political economy in support of good water services can occur in a wide range of circumstances, including in: • Countries with low incomes. Burkina Faso’s gross domestic product (GDP) per capita is US$720, and Uganda’s is US$677. Building stakeholder confidence has been integral to the • Countries with uncompromising governance improvement strategies of all case study utilities – from call environments. When Uganda’s reforms started in 1998 centers to deal with complaints and follow ups, to forging the country was still recovering from decades of brutal alliances with political leaderships, city stakeholders and dictatorship and civil war. Niger ranks 20th in Africa on customer groups. (Photo Credit: eThekwini Water). the World Bank government effectiveness indicator. • Cities of all sizes, ranging from 150,000 (Nyeri) to more than 3.5 million (Dakar). These cases show that alignment of the political economy • Countries stressed for water resources. Burkina Faso of the water sector with service to the poor is possible in has renewable internal freshwater resources per capita a range of contexts. Ouagadougou is perhaps the most of just 711 m3, well below the Democratic Republic impressive case, being a poor, arid city with one of the of the Congo’s 12,020 m3 per capita, for example. fastest growth rates in the region. This case provides hope • Cities with rapid population growth. Ouagadougou that good service to the poor is possible in cities of all sizes grew at an average of 7.5 percent per year from 2000 and income levels, and a wide range of endowments in to 2015. terms of water, financial, and institutional resources. 30 V. Financing the Extension of Good Service to the Poor To improve service to the urban poor, utilities in the cases respectively). Nearly all the poor not served by direct studied had to invest in network expansion and bulk water connections are served by public taps. Water in these supply schemes that often cost hundreds of millions of cities is available 24 hours a day and is safe to drink. In dollars. This section explores two questions: how did these Kampala and Ouagadougou, most of the poor are able to utilities successfully finance improved service, and how obtain safe water from public taps. Overall piped water could typical utilities in Africa do the same? access rates for the poor in these cities are 78 percent and 90 percent, respectively. These service improvements are 5.1 How successful utilities financed summarized in Table 5.1. improved service to the poor In the 1990s, water service in four African cities–Nyeri, Delivering quality water service depends on making Kampala, Ouagadougou, and Dakar–looked very different connections throughout the city. The utilities serving from today.42 Water was rationed and typically unsafe to these four cities, plus the utility in eThekwini (Durban), drink. In Kampala and Ouagadougou, less than half the have increased the number of water connections in the city population had access to piped water. Access rates for city by an average of 93 percent since 2006.43 Growth the poor are unknown, but likely even lower. in connections–represented as an index with a base year (2006) equal to 100–is shown in Figure 5.1. The number Today, these cities are exemplars of good water service to to the right of each line is the value of the index in the the poor. In Dakar and Nyeri, most poor have access to most recent year for which data are available.44 water piped to the premises (75 percent and 88 percent, 42 eThekwini is not analyzed in this section due to unavailability of capital expenditure 43 2006 was chosen as a base year because it is the earliest year for which data on con- and financing data. nections are available for all cities. In all cases, the reform began before 2006. 44 Subtracting 100 from this number gives the percentage growth in connections since 2006. Table 5.1: Improved water supply service to the poor City Situation in the 1990s Situation today Nyeri • Water rationed and unreliable • 88% of the poor have access to piped water to the • Unsafe to drink premise, and 2% use public taps • Utility did not serve informal settlements on • Water is available 24 hours a day and is safe to steep slopes surrounding urban periphery drink Kampala • Less than half the city population had access • 78% of the poor have piped water to piped water (1998) Dakar • Water shortage • 75% of poor have water piped to their premises, • 20% had no access to piped water at all, while with a further 22% accessing standpipes others relied on standpipes Ouagadougou • Water shortage • 90% of the poor have piped water • Half the population had access to piped water (2000) Source: Household surveys (see Table D.1 for list of sources). 31 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Financing the Extension of Good Service to the Poor Figure 5.1: Growth in water connections in case study cities Source: Utility research (see Table D.2 for full list of sources). These utilities expanded connections in their cities at rates 5.2. For context, the change in the absolute number of faster than population growth. This can be measured by connections per 100 people in the city is shown as well. connection density, the number of connections per 100 people in the city. These five utilities have increased the In addition to expanding access, most of the cities invested number of water connections per 100 people in the city heavily in increasing bulk water supply and treatment. by an average of 40 percent since 2006, as shown in Figure Figure 5.2: Growth in connection density (connections per 100 people in city) Source: Utility research (see Table D.2 for full list of sources). 32 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Financing the Extension of Good Service to the Poor Financing solutions implemented for four utilities: NWSC (Kampala), NYEWASCO (Nyeri), How did these utilities finance such service improvements? ONEA (Ouagadougou), and SDE/SONES (Dakar).45 In every case, the utilities were on-lent donor finance on concessional terms, with low interest rate and long grace NYEWASCO periods. The loans were primarily repaid with operating For NYEWASCO to increase access and improve service, cash, generated by improving operating efficiency (thus KShs 1.2 billion (US$19 million) in capital expenditure minimizing costs) and expanding access (thus increasing was required from 2005 to 2014. Capital expenditure for revenue). Some of the finance was given as an equity network expansions and other major projects was primarily contribution or another form of grant. Table 5.2 shows financed by loans, grants, and net cash from operating the sources and amounts of capital expenditure financing activities, as shown in Figure 5.3. 45 The NWSC, ONEA, and SDE/SONES are national utilities. The figures presented in the table are for the entire service area, not just the city of interest. Table 5.2: Sources and amounts of capital investment financing Utility Estimated total Capital investment per Percent grant- Percent financed Percent (years) capital investment person served per year financed by internal cash financed (US$ million) (US$/ person/year) flow by loans NYEWASCO (2005–2014) 19 22 5% 14% 80% NWSC (2002–2011) 97 4 28% 52% 16% ONEA (2002–2013) 600 23 52% 10% 29% SDE/SONES (1996–2013) 770 10 29% 23% 47% Notes: For the NWSC, total capital expenditure was estimated using cash flow statements. Cash outflow was summed from the following financing activities: Capital Work-in-Progress; Purchase of Property, Plant and Equipment; and Purchase of Computer Software. For ONEA and SDE/SONES, investment data were provided by the World Bank. Totals for the NWSC, ONEA, and SDE/SONES are for the entire service area, not solely the city of focus. Source: Utility research (see Table D.2 for full list of sources). Figure 5.3: Sources and amounts of capital expenditure financing (KShs billion) Notes: All amounts in nominal Kenyan Shillings (KSh). The 2008 financial year was 18 months long, running from January 2007 to Jun 2008. For this reason, 2007 does not appear in the graph above. (Prior to 2007, NYEWASCO’s financial year ran from January to December.) Source: NYEWASCO Financial Statements. 33 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Financing the Extension of Good Service to the Poor From 2005 to 2008, NYEWASCO financed 94 percent Table 5.3: Summary of efficiency of total capital expenditure through a KShs1.1 billion indicators, NYEWASCO (US$18.2 million) loan from the KfW for network Indicator 2006 2014 rehabilitation and extension. The loan was facilitated by the Collection efficiency 98% 100% Government of Kenya, which borrowed in Euros, and on- Staff productivity* 9 3 lent to NYEWASCO in Kenyan Shillings. The other loan terms were an interest rate of 2.5 percent, a tenor of 30 Nonrevenue water 42% 19% years, and a grace period of eight years. The loan financed a * Staff per 1,000 connections. new production facility and network expansion, as well as Source: Utility research (see Table D.2 for full list of sources). rehabilitation and equipment for leak detection and repair. Following that, from 2009 to 2014, 57 percent of capital expenditure was financed through cash from operating activities (US$2.4 million). Another 33 percent was NYEWASCO maintained a high collection rate through financed through grants from the Water Services Trust continued collection of long over-due arrears, as well Fund (US$0.7 million). as rigorous collection of current bills. To collect from government customers, NYEWASCO makes a point of At the same time, NYEWASCO paid more than US$3.6 knowing when government customers will receive cash million (in 2014 real terms) in debt service payments from from the budget, and follows up on payment at that 2010 to 2014. Cash from operations (at US$6.0 million time. To ensure collections from the police compound, in 2014 real terms) was sufficient to service the debt and NYEWASCO strategically re-laid pipes so that the contribute to other capital projects. Figure 5.4 shows the net compound could be cut off for nonpayment from within cash from operations compared to debt service payments. the utility headquarters. This innovative approach stopped the harassment of utility staff that had previously prevented To service the debt, and to finance additional capital it from enforcing payment. expenditure, NYEWASCO improved operating cash flow through greater efficiency. To improve efficiency, NRW In addition, increased sales, due to expansion and a major was reduced from 42 percent in 2006 to 19 percent in bulk water supply project, also helped increase operating 2014. During the same period, staff costs as a percentage cash. From 2006 to 2009, water production increased from of revenue fell from 44 percent to 31 percent, in part due an average of 10.2 MLD (3.7 million m3/year) to an average to increased labor productivity–there were just three staff of 14.3 MLD (5.2 million m3/year)–an average increase of per 1,000 connections by 2014. Meanwhile, the collection 12 percent per year. During the same period, active water ratio stayed high at around 100 percent. connections increased from 9,863 to 13,661–an average Figure 5.4: Capital expenditure financing and debt service payments Source: NYEWASCO Financial Statements. 34 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Financing the Extension of Good Service to the Poor increase of 11 percent per year. This led to a 50 percent was obtained for financing the extension of the Ggaba increase in the volume of water sold during the period. intake plant, which supplies water to Kampala city and the surrounding areas. This loan is being serviced from NWSC operating cash flow. For the NWSC to increase access and improve service in Kampala and in other cities, more than UShs171 billion Increasing operating cash flow was key to achieving (US$97 million) of capital expenditure was required from expansion in service. The NWSC increased collection 2002 to 2011. While approximately 28 percent was grant- efficiency (from 85 percent in 2001 to 95 percent in 2011), financed, 52 percent was financed by internal cash flow reduced NRW (from 43 percent in 2001 to 33 percent and 16 percent from loans. A small portion was financed in 2011), and increased labor productivity by limiting by other sources–this includes cash from nonoperating staff growth as connections increased (see Table 5.4). Real activities, such as proceeds from disposal of property, tariffs increased at a modest 3 percent per year. Together, plant, and equipment. The NWSC’s sources of finance are these factors provided the operating cash surplus used to depicted in Figure 5.5. finance investment and repay debt. Early in this period, the NWSC had debts to the government that it could not service. The government agreed to a Table 5.4: Summary of efficiency moratorium on debt service for a period–then, in 2007, indicators, NWSC the government converted all outstanding loans from Indicator 2001 2011 development partners (UShs 85 billion, or US$47 million) into equity. Collection efficiency 85% 95% Staff productivity* 16 6 Since then, the NWSC has borrowed from commercial Nonrevenue water 43% 35% banks and is repaying loans from operating cash flow. For * Staff per 1,000 connections. instance, in 2010, a commercial loan of US$2 million Source: Utility research (see Table D.2 for full list of sources). Figure 5.5: Sources and amounts of capital expenditure financing (UShs billion) Notes: All values are in nominal Ugandan shillings (UShs). The exchange rate in 2011 was UShs2,340 to US$1. Data were unavailable for 2008 so capital expenditure for that year is not included in the total. Source: NWSC Financial Statements. 35 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Financing the Extension of Good Service to the Poor ONEA Operating cash flow was key to the expansion, allowing For ONEA to increase access and improve service in ONEA to service the debt, and to invest directly. Forty- Ouagadougou and other cities, investment totaling about eight percent of capital expenditure was funded directly US$600 million was required from 2002 to 2013.46 While or indirectly, from free cash from operations. Drivers approximately 52 percent was grant-financed, 19 percent of increased cash flow included: increasing collection was financed by own cash and 29 percent from loans. efficiency (78 percent in 2002 to 97 percent in 2013), ONEA can service its debt from its operating cash flows. maintaining low levels of NRW (around 18 percent), and Figure 5.6 shows the sources and amounts of investment increasing labor productivity (from eight staff per 1,000 financing during this period. connections to three staff per 1,000 connections) by limiting staff growth as connections increased. A contract Among the most important initiatives was the Ouagadougou with Veolia (a specialized water operator) to help manage Water Supply Project (US$269 million, 2001–2007). the utility’s commercial function was useful in achieving The World Bank lent US$70 million to the Government some of these improvements. of Burkina Faso. US$28 million was on-lent to ONEA, and the remaining US$42 million was given as an equity SDE/SONES contribution. The interest rate was 5.4 percent and the tenor For the SDE and SONES to increase access and improve was 20 years (including a 10-year grace period).47 Other service in Dakar and other cities, investment totaling about development partners provided the remaining finance, also US$770 million was required from 1996 to 2013.48 This via loans to the government. amounts to about US$10 per person served per year. Figure 5.7 shows the sources and amounts of investment financing The program included construction of the Ziga dam, during this period. While approximately 29 percent was Boudtenga reservoir (5,400 m3), a water treatment plant grant-financed, 23 percent was financed by own cash and 47 and pumping station, and extension of the distribution percent from loans from development partners. The loans network. Increased water production led to near perfect from development partners are repaid from tariff revenue, water supply reliability (23 hours per day). which is allocated to SONES for this purpose. 46 Figure quoted in 2013 dollars. 48 Figure quoted in 2013 dollars. 47 World Bank. 2001. “Project Appraisal Document, Ouagadougou Water Supply Project,” (2001): 12. Figure 5.6: Sources and amounts of investment, ONEA Source: Utility research (see Table D.2 for full list of sources) 36 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Financing the Extension of Good Service to the Poor Figure 5.7: Sources and amounts of capital expenditure financing, SDE/SONES Source: Utility research (see Table D.2 for full list of sources). Two major programs implemented during this period were Strategies for improving operating cash flow the Senegal Water Project (US$223 million, 1996–2004) Generating operating cash allowed utilities to repay the loans and the Long-Term Water Project (US$255 million, 2002– that financed network expansion and service improvement. 2009).49 The World Bank was a financier for both projects, A small portion of operating cash also directly financed providing US$85 million and US$146 million respectively, investment. Operating cash flow is defined as the net cash with interest rates around 6 percent and tenors of 20 years generated from normal business operations. There are three (including five-year grace periods).50 main ways to increase operating cash flow: • Improving operating efficiency, in particular by The Senegal Water Project financed urgent work to increase increasing the collection ratio and reducing NRW water supply in Dakar, including boreholes, expansion of a (particularly commercial losses). treatment plant, and leakage reduction. A private operator • Expanding access, thus increasing sales and therefore was brought in through this project as well. Signing a PPP revenue. contract satisfactory to the World Bank was a condition of • Raising real tariffs, thus increasing revenue. the World Bank loan. The Long-Term Water Project financed the Keur Momar Sarr water treatment plant (in 2005, 65,000 It is important to note that expanding access and raising m3/d; upgraded in 2008 to 130,000 m3/d) along with a tariffs alone may not lead to positive operating cash flow. continued expansion in the distribution network. Operating efficiency is also important. For instance, if a utility has a collection ratio of just 80 percent, 20 percent of earned revenue is not collected as cash. If this US$231 million utility could improve its collection ratio to 95 percent, even without any increase in total revenue, cash collected Financing from the World Bank towards the implementation of the would increase by 18.75 percent.51 Senegal Water Project (US$223 million, 1996–2004) and the Long- Term Water Project (US$255 million, 2002–2009), providing US$85 million and US$146 million respectively. How the utilities increased operating cash in practice is shown in Table 5.5. For the NWSC and ONEA, 49 World Bank, Implementation Completion Report, Senegal Water Project; World Bank, increasing collection ratios were particularly important. Implementation Completion Report, Long Term Water Project. Note: These projects included sanitation components which is overseen by ONAS, not SDE or SONES. 50 World Bank, Staff Appraisal Report, Senegal Water Project, p. iv; World Bank, Project Appraisal Document: Long Term Water Project, 20. 51 15 percent divided by 80 percent equals 18.75 percent. 37 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Financing the Extension of Good Service to the Poor Table 5.5: Strategies for improving operating cash flow Utility Increase in collection Decrease in NRW Annual increase in Annual change in real ratio connections average tariffs NYEWASCO 98% to 100% 42% to 19% 10% (2006 to 2014) -8% (2010 to 2014) (2006 to 2014) (2006 to 2014) NWSC 85% to 95% 43% to 35% 15% (2001 to 2011) +3% (2002 to 2009) (2001 to 2011) (2001 to 2011) -1% (2009 to 2011) ONEA 78% to 97% Averaged 17% 15% (2002 to 2013) -2% (2005 to 2013) (2002 to 2013) (2002 to 2013) SDE/SONES Averaged 97% 30% to 20% 6% (1996 to 2013) +2% (1997 to 2009) (1996 to 2013) (1996 to 2013) -1% (2009 to 2013) Source: Utility research (see Table D.2 for full list of sources) At NYEWASCO and SDE/SONES, collection ratios were Figure 5.8 compares the operating efficiency of 11 utilities always quite high–these utilities maintained their high (six with relatively good service and five with typical service, collection ratios and focused on significantly reducing selected based on data availability). The figure shows that NRW. All four utilities increased total connections at the utilities in cities with relatively good service to the poor impressive rates–notably, for more than a decade, both the (represented by blue trend lines) are more efficient than the NWSC and ONEA achieved annual average increases of utilities in cities with typical service to the poor (represented 15 percent. by gray trend lines). Real tariff increases, if any, were moderate. The NWSC had Comparing longitudinal trends in connectedness the highest average annual real tariff increase (3 percent per (connections per 100 people in city) among utilities results year during the 2002 to 2009 period). From 1997 to 2009, in a similar finding (see Figure 5.9). Utilities in cities with the SDE’s real average tariffs increased by about 2 percent relatively good service to the poor have higher connectedness per year. In recent years, real average tariffs have actually than utilities in cities with typical service to the poor. As fallen at all four utilities. This comparison shows that, while shown in the previous section, expansion of access in the strategies differed slightly from utility to utility, improving relatively good cities has certainly contributed to improved operating efficiency and expanding access (by using loan operating cash flow. money) helped improve operating cash flow. Real average tariff increases were not always necessary.52 This comparison raises the question of whether the typical African cities could finance improved service to the poor 5.2 How typical utilities could finance by adopting similar strategies. To test this proposition, the improved service to the poor research team analyzed whether Mombasa would be able to Four cities with good service to the poor–Nyeri, improve service to the poor using a financing strategy based Ouagadougou, Kampala, and Dakar–broadly used similar largely on loans serviced from operating cash flow. financing strategies. Concessional loans, along with some grants, financed most capital expenditure early on. By For Mombasa Water Supply and Sanitation improving operating efficiency and increasing sales, utilities Company (MOWASCO) to increase access to 100 percent generated the cash needed to service the debt. Over time, by 2030, while ensuring adequacy of supply, more than free operating cash also directly financed new capital US$1.0 billion in capital expenditures would be required. expenditure and allowed additional borrowing. MOWASCO is currently not financially viable. Reasons for this poor financial performance include low collection rates 52 Adequate tariffs are critical for good service provision, as discussion in Section 6. (80 percent in 2014), high NRW (52 percent from 2013 to 38 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Financing the Extension of Good Service to the Poor Figure 5.8: Comparison of utility operating efficiency * Weighted average of NRW, staff productivity, and collection ratio indices. Source: Utility research (see Table D.2 for full list of sources). 2015), and not supplying enough water to meet demand. • Increasing tariffs by just 1.5 percent per annum in real Despite this, MOWASCO could finance the US$1.0 billion terms. needed to reach full coverage by: • Borrowing a total of US$1 billion at an interest rate • Progressively increasing collection efficiency to 96 of 0.75 percent, with a 10-year grace period, and a percent by 2025. loan tenor of 30 years. • Progressively reducing NRW to 25 percent by 2030. • Cutting operating expenses per m3 supplied by 10 As Table 5.6 shows, these improvements are all within the percent by 2025. range achieved by other Sub-Saharan African water utilities. Figure 5.9: Comparison of utility connectedness (connections per 100 people in city) Source: Utility research (see Table D.2 for full list of sources). 39 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Financing the Extension of Good Service to the Poor Table 5.6: Performance of successful utilities Utility Country Collection ratio NRW Decrease in real opex/m3 supplied SEEN Niger 100% 16% 17% (2005–2014) Nyeri Kenya 100% 18% 11% (2005–2014) ONEA Burkina Faso 96% (avg) 17% 2% (2005–2013) SDE Senegal 97% (avg) 20% - MOWASCO assumptions for financing 96% 25% 10% (2015–2025) Source: Full list of sources for utility data included in Appendix B.2. Figure 5.10 shows the free cash from operations that County, and management effectiveness in MOWASCO. It MOWASCO could generate if these performance levels were certainly does not prove that all cities in Africa can self- reached. This would be sufficient to service the debt taken finance the provision of universal access to quality water on to make the investment. The projected improvement in services. What it does do is provide the hope that, if the service is also depicted in Figure 5.10. financial techniques that worked in the five case study cities are deployed in other cities in Africa, there is a reasonable What does this analysis of MOWASCO tell us? It does prospect of extending service to the poor. Mombasa was not say that Mombasa will soon solve its water supply not selected for analysis because it is a particularly easy problems. That depends on political economy in Mombasa case. On the contrary, providing water service in Mombasa Figure 5.10: A financing strategy for universal access 40 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Financing the Extension of Good Service to the Poor is challenging. Given the city’s massive service problems • Generating positive free cash from operations, and investment needs, the analysis might have shown that through a combination of ensuring cost-effectiveness massive public subsidies are needed if universal, affordable and adequate tariffs. Cost-effectiveness was achieved water supply is to be provided. That the financial analysis in large part by limiting NRW, collecting bills did not show this is some evidence that what worked effectively, and boosting labor productivity. Adequate financially in five African cities can be extended to other tariffs are those which generate enough revenue– cities in Africa. across all customer classes–to fully cover operating and maintenance costs and generate a surplus for 5.3 Key points investment. The operating surplus was typically in the The water supply networks and production facilities in 20 percent to 40 percent range. African cities need substantial investments, both for new • Ensuring affordability for poor people through service and infrastructure and for the upgrading of existing facilities. A tariff offerings specifically designed to meet their needs combination of grants and concessional loans have assisted (these are discussed in detail in the following section). utilities that have reached the poor, but the utilities have also • Using the free cash generated from operations to fund worked hard at improving operating efficiency, especially infrastructure investment. This was done by taking reducing NRW, improving productivity, and increasing out loans which could be serviced with the cash from collection rates. This made them able to raise new revenue, operations, and direct investment of the cash left after thereby more operating cash, and enabling them to pay debt service into additional infrastructure. Across the back loans and finance more capital investment. cases, the share of infrastructure investment financed from operating cash-flows ranged from 10 percent There is a perceived trade-off between cost recovery and to about 50 percent of the total, and from US$2.4 affordability, but the case studies show that this is not million in Nyeri to US$180 million in Dakar. inevitable. Utilities in the case studies have helped lower • Taking advantage of grants from the national the cost of connections through lower tariffs for the poor, government for capital investment, to cover or by enabling customers to pay back connection costs infrastructure investment that could not be financed over time. New technologies like cellphone meter readings from operating cash flow. These grants were typically and payments, and the introduction of prepaid meters higher at the start of the reform, and tapered off as the have also opened ways for poor households to spread out utility generated free cash. Grants as a share of total water payments into affordable portions. The net result capital expenditure ranged from 5 percent in Nyeri to has been that investment needs and operational costs have about 50 percent in Ouagadougou been proven not to be inevitably prohibitive to extending and improving services to poor people. The big question is whether similar strategies can work in other African cities. An analysis of Mombasa shows that This does not mean that serving the poor is profitable at least one African city with severe water access problems for utilities. As the next section describes, many utilities could achieve universal access if it adopted the techniques charge people less than the average cost of service. At the described above. For other cities, the answer will vary. Some extreme, eThekwini provides enough water to meet basic cities may able to achieve universal access entirely from their needs for free for poor people. Nevertheless, all five of own resources. These will typically be cities endowed with the case study utilities have found ways to be profitable higher income levels, and utility networks that are already at the utility level, even if they supply services to poor relatively extensive. The poorest cities with the greatest people at below cost. needs may need grants if they are to achieve widespread affordable access quickly–just as Ouagadougou did. It is In summary, in the five cities studied, extension of access safe to say, though, that the financial techniques used in the to the poor was made possible, and affordable, through the five case studies will always help to extend access faster, and following strategies: sustain it longer, for any given level of grant provision. 41 VI. Practical Techniques to Achieve Widespread and Affordable Access The previous sections highlighted that financing network 6.1 Techniques to overcome financial expansion is within the grasp of many African cities, political barriers to access economy acumen or enabling environments can open It is often assumed that there is a trade-off between cost opportunities to serve the poor, and there is a suit of technical recovery and affordable service. If utilities can generate the possibilities to extend services. They also show that serving the operating cash needed to finance investment, it is expected poor is unlikely if it merely an addition to a utility–instead that tariffs must be high–perhaps prohibitively so. There it becomes more likely that utilities will be able to serve is a logic to this. Utilities need to recover their operating poor people better if they are well-managed utilities more costs and some of their capital costs if they are to finance generally. Not only does being a good utility bring financial expansion and improvement of service to the poor. However, and technical capacity, but it also may open the scope for the poor often cannot afford to pay a typical cost recovery taking risks and innovating. The discussion here continues to tariff or a standard utility connection fee. These become focus on the case study utilities, but it also shows with a few financial barriers to access. references encouraging examples of other utilities grappling with, and innovating to improve, services. Nevertheless, in the cities where access for the poor is relatively good, utilities have managed to combine cost- However, even when these bigger issues seem resolved, recovery with affordability. Indeed, their strategy has been to service providers often need to overcome many other barriers use cost recovery to finance the infrastructure that provides the poor face in accessing water. This section examines the access, while using other strategies to ensure affordability practical techniques that the case study cities and utilities for the poor. This section reviews some of the strategies used to achieve widespread, affordable access. that allowed these cities to transcend the trade-off between affordability and financing access. To provide convenient access for poor customers, some utilities focus on water piped to the premises, while others at least endeavor locating public taps near where people live, and ensuring that the unit price of water is affordable. (Photo Credit: ONEA, Ouagadougou, Burkina Faso). 42 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Practical Techniques to Achieve Widespread and Affordable Access 6.1.1 Affordability of access connections Utilities need to build these taps near where people live, The first financial barrier to address is the connection and ensure the unit price charged is affordable. Niamey, charge. NYEWASCO has kept connection costs down where 78 percent of the poor access water through a public for low income consumers. Currently, a low income tap, is doing well to ensure that standpipes are built near family needs to pay KShs 3,100 (about US$35) to get a the poor. The average roundtrip time to collect water for connection. Of this payment, KShs 2,000 is a deposit. the poor is just 12 minutes (see Figure 3.4). Kampala, Recognizing that even this level of upfront payment is a where 71 percent of the poor access water through a barrier, NYEWASCO has decided to reduce the deposit to public tap, is still working on enabling convenient access. KShs 1,500. Other utilities in Kenya, such as Nairobi, have The average roundtrip time is 21 minutes. The NWSC is followed similar approaches, and although it is too early addressing this by providing more water kiosks in more to measure the results, this shows an openness to learn as areas, providing more connections to the premise in poor service providers across Kenya struggle to deal with the areas, and charging the ‘kiosk’ water rate to individual demands of urbanization and informal settlements. customers whose connections serve multiple households. In Kampala, the NWSC has operated an Affordable 6.1.2 Ensuring affordability of on-going service Connections Policy since 2004. Connection fees were Once connected to the network, or conveniently able reduced from US$75 to US$35. Moreover, the NWSC to access a public tap, the poor worry about paying for took on responsibility for making a service connection on-going service. In many African cities, incomes of of up to 50 meters from the NWSC supply point. The poor families are not sufficient to pay for a basic needs NWSC also maintains the service line, up to the meter on quantity of water at average cost. In addition, from a cash the customer’s premises. management perspective, paying a full month’s water bill at once is challenging, especially if consumption is With the help of development partners, SONES provided higher than expected. Moreover, customers know the more than 200,000 social connections since 1996–most consequences of nonpayment are high–disconnection of these were located in Dakar.53 The connection costs for itself comes with additional costs. Even once the debt is poor households are subsidized by the water rates, and paid, a reconnection fee is often charged before service amounts to just over 20 percent (US$31) of the actual can resume. cost of a connection. Subsidizing connections enables beneficiaries to benefit from a higher quality and quantity To ensure affordability of on-going service for the poor, of water services, which the authorities see as pivotal to utilities have adopted many strategies. Some sell a basic improving public health in peri-urban neighborhoods needs quantity of water at a below average tariff, and sell where many poor people reside. water above the ‘basic needs’ quantity at an above average tariff (known as a progressive or rising block tariff structure). ONEA has also implemented social connection policies, They also cross-subsidize nonresidential consumption, by and furthermore pays for the construction materials in charging nonresidential customers more than residential informal settlements (see Section 6.2.2). customers. Some have implemented innovative billing and payment systems to help customers manage their cash This said, most poor people in Africa access water from flow. Others offer volume restricted connections, to allow public taps (see Figure 3.1). There is no “connection households to monitor consumption during the month. charge” to access a public tap, but time spent collecting water (which could be used for other purposes) is often a Progressive tariffs for residential, piped to premise connections prohibiting factor. Under a progressive–or ‘rising block’–tariff structure, a monthly ‘basic needs’ quantity of water is sold at a low 53 World Bank. 2011. ‘Water Supply and Sanitation in Senegal: Turning Finance into tariff. The equity objective is to provide affordable water to Services for 2015 and Beyond.’ World Bank (2011): 22. Accessed April 28, 2016, https://www.wsp.org/sites/wsp.org/files/publications/CSO-senegal.pdf. low-volume customers, who are typically poor. 43 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Practical Techniques to Achieve Widespread and Affordable Access However, it has been questioned whether rising block Figure 6.1 shows the SDE subsidizes consumption at tariffs actually achieve this objective in practice. Often, low amounts by charging much higher tariffs for what it consumption at the lowest block is cross-subsidized for all deems to be excessive consumption. Senegal’s increasing customers, not solely the poor.54 In cities where the poor do block tariff structure has a subsidized social tariff for not have water piped to the premises at all–due to limited consumption below 20 m3 (FCFA202, or US$0.40) per network expansion or high connection charges–solely the bimester.56 There is also a regular tariff for consumption well-off benefit.55 These issues highlight how rising block from 21 m3 to 40 m3 (FCFA697.97, or US$1.39), tariffs are often misused, failing to benefit the poor. and a “dissuasive” tariff for consumption above 40 m3 (FCFA878.35, or US$1.75). The dissuasive tariff is In the cities where service for the poor is good, however, designed to deter excessive water use. It can be seen that the utilities have used rising block relatively tariffs well. the social tariff is less than a third of the regular tariff, and eThekwini, for instance, offers the initial block of free water less than a quarter of the dissuasive tariff. Only the social only to poor households. In Dakar, 75 percent of the poor tariff–and the standpipe tariff (FCFA366, or US$0.73)– have access to piped water to premise, and consume low are below the average tariff (FCFA494, or US$1.08). quantities–likely to ensure their consumption falls within the lowest block. The same logic applies for Nyeri, where 88 eThekwini also employs a rising block tariff, with a number percent of the poor have access to piped water to premise. of distinctive features. Most notably, the first block of water Residential tariff structures for eThekwini and SDE/SONES is free, in accordance with South Africa’s ‘free basic water’ (Dakar) are depicted in Figure 6.1. Additional information policy. When eThekwini introduced the policy, the free about the strategies they use follows. amount was set at 6 m3 per month per household. After residents of poor communities made it known that this was not enough to meet basic needs, the amount was increased 54 Banerjee, Sudeshna, Vivien Foster, Yvonne Ying, Heather Skilling, and Quentin to 9 m3 per month per household. Wodon. 2010. “Cost Recovery, Equity, and Efficiency in Water Tariffs: Evidence from African Utilities,” World Bank. Komives, Kristin, Vivien Foster, Jonathan Halpern, and Quentin Wodon, 2005. 55 Water, electricity, and the poor: who benefits from utility subsidies? World Bank. 56 Every two months Figure 6.1: Comparison of progressive tariff structures (eThekwini, SDE/SONES) * First block applies to households whose dwelling has a capital value of less than 250,000 Rand (US$20,500). Sources: eThekwini Water and Sanitation, ‘Water Tariffs 2015/2016’; SONES. 44 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Practical Techniques to Achieve Widespread and Affordable Access In South Africa, the cost of providing this free water Government customers pay even more–their tariff is allowance is partially supported by a national government FCFA1,868.88/m3 (US$3.72/m3). The high tariff for unconditional grant indexed to local poverty levels. government was introduced in 2007 as a way to keep However, the grant does not cover the full cost. To limit residential tariffs stable while ensuring cost recovery for the financial burden on the municipality, the municipality the utility. The relatively high tariffs charged to commercial provides free water only to households whose dwelling and government customers allow the lower blocks of the has a capital value of less than ZAR250,000 (about residential tariff to be maintained at affordable levels. US$20,500), or whose houses are built of materials that This also enables the standpipe tariff, which is kept low to indicate poverty. This innovative combination of an initial promote affordability. low cost or free block, with targeting, improves the utility’s ability to finance investment, since it avoids subsidizing The NWSC’s tariff structure (see Figure 6.3) is another middle class and wealthy households. Households with example of cross-subsidization. Domestic customers are property values greater than ZAR250,000 pay US$1.16/ charged less than nondomestic customers, no matter how m3 for the first 9 m3/month. much they consume. The NWSC charges households US$0.77/m3, a rate slightly below the weighted average Cross-subsidized tariffs for residential customers tariff (US$0.82/m3). The standpipe tariff is even lower Another way to achieve sustainable cost recovery, while at US$0.47/m3. This is vital to the NWSC’s strategy of ensuring affordability, is to charge more for water to affordable access, since most poor people in Uganda get those customer categories that are better able to pay. water from standpipes and kiosks. To compensate for the Such cross-subsidies between customer classes have been low rates charged to households, the tariff for institutional important to Dakar and Kampala’s ability to finance and commercial customers is above the average, at service expansion while maintaining affordability. The US$0.94/m3 and US$1.16m3, respectively. However, to SDE/SONES charges commercial customers a high tariff encourage large users to stay on the system, the tariff for of FCFA878.35 (US$1.75). This is higher than the first commercial consumption above 1,500 m3 per month (at two blocks of residential tariff, and equal to the highest US$0.93) is lower than the tariff for consumption below block of the residential tariff. that amount (US$1.16). Figure 6.2: SDE/SONES tariff structure, 2014 Source: ‘Grille tarifaire, 5éme Bimestre 2014’. 45 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Practical Techniques to Achieve Widespread and Affordable Access Figure 6.3: NWSC tariff structure, 2015 Source: NWSC Tariffs, November 1, 2015. Share of nonresidential customers in consumption than the average tariff, but not too high. To keep large users by volume on the system, nonresidential tariffs must be set lower than The NWSC’s ability to finance access and maintain the cost of alternate water sources for large users. affordability has been aided by a growth in the share of commercial consumption from 21 percent to 33 percent Helping customers manage cash flow (by volume), as commercial consumption by volume has Many poor households are able to pay for service, but do risen by an average of 11.1 percent per year.57 These trends not always have cash available when the bill is due. They indicate that charging commercial customers at rates above value payment options that allow them to pay when cash the average tariff has not resulted in reduced consumption. is available. Moreover, the money available for water may The decreasing block tariff for commercial customers could fluctuate from month to month. Poor customers value be a reason the cross-subsidization mechanism continues to options that allow them to know how much the water they work well. In contrast, for the SDE/SONES nonresidential are using is costing them, so they can keep consumption in consumption (as a share of total) has decreased over time, line with ability to pay on a day-by-day basis. Traditional from 33 percent of all water billed by volume (2004) to 7 utility meters and monthly billing cycles do not allow this. percent (2013).58 One of the most prominent and biggest innovations in water utility payment systems in recent years has been prepaid The takeaway from these cases is that cross-subsidization meters. Prepaid meters are perceived to be an attractive works as long as certain principles are followed. The average option for water utilities because they eliminate credit tariff should be sufficient for cost recovery. To ensure risk. The World Bank studied the use of prepaid meters in affordability for households, residential tariffs can be set several Sub-Saharan African countries to better understand lower than the average tariff, but rates should be no lower the benefits and costs in practical terms across their three than needed for affordability, and should apply only to main applications–for water dispensers (public standpipes), amounts required to meet basic needs. If possible, targeting individual connections (yard taps or house connections), can be used to limit access to subsidized water to those who and institutional/commercial customers.59 really need it, as is done in eThekwini. Commercial tariffs or tariffs for high levels of consumption can be set higher 59 Heymans, C., K. Eales, and R. Franceys. 2014. The Limits and Possibilities of Pre- 57 International Benchmarking Network for Water and Sanitation Utilities (IBNET). paid Water in Urban Africa: Lessons from the Field, WSP, World Bank, Washington. 58 IBNET. The discussion on prepaid meters draws on this report, not the case studies. 46 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Practical Techniques to Achieve Widespread and Affordable Access However, prepaid water dispensers are no less prone to ‘capture’ than any other valuable resource. In Kampala, some landlords deny prepaid customers access to ‘their’ meter unless they pay a premium, despite the agreement they sign with the utility that commits them to allow any customer access to the meter installed on or adjacent to their property. Some landlords insist on selling the water themselves with a substantial mark-up. The study found, however, that the economics of prepaid meters are marginal for both water dispensers and individual domestic connections. The financial analysis showed that a ‘typical’ utility in Africa would, in fact, make a loss on prepaid meters in these two contexts. However, the economics of prepaid meters are much stronger for commercial and institutional applications with high consumption volumes, and prepaid meters have been used successfully in Lusaka to get government departments to pay for water, a significant benefit. Prepaid water systems are not a ‘turnkey’ technical solution. All service providers studied found that they had underestimated the requirements to manage prepaid systems effectively. Managing prepaid systems is more demanding than conventional meters and billing, with greater complexity and system fragility as a result of the interdependent electronic, mechanical, software, and information system components. The additional demands Prepaid Metering: Although a more expensive technology, and add to the costs of prepaid systems for the service provider not always reliable, research shows many poor customers beyond the initial purchase price. see prepaid systems as allowing them to consume in line with their ability to pay on a day-to-day basis. Water utilities should carefully consider the benefits relative (Photo Credit: Kathy Eales). to the full management implications and costs of prepaid systems for their specific contexts. For both financial and practical reasons, it makes sense to implement prepaid The study found that prepaid water dispensers in Kampala, meters for commercial and institutional customers first Nairobi, and Nakuru have resulted in a sharp fall in what before extending their application to other areas. people without their own connections pay for water. Customers now get more water for less money, because they Other than prepaid meters, cities with good service to the receive the benefit of a lifeline tariff directly. In Kampala, poor have also developed a number of other interesting a 20-liter jerrycan costs just fewer than UGX25 (Ugandan options to assist poor families with cash flow management shillings) (US$0.01) from a prepaid standpipe. This works and staying connected. Some have adopted technical out at 55 percent of the cost from a house connection, and innovations to help low income customers keep their substantially less than the UGX200 to UGX500 and more consumption in line with their ability to pay, and to pay that water vendors and resellers charge. small amounts when money is available, rather than larger 47 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Practical Techniques to Achieve Widespread and Affordable Access amounts dictated by the utilities’ billing cycle. eThekwini’s While water sector professionals in cities need to contribute water bailiffs are one such example. to plans for holistic solutions to poverty, they also know that these plans may take decades to bear fruit. In the NYEWASCO, like other Kenyan water utilities, has adopted meantime, they have to find ways to serve the poor as best a technology to allow customers to read their own meters as they can. Approaches used by utilities to overcome each of frequently as they like, and then pay using mobile money these barriers in the short-term are summarized below. (on Kenya’s MPESA platform). This allows customers to monitor their own consumption and its costs, and to pay 6.2.1 Prepay water dispensers that cut the monopoly when cash is available. Regular meter-readings by the utility power of water vendors in Nairobi and Kampala cross-check readings done by customers. This costs less to In some Nairobi and Kampala slums, piped water service implement than prepaid meters. was lacking and households relied on informal water vendors. Due to lack of competition (enforced by violence to deter competitors) these vendors were able to charge high Another innovation is eThekwini’s electronic flow limiter, prices. In addition, slum landlords that had connections which monitors consumption. When the daily free water would charge tenants excessive amounts for water. (Nairobi allowance is reached, the flow-limiter slows the flow of is not a case study city, but is included because of its success water to a trickle. Originally it was developed as a way to in improving services to low income communities by using ensure that customers with poor payment records received water dispensers.) free basic water without running up further charges. Now low income customers with piped connections request flow For various reasons, the Nairobi City Water and Sewerage limiters to be installed, so they can avoid going over their Company (NCWSC) was not able to install piped daily allowance and incurring charges they may not be able networks throughout the slum. In Kampala, the NWSC to pay. This device could also be useful in systems that do also concluded that it can bring water to more people more not have free water allowances. It would allow families to rapidly through standposts, and that prepaid systems then limit their bills to a known amount, or to remain within the facilitate better cost recovery while helping poor customers first block of a rising block tariff. to control their expenses better. As an interim solution to mitigate these issues, these two utilities placed prepayment 6.2 Techniques to overcome nonfinancial water dispensers at regular intervals in and near informal barriers to good service areas where they already had a network. Those who live The obvious barriers to serving the poor are financial. Less near the dispensers can access water more readily, at a obvious barriers to serving the poor are legal, social, and fixed and reasonable price. In areas still primarily served technical in nature. This section examines how African by water vendors, prices have fallen due to competition cities have overcome three nonfinancial barriers to access from the dispensers. for poor people: water mafia and other on-sellers charge excessive mark-ups, informal land tenure does not allow 6.2.2 Small providers as utility agents serve informal service provision, and pattern of settlement or topography areas in Ouagadougou makes conventional network designs infeasible. ONEA, in Ouagadougou, does not have a formal mandate to serve informal areas, as inhabitants of these areas lack legal These issues are clearly multisectoral. From the perspective title to the land they occupy. Determined to nevertheless of the city as a whole, long-term solutions may lie in the ensure access to water for the poor families living in informal reform of land and housing markets. Release of adequate settlements, the utility has delegated service provision to land for housing, provision of title to the land and services small entrepreneurial providers. This relationship is shown to the plots, adequate transport for travel to jobs, fairer in Figure 6.4. tenancy laws–all these things should be considered. 48 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Practical Techniques to Achieve Widespread and Affordable Access Figure 6.4: Delegation of water service in informal settlements in Ouagadougou Source: Sawadogo, Dieudonné, ‘Delivering city-wide WASH services: reaching informal settlements in Ouagadougou, Burkina Faso’, August 2015, PowerPoint presentation. As shown in Figure 6.4, ONEA provides water to a meter plans to serve 80 percent of its service area population (shown as ‘M’ in the figure) at the edge of the settlement. with piped water to premises. The private operator manages the network within the settlement itself. These providers typically start in business The same approach has been applied in several other operating a water kiosk. However, they can then expand cities. In Maputo, Mozambique, for example, data show their delivery options, supplying multiple water kiosks that more people in the metropolitan area are being served (BF in the figure) and making connections to individual through small providers than by the formal utility. Box premises. Connection materials are provided by ONEA 6.1 captures the essence of this interesting case. It shows and the operators’ installation costs are reimbursed to that even though utilities may be–or even need to be–at encourage them to make connections. The operators are the core of an urban service delivery system, they could able to apply a small margin on the price of water paid to benefit from working with others to serve more people. ONEA. However, ONEA controls the prices that the small It also indicates that obtaining water in this way may providers charge, reducing the risk of small providers using cost poor people more, as the prices charged by the small local monopoly power to on-sell water at excessive prices. providers have been considerably higher than the tariffs of the public provider. By the end of 2014, there were 7,578 connections in five informal settlements of Ouagadougou. Figure 6.5 shows 6.2.3 Ensuring access in hard-to-reach areas in Nyeri access to piped water to premise in informal settlements in and eThekwini (Durban) Ouagadougou. Those settlements shaded blue, indicating In Nyeri, informal settlements are typically located on steep at least 90 percent access, are the pilot areas for this slopes, making construction of conventional networks initiative. The areas shaded in red, indicating low access technically infeasible. To serve the poor in these areas, to piped water to premise, will follow. By 2030, ONEA NYEWASCO first opened water kiosks from which small 49 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Practical Techniques to Achieve Widespread and Affordable Access Figure 6.5 : Access to piped water to premise in Ouagadougou’s informal settlements Legend Piped Network Piped Water to Premise Percent Access 0% - 10% 91% - 100% 0 1.5 3 ± 6 Kilometers Piped Network and Access to Piped Water within Premise in Informal Settlements City Characteristics Ouagadougou, Burkina Faso Data Sources: (1) L'Office national de l'eau et de l'assainissement (ONEA) (2) WaterAid Map Authors: Baker, Homann & Krishnan Date: 12/1/2015 Figure 7 Esri, HERE, DeLorme, MapmyIndia, © OpenStreetMap contributors, and the GIS user community private operators could sell water to people living in the while ensuring that very poor households have a water communities. However, the utility was not content with this supply adequate for their basic needs. However, the newly solution. It wanted as many households as possible to have urbanizing communities on the periphery of the more water piped to their premises. To achieve this, NYEWASCO formal urban areas are very dynamic–many families only provided a connection at a ‘meter bank’ at the boundary of stay in a community for a few months before moving on, the community. The utility then provided flexible pipes to and new dwellings are constantly being built. This resulted households that wished to connect from the meter bank to in the private connection pipes getting built over, people their home. This approach has been successful–88 percent forgetting where the pipes are, and new neighbors tapping of the poor have access to piped water to premises. into existing pipes, taking water that other households were being charged for. eThekwini’s response has been to eThekwini also experimented with offering a version move away from meter banks in informal and transient of meter banks in its peri-urban areas, characterized by communities and it is now aiming to provide individual lower density in a more rural setting within the municipal volume-restricted connections to each household. boundary. Tanks located at household premises are filled once a day with water needed to meet the basic needs of the 6.2.4 Do Pro-Poor Units help in serving the poor? household. The flow in the low pressure HDPE supply pipe Traditional utilities are well adapted to serving customers to the tank is controlled electronically at the meter bank. in formal areas of the city–customers with addresses and This prevents the theft of water that would occur if there regular monthly incomes. For the traditional utility, serving were continuously pressurized pipes in the community, poor customers with low and erratic incomes, living 50 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Practical Techniques to Achieve Widespread and Affordable Access Box 6.1: The role of private providers in Greater Maputo Maputo is not one of the case study cities. However, it illustrates how small scale providers can serve the poor in a situation where the utility is struggling. It complements the description of how ONEA worked with small scale providers in Ouagadougou. The metropolitan area of Maputo has between 1.7 million to 2 million inhabitants. The vast majority of households get their water from a yard tap or house connection. Maputo’s 500 or so water operators provide piped water services to about 191,000 households. This is more than the 185,000 households served by Águas da Região de Maputo (the main utility). The private providers range in size, with at least one serving 12,000 connections. The tariffs charged by private providers are higher than those charged by the main utility, averaging 35 Metical per m3, compared to the main utility’s charge of 14 Metical for the first social tariff block, 19 Metical for the next, and 25 Metical thereafter. Private water providers developed because of the lack of service and poor reliability of the public provider. Private households drilled boreholes for themselves, some then helped their neighbors and translated this into a larger business serving more people. Private operators now provide competing networks even in areas where a public network already exists. In some areas the survey found that private providers were more reliable, provided a service for longer hours compared with the public provider. More recently, the public provider has experienced difficulties with the adequacy of its water supply and has reduced its hours of supply due to maintenance problems. The survey also found that private providers offer more flexible terms for payment. For these reasons, many customers prefer using a private provider rather than a public provider even when there is a choice and notwithstanding the higher tariff. Private providers are 100 percent self-financed. Private water providers are practical in Maputo due to the readily available groundwater found in wide-ranging and shallow aquifers. This is a safe source of water in many, but not all, areas. Risks involve saline intrusion as the water table is lowered through over-extraction and contamination of the source with nitrates and pathogens from human waste as a result of inadequate or poorly maintained sanitation and wastewater infrastructure. The government has taken the initiative to both recognize and regulate private operators. In areas where groundwater quality is a problem, special regulations will apply, and the public provider will sell bulk piped water from the public network to the private water providers in these areas. All private providers will need to be licensed in terms of a Decree made by the Minister’s Council in October 2015, yet to be published, and licenses will address issues of mandate (area of service), water quality, and price. It is anticipated that this system, in the process of being established, will reduce conflicts between the public and private providers. Licenses will be issued by the local municipalities or district government and the system will be overseen by the national water regulator, CRA. A key challenge will be to establish capacity at the local and district government to issue licenses, and for this reason CRA will provide training and guidance. Sources: Personal communication, 2016; SUWASA, 2015: Sustainable Water and Sanitation for Africa: Final Report, USAID. in informal areas, with no address to send a bill to, is a poor customers and informal areas. Box 6.2 discusses this problem. Utilities in our sample have had mixed experiences polemic, with reference to a few experiences in the broader of creating a special unit to help them adapt to serving utility fraternity, not only the detailed case studies. 51 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Practical Techniques to Achieve Widespread and Affordable Access Box 6.2: Pro-Poor Units: Drivers for serving the poor? African water utilities with limited piped works, and with a mandate or requirement to recover their costs, are often perceived to predominantly serve businesses and the wealthier residents. In an effort to reorient water and sanitation utilities to meet the needs of poor people in large urban settlements, development agencies have encouraged utilities to set up dedicated Pro-Poor Units within their utility structures.60 Dedicated Pro-Poor Units are motivated on the grounds that service delivery to the poor in Africa’s large cities requires a special approach due to the challenges presented by unclear land tenure, unplanned layout, overcrowding, and a lack of accurate data. It is also argued that Pro-Poor Units are necessary to counter a perceived lack of political will on the part of utility owners and/or managers to serve the poor. In Nairobi, the utility also intends using the informal settlements unit to house “one-stop” facilities that would make it easier for the residents of these settlements to interface with the utility on all service related issues. Examples of Pro-Poor Units have been in Kampala (National Water and Sewerage Corporation), the Nairobi City Water and Sewerage Company (Kenya), Dar es Salaam Water and Sewerage Authority (Tanzania), and Lusaka Water and Sewerage Corporation (Zambia), among others. The National Water and Sewerage Corporation Pro-Poor Unit supports branches in the city of Kampala with informal settlements and works with headquarters and development partners to implement capital works programs targeting the urban poor. The Nairobi City Water and Sewerage Company’s Pro-Poor Unit was created mainly as a liaison unit based at headquarters. However, the utility found that branch office staff were reluctant to serve the poor and the unit has now been established as a seventh ‘region’ to give it greater status with respect to investment and operations. The Lusaka Water and Sewerage Company coordinates the implementation and operations of services in peri-urban and informal settlements. Dar es Salaam Water and Sewerage Authority implements and supervises the community-managed water and sanitation schemes; and is also responsible for DAWASA public relations and implementing the resettlement action plan. This study has shown that some utilities have achieved very extensive access by poor households to piped water and that this is achieved through building an extensive network within reach of most households. This requires the pro-poor strategy to be embedded within the overall strategy of the utility. The risk with Pro-Poor Units is that their activities may become isolated and marginalized from the rest of the utility. Nevertheless, these units can play an important role in catalyzing change, especially as a transitional mechanism in a utility reform process. In the National Water and Sewerage Corporation, for example, the setting up of a Pro-Poor Unit was one of a number of strategies used to kick-start a new initiative to improve access to water by poor people.61 In eThekwini, a dedicated unit was set up to initiate the expansion of piped water services into rural areas. After five years the activities of the unit were fully incorporated (“mainstreamed”) into the normal business of the utility. 60 Setting up Pro-Poor Units to improve service delivery. Lessons from water utilities 61 World Bank and the International Bank for Reconstruction and Development. in Kenya, Tanzania, Uganda, and Zambia. Water and Sanitation Program Field Note. 2014. “Do pro-poor policies increase water coverage? An analysis of service delivery in September 2009. Kampala’s informal settlements.” 52 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Practical Techniques to Achieve Widespread and Affordable Access Pro-poor units may improve data on poor customers, and thereby enable utilities to plan for and respond more effectively to service needs and problems. (Photo Credit: Kathy Eales). 6.3 Key points Service provided to the poor might be loss-making if Utilities in the cities that serve the poor relatively well analyzed on a stand-alone basis. The utilities in the five have taken special initiatives to help poor customers to cases do not analyze it that way. They see their mission as overcome financial and nonfinancial barriers to safe, providing service to all. Thus they try to provide all poor reliable, and affordable access. The basis for this has been households with a good quality service that they can afford, that the utility has been well-managed generally, so that while earning from the totality of their customers enough it is customer-oriented, held accountable, and willing revenue to cover their operating and maintenance costs and to innovate and take risks. The specific techniques are generate a surplus for investment. summarized in Table 6.1. These cities and utilities have also tackled nonfinancial The five utilities studied have, for example, implemented issues. They have not backed down when informal land rising block tariff structures targeted on poor households, tenure disallowed service provision, or physical conditions and cross-subsidized residential consumption by charging made network extension difficult. Instead, they improvised higher tariffs to commercial customers. The average tariff with new technology and institutional arrangements. They was thus kept sufficient for cost recovery, while targeting have taken counter-measures against water mafia and other subsidized water to those who really need it. on-sellers that inflated the cost of water. 53 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Practical Techniques to Achieve Widespread and Affordable Access Table 6.1: Summary of techniques used to achieve affordable access Barrier Description of technique Where used (examples) Financial barriers Connection costs Keep connection charges low NYEWASCO (Nyeri) has reduced its up-front payment for a new are unaffordable connection to US$26. Of this, about half is a refundable deposit against future bills. Only low income customers benefit from this low charge. NWSC (Kampala) cut its up-front charge to US$35. The NWSC also covers the cost of making a service connection of up to 50 meters. This low charge is only available to low income consumers. Provide convenient standposts and The NWSC uses standpipes to serve 71 percent of poor people shared taps in Kampala. Roundtrip times average just 21 minutes. The NWSC (so customers can access water has also made it easier for households to serve their neighbors without paying connection fees) by charging a lower tariff to any connection that serves more than two or three households. Niamey serves 78 percent of poor people through standpipes. By building standpipes in convenient locations, roundtrip time has been reduced to 12 minutes. (Niamey is not a case study city, but is one of the seven cities with relatively good service for the poor.) Monthly bills are Charge poor households lower eThekwini (Durban) does not charge poor households for up to 9 unaffordable tariffs for a basic needs level of m3 per month. consumption SDE (Dakar) charges $0.40 per m3 for consumption below 10 m3 per month. Keep standpipe charges low Kampala (NWSC) charges $0.47 per m3 for standpipe consumption. Provide information and flexible NYEWASCO (Nyeri) offers customers a mobile app that lets payment options to help customers them read their own meter at any time, and get an updated bill. manage consumption and cash flow Customers can also pay some or all of their bill at any time, using mobile money. (Other Kenyan utilities, including Nairobi, have similar systems.) eThekwini (Durban) uses electronic flow-limiters that cut off supply when the free water quantity for the day has been reached. This helps customers avoid incurring bills entirely. Nonfinancial barriers Monopoly power of Water dispensers (standpipes that NWSC (Kampala) placed prepaid water dispensers (automatic water vendors and operate automatically to dispense water vendors) in poor communities. landlords water in exchange for electronic NCWSC (Nairobi) placed prepaid water dispensers at frequent payment) intervals along the edge of slums. (Nairobi is not a case study city, but its experience is shared because it has been successful in driving down prices charged by informal vendors in slums through use of water dispensers.) Legal restrictions Use of small scale providers as ONEA (Ouagadougou) uses small scale providers as its agents to on the utility serving agents of the utility serve informal settlements. These agents buy water from ONEA, properties in build small networks into the informal communities, and on- informal settlements sell the water through standpipes and individual connections in these communities. ONEA gives the small providers the material to build out these connections, and also regulates the price at which the small providers may sell. 54 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Practical Techniques to Achieve Widespread and Affordable Access Barrier Description of technique Where used (examples) Terrain, settlement Provision of meter banks on the NYEWASCO (Nyeri) serves poor settlements on steep slopes pattern, and lack edge of the settlement, from which where it cannot build by offering connections at the edge of of streets makes households can make their own the settlement. Several meters are placed at one location, installing utility connections and households run a pipe from the meter to their premises. connections in NYEWASCO also provides the piping materials to the low income low income areas households. difficult Utility processes Pro-Poor Units NWSC (Kampala) has a Pro-Poor Unit. This unit helps the and mindset are not operating branches of the utility to serve poor communities suitable for serving appropriately, including helping with the installation of networks poor communities poor communities. eThekwini (Durban) developed a Pro-Poor Unit, then mainstreamed the pro-poor thinking into the entire utility. 55 VII. Conclusion Africa needs new approaches to urban water that utilities can serve the poor well, and indeed are far and supply away the most important providers of water to the poor in Access rates for water piped to the premises in urban Africa the cities studied. Even though they may in some situations are falling, while access rates to improved water are rising only have to partner with other institutions, there is no obvious marginally. Meanwhile, the urban population is expected to alterative to dedicated, professional water utilities as the triple by 2050, increasing from 340 million people today instruments to develop and manage networks on scale, as is to over 1 billion people. Demand for water will rise, water required in Africa’s rapidly growing cities. resources will become more expensive, and financing needs will increase significantly. How can the SDGs–which target Financing good water service “universal and equitable access to safe and affordable drinking To turn around service to the poor, utilities need substantial water for all”–be met in this context? investment to build distribution networks and production facilities. At the start of their turnaround, most of the Provision of good water service to the poor in utilities studied have needed concessional loans and some African cities is possible grants. They then quickly worked to improve operating The analysis here offers many lessons from those cities efficiency–by reducing NRW, improving staff productivity, that have already made significant progress towards the and increasing collection rates–and then to expand access challenging goal of good services to the poor. Household (thus also increasing revenue). Increased operating cash survey data show that more than 90 percent of the poor then allowed them to service the concessional loans, and have access to improved water across a range of African to finance additional capital investment. Using similar cities, and that in seven such cities included in this study, financing strategies, typical African cities may be able water is also available for at least 18 hours per day. The to serve the poor well, as the evidence for Mombasa in report classified the combination of such coverage and Section 6 shows. frequency of supply as providing relatively good service to the poor. The experience of these cities show that the Ensuring good service is affordable provision of water supply is possible in rather difficult urban To service the debt used to expand access, utilities need to environments: densely populated, rapidly urbanizing, and recover their costs. To serve the poor well, utilities need located in countries with limited freshwater resources, to also ensure access is affordable. This is often portrayed relatively low household income levels, and not necessarily as a trade-off, yet good utilities transcend the traditional conducive governance environments. trade-off between cost recovery and affordability. They minimize up-front connection charges, and provide a In the cities where the poor are served well, traditional basic needs quantity of water at a low tariff. They also utilities are the main service providers to the poor, and use innovative technologies–such as prepaid and limited some of them have ventured into partnerships with others– flow meters or cellphone-based payment platforms–to such as small providers–to address areas of service delivery help poor households to manage their water payments in where they lacked capacity, experience or legal mandates line with their cash flows, and manage consumption to an to operate. These utilities have in common that they are affordable level. effectively managed, having high operating cost coverage ratios and scoring well on other measures of efficiency and Utilities also innovate to overcome nonfinancial barriers. cost effectiveness. The research shows that making the utility Prepayment water dispensers also have this effect–often used work well is a core part of achieving improved services in in convenient locations in informal settlements, to compete cities. This is not to say that complementary strategies with water vendors and drive down their prices. Some cannot also provide good service to the poor, but rather utilities delegate service provision to small entrepreneurs in 56 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Conclusion informal areas where they lack a formal mandate because and leadership–delivered by bureaucratic entrepreneurs– inhabitants lack legal titles to the land they occupy. which are needed in all phases of reform. To turn around poor service, cities must work Over time, these three factors help foster virtuous circles, with the grain of the political economy strengthening utility capacity and service to the poor even While good service for poor households is possible, it is not in inauspicious contexts. Reforms can persist and deepen. the norm. The dysfunctional political economy in many For instance, political and bureaucratic opponents buy in African countries, related to poor governance arrangements, to reforms. Tariffs can become less politicized when utilities often leads to a ‘low equilibrium trap’ at the utility level. operate from a position of strength and a record of success Those in power choose predation, using the water utility as a on which to justify increased charges. In this stage too, vehicle from which to extract public resources for personal development partners can play a role through disseminating or political ends. comparative knowledge, but also by continuing to finance the utility whose performance is improving, to sustain and Why then, in the context of the general political protect good service. environment and poor governance arrangements, do some African utilities serve the poor well? While difficult, reform As good service is achieved, however, the risk of predation within a dysfunctional political environment is possible. In does not disappear. In fact, as utilities become more the five cities analyzed in detail, there were three factors financially successful, predation could well increase. Formal typically involved in the starting of reform–a catalytic governance rules and managerial competence may be useful event such as a water supply crisis, an exemplary utility instruments to keep this at bay, but the more enduring way leader, and a secure political leader (see diagram below). to counter predatory behavior is to develop networks of The confluence of these factors is determined by the local internal and external alliances. However, if those predators political economy, and cannot easily be influenced by come from erstwhile government allies–whether seeking outsiders. However, where the political economy conditions rents or political opportunism–restraining them can prove are favorable for reform, financial and technical support are challenging. Changes in political or professional leadership helpful. Development partners, noticing these conditions can also increase the risks of predation. Again, this issue can are in place, can play a role by providing initial finance for be addressed by implementing systems involving multiple the turnaround. stakeholders, which will protect the utility from predation while promoting accountability. Equally critical, embedded To build momentum for reform, three factors–internal alliances with domestic stakeholders, ‘not among the voters competence, external alliances, and institution-building– of one party or one section, not in a single class or interest are key. Which to tackle first depends on context. group, but in multiple and diverse political affiliations’ may Successful reforms are problem-focused, well-timed, and best afford utilities, and their leaders, sufficient leverage and adaptive. They are not necessarily direct applications of autonomy to sustain the process of reform.62 ‘best practices’, but rather are creative next steps that allow the reform process to continue accelerating, with sensitivity Furthermore, formal governance arrangements should to the specific context. aim to embed relations of accountability with external stakeholders who benefit from success, and to support At the managerial level, sustainability requires building a professionalism in the utility’s culture. However, institutional deep management bench, and a culture of joint planning, form in itself is no guarantee of success–and neither is being accountability, and transparency. Management reforms and public, nor being private, a fundamental precondition for institutional culture and capacity-building, an essential success. The successful cases analyzed here have ranged from component of both the ‘building momentum’ and corporatized public to private water service providers, and ‘sustaining reform’ phases, are important dimensions of the to water or engineering departments that remained part of process. They are also outgrowths of effective management 62 Carpenter (2001: 5). 57 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Conclusion municipalities. They have also included national and more One thing is clear though, given the limited concessional localized institutions. loan and grant finance and the high investment needs–as operating cost recovery improves, utilities would gain from Role of development partners transitioning to blended or commercial financing, rather So where should development partners focus scarce than remaining reliant on grant financing. concessional loans and grant finance? The political economy analysis does not suggest a fully replicable model–the local On the other hand, where promising conditions are not in context must always be considered. Rather, the research place, funding the utility may not lead to sustainable services shows that good reforms have developed from the right for poor. Capacity building and reforms that can improve combination of political economy dynamics. It is advised governance–for instance, improving transparency and that would-be reformers, in politics or at the utility-level, if promoting information sharing–would be more practical given an opening through an exploitable catalytic moment, in these cases, in the hope that the country will develop use that opening to build momentum. Development towards conditions where transition to a commitment and partners, spotting political or technical champions, should capacity to serve the urban poor is more likely. back them. Development partners can support reforms through the Under these conditions, finance could be phased in, with three phases–starting, building, and sustaining. They emergency needs addressed first. Additional funding cannot initiate nor control the reform, however. Successful to improve management effectiveness and operations, utilities must master the complex interdependence between perhaps coupled with technical assistance from private the technical, the financial, and the institutional to work operators, could come next. A large financing deal may with challenging political environments to better serve the wait until the commitment to reform appears credible. Yet, poor in their cities. Political leaders can win by partnering development partners may also need to take some risks with strong technical leaders to build effective institutions before improvements are seen. There is no universal strategy. that earn the support–and the votes–of the poor. 58 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms Appendixes 59 Contents Appendices Appendix A: Case studies: How Cities Turned Around Water Service for the Poor..............................................61 Appendix B: Suggestions for Further Research..................................................................................................90 Appendix C: Perspectives on Alternative or Supplementary Service Providers.....................................................91 Appendix D: Utility Management Effectiveness Index..........................................................................................92 Appendix E: Longitudinal Utility Data..................................................................................................................94 Appendix F: Data Sources...............................................................................................................................129 Appendix G: References..................................................................................................................................132 Tables Table C.1: Perspectives on alternative or supplementary service providers......................................................91 Table D.1: Utility management effectiveness index..........................................................................................92 Table D.2: Inputs for utility management effectiveness index...........................................................................93 Table F.1: Household survey data sources...................................................................................................129 Table F.2: List of data sources for utility-reported data..................................................................................130 Figures Figure A.1: Dakar provides water piped to the premises to (almost) all..............................................................62 Figure A.2: Operating expenses and tariffs, operating cost recovery ratio.........................................................64 Figure A.3: The number of connections in eThekwini increased by 35 percent over 11 years with some experimentation in alternatives.......................................................................................................67 Figure A.4: Operating cost coverage, eThekwini Water and Sanitation Unit.......................................................68 Figure A.5: Access to piped water by poor, Kampala.......................................................................................73 Figure A.6: Achieving financial self-sufficiency at NWSC...................................................................................75 Figure A.7: Sources and amounts of capital expenditure financing....................................................................81 Figure A.8: Cash from operations compared to debt service payments............................................................81 Figure A.9: Nonrevenue water, staff productivity (NYEWASCO).........................................................................82 Figure A.10: Cost recovery and real average water tariff over time......................................................................82 Figure A.11: Formal and informal settlements in Ouagadougou..........................................................................85 Figure A.12: Delegation of water supply to private operators in informal areas....................................................87 60 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix A: Case studies: How Cities Turned Around Water Service for the Poor Appendix A: Case studies: How Cities Turned Around Water Service for the Poor A.1 Dakar, Senegal In 1994, Dakar, the capital of Senegal, was running out This turnaround was made possible by an innovative of water to serve its 2 million inhabitants. Poor people in scheme to bring in a private firm to operate water supplies. the city mostly relied on standpipes. Twenty percent of the The private firm increased cost effectiveness, generating population had no access to piped water at all, relying on free cash flows. These cash flows serviced the multilateral water vendors and shallow wells. Water was rationed to 16 development bank loans that financed expansion of water hours a day. A growing population and depleting water production and extension of service to poor areas. resources spelled imminent crisis. The reform was planned with the assistance of the World Twenty years later, 75 percent of poor people have water Bank, which also provided much of the financing. piped to their premises. Standpipes serve almost all the Government buy-in was made easier by the relatively secure remainder. Overall, 98 percent of poor people access piped position of the President, a reform plan that preserved jobs, water supply, which is available nearly 24 hours per day. The and public sector control of the investment program, and investment and operating costs of these improvements have a belief that the plan could end a water crisis and benefit been covered largely from tariffs charged for sales of water. poor constituents. Service before In 1994, the utility was only able to supply 60 percent of demand. Public standpipes were the main provision the reforms made for poor people. Fourteen percent of total population of the city relied on standpipes, drawing just 20 lpcd–barely enough to meet basic needs. A further 20 percent of the population had no access to piped water at all, relying on water vendors and wells. Supply from the utility was limited to 16 hours per day, with low pressures and contamination of the piped supply in some areas. Catalyst for Impending crisis catalyzed reform. Inadequate water supplies threatened a city with population growth reform running at over 5 percent per annum. Ground resources near the city were depleted. The water table was falling as much as 1.5 meters per year in some areas. Saline intrusion threatened to ruin the aquifer. Shallow wells, relied on by poor people and others, were being contaminated with seepage from latrines.1 The only sustainable solution was construction of a new system to bring water from the Senegal River, about 250 km away. This system was expected to take eight to 10 years to complete. In the meantime, additional boreholes, expansion of a treatment plant, and leakage reduction works were urgently needed to prevent further deterioration. This was expected to cost US$162.5 million (in 1995).2 Actors driving The government turned to the World Bank for assistance. World Bank staff believed that private operation the reforms of the system was needed to make the utility efficient and made finance conditional on this.3 The President of Senegal at that time, Abdou Diouf, was in a relatively secure position. His party had been in power since 1960. Potential opponents were being included in government to build ownership of the reform across the political spectrum. It has been argued that this approach was supported by donors because the resulting political consensus made it easier to enact economic reforms in a number of important areas.4 1 World Bank. 1995. Staff Appraisal Report. Water Sector Project, 2–4. 3 World Bank. 1995. Staff Appraisal Report, paragraph 8.2 2 World Bank. 1995. Staff Appraisal Report, 24. 4 Beck, Linda. 1999. “Senegal’s Enlarged Presidential Majority: Deepening Democ- racy or Detour?” In State, Conflict & Democracy in Africa, edited by Richard Joseph, 214. Boulder: Lynne Rienner. 61 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix A: Case studies: How Cities Turned Around Water Service for the Poor Course of action A comprehensive long-term plan was developed. The first stage of the plan focused on quick increases in decided on production and extension of service into some poor areas. The second stage involved the much larger Keur Momar Sarr water treatment plant (in 2005, 65,000 m3/d; upgraded in 2008 to 130,000 m3/d) along with a continued expansion in the distribution network into poor areas. The infrastructure expansion plan was anchored on institutional reforms. The water supply service of the state-owned utility was divided into two companies: the SDE became the company responsible for operating and maintaining the infrastructure, providing service to customers, and collecting revenue. The infrastructure assets were transferred to SONES, a new publicly-owned company responsible for planning and financing new investments. A controlling stake in the SDE was to be transferred to an international water operator through a competitive tender. Sanitation services were spun out from the water utility into a new public entity (ONAS), since sanitation needed continued subsidies. Technical Production capacity was increased by about 30 percent (up 60 MLD) in the first phase through expansion solutions of an existing treatment works and an emergency program of drilling new boreholes. In the second phase, implemented the 130 MLD Keur Momar Sarr water treatment plant (near the Senegal River) and associated transmission and treatment works were completed. With more water available the system was able to meet the city’s demand for water. Another important technical solution was network rehabilitation and active leakage control. In the first phase (in which production capacity was still not adequate to meet demand) 100 km of pipe were rehabilitated, along with 25,000 connections.5 Between 1995 and 2013, the distribution network in the SDE/SONES service area was more than doubled by adding 4,800 km of pipe. The number of connections in Dakar was almost doubled, from around 189,000 (2002) to around 368,000. In densely populated areas, most poor families have water piped to their premises, while in less densely populated areas standpipes are more common. Figure A.1 shows how the network extends into poor areas (outlined in red). In the majority of these poor areas, at least 80 percent of households have water piped to their premises. L 5 World Bank. 2004. Implementation Completion Report. Senegal Water Project, 10. I P Figure A.1: Dakar provides water piped to the premises to (almost) all Legend Malika Piped Network Poorest 40% Keur Massar Improved Water Sources Percent Access 81% - 90% 91% - 100% Rufisque Malika Keur Massar ± Rufisque Ngor Yoff Ngor Daliford 0 2 4 8 Yoff Daliford Kilometers Mbao Mbao Access to Improved Rufisque Est Rufisque Est Water Sources Hann Bel-Air Water Access (2014) Hann Bel-Air Dakar, Senegal Data Sources: (1) ANSD (2) SONES Plateau D Map Authors: Baker, Homann & Krishnan ( ( Plateau M B Date: 12/18/2015 Figure 3 Esri, HERE, DeLorme, MapmyIndia, © OpenStreetMap contributors, and the GIS user community 62 Da Esri, HERE, DeLorme, MapmyIndia, © OpenStreetMap contributors, and the GIS user community Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix A: Case studies: How Cities Turned Around Water Service for the Poor Commercial In 1995, most poor people served by the utility got water from standpipes. Earlier in the 1990s, the standpipes solutions had been given to small entrepreneurs to operate. This was done to reduce waste of water. The standpipe implemented operators sold water by filling containers, and paid the utility for the water used as recorded by a meter. Standpipes are still an important method for delivering service to poor people, with 22 percent of poor people in Dakar getting water this way. Senegal’s increasing block tariff structure has a subsidized social tariff (tranche sociale) for levels of consumption below 20 m3 (US$0.40) per bimester.6 There is also a regular tariff for consumption from 21 m3 to 40 m3 (US$1.39), and a “dissuasive” tariff for consumption above 40 m3 (US$1.75). The dissuasive tariff is designed to be a disincentive for excessive water use. It can be seen that the tariff for household consumption of less than 20 m3 per 60 days is less than a third of the regular tariff, and less than a quarter of the tariff for consumption in the top block. A family of six consuming 50 liters each per day would pay US$3.62 on average each month. Bills are sent every two months based on meter readings, and the SDE can cut off water supply for nonpayment. The standpipe tariff is US$0.73 per m3–this is between the social and regular tariff for domestic piped connections. A disadvantage of rising block tariffs is that they can discourage utilities from serving poor neighborhoods. This is because the utility will expect most customers to pay only the lowest tariff. If this is lower than the cost of service, the utility will lose money–or have to put up tariffs to other customers. The affermage contract for Dakar gets around this problem by paying the private operator the same tariff for every unit of water sold. A more unusual kind of subsidy was introduced in 2007. The cost recovery model indicated that the average tariff should be increased, so that the operator’s tariff and SONES debt service requirements could be covered. Rather than increase all tariffs equally, the government agreed that the tariffs for government customers should be increased by 70 percent, while the tariffs for other customers remained constant. In 2009, the government tariff was raised again, to allow other tariffs to remain constant. In 2015 this arrangement was reversed, necessitating a 4 percent increase in the tranche sociale and a 9 percent increase in the rates to other users. Financing Total urban water sector investments from 1996 to 2013 totaled US$770 million.7 About half was financed solutions by loans from donors. Two major programs implemented were the Senegal Water Project (US$223 million, implemented 1996-2004) and the Long-Term Water Project (US$255 million, 2002– 2009).8 The World Bank was a major financier for both projects, providing US$85 million and US$146 million, respectively. The interest rates on these loans were around 6 percent. Tenors were 20 years (including five-year grace periods).9 The cost of finance was kept down because the investments were the responsibility of the publicly-owned SONES. This meant standard donor rates were applied. (If investment had been the responsibility of the private operator, the cost of the finance would have been higher.) These loans had to be repaid. The private operating company (SDE) was limited by its contract to charging a fixed operator tariff. For the sector to be financially self-sufficient, the average retail tariff paid by customers had to cover the operator tariff, plus an amount to service the debt taken on by SONES. Keeping the tariff at affordable levels was a challenge. Thus it was important to keep the operating cost component of the tariff to the lowest possible level. The decision to bring in a private operator was made in part because analysis suggested that this would result in lower tariffs. The winning bid (from SAUR) was for an operator tariff that was 60 percent of the tariff the public utility had been charging customers. 6 Every two months. 9 World Bank, Staff Appraisal Report, Senegal Water Project, iv; World Bank, Project 7 World Bank. 1995. Staff Appraisal Report. Appraisal Document: Long Term Water Project, 20. 8 Implementation Completion Report, Senegal Water Project; Implementation Completion Report, Long Term Water Project. Note: These projects included sanitation components which is overseen by ONAS, not SDE or SONES. 63 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix A: Case studies: How Cities Turned Around Water Service for the Poor To make money within the agreed operator tariff, the SDE had to maximize cost effectiveness. It maintained a collection rate of 98 percent, reduced NRW from 29 percent to 20 percent, and tripled staff productivity, moving from seven staff per 1,000 connections to just two staff per 1,000 connections. Because of these gains, average operating expenditures in real terms declined slightly over the period. Fairly modest real tariff increases (see Figure A.2) then allowed the sector to generate a substantial operating surplus which could be used to service debt, or to pay for capital expenditure directly. Figure A.2: Operating expenses and tariffs, operating cost recovery ratio Source: Utility research (see Table F.2 for full list of sources). Management The most important management solution was to bring in an international private operator. That operator, in solutions turn, brought in modern management systems and techniques, and skilled managers. The operator trained implemented local staff, so the company is now run almost entirely with national staff, at high levels of efficiency. Profit incentives for owners, in turn, led to performance-oriented management for the staff. Institutional and The management model involved a significant realignment of roles and responsibilities in the sector. governance Operational autonomy was conferred on the privatized operating company, which was controlled through a solutions long-term affermage contract that specified what the operator had to achieve, and controlled what it could charge. Reflections and conclusions Results of the By 2014, 75 percent of poor people in Dakar had water piped to their premises, with a further 22 percent turnaround accessing standpipes. Compare this with 20 years previously, when most poor people with piped water got it from public taps and 20 percent of the entire population–most of them poor–had no access to piped water at all. Across the city as a whole, 86 percent of people have water piped to their premises. Water supply is reliable and has been available 24 hours per day at constant pressure across most of the system for most of the period since 2007. That said, water demand is once again running ahead of supply. Some residents complain that water does not reach the higher stories, or is not available in certain areas at certain times. The significant increase in access and service seen in Dakar was achieved without burdening Senegal’s stretched public finances. Finance for investment was available on concessional terms because of the credible institutional model adopted. Debt service on the loans was covered out of the utility’s cash flow because the private operator managed to run the system cost-effectively. 64 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix A: Case studies: How Cities Turned Around Water Service for the Poor How a mandate A comprehensive reform plan was prepared with the assistance of the World Bank. The plan was acceptable for the reform to stakeholders because workers were guaranteed employment under the new arrangements, tariff increases was secured were limited, and public sector managers retained control of the investment program. The lack of other financing options, and World Bank insistence on the reforms as a condition of financing, contributed to the Government’s acceptance of the approach. Acceptance was made easier by the cooption of potential political opponents into the Government at the time. Another factor which may have influenced the reform direction was experience with private management of the utility prior to 1971 by the private operator Generale des Eaux (later renamed Veolia Environment). In addition, SAUR, the private operator that won the tender, had been providing technical assistance to the public utility for some years prior to the reforms. Phasing the The reform in Dakar followed a blueprint laid out at the start. Quick measures to increase water availability reforms and extend service to poor areas were prioritized. Institutional changes were implemented in parallel. While these measures were taking effect, planning for the big increase in water supply the city needed was already under way. The Dakar turnaround is remarkable for its institutional continuity. It has continued in the same basic form for almost 20 years, surviving a change in administration when opposition leader Wade was elected President in 2000. It also continued through the next change in administration in 2012. Sustaining Supply is once again exceeding demand in Dakar. Major new production sources are being developed, but performance there are concerns they will not come on-stream in time. In 2013, a transmission main that carries water from a source 250 km out of Dakar ruptured. This main, which supplies 40 percent of the city’s water, took three weeks to restore to service. The resulting shortages led to outbreaks of social unrest. Some stakeholders suggest that SONES has not been sufficiently proactive in infrastructure development and major rehabilitation works. Indeed, the government recently gave the SDE responsibility to implement some urgent expansions in supply. The affermage model depends for its sustained success on the continued strong performance from the public sector entity in charge of infrastructure development. If this is lacking, the model will need to evolve toward greater private sector responsibility or risk failing to deliver. Another risk to sustainability in Dakar is the periodic expiration of the affermage contract. While the limited term provides the government with flexibility, it also builds in a discontinuity. Every 10 years–more or less– the system has to figure out whether to rebid the contract, bring operations back into the public sector, or negotiate an extension. Despite these risks, stakeholders in Dakar say that the system is generally working well. Accountability is institutionalized through contracts. Reliance on donor finance for investment promotes good governance and limits predation in SONES. As a private company, the SDE is able to insulate its operations from political predation to a great extent. The fact that the SDE and SONES can each perform the other’s functions to some extent provides tension, but also resilience. The SDE has stepped up where SONES has appeared to lag, but this in turn is likely to encourage the latter to improve its performance so it can earn back the territory. 65 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix A: Case studies: How Cities Turned Around Water Service for the Poor A.2 eThekwini (Durban), South Africa When apartheid ended in South Africa, local governments autonomy, and develop institutional capability to provide strove to undo the racial inequalities which had forced effective public services. the black majority to live in townships spread around the periphery of the more affluent suburbs reserved for white Over an 11-year period (2004 to 2015) the number of people. These townships had poor urban services compared connections increased by over 129,000 (from 369,000 to to the traditionally white areas. 498,000), an increase of 35 percent at an annual rate of 2.8 percent. The network length nearly doubled from 6,998 km In the Durban area, over 46 separate municipal councils in 2000 to 13,900 km in 2015. The 2011 Census showed were amalgamated over a period of six years (ending that 80 percent of the population had access to water on in 2000), to form a single metropolitan municipality, their premises and a further 17 percent had access to an eThekwini. The Durban water department now had the improved piped water source, giving a total access of 97 responsibility to manage and extend services over a much percent. About 55 percent of the poorest 40 percent of the wider and poorer area. The service area was extended by 68 population have access to water piped to the premises, and percent, adding a further 9 percent to the population of a further 38 percent to a public tap–that is, 93 percent of 3.6 million people. The far-reaching and complex political poor people have access to a safe piped water source. and institutional changes brought about by the transition to a universal democratic franchise created the space For the 97 percent of households using piped water, water for ‘public entrepreneurship’ whereby public managers is available 24 hours a day and is safe to drink. The city were able to build external alliances (within a broad received the highest award (‘blue drop’) by the national water and progressive political mandate), establish managerial ministry for its management of drinking water quality. Service before Reliable data on the situation before 1995 are not available. White people living in and near the city center the reforms (the original Durban municipality) received on-premise, good quality, 24x7 water. However, black people living in townships received an unreliable service and rural households received no service at all. Catalyst for The democratic transition in South Africa in 1994, from an exclusive race-based franchise to universal reform franchise, and the subsequent reform of local government to amalgamate previously racially separate local governments, created the catalyst for the reform of Durban water. Actors driving Neil Macleod, head of the Durban water department, an engineer with a Master’s in Business Administration, the reforms took up the challenge of extending a good service equitably across a wider and poorer area at the commencement of the political changes in the early 1990s. He was able to work within a stable and enabling political context created by the Executive Mayor, Obed Mlaba (1996–2011), and later with an effective City Manager, Mike Sutcliffe (2002–2012), to establish and sustain the Water and Sanitation unit as a ‘semi- autonomous’ (from a management perspective) entity within the municipal administration over 20 years. Course of action The extension of the service boundary involved taking on additional preexisting service areas (as well as decided on previously unserved areas), and the incorporation of operations and staff from other municipalities. This required normalization of staff conditions across the operations and the creation of a single corporate culture and identity. At the same time, new investments were required and existing services needed to be rehabilitated. Technical A key challenge was to provide universal but affordable services to all people in the service area. A solutions conventional full-pressure system with individual connections and meters for each household was implemented considered too expensive, given the high needs, and the relatively low densities and low income levels on the urban periphery (where most of the population without a piped water supply lived). eThekwini solved this problem by providing a full-pressure metered connection to a local off-take for a group of houses (typically 15 households) and then reticulating through separate inexpensive and flexible 66 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix A: Case studies: How Cities Turned Around Water Service for the Poor small diameter HDPE pipes at low pressures to a 200 liter ground tank at each house. An electronic bailiff unit was developed that automatically filled each tank overnight. This fulfilled the City’s commitment to provide a free basic service in an affordable way. (Households could opt for a higher level of service but would have to pay for this.). About 5,000 units were installed (see Figure A.3). In time, difficulties were experienced with the management of this system. The population in the peri-urban areas grew more rapidly than anticipated and new houses were constructed over existing pipes. New residents made illegal connections to their neighbors’ pipes. In response, the Water Department innovated again, developing an electronic flow-limiter. Poor households are now offered a full pressure connection. The electronic flow-limiter assists households to keep their consumption within the free basic amount of 300 liters per day by limiting flow to a trickle when this amount is exceeded. Figure A.3: The number of connections in eThekwini increased by 35 percent over 11 years with some experimentation in alternatives Source: Utility research (see Table F.2 for full list of sources). Commercial eThekwini adopted a policy of free basic water in 2000. At first, the free basic amount was calculated on solutions the basis of 25 liters per person per day, a minimum requirement for health specified by the World Health implemented Organization. A monthly amount was calculated assuming eight persons per household to give 200 liters per day, or 6 m3 per month. This was provided to all consumers, due to the difficulty in identifying only the poor households. Following feedback from consumer forums, the free basic water allowance was increased to 300 liters per household per day in July 2008. Households with restricted water services (ground tanks or flow limiters as described above) had the daily amount delivered to them increased by the utility. For customers on regular metered service, consumption in excess of the free water limit was charged at the standard tariff. Tariffs escalate with increased usage, with the top marginal tariff (for consumption above of 45 m3 per month) set at the long run marginal cost of supply. This is above the average historic cost and is both efficient and surplus generating. In July 2012, eligibility for free basic water was restricted to customers whose property value was less than ZAR250,000 (about US$20,000). Customers with higher value houses (an indicator of wealth) are now required to pay the regular tariff for all water consumption. 67 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix A: Case studies: How Cities Turned Around Water Service for the Poor Financing Immediately after the democratic elections, the municipality focused its investments on expanding access solutions to poor households. These initiatives were supported across South African municipalities by the national implemented government through capital and operating grants. Initial grants were project specific and were part of the national Reconstruction and Development Program. These grants were transformed into sector-based programmatic grants. The multiple grants were then aggregated into a single capital grant called the Municipal Infrastructure Grant and a single operating grant, the Equitable Share Grant. In addition to the grant money from national government, the municipality has raised commercial finance. For example, it established a EUR50 million facility with the European Investment Bank for infrastructure investment in 2010. The utility has maintained a positive operating cost recovery ratio (see Figure A.4) by implementing cost- recovery tariffs, and has consistently achieved a high cash collection rate (averaging over 100 percent over the last few years as a result of successful arrears debt collection).10 This means that the utility is able to cover all its operating and maintenance costs from the tariff, and generate free cash flow which is available to the municipality to service the debt raised to finance new investment. Figure A.4: Operating cost coverage, eThekwini Water and Sanitation Unit Source: Utility research (see Table F.2 for full list of sources). Management eThekwini developed a people-focused and trust-based organizational culture in which staff contributions solutions were recognized and valued. The eThekwini management has consistently emphasized that it regards implemented people as the most important asset in the organization. The Managing Director (MD) adopted a personal and engaging management style, being quick to give visible recognition to good staff performance through awards. eThekwini established a customer management unit with equal status to other major management units such as finance and the engineering functions. The MD built trust in the management team and communicated effectively with staff, frequently visiting staff in the field. He fostered a positive attitude by encouraging people to propose solutions to the problems they brought to management. Perhaps unusually for a department within a municipality, he encouraged talking rather than writing, with minimum use of memoranda, minutes, and lengthy reports. The MD negotiated a performance contract with the City Manager linked to the water unit’s overall performance. This contract made 25 percent of his remuneration dependent on meeting performance targets. These performance agreements were applied to the senior management team. 10 The water unit produces a set of income statements for water that are then amalgamated with the City’s finances (all services). Separate cash flow and balance sheets are not available for water and the separate income statement for water is not published. The income statement contains an accounting entry with an estimate of the opportunity cost of revenue forgone by providing free basic water (included as negative revenue). This item has been removed from the income statement, so that the revenue used here is the actual amount billed. 68 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix A: Case studies: How Cities Turned Around Water Service for the Poor Institutional and The MD protected management and staff from outside interference in operations while demanding governance accountability, thereby building staff loyalty and promoting staff retention. The MD’s response to inappropriate solutions requests was to insist these were put in writing. The MD was thus able to establish and sustain the Water and Sanitation unit as a ‘semi-autonomous’ (from a management perspective) entity within the municipal administration. The unit had a distinctive corporate culture (it had its own separate head-office), ring-fenced finances (it had a separate income statement), was responsible for meter reading, and had its own financial and human resources staff involved in the day-to-day management of the business. The MD was able to ‘showcase’ the effectiveness of the unit, winning political support for the unit’s investment proposals, annual budgets and on-going management autonomy, creating a virtuous cycle– management effectiveness wins trust, trust enables autonomy, autonomy enables effective management. It is this virtuous circle that resulted in eThekwini Water and Sanitation being named the 2014 winner of the Stockholm Industry Water Award, for its transformative and inclusive approach to providing water and sanitation services. Reflections and conclusions Results of the The results of the reforms are clear: 55 percent of the poorest 40 percent of households have water piped turnaround to the premises. Those poor households that do not have water piped to the premises get safe, reliable water from standpipes. Overall, 97 percent of the population have access to piped water, with 80 percent having water piped to the premises. This water is safe to drink and available 24 hours a day. Affordability was enabled by providing 300 liters per day of free water for poor households. How a mandate The national and local political reforms in 1994 established a broad mandate for reform, with a focus on for the reform extending services to the poor. The 1996 Constitution included a right to water. The constitutional and policy was secured mandate was given practical effect through the key reform actors at the local level–the executive mayor, the municipal mayor, and the head of the water and sanitation department. Somewhat unusually, Neil Macleod, a white male, was retained as the head of water (a position he held for over 20 years).11 Early success and effectiveness was likely to have been a key factor in convincing the city manager (a new appointee) and the executive mayor (a politician) of the merits are retaining the existing water manager. Macleod himself said that he ‘ran faster than the others’. Macleod anticipated the forthcoming political changes and acted early to reorient the water service towards a pro-poor focus even before the first democratic elections in 1994. New politicians were eager to be seen to be doing something quickly and it was thus convenient to back a manager who had already demonstrated he was being effective in line with the new political agenda. Phasing the The MD was able to ‘showcase’ the effectiveness of the unit in providing services to poor people early on reforms (even before the formal political changes in 1994), and was thus able to win political support for the unit’s investment proposals, annual budgets, and on-going management autonomy. The turnaround began with institutional reforms to incorporate multiple municipal water departments into a single metropolitan water department within the metropolitan government. The utility moved rapidly to incorporate the various township utilities into its management structure. It focused on expanding and improving service, and constantly innovated institutionally. Some elements of the design were agreed early in the process–such as ring-fencing the utility’s finances and performance pay for the MD. Others, such as the free water allowance and how it would be paid for, were developed “on the hoof”, and adapted in response to customer feedback. In many ways, the institutional structure remained very light–a department within the metropolitan municipality. There is no independent regulator, no long-term performance contract, and no separate legal corporate entity with a Board of Directors. 11 In other cities, senior managers have held their positions for much shorter periods. The City of Tshwane, for example, has had five city managers in 12 years. 69 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix A: Case studies: How Cities Turned Around Water Service for the Poor Sustaining The water department remains within the overall municipal structure and is accountable to the municipal performance manager, the executive mayor, and the elected councilors. The council passes budgets annually and the department must report on its performance and achievements each year. This is an indirect accountability model operating at the local municipal level. Durban Water must also meet national minimum standards regulated by the national finance, local government, and water ministries–indirect accountability operating at the national level. The water unit holds customer focus groups, undertakes surveys of customer satisfaction and manages an effective call center that responds to queries and complaints. These more direct feedback mechanisms complement the indirect accountability processes. eThekwini developed a strong second tier of managers and the transition to a new manager has been apparently seamless. The strong performance culture within the organization could play an important role in sustaining the organization going forward. However, time will tell to what extent the formal structures and organizational culture are sufficient to sustain performance over time and resist predatory pressures being experienced in other state institutions in South Africa.12 12 See, for example, Levy, Brian, Alan Hirsch, and Ingrid Woolard. 2015. “Governance and Inequality: Benchmarking and Interpreting South Africa’s Evolving Political Settlements.” ESID Working Paper No. 51. 70 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix A: Case studies: How Cities Turned Around Water Service for the Poor A.3 Kampala, Uganda Uganda’s capital Kampala has a population of over 1.6 The turnaround was backed by the President of the country million. In 1998, when the city population was closer to as part of a donor-driven program of parastatal reform. 1 million, less than half the population had access to safe A charismatic new MD of the national utility (National water sources. Access for poor families was not tracked Water and Sewerage Corporation, hereafter NWSC) but was certainly lower than this, with most poor families helped drive this reform. He created a management system living in slums and relying on contaminated springs, wells, based on autonomy, accountability, and incentives, which and water bodies. Now, 83 percent of the population has built on the successes achieved in the first six years of the access to relatively convenient and safe piped water.13 More turnaround (1998–2004) through management contracts importantly, this access is almost evenly distributed, with with international firms. His strategy involved investment, 78 percent of poor families having access to piped water. cost-effectiveness, and financial viability as essential Five years after the retirement of the utility leader that elements of extending access to everybody, including the drove those initial reforms and service improvements, the poor. A Pro-Poor Unit, massive expansion in the number commitment to constantly improving performance has of public water points, and introduction of prepaid public also been sustained under the new leadership of the current water points, have all helped in bringing better quality MD, Silver Mugisha. This includes a new “Water for All” water to poor people in Kampala. program aimed at reaching the poor, both in Kampala and in a larger number of small towns that have previously been served by small providers. Service before A precise picture of service for poor people in Kampala in 1998 is not possible from the available data. the reforms For the population as a whole, access to piped water (from private connections and public taps) was 48 percent. Most of those without piped water drew water from boreholes and protected springs. More than half of those with access to piped water got it from a standpipe or kiosk. Data from 2004 (by which time improvements had already started) show that only around 22 percent of households had water piped to the premises. Among the poor, less than 48 percent would have had access to piped water. Around 44 percent of families in Kampala lived in slums, where piped water services were very limited.14 Many of these slums were built on private land, on which the utility was generally not able to offer either individual connections or standpipe service.15 Moreover, the poor families predominantly lived in the low-lying and swampy valleys between Kampala’s hills. In these areas, wells, springs, and water bodies provided alternative sources of water which many poor people relied on for reasons of convenience and cost.16 In fact, when households in Uganda that were not using safe sources were asked why not, 55 percent cited cost as the main reason, while most of the rest cited inadequate sources or the long distance to safe sources.17 Unfortunately, much of this water was contaminated, often as a result of infiltration from latrines.18 Contaminated water sources contributed to cholera outbreaks from late December 1997 to March 1998.19 13 DHS, 2011. 14 This estimate is for 2002, but the number would have been similar four years earlier. The World Bank. 2014. Do Pro-Poor Policies Increase Water Coverage: An analysis of service delivery in Kampala’s Informal Settlements, 12–13. The World Bank. 15 World Bank. 2014. Do Pro-Poor Policies Increase Water Coverage, 13. Fifty-two percent of land in Kampala is owned by the Kingdom of Buganda. Many slums have grown up on these lands. 16 World Bank. 2014. Do Pro-Poor Policies Increase Water Coverage, 14. 17 Uganda Bureau of Statistics, “National Survey of Service Delivery 2004”, Table 5.9. 18 Kayaga, S., J. Fisher, and R. Franceys. 2009. “Improved access to urban water services in Uganda. Proceedings of the ICE: Municipal Engineer” Volume 162, Issue 3 (2009): 165–170. Legros, D., M. McCormick, C. Mugero, M. Skinnider, D.D. Bek’Obita, and S.I. Okware. 2000. “Epidemiology of Cholera Outbreak in Kampala, Uganda.” East Africa 19 Medical Journal Volume 77, Issue 7 (2000): 347–349. 71 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix A: Case studies: How Cities Turned Around Water Service for the Poor Catalyst for The catalyst for reform in Kampala was the relationship between the Museveni administration, international reform financial institutions (IFIs), and political elites in country. In the 1990s President Museveni and his National Resistance Movement (NRM) were supported by donors.20 This combination of local leadership and international support enabled consolidation of service and governance reform after a long spell of dictatorship and civil war. In early 1995, President Museveni committed his government to privatizing 85 percent of Uganda’s public enterprise sector by the end of 1997.21 This commitment helped paved the way for a debt forgiveness deal worth US$2 billion to Uganda, being developed by the International Monetary Fund (IMF) and World Bank.22 Yet at the same time, many in the NRM resisted foreign private ownership of important Ugandan companies. The tension between a desire to please donors by embracing market-oriented economic governance on the one hand–and the desire for national, social control of key entities on the other–shaped the reform path in the urban water sector. Donors recommended private participation in water services to deliver sustainable improvements, noting that “‘Over the last 10 years, the GOU [Government of Uganda] in partnership with the World Bank and other donors has made significant investments (over US$100 million) in the urban water and sewerage sector. These investments have contributed immensely to rehabilitating the existing infrastructure under the NWSC management. Unfortunately, these investments have not been matched with the necessary efficient commercial and financial management capacity that can ensure the delivery of sustainable services in the medium to long-term.’”23 The Government of Uganda agreed to bring in an international company to operate the NWSC under a management contract. At the same time, utility and political leaders in Uganda began a search for a Ugandan, public sector solution to the performance problems in the NWSC. Course of action In 1997 the principles were set: commercialized management of the utility, continued donor investment, decided on and the aim of improving service to everyone. In Kampala a management contract with Gauff, a German company, was signed in 1998. Yet there was no overall blueprint. The management contract was for just three years–an interim step. The next stage of institutional development was left open. Regulatory matters such as setting of tariffs and service standards were left largely to the utility. There was no engineering masterplan that guided service in the city. Much was left open to be solved as new information and challenges emerged. Actors driving the At a political and policy level the actors driving the reforms were President Museveni and several donors. In reforms 1998, a powerful new player entered with the appointment of William Muhairwe as MD of the NWSC. From then on, the strategy in Kampala’s water sector was a product of Muhairwe’s drive and enthusiasm, backed up by President Museveni’s all-powerful support. While donors initially advanced management reform, the model that developed after 2004 was a new approach that combined autonomy incentives and accountability with public sector ownership and a Ugandan management team. 20 Ottaway, Marina, 2000. Africa’s New Leaders: Democracy or State Reconstruction? Washington, DC: Carnegie Endowment for International Peace. 21 Tangri, Robert K. 1999. The Politics of Patronage in Africa: parastatals, privatization and private enterprise, 53. Oxford. 22 International Monetary Fund. “HIPC Debt Relief for Uganda Increased to a Total of US$2 Billion: Additional Relief Vital for Uganda’s Poverty Reduction Programs” International Monetary Fund February 8, 2000. Accessed September 18, 2015. https://www.imf.org/external/np/sec/pr/2000/pr0006.htm 23 World Bank. “Aid Memoiré Document, 1998, Project Evaluation Report to NWSC Management,” cited. Mugisha and Berg. 2008. 72 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix A: Case studies: How Cities Turned Around Water Service for the Poor Technical Water production for Kampala nearly doubled from 2002 to 2015, increasing from 34 million m3/year to solutions 63 million m3/year. In 2006, the Gaba III water treatment plant project was completed–this raised water implemented production for Kampala and nearby areas by 80,000 m3/day. To extend service throughout the city, the NWSC more than doubled the total length of the network, adding around 1,300 km of pipe in Kampala between 2004 and 2013. Network extensions allowed the total number of connections to be more than quadrupled, from around 45,000 in 2002 to over 200,000 in 2015.24 In planning its network extension, the NWSC used geographic information systems (GIS) to identify poor communities and ensure service reached them.25 Figure A.5 shows rates of access to piped water in neighborhoods of Kampala for which sufficient data were available. Those areas surrounded by a red line are where the poorest 40 percent are estimated to live. The map shows that in many poor areas, access to piped water is greater than 90 percent. However, in some areas in the Northeast, access to piped water by the poor appears very low. Figure A.5: Access to piped water by poor, Kampala 24 Connections include public standpipes and all connection types (domestic, industrial, and institutional). 25 World Bank. 2014. Do Pro-Poor Policies Increase Water Coverage, 13. 73 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix A: Case studies: How Cities Turned Around Water Service for the Poor Standpipes serving water kiosks were chosen as the main way of providing water services to low income communities. This kept costs low, compared to making individual connections, and also enabled service to poor people in informal settlements on private land where the NWSC could not offer individual service. The NWSC has more than tripled the number of standpipes in its service area since 2003, bringing the total number to around 7,700 in 2013. Water is available for 18 hours per day on average across the network. Results of water quality tests have been provided in the publicly available Annual Reports since 2010. As of 2013, in the 23 towns the NWSC served at the time, at least 97 percent of samples passed E-coli tests. In the same year, 22 of 23 towns produced water that met the National Standard for turbidity (less than 5 NTU). Commercial In 2004 the NWSC changed its commercial and services delivery strategies to better align them with the solutions government’s objective that 100 percent of the urban population should have safe, sustainable, water implemented services within easy reach.26 Around 7 percent of poor households in Kampala have water piped to their premises.27 To promote such access, the NWSC has operated an Affordable Connections Policy since 2004. Connection fees were reduced from US$75 to US$35. Moreover, the NWSC took on responsibility for making a service connection of up to 50 meters from the utility’s supply point. The NWSC also maintains the service line, up to the meter on the customer’s premises. By taking responsibility for constructing and maintaining the line, the NWSC has not only helped customers but also increased the quality of the service lines, reducing nonrevenue water. The connection subsidy is covered from a 10.7 percent surcharge on most consumption. The tariff for domestic consumption is around US$0.77/m3. Standpipes and shared taps are the main source of water for poor people in Kampala, with 71 percent of poor families relying on them in 2011.28 These shared taps have provided a huge increase in access to safe water, compared with what prevailed in 1998. Most of these taps serve kiosks operated by someone who has paid for the connection and then on-sells the water. Some kiosks are municipal, some are run by community groups, others are private. The NWSC charges $0.47/m3 for water dispensed at standpipes. This is below the domestic tariff. The subsidy is justified on the basis that standpipes are mostly used by poor people. However, some of the kiosks charge a significant mark-up for the water. Once the water is dispensed in 20 liter jerrycans at the kiosk, the effective tariff can be US$1/m3or more. To make access to shared taps more widespread and more affordable, the NWSC innovated by:  Removing the monthly fixed charge for some standpipes in poor areas. This reduced the cost of water from these taps by about 10 percent. This change was made after the President pledged it during the 2006 election campaign.  Recognizing that some individual yard taps are, in fact, shared water points. The NWSC monitors consumption of water per connection in low income areas. When consumption is unusually high, staff from the NWSC’s Pro-Poor Branch visit. If it is clear that the yard tap is serving more than two or three households, then the tariff for public water points is applied.  Using prepaid public water points. These are standpipes which automatically dispense water when a customer inserts an electronic token. By cutting out the middleman (the kiosk operator) these prepaid water points ensure that poor customers can access water at the tariff set by the NWSC, avoiding mark-ups. 26 The World Bank. 2014. Do Pro-Poor Policies Increase Water Coverage, 21. Material in this subsection is derived from that publication unless otherwise noted. 27 DHS, 2011. 28 DHS, 2011. 74 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix A: Case studies: How Cities Turned Around Water Service for the Poor Financing The infrastructure and service improvements described above required about US$100 million in capital solutions expenditure between 2002 and 2011.29 Only 28 percent of this was grant-financed. About half was financed implemented from operating cash flows. In 2010, a commercial loan of US$2 million was obtained for financing the extension of the Gaba intake plant, which supplies water to Kampala city and the surrounding areas. This loan is being serviced from operating cash flow. When the reforms started, the NWSC had debts to the government that it could not service. Most of the indebtedness resulted from loans from IFIs to the Government of Uganda that had been on-lent to the utility. The government agreed to a moratorium on debt service for a period, which gave the NWSC financial breathing space. Then, in 2007, the government converted the outstanding balance of US$47 million into equity–effectively forgiving the debt. Achieving financial self-sufficiency was core to the service improvement strategy from the start. The NWSC’s ability to earn a surplus over operating costs has enabled expansion of service to poor neighborhoods to proceed independently of availability of funding from the government budget. As Figure A.6 illustrates, the NWSC has had an operating surplus of at least 18 percent of operating costs since 2002 (bottom right panel). Figure A.6 shows a positive continuation of these trends in these developments, and a new drive is in progress to address NRW with fresh dedication. Figure A.6: Achieving financial self-sufficiency at NWSC Note: Figures are for the NWSC as a whole, not just its Kampala service territory. Source: Utility research (see Table D.2 for full list of sources). 29 Figure quoted in 2011 US dollars. Data are from NWSC financial statements and annual reports. 75 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix A: Case studies: How Cities Turned Around Water Service for the Poor To achieve this surplus, the NWSC has driven NRW down from 50 percent in 1998, with its continued efforts helping getting it to around 32 percent in 2015. Collection efficiency has also improved considerably, from 85 percent in 2001 to over 95 percent since 2010. Labor productivity has skyrocketed from 36 staff per 1,000 connections to just six staff per 1,000 connections now. Until 2002 this improvement was achieved by reducing staff numbers, which brought the ratio down to 10. Since then, productivity gains have been achieved by adding connections quickly, while holding increases in staff numbers to a lower rate. Adequate tariffs have been equally important to the NWSC’s ability to finance expansion of service to the poor. To protect real revenue from the effects of inflation, the NWSC’s tariff has automatically increased with inflation since 2004.30 Cross-subsidies are important in the NWSC’s strategy to combine financial stability with affordability. The average water and sewer tariff of US$1.06 – adequate for cost recovery – is higher than the domestic tariff of US$0.77, which is in turn is higher than and the standpipe tariff of US$0.47. Overall cost recovery is achieved by charging large users more than domestic customers. However, to keep large users on the system–and so contributing to revenue–the tariff rate for commercial consumption above 1,500 m3 per month is lower than for consumption below that rate (US$0.93/m3 compared with US$1.16/m3). In 2015, eligibility for subsidized connections was restricted to only poor customers. By charging better off customers the full cost of a connection, the NWSC hopes to increase the cash available to expand service to poor customers.31 Institutional and A new law in 1995 corporatized the NWSC, giving it a fairly standard set of corporate powers and governance responsibilities. The corporation is governed by a Board of Directors appointed by the Minister responsible solutions for water. In 1998 a new Board and a reforming MD, Muhairwe, were appointed. Muhairwe replaced Hillary Onek (who later became Minister of Internal Affairs) as MD “at a time when the company was almost collapsing due to poor management, poor services and debt”.32 Shortly before Muhairwe and the Board were appointed, Gauff, an international engineering firm, had been brought in on a three-year contract to run the NWSC’s Kampala operations (more than 70 percent of the company). After a gap of a couple of years, another management contract was awarded, this time to ONDEO. In parallel with these ‘external’ management contracts, Muhairwe developed a series of internal management contracts. The first of these were ‘area performance contracts’ which applied to the areas outside Kampala not managed by the external management contractors. The NWSC managers for these areas were given performance targets, and bonuses for meeting the targets. Later, a similar concept called ‘Internally Delegated Management Contracts’ was applied to the management of the NWSC’s Kampala service area also. In 2003 a Performance Contract was agreed between the government and NWSC. The Performance Contract set out targets for the utility, including a requirement to develop plans, and funding for network expansion. The contract was essentially a corporate strategic plan with quantifiable targets and milestones, setting out the agreed way forward.33 By 2004 then, the NWSC had forged for itself a distinctive governance structure. Contracts agreed with the national government set out service and efficiency indicators. The utility was financially and managerially autonomous. Performance targets and managerial autonomy were transmitted down the organization from the MD to unit managers using internal management contracts. Incentives for good performance were provided through worthwhile bonuses. 30 The Water Act (General Rates) Instrument, 2006. 31 Mugisha, Silver. 2015. “Water for all: Delivering the promise.” National Water and Sewerage Corporation. Matsiko, Haggai. 2011. “Behind the Scenes at Muhairwe’s Exit.” The Independent (Uganda). http://independent.co.ug/business/business-news/4897-behind-the-scenes-at- 32 muhairwes-exit 33 Performance Contract between the Government of the Republic of Uganda and the National Water and Sewerage Corporation, October 17, 2003. 76 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix A: Case studies: How Cities Turned Around Water Service for the Poor The appointment of quality technocratic leadership in the senior management of the NWSC was instrumental in enabling reforms to be sustained and further reforms introduced. It has also helped securing political autonomy and instituting rigorous performance metrics. After Muhairwe retired, his place was taken by Dr. Eng. Silver Mugisha, a member of the management team developed during these earlier transformative reforms. There has also been willingness to try new approaches and renew priorities. In the three years or so since his appointment, Mugisha has been quite assertive in driving a firmer agenda on the NWSC’s role in smaller towns than was the case under the previous MD. He has also placed a renewed emphasis on revenue collection, which–according the utility’s data–has doubled. There has been a renewed emphasis on NRW reduction activities, and notably a new drive for increased community and other key stakeholder involvement, customer outreach, and responsiveness. This has included upgrading the call center and digital (social media) systems. Management During the period of the external management contracts, substantial progress was made in reviving the solutions NWSC as an effective service provider. NRW fell from 50 percent in 1998 to 43 percent by 2001 while implemented connections rose from 51,000 to 66,000. GIS and other IT systems were improved. In parallel, Muhairwe focused on building a performance culture within the organization. An initial 100- Days program crash program (February 1999 to May 1999) set demanding short-term targets which the management team stormed toward, shocking the system out of its lethargy. Other initiatives followed, including consultative strategic planning to build a shared sense of purpose among all staff.34 In 2006 a Pro-Poor Unit was created to bring a focus on extending service to poor areas. This unit, with seven multidisciplinary staff, takes up issues in effectively serving the poor with the commercial and operational branches of the Corporation.35 Under the current MD, Mugisha, the cultivation of a professional staff and an institutional culture of efficiency and accountability remains firmly on the organization’s agenda. With government support for the NWSC’s new “Water for All” program, this has meant an improvement in getting government institutions to meet their own debt obligations to the utility. Reflections and conclusions Results of the Thanks to the NWSC’s turnaround, 78 percent of poor households in Kampala now have access to piped turnaround water services. The cost of water has come down for many poor people. Across the network, water is available on average for 18 hours per day. All operating costs, and much of the capital costs of expansion to poor customers, are covered by resources generated by the utility itself. This resource generation is made possible by the NWSC’s cost effective operations, coupled with a tariff that allows for cost recovery overall. How a mandate Institutional reforms to create an effective utility were recommended by donors, and supported by a for the reform President with a secure political position. However, the path the reforms took was not that envisaged by was secured the donors. Rather, it was a homegrown model crafted by a savvy utility MD. This model transcended the dilemma of ‘efficiency or local control’ by adapting models designed to bring efficiency through foreign management control to work with local, public sector control. In this way it appealed to local political elites while also being acceptable to international donors. The model has delivered sustained success in expanding access for poor people for over more than 15 years. 34 Mugisha, Silver, Sanford V. Berg, and William T. Muhairwe. 2006. “Using Internal Incentive Contracts to Improve Water Utility Performance: The Case of Uganda’s NWSC,” 3. 35 World Bank. 2014. Do Pro-Poor Policies Increase Water Coverage, 21. Material in this sub-section is derived from that publication unless otherwise noted, 26. 77 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix A: Case studies: How Cities Turned Around Water Service for the Poor Phasing the The reforms were fluid and adaptive. The initial concern was to deliver water to existing customers while reforms staunching financial losses. Competing models of local and international management were tried in parallel. Pragmatic solutions including tariff indexation, financial restructuring, and a performance contract rather than a regulator, were developed. Much attention was given to the management aspects of creating a successful utility, from the original 100-day program to eventual construction of a training center to boost skills systematically. After the utility’s finances and management were stabilized in 2004, attention shifted to improving service to the poor, with creation of a Pro-Poor Unit, and development of commercial and technical solutions to improve access for poor households. Sustaining As the NWSC has become successful, generating increased revenue, positive cash flows, and deploying performance a large capital expenditure program, the temptation for predation must increase. The utility management team told us success in serving the poor breeds its own protection, since “In Kampala, the poor vote.” The utility also cites the barazas (community meetings) as ensuring it remains accountable while building community constituencies of support. A strong culture, coupled with training and management succession planning, support continued professionalism. The formal mechanism of automatic tariff indexation helps to protect against populist short-termism, preserving the utility’s ability to continue to fund expansion of service to the poor. 78 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix A: Case studies: How Cities Turned Around Water Service for the Poor A.4 Nyeri, Kenya In Nyeri–a city of almost 150,000 people in Central In 1995 the Nyeri Council, led by Mayor Jackson Wanjage, Kenya–88 percent of poor households have water piped decided the town deserved better. The Municipal Engineer to their premises by the municipal utility (NYEWASCO). (Joseph Nguiguti) provided the technical expertise. Two percent use public taps, while 7 percent rely on other Together, they persuaded a donor to let them join an existing improved sources, giving a total of 96 percent access to program in Kenya supporting the creation of municipal- improved water services.36 owned public companies with autonomous management and a commercial orientation. The new company faced For the 90 percent of poor households using piped water, two immediate problems–inadequate supply and high water is available 24 hours a day, and is safe to drink. In losses. While in the process of negotiating a loan for new fact, Nyerians travelling to Nairobi and other towns in investments to fix the supply problem, the new company Kenya may carry containers of NYEWASCO water for focused on improving cash flows and reducing losses. their journey, so great is the trust in the utility. Things really started to turn around in the early 2000s when It was not always like this. In the early 1990s, water NYEWASCO secured a loan from the KfW to finance was provided by the then municipality. It was rationed, new production capacity. Thanks to sound technical and unreliable, and unsafe to drink. The network did not serve commercial management, NYEWASCO has been able to the growing informal settlements built on steep slopes extend the network and service the loan through improved around the urban periphery. cash flows, made possible by providing a better service. Social tariffs ensure water is affordable. Kiosks and meter 36 Baseline: “State of the City” survey (Kenya), 2012–2013. banks serve people in informal settlements where laying conventional distribution networks is not feasible. Service before Reliable data on the situation before 1995 are not available. Those who lived through the period recall that the reforms infrastructure investment had not kept up with population growth. Water was rationed by limiting supply to a few hours per day–and this water was not safe to drink. Informal settlements were not served at all. Catalyst for There was no donor conditionality, technical assistance or national government intervention to catalyze reform reform. Rather, the then local government became convinced that the poor service was not in keeping with the town’s needs, and that it must be possible to do better. Actors driving the The political impetus for reform came from the visionary and charismatic Mayor, Jackson Wanjage, who reforms had political support from the municipal council. Technical and managerial input came from a capable and motivated Municipal Engineer, Nguiguti, who at the time was responsible for water supply because of his role in running all of the engineering services in the municipality. Course of action In 1995 the town administration united around the objective of providing safe reliable water to all citizens. decided on Lacking money and utility management expertise, the decision was made to seek external support. The mayor and municipal engineer successfully lobbied to be included in a German Technical Cooperation (GTZ) program assisting with water utility commercialization. This helped develop the next stage of the reform vision–to build an efficient utility, and raise finance to expand production and distribution. Technical The number one problem was that water production was not adequate to supply the needs of the growing solutions population. A loan from the KfW was negotiated to finance a new bulk water supply scheme. However, implemented donor sanctions during the Moi administration meant it could not be disbursed. NYEWASCO’s own engineering team improvised ways to boost production from the existing infrastructure above design capacity, increasing supply from 6 MLD to 8 MLD. The team also started to get leakage under control. More production and less leakage meant that hours of supply could be increased. 79 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix A: Case studies: How Cities Turned Around Water Service for the Poor In 2003 the KfW loan was finally disbursed, allowing NYEWASCO to construct a new 21 MLD capacity plant. The new water source brought an end to water rationing. It also reduced pumping and treatment costs because the new facility was located above the town, near the water source. With the new facility on line, NYEWASCO was able to expand its distribution network, taking advantage of a grace period in loan repayment and increased cash flows arising form a better service. Between 2006 and 2015, 752 km of pipe were added to the network, more than quadrupling total network length. This allowed more than 13,000 additional households to be connected. During the same period, the number of public water kiosks was increased from 3 to 16. Networks were extended to all parts of the town, both rich and poor. In informal communities, after the installation of public kiosks, households were offered private connections. Where conventional service pipes could not be laid for technical reasons, meter banks were installed. NYEWASCO then assisted households in making flexible pipe connections from the meter bank to their home. Commercial To assist poor households in connecting, Nyeri has kept the upfront costs of connection low. Until recently, solutions households had to pay KShs 3,100 (about US$35), of which KShs 2,000 is a deposit. Recognizing that even implemented this amount is a barrier, NYEWASCO recently reduced the deposit by 500 KShs. The tariff is rising block. Consumption up to 10 m3/month is KShs 32/m3 (US$0.36/m3). A poor household of six people consuming 50 liters each per day would have to pay KShs 284 per month (US$3.23). Industrial and commercial customers, and residential customers consuming in the higher blocks, pay cost- reflective tariffs. Businesses such as tourist lodges pay tariffs six times higher than people in slums. These cross-subsidies allow the utility to finance universal service. The average water tariff across all customer classes is KShs 58/m3 (US$0.65/m3). All customers are metered. NYEWASCO uses mobile technology to help poor customers manage their cash flow by paying in small and irregular increments, as water is consumed or as cash is available. Customers can read their meter at any time and use a mobile phone to communicate the reading to the utility billing system. The customer can then pay using MPESA, Kenya’s widely used mobile money system. Financing When Nyeri embarked on its turnaround strategy, it had no funds to invest in infrastructure. After embarking solutions on the GTZ-supported commercialization strategy, NYEWASCO was able to secure a loan from the KfW. implemented The loan was facilitated by the Government of Kenya, which borrowed in Euros, and on-lent to NYEWASCO as a KShs 1.1 billion (US$18 million) loan. The other loan terms were an interest rate of 2.5 percent, a tenor of 30 years, and a grace period of eight years. This loan financed the new production facility and some initial network expansion, as well as work on rehabilitation and equipment for leak detection and repair. Apart from the KfW loan, other sources of finance were cash generated from operating activities (all of which went on debt service or was plowed back into the business), and a total of US$0.7 million in grants from Kenya’s Water Services Trust fund which assisted with expanding distribution. Financing sources over time are shown in Figure A.7. NYEWASCO could not have secured this loan if it had not made possible by continued collection of long over-due convinced the KfW of its ability to operate cost-effectively arrears, as well as rigorous collection of current bills. To and to service the debt. To actually service the debt, the collect from government customers, NYEWASCO makes a utility needed to generate enough cash from operations point of knowing when government customers will receive to cover its debt service obligations once the grace period cash from the budget, and follows up on payment at that ended in 2009. As Figure A.8 shows, this was achieved. time. To ensure collections from the police compound, the utility strategically re-laid pipes so that the compound To ensure that it could service the debt, NYEWASCO could be cut off for nonpayment from within the utility focused on improving operating efficiency. By 2006 it had headquarters. This innovative approach stopped the already got its collection rate up to 98 percent. From then harassment of utility staff that had previously prevented it on the collection rate averaged more than 100 percent– from enforcing payment. 80 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix A: Case studies: How Cities Turned Around Water Service for the Poor Figure A.7: Sources and amounts of capital expenditure financing Note: The 2008 financial year ran from January 2007 to June 2008. Prior to 2007, NYEWASCO’s financial year ran from January to December. For this reason, 2007 does not appear in the graph above. Source: NYEWASCO Financial Statements. Figure A.8: Cash from operations compared to debt service payments Source: NYEWASCO Financial Statements. 81 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix A: Case studies: How Cities Turned Around Water Service for the Poor Figure A.9: Nonrevenue water, staff productivity (NYEWASCO) Source: NYEWASCO. As Figure A.9 shows, NRW was 42 percent in 2006, and These increases in operating efficiency have allowed then went up as new supply came on line. NRW reached a NYEWASCO to consistently achieve an operating cost high of 48 percent before NYEWASCO was able to get it coverage ratio of between 1.26 and 1.54 every year since under control and drive it down to its current levels of 18 2009, even as average tariffs declined in real terms (see percent. During the same period, staff productivity tripled, Figure A.10). It should be noted that NYEWASCO is because the utility was able to hold staff numbers roughly concerned that the declining real tariff could threaten its constant while tripling customer numbers. ability to finance asset replacement in the future. Figure A.10: Cost recovery and real average water tariff over time Source: NYEWASCO. 82 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix A: Case studies: How Cities Turned Around Water Service for the Poor Management To drive the efficiency gains that allowed universal service to be financed, NYEWASCO adopted modern solutions management techniques. All staff are involved in strategic planning. Computerized systems are used. The implemented systems are transparent and audited, reducing risk of malfeasance. Staff are provided with technical and managerial training. All of this has built a strong corporate culture which now resists predation or malfeasance. Institutional and At the start of the commercialization program, the utility–which had been a department in the municipal governance government–was corporatized, becoming NYEWASCO (a wholly municipal-owned company established solutions in terms of private company law in Kenya). Utility staffing, accounts, and finances are ring-fenced. Kenya promulgated a new Constitution in 2010, and since 2013, water services have been a devolved function. This means that Nyeri Water and Sewerage Company is currently owned by the County Government of Nyeri. However, since the national Water Act of 2002 has not yet been repealed, NYEWASCO still has a Service Provision Agreement (SPA) with the nationally owned asset holding Tana Water Services Board. NYEWASCO reports and accounts together with the County Government of Nyeri to the national regulator, WASREB, on an annual basis. The utility is therefore accountable to the county government, but also to WASREB, Kenya’s national water regulator. WASREB sets tariffs and minimum services standards through a process divorced from local politics. WASREB also monitors and publishes key performance indicators for NYEWASCO and all other Kenyan utilities, and it oversees and enforces minimum good governance processes related to the selection of board members, their security of tenure, and related matters. The continuity of the management team is a notable feature, with the capable MD retaining his position from the commencement of the company until 2014. After a prolonged interim arrangement, a new MD took office in 2015. Reflections and conclusions Results of the The results of the reform are clear: near-universal access to piped water (94 percent), 96 percent of poor turnaround households with access to improved water services, 88 percent of poor households with water piped to their home, water that is safe to drink, and availability of 24 hours a day. All this was achieved with almost no financial support from the local or national government. The utility’s operating efficiency is among the best in Africa, which enabled it to finance near universal service while preserving affordable tariffs and minimizing subsidies. How a mandate Civic pride, a strong political leader (the mayor), and a dedicated and capable utility engineer were the for the reform key ingredients that came together to put Nyeri on the path toward universal high-quality water service. was secured From a political economy perspective, it probably helped that the utility at the time was small and starved of funds. Its small size and weak financial standing likely minimized vested interests in continuation of the status quo–low coverage and intermittent service–for patronage, corruption, and other forms of predation. In other words, it seems the political benefits of improved service outweighed the political costs from loss of rent-seeking opportunities. It may also have helped that Nyeri is a relatively ethnically and culturally homogenous town, minimizing inter-group rivalries that could otherwise have interfered with a drive for universal service and the use of cross-subsidization to achieve it. Phasing the The turnaround began with managerial reforms to increase efficiency. Governance reforms to create a ring- reforms fenced utility followed. Throughout, the utility’s goal was to raise the finance needed to improve and extend service. However, securing the finance took several years, and disbursement was delayed because of a period of poor relations between Kenya and IFIs. In the interim, the utility improvised low-budget solutions to improve service. These service improvements, together with a policy of avoiding layoffs and building corporate culture, secured continuing stakeholder support for the turnaround plan. Once finance was available, the first priority was to bring a big new production plant on line. This immediately improved service to existing customers. Leakage went up initially, but that did not matter in the short term: ample water was available, and pumping and treatment costs had been reduced. The next priorities were to get leakage under control, and to extend the network. In parallel, NYEWASCO knew it was urgent to increase operating cash flows. The grace period on the loan created just enough time in which to do this. 83 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix A: Case studies: How Cities Turned Around Water Service for the Poor Sustaining NYEWASCO’s success in growing, securing finance, and generating positive cash flow now make it a performance tempting target for predation by politicians and corruption from within. The utility protects itself from predation by building strong community support. It not only provides good service, but also has highly visible commitments which create credibility with the community, such as fixing pipe bursts in less than 24 hours. NYEWASCO is transparent, holding regular barazas (similar to town hall meetings) and an annual open day in which customers are invited to tour its offices and facilities. As a result, customers–who are also voters–are fiercely protective of NYEWASCO and its autonomy. For instance, when the local radio station reported that a political attempt to interfere in NYEWASCO’s management could be afoot, many citizens called the utility asking what they needed to do to stop such interference. At NYEWASCO, an internal culture of performance, transparency, and accountability creates an esprit de corps that also militates against corruption and predation. 84 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix A: Case studies: How Cities Turned Around Water Service for the Poor A.5 Ouagadougou, Burkina Faso A severe water shortage and financial difficulties in the well-calibrated financial model, agreed with a stakeholder 1990s led to an initiative to reform the utility responsible body that is formally engaged in the process, and used to for providing water to Ouagadougou in Burkina Faso. In inform both planning and tariffs, has been an important the context of a structural adjustment program for the contributing factor. country, the financiers proposed that the government lease ONEA’s operations to a private operator as part ONEA has a mandate to serve formal areas only; however, of a broader set of financial conditions. An alternative 25 percent of the 2.5 million people in Ouagadougou live was negotiated, namely, to establish a corporatized in informal settlements (Figure A.11). The expansion of government-owned company operating under public Ouagadougou’s water supply system enabled ONEA to management. A private consortium was contracted to expand services to informal areas even though it did not have provide commercial and financial management support a formal mandate to do so. The management of the network services to the public utility in the period 2001 to 2006. going into the informal settlements is delegated to private Improvements in service coverage and performance have operators who construct the network and connections and been both dramatic and sustained over a period of 15 years. sell water to customers. This is resulting in much improved Starting initially with an investment in water production services in these areas with many households receiving a capacity, the utility subsequently focused on expanding connection at their premises. the distribution network, resulting in an increase in the number of connections from 42,000 in 1996 to 330,000 in 2014, a growth of 11 percent per year. As a result, 90 percent of the poorest 40 percent of households living in Ouagadougou (and 94 percent of the total population) now have access to a piped water service. Only half of the total population had access to individual or communal piped water service before 2000. These achievements are particularly noteworthy in a country with a per capita GDP of US$713 (2014) and available freshwater resources of only 711 m3 per person per annum (2014). After an initial injection of equity, grants, and loans, the expansion of services was achieved through net positive cash flows. These were in turn enabled by a combination of improved management and efficiency and increased tariffs. These improvements were reflected in increased revenue collection (from 85 percent to 97 percent), greater staff productivity (from over eight to below three), and maintaining NRW at 18 percent. ONEA has maintained financial viability with net cash flows (cash revenues less cash expenditure for operations) comfortably exceeding the debt servicing requirements. A (Photo Credit: ONEA, Ouagadougou, Burkina Faso). 85 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix A: Case studies: How Cities Turned Around Water Service for the Poor Service before The population of Ouagadougou had doubled in the period 1985 to 2000 (to about 1 million people), but the reforms the development of water services infrastructure had not kept pace with increasing demand. In 2000, only half of the population had access to piped water (through individual taps or communal standpipes) and 60 percent of the service areas experienced severe water shortages during the three hottest months of the year. ONEA had 29,000 customers in Ouagadougou. Staff productivity was low at 10 staff per 1,000 connections. Cash collections were also below par at 85 percent from private customers and 65 percent from government customers. Up until 1996, ONEA was mired in serious management and administrative problems–it had stopped producing financial statements and, until 1998, its auditors were unable to verify its accounts. Water was sold below the cost of production and distribution, and the service provider was in serious financial difficulty. Finances for the expansion of services were not available.37 Catalyst for The reforms arose as a result of a deepening crisis. There was not enough water to provide to residents reform in Ouagadougou and severe shortages were a regular feature, especially in the hot summer months. The government did not have the financial resources to remedy this situation and turned to donors for assistance. Burkina Faso entered into a structural adjustment program in 1990 to address its dire economic and financial situation. Engagement between the government and donors throughout the 1990s ultimately led to an investment project, conditional on associated reforms of the utility commencing in 2001. Actors driving the Salif Diallo, Minister of Environment and Water (1995 to 1999), negotiated the water sector reforms with reforms the World Bank and other donors. Mamadou Lamine Kouate was the Director-General (MD) of ONEA over the period 1995 to 2005 and led the reforms of the utility. Both resisted the World Bank’s recommendation to introduce a private operator to manage the service, along the lines of the successful reform which had been implemented in Senegal. This tension produced a different solution–ONEA remained a government- owned, limited liability company. Through a performance-based service contract, deputy managers for the commercial and finance functions were brought on from Veolia. A Supervision Committee–comprising representatives of consumers, government, nongovernmental organizations  (NGOs), and donors–was established to oversee ONEA’s and the government’s performance against the Contract Plan. Kouate proved to be a capable leader and manager of the utility reforms (as evidence in the impressive results). The undertakings made by Diallo on behalf of the government–including, for example, a government commitment to pay for water–proved to be credible. Course of action Investments in increasing the supply of water to Ouagadougou were urgently needed. Funds were committed decided on in 1996 by 11 donors, including the World Bank who contributed US$70 million of the more than US$200 million water infrastructure and reform project. Conditions related to the funding were negotiated with the Government of Burkina Faso. Technical assistance from a private operator, Veolia, was provided through a service contract in the period 2001 to 2006 coinciding with investments in infrastructure, to run ONEA’s commercial, accounting, and financial operations. Technical Investments were made in additional supply capacity for Ouagadougou with the Ziga water supply project solutions (including a new 200 million m3 dam) and in extending the network–171 km of secondary network and 1,437 implemented km of tertiary network were constructed. A key challenge was to provide piped water services to people living in informal settlements (where some 25 percent of the city population lived). Stakeholders agreed on a new vision of water supply to informal areas in 2004. Pilot projects were conducted in five suburbs of Ouagadougou from 2009. This involved the construction of the network and supply of water to a meter at the edge of the pilot informal settlements. The management of the network going into the informal settlements was delegated to private operators, as was the construction of the network and connections (using flexible HDPE piping and meters) and the retail to customers. Private household connections are an important part of the service mix as this enables the private operator to sell more water and hence increase its own revenue. Water is sold on the same terms as are provided by ONEA. Connection materials are provided by ONEA and the operators’ installation costs are reimbursed to encourage them to make connections. By the end of 2014, there were 7,578 connections 37 World Bank. 2001. Project Appraisal Document, Ouagadougou Water Supply Project. 86 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix A: Case studies: How Cities Turned Around Water Service for the Poor in the five informal settlements of Ouagadougou and the program is being extended to other informal areas. ONEA is the lead agency in the operation and an inclusive stakeholder consultation framework has been established. ONEA monitors water quality using a central laboratory with modern facilities and regional laboratories. The National Public Health Laboratory under the Ministry of Health carries out daily checks and its results are included in the calculation of the indicators and in the water quality reports. Figure A.12: Delegation of water supply to private operators in informal areas Source: ONEA. Commercial Connection fees were reduced and are subsidized for both ONEA’s direct customers as well as customers solutions getting water from private operators in informal areas. ONEA sells water to the private operators at a implemented wholesale price which allows operators to apply a margin while keeping the retail price at the same level as ONEA’s retail price in other service areas. The financial model ensures that ONEA maintains a financial equilibrium with free cash flow to repay capital loans and invest in the network. ONEA’s own direct customers benefit from a social tariff for the first 8 m3 per month. Apart from these social tariff arrangements, the tariff is set to recover costs on average. ONEA’s tariff for the first consumption block is just 18 percent of the tariff for consumption in excess of 30 m3. Financing An investment of more than US$200 million was made by 11 development banks and donors (primarily for solutions a new dam, transmission and distribution pipelines, and related infrastructure). The World Bank contributed implemented US$70 million through an IDA loan to the government, of which US$42 million was transferred to ONEA in the form of an equity contribution, and the remaining US$28 million on-lent to ONEA with a maturity of 20 years, including 10 years of grace period for the principal, and at an annual interest rate of 5.4 percent. The introduction of the financial equilibrium model has been a key factor in the achievement of a financially sustainable provider and sector. The model is used to set the annual tariff through a transparent process agreed with stakeholders. Ongoing network expansion has been financed through free cash flow. 87 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix A: Case studies: How Cities Turned Around Water Service for the Poor Management The Director-General had a tenure of 11 years, and along with other senior staff of long standing, provided solutions stability in the management of the utility. ONEA’s accounts were certified by an international audit firm implemented for the very first time in 1998. The first strategic plan was developed for the period 2004 to 2008. ISO certification was achieved in 2009. NRW is well managed and has been in the range 17 percent to 19 percent over the last 10 years. ONEA has consistently achieved high cash collections of more than 95 percent since 2006, including from government. (Government is obliged by the performance contract to settle its water bills.) Institutional and The Minister of Water issued a mandate to the Director General of ONEA to establish the Ziga Project Unit. governance In 1997, the Director General of ONEA delegated powers to the Ziga Project Unit, granting it administrative solutions and financial autonomy. In view of the close ties between the project (with a large infrastructure focus) and the implementation of sector reform, the head of the unit played an increasingly important role and enjoyed excellent relations with the Director General of ONEA. The utility reform component of the project was designed with a focus on the finances, with three key elements: instituting a performance contract between government and the utility, with audits; the development and institutionalization of a multiyear financial equilibrium model (ensuring the sector was adequately funded); and the creation of a path of financial recovery with improved cost controls and efficiency and tariffs that recovered costs. A private operator was contracted to assist with the financial and commercial aspects of the report and played a key part in all three. Financial equilibrium was achieved a few years into the reforms. The audited performance contract between the government and the utility was a key innovation. An international technical auditor was used to audit ONEA’s performance and the service contract for the first time in 1999. The audited performance is published and is used in the stakeholder processes, together with the financial equilibrium model, to agree on targets and to set tariffs. A Supervision Committee was also established to supervise ONEA’s activities. This Committee includes representatives from the government, technical and financial partners, municipalities, and consumers. At its annual meetings, it examines the implementation of the investment program for the attainment of the agreed service targets and ONEA’s management and performance report. The Committee makes recommendations on the sector’s financial stability, particularly whether price adjustments are needed, based on the analysis of the technical auditor who audits all the reports submitted to the Committee in advance. 88 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix A: Case studies: How Cities Turned Around Water Service for the Poor Reflections and conclusions Results of the The results of the reforms are clear: 90 percent of the poorest 40 percent of households living in Ouagadougou turnaround (and 94 percent of the total population) now have access to a piped water service whereas only half of the total population had access to a piped service before 2000. The network length increased by a factor of 2.5 times over the period 2005 to 2014. The number of connections per 100 people in the service were more than doubled. The water is safe to drink. Service reliability has improved and a 24x7 supply has been achieved in the period 2004 to 2008 and 2011 to 2012.38 Connections were made affordable through a social connection fee policy, and more poor people living in informal settlements are also being served with affordable piped water at or near their houses each year. A social tariff keeps the first 8 m3 of water provided affordable for on-premises connections. Tariffs are cost reflective and the utility is efficiently managed to support and sustain the above. How a mandate The reform mandate arose from a crisis in the water supply coinciding with a financial crisis, both for the for the reform country and the utility. The country needed financial assistance and negotiated access to finance and was secured a related set of conditions with international development partners (the World Bank and others), which included the reform of the utility. The mandate was given effect by two key actors, the Minister responsible for water and the Manager in charge of the utility, with the support for the financiers and a private consortium (for financial management support). This was not a straightforward process and took several years. There were disagreements between the World Bank and the government on what constituted necessary reforms. The World Bank wanted to implement a lease contract with a private operator. The government wanted to retain public ownership and management. A compromise was reached with the use of a private operator to run the financial and commercial functions of the utility for a period of time. The decision to sponsor the subsidized connection program together with an effective communications strategy facilitated buy-in by customers and other stakeholders to the reforms. Phasing the Initial attention was given to the financing of an expanded water supply and then expanding the water reforms network. Financial management support was provided by a private consortium that was competitively procured. Improved efficiencies softened the necessary tariff increases. An audited contract between the government and the utility clarified accountability and was used to set appropriate performance targets. The achievement of financial equilibrium was phased in. A capable manager was in place throughout a key phase of the reform period–1995 to 2005–and a strong management team built so that the succession was smooth. Later in the reform process (2009), the utility innovated to expand water supply to the rapidly growing informal settlements using the expanded supply and the extended network as the necessary backbone to achieve this. It developed alternative institutional arrangements, making use of small private operators, to overcome the problem of its limited formal and legal mandate. Sustaining The reforms are remarkable in the extent to which they have been sustained, with continued expansion of performance service to serve the poor, supported by ongoing excellent utility performance and financed by the utility’s own cash flows in a country with limited economic means, all over a period of close to 20 years since the onset of the reforms. ONEA has managed to develop and sustain a culture of performance within the organization that is supported by management autonomy and a formal performance contract with government, that is, in turn, overseen by a formalized stakeholder body. These three ingredients–strong performance culture, formal rules of accountability, and a formalized alliance of stakeholders–all appear to reinforce each other to sustain and extend the achievements of the reforms. 38 Since then reliability has declined to 21 hours in 2015 due to supply limitations. Supply constraints are being addressed through new investments and the situation is anticipated to be alleviated in 2017. 89 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix B: Suggestions for Further Research Appendix B: Suggestions for Further Research The important deficiencies in data available, and the Better yet, if spatially disaggregated household survey limitations of the methodology used, mean that additional data were widely available–and utilities employed work on this topic would be valuable. This work should better GIS mapping capabilities–the extent to which add detail to the messages established in this report, test network extent drives service to the poor could be better their limits, and add new ideas beyond those which can be understood. This knowledge could help utilities with gleaned from the cases studied for this report. citywide investment planning. Better data are needed Additional cases should be studied To better understand access and service levels for the poor, Five case studies have illuminated many aspects about much more household survey data are needed. Household how cities serve the poor relatively well in their particular survey data need to be available across many more cities, contexts. Additional cases would add to this rich body of and need to cover more variables–such as reliability, knowledge. In particular, political economy analysis for adequacy, and affordability. This robust dataset would allow cities that do not serve the poor well could uncover what for the use of econometric analysis–like that applied by the drives bad service to the poor, and how that contrasts with forthcoming Performance of Water and Wastewater Utilities the political economy dynamics in cities with relatively in Africa study–to examine the extent to which utility good service. management effectiveness and cost recovery drive service to the poor across the region as a whole. Simply studying more cities that are relatively successful in serving the poor would uncover more useful techniques that Utilities will need to produce and publish better data could be considered for use elsewhere. It is quite possible as well. Audited, unqualified accounts will give more that in a wider range of cities, there would be some that accurate cost recovery figures. Universal metering will provide good service to the poor largely through small scale result in more accurate NRW levels. On the service side, providers, or through utilities that were not managed well. more accurate estimates of coverage rates are needed, as Knowledge of why these alternative methods have worked, is a disaggregation of coverage rates by customer type and in what contexts, could inform the reform plans of (residential, commercial or government), location (major typical cities looking to serve the poor better. city versus smaller town), and income (so that utility service to the poor can be understood). 90 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix C: Perspectives on Alternative or Supplementary Service Providers Appendix C: Perspectives on Alternative or Supplementary Service Providers Table C.1: Perspectives on alternative or supplementary service providers Features, by Dependent Independent technology i) Piped Networks System Operator buys water in bulk from Operator develops own water sources (wells or boreholes) utility and develops distribution and connect network to households and other users sub-networks connected directly to households, institutions and public kiosks stand posts Organization Private company or individual, Sole proprieter, cooperative, private land and housing community organization or developer, water user association, community-based neighborhood association organization. Regulatory Issues Contract with utility, business license, Groundwater abstraction permits, title deeds, resale customer agreements, bulk rates, permits licenses, water quality testing, business licenses, customer tariffs rights to own infrastructure and/or to lay networks in public rights of way ii) Point Sources System Kiosk or standpost connected to the Water point linked to own source (well or borehole, utility network (could be household underground or above-ground storage tank) installed supply); buying water in bulk - at a privately and operated on a for-profit basis. Water may be special tariff - or at household tariff purchased from a tanker Organization Individual, enterprise, self-help group Neighborhood association, micro-enterprise, community based organizations Regulatory Issues Contract with utility, license/permit, Groundwater abstraction permits, license, tariff structure, customer tariff, bulk purchase price, water quality testing performance incentives iii) Mobile Distributors System Tankers or truckers obtain water in bulk Tankers, truckers and carters develop source or obtain from the utility (or municipal supply) water from a private well for distribution to households; and deliver it directly to the customer, public utility water storage tanks, communal cisterns, or including public utility water storage institutions tanks, communal cisterns, or individual households and institutions Organization Sole proprietor, tanker association, Sole proprietor, tanker association, lessee, informal sector lessee, informal sector Regulatory Issues Transport license, business license, Transport license, business license, water quality, tanker cleanliness, bulk rate, utility abstraction permit contract, customer tariff Source: Kariuki, Mukami, and Jordan Schwartz. 2005. “Small-Scale Private Service Providers of Water Supply and Electricity: A Review of Incidence, Structure, Pricing and Operating Characteristics.” World Bank Policy Research Working Paper 3727. 91 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix D: Utility Management Effectiveness Index Appendix D: Utility Management Effectiveness Index To measure utility management effectiveness ultimately a judgment call, indicator choice and weighting quantitatively, an index was developed (shown in Table are based on a synthesis of what is used by the water D.1). The indicators included in the index are those that regulators in Kenya, Zambia, and Tanzania. are widely available for a large set of utilities.39 While Using this index, management effectiveness of the utilities serving the 17 cities in our sample was calculated, as shown 39 For instance, NRW as a percentage of water supplied is the most widely used and understood measure of NRW; however, NRW is best represented as cubic meters per in Table D.2. Years were selected to match the year for connection per hour. which household survey data were available. Table D.1: Utility management effectiveness index Indicator Scoring (max 100 in each subcategory) Weight Operating cost coverage  If < 1.00, score = 0; =1.00, score = 50; ≥ 1.20, score = 100 30 Collection ratio Collection ratio (%) * 100 If collection ratio > 100%, score = 100 NRW  If NRW < 20%, score = 100; If NRW > 60%, score = 0  Otherwise, score = [ (NRW – 20%) / 40% ] * 100 Staff/1,000 conn.  If staff/1,000 conn. ≤ 4, score = 100; ≥ 12, score = 0  Otherwise, score = [ (staff/1,000 conn.) – 4 ] / 8 * 100 TOTAL 100 Source: An adaptation of the indices and weights used by WASREB, NWASCO, and EWURA. 92 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix D: Utility Management Effectiveness Index Table D.2: Inputs for utility management effectiveness index City Year OCCR Coll. ratio Staff prod. NRW ME index Accra 2008 1.15 93% 7.7 50% 66 Addis Ababa 2014 1.09 102% 5.1 37% 78 Dakar 2014 1.33 98% 2.6 20% 100 Dar es Salaam 2012 0.83 78% 6.2 50% 41 Durban 2015 1.30 98% 3.5 39% 88 Hargeisa 2011 0.90 95% 19.0 24% 43 Kaduna 2011 0.23 54% 16.7 30% 30 Kampala 2011 1.31 95% 5.9 35% 85 Kinshasa 2014 0.63 63% 12.1 43% 25 Lusaka 2014 1.26 96% 9.8 42% 69 Maputo 2011 1.03 123% 4.6 52% 67 Mombasa 2013 1.04 80% 9.3 49% 50 Nairobi 2013 1.05 78% 5.0 42% 68 Niamey 2012 1.13 119% 4.8 15% 92 Nyeri 2013 1.40 102% 4.2 24% 97 Ouagadougou 2010 1.24 96% 3.6 19% 99 Tanga 2012 0.98 100% 5.2 28% 62 OCCR = Operating cost coverage ratio. Coll. ratio = Collection ratio. Staff prod. = Staff per 1,000 water and sewer connections. NRW = Nonrevenue water. ME index = Management effectiveness index. Source: Utility research (see Table F.2 for full list of sources). 93 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data Appendix E: Longitudinal Utility Data This appendix contains longitudinal performance graphs • Total amount supplied into network and total for the utilities serving the 17 cities. Graphs were created amount sold; for 12 key indicators: • NRW; • Water coverage; • Total staff numbers; • Population served versus service area population; • Staff productivity; • Water connections; • Collection ratio; and • Connection density (connections per 100 people • Operating cost coverage. in city); • Hours of service per day; Blank graphs indicate that data were unavailable for • Network length; that indicator, for that utility. Data sources are listed in Appendix F.2. 94 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data E.1 Addis Ababa Water and Sewerage Authority (AAWSA) 95 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data 96 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data E.2 Aguas da Regiao de Maputo, Mozambique 97 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data 98 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data E.3 Dar es Salaam Water and Sewerage Corporation (DAWASCO), Tanzania 99 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data 100 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data E.4 eThekwini Water and Sanitation Unit, South Africa 101 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data 102 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data E.5 Ghana Water Company, Ltd. (GWCL), Ghana 103 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data 104 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data E.6 Hargeisa Water Agency (HWA), Somaliland 105 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data 106 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data E.7 Kaduna State Water Board (KSWB), Nigeria 107 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data 108 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data E.8 Lusaka Water and Sewerage Company (LWSC), Zambia 109 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data 110 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data E.9 Mombasa Water Supply and Sanitation Company (MOWASCO) 111 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data 112 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data E.10 Nairobi City Water and Sewerage Company (NCWSC) 113 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data 114 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data E.11 National Water and Sewerage Corporation (NWSC), Uganda 115 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data 116 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data E.12 Nyeri Water and Sewerage Company (NYEWASCO) 117 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data 118 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data E.13 l’Office national de l’eau et de l’assainissement (ONEA) 119 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data 120 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data E.14 REGIDESO, Democratic Republic of the Congo 121 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data 122 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data E.15 Sénégalaise des Eaux (SDE) and Société Nationale des Eaux du Sénégal (SONES), Senegal 123 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data 124 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data E.16 Société d’Exploitation des Eaux du Niger (SEEN), Niger 125 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data 126 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data E.17 Tanga Urban Water Supply and Sewerage Authority (Tanga UWASA), Tanzania 127 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix E: Longitudinal Utility Data 128 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix F: Data Sources Appendix F: Data Sources Sources of data used in this study are listed in this Appendix. F.1 Household survey data Data on access and convenience were obtained from household surveys. The surveys used are listed in Table F.1. Table F.1: Household survey data sources City Source Year Accra DHS 2008 Addis Ababa DHS 2014 Dakar DHS 2014 Dar es Salaam DHS 2012 Durban Census 2015 Hargeisa MICS 2011 Kaduna MICS 2011 Kampala DHS 2011 Kinshasa DHS 2014 Lusaka DHS 2014 Maputo DHS 2011 Mombasa Kenya City Baseline Survey 2012-2013 Nairobi Kenya City Baseline Survey 2012-2013 Niamey DHS 2012 Nyeri Kenya City Baseline Survey 2012-2013 Ouagadougou DHS 2010 Tanga DHS 2012 129 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix F: Data Sources F.2 Utility data The list of data sources used for utility-reported data is included in Table F.2. Table F.2: List of data sources for utility-reported data Utility List of data sources AAWSA  AAWSA Aguas Maputo  CEO Self-Assessment  CRA  IBNET DAWASCO  EWURA  IBNET eThekwini  eThekwini Water and Sanitation Unit  eThekwini financial statements Ghana WCL  Ghana WCL  IBNET Hargeisa Water and  Hargeisa Water and Sewerage Authority Sewerage Authority Kaduna State Water  Kaduna State Water Board Board Lusaka WSC  NWASCO  Lusaka WSC financial statements  IBNET MOWASCO  MOWASCO  WASREB Nairobi WSC  CEO Self-Assessment  WASREB NWSC  NWSC Annual Reports (with financial statements)  NWSC Strategic Plan, 2013–2018 NYEWASCO  NYEWASCO  NYEWASCO financial statements ONEA  Sawadogo, Dieudonne. “Delivering City-Wide Wash Services-Reaching informal settlements in Ouagadougou, Burkina Faso.” Presentation, August 2015  Marin, Philippe, Matar Fall, and Harouna Ouibiga. “Corporatizing a water utility: A successful case using a performance-based service contract for ONEA in Burkina Faso.” GridLines note no. 53 (March 2010), PPIAF  CEO Self-Assessment  IBNET 130 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix F: Data Sources Utility List of data sources Regideso  Tribeche. “Democratic Republic of Congo: Evolution institutionnelle de secteur de l’eau en milieu urbain: Investigations preliminiaires.” 2015  IBNET  Étude de l’effeicacite de la Regideso, du cadre institutionnel et de l’organization du secteur de l’eau urbaine en RDC: Phase 1, Diagnostic SEEN  CEO Self-Assessment SDE  “Indicateurs Senegal (18 08)”, Excel spreadsheet provided by the World Bank Dakar office  Public-Private partnerships for Urban Water Utilities: The Senegalese experience, AFD. PowerPoint presentation, 2011  Trémolet, Sophie and Castalia. Case Study on Senegal’s Water and Sanitation Sector Economic Regulation, May 2006, and interviews with SDE management team, November 2015 Tanga UWASA  Tanga UWASA  EWURA  IBNET 131 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix G: References Appendix G: References Banerjee, Sudeshna, Vivien Foster, Yvonne Ying, Heather Howard, G., and Bartram, J. 2003. “Domestic water Skilling, and Quentin Wodon. 2010. “Cost Recovery, quantity, service level and health.” Geneva: World Equity, and Efficiency in Water Tariffs: Evidence from Health Organization. African Utilities.” The World Bank. Kariuki, Mukami, and Jordan Schwartz. 2005. “Small- Beck, Linda. 1999. “Senegal’s Enlarged Presidential Scale Private Service Providers of Water Supply and Majority: Deepening Democracy or Detour?” In State, Electricity: A Review of Incidence, Structure, Pricing Conflict & Democracy in Africa, edited by Richard and Operating Characteristics.” World Bank Policy Joseph. Boulder: Lynne Rienner. Research Working Paper 3727. Bratton, Michael, and Carolyn Logan. 2013. “Voters but Kayaga, S., J. Fisher, and R. Franceys. “Improved Access not yet Citizens.” In Voting and Democratic Citizenship to Urban Water Services in Uganda Proceedings of the in Africa, edited by Michael Bratton. Boulder: Lynne ICE: Municipal Engineer.” ICE Publishing 162 (2009): Rienner. 165–170. Accessed April 28, 2016. doi: 10.1680/ Carpenter, Daniel P. 2001. “The Forging of Bureaucratic muen.2009.162.3.165. Autonomy: Reputations, Networks, and Policy Keefer, Philip. 2004. “What Does Political Economy Tell Innovation in Executive Agencies, 1862-1928.” Us About Economic Development – and Vice Versa?” Princeton. The World Bank, WPS 3250. Castalia Strategic Advisors. 2014. Barriers to Infrastructure Komives, Kristine, Vivien Foster, Jonathan Halpern, and Service Delivery in Sub-Saharan Africa and South Asia. Quentin Wodon, with support from Roohi Abdullah. Evans, Barbara, Jamie Bartram, Paul Hunter, Ashley 2005. “Water, Electricity, and the Poor: Who Benefits Rhoderick Williams, Jo-Anne Geere, Batsi Majuru, from Utility Subsidies?” The World Bank. Laura Bates, Michael Fisher, Alycia Overbo, and Wolf- Kooy, Michelle, and Daniel Harris. 2012. “Political Peter Schmidt. 2013. Public Health and Social Benefits economy analysis for operations in water and sanitation: of at-house Water Supplies. a guidance note.” ODI. Fosu, Augustin. 2013. “Institutions and African Economies: Legros, D., M. McCormick, C. Mugero, M. Skinnider, D.D. An Overview.” Journal of African Economies 22, 4: 491– Bek’Obita, and S.I. Okware. 2000. “Epidemiology of 498. Cholera Outbreak in Kampala, Uganda.” East African Harris, Daniel, and Michelle Kooy. 2012. “Political Medical Journal 77 (7): 347–349. Economy Analysis for Operations in Water and Levy, Brian. 2014. “Working with the Grain: Integrating Sanitation: A Guidance Note.” ODI. Governance and Growth in Development Strategies.” Hawkins, Peter, Isabel Blackett, and Chris Heymans. 2013. Oxford University Press. “Delivering Sanitation to the Urban Poor: A Scoping Levy, Brian, Alan Hirsch, and Ingrid Woolard. 2015. Study.” The World Bank. “Governance and Inequality: Benchmarking and Heymans, C., K. Eales, and R. Franceys. 2014. “The Limits Interpreting South Africa’s Evolving Political and Possibilities of Prepaid Water in Urban Africa: Settlement.” Effective States and Inclusive Development Lessons from the Field.” Water Sanitation Program (ESID) Working Paper No. 51. (WSP). Washington: The World Bank. 132 Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms | Appendix G: References Manghee, Seema, and Alice Poole. 2012. “Approaches to Strande, Linda, Mariska Ronteltap, and Damir Brdjanovic. Conducting Political Economy Analysis in the Urban 2015. “Faecal Sludge Management: Systems Approach Water Sector.” The World Bank. for Implementation and Operation.” London: IWA Matsiko, Haggai. 2011. “Behind the Scenes at Muhairwe’s Publishing. Exit.” The Independent (Uganda). http://independent. Tangri, Robert K. 1999. “The Politics of Patronage in co.ug/business/business-news/4897-behind-the- Africa: parastatals, privatization and private enterprise.” scenes-at-muhairwes-exit. Oxford. Mugisha, Silver, Sanford V. Berg, and William T. Muhairwe. Trémolet, Sophie, and Castalia Strategic Advisors. 2006. 2006. “Using Internal Incentive Contracts to Improve “Case Study on Senegal’s Water and Sanitation Sector Water Utility Performance: The Case of Uganda’s Economic Regulation.” NWSC.” March 29 (Draft paper). Van den Berg, Caroline, and Alexander Danilenko. Muhairwe, William. 2010. Making Public Enterprises Work. Forthcoming. “Performance of Water and Wastewater IWA Publishing. Utilities in Africa.” The World Bank. North, Douglass C. 1990. Institutions, Institutional WHO (World Health Organization). Progress on Change, and Economic Performance. Cambridge: Sanitation and Drinking Water: 2015 Update and Cambridge University Press. MDG Assessment. Accessed April 28, 2016. http:// Olson, Mancur. 1982. The Rise and Decline of Nations: apps.who.int/iris/bitstream/10665/177752/1/978924 Economic Growth, Stagflation, and Social Rigidities. Yale 1509145_eng.pdf. University Press. World Bank. 2014. “Do Pro-Poor Policies Increase Water Ottaway, Marina. 2000. Africa’s New Leaders: Democracy Coverage: An analysis of service delivery in Kampala’s or State Reconstruction? Washington, DC: Carnegie Informal Settlements.” The World Bank. Endowment for International Peace. World Bank. Forthcoming. “Fecal Sludge Management: Schwartz, Klaas. 2008. “The New Public Management: Diagnostics and Decision Support Tools for Service The future for reforms in the African water supply and Delivery in Poor Urban Areas.” The World Bank. sanitation sector?” Utilities Policy Volume 16 Issue 1: World Bank. Water Supply and Sanitation in Senegal: 49–58. Turning Finance into Services for 2015 and Beyond. Steinmo, Sven, Kathleen Thelen, and Frank Longstreth. Accessed April 28, 2016. https://www.wsp.org/sites/ 1992. Structuring Politics: Historical Institutionalism wsp.org/files/publications/CSO-senegal.pdf. in Comparative Analysis. Cambridge: Cambridge University Press. 133 August 2016 Water and Sanitation Program The World Bank Group Delta Centre, Menengai Road, Upper Hill P.O. Box 30577 - 00100 Nairobi, Kenya Phone: +254 20 293-6000 E-mail: wspaf@worldbank.org Web site: www.worldbank.org/water Design and Layout by Eric Lugaka