60804 DIREC TIONS IN DE VELOPMENT Infrastructure Africa's Water and Sanitation Infrastructure Access, Affordability, and Alternatives Sudeshna Ghosh Banerjee and Elvira Morella Africa's Water and Sanitation Infrastructure Africa's Water and Sanitation Infrastructure Access, Affordability, and Alternatives Sudeshna Ghosh Banerjee and Elvira Morella Vivien Foster and Cecilia Briceño-Garmendia, Series Editors © 2011 The International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org All rights reserved 1 2 3 4 14 13 12 11 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 in this volume do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The bound- aries, colors, denominations, and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this 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; Internet: 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. ISBN: 978-0-8213-8457-2 eISBN: 978-0-8213-8618-7 DOI: 10.1596/978-0-8213-8457-2 Library of Congress Cataloging-in-Publication Data Africa's water and sanitation infrastructure: access, affordability, and alternatives / editors, Sudeshna Ghosh Banerjee, Elvira Morella. p. cm. Includes bibliographical references. ISBN 978-0-8213-8457-2 -- ISBN 978-0-8213-8618-7 (electronic) 1. Water utilities--Africa. 2. Water-supply--Economic aspects--Africa. 3. Sanitation-- Economic aspects--Africa. 4. Sewage disposal--Economic aspects--Africa. I. Banerjee, Sudeshna Ghosh, 1973- II. Morella, Elvira, 1976- III. World Bank. HD4465.A35.A47 2011 363.6'1096--dc22 2010047886 Cover photograph: Arne Hoel / The World Bank Cover design: Naylor Design Contents About the AICD xvii Series Foreword xix Acknowledgments xxi Abbreviations xxvii Chapter 1 The Elusiveness of the Millennium Development Goals for Water and Sanitation 1 A Timely Synthesis 3 Data Sources and Methodologies 4 Key Finding 1: Wide Differences in Patterns of Access to Water 10 Key Finding 2: Equally Wide Differences in Patterns of Access to Sanitation 13 Key Finding 3: High Costs, High Tariffs, and Regressive Subsidies 16 Key Finding 4: The Stark Challenge of Financing the MDG 18 Key Finding 5: Institutional Reform for Better Water Sector Performance 24 A Multidimensional Snapshot of WSS in Africa 27 v vi Contents Annex 1.1 Surveys in the AICD DHS/MICS Survey Database 28 Annex 1.2 Surveys in the AICD Expenditure Survey Database 29 Annex 1.3 Introducing a Country Typology 30 Notes 31 Bibliography 31 Chapter 2 Access to Safe Water: The Millennium Challenge 33 The Importance of Wells and Boreholes in Water Supply 33 Low Access to Piped Water. . . for Various Reasons 37 Multiple Players in the Urban Water Market 42 The Role of Wells, Boreholes, and Surface Water in the Rural Water Market 49 Steep Growth of Wells and Boreholes as Sources of Water 52 Notes 59 References 59 Chapter 3 Access to Safe Sanitation: The Millennium Challenge 63 The Predominance of On-Site and Traditional Pit Latrines 63 The Sanitation Challenge across Countries 66 Steep Increases in the Use of Traditional Pit Latrines 69 Good Progress in a Handful of Countries 71 References 81 Chapter 4 Improving the Organization of the Water and Sanitation Sectors 83 The Heterogeneity of the Urban Water Market 84 Varied Institutional Models for Nonpiped Services in the Urban Water Market 102 Many Levels of Government Players in the Rural Water Market 110 Many Players with No Clear Accountability in the Sanitation Market 115 Notes 120 References 120 Contents vii Chapter 5 Urban Water Provision: The Story of African Utilities 123 Access to Utility Water 124 The Pace of Expansion of Utility Water Coverage 126 Water Production Capacity Varies from Country to Country 128 Two-Part Tariff Structures for Piped Water 128 Sewerage Charges Linked to Water Bills 131 Modest Water Consumption by End Users 132 Substantial Water Losses in Distribution System 136 Difference in Quality of Service among Country Groups 138 Technical Efficiency and Effective Management of Operations 139 Financial Efficiency and the Alignment of Operations and Finances 142 The High Cost of Inefficiencies in Operations and Pricing 146 The Role of Institutions in Improving Performance 153 Annex 5.1 Utilities in the AICD WSS Database 158 Notes 159 References 159 Chapter 6 Cost Recovery, Affordability, and Subsidies 161 Average Monthly Spending on Water 161 Wide Price Variations among Service Providers in the Urban Water Market 162 Two-Part Tariffs and the Small Consumer 168 Paying for Water: How Common? 171 Recovering Operating Costs: Affordable 173 The High Cost of Connecting to Water and Sanitation Services 176 The Cost of Subsidizing Capital and Operating Expenses 179 Poor Targeting of Utility Subsidies 180 Connection Subsidies as a Viable Alternative 184 Annex 6.1 Methodology for Estimating the Annual Gross Profit and the Annual Cross-Subsidy between Household Consumers and Standpipes Captured by Standpipe Operators in a City 187 viii Contents Notes 188 References 189 Chapter 7 Spending Needed to Meet Goals in Water and Sanitation 191 The Challenge of Expanding Coverage 191 The Unit Cost of Service Provision across Countries 197 To Close the MDG Coverage Gap 202 Annex 7.1 Unit Cost Matrix Model: A Methodology for Estimating Nonstandardized Unit Costs of Network Assets 209 Annex 7.2 Methodology for Quantifying Rehabilitation and O&M Needs 213 Notes 214 References 214 Chapter 8 Bridging the Funding Gap 215 Current Spending on Water and Sanitation 215 Poor Budget Execution by the WSS Sector 220 Even after Efficiency Savings, a Persistent Funding Gap 224 Limited Scope for Raising Additional Finance 231 Promising Ways to Increase Funds 239 Other Ways to Reach the MDG 240 Notes 248 References 248 Chapter 9 Policy Options for the Water and Sanitation Sectors 251 Policy Options for the Water Sector 251 Policy Options for the Sanitation Sector 260 Reference 267 Appendix 1 Access to Water Supply and Sanitation Facilities 269 Appendix 2 Institutions in the Water and Sanitation Sector 293 Appendix 3 Performance Indicators of Selected Water Utilities 323 Contents ix Appendix 4 Tariffs 349 Appendix 5 Affordability of Water and Sanitation 365 Appendix 6 Funding Gap for Water Supply and Sanitation 371 Index 387 Boxes 2.1 The Problem of Shrinking Households 38 2.2 Coverage, Access, and Hookup Rates: Relationships and Definitions 40 2.3 Legalizing Household Resellers in Côte d'Ivoire 48 2.4 The Opportunity Cost of a Distant Water Supply 50 3.1 Ethiopia's Success with a Community-Led Program 77 4.1 Senegal's Successful Experience with Private Sector Participation 88 4.2 Regulation in Water Reseller Market in Abidjan 109 4.3 Issues Constraining Rural Water Supply in Cross River State, Nigeria 111 5.1 Burkina Faso's Sanitation Tax 134 5.2 Methodology for Estimation of Hidden Cost 147 5.3 Uganda's Successful Case of State-Owned Enterprise Reform 156 6.1 Piped Water Delivered through Public Standposts in Kigali, Rwanda 165 7.1 The Construction Index Factor 201 7.1A Unit Costs of Infrastructure Projects Study 211 Figures 1.1 JMP and AICD Estimates of the Prevalence of "Improved" Water Supply and Sanitation 6 1.2 The Sanitation Ladder 7 1.3 Dependence on Surface Water in Urban and Rural Areas, 1990s versus Early 2000s 14 1.4 Coverage of Water Services, by Income Quintile 15 1.5 Coverage of Sanitation Services, by Income Quintile 16 1.6 Annual Growth in the Use of Sanitation Types, 1990­2005 17 x Contents 1.7 Affordability of Full-Cost-Recovery Tariffs in Low-Income Countries 19 1.8 Hidden Costs and Institutions 25 2.1 African Households' Access to Various Forms of Water Supply 35 2.2 Extent of Access to Piped Water through Household Connection, by GDP and Urbanization Rate 37 2.3 Country Scatter Plot of Current Access Rates for Piped Water and Demand-Side Factors in Coverage Deficit 42 2.4 Working Status of Rural Water Points 52 2.5 Annualized Change in Coverage of Various Water Sources, 1995­2005 55 2.6 Annualized Change in Coverage of Various Water Sources, 1995­2005 56 2.7 Four Solid Performers in Expanding Access to Safe Water, 1995­2005 58 3.1 Population That Has Wastewater Connection in the Utility Service Area 64 3.2 Population Sharing Water and Toilet Facilities 65 3.3 Access Patterns across Income Quintiles 66 3.4 Annualized Growth in Coverage in Urban and Rural Areas, 1995­2005 70 3.5 Growth in Access by Mode and Quintile 72 3.6 Successful Examples from Up and Down the Sanitation Ladder, 1995­2005 75 4.1 Range of Institutional Arrangements in Water Service Provision 84 4.2 Country Ranking and Prevalence of Key Reform Activities 90 4.3 Year of Establishment of Regulatory Agencies 92 4.4 Understanding Performance in Regulatory Autonomy 93 4.5 Prevalence and Key Attributes of the Social Accountability Index 95 4.6 Country Ranking and Prevalence of Key Attributes of Regulation 96 4.7 Legal Status and Ownership Structure of Water Utilities 97 4.8 Performance in Managerial Autonomy 98 4.9 Performance Monitoring 99 Contents xi 4.10 Country Ranking and Prevalence of Key Attributes of the SOE Governance 101 4.11 Solid Country Performances 102 4.12 Utility Direct Management Models 104 4.13 Delegated Management Models 106 4.14 Responsibility for Maintenance and Monitoring of Rural Water Points 113 4.15 Country Ranking and Prevalence of Key Attributes for the Rural Reform Index 114 4.16 Responsibilities for On-Site Sanitation Functions 116 4.17 Urban Utilities' Responsibility over On-Site Sanitation and Wastewater Management 116 4.18 On-Site Sanitation Index 118 5.1 Expansion of Utility Water Coverage 127 5.2 Variations in Tariff Structures 131 5.3 Utility Prices and Charges 132 5.4 Frequency Distribution of Nonrevenue Water 137 5.5 Cross-Plots between NRW and Other Variables 137 5.6 Effective Tariffs at Various Consumption Levels 144 5.7 Reported versus Implicit Collection Ratios 146 5.8 Utility Inefficiencies as Percentage of Total Utility Revenue 149 5.9 Utility Efficiency Affects Access Expansion and Water Quality 152 5.10 Hidden Costs and Institutions 153 6.1 Spending on Water Services 162 6.2 Comparison of Official and Retail Standpost and Small Piped Consumer Prices 164 6.3 Price by Water Service Provider 167 6.4 Average Water Tariffs for Africa at Different Consumption Levels 169 6.5 Utilities Charging Higher Effective Prices to Small Consumers 170 6.6 Connection and Payment, by Consumer Categories 171 6.7 Nonpayment Rates of Water Services 172 6.8 Share of Average Urban Household Budget Required to Purchase Subsistence Amounts of Piped Water, by Continental Income Quintiles 175 6.9 Formal Water Connection Cost 177 xii Contents 6.10 Subsidy Needed to Maintain Affordability of Water Services 180 6.11 Overall Targeting Performance () of Utility Subsidies 182 6.12 Access Factors and Subsidy Design Factors Affecting Targeting Performance 183 6.13 Potential Targeting Performance of Connection Subsidies under Various Scenarios 185 7.1 Water MDG Gap, 2006 192 7.2 Sanitation MDG Gap, 2006 194 7.3 Population Split across Water and Sanitation Modalities Given Current and Target Coverage by 2015 under the Base Scenario Assumptions 197 7.4 Urban-Rural Split of Spending Needs 204 7.5 Africa's Water and Sanitation Needs by Country 207 8.1 Water and Sanitation Spending from All Sources as a Percentage of GDP, Annual Averages by Functional Category, 2001­05 218 8.2 Water and Sanitation Capital Investment as a Percentage of GDP, by Funding Source, Annualized Averages for 2001­05 220 8.3 Split Investment Responsibilities between Governments and Public Enterprises 223 8.4 Potential Efficiency Gains from Different Sources 227 8.5 Water Infrastructure Funding Gap 228 8.6 Aid Commitments for Water Supply and Sanitation as a Percentage of GDP, 2001­05 235 8.7 Costs of Capital by Funding Source, 2001­05 238 8.8 Spending Needs by Country Type under Different Time Horizons 241 8.9 Annual Spending Needs over Different Time Horizons, by Country Type 242 8.10 Spending Needs by Country under Different Level- of-Service Assumptions 245 8.11 Overall Spending Needs by Country Groups under Different Service Assumptions 247 Tables 1.1 Regional Progress toward the MDG Drinking Water Target 2 1.2 Regional Progress toward the MDG Sanitation Target 3 1.3 Definition of Coverage of Improved Water 5 Contents xiii 1.4 Modules of AICD WSS Survey 8 1.5 Utilities Analyzed in This Report, by Categories 10 1.6 Evolution of Water Supply Coverage in Africa, by Source 11 1.7 Services Provided by Utilities in Their Service Areas 12 1.8 Quality of Services Provided by Utilities in Their Service Areas 12 1.9 Average Price for Water Service in 15 Largest Cities, by Type of Provider 13 1.10 Patterns of Access to Sanitation in Africa 15 1.11 Overall WSS Spending Needs 20 1.12 Breakdown of Spending Needed to Meet MDGs in WSS, by Spending Category and Country Group 21 1.13 Spending by Functional Category, Annualized Average Flows, 2001­05 22 1.14 Funding Gap 23 1.15 Overview of Impact of Private Sector Participation on Utility Performance 26 2.1 Coverage Rate of Water Supply 36 2.2 Coverage Rate of Water Supply, by Country Income and Urbanization Status 37 2.3 Water-Service Coverage in Urban Africa and Share of Coverage Deficit Attributable to Demand-Side Factors 41 2.4 Patterns of Urban Access to Water 43 2.5 Water Supply in Africa's Largest Cities, by Source 45 2.6 Working Status of Standposts in the Largest Cities in Africa 47 2.7 Patterns of Access across Countries in Rural Areas 51 2.8 Evolution of Water-Supply Sources, 1990­2005 54 2.9 Annualized Change in Coverage by Water Source and by Country, 1995­2005 57 3.1 Patterns of Access to Sanitation 65 3.2 Patterns of Access to Flush Toilets and Alternatives 67 3.3 Patterns of Access to Sanitation across Countries 69 3.4 Annualized Change in Coverage, 1995­2005 74 3.5 Annualized Change in Coverage by Modality and by Country, 1990­2005 79 4.1 Urban Reform, Regulation, and the SOE Governance Index 86 4.2 Regulatory Roles in the Urban Water Sector 91 xiv Contents 4.3 Standpipe Management 104 4.4 Regulation of Household Water Resellers 109 4.5 Regulation of Water Tankers 110 4.6 Stakeholder Involvement in Rural Water Activities 112 4.7 Management of Latrines 119 5.1 Comparison of Coverage Statistics for Water, Based on Utility Data versus Household Surveys 125 5.2 Overview of Access Patterns in the Utility Service Area 126 5.3 Water Production per Capita in the Utility Service Area 129 5.4 Structure and Level of Wastewater Tariffs 133 5.5 Indicators of Demand Management Calculated across Utilities with Metering Ratios above 50 Percent 136 5.6 Indicators of Service Quality 139 5.7 Indicators of Operational Efficiency 140 5.8 Utility Cost Structures 141 5.9 Utility Financial Ratios 143 5.10 Hidden Cost of Inefficiencies 150 5.11 Overview of Impact of Private Sector Participation on Utility Performance 155 6.1 Monthly Household Budget 163 6.2 Prices by Alternate Water Service Provider 166 6.3 Comparison of Water Tariffs in Africa and Other Global Regions at Various Levels of Consumption 168 6.4 Reference Points for the True Cost of Infrastructure Services 174 6.5 Share of Urban Households Whose Utility Bill Would Exceed 5 Percent of the Monthly Household Budget at Various Prices 176 6.6 Cost of Facility as Percentage of Monthly Household Budget in Senegal 179 7.1 Additional Population to Be Served by 2015 195 7.2 Unit Costs of Water Network Infrastructure Services by Location in the Median Country 199 7.3 Unit Costs of Wells and Boreholes 200 7.4 Unit Costs of On-Site Sanitation Services 201 7.5 Overall Water and Sanitation Spending Needs 203 7.6 Split of Spending Needs by Category 205 7.1A Population Density across Urban and Rural Typologies (Number of People per Square Kilometer) 210 Contents xv 8.1 Spending by Functional Category, Annualized Average Flows, 2001­05 216 8.2 Capital Investments of the Most Important Players, Annualized Average Flows, 2001­05 219 8.3 Annual Budgetary Flows, Annualized Averages, 2001­05 221 8.4 Public Infrastructure Spending by Institution in the WSS Sector, 2001­05 222 8.5 Average Budget Variation Ratios for Capital Spending 223 8.6 Potential Gains from Greater Efficiency 225 8.7 Funding Gap 230 8.8 Size and Composition of the Annual Funding Gap by O&M and Capital Expenditure 231 8.9 Net Change in Central Government Budgets, by Economic Use 232 8.10 Outstanding Financing Stock for Water and Sanitation Infrastructure, as of 2006 237 8.11 Time Needed to Meet the MDG Targets with Today's Budget Envelopes 243 8.12 Spending Needs to Meet the MDG Targets under Different Level-of-Service Scenarios 246 8.13 Funding Gaps under Base and Pragmatic Scenarios 248 A1.1 Piped Water 270 A1.2 Standposts 272 A1.3 Wells/Boreholes 274 A1.4 Surface Water 276 A1.5 Septic Tank 278 A1.6 Improved Latrine 280 A1.7 Traditional Latrine 282 A1.8 Open Defecation 284 A1.9 Annualized Change in Water Access: National 286 A1.10 Annualized Change in Water Access: Urban 287 A1.11 Annualized Change in Water Access: Rural 288 A1.12 Annualized Change in Sanitation Access: National 289 A1.13 Annualized Change in Sanitation Access: Urban 290 A1.14 Annualized Change in Sanitation Access: Rural 291 A2.1 Specification of Urban Water Reform Index 294 A2.2 Urban Water Reform Index 296 A2.3 Specification of Regulation Index 300 A2.4 Regulation Index 302 A2.5 Specification of SOE Governance Index 306 xvi Contents A2.6 SOE Governance Index 309 A2.7 Specification of Rural Water Reform Index 318 A2.8 Rural Water Reform Index 318 A2.9 Specification of On-Site Sanitation Index 319 A2.10 On-Site Sanitation Index 320 A3.1 Access to Utility Water 324 A3.2 Distribution Infrastructure 328 A3.3 Treatment 336 A3.4 Staffing 340 A3.5 Financial Performance 344 A4.1 Structure of Domestic Tariffs 350 A4.2 Domestic Tariffs at Various Levels of Consumption 352 A4.3 Cost Recovery at Various Levels of Consumption 354 A4.4 Structure of Nondomestic Tariffs 356 A4.5 Structure of Sanitation Tariffs (Only Utilities with Wastewater Responsibility) 358 A4.6 Structure of Standpost Tariffs 359 A4.7 Scorecard on Efficiency, Equity, and Cost Recovery 361 A5.1 Contribution of Food to Total Spending 366 A5.2 Spending on Water Services 367 A5.3 Affordability of Piped Water at 5 Percent Budget Threshold for Urban Households 369 A6.1 Water and Sanitation Expansion and Rehabilitation 372 A6.2 Indicative Water and Sanitation Spending Needs 374 A6.3 Existing Financing Flows to Water and Sanitation Sectors 376 A6.4 Annual Budgetary Flows (not Traced) 378 A6.5 Public Infrastructure Spending by Sector and Institution 380 A6.6 Size and Composition of Funding Gap 382 A6.7 Reducing the Funding Gap 384 About the AICD This study is a product of the Africa Infrastructure Country Diagnostic (AICD), a project designed to expand the world's knowledge of physical infrastruc- ture in Africa. The AICD provides a baseline against which future improvements in infrastructure services can be measured, making it possible to monitor the results achieved from donor support. It also offers a more solid empirical foundation for prioritizing invest- ments and designing policy reforms in the infrastructure sectors in Africa. The AICD was based on an unprecedented effort to collect detailed economic and technical data on the infrastructure sectors in Africa. The project produced a series of original reports on public expenditure, spend- ing needs, and sector performance in each of the main infrastructure sectors, including energy, information and communication technologies, irrigation, transport, and water and sanitation. The most significant findings were synthesized in a flagship report titled Africa's Infrastructure: A Time for Transformation. All the under- lying data and models are available to the public through a Web portal (http://www.infrastructureafrica .org), allowing users to download customized data reports and perform various simulation exercises. The AICD was commissioned by the Infrastructure Consortium for Africa following the 2005 G-8 Summit at Gleneagles, which flagged the importance of scaling up donor finance to infrastructure in support of Africa's development. The first phase of the AICD focused on 24 coun- tries that together account for 85 percent of the gross domestic product, population, and infrastruc- ture aid flows of Sub-Saharan Africa. The countries were Benin, Burkina Faso, Cape Verde, Cameroon, Chad, Democratic Republic of Congo, Côte d'Ivoire, xvii xviii About the AICD Ethiopia, Ghana, Kenya, Lesotho, Madagascar, Malawi, Mozambique, Namibia, Niger, Nigeria, Rwanda, Senegal, South Africa, Sudan, Tanzania, Uganda, and Zambia. Under a second phase of the project, coverage was expanded to include the remaining countries on the African continent. Consistent with the genesis of the project, the main focus was on the 48 countries south of the Sahara that face the most severe infrastructure chal- lenges. Some components of the study also covered North African countries to provide a broader point of reference. Unless otherwise stated, therefore, the term "Africa" is used throughout this report as a shorthand for "Sub-Saharan Africa." The AICD was implemented by the World Bank on behalf of a steering committee that represents the African Union, the New Partnership for Africa's Development (NEPAD), Africa's regional eco- nomic communities, the African Development Bank, and major infrastructure donors. Financing for the AICD was provided by a multidonor trust fund to which the main contributors were the Department for International Development (United Kingdom), the Public Private Infrastructure Advisory Facility, Agence Française de Développement, the European Commission, and Germany's Kreditanstalt für Wiederaufbau (KfW). The Sub-Saharan Africa Transport Policy Program and the Water and Sanitation Program provided technical support on data collection and analysis pertaining to their respec- tive sectors. A group of distinguished peer reviewers from policy-making and academic circles in Africa and beyond reviewed all of the major outputs of the study to ensure the technical quality of the work. Following the completion of the AICD project, long- term responsibility for ongoing collection and analysis of African infrastructure statistics was transferred to the African Development Bank under the Africa Infrastructure Knowledge Program (AIKP). A second wave of data collection of the infrastructure indicators analyzed in this volume was initiated in 2011. Series Foreword The Africa Infrastructure Country Diagnostic (AICD) has produced con- tinent-wide analysis of many aspects of Africa's infrastructure challenge. The main findings were synthesized in a flagship report titled Africa's Infrastructure: A Time for Transformation, published in November 2009. Meant for policy makers, that report necessarily focused on the high-level conclusions. It attracted widespread media coverage feeding directly into discussions at the 2009 African Union Commission Heads of State Summit on Infrastructure. Although the flagship report served a valuable role in highlighting the main findings of the project, it could not do full justice to the richness of the data collected and technical analysis undertaken. There was clearly a need to make this more detailed material available to a wider audience of infrastructure practitioners. Hence the idea of producing four technical monographs, such as this one, to provide detailed results on each of the major infrastructure sectors--information and communication technologies (ICT), power, transport, and water--as companions to the flagship report. These technical volumes are intended as reference books on each of the infrastructure sectors. They cover all aspects of the AICD project relevant to each sector, including sector performance, gaps in financing and efficiency, and estimates of the need for additional spending on xix xx Series Foreword investment, operations, and maintenance. Each volume also comes with a detailed data appendix--providing easy access to all the relevant infrastructure indicators at the country level--which is a resource in and of itself. In addition to these sector volumes, the AICD has produced a series of country reports that weave together all the findings relevant to one par- ticular country to provide an integral picture of the infrastructure situa- tion at the national level. Yet another set of reports provides an overall picture of the state of regional integration of infrastructure networks for each of the major regional economic communities of Sub-Saharan Africa. All of these papers are available through the project web portal, http://www.infrastructureafrica.org, or through the World Bank's Policy Research Working Paper series. With the completion of this full range of analytical products, we hope to place the findings of the AICD effort at the fingertips of all interested policy makers, development partners, and infrastructure practitioners. Vivien Foster and Cecilia Briceño-Garmendia Acknowledgments This book was coauthored by Sudeshna Ghosh Banerjee and Elvira Morella with support from Carolina Dominguez, under the overall guid- ance of series editors Vivien Foster and Cecilia Briceño-Garmendia. All are with the World Bank. The book draws upon a number of background papers that were pre- pared by World Bank staff and consultants, under the auspices of the Africa Infrastructure Country Diagnostic (AICD). Key contributors to the book on a chapter-by-chapter basis were as follows. Chapter 1 Contributors Sudeshna Ghosh Banerjee, Cecilia Briceño-Garmendia, Tarik Chfadi, Amadou Diallo, Carolina Dominguez, Vivien Foster, Sarah Keener, Manuel Luengo, Elvira Morella, Taras Pushak, Heather Skilling, Clarence Tsimpo, Helal Uddin, Quentin Wodon. Key Source Documents Banerjee, S., H. Skilling, V. Foster, C. Briceño-Garmendia, E. Morella, and T. Chfadi. 2008. "Ebbing Water, Surging Deficits: Urban Water Supply in Sub-Saharan Africa." AICD Background Paper 12, World Bank, Washington, DC. xxi xxii Acknowledgments Banerjee, S., Q. Wodon, A. Diallo, T. Pushak, H. Uddin, C. Tsimpo, and V. Foster. 2008. "Access, Affordability and Alternatives: Modern Infrastructure Services in Sub-Saharan Africa." AICD Background Paper 2, World Bank, Washington, DC. Briceño-Garmendia, C., K. Smits, and V. Foster. 2008. "Financing Public Infrastructure in Sub-Saharan Africa: Patterns and Emerging Issues." AICD Background Paper 15, World Bank, Washington, DC. Keener, S., M. Luengo, and S. G. Banerjee. 2009. "Provision of Water to the Poor in Africa: Experience with Water Standposts and the Informal Water Sector." AICD Working Paper 13, World Bank, Washington, DC. Morella, E., V. Foster, and S. Banerjee. 2008. "Climbing the Ladder: The State of Sanitation in Sub-Saharan Africa." AICD Background Paper 13, World Bank, Washington, DC. Chapter 2 Contributors Sudeshna Ghosh Banerjee, Cecilia Briceño-Garmendia, Tarik Chfadi, Amadou Diallo, Vivien Foster, Sarah Keener, Manuel Luengo, Elvira Morella, Taras Pushak, Heather Skilling, Clarence Tsimpo, Helal Uddin, Quentin Wodon. Key Source Documents Banerjee, S., H. Skilling, V. Foster, C. Briceño-Garmendia, E. Morella, and T. Chfadi. 2008. "State of the Sector Review: Rural Water Supply." AICD Working Paper, World Bank, Washington, DC. Banerjee, S., Q. Wodon, A. Diallo, T. Pushak, H. Uddin, C. Tsimpo, and V. Foster. 2008. "Access, Affordability and Alternatives: Modern Infrastructure Services in Sub-Saharan Africa." AICD Background Paper 2, World Bank, Washington, DC. Keener, S., M. Luengo, and S. G. Banerjee. 2009. "Provision of Water to the Poor in Africa: Experience with Water Standposts and the Informal Water Sector." AICD Working Paper 13, World Bank, Washington, DC. Chapter 3 Contributors Elvira Morella, Sudeshna Ghosh Banerjee, Amadou Diallo, Vivien Foster, Taras Pushak, Clarence Tsimpo, Helal Uddin, Quentin Wodon. Key Source Documents Banerjee, S., Q. Wodon, A. Diallo, T. Pushak, H. Uddin, C. Tsimpo, and V. Foster. 2008. "Access, Affordability and Alternatives: Modern Infrastructure Services in Sub-Saharan Africa." AICD Background Paper 2, World Bank, Washington, DC. Acknowledgments xxiii Morella, E., V. Foster, and S. Banerjee. 2008. "Climbing the Ladder: The State of Sanitation in Sub-Saharan Africa." AICD Background Paper 13, World Bank, Washington, DC. Chapter 4 Contributors Maria Vagliasindi, Cecilia Briceño-Garmendia, Tarik Chfadi, Vivien Foster, Sarah Keener, Manuel Luengo, John Nellis, Heather Skilling, Sudeshna Ghosh Banerjee. Key Source Documents Vagliasindi, M., and J. Nellis. 2009. "Evaluating Africa's Experience with Institutional Reforms for the Infrastructure Sectors." AICD Working Paper 23, World Bank, Washington, DC. Banerjee, S., H. Skilling, V. Foster, C. Briceño Garmendia, E. Morella, and T. Chfadi. 2008. "Ebbing Water, Surging Deficits: Urban Water Supply in Sub-Saharan Africa." AICD Background Paper 12, World Bank, Washington, DC. Keener, S., M. Luengo, and S. G. Banerjee. 2009. "Provision of Water to the Poor in Africa: Experience with Water Standposts and the Informal Water Sector." AICD Working Paper 13, World Bank, Washington, DC. Chapter 5 Contributors Cecilia Briceño-Garmendia, Sudeshna Ghosh Banerjee, Tarik Chfadi, Vivien Foster, John Nellis, Heather Skilling, Karlis Smits, Maria Vagliasindi, Quentin Wodon, Yvonne Ying. Key Source Documents Banerjee, S., V. Foster, Y. Ying, H. Skilling, and Q. Wodon. 2008. "Cost Recovery, Equity and Efficiency in Water Tariffs: Evidence from African Utilities." AICD Working Paper 7, World Bank, Washington, DC. Banerjee, S., H. Skilling, V. Foster, C. Briceño-Garmendia, E. Morella, and T. Chfadi. 2008. "Ebbing Water, Surging Deficits: Urban Water Supply in Sub-Saharan Africa." AICD Background Paper 12, World Bank, Washington, DC. Briceño-Garmendia, C., K. Smits, and V. Foster. 2008. "Financing Public Infrastructure in Sub-Saharan Africa: Patterns and Emerging Issues." AICD Background Paper 15, World Bank, Washington, DC. Vagliasindi, M., and J. Nellis 2009. "Evaluating Africa's Experience with Institutional Reforms for the Infrastructure Sectors." AICD Working Paper 23, World Bank, Washington, DC. xxiv Acknowledgments Chapter 6 Contributors Vivien Foster, Sudeshna Ghosh Banerjee, Tarik Chfadi, Amadou Diallo, Sarah Keener, Manuel Luengo, Taras Pushak, Heather Skilling, Clarence Tsimpo, Helal Uddin, Quentin Wodon, Yvonne Ying. Key Source Documents Banerjee, S., V. Foster, Y. Ying, H. Skilling, and Q. Wodon. 2008. "Cost Recovery, Equity and Efficiency in Water Tariffs: Evidence From African Utilities." AICD Working Paper 7, World Bank, Washington DC. Banerjee, S., Q. Wodon, A. Diallo, T. Pushak, H. Uddin, C. Tsimpo, and V. Foster. 2008. "Access, Affordability and Alternatives: Modern Infrastructure Services in Sub-Saharan Africa." AICD Background Paper 2, World Bank, Washington, DC. Keener, S., M. Luengo, and S. G. Banerjee. 2009. "Provision of Water to the Poor in Africa: Experience with Water Standposts and the Informal Water Sector." AICD Working Paper 13. World Bank, Washington, DC. Morella, E., V. Foster, and S. Banerjee. 2008. "Climbing the Ladder: The State of Sanitation in Sub-Saharan Africa." AICD Background Paper 13, World Bank, Washington, DC. Chapter 7 Contributors Elvira Morella, Africon. Key Source Documents Africon. 2008. "Unit Costs of Infrastructure Projects in Sub-Saharan Africa." AICD Background Paper 11, World Bank, Washington, DC. Chapter 8 Contributors Cecilia Briceño-Garmendia, Carolina Dominguez, Vivien Foster, Nataliya Pushak, Karlis Smits. Key Source Documents Briceño-Garmendia, C., K. Smits, and V. Foster. 2008. "Financing Public Infrastructure in Sub-Saharan Africa: Patterns and Emerging Issues." AICD Background Paper 15, World Bank, Washington, DC. Foster, Vivien, William Butterfield, Chuan Chen, and Nataliya Pushak. 2008. "Building Bridges: China's Growing Role as Infrastructure Acknowledgments xxv Financier for Sub-Saharan Africa." Trends and Policy Options 5, Public-Private Infrastructure Advisory Facility, World Bank, Washington, DC. Chapter 9 This chapter is a synthesis of our findings. None of this research would have been possible without the generous collaboration of government officials in the key sector institutions of each country, as well as the arduous work of local consultants who assembled this information in a standardized format. In addition, thanks are due to the staff of the Water and Sanitation Program (WSP), who have collabo- rated in the data collection process and rallied to support this endeavor in their respective countries by facilitating meetings, sharing documenta- tion, and completing data collection where needed. Dennis Mwanza, for- mer urban water supply thematic leader at WSP-Africa, has been an active partner in supporting this process from its inception. Piers Cross, Vivian Castro, Valentina Zuin, and Jean Doyen, also of WSP-Africa, have been invaluable partners in this effort, ensuring that messages are aligned with real-world knowledge of the sector and represent the on-the-ground realities in Africa. Finally, the project team is grateful to the team of local consultants, sourced from PricewaterhouseCoopers-Africa (PwC) and independent local consultants, who carried out this exercise in the field amid all the problems associated with gathering data at this scale. A list of these partners appears below. Afua Sarkodie ably led the PwC team and pro- vided oversight and support to the local consultants. Collaborating institutions (Water and Sanitation Program, Local consultants Country the World Bank) or other partners Benin Sylvain Migan Burkina Faso Seydou Traore, Christophe Prevost Cameroon Astrid Manroth Cape Verde Sandro de Brito Chad Yao Badjo Kenneth Simo Congo, Dem. Rep. Georges Kazad Henri Kabeya Côte d'Ivoire Emmanuel Diarra Eric Boa (continued next page) xxvi Acknowledgments Collaborating institutions (Water and Sanitation Program, Local consultants Country the World Bank) or other partners Ethiopia Belete Muluneh Yemarshet Yemaneh Mengistu Ghana Ventura Bengoechea Afua Sarkodie Kenya Dennis Mwanza, Japheth Mbuvi, Ayub Osman, Peter Njui Vivian Castro Lesotho Jane Walker Peter Ramsden Madagascar Christophe Prevost Gerald Razafinjato Malawi Midori Makino, Bob Roche Caroline Moyo Mozambique Jane Walker, Luiz Tavares, Carla Barros Costa Valentina Zuin Namibia Birgit de Lange, Peter Ramsden Niger Ibrah Sanoussi, Matar Fall Nigeria Joe Gadek, Hassan Kida Mohammed Iliyas Rwanda Bruno Mwanafunzi, Christophe Charles Uramutse Prevost Senegal Pierre Boulanger Ndongo Sene South Africa Peter Ramsden Sudan Solomon Alemu A. R. Mukhtar Tanzania Francis Ato Brown, Nat Paynter Kenneth Simo Uganda Samuel Mutono Zambia Barbara Senkwe Caroline Moyo The work benefited from widespread peer review by colleagues within the World Bank, notably Robert Roche, Alexander McPhail, Caroline van den Berg, Christophe Prevost, Ventura Bengoechea, Dennis Mwanza, and Meike van Ginneken. The external peer reviewer for this volume, Sophie Tremolet, provided constructive and thoughtful comments. The compre- hensive editorial effort of Steven Kennedy is much appreciated. Abbreviations ADAMA Nazareth Water Company, Ethiopia ADeM Águas de Moçambique AICD Africa Infrastructure Country Diagnostic AREQUAPCI association of water resellers AWSA Addis Ababa Water Services Authority, Ethiopia BOCC basket of construction components BWB Blantyre Water Board, Malawi CAPEX capital expenditure CFA Communauté Financière Africaine Franc CBO community-based organization CEMAC Central African Economic and Monetary Community COMESA Common Market for Eastern and Southern Africa CRWB Central Region Water Board, Malawi DAWASCO Dar es Salaam Water and Sewerage Company, Tanzania DBT direct block tariff DHS demographic and health survey DUWS Dodoma Urban Water and Sewerage Authority, Tanzania EAC East African Community ECOWAS Economic Community of West African States FCT Federal Capital Territory, Nigeria xxvii xxviii Abbreviations GDP gross domestic product GNI gross national income GRUMP Global Rural-Urban Mapping Project GWC Ghana Water Company HCI high conflict index IBNET International Benchmarking Network for Water and Sanitation Utilities IBT increasing block tariff ICP International Comparison Program IDA International Development Association IDAMC Internally Delegated Area Management Contract IFRS International Financial Reporting Standards JIRAMA Jiro sy Rano Malagasy, Madagascar JMP Joint Monitoring Programme KIWASCO Kisumu Water and Sewerage Company, Kenya LCI low conflict index LWB Lilongwe Water Board, Malawi LWSC Lusaka Water and Sewerage Company, Zambia MCI medium conflict index MDG Millennium Development Goal MICS multiple-indicator cluster survey MSNE Mauritania Société Nationale d'Eau et d'Electricité MWSA Mwanza Water and Sewerage Authority, Tanzania MWSC Mombasa Water and Sewerage Company, Kenya NGO nongovernmental organization NRW nonrevenue water NWASCO Nairobi Water and Sanitation Company, Kenya NWC National Water Company NWSC National Water and Sewerage Company, Uganda NWSC Nkana Water and Sewerage Company, Zambia O&M operations and maintenance ODA official development assistance OECD Organisation for Economic Co-operation and Development ONAS Office National de l'Assainissement du Sénégal ONEA Office Nationale des Eaux et d'Assainissement, Burkina Faso OPEX operating expenditure PPI private participation in infrastructure PPIAF Public-Private Infrastructure Advisory Facility Abbreviations xxix PPP purchasing power parity PSP private sector participation PwC PricewaterhouseCoopers-Africa REGIDESO Régie de Production et de Distribution d'Eau RUWATSSA State Rural Water Supply and Sanitation Agency, Nigeria SADC Southern African Development Community SDE Sénégalaise des Eaux SEEG Société d'Electricité et d'Eaux du Gabon SEEN Société de Exploitation des Eaux du Niger SNEC Société National des Eaux du Cameroon SODECI Société de Distribution d'Eau de Côte d'Ivoire SOE state-owned enterprise SONEB Société Nationale des Eaux du Benin SPEN Société de Patrimoine des Eaux du Niger STEE Société Tchadienne d'Eau et d'Electricité, Chad SWSC Southern Water and Sewerage Company, Zambia TdE Togolaise des Eaux UNICEF United Nations Children's Fund VIP ventilated improved pit WASA Water and Sanitation Authority, Lesotho WB Water Board WHO World Health Organization WSP Water and Sanitation Program WSP-SA Water and Sanitation Program­South Asia WSS water supply and sanitation WUC Water Utilities Corporation, Botswana CHAPTER 1 The Elusiveness of the Millennium Development Goals for Water and Sanitation The welfare implications of safe water cannot be overstated. Infectious diarrhea and other serious waterborne illnesses are leading causes of infant mortality and malnutrition. Their impact extends beyond health to the economic realm in the form of lost work days and school absen- teeism. It is estimated that meeting the Millennium Development Goal (MDG) for access to safe water1 would produce an economic benefit of US$3.1 billion (in 2000 dollars) in Africa, a gain realized by a combina- tion of time savings and health benefits. The cost-benefit ratio is about 11, which suggests that the benefits derived from access to safe water are far greater than the costs of providing it (Hutton and Haller 2004). Similarly, sanitation makes a key contribution to public health, par- ticularly in densely populated areas. Adequate sanitation is defined as any private or shared, but not public, facility that guarantees that waste is hygienically separated from human contact (JMP 2000). Adequate sanitation reduces the risk of a broad range of diseases--including res- piratory ailments, malaria, and diarrhea--and reduces the prevalence of malnutrition. Access to this standard of sanitation produces direct health gains by preventing disease and delivering economic and social benefits. It is estimated that a reduction in diarrheal illness would pro- duce a gain of 99 million days of school and 456 million days of work 1 2 Africa's Water and Sanitation Infrastructure for the working population ages 15­59 in Africa. The workdays alone rep- resent economic benefits equal to as much as US$116 million (Hutton and Haller 2004). The international adoption of the MDGs in 2000 created a new framework for focusing poverty reduction efforts on the indicators that are most meaningful for economic development. The MDGs have called attention to deficiencies in the quantity and quality of water supply and sanitation (WSS). MDG 7 calls for ensuring environmental sustainability and--relevant to this book--reducing by half the number of people without sustainable access to safe drinking water and improved sanitation. Although the world overall is on track to meet the MDG drinking water target, Africa lags. The gap is most acute in Sub-Saharan Africa, where only 58 percent of the population enjoys access to safe drinking water, and the gap is widening, as the increasingly urban popu- lation places a greater strain on existing service providers (table 1.1). Of the 828 million people in the world whose water sources remain unimproved, 37 percent live in Sub-Saharan Africa. According to projec- tions, 300 million people--almost 38 percent of Sub-Saharan Africa's population, or half the number of people who presently have access to improved water--will need to be covered to meet the MDG target (JMP 2008). The world is not on track to meet the MDG sanitation target. More than 2.5 billion people remain without improved sanitation worldwide; of that total, 22 percent, or more than half a billion people, live in Africa. A reported 221 million people in Africa still defecate in the open, the second- largest total for any region after South Asia. Access to improved sanitation Table 1.1 Regional Progress toward the MDG Drinking Water Target Drinking water Coverage coverage (%) needed to be on MDG target Region 1996 2006 track in 2006 (%) coverage (%) Progress Sub-Saharan Africa 49 58 65 75 Off track North Africa 88 92 92 94 On track Latin America and the Caribbean 84 92 89 92 On track East Asia 68 88 78 84 On track South Asia 74 87 82 87 On track Southeast Asia 73 86 82 87 On track West Asia 86 90 90 93 On track Source: JMP 2008. The Elusiveness of the Millennium Development Goals for Water and Sanitation 3 Table 1.2 Regional Progress toward the MDG Sanitation Target Sanitation Coverage coverage (%) needed to be on MDG target Region 1990 2006 track in 2006 (%) coverage (%) Progress West Asia 79 84 86 90 On track Latin America and the Caribbean 68 79 78 84 On track North Africa 62 76 74 81 On track Southeast Asia 50 67 64 75 On track East Asia 48 65 65 74 On track South Asia 21 33 46 61 Off track Sub-Saharan Africa 26 31 50 63 Off track World 54 62 69 77 Off track Source: JMP 2008. has increased only modestly in Sub-Saharan Africa, from 26 percent of the total population in 1990 to 31 percent in 2006. To be on track with the MDG's sanitation benchmark, improved sanitation coverage should have been at 50 percent of the population in 2006. To meet the MDG sanitation target, the current number of people with improved sanita- tion in Africa needs to more than double, from 242 million in 2006 to 615 million in 2015. Unless the current trend changes, Sub-Saharan Africa will definitely not meet the sanitation target (table 1.2). A Timely Synthesis With only five years remaining until the MDG deadline in 2015, it is essential to take stock of the status of the WSS sectors, analyze their achievements and shortcomings in Sub-Saharan Africa, and identify the sector characteristics that either advance or inhibit the population's ability to access service. Governments have adopted WSS reforms and attracted investments to build dynamism in the sectors and to enhance performance outcomes. These initiatives have been critical to developing implementation capacity and to establishing innovative forms of serv- ice delivery. Building on background work carried out under the auspices of the Africa Infrastructure Country Diagnostic (AICD) and presented by Foster and Briceño-Garmendia (2009), this volume integrates a wealth of primary and secondary information to present a quantitative snapshot of the state of the WSS sectors in Africa, including the current status of 4 Africa's Water and Sanitation Infrastructure access and coverage trends. It explains institutional and governance struc- tures and utility performance and articulates the volume and quality of financing available over time for WSS. The volume also evaluates the chal- lenges to the WSS sectors and explores the factors that might explain the expansion of coverage. Finally, it endeavors to estimate spending needs for WSS, with a target of meeting the MDG goal, and compares those needs with the existing financing envelopes, disaggregated into shares that can be recouped through efficiency improvements and gaps that would remain even if all feasible efficiencies were achieved. The directions for the future draw on lessons learned from experiences around the continent and present the menu of choices available to African countries. Data Sources and Methodologies Monitoring the progress of infrastructure sectors such as water supply has been a significant by-product of the MDGs, and serious attention and funding have been devoted in recent years to developing monitoring and evaluation systems in countries around the world. The Joint Monitoring Programme (JMP) on WSS is an institutional endeavor by the World Health Organization and the United Nations Children's Fund to system- atically track progress toward the WSS MDG. The JMP's monitoring introduced the concept of improved and unimproved WSS and catego- rized WSS sources according to the typology shown in table 1.3. The JMP and AICD Methodologies AICD used a body of household surveys similar to that of the JMP-- demographic and health surveys (DHSs), multiple-indicator cluster surveys (MICSs), and income/expenditure surveys--but the JMP has adopted special rules for use when the exact disaggregation across WSS modalities is not available in the surveys. Those rules apply most often to the largest sources of WSS, namely, wells or boreholes and traditional pit latrines. The JMP statistics apportion 50 percent of wells or boreholes to the protected or "improved" category and the remainder to the unpro- tected or "unimproved" category. Similarly, covered pit latrines are placed in the "improved" category, and the unprotected in the "unimproved" cat- egory. In the AICD analysis, the information available in the survey has been taken at face value without any adjustment. Therefore, only the household connections to piped water and piped water delivered through public standposts constitute the "improved water" category, and flush toilets and ventilated improved pit (VIP) latrines are included in the The Elusiveness of the Millennium Development Goals for Water and Sanitation 5 Table 1.3 Definition of Coverage of Improved Water JMP category AICD category Primary source of water supply Piped water into dwelling or yard Improved Improved Public tap or communal standpipe, Improved Improved standposts, or kiosks Wells or boreholes, hand pumps, or Improved/unimproved Unimproved rainwater Surface water (for example, lake, river, Unimproved Unimproved pond, dam, or spring) Vendors or tanker trucks Unimproved Unimproved Other (for example, bottled water) Unimproved Unimproved Primary source of sanitation Flush toilet to network or septic tank Improved Improved VIP latrine, SanPlat, or basic pits with slab Improved Improved Traditional pit latrine Improved/unimproved Unimproved Bucket or other container Unimproved Unimproved Other Unimproved Unimproved No facility (nature or bush) Unimproved Unimproved Source: Banerjee, Wodon, and others 2008. Note: VIP = ventilated improved pit. "improved sanitation" category. Further, the DHSs describe access to san- itation without discriminating between on-site sanitation and use of sew- erage facilities, so that both are included in the flush toilet category. Most of these flush toilets, however, use septic tanks rather than sewer connec- tions. For this reason, this study assumes that the DHS information relat- ing to flush toilets refers to septic tanks. Owing to these methodological differences, the JMP and AICD figures differ on improved water and improved sanitation. Not surprisingly, the differences are more pronounced in rural areas, where wells/boreholes and traditional pit latrines are the most prevalent forms of WSS sources (figure 1.1). In this volume, we focus above all on what lies within each of the improved and unimproved categories, rather than on the aggre- gates. Further, JMP uses methodologies that usually differ from method- ologies used by each country to evaluate coverage. In most cases, national statistics would show higher coverage figures than does JMP. Sanitation can be provided on numerous distinct levels that can be graphically represented as rungs on a ladder. Starting from open defeca- tion, the successive increments are traditional latrines (various kinds of pits), improved latrines (including SanPlat, VIP latrines, and basic pits 6 Africa's Water and Sanitation Infrastructure Figure 1.1 JMP and AICD Estimates of the Prevalence of "Improved" Water Supply and Sanitation a. Water supply b. Sanitation 90 45 80 40 percentage of population percentage of population 70 35 60 30 50 25 40 20 30 15 20 10 10 5 0 0 JMP AICD JMP AICD national rural urban Sources: Banerjee, Wodon, and others 2008; JMP 2008. with slabs), and flush toilets (connected to either a septic tank or a water- borne sewage network). The higher rungs of the ladder carry higher unit costs and lower levels of perceived health risk (figure 1.2). This concept carries over to water, but not as clearly, because the sources cannot be ranked on the basis of quality or cost. It is evident, however, that surface water represents the bottom rung, and household connections to piped water and piped water delivered through public standposts are at the upper end of the ladder. What comes out very clearly in the literature is that the distance to the water source makes a substantial difference to health outcomes and time savings. Data Sources The analysis presented in this book is based on three primary databases that underlie three AICD background papers--Banerjee, Skilling, and others (2008), Banerjee, Wodon, and others (2008), and Morella, Foster, and Banerjee (2008). These background papers are referred to through- out this volume. Household surveys: AICD DHS/MICS Database and Expenditure Database. The results from household surveys are used extensively in chapters 2 and 3. The first, the AICD DHS/MICS database, was used to analyze the current status and access trends presented in this volume; it is a composite of 63 DHS and MICS data sets. Thirty countries in Africa The Elusiveness of the Millennium Development Goals for Water and Sanitation 7 Figure 1.2 The Sanitation Ladder cost flush toilet/septic tank improved latrines (VIP, chemical, SanPlat) not acceptable traditional latrine Open defecation Fixed-place defecation Source: Authors. Note: VIP = ventilated improved pit. have had at least one DHS conducted since 1990, and 24 have at least two DHS data points between 1990 and 2005, which enables trend analysis. Second, the AICD expenditure survey database includes the most recent household-level expenditure surveys for 30 African countries during the period from 1997 to 2005. This database incorporates sur- veys modeled after the Living Standards Measurement Surveys. These surveys provide a wealth of information on use of and payment for infra- structure services, as well as offering data on household assets and expen- diture patterns. Known by different names in different countries, these surveys are carried out by country governments to reflect local nuances and priorities. Therefore, their infrastructure modules often are not harmo- nized or comparable, and standardization techniques have been employed to permit continentwide inferences (annexes 1.1 and 1.2). AICD Water Supply and Sanitation Survey. This survey was carried out in two phases and administered to line ministries, sector institutions, and water utilities with a view to capturing institutional and performance variables. Seven modules of data were collected for each country, of which five are qualitative and two are quantitative. The focus of each module is reflected in table 1.4. The data were collected in two phases (2007 and 2009) and from two distinct sources (AICD and the International Benchmarking Network for Water and Sanitation Utilities [IBNET]). AICD's data collection in 24 countries in 2007 resulted in a comprehensive data set covering 51 utilities. AICD's 2009 flagship report (Foster and Briceño-Garmendia 2009) was 8 Table 1.4 Modules of AICD WSS Survey Data collection Data collection Module Description unit source and coverage Topics in questionnaire Module 1: Institutional Qualitative Country AICD Phase I Legal framework, sector organization, regulatory framework, and regulatory regulatory process, tariff adjustment, private participation Module 2: Rural water Qualitative Country AICD Phase I Sector organization, service characteristics Module 3: Governance Qualitative Utility AICD Phase I Ownership, board structure, performance contract, performance monitoring and disclosure, finance, labor Module 4: Sanitation Qualitative Country AICD Phase I Sector organization, service characteristics Module 5: Small-scale Qualitative Largest city AICD Phase I Point sources, mobile sources independent providers Module 6: Operational Quantitative Utility AICD Phase I, II, IBNET Access, quality of service, operational performance, financial and financial performance Module 7: Tariff schedules Quantitative Utility AICD Phase I Currently effective tariff schedule Source: Authors. Note: IBNET = International Benchmarking Network for Water and Sanitation Utilities. The Elusiveness of the Millennium Development Goals for Water and Sanitation 9 based on this data-collection effort. In 2009, AICD carried out a second round of data collection in three more countries and covering three additional utilities, but this information was restricted to operational and financial performance only (module 6). The AICD data set was integrated with that of IBNET, which collected operational and finan- cial performance data (module 6) for 32 more utilities. The upper bound of the data set covers 32 countries and 86 utilities; the lower bound covers 24 countries and 51 utilities. Different modules underpin the individual chapters in this volume. For instance, chapter 2 draws on modules 2 and 5 to elaborate on the current state of the formal, informal, and rural water markets. Chapter 3 employs the questions in module 4 to present the sanitation snapshot. Chapter 4 draws on modules 1 and 3, which contain questions detailing the institu- tional environment of the WSS sectors. Quantitative data were captured to develop an understanding of the financial, technical, and operational performance of the selected utilities (module 6). Utilities were asked to provide data for the 10-year period from 1995 to 2005, but because older data were rarely available, the emphasis shifted to collecting data from the five-year period from 2000 to 2005. Chapter 5 is based on the oper- ational and financial time-series data on utility performance contained in module 6. The information presented in tariff schedules (module 7) is used in chapters 5 and 6. AICD fiscal database. The country-level analysis of the volumes, pat- terns, and composition of financial resources for WSS draws on the AICD fiscal database used extensively in chapter 8. That database, which cap- tures information on public spending in the infrastructure sectors of 25 countries, is a unique attempt to document in a standardized manner the levels and patterns of public spending for infrastructure, including WSS. If one uses the database, it is possible to comparisons across sectors and ensure consistency over time. Financing flows within public spending are defined as including tax revenue or user charges channeled through both on-budget (central and local governments) and off-budget mecha- nisms (state-owned enterprises and special funds). Country Categories The performance of the utilities is evaluated across various functional and financial dimensions and presented in selected country groupings consistent with the method used in Foster and Briceño-Garmendia (2009), described in annex 1.3. The country groupings are based on (1) income and fragility (middle-income, low-income fragile, low-income, 10 Africa's Water and Sanitation Infrastructure Table 1.5 Utilities Analyzed in This Report, by Categories IBNET AICD Total Sub-Saharan Africa 32 54 86 Income group Low-income, fragile 1 2 3 Low-income, nonfragile 20 26 46 Middle-income 1 11 12 Resource-rich 10 15 25 Regional economic community CEMAC 1 3 4 COMESA 10 19 29 EAC 17 8 ECOWAS 3 14 SADC 27 26 Water availability High water scarcity 2 30 32 Low water scarcity 30 24 54 Source: Authors. Note: CEMAC = Central African Economic and Monetary Community; COMESA = Common Market for Eastern and Southern Africa; EAC = East African Community; ECOWAS = Economic Community of West African States; SADC = Southern African Development Community. and resource-rich), (2) water scarcity (high, low), and (3) regional eco- nomic community (EAC, ECOWAS, CEMAC, COMESA, and SADC) (table 1.5). The utilities are further distinguished by size (small, large). Key Finding 1: Wide Differences in Patterns of Access to Water In rural areas, reliance on surface water remains prevalent, and bore- holes are the principal improved source of water. The share of the population relying on surface water fell sharply in the 1990s, from 50 percent to just more than 40 percent, where it has remained for the past five years (table 1.6). Boreholes are the main source of improved water, accounting for a further 40 percent of the population. Access to piped water and standposts is very low, barely increasing over the period 1990­2005. Indeed, in many countries, less than 1 percent of the rural population receives piped water. It is striking that in more urbanized countries, access to piped water and standposts in rural areas is substantially higher. In urban areas, coverage of piped water fell markedly over the period 1990­2005 owing to rapid population growth. At close to 40 percent, however, it is still the single largest source of urban water. Coverage of The Elusiveness of the Millennium Development Goals for Water and Sanitation 11 Table 1.6 Evolution of Water Supply Coverage in Africa, by Source (percent) Well and Piped supply Standposts boreholes Surface water Period Urban Rural Urban Rural Urban Rural Urban Rural 1990­95 50 4 29 9 20 41 6 50 1995­2000 43 4 25 9 21 41 5 41 2001­05 39 4 24 11 24 43 7 42 Source: Banerjee, Wodon, and others 2008. standposts saw a similar decline, but that of boreholes rose, so that each represented about 24 percent of the urban population in 2005. Overall, about two-thirds of the urban populace depends on utility water. The lower coverage of standposts compared with piped water is particularly striking, given the relatively low cost of standposts and the pressure to expand services rapidly. Reliance on surface water, at 7 percent of the urban population, changed little between 1990 and 2005. Utilities are the central actors responsible for water supply in urban areas. In the middle-income countries they are essentially the only play- ers, reaching about 98 percent of the urban population, the vast majority through private piped-water connections. In low-income countries only 68 percent of urban residents benefit from utility water, fewer than half through private piped connections (table 1.7). For the rest, informal shar- ing of connections through resale between neighbors (15 percent of the urban population) is almost as prevalent as formal sharing through stand- posts (19 percent of the urban population). Utilities report providing about 20 hours per day of service (table 1.8). They typically produce just more than 200 liters per customer served, though the amount for middle-income countries is about twice that for low-income countries. If the total water production of the utilities could be evenly distributed to the entire population residing in the utility ser- vice area, it would amount to 74 liters per capita a day, just about ade- quate to meet basic human needs. Urban households that do not benefit from utility water rely on sev- eral alternatives. The rapid expansion of boreholes in urban areas has already been noted. Water vendors, another alternative, may sell water obtained from utilities, boreholes, or surface sources from either trucks and carts or, less frequently, through private distribution networks. Water vendors account for only 3 percent of the African urban market, 12 Africa's Water and Sanitation Infrastructure Table 1.7 Services Provided by Utilities in Their Service Areas (percent) Access by Access by private sharing Access to residential neighbors' utility water piped-water Access by private by some connection standpost connection modality Sub-Saharan Africa 44.3 13.0 21.7 64.0 Low-income countries 42.2 23.2 22.5 68.6 Low-income, fragile countries 25.6 2.2 41.0 56.0 Resource-rich countries 30.3 15.8 7.4 48.8 Middle-income countries 88.0 9.7 0.3 97.8 Source: Banerjee, Skilling, and others 2008. Table 1.8 Quality of Services Provided by Utilities in Their Service Areas Availability of water Quality of supply Water production Water production per customer Continuity per resident in served by utility Samples of water the utility service in service area passing service area (liters per (liters per capita chlorine test (hours capita per day) per day) (%) per day) Sub-Saharan Africa 116.4 162.9 87.9 19.6 Low-income countries 66.0 130.2 92.8 19.0 Low-income, fragile countries 35.7 76.5 75.3 18.2 Resource-rich countries 140.5 208.8 78.1 18.4 Middle-income countries 208.8 233.6 97.2 24.0 Source: Banerjee, Skilling, and others 2008. rising to 7 percent in West Africa. In some countries, however, their con- tribution to urban water supply is much larger: Nigeria (10 percent), Chad (16 percent), Niger (21 percent), and Mauritania (32 percent). In 15 large cities in Africa, the cost of vendor water, particularly when transported directly to the household, can be 2­11 times more expen- sive than having a household connection (table 1.9). This high willing- ness to pay for vendor water is a potential revenue source that the utilities are typically unable to capture. Wells and boreholes are by far the fastest-growing source of improved water in both urban and rural areas. Service expansion shows a similar overall pattern in both cases: The absolute number of people depending The Elusiveness of the Millennium Development Goals for Water and Sanitation 13 Table 1.9 Average Price for Water Service in 15 Largest Cities, by Type of Provider House Small piped Household Water Water connection network Standpost reseller tanker vendor Average price (US$ per cubic meter) 0.49 1.04 1.93 1.63 4.67 4.00 Markup over house connection (%) 100 214 336 402 1,103 811 Source: Keener, Luengo, and Banerjee 2009. on surface water continues to grow, a grim statistic in its own right (figure 1.3). Across the board, wells and boreholes are expanding coverage much more rapidly than all the utility-based alternatives put together. Within the purview of the utility, access to standposts seems to be grow- ing faster than piped water. However, the combined growth rates of the various improved forms of water in urban areas (less than 1 percent a year) still fall short of population growth (more than 4 percent a year). Access to improved water sources is highly inequitable across the income distribution (figure 1.4). Access to piped water and standposts is heavily concentrated among the more affluent segments of the popula- tion, typically in urban areas. The poorest 40 percent of the population, by contrast, depends on surface water and on wells and boreholes in almost equal measure. Only 10 percent of African households in the bot- tom 60 percent of population are covered by piped supply. For the middle- income countries, access to piped water and standposts among the poorest quintiles is substantially higher than in the low-income countries. Key Finding 2: Equally Wide Differences in Patterns of Access to Sanitation Traditional pit latrines are by far the most common facility in both urban and rural areas, but more than a third of the population--mostly in rural areas--still defecates in the open (table 1.10). Improved sanitation (sep- tic tanks and improved latrines) reaches less than 20 percent of Africa's population, and less than 10 percent in rural areas. Coverage of improved latrines is no greater than that of septic tanks, despite the significant cost difference between them. Only 10 percent of the population uses a sep- tic tank; coverage in rural areas is practically negligible. In urban areas, septic tanks are much more common than improved latrines, and less than 10 percent of the population practices open defecation. 14 Africa's Water and Sanitation Infrastructure Figure 1.3 Dependence on Surface Water in Urban and Rural Areas, 1990s versus Early 2000s a. Urban 2.0 annualized increase in access (% population) 1.5 1.0 0.5 0.0 ly s s er st le pp at o ho w dp su re ce an d bo rfa pe st ls/ pi su el w b. Rural 2.0 annualized increase in access (% population) 1.5 1.0 0.5 0.0 ­0.5 ly ts s er le pp s at po ho w su re d ce an d bo rfa pe st ls/ pi su el w Source: Banerjee, Wodon, and others 2008. Waterborne sewerage systems are rare in Africa. Only half of the large cities operate a sewerage network at all, and only in Namibia, South Africa, and the exceptional case of Senegal do some of the utilities cov- ering the largest cities provide universal sewerage coverage. Little more than half of the households with piped water also have flush toilets, which are often connected to septic tanks rather than to sewers. Patterns of access to sanitation vary dramatically across income groups. Open defecation is widely practiced in the lowest income quintile and not practiced at all in the highest. Conversely, improved latrines and septic tanks, virtually nonexistent among the poorest quintiles, are used by only 20­30 percent of the population in the The Elusiveness of the Millennium Development Goals for Water and Sanitation 15 Figure 1.4 Coverage of Water Services, by Income Quintile 60 50 % population 40 30 20 10 0 poorest second third fourth richest quintile household piped water standposts wells/boreholes surface water Source: Banerjee, Wodon, and others 2008. Table 1.10 Patterns of Access to Sanitation in Africa (percentage of population) Area Open defecation Traditional latrine Improved latrine Septic tank Urban 8 51 14 25 Rural 41 51 5 2 National 34 52 9 10 Source: Banerjee, Wodon, and others 2008. richest. Access to improved latrines parallels that of septic tanks, suggesting that despite their lower cost, improved latrines remain something of a luxury, with little success in penetrating the middle of the income distribution. More important, the minimal presence of improved sanitation across poorer groups highlights a crucial issue-- that high average rates of coverage do not help the most vulnerable populations. Traditional latrines are by far the most egalitarian form of sanitation, accounting across income ranges for about 50 percent of households (figure 1.5). Traditional latrines are not only are the most common form of sanita- tion in Africa, but they are also the fastest growing. In recent years they have been used by an additional 2.8 percent of the population each year in urban areas and an additional 1.8 percent in rural areas, more 16 Africa's Water and Sanitation Infrastructure Figure 1.5 Coverage of Sanitation Services, by Income Quintile 100 80 % population 60 40 20 0 poorest second third fourth richest quintile septic tank improved latrine traditional latrine open defecation Source: Morella, Foster, and Banerjee 2008. than twice the rate of expansion of septic tanks and improved latrines combined (figure 1.6). Growth in the use of traditional latrines is concen- trated among the poorer quintiles and of improved latrines and septic tanks among the richer quintiles. Because the MDG target focuses on the two most improved sanitation options, the expanding use of traditional latrines does not always fully register in policy discussions. Meanwhile, the prevalence of open defecation in Africa has finally begun to decline, albeit at a very modest pace. Key Finding 3: High Costs, High Tariffs, and Regressive Subsidies African water utilities operate in an environment of high costs, with two-thirds of the utilities operating in 2005 within the cost band of $0.4 to $0.8/m3. Since then, costs have continued to rise in nominal terms. The high average cost of operations and maintenance (O&M) in Africa is somewhat misleading, driven as it is by the high cost of pro- viding services in the middle-income countries of South Africa and Namibia, which is more than $1, because it includes the cost of purchas- ing bulk water. Overall, Africa's experience in recovering operating costs is positive, with many utilities setting tariffs at levels high enough to recoup O&M costs. In fact, African tariffs are highest among the devel- oping regions, with the operating ratio very close to 1 mainly because utilities spend everything they collect and nothing over that. Thus, they are not adequately funding either capital expenditures or rehabilitation or maintenance. The Elusiveness of the Millennium Development Goals for Water and Sanitation 17 Figure 1.6 Annual Growth in the Use of Sanitation Types, 1990­2005 (percent) a. Urban areas 3.0 2.5 annual change in coverage 2.0 (% population) 1.5 1.0 0.5 0.0 ­0.5 ne ne n nk tio ta tri tri ca la l la ic fe pt ed na de se ov tio en pr di op im tra b. Rural areas 3.0 2.5 annual change in coverage 2.0 (% population) 1.5 1.0 0.5 0.0 ­0.5 e ne n nk rin tio ta tri t ca la l la ic fe pt d na de e se ov tio en pr di op im tra Source: Morella, Foster, and Banerjee 2008. Full cost recovery is far off. Only four utilities in middle-income coun- tries achieve their capital cost recovery at an average level of consump- tion of 10 m3/month. It is only in the last block of the increasing block tariff structure that prices are set with an eye to cost recovery because of the widespread perception that recouping capital costs from consumers is not feasible because of the limited budgets of African households. 18 Africa's Water and Sanitation Infrastructure In most countries of the region, utilities' capital costs have been almost entirely subsidized by the state or by donors, but the subsidies are highly regressive, especially those to residential consumers in urban areas. Across the bottom half of the income distribution, barely 10 percent of households have access to piped water. Indeed, more than 80 percent of households with piped water come from the top two quintiles of the income distribution. Because poorer households are almost entirely excluded, they cannot benefit from subsidies embedded in prices for piped water. In many cases, targeting performance is fur- ther exacerbated by poor tariff design, with widespread use of mini- mum charges and rising block tariffs that provide large blocks of highly subsidized water to all consumers. Tariffs high enough to provide full capital cost recovery should be affordable for half of the population in Africa--and for about 40 percent of the population in low-income countries--but not for the remainder. Assuming household average consumption of 10 cubic meters a month (or about 65 liters per capita a day), a monthly utility bill under full-cost- recovery pricing of $1 would be about $10. Based on an affordability threshold of 5 percent of household income, full-cost-recovery tariffs would prove affordable for 40 percent of the population in low-income countries (figure 1.7). With about 10 percent of the national population already enjoying a direct water connection, an additional 30 percent of the population could be connected to water service and be able to pay for it. Most of the remaining 60 percent of the population would be able to afford bills of about $6 a month. Key Finding 4: The Stark Challenge of Financing the MDG The overall price tag for reaching the MDG target for access to WSS is estimated at $22.6 billion per year, or 3.5 percent of Africa's gross domestic product. Most of that sum is related to the water sector, which is estimated to require allocations up to $17 billion per year, or 2.7 percent of Africa's gross domestic product (GDP) (table 1.11). The cost of new infrastructure is the largest share, requiring allocations of up to 1.5 percent of Africa's GDP every year, or 43 percent of overall spending. O&M needs are the next largest category, standing at 1.1 percent of Africa's GDP, or 31 percent of overall costs. Rehabilitation of existing assets requires lower yet substantial allocations--up to 0.9 percent of Africa's GDP--accounting for one-fourth of the overall needs. The Elusiveness of the Millennium Development Goals for Water and Sanitation 19 Figure 1.7 Affordability of Full-Cost-Recovery Tariffs in Low-Income Countries % households able to spend more than 5% of household budget 120 100 80 60 40 20 0 2 3 4 6 8 10 12 14 16 US$/month cost-of-recovery bill, low bound cost-of-recovery bill, medium bound cost-of-recovery bill, upper bound low-income countries Source: Adapted from Banerjee, Wodon, and others 2008. The composition of spending needs differs between middle- and low- income countries (table 1.12). Low-income countries (fragile or nonfrag- ile) and resource-rich countries show much similarity, with costs divided almost equally among expansion, rehabilitation, and maintenance. Conversely, middle-income countries focus more on maintenance, which accounts for half of the overall spending needs, but the high coverage rates and relatively lower rehabilitation backlog make infrastructure expansion and rehabilitation less of a priority. The affordability of meeting the MDG challenge appears to correlate strongly with a country's income. Halving the share of the population that lacks access to WSS services by 2015 is estimated to require only 1.5 percent of middle-income countries' GDP per year. Resource-rich countries would have to invest twice as much annually--3 percent of their GDP. The bill becomes prohibitively expensive for low-income countries, which would have to allocate at least 7 percent of their GDP to WSS every year to meet the goal. The burden would be even higher for fragile states: almost 12 percent of GDP each year. As of 2005, Sub-Saharan Africa spends about $7.9 billion a year (1.2 percent of the region's GDP) on WSS--about a third of what is required if the MDG is to be met. In absolute terms, spending levels vary significantly across the country groups (table 1.13): Middle-income coun- tries spend $2.6 billion, followed by low-income countries ($1.8 billion), 20 Table 1.11 Overall WSS Spending Needs Share of GDP (%) $ million per year CAPEX CAPEX Total Total Total Total Expansion Rehabilitation CAPEX O&M needs Expansion Rehabilitation CAPEX O&M needs Water 1.13 0.68 1.80 0.89 2.69 7,225 4,327 11,553 5,686 17,239 Sanitation 0.41 0.21 0.62 0.22 0.84 2,617 1,352 3,969 1,432 5,401 Total 1.54 0.89 2.42 1.11 3.53 9,843 5,679 15,522 7,118 22,640 Source: Authors' calculations. Note: CAPEX = capital expenditure. Table 1.12 Breakdown of Spending Needed to Meet MDGs in WSS, by Spending Category and Country Group Share of GDP (%) $ million per year CAPEX Total CAPEX Total New Total spending New Total spending investment Rehabilitation CAPEX O&M needs investment Rehabilitation CAPEX O&M needs Sub-Saharan Africa 1.5 0.9 2.4 1.1 3.5 9,843 5,679 15,522 7,118 22,640 Resource-rich countries 1.3 0.8 2.1 0.8 2.9 2,864 1,741 4,605 1,759 6,364 Middle-income countries 0.4 0.4 0.7 0.7 1.5 1,034 951 1,985 1,991 3,976 Low-income, fragile countries 5.9 2.7 8.5 3.3 11.8 2,208 1,006 3,213 1,223 4,437 Low-income, nonfragile countries 3.4 1.8 5.1 1.9 7.1 3,714 1,968 5,682 2,128 7,810 Source: Authors' calculations. Note: CAPEX = capital expenditure. 21 22 Africa's Water and Sanitation Infrastructure Table 1.13 Spending by Functional Category, Annualized Average Flows, 2001­05 Share of GDP (%) $ million per year Total Total Total Total O&M CAPEX spending O&M CAPEX spending Sub-Saharan Africa 0.5 0.7 1.2 3,112 4,778 7,890 Low-income, fragile countries 0.3 0.8 1.1 128 313 441 Low-income, nonfragile countries 0.3 1.4 1.7 307 1,533 1,840 Middle-income countries 0.7 0.2 1.0 1,996 641 2,637 Resource-rich countries 0.1 0.7 0.8 188 1,564 1,753 Sources: Briceño-Garmendia, Smits, and Foster 2008 for public spending; PPIAF 2008 for private flows; Foster and others 2008 for financiers from outside the Organisation for Economic Co-operation and Development. Note: CAPEX = capital expenditure. and resource-rich countries ($1.7 billion); fragile states spend about $0.5 billion in capital investment and O&M. Expressed as a percentage of GDP, infrastructure spending fluctuates widely across different country groups. Low-income countries and fragile states spend 1.1 and 1.7 percent of their GDP, respectively, whereas middle-income coun- tries and resource-rich countries spend 1 percent or less of theirs (1.0 and 0.8 percent, respectively). The composition of spending also varies substantially across country groups. Middle-income countries allocate 80 percent of WSS spending to maintenance, reflecting the fact that they have already built much of the infrastructure needed. By contrast, the other country groups allocate no more than 30 percent to this item. Therefore, resource-rich countries, low-income countries, and fragile states spend 70 to 90 percent of their budgets for WSS infrastructure on capital investments. Although this reflects their need to build new facilities, a danger looms of neglecting the maintenance needs of the limited network that is available. Inefficiencies of various kinds (incomplete execution of budgets, oper- ational inefficiencies, and underpricing) total an estimated $2.9 billion a year (0.5 percent of GDP). Eliminating those efficiencies would provide a large share of the additional funds needed to achieve the MDG. Even if all the efficiency gains were realized, however, a funding gap would remain. Subtracting existing spending and potential efficiency gains from the spending needed to reach the MDG leaves an annual funding gap of about $11.9 billion a year, or 1.8 percent of GDP (table 1.14). Although the gap is widest for capital investment ($8.6 billion), a large shortfall also exists for O&M. Table 1.14 Funding Gap ($ million per year) Sources of inefficiency Spending Gain from traced eliminating Underexecution Operating (Funding gap) Total needs to needs inefficiencies of budget inefficiencies Underpricing or surplus Sub-Saharan Africa ­22,640 7,890 2,877 168 1,259 1,450 ­11,873 Low-income, fragile countries ­4,531 441 471 6 106 358 ­3,620 Low-income, nonfragile countries ­7,810 1,840 685 39 265 381 ­5,285 Middle-income countries ­3,987 2,637 1,037 8 492 537 ­312 Resource-rich countries ­6,364 1,753 522 137 172 214 ­4,089 Source: Briceño-Garmendia, Smits, and Foster 2008. 23 24 Africa's Water and Sanitation Infrastructure The smallest funding gap is found in middle-income countries, where inefficiencies are greatest. After tackling the inefficiencies, middle- income countries would face a negligible funding gap of $0.3 billion, most of which could be realized by reallocating resources from O&M to capital expenditure or from another infrastructure sector. The largest funding gap remains in low-income countries (nonfragile), which account for about half of the total funding gap for Sub-Saharan Africa ($5.3 billion). In the aggregate, the region needs to increase capital investment in water infrastructure by 1.3 percent of GDP. Low-income, nonfragile countries need to invest an additional 3.3 percent and fragile states an additional 6.8 percent. Key Finding 5: Institutional Reform for Better Water Sector Performance Many African governments have reformed their WSS systems in the past two decades to provide better services for their citizens. Countries that have pursued institutional reforms have built more efficient and effective sector institutions and achieved faster expansion of higher quality ser- vices. The potential dividend of such efforts is large, because addressing utility inefficiencies alone could make a substantial contribution to clos- ing the sector funding gap in many countries. Utilities that have decen- tralized their WSS services or adopted private sector management have done a better job of eliminating inefficiencies and other hidden costs than those that have not. Unbundling of services can also be beneficial, but unbundling is rare in Africa and exclusively concentrated in middle- income countries, whose superior performance can be explained for many other reasons. At the same time, higher levels of regulation and bet- ter governance of utilities (often accompanied by corporatization) are associated with lower efficiency (figure 1.8). The reform agenda has had two major thrusts: increasing private par- ticipation and improving governance from within. Private sector participation has helped to improve utility performance, with Senegal being particularly noteworthy. Management contracts awarded to private operating companies, being relatively short-term instruments, have had a material effect on improving revenue collection and service continuity, but they have not had much of an impact on more intractable issues, such as reducing unaccounted-for water and expanding access. Lease contracts have drastically improved access and boosted oper- ational efficiency, but, except in Côte d'Ivoire, the associated investments The Elusiveness of the Millennium Development Goals for Water and Sanitation 25 Figure 1.8 Hidden Costs and Institutions 200 hidden costs (% total billings) 150 100 50 0 n g ce n n en r em cto io tio tio lin an at t nd la a ag se rn liz tiz gu bu ve an te tra ra re m riva un go po n gh ce or gh p hi de Ec hi SO yes no Source: Banerjee, Skilling, and others 2008. Note: SOE = state-owned enterprise. have been publicly financed. The lease contracts in Guinea and in Maputo have been affected by a lack of coordination between the private contractor and the government, which has stalled progress in some key areas, such as unaccounted-for water. Overall, private sector contracts accounted for almost 20 percent of the increase of household connections in the region, twice the amount that would be expected given their mar- ket share of only 9 percent (table 1.15). However, half of these gains were made in Côte d'Ivoire alone (which has been adversely affected since the onset of civil war in 2002). About half of the countries (mainly anglophone) have established ded- icated regulatory agencies for the water sector, although a significant number of these have not adopted private sector participation. Conversely, a number of francophone countries with private participation have adopted regulatory frameworks contractually, without establishing an independent regulatory agency. No evidence seems to support the superiority of any one of these two approaches. Even where explicit reg- ulatory frameworks have been established, these typically meet only around half of the corresponding good practice criteria. However, evi- dence for the links between introducing an independent regulator and improving performance is negligible for the water sector. Similarly, no conclusive evidence is found of the superiority of regulation by contract over the traditional form of regulation by agency. 26 Table 1.15 Overview of Impact of Private Sector Participation on Utility Performance Unit change in performance before and after private participation Household Improved Service Unaccounted- Collection Labor Country or city Contract connections water continuity for water ratio productivity Gabon Concession +20 ­8 contract Mali +15 +29 ­14 Côte d'Ivoire Lease contract +19 +22 +2.6 Guinea or affermage +27 ­0 Maputo +2 +10 ­1 +24 Niger +9 +3 ­5 +3.2 Senegal +18 +17 ­15 +2.8 Johannesburg Management ­0 +10 Kampala contract +6 ­2 +12 Zambia +5 ­28 +19 Source: Adapted from Marin 2009. Note: Blank cells denote missing data; household connections and improved water are measured as additional percentage points of households with access; service continuity is measured as additional hours per day of service; unaccounted-for water is measured as lower percentages of lost water; collection ratio is measured as additional percentage points of collection; and labor productivity is measured as additional thousands of connections served per employee. The Elusiveness of the Millennium Development Goals for Water and Sanitation 27 Of governance reforms that appear to be the most important drivers of higher performance, two are especially promising: performance contracts with incentives and independent external audits. For instance, Uganda's water company has had success using a performance contract that offers incentives for good performance and improves accountability. The intro- duction of independent audits has also positively affected efficiency. A Multidimensional Snapshot of WSS in Africa What policies are appropriate to deal with the state of the sectors just reviewed? How can WSS services be improved and access to them widened to include more of the continent's people? No recipe book neatly lays out the steps that each country should adopt to enhance cov- erage. In fact, the challenge of expanding access differs immensely across Sub-Saharan Africa, and so do the explanations for mixed performance. The rest of this volume presents a snapshot of sector performance, financing resources, and institutional, regulatory, and governance frame- works that is meant to augment our understanding of specific country experiences, help define barriers and constraints, measure resources and capacities, and identify opportunities for improvement. Chapters 2 and 3 set the stage by presenting access trends and market structures in water and sanitation sectors, respectively. Chapter 4 dis- cusses the sector's organization and regulatory arrangements. An analysis of performance variables in urban water utilities follows in chapter 5. Tariff structures, subsidy mechanisms, and affordability themes are intro- duced in chapter 6. Chapters 7 and 8 present financing arrangements for WSS, estimate the amounts that will have to be spent to achieve the MDG targets for access to WSS, and calculate the gap between available financing and the amounts needed. Finally, chapter 9 provides menu of options that may be used to bridge the funding gap in water and sanita- tion. These concluding chapters also review policy options. The chapters are supported by a comprehensive set of tabular appen- dixes that present the information base generated from AICD's extensive data-collection and data-processing efforts. Six sets of tables follow: Appendix 1 deals with access to WSS services (chapters 2­3). Appendix 2 relates to the institutional landscape (chapter 4). Appendix 3 is concerned with the technical and financial performance of water utilities (chapter 5). Appendix 4 relates to utility tariffs. Appendix 5 explores the affordability of WSS services (chapter 6). Appendix 6 deals with investment needs and the gap between those needs and available resources (chapters 7­8). 28 Africa's Water and Sanitation Infrastructure Annex 1.1 Surveys in the AICD DHS/MICS Survey Database Included in the Available observations Year of survey trend Country 1990­95 1996­2000 2001­05 DHS MICS analysis Benin 1996, 2001 X Burkina Faso 1993, 1999, 2003 X Cameroon 1991, 1998, 2004 X Central African Republic 1995 Chad 1997, 2004 X Comoros 1996 Congo, Dem. Rep. 2000 X Congo, Rep. 2005 Côte d'Ivoire 1994, 1999 X Ethiopia 2000, 2005 X Gabon 2000 Ghana 1993, 1998, 2003 X Guinea 1999, 2005 X Kenya 1993, 1998, 2003 X Lesotho 2005 2000 X Madagascar 1992, 1997, 2004 X Malawi 1992, 2000, 2004 X Mali 1996, 2001 X Mauritania 2001 Mozambique 1997, 2003 X Namibia 1992, 2000 X Niger 1992, 1998 X Nigeria 1990, 1999, 2003 X Rwanda 1992, 2000, 2005 X Senegal 1993, 1997, 2005 X South Africa 1998 Sudan 2000 Tanzania 1992, 1999, 2004 X Togo 1998 Uganda 1995, 2001 X Zambia 1992, 1996, 2002 X Zimbabwe 1994, 1999 X Source: Banerjee, Wodon, and others 2008. Note: DHS = demographic and health survey, MICS = multiple-indicator cluster survey. The Elusiveness of the Millennium Development Goals for Water and Sanitation 29 Annex 1.2 Surveys in the AICD Expenditure Survey Database Questions Questions Sample on water on Country Type and year of survey size supply sanitation 1 Angola Integrated Expenditure Survey 2000 10,116 Yes No 2 Benin Core Welfare Indicators 5,350 Yes Yes Questionnaire 2002 3 Burkina Faso Core Welfare Indicators 8,500 Yes Yes Questionnaire 2003 4 Burundi Priority Survey 1998 6,668 Yes No 5 Cameroon Enquête Camerounaise auprès 4,584 Yes Yes des ménages II 2001 6 Cape Verde Integrated Expenditure Survey 2001 -- Yes Yes 7 Chad Enquête sur la consommation et le 10,992 Yes Yes secteur informel au Tchad 2002 8 Congo, Dem. Integrated Expenditure Survey 2005 10,801 Yes Yes Rep. 9 Congo, Rep. Enquête Congolaise auprès des 12,097 Yes Yes ménages pour l'évaluation de la pauvreté 2005 10 Côte d'Ivoire Integrated Expenditure Survey 2002 5,002 Yes Yes 11 Ethiopia Welfare Monitoring Survey 2000 16,672 Yes Yes 12 Gabon Core Welfare Indicators 7,902 Yes Yes Questionnaire 2005 13 Ghana Ghana Living Standards Survey 5,991 Yes Yes 1998/99 14 Guinea-Bissau Core Welfare Indicators 3,216 Yes Yes Questionnaire 2002 15 Kenya Welfare Monitoring Survey 1997 10,874 Yes Yes 16 Madagascar Enquête prioritaire des 5,081 Yes Yes ménages 2001 17 Malawi Integrated Household Survey 2003 11,280 Yes Yes 18 Mauritania Enquête permanente sur les 5,865 Yes Yes conditions de vie des ménages 2000 19 Morocco Integrated Household Survey 2003 5,129 Yes Yes 20 Mozambique Inquérito aos agregados familiares 8,703 Yes Yes sobre orçamento familiar 2002/03 21 Niger Integrated Household Survey 2005 6,690 Yes Yes 22 Nigeria Nigeria Living Standards Survey 2003 19,158 Yes Yes 23 Rwanda Enquête intégrale sur les conditions 6,420 Yes Yes de vie des ménages (avec module budget et consommation) 1999 24 São Tomé Enquête sur les conditions de vie 6,594 Yes Yes and Príncipe des ménages 2000 (continued next page) 30 Africa's Water and Sanitation Infrastructure Questions Questions Sample on water on Country Type and year of survey size supply sanitation 25 Senegal Integrated Expenditure Survey 2001 2,418 Yes Yes 26 Sierra Leone Integrated Household Survey 2003 3,713 Yes Yes 27 South Africa Integrated Expenditure Survey 2000 26,263 Yes Yes 28 Tanzania Household Budget Survey 2000 22,207 Yes Yes 29 Uganda National Household Survey 2002 9,710 Yes Yes 30 Zambia Living Conditions Monitoring 9,715 Yes Yes Survey 2002 Total 267,711 30 28 Source: Banerjee, Wodon, and others 2008. Note: -- = not available. Annex 1.3 Introducing a Country Typology Africa's numerous countries face widely diverse economic situations. Understanding that structural differences in countries' economies and institutions affect their growth and financing challenges as well as their economic decisions (Ndulu and others 2007), this chapter introduces a four-way typology to organize the rest of the discussion. This typology provides a succinct way of illustrating the diversity of infrastructure financing challenges faced by different African countries. Middle-income countries have a gross domestic product (GDP) per capita in excess of $745 but less than $9,206. Examples include Cape Verde, Lesotho, and South Africa (World Bank 2007). Resource-rich countries are countries whose behaviors are strongly affected by their endowment of natural resources (IMF 2007).2 Resource- rich countries typically depend on minerals, petroleum, or both. A coun- try is classified as resource rich if primary commodity rents exceed 10 percent of GDP. (South Africa is not classified as resource intensive, using this criterion.) Examples include Cameroon, Nigeria, and Zambia. Fragile states are low-income countries that face particularly severe development challenges, such as weak governance, limited administrative capacity, violence, or the legacy of conflict. In defining policies and approaches toward fragile states, different organizations have used differ- ing criteria and terms. Countries that score less than 3.2 on the World Bank's Country Policy and Institutional Performance Assessment belong to this group. Fourteen countries in Africa are in this category. Examples include the Democratic Republic of Congo, Côte d'Ivoire, and Sudan (World Bank 2005). The Elusiveness of the Millennium Development Goals for Water and Sanitation 31 Other low-income countries compose a residual category of countries with GDP per capita below $745 and that are neither resource-rich nor fragile states. Examples include Benin, Ethiopia, Senegal, and Uganda. Notes 1. See United Nations, "Millenium Development Goals, http://www.un .org/millenniumgoals/. 2. See also Paul Collier and Stephen O'Connell, draft chapter (2006) for the synthesis volume of the African Economic Research Consortium's Explaining African Economic Growth project, Oxford University and Centre for Study of African Economies, and Swarthmore College and Centre for Study of African Economies. Bibliography Banerjee, S., H. Skilling, V. Foster, C. Briceño-Garmendia, E. Morella, and T. Chfadi. 2008. "Ebbing Water, Surging Deficits: Urban Water Supply in Sub- Saharan Africa." AICD Background Paper 12. World Bank, Washington, DC. Banerjee, S., Q. Wodon, A. Diallo, N. Pushak, H. Uddin, C. Tsimpo, and V. Foster. 2008. "Access, Affordability and Alternatives: Modern Infrastructure Services in Sub-Saharan Africa." AICD Background Paper 2. World Bank, Washington, DC. Briceño-Garmendia, C., K. Smits, and V. Foster. 2008. "Financing Public Infrastructure in Sub-Saharan Africa: Patterns and Emerging Issues." AICD Background Paper 15. World Bank, Washington, DC. Esrey, S. A., J. B. Potash, L. Roberts, and C. Shiff. 1991. "Effects of Improved Water Supply and Sanitation on Ascariasis, Diarrhea, Dracunculiasis, Hookworm Infection, Schistosomiasis and Trachoma." Bulletin of the World Health Organization 69 (5): 609­21. Foster, Vivien, and Cecilia Briceño-Garmendia, eds. 2009. Africa's Infrastructure: A Time for Transformation. Paris and Washington, DC: Agence Française de Développement and World Bank. Foster, Vivien, William Butterfield, Chuan Chen, and Nataliya Pushak. 2008. "Building Bridges: China's Growing Role as Infrastructure Financier for Sub- Saharan Africa." Trends and Policy Options 5, Public-Private Infrastructure Advisory Facility, World Bank, Washington, DC. Hutton, G., and L. Haller. 2004. "Evaluation of the Costs and Benefits of Water and Sanitation Improvements at the Global Level." World Health Organization, Geneva. 32 Africa's Water and Sanitation Infrastructure IMF (International Monetary Fund). 2007. "Regional Economic Outlook: Sub- Saharan Africa." International Monetary Fund, Washington, DC. JMP (Joint Monitoring Programme). 2000. Global Water Supply and Sanitation Assessment 2000 Report. Geneva: World Health Organization; New York: United Nations Children's Fund. ------. 2008. Progress on Drinking Water and Sanitation. Special Focus on Sanitation. Geneva: World Health Organization; New York: United Nations Children's Fund. Keener, S., M. Luengo, and S. G. Banerjee. 2009. "Provision of Water to the Poor in Africa: Experience with Water Standposts and the Informal Water Sector." AICD Working Paper 13. World Bank, Washington, DC. Marin, P. 2009. "Public Private Partnerships for Urban Water Utilities: A Review of Experiences in Developing Countries." Trends and Policy Options 8, Public-Private Infrastructure Advisory Facility and World Bank, Washington, DC. Morella, E., V. Foster, and S. Banerjee. 2008. "Climbing the Ladder: The State of Sanitation in Sub-Saharan Africa." AICD Background Paper 13. World Bank, Washington, DC. Ndulu, Benno J., Stephen A. O'Connell, Robert H. Bates, Paul Collier, and Charles C. Soludo, eds. 2007. The Political Economy of Economic Growth in Africa, 1960­2000. Volume 1. Cambridge: Cambridge University Press. PPIAF (Public-Private Infrastructure Advisory Facility). 2008. "Private Participation in Infrastructure Project Database." http://ppi.worldbank.org/. World Bank. 2005. "Infrastructure Finance for Africa--A Strategic Framework." Unpublished concept note, World Bank, Washington DC. ------. 2007. "DEPweb Glossary." Development Education Program, World Bank, Washington, DC. http://www.worldbank.org/depweb/english/modules/ glossary.html#middle-income. CHAPTER 2 Access to Safe Water: The Millennium Challenge The water landscape in Africa is characterized by discrepancies within and among countries. Some countries are closer than others to achieving the water target spelled out in the Millennium Development Goals (MDGs).1 In this chapter, we present the recent evolution and current status of water service in Africa, focusing on the underlying markets-- urban formal, urban informal, and rural--each with their unique attrib- utes and players. Some countries emerge as robust performers in expanding coverage in urban and rural areas, whereas others have remained stagnant or fallen behind in serving their population. The Importance of Wells and Boreholes in Water Supply Less than one-third of African households have reached the top parts of the ladder. About 15 percent of African households receive piped water through household connections; another 15 percent receive it through standposts. Wells and boreholes cover 37 percent of households, making them the most prevalent form of water supply in the region. Much of the remainder of the population relies on surface water. Operating mainly in urban areas, water vendors serve about 2 percent of households. 33 34 Africa's Water and Sanitation Infrastructure Rates of water supply coverage show tremendous heterogeneity from one country to another. The variation in household piped-water coverage is wide--from 2 percent in Uganda to about 60 percent in South Africa. In most countries, piped water reaches less than 20 percent of households (figure 2.1a). Only three countries--Gabon, Senegal, and South Africa-- can claim a piped-water coverage rate of more than 40 percent. The cov- erage of wells/boreholes and surface water reveal even greater variation (figure 2.1b). The low rate of piped-water coverage reflects Africa's relatively low rate of urbanization. Piped-water coverage in rural areas is several magni- tudes lower than in urban areas. Only 4 percent of rural households in Africa receive piped water, compared with 38 percent in urban areas. When public standposts are included, more than 60 percent of Africa's urban households have access to some kind of utility provided water. In rural areas, wells and boreholes and surface water predominate. More than 80 percent of Africa's rural households receive their water from these sources. Richer households are much more likely to enjoy access to piped water than are poorer households. On the water-supply ladder, rising income is associated with piped water and a declining dependence on wells, bore- holes, and surface water. In the lowest three quintiles of the wealth dis- tribution, access to piped water through a household connection is well below 10 percent, with negligible coverage of the poorest households (table 2.1). Even in the fourth quintile, access to piped water within the household is less than 20 percent, whereas for the richest quintile it is close to 50 percent--still far from universal (and highly variable across countries). Most of the countries in the sample of the Africa Infrastructure Country Diagnostic (AICD) are low-income countries with a per capita gross domestic product (GDP) of less than $1,000 per year, but the sam- ple also includes several middle-income countries: Cape Verde, Gabon, Lesotho, Namibia, and South Africa. The degree of urbanization varies widely in Africa--from 12 percent in Uganda to 80 percent in Gabon-- the average being about 35 percent. Both income and urbanization are directly correlated with access to safe water. Higher incomes make safe water more affordable, and the greater population densities associated with urbanization help to reduce the cost of expanding access to modern services. Access to piped water is four times greater in middle- than in low- income countries and three times greater in the most urbanized Access to Safe Water: The Millennium Challenge 35 Figure 2.1 African Households' Access to Various Forms of Water Supply a. Rate of household access to different forms of water supply 100 80 % households 60 40 20 0 er ts s er s le or os at at ho nd w w dp re ve d ce an bo pe rfa st ls/ pi su el w less than 10% between 10% and 20% between 20% and 40% above 40% b. Households served by different modes of water supply (median, plus minimum/maximum country rates) 80 70 72% 68% 60 59% % households 50 48% 40 42% 30 20 19% 18% 16% 14% 10 7% 4% 0 2% 1% 0% er ts es er s or os at at ol nd w w dp eh ve ed ce or an rfa /b p st pi ls su el w Source: Banerjee, Wodon, and others 2008. 36 Africa's Water and Sanitation Infrastructure Table 2.1 Coverage Rate of Water Supply (percent) Quintile Quintile Quintile Quintile Quintile Overall Rural Urban 1 2 3 4 5 Household piped water 15 4 38 0 3 7 18 46 Standposts 15 10 25 6 11 13 20 23 Wells/ boreholes 37 43 24 44 46 42 35 20 Surface water 30 41 7 49 39 34 23 7 Vendors 2 1 4 1 1 2 2 2 Source: Banerjee, Wodon, and others 2008. economies than in the least, and recourse to surface water is about twice as prevalent in the low-income and least urbanized countries than in the middle-income and most urbanized countries (table 2.2). These patterns hold across urban and rural service segments, and across the different quintiles of the distribution of spending on water service as well. Thus, in more highly urbanized countries, even the rural population is substan- tially better off. Nevertheless, even in middle-income and urbanized countries, the benefits of access are largely confined to the top three quin- tiles of the distribution, with too many in the bottom two quintiles still without access to safe water. In the vast majority of countries the distribution of access is even more unequal than the distribution of income, exacerbating inequalities in soci- ety as a whole. Furthermore, the distribution of new connections result- ing from the service expansions that have occurred in recent years is also more unequal than income. It appears, therefore, that the benefits of cur- rent access and new extensions tend to accrue to the better-off. This may be because access rates in Africa remain low even among the wealthier segments of the population, so it makes business sense for the utilities to initially concentrate their expansion efforts (Diallo and Wodon 2005). Even if one controls for income and urbanization, some countries stand out as having much higher (or lower) levels of coverage than might be expected, and these cases merit closer study (figure 2.2). As seen in figure 2.2, Cameroon and Ghana have relatively high incomes and high rates of urbanization, yet their piped-water coverage is relatively small, suggesting underperformance. Senegal, by contrast, has coverage that compares favorably with that of peers at similar (and even greater) levels of income and urbanization. Sitting just to the left of Senegal, on the 50 percent urbanization line, Nigeria stands out as having low levels of Access to Safe Water: The Millennium Challenge 37 Table 2.2 Coverage Rate of Water Supply, by Country Income and Urbanization Status (percent) Household Public Well or Surface Population weighted piped water standpost borehole water Vendors By country income Middle 44 22 13 18 1 Low 11 14 40 32 2 By urbanization level Low 7 16 36 39 1 Medium 17 12 35 33 0 High 21 15 40 19 4 Source: Banerjee, Wodon, and others 2008. Note: Urbanization level: low (0­30 percent), medium (30­40 percent), and high (> 40 percent). Figure 2.2 Extent of Access to Piped Water through Household Connection, by GDP and Urbanization Rate 70 60 Cameroon urbanization rate (%) 50 Ghana 40 Senegal 30 20 10 Lesotho Uganda 0 0 1,000 2,000 3,000 4,000 GDP per capita PPP, US$ Source: Banerjee, Wodon, and others 2008. Note: PPP = purchasing power parity. piped-water coverage relative to peers. Benin, a strong performer on piped-water access, provides a good contrast with Nigeria. Zambia, too, per- forms reasonably well on access to piped water, relative to its per capita national income and rate of urbanization. Low Access to Piped Water. . . for Various Reasons Access to piped water is low in most of Africa and has not expanded sub- stantially in recent years. The main reasons are rapid population growth 38 Africa's Water and Sanitation Infrastructure and shrinking household size (box 2.1), two trends that continually increase the size of the unserved population and challenge the capacities of weak and underfunded utilities to expand connections to growing numbers of households. Box 2.1 The Problem of Shrinking Households As incomes rise, African households are getting smaller. Urbanization, lower fertil- ity, and greater economic resources all allow nuclear families to disengage from extended households because they no longer need the economies of scale pro- vided by larger households. In Benin, for example, the average household size decreased from 6.0 in 1996 to 5.2 in 2001. Shrinking household size exerts a strong effect on the need for new water- supply connections, sometimes canceling out the effect of slower population growth. For that reason, the new-connection needs of richer countries may equal or outstrip those of poorer countries. There is a wide cross-country dispersion in the relative growth rates of popu- lation versus the number of households. For the AICD sample as a whole, howev- er, the average rate of population growth is 2.5 percent, and the average increase in the number of households is 3.2 percent, so the trend toward smaller house- hold sizes represents almost one-third (0.7 percent) of the new connections needed to keep access rates constant (Diallo and Wodon 2007). In a few countries, by contrast, household size has increased. Typically this occurs during hard times, as households join forces to cope with deterioration in their living conditions. Rates of Change in Number of Households and Population, Selected Countries Difference between annual household growth and population growth Countries Higher than 2 percent Benin, Namibia, Zimbabwe Between 1 and 2 percent Cameroon, Guinea, Mali, Nigeria Between 0 and 1 percent Burkina Faso, Côte d'Ivoire, Kenya, Madagascar, Malawi, Niger, Rwanda, Senegal, Tanzania, Zambia Less than 0 percent Chad, Ethiopia, Ghana, Mozambique, Uganda Source: Banerjee, Wodon, and others 2008. Access to Safe Water: The Millennium Challenge 39 The challenge of reaching universal access to safe sources of water is typically understood as a supply-side problem of rolling out infrastructure networks to increasingly far-flung populations, entailing major invest- ments. However, even in densely populated urban areas, where infrastruc- ture is already present or easy to expand, service coverage is by no means universal. Part of the access problem therefore appears to be related to demand-side barriers that prevent households from hooking up to avail- able services. In addition to high connection charges that make hookups unaffordable, demand-side barriers include illegal land tenure, which dis- qualifies households from connecting, and a variety of other social and economic factors that may deter households from becoming utility clients. Household surveys can be used to explore the reasons why a house- hold might elect not to connect to the water-supply network. Samples are based on geographic clusters that at least for urban areas are physically small, amounting to no more than a few city blocks. It is therefore possi- ble, at least in urban areas, to study the extent to which people who lack access to infrastructure live in clusters where infrastructure is available (as indicated by the fact that some of their immediate neighbors are con- nected). The resulting analysis gives us a sense of the degree to which low access to services is driven by supply-side issues (infrastructure networks not reaching the areas where people live) or by demand-side issues (peo- ple not connecting to available infrastructure networks). The building blocks of the analysis are presented in box 2.2. The novelty of this approach is that we break down the traditional measure of household coverage into two components (using the method of Foster and Araujo 2004 and Komives and others 2005). The first compo- nent, which we call access, is the percentage of the population that lives in a cluster where at least one household has service coverage, indicating that the infrastructure is physically proximate and that households probably have an opportunity to connect. The second component, which we call hookup, is the percentage of the population living in clusters where the opportunity to connect to the service is available. Using these two concepts, we can estimate the percentage of the unserved population that constitutes a supply-side deficit (meaning that they are too far from the network to make a connection until the network is expanded to reach them) versus a demand-side deficit (meaning that something other than distance from the network is preventing them from taking up the service). The optimal policy response to the two conditions is very different-- hence the importance of making the distinction. The solution to a supply- side deficit is to make further investments to extend the geographic reach 40 Africa's Water and Sanitation Infrastructure Box 2.2 Coverage, Access, and Hookup Rates: Relationships and Definitions Coverage rate = number of households using the service / total number of households Access rate = number of households living in communities or clusters where service is available / total number of households Hookup rate = number of households using the service / number of households living in communities where service is available Coverage = access rate x hookup rate Unserved population = 100 ­ coverage rate Pure demand-side gap = access rate ­ coverage rate Supply-side gap = unserved population ­ pure demand-side gap Pure supply-side gap = supply-side gap x hookup rate Mixed demand and supply-side gap = supply-side gap x (100 ­ hookup rate) Proportion of deficit attributable to demand-side factors only = pure demand- side gap / unserved population Proportion of deficit attributable to supply-side factors only = pure supply-side gap / unserved population Proportion of deficit attributable to both demand- and supply-side factors only = mixed demand and supply-side gap / unserved population Source: Foster and Araujo 2004. of the network. The solution to a demand-side deficit is to make policy changes that address barriers to service take-up, such as high connection charges or illegal tenure. For various reasons, it could be questioned whether everyone in a given geographic cluster really has the opportunity to connect. First, even in a small cluster, some residents may live too far from the network to con- nect. Second, the network may not have the carrying capacity required to service all residents in a particular geographic cluster without further investment and upgrade. Third, even if a household is physically close to a network with adequate carrying capacity, it may choose not to connect because it has an acceptable alternative (such as a borehole). Access to Safe Water: The Millennium Challenge 41 Diallo and Wodon (2007) use a statistical approach to correct for these problems. They simulate the maximum connection rate obtainable in any primary sampling unit based on that of the richest households in that area. If less than 100 percent of the richest households (which are assumed to be able to play) are connected, something other than demand-side barriers is probably at work. The methodology is less applicable to rural areas because the clusters tend to be larger and population densities much lower. Rates of access to piped water in urban areas of Africa exceed cover- age rates by 30 to 40 percentage points (table 2.3). Indeed, access rates are as high as 70 to 90 percent, which means that the vast majority of the urban population, even in low-income countries, lives in relatively close proximity to existing water networks. Hookup rates are another story: They are significantly higher in middle-income than in low-income coun- tries. The proportion of the coverage deficit that is attributable to demand-side factors, adjusted using the method of Diallo and Wodon (2007), is 14 percent in the low-income countries (meaning that one in seven urban residents elects not to connect to the available service) and 36 percent in the middle-income countries. Without the adjustment, the share of the coverage deficit attributable to demand-side factors appears much larger. When coverage is examined by country, one sees a very strong relation- ship between the level of access (that is, the share of the population liv- ing in areas where piped-water service is available) and the size of the demand-side deficit (figure 2.3). That relationship is intuitively satisfying because, as rates of access rise with expansion of infrastructure network, Table 2.3 Water-Service Coverage in Urban Africa and Share of Coverage Deficit Attributable to Demand-Side Factors (percentage of urban households) Proportion of coverage Decomposition of coverage deficit attributable to (1) (2) (1) x (2) demand-side factors Access Hookup Coverage Unadjusted Adjusted Country income Low 68 42 31 58 14 Middle 91 74 69 61 36 Urbanization level Low 76 42 33 65 20 Medium 76 56 46 63 8 High 71 49 34 55 45 Source: Banerjee, Wodon, and others 2008. 42 Africa's Water and Sanitation Infrastructure Figure 2.3 Country Scatter Plot of Current Access Rates for Piped Water and Demand-Side Factors in Coverage Deficit 100 90 demand-side problem (%) 80 70 60 50 40 30 20 10 0 30 40 50 60 70 80 90 100 access rate (%) Source: Banerjee , Wodon, and others 2008. demand-side factors come to assume a greater role in the remaining cov- erage deficit. One also observes, however, substantial variation across countries in the size of the adjusted coverage deficit that is due to demand-side factors--from less than 5 percent in Burkina Faso, the Central African Republic, Chad, Ethiopia, Mozambique, Rwanda, Tanzania, and Uganda, to more than 50 percent in Côte d'Ivoire, the Republic of Congo, Gabon, Senegal, and Zambia. We have already noted the importance of distinguishing between demand- and supply-side factors when making policies to increase access. The demand-side problems are comparatively more deep-rooted and are directly related to the consumer's income and ability to pay. The supply- side problems are related to the utilities' investments in its network and to expand its consumer base. The ability to do so depends on the strength of its revenue: If the volume of high-value industrial and residential con- sumers is low in the consumer mix, the utilities will find it difficult to generate adequate funds to invest in network expansion. Multiple Players in the Urban Water Market Our analysis of patterns of access to water in urban areas reveals three categories of countries (table 2.4). The first comprises countries in which a large share of the urban population obtains water through wells and Access to Safe Water: The Millennium Challenge 43 Table 2.4 Patterns of Urban Access to Water (percent) Dominant Range of Average modality prevalence prevalence Countries Piped water 28­93 57 Benin; Comoros; Congo, Dem. Rep.; Congo, Rep., Côte d'Ivoire; Ethiopia; Gabon; Kenya; Lesotho; Mauritania; Namibia; Senegal; South Africa; Togo; Zambia; Zimbabwe Standposts 37­53 43 Burkina Faso, Cameroon, Central African Republic, Ghana, Guinea, Madagascar, Malawi, Mozambique, Niger, Rwanda, Tanzania, Uganda Wells/ 33­48 39 Chad, Mali, Nigeria, Sudan boreholes Source: Banerjee, Wodon, and others 2008. boreholes, while other improved sources also provide substantial cover- age. The second comprises countries where the majority of the urban population depends on public standposts. The third group comprises countries where the majority of the urban population has piped water from household connections. Urban households that lack a piped-water connection have several alternative sources from which to choose: public standposts, water kiosks, vendors (or resellers) of water, rainwater harvesting, shallow wells, and surface water. Although the ability of these alternative suppliers to pro- vide adequate service to the unconnected population is debated, their operations recently have come to be better understood (Collignon and Vézina 2000; Kariuki and others 2003; Kariuki and Schwartz 2005; Keener and Banerjee 2005). These providers have come to occupy an important place in urban Africa, particularly in dense periurban areas and in postconflict economies. In these areas, the formal sector's ability to deliver services is continually challenged, and an informal market has emerged to fill the gap. Household connection rates are directly linked to the strength of the informal market. Not surprisingly, the percentage of unconnected house- holds served by water tankers or water vendors is higher in countries where household connection rates are lower. In countries with very low rates (less than 30 percent) of household connection, 13 percent of the unconnected urban population, on average, relies on water trucks or water vendors. In countries with low to medium (30­60 percent) rates of 44 Africa's Water and Sanitation Infrastructure household coverage, just 4 percent of the unconnected urban population relies on water trucks or vendors. For countries with medium to high (> 60 percent) rates of household coverage rates, only 2 percent of the unconnected urban population relies on water trucks or vendors. In an analysis of data available from Africa's 24 largest cities from the AICD Water Supply and Sanitation (WSS) Survey, we found that public standposts are the principal source of water for unconnected households. Average standpost coverage in the cities studied was 28 percent, but standposts supply water to up to 53 percent of unconnected households (table 2.5). The actual coverage of public standposts may be lower, however, than suggested by the foregoing figures, which are derived from official data reported by utilities and governments. Several independent sectoral sur- veys assessed the coverage provided by standposts and other alternative providers in a way that made it possible to compare the results with offi- cial statistics. In Maseru, the capital of Lesotho, for example, data from an official multiple-indicator cluster survey revealed that about half of the urban population lacked a piped-water connection and that the utility assumed that this segment was reliant on its free public standposts. But an earlier, more detailed sectoral survey undertaken in Maseru in 2002 showed that coverage by free public standposts was as low as 16 percent of the population, with the coverage among the unconnected falling from 100 percent to 24 percent (Hall and Cownie 2002).2 It is unlikely that the three-year lag between surveys accounts for the stark differences in these numbers. In fact, utility data deviate from household survey data in estimating standpost coverage. Most utilities calculate that coverage by multiplying the number of existing standposts by a "standard" number of users (usu- ally 300 to 500).3 The resulting estimates can be very inaccurate, how- ever, because they do not take into account the factors that affect the real usage of standposts--such as their location relative to population, water pressure, operating hours, and even whether a given standpost is actually working. In Ouagadougou, for example, the number of people relying on standposts was often calculated using a multiplier of 700 people per standpost. After detailed field studies showed actual coverage to be much lower, the utility reduced its standard number of users from 700 to 300 people per standpost.4 About one out of five standposts in Africa is in poor working condi- tion. In some places, the figures are much worse. In Kinshasa, for instance, only 21 percent of the standposts are in good working condition Table 2.5 Water Supply in Africa's Largest Cities, by Source Household Standpipes/ Water Household Water Small piped connection kiosks tankers resellers vendors networks Country Largest city (%) (%) (%) (Yes/No/%) (Yes/No) (Yes/No) Benin Cotonou 31 -- n.a. Yes No Yes Burkina Faso Ouagadougou 34 61 n.a. No 5 No Ethiopia Addis Ababa 39 40 n.a. Yes Yes No Mozambique Maputo 26 26 n.a. 26 Yes 12 Niger Niamey 31 21 n.a. No 10 No Nigeria Kaduna 48 2 -- Yes Yes No Rwanda Kigali 35 51 3.21 10 No No Senegal Dakar 77 19 n.a. Yes No No South Africa Johannesburg 88 12 0.24 No No No Congo, Dem. Rep. Kinshasa 36 -- n.a. Yes No Yes Ghana Accra 56 -- -- Yes Yes No Kenya Nairobi 51 41 -- No 8 9 Lesotho Maseru 33 16 1.00 31 5 No Malawi Blantyre 47 -- n.a. Yes No No Namibia Windhoek 73 20 n.a. No No No Sudan Greater Khartoum 27 0.11 0.43 Yes 60 No Zambia Lusaka 27 58 n.a. Yes Yes No Cape Verde Praia 34 60 6.30 No No No Chad N'Djamena 22 -- -- Yes Yes Yes Côte d'Ivoire Abidjan 65 -- n.a. Yes No Yes Madagascar Antananarivo 42 34 n.a. Yes 8 Yes (continued next page) 45 46 Table 2.5 (continued) Household Standpipes/ Water Household Water Small piped connection kiosks tankers resellers vendors networks Country Largest city (%) (%) (%) (Yes/No/%) (Yes/No) (Yes/No) Tanzania Dar es Salaam 29 4 2.00 35 2 Yes Uganda Kampala 30 5 -- Yes Yes Yes Average 43 28 2.20 -- -- -- Median 35 21 2 -- -- -- Minimum 22 0.11 0 10 2 6 Maximum 88 61.0 6 35 60 12 Number of countries with All All 11/23 (48) 17/23 (74) 14/23 (61) 9/23 (39) relevant presence Source: Keener, Luengo, and Banerjee 2009. Note: For the unconnected market, the data obtained from independent studies have been highlighted. The remaining data come from utility and government sources. n.a. = not applicable, -- = not available. Access to Safe Water: The Millennium Challenge 47 (table 2.6). In many cities where standposts tend to be in poor working condition, vendors sell water door to door or from household connec- tions. In such cases, although people may occasionally obtain their water directly from the standpost, they also get it from vendors who make it their business to transport water from operating standposts. The growing role of household resellers is usually hidden in household surveys, because it is illegal to sell water in many countries, and house- holds are unwilling to admit to engaging in proscribed activities. However, the results of the module of AICD's WSS Survey devoted to small-scale independent providers (module 5) reveal that household reselling is a common occurrence in 70 percent of the countries studied-- despite being prohibited in 24 percent of the countries in which it is prevalent. In Maputo, for instance, one-third of the unconnected obtain their water from neighbors (Boyer 2006). Similarly, in Maseru, household resellers provide water to 31 percent of the population and to almost half of the unconnected (Hall and Cownie 2002). Table 2.6 Working Status of Standposts in the Largest Cities in Africa (percent) Population depending on Share in good Share free Country Largest city standposts working order of charge Sudan (HCI) Greater Khartoum 0.11 100 0 Congo, Dem. Rep. (HCI) Kinshasa n.a. 21 -- Mozambique (MCI) Maputo 26 58 0 Rwanda (MCI) Kigali 51 75 0 Namibia (MCI) Windhoek 20 100 100 Lesotho (LCI) Maserua 16 48 100 Kenya (LCI) Nairobi 41 89 0 Nigeria (LCI) Kaduna 2 55 96 Benin (LCI) Cotonou n.a. 100 0 Burkina Faso (LCI) Ouagadougou 61 100 0 Cape Verde (LCI) Praia 60 100 0 Niger (LCI) Niamey 21 98 0 Zambia (LCI) Lusaka 58 97 0 Malawi (LCI) Blantyre n.a. 90 0 Madagascar (LCI) Antananarivo 34 82 40 Average 32.40 81 24 Source: Keener, Luengo, and Banerjee 2009. Note: Data obtained from independent studies have been highlighted. The remaining data come from utility and government sources. HCI = high conflict index, MCI = medium conflict index, LCI = low conflict index, n.a. = not applicable, -- = not available. a. A negligible percentage of the standpipe/kiosk coverage is paid. 48 Africa's Water and Sanitation Infrastructure Legalizing household resale of water could be beneficial in expanding access to safe water, as demonstrated in Côte d'Ivoire (box 2.3). Other alternatives to piped water are offered by small-scale service providers who sell water from boreholes, wells, and other nonnetwork sources. In the past decade, water vendors, such as standpost operators, have gained some attention from the development community. Overall, vendors serve only 4 percent of urban Africa, but in some countries they play a prominent role. In Mauritania, 32 percent of urban residents depend on vendors. Vendors serve more than 5 percent of urban house- holds in Burkina Faso, Chad, Niger, Nigeria, and Tanzania. Water truckers tend to supply high- and middle-income households. They are especially visible in cities where the piped-water service is very poor in reach and reliability, such as Dar es Salaam, Kampala, and Nairobi. Truckers are present in half of the cities considered in this study, but their market share is limited (between 0.2 and 6.5 percent). In some cities, such as Accra and Luanda, water tankers not only supply directly to upper- and middle-income households but also play a key role in the Box 2.3 Legalizing Household Resellers in Côte d'Ivoire To make it easier for the poor to receive safe water, Côte d'Ivoire legalized house- hold resellers in informal settlements. Legalization enabled the water utility, Société de Distribution d'Eau de Côte d'Ivoire (SODECI) to indirectly influence the price and quality of water sold in these areas. It issued about 1,000 licenses to water resellers, many of whom have invested in last-mile network extensions to cater to demand in poor neighborhoods. SODECI reduces the risk of nonpayment by charging a high deposit (about $300) and invoicing resellers monthly. But the scheme faces implementation challenges. Household resellers pay SODECI twice--in the form of reseller payments and a price markup for network extensions. Furthermore, there is no special tariff for household resellers; they pay the high consumer tariff, so the incentive to become a household reseller is limited. An association of water resellers called AREQUAPCI that includes members licensed by SODECI has successfully worked out a deal to buy water at the same preferential rate as standpipe operators. Source: Collignon and Vézina 2000. Access to Safe Water: The Millennium Challenge 49 supply chain. Because of the limited extent of the piped network, many kiosks depend on water supplied by tankers. Small piped-water networks are relatively new in the urban landscape. In 40 percent of the largest cities, small, secondary water networks are operated by independent providers. These may be connected to the main city network (as in Abidjan, Cotonou, and Nairobi) or completely sepa- rated from the city network (as in Kampala, Maputo, and Nairobi). Even then, their coverage is marginal, at 12 percent in Maputo and 9 percent in Nairobi. The Role of Wells, Boreholes, and Surface Water in the Rural Water Market In most countries, wells and boreholes remain the most important source of water in rural Africa. Surface water is the second most important source, extending to more than 30 percent of the rural population in half of the sample countries. In no country in our sample does piped supply extend to more than 25 percent of the rural population. In fact, only in Namibia and South Africa does piped water reach more than 15 percent of the rural population, and in 7 out of 10 African countries it reaches less than 5 percent. Also, water collection imposes an enormous burden on households, primarily on women and children (box 2.4). Taking water closer to people promises enormous gains from health and time savings even if the opportunity cost of time is severely discounted. Our analysis of access patterns at the rural level reveals three cate- gories of countries (table 2.7). The first comprises those in which most of the rural population depends on surface water, the second those in which most rural dwellers obtain water through wells or boreholes, and the third group countries in which the rural population tends to rely on standposts. The challenges in rural water supply management are many, but per- haps the most important is sustaining the service. Governments struggle to enhance access to safe water and to maintain existing facilities, but low capacity at the local level hobbles water supply management, because inadequate maintenance leads to frequent breakdowns and cuts the use- ful life of equipment obtained with scarce resources. The need for new or rehabilitated systems widens the gap between available funding and the funding needed to meet the water MDG. In many countries, more than a third of rural water systems are not working at any given time (figure 2.4). Senegal, where 85 percent of rural water facilities are functioning, is the top performer, providing a stark 50 Africa's Water and Sanitation Infrastructure Box 2.4 The Opportunity Cost of a Distant Water Supply Fetching water from outside the home is an activity dominated by women and girls. Blackden and Wodon (2006) compute that more than two-thirds of the 6 million hours that Ghanaians spent fetching water in 1992 were spent by women. If access to water were more convenient, those hours might be spent on educa- tion or other productive purposes. Providing African households with reasonable access to water would bring significant gains in productivity, health, and welfare. On average, urban households that lack private water connections live about 500 meters from their water source, while in rural areas the average distance is closer to one kilometer. Some 20 percent of urban households and 30 percent of rural households live more than one kilometer from their water source. Distance of Households from Water Source in Selected Countries (Percentage of Households) 70 60 50 percent 40 30 20 10 0 urban rural in dwelling < 1 km 1­2 km > 2 km Patterns of access vary from country to country, but, on average, urban house- holds have more convenient access to water than do rural households. For in- stance, 53 percent of rural households in Tanzania live more than two kilometers from their water source. At the other extreme are Madagascar, Nigeria, and South Africa, where less than 2 percent of rural households live more than two kilometers from their water. Even in urban areas, water can be far away. In urban Mauritania, for example, 66 percent of households live more than two kilometers away from their water source. In urban Ghana and Sierra Leone, the corresponding figure is 53 percent. In comparison, less than 5 percent of households in urban areas in the Democratic Republic of Congo, the Republic of Congo, Ethiopia, Morocco, (continued next page) Access to Safe Water: The Millennium Challenge 51 Box 2.4 (continued) Niger, Nigeria, South Africa, Uganda, and Zambia live more than two kilometers from their water source. Household surveys allow us to measure changes in the time households spend fetching water. Since 1990, the average time spent fetching water for household consumption has remained virtually unchanged, at 45 to 50 minutes (round trip). In some countries, more time is spent at the task. Households in Ethiopia, Mozambique, Tanzania, and Uganda spend more than one hour each day fetching water for household consumption. In Ethiopia, Mozambique, Tanzania, and Uganda, moreover, the amount of time has increased over the years. These are also countries where more than 90 percent of households fetch water from outside their dwelling. Source: Banerjee, Wodon, and others 2008. Table 2.7 Patterns of Access across Countries in Rural Areas (percent) Dominant Range of Average modality prevalence prevalence Countries Standposts 28­93 57 Lesotho, South Africa Wells/ 41­87 62 Benin, Burkina Faso, Cameroon, Central boreholes African Republic, Chad, Comoros, Côte d'Ivoire, Ghana, Guinea, Malawi, Mali, Mauritania, Mozambique, Namibia, Niger, Nigeria, Senegal, Sudan, Tanzania, Togo, Uganda, Zambia, Zimbabwe Surface 56­87 65 Congo, Dem. Rep.; Congo, Rep.; Ethiopia; water Gabon; Kenya; Madagascar; Rwanda Source: Banerjee, Wodon, and others 2008. contrast with the Democratic Republic of Congo, where no more than 40 percent of rural water equipment is in working order. A significant number of rural water facilities are in need of rehabilitation at any given time--more than half in the Democratic Republic of Congo, Madagascar, and Malawi. Evidence from Ethiopia suggests that mechanized boreholes are more likely to be nonfunctional than springs and hand pumps, probably for lack of a reliable supply chain of replacement parts (Water and Sanitation 52 Africa's Water and Sanitation Infrastructure Figure 2.4 Working Status of Rural Water Points 100 80 60 percent 40 20 0 l r n da re o n da ad i o . ga aw ep ca th as da ni oi an an Ch as ne .R Be aF so al Su Iv Ug Rw ag M m Se d' Le in De ad te rk Cô Bu M o, ng Co nonfunctioning rural water points rural points in need of rehabilitation Source: Banerjee, Skilling, and others 2008. Program 2006). Field research from Ghana, Kenya, Uganda, and Zambia reveals that the supply-chain problem also affects hand pumps, because of factors specific to the African rural water realm--among them the sep- aration of pumps from other machines requiring spare parts, low pump density, poor choice of technology, restrictive maintenance systems, and relatively poor and immobile end users (Harvey and Reed 2006) Analysis of 25 studies across 15 countries in Africa has clarified the division of responsibility in the supply chain for spare parts. Governments and donors are responsible for managing the chain, but public and private sec- tor entrepreneurs are important players as well. One thing is clear: Depending on the private sector alone to supply spare parts is unlikely to be sustainable because of the low population density and income level of many rural areas (Water and Sanitation Program 2006). Steep Growth of Wells and Boreholes as Sources of Water The dynamics of service expansion reveal a similar overall pattern in both urban and rural areas. Across the board, the use of wells and boreholes is expanding more rapidly than all the utility-based alternatives put together. Water supply has evolved differently in Africa's urban areas than in its rural areas. Utilities have been unable to keep pace with the rising demand Access to Safe Water: The Millennium Challenge 53 for water in urban areas, with the result that piped-water coverage has declined over the past decade. In the mid-1990s, 43 percent of urban African households received piped water; by the early 2000s, the figure had slipped to 39 percent (table 2.8). The situation with urban standposts is similar, with a decline from 29 percent to 24 percent over the past 15 years. The decline occurred because the combined growth rates of improved sources of water in urban areas (less than 1 percent a year) fell short of pop- ulation growth (more than 4 percent a year). The decline in piped water has been matched by a rise in the prevalence of wells and boreholes, as well as slight increases in the use of surface water and water vendors in urban areas. By contrast, the situation in rural areas has improved, though from a low baseline. More rural dwellers now have access to standposts, wells, and boreholes than they did in the early 1990s. Most important, depend- ence on surface water has declined substantially--from 50 percent to 42 percent in rural areas and from 41 percent to 33 percent overall. To learn how households have moved from one source to another, we analyzed household surveys completed for the time periods 1995 to 2000 and 2001 to 2005. Our analysis used two indicators: annualized change in coverage (expressed as a percentage of the population) and absolute annual change in population coverage. The first indicator is defined as the number of people who gain coverage to each water source each year, divided by the population in the end year. The second indicator is the absolute number of people who move into or out of a specific source each year during the time period. Each year during the decade from 1995 to 2005, about 400,000 peo- ple were added to the rolls of those who receive piped water (figures 2.5 and 2.6). In other words, the absolute number of people who gained piped water obtained through a household connection was that much higher than the number of those who lost it (or who were born into households without piped water). Most of the change came from net- work expansions in Ethiopia, Côte d'Ivoire, and Senegal--partially offset by contractions of coverage in Nigeria and Tanzania. Ethiopia annually moved about 300,000 people to piped-water service between 1995 and 2005, whereas Nigeria lost about 700,000 people from piped water in the same 10-year period. Other sources--chiefly standposts, wells and boreholes, and surface water--recorded an increase in use. The rise in the number of people using surface water is primarily due to changes in the Democratic Republic of Congo, where more than 4 million people each year are added to the rolls of surface-water users. 54 Table 2.8 Evolution of Water-Supply Sources, 1990­2005 (percentage of population using source) Household connection to piped water Standposts Wells and boreholes Surface water Urban Rural Overall Urban Rural Overall Urban Rural Overall Urban Rural Overall 1990­95 50 4 18 29 9 15 20 41 37 6 50 41 1995­2000 43 4 17 25 9 15 21 41 38 5 41 31 2001­05 39 4 17 24 11 16 24 43 41 7 42 33 Source: Banerjee, Wodon, and others 2008. Access to Safe Water: The Millennium Challenge 55 Figure 2.5 Annualized Change in Coverage of Various Water Sources, 1995­2005 a. Absolute annual change in population coverage, urban 8.0 population gaining access annually (millions) 6.0 4.0 2.0 0 ­2.0 piped water standposts well/boreholes surface water ­4.0 b. Absolute annual change in population coverage, rural 3.5 3.0 population gaining access annually (millions) 2.5 2.0 1.5 1.0 0.5 0 ­0.5 ­1.0 ­1.5 piped water standposts well/boreholes surface water Source: Banerjee, Wodon, and others 2008. A few outliers emerge as exceptions to the generally mediocre picture. Senegal stands out as having the largest average annual gain in piped-water coverage, adding almost 2 percent of its population each year, immedi- ately followed by Benin (table 2.9). By contrast, the Democratic Republic of Congo, Malawi, Nigeria, Rwanda, Tanzania, and Zambia reduced their 56 Africa's Water and Sanitation Infrastructure Figure 2.6 Annualized Change in Coverage of Various Water Sources, 1995­2005 a. Urban 2.0 annualized increase in access (% population) 1.5 1.0 0.5 0 ly s s er st le pp at o ho w dp su re ce an d bo rfa pe st ls/ pi su el w b. Rural 2.0 annualized increase in access (% population) 1.5 1.0 0.5 0 ly ts s er le pp os at ho ­0.5 w dp su re ce an d bo rfa pe st ls/ pi su el w Source: Banerjee, Wodon, and others 2008. coverage between the late 1990s and the early 2000s. In the case of pub- lic standposts, Mali has achieved the most accelerated expansion, fol- lowed by Benin. On the opposite side of the spectrum, Lesotho, Malawi, and Nigeria recorded reductions in access to standposts. Uganda was by far the leader in enhancing well and borehole coverage, adding almost 7 percent of its population each year. Another way to assess national water-supply performance is to rank countries in terms of their success in reducing reliance on surface water. From this angle, the progress is far from dramatic. Uganda also stands out for moving almost 3 percent of its population away from surface water every year, immediately followed by Lesotho. In other countries, less than 2 percent of the population has moved away from surface water every year, although reliance on surface water has actually risen in the Democratic Republic of Congo, as noted, and in several other countries. Table 2.9 Annualized Change in Coverage by Water Source and by Country, 1995­2005 (percentage of population accessing source) Household piped water Standposts Wells/boreholes Surface water Senegal 1.98 Mali 2.14 Uganda 6.53 Uganda ­2.75 Benin 1.78 Benin 1.88 Lesotho 3.75 Lesotho ­2.45 Zimbabwe 1.69 Burkina Faso 1.40 Nigeria 3.60 Mozambique ­1.81 Côte d'Ivoire 1.47 Tanzania 1.36 Mozambique 2.95 Namibia ­1.19 Namibia 1.43 Madagascar 1.18 Malawi 2.69 Cameroon ­0.99 Mali 0.83 Congo, Dem. Rep. 1.12 Rwanda 2.51 Ghana ­0.88 Burkina Faso 0.69 Ethiopia 1.09 Ghana 2.09 Côte d'Ivoire ­0.72 Cameroon 0.45 Côte d'Ivoire 0.94 Niger 1.55 Zimbabwe ­0.42 Ethiopia 0.44 Cameroon 0.81 Guinea 1.45 Benin ­0.40 Niger 0.27 Uganda 0.71 Cameroon 1.33 Nigeria ­0.39 Ghana 0.26 Namibia 0.57 Côte d'Ivoire 1.16 Ethiopia ­0.36 Chad 0.23 Niger 0.52 Namibia 1.06 Guinea ­0.16 Mozambique 0.16 Guinea 0.49 Zambia 0.95 Senegal ­0.07 Uganda 0.12 Senegal 0.44 Ethiopia 0.89 Niger ­0.02 Kenya 0.09 Rwanda 0.33 Tanzania 0.82 Mali 0.28 Guinea 0.08 Ghana 0.30 Chad 0.73 Tanzania 0.50 Madagascar 0.03 Mozambique 0.28 Zimbabwe 0.52 Madagascar 0.53 Lesotho 0.00 Chad 0.26 Kenya 0.40 Malawi 0.60 Malawi ­0.09 Kenya 0.20 Madagascar 0.25 Zambia 0.66 Zambia ­0.13 Zambia 0.19 Benin ­0.39 Kenya 0.86 Congo, Dem. Rep. ­0.16 Zimbabwe ­0.02 Mali ­0.57 Chad 1.20 Rwanda ­0.39 Malawi ­0.31 Burkina Faso ­0.77 Rwanda 1.81 Nigeria ­0.57 Lesotho ­0.47 Senegal ­0.99 Burkina Faso 2.31 Tanzania ­1.01 Nigeria ­0.66 Congo, Dem. Rep. ­4.75 Congo, Dem. Rep. 7.53 57 Source: Banerjee, Wodon, and others 2008. 58 Africa's Water and Sanitation Infrastructure Although most of Sub-Saharan Africa is not on track to reach the water MDG by 2015, a handful of countries have made remarkable progress in expanding access to improved sources of water, and at a rate that substantially exceeds their peers. This group includes Benin, Burkina Faso, Mali, and Senegal, which have moved a substantial share of their population to improved sources of piped-water connections or stand- posts. Lesotho, Nigeria, and Uganda have experienced the largest gains in expanding well and borehole coverage. The performance of four of those countries is tracked in figure 2.7. Figure 2.7 Four Solid Performers in Expanding Access to Safe Water, 1995­2005 a. Benin b. Senegal 10 10 % population gaining access every year % population gaining access every year 8 8 6 6 4 4 2 2 0 0 er s s er er ts s er st le le ­2 os at at at at ­2 po ho ho w w w w dp re re nd d ce d ce an bo bo pe pe a rfa fa st st ls/ ls/ pi pi r su su el el w w c. Mali d. Uganda 10 10 % population gaining access every year % population gaining access every year 8 8 6 6 4 4 2 2 0 0 ­2 ­2 er s s er st le at at po ho w w er ts s er re d le d ce ­4 os an at at bo ho pe rfa w w dp st re ls/ pi d ce su an bo pe el rfa st w ls/ pi su el w Source: Banerjee, Wodon, and others 2008. Access to Safe Water: The Millennium Challenge 59 When we analyze rural and urban spaces in isolation, other leaders emerge. Benin, Namibia, and Senegal each have managed to move more than 1 percent of their rural population to piped water supplied through household connections. Benin has also succeeded in raising standpost access in rural areas. Mali has provided standpost access to an additional 3 percent of its rural population each year. The biggest success story in well and borehole coverage is Uganda, where slightly more than 7 percent of the rural population has converted to this source of water each year. As noted, Uganda is also a major success story in reducing dependence on surface water. In the urban water space, Ethiopia stands out as having achieved the largest average annual gain in household connections to piped water, adding almost 5 percent of its urban population each year, immediately followed by Côte d'Ivoire. By contrast, the Democratic Republic of Congo, Malawi, Nigeria, Rwanda, Tanzania, and Zambia slipped in their urban piped-water connections between the late 1990s and the early 2000s. In the case of pub- lic standposts, Uganda achieved the most accelerated expansion in urban areas, followed closely by Burkina Faso (which also did well with household connections). On the opposite side of the spectrum, Côte d'Ivoire, Lesotho, and Nigeria recorded urban dwellers' declining access to standposts. Nigeria, Malawi, and Rwanda were by far the leaders in enhancing well and borehole coverage, more than 4 percent of its urban population each year. Notes 1. With a target date of 2015, MDG number 7 calls for ensuring environmental sustainability and--central to this analysis--reducing the number of people without sustainable access to safe drinking water by half. 2. These figures are likely to have changed. Since this study was completed, the water utility has undertaken, with apparent success, a new program focusing on token-run standposts. 3. For all the cities for which we could only rely on the utility's information, cov- erage was calculated this way. 4. Personal communication with Seydou Traore, Water and Sanitation Program, on September 25, 2007. References Banerjee, S., H. Skilling, V. Foster, C. Briceño-Garmendia, E. Morella, and T. Chfadi. 2008. "State of the Sector Review: Rural Water Supply." AICD Working Paper, World Bank, Washington, DC. 60 Africa's Water and Sanitation Infrastructure Banerjee, S., Q. Wodon, A. Diallo, N. Pushak, H. Uddin, C. Tsimpo, and V. Foster. 2008. "Access, Affordability and Alternatives: Modern Infrastructure Services in Sub-Saharan Africa." AICD Background Paper 2, World Bank, Washington, DC. Blackden, C. M., and Q. Wodon, eds. 2006. Gender, Time Use, and Poverty in Sub- Saharan Africa. Washington, DC: World Bank. Boyer, A. 2006. "Survey of Household Water Resale Activity in Peri-Urban Maputo: Preliminary Discussion of Findings." Water and Sanitation Program, Maputo, Mozambique. Collignon, B., and M. Vézina. 2000. "Independent Water and Sanitation Providers in African Cities: Full Report of a Ten-Country Study." Water and Sanitation Program, Washington, DC. Diallo, A., and Q. Wodon. 2005. "A Note on Access to Network-Based Infrastructure Services in Africa: Benefit and Marginal Incidence Analysis." Unpublished paper, Africa Poverty Department, World Bank, Washington, DC. ------. 2007. "Demographic Transition Towards Smaller Household Sizes and Basic Infrastructure Needs in Developing Countries." Economics Bulletin 15 (11): 1­11. Foster, V., and M. C. Araujo. 2004. "Does Infrastructure Reform Work for the Poor? A Case Study from Guatemala." Policy Research Working Paper 3185, World Bank, Washington, DC. Hall, D., and D. Cownie. 2002. "Ability and Willingness to Pay for Urban Water Supply: An Assessment of Connected and Unconnected Households in Maseru, Lesotho." Africa Urban and Water Department, World Bank, Washington, DC. Harvey, P. A, and R. A. Reed. 2006. "Sustainable Supply Chains for Rural Water Supplies in Africa." Engineering Sustainability 159 (1): 31­39. Kariuki, M., B. Collignon, B. Taisne, and B. Valfrey. 2003. "Better Water and Sanitation for the Urban Poor: Good Practice from Sub-Saharan Africa." Water Utility Partnership for Capacity Building, Abidjan. Kariuki, M., and J. Schwartz. 2005. "Small-Scale Private Service Providers of Water and Electricity: A Review of Incidence, Structure, Pricing and Operating Characteristics." Policy Research Working Paper 3727, World Bank, Washington, DC. Keener, S., and S. G. Banerjee. 2005. "Measuring Consumer Benefits from Utility Reform: An Exploration of Consumer Assessment Methodology in Sub- Saharan Africa." Africa Post-Conflict and Social Development Department, World Bank, Washington, DC. Keener, S., M. Luengo, and S. G. Banerjee. 2009. "Provision of Water to the Poor in Africa: Experience with Water Standposts and the Informal Water Sector." AICD Working Paper 13, World Bank, Washington, DC. Access to Safe Water: The Millennium Challenge 61 Komives, K., V. Foster, J. Halpern, and Q. Wodon. 2005. "Water, Electricity, and the Poor: Who Benefits from Utility Subsidies?" Water and Sanitation Unit, World Bank, Washington, DC. Water and Sanitation Program. 2006. "Spare Parts Supplies for Handpumps in Africa: Success Factors for Sustainability." Field Note 15, Africa Region, World Bank, Nairobi. CHAPTER 3 Access to Safe Sanitation: The Millennium Challenge To meet the sanitation target articulated in the Millennium Development Goals (MDGs), the number of people with proper sanitation in Africa needs to more than double--from 242 million in 2006 to 615 million in 2015. Some countries are closer to meeting the target than others. This chapter focuses on sanitation coverage trends in Africa with an analysis of progress made in the past decade. It then goes on to identify the countries that have managed to raise a substantial population up from the lower end of the sanitation ladder. The Predominance of On-Site and Traditional Pit Latrines Waterborne sewerage systems are rare in Africa. Only half of Africa's large cities have sewerage networks, and only Namibia, Senegal, and South Africa provide universal sewerage access. Sewerage networks that reach just about 10 percent of the population within the service area, such as those in Côte d'Ivoire, Kenya, Lesotho, Madagascar, Malawi, and Uganda, are more typi- cal. Little more than half the households with piped water also have flush toilets, which are often connected to septic tanks rather than to sewers. This is not surprising given that development of waterborne sewerage networks generally lags substantially behind the evolution of the piped- water networks on which they depend. In the low-income countries of 63 64 Africa's Water and Sanitation Infrastructure Africa, only 15 percent of the population enjoys private connections to piped-water networks, and this already places a low ceiling on the poten- tial for waterborne sewerage (figure 3.1). Sanitation is predominantly on-site and typically takes the form of traditional pit latrines. Half of the population uses traditional latrines, and the rate of use is approximately equal in both urban and rural areas. Overall, one-third of the population practices open defecation. Curiously, the number of improved latrines is not much greater than that of septic tanks, despite a significant cost difference between the two. An urban-rural divide emerges when access to improved sanitation is considered. In rural areas, 41 percent of the population continues to practice open defecation, and improved sanitation modalities reach less than 10 per- cent. Conversely, in urban areas, 39 percent has access to improved modalities such as improved latrines or septic tanks, and less than 10 per- cent practices open defecation. Africa's low overall access rates to improved sanitation are partly due to negligible service coverage in rural areas, where most people still reside (table 3.1). Traditional latrines are the most common sanitation option in Africa, but the health benefits they provide depend on how they are constructed and used. Even basic latrines can provide protection if they are covered and emptied in a timely fashion, and if people wash their hands after use. Conversely, improved latrines will not provide sanitary protection if peo- ple do not use them properly, or do not use them at all. Figure 3.1 Population That Has Wastewater Connection in the Utility Service Area 100 90 80 % population 70 60 50 40 30 20 10 0 ro , Se ia un D out gal lit S, T rica a ut ica n) Na a M N th A a ici SC frica SO SW Nam a CI , Z ia KI te d bia CO ire A, , Za a NW dag ia , U ar M A, L da rp , T ho ia kin an o ek bu out ani ba ity, fric i bi AM SC eny as M NAS mib ou ib DE SC ib b ON ratio zan SC asc d n AS 'lvo Co SA sot et un , So Afr ne lit am , C am M m aF f m AS ga , B Su kw ho Jo y, S nz ici UW h A A JIR LW , K an O Na e pa , Z h (D pa h a EA n, y, y, a ur ô lit ,S ,S un W l pa W W W i M M rg o ici ici et pa un ur n M er w at ay ti ro To ka M W sB pe ha n um he ind vi ei Os Ca al st in to eT W W en ar ak Kh Dr Source: Morella, Foster, and Banerjee 2008. Access to Safe Sanitation: The Millennium Challenge 65 Table 3.1 Patterns of Access to Sanitation (percent) Open Traditional Improved defecation latrine latrine Septic tank Urban 8 51 14 25 Rural 41 51 5 2 Overall 31 51 8 10 Source: Banerjee and others 2008. Figure 3.2 Population Sharing Water and Toilet Facilities 50 45 40 35 % countries 30 25 20 15 10 5 0 0­20 20­40 40­60 60­80 population (%) Source: Morella, Foster, and Banerjee 2008. In urban areas, sanitation facilities are typically shared among multiple families. Household surveys focus only on formal service provision and do not take into account informal sharing between households. In urban areas, more than 40 percent of households report sharing toilet facilities with other households (figure 3.2). In Benin, Burkina Faso, the Democratic Republic of Congo, Ghana, Guinea, and Madagascar, more than half of households share toilet facilities. In Ghana--where compound housing is commonplace--as many as 80 percent of urban dwellers share water and sanitation facilities with other households. This practice suggests that peo- ple lose time waiting to access facilities and may also pay significant sur- charges to the facility owners. Shared facilities are often poorly maintained, which poses health risks and may discourage use. 66 Africa's Water and Sanitation Infrastructure Figure 3.3 Access Patterns across Income Quintiles 100 90 80 70 % population 60 50 40 30 20 10 0 Q1 Q2 Q3 Q4 Q5 quintile septic tank improved latrine traditional latrine open defecation Source: Banerjee, Wodon, and others 2008. Patterns of sanitation access vary dramatically across the socioeco- nomic spectrum. As might be expected, open defecation is more widely practiced by those in the lowest income groups, where it accounts for half of the population and declines steadily toward zero prevalence in the highest income groups. Conversely, the poorest half of the population has virtually no access to improved latrines and septic tanks; even among the richest strata, barely 20 to 30 percent of households have such access (figure 3.3). The figures indicate that although improved latrines cost less than septic tanks, they remain something of a luxury, even for the middle-income groups. As well, although high average rates might suggest comprehensive coverage, the numbers are somewhat misleading, because people in higher income groups are generally the ones benefiting from these sanitation improvements, and those in the more vulnerable popula- tions are left without adequate coverage. Finally, traditional latrines are by far the most egalitarian form of sanitation, used in about 50 percent of households across all income ranges. The Sanitation Challenge across Countries In most countries, well below 10 percent of the population has septic tanks and less than 20 percent has improved latrines. The difference is made up, in varying degrees, by traditional pit latrines and/or open defecation. Access to Safe Sanitation: The Millennium Challenge 67 Fifty-one percent of the population uses pit latrines, and this number remains remarkably constant between urban and rural areas and across the socioeconomic spectrum. In Malawi, Tanzania, and Uganda, as much as 80 percent of the population uses traditional pit latrines. These general patterns masks huge differences in access to different modalities of sani- tation throughout the African countries (table 3.2). In all countries, the patterns of access between urban and rural areas differ greatly. In Zimbabwe, 95 percent of urban residents use septic tanks, but rural coverage is less than 2 percent. In Namibia, Senegal, and South Africa more than 50 percent of the urban population has access to septic tanks; the numbers in rural areas range from 14 percent (Senegal) to 6 percent (South Africa). Burkina Faso has the best coverage of improved latrines in urban areas, where 70 percent of the population uses this type of facility. Yet, in rural areas, coverage is 10 times smaller, down Table 3.2 Patterns of Access to Flush Toilets and Alternatives (percentage of households, population-weighted average) Improved Traditional Open Septic tank latrine latrine defecation By time period (national) Early 1990s 9 6 50 46 Late 1990s 9 7 47 37 Early 2000s 10 9 52 34 By location Rural 2 5 52 41 Urban 28 14 49 8 By quintile First 0 0 50 49 Second 1 2 54 41 Third 4 6 57 32 Fourth 12 11 54 23 Fifth 34 19 40 6 By country income group Low 7 8 52 33 Middle 33 8 41 13 By subregion East Africa 4 4 56 35 West Africa 12 8 48 33 Southern Africa 23 11 36 28 Central Africa 3 13 65 18 Source: Banerjee, Wodon, and others 2008. Note: The total on trend analysis (by time period) may not add to 100 because a balanced panel has been taken in the three time periods. 68 Africa's Water and Sanitation Infrastructure to 7 percent. In Zimbabwe, the unserved population in urban areas is close to zero, as opposed to more than 40 percent in rural areas. In all countries, urban sanitation coverage generally exceeds national averages. In many countries, most of the urban population enjoys septic tanks and improved latrines, while less than 20 percent practices open defecation. Conversely, most of the population in rural areas uses traditional latrines, and no more than 15 percent of the rural population has septic tanks. Open defecation remains common in rural areas, and more than 50 per- cent of the population in half the countries engages in this practice. In a few countries--Benin, Burkina Faso, Chad, Namibia, and Niger--nearly all the rural populations still practice open defecation. Countries can be categorized in three ways, based on their urban san- itation coverage. The first group includes countries where most of the urban population--between 50 and 90 percent--rely on traditional latrines. This is the case of the Central African Republic, Chad, the Comoros, the Democratic Republic of Congo, the Republic of Congo, Ethiopia, Guinea, Lesotho, Malawi, Mali, Mauritania, Mozambique, Nigeria, Sudan, Tanzania, and Uganda. The second group comprises countries where most of the urban population--from one-third to one- half--use improved latrines, along with a significant percentage--20 to 40 percent--who use traditional latrines. This is the case of Benin, Burkina Faso, Cameroon, Ghana, Madagascar, Niger, and Rwanda. The third group includes countries where at least one-third and up to 95 per- cent of the urban populations have septic tanks, although in some coun- tries up to 45 percent still use traditional latrines. This is the case of Côte d'Ivoire, Gabon, Kenya, Namibia, Senegal, South Africa, Zambia, and Zimbabwe. Similarly, countries can be categorized based on coverage in rural areas, but a different group of countries emerge. The first category includes countries where more than 50 percent of the rural population still prac- tices open defecation: This is the case of Benin, Burkina Faso, Chad, Côte d'Ivoire, Ethiopia, Mauritania, Mozambique, Namibia, Niger, and Sudan. The second category includes countries where most use traditional latrines. This is the largest group, including Cameroon, the Comoros, the Democratic Republic of Congo, the Republic of Congo, Gabon, Ghana, Guinea, Kenya, Malawi, Mali, Nigeria, South Africa, Tanzania, Uganda, and Zambia. In the third category, an increasing number of people use improved latrines, although many still use traditional latrines and practice open defecation. This is the case in the Central African Republic, Lesotho, Madagascar, Rwanda, Senegal, and Zimbabwe (table 3.3). Access to Safe Sanitation: The Millennium Challenge 69 Table 3.3 Patterns of Access to Sanitation across Countries (percent) Dominant Range of Average modality prevalence prevalence Countries Urban Flush toilet 30­95 58 Côte d'Ivoire, Gabon, Kenya, Namibia, Senegal, South Africa, Zambia, Zimbabwe Improved latrine 29­67 50 Benin, Burkina Faso, Cameroon, Ghana, Madagascar, Niger, Rwanda Traditional latrine 45­87 68 Central African Republic; Chad; Comoros; Congo, Dem. Rep.; Congo, Rep.; Ethiopia; Guinea; Lesotho; Malawi; Mali; Maurita- nia; Mozambique; Nigeria; Sudan; Tanza- nia; Uganda Rural Improved latrine 11­44 25 Central African Republic, Lesotho, Madagascar, Rwanda, Senegal, Zimbabwe Traditional latrine 50­83 71 Cameroon; Comoros; Congo, Dem. Rep.; Congo, Rep.; Gabon; Ghana; Guinea; Kenya; Malawi; Mali; Nigeria; South Africa; Tanzania; Uganda; Zambia Open defecation 54­94 74 Benin, Burkina Faso, Chad, Côte d'Ivoire, Ethiopia, Mauritania, Mozambique, Namibia, Niger, Sudan Source: Banerjee, Wodon, and others 2008. Steep Increases in the Use of Traditional Pit Latrines Not only are traditional latrines the most common form of sanitation in Sub-Saharan Africa, but they have also been by far the fastest growing one since 1995. Annualized reports show that an estimated 2.8 percent of the urban population and 1.8 percent of the rural population gains access to traditional latrines each year (figure 3.4). This is a much faster rate of growth than expansion of septic tanks and improved latrines together. Given that the MDG target focuses on the two higher-end improved sani- tation options (septic tank, improved latrines), this rapid increase in the number of traditional latrines is not always fully recognized in policy dis- cussions. Expansion rates of improved latrines and septic tanks are four times faster in urban areas than in rural areas. Another piece of good news is that the prevalence of open defecation in Sub-Saharan Africa has finally begun to decline, albeit at a very modest pace. Approximately 0.3 percent 70 Africa's Water and Sanitation Infrastructure Figure 3.4 Annualized Growth in Coverage in Urban and Rural Areas, 1995­2005 a. Urban areas 3.0 2.5 % population gaining access every year 2.0 1.5 1.0 0.5 0 ­0.5 e e n n k n n io ta tri tri at la l la ic ec pt ed f na de se ov tio en pr di op im tra b. Rural areas 3.5 3.0 % population gaining 2.5 access every year 2.0 1.5 1.0 0.5 0 ­0.5 e ne n nk rin tio ta tri t ca la l la ic fe pt d na de e se ov tio en pr di op im tra Source: Morella, Foster, and Banerjee 2008. of the urban population has been moving away from open defecation each year into some form of sanitation service, and the corresponding fig- ure for the rural population is 0.1 percent. Expansion rates across income groups show that the poorest have little access to the best forms of sanitation. The expansion of septic tanks is concentrated in the middle- and upper-income quintiles, reaching a peak in the third quintile, well beyond the growth in the fifth quintile. Although people in all income groups have gained better access to improved latrines, those in the highest income groups have Access to Safe Sanitation: The Millennium Challenge 71 benefited the most. Those in all income groups have also gained more access to traditional latrines; however, this gain has been greatest for those in the lower-income groups. The number of people practicing open defecation decreases only in the second quintile of the distribu- tion (figure 3.5). The geographical distribution of improved sanitation modes shows the rates of development in various countries over the years. Nigeria and Senegal account for much of the increased septic tank coverage, 35 percent and 17 percent, respectively, mainly due to their sizes. Burkina Faso, Madagascar, and Rwanda account for much of the improved latrine growth. For traditional latrines, Nigeria and Ethiopia account for 51 percent of new users. Despite these improvements, the largest populations (70 million people) still practice open defecation in Ethiopia and Nigeria. Good Progress in a Handful of Countries A handful of African countries have been making impressive gains in sanitation since 1990. Although the improvements in these countries may still be too small and too late to meet the sanitation MDG, the successful cases could provide valuable lessons for other countries in the region. The following analysis highlights countries that have had the greatest changes in access to different levels of sanitation. This list was dominated by some of the larger countries, such as Ethiopia and Nigeria, where, as a result of their sizes, even relatively modest percentage changes had major results. In this section, the focus is on countries that have achieved large percentage gains relative to the size of their populations. This signals successful experience, although in the case of the smaller countries this does not prove to be material at the regional level. Any country moving more than 2 percent of its population up any of the rungs of the sani- tation ladder each year can be considered to be making noteworthy progress (table 3.4). Several solid performers emerge. In the case of septic tanks, Senegal stands out as having by far the largest average annual gain, as more than 3 percent of its population gains access to septic tanks each year. As a result, the number of people using a septic tank in Senegal has increased from 9 percent to 36 percent from 1997 to 2005 (figure 3.6). By contrast, Lesotho, Madagascar, and Zambia show declining septic tank coverage from the late 1990s and the early 2000s. 72 Figure 3.5 Growth in Access by Mode and Quintile a. Open defecation, 1996­2000 b. Open defecation, 2001­05 100 100 80 80 percent percent 60 60 40 40 20 20 0 0 Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 quintile quintile c. Traditional latrine, 1996­2000 d. Traditional latrine, 2001­05 100 100 80 80 percent percent 60 60 40 40 20 20 0 0 Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 quintile quintile e. Improved latrine, 1996­2000 f. Improved latrine, 2001­05 100 100 80 80 percent percent 60 60 40 40 20 20 0 0 Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 quintile quintile g. Septic tank, 1996­2000 h. Septic tank, 2001­05 100 100 80 80 percent percent 60 60 40 40 20 20 0 0 Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 quintile quintile Source: Morella, Foster, and Banerjee 2008. 73 74 Table 3.4 Annualized Change in Coverage, 1995­2005 (percentage of population per year) Septic tank Improved latrine Traditional latrine Open defecation Senegal 3.50 Madagascar 6.46 Côte d'Ivoire 4.10 Ethiopia ­2.30 Zimbabwe 1.51 Rwanda 4.59 Uganda 3.96 Zimbabwe ­1.37 Mali 1.02 Burkina Faso 4.43 Ethiopia 3.92 Mozambique ­1.25 Namibia 1.00 Benin 2.53 Congo, Dem. Rep. 3.63 Madagascar ­0.84 Ghana 0.70 Zimbabwe 1.13 Nigeria 2.84 Senegal ­0.84 Nigeria 0.63 Cameroon 0.95 Mozambique 2.79 Guinea ­0.55 Benin 0.48 Mali 0.81 Malawi 2.61 Mali ­0.43 Cameroon 0.38 Lesotho 0.64 Guinea 2.09 Cameroon ­0.29 Ethiopia 0.37 Ghana 0.61 Mali 1.36 Côte d'Ivoire ­0.14 Burkina Faso 0.34 Tanzania 0.57 Zambia 1.08 Congo, Dem. Rep. ­0.05 Tanzania 0.25 Kenya 0.48 Chad 0.90 Malawi ­0.04 Chad 0.23 Guinea 0.34 Ghana 0.79 Rwanda 0.20 Malawi 0.17 Niger 0.32 Kenya 0.77 Nigeria 0.34 Uganda 0.10 Namibia 0.30 Niger 0.63 Namibia 0.35 Côte d'Ivoire 0.08 Congo, Dem. Rep. 0.26 Cameroon 0.57 Uganda 0.38 Kenya 0.05 Zambia 0.20 Zimbabwe 0.52 Zambia 0.42 Guinea 0.04 Uganda 0.20 Tanzania 0.52 Ghana 0.61 Congo, Dem. Rep. 0.04 Mozambique 0.17 Namibia 0.15 Tanzania 0.63 Niger 0.00 Malawi 0.16 Senegal 0.03 Kenya 0.82 Rwanda 0.00 Ethiopia 0.12 Rwanda ­0.44 Benin 0.90 Mozambique 0.00 Chad ­0.52 Lesotho ­0.48 Burkina Faso 1.04 Madagascar ­0.01 Nigeria ­0.68 Benin ­1.08 Lesotho 1.05 Lesotho ­0.09 Côte d'Ivoire ­1.20 Burkina Faso ­2.25 Chad 1.60 Zambia ­0.12 Senegal ­1.29 Madagascar ­3.69 Niger 1.81 Source: Morella, Foster, and Banerjee 2008. Access to Safe Sanitation: The Millennium Challenge 75 Figure 3.6 Successful Examples from Up and Down the Sanitation Ladder, 1995­2005 a. Ethiopia: Getting onto the bottom rung b. Senegal: Mainstreaming septic tanks 8 8 annualized change in coverage annualized change in coverage 6 6 by mode (% population) by mode (% population) 4 4 2 2 0 0 ­2 ­2 ­4 ­4 ­6 ­6 t t e n t t e n le la le la n n tio tio nP nP tri tri i i to to ca ca l la l la Sa Sa sh sh fe fe na na nd nd flu flu de de tio tio l, a l, a en en di di ica ica op op tra tra em em ch ch P, P, VI VI c. Madagascar: upgrading latrines d. Côte d'Ivoire: upgrading latrines 8 8 annualized change in coverage annualized change in coverage 6 6 by mode (% population) by mode (% population) 4 4 2 2 0 0 ­2 ­2 ­4 ­4 ­6 ­6 t at ne n t t ne n ile ile la tio tio Pl nP tri tri to to n ca ca la la Sa Sa sh sh fe fe l l na na nd nd flu flu de de tio tio l, a l, a en en di di ica ica op op tra tra em em ch ch P, P, VI VI Source: Morella, Foster, and Banerjee 2008. Note: VIP = ventilated improved pit. In the case of improved latrines, Burkina Faso, Madagascar, and Rwanda stand out as having achieved accelerated expansion. In Madagascar, about 7 percent of the population has gained improved latrine coverage every year; in Burkina Faso and Rwanda, the corresponding figure exceeds 4 per- cent. In the Democratic Republic of Congo, Côte d'Ivoire, Ethiopia, and 76 Africa's Water and Sanitation Infrastructure Uganda, more than 3 percent of the population has gained access to tradi- tional latrines every year. Another way to quantify success is to identify the countries that have had the most rapid reductions in the number of people practic- ing open defecation. Ethiopia has had the biggest reduction: between 2000 and 2005, the share of the population without access to any form of sanitation dropped from 82 percent to 62 percent. Mozambique and Zimbabwe immediately follow: more than 1 percent of their pop- ulations have stopped the practice of open defecation every year. Nigeria, which has made impressive gains in many areas of sanitation improvement, has not had such rapid reduction in its open defecation rate. Conversely, Senegal continues showing a salient performance on septic tank coverage. Individual countries are focused on different goals, based on their cur- rent levels of sanitation coverage, and the strategies they employ have dis- tinct financial and health implications. In Ethiopia, for example, the main goal is to reduce the practice of open defecation by getting people onto the bottom rung of the sanitation ladder. Ethiopia therefore adopted a culturally appropriate formula rather than simply spending money on hardware, which yielded significant results (box 3.1). Countries such as Burkina Faso and Madagascar are focused on upgrading services for people who are already engaged in some kind of basic sanitation practice. In Senegal, the aim is to move people from the middle to the top of the ladder by building more septic tanks. A similar analysis of a country's performance can be conducted at the urban and rural levels. For example, Senegal appears to have made great strides in septic tank coverage when looking at figures for urban areas, but the increase in rural areas is much less remarkable. The same applies to Zimbabwe, where the noteworthy expansion in urban areas is offset by the decline in septic tank coverage in rural areas. Also, in Burkina Faso there has been little improvement in latrine access in large cities, and the results in rural areas are 10 times smaller. Conversely, Côte d'Ivoire and Uganda show similar progress in traditional latrine coverage across urban and rural areas, as does Ethiopia in reducing the practice of open defecation (table 3.5). Access to Safe Sanitation: The Millennium Challenge 77 Box 3.1 Ethiopia's Success with a Community-Led Program The southern region of Ethiopia--home to diverse cultures and scores of ethnic groups--has a population of 15 million, much larger than many African countries. Population density varies, peaking at 1,100 people per square kilometer in the Wanago district. In early 2003, access to on-site sanitation was lower than 13 percent, below the national average of 15 percent (see figure). Traditional latrines were most prevalent but scarcely used, poorly maintained, smelly, and dangerous to children and ani- mals. Meanwhile, population expansion, growing household densities, and defor- estation were combining to reduce private options for open defecation. Latrine Construction 2002/03 and 2005/06 On-site sanitation in southern region 100 88.8% 80 78.8% % increase 60 51.7% 40 20 12.8% 0 2002/03 2003/04 2004/05 2005/06 annual increase in latrines with 2005/06 projection The Southern Regional Health Bureau, charged with promoting sanitation and hygiene by the national Ministry of Health, applied a community-led total sanitation approach, including zero subsidies but allowing the community to devise its own innovative and affordable models. With a modest but dedicated sum of money, a mass communication cam- paign was launched using the slogan "Sanitation is everyone's problem and everyone's responsibility." It promoted sustainable and affordable sanitation by creating awareness and encouraging self-financing across all income quintiles. Close collaboration with all stakeholders created advocacy consensus building and capacity building, promotion (by community volunteers), and supportive supervision. (continued next page) 78 Africa's Water and Sanitation Infrastructure Box 3.1 (continued) At the household level, women were identified as the main drivers of latrine construction. At public consensus-building meetings, they complained about how open defecation directly affects their lives, highlighting the health risks of contact with feces in the banana plantations and in the fields where they collect fodder for cattle. They also complained of the bad smell and embarrassment of seeing people defecate in the open space. Featured stories cited shame as an important factor in consensus building and a strong motivator for latrine con- struction. Volunteer community health promoters went house to house across villages with health extension workers and members of the subdistrict health committee to persuade households to build latrines, and then they supervised construction. Alongside other gains in public health, pit latrine ownership rose from less than 13 percent in September 2003 to more than 50 percent in August 2004. By August 2005, it had reached 78 percent, and a year later was on track to reach 88 percent. Source: Reproduced from Water and Sanitation Program 2008. Table 3.5 Annualized Change in Coverage by Modality and by Country, 1990­2005 (percent) Septic tank Improved latrine Traditional latrine Open defecation Urban Senegal 5.7 Burkina Faso 17.2 Nigeria 5.1 Malawi 1.0 Zimbabwe 3.0 Madagascar 8.5 Congo, Dem. Rep. 4.7 Rwanda 0.4 Mali 2.3 Rwanda 6.1 Côte d'Ivoire 4.5 Namibia 0.4 Namibia 1.8 Benin 5.3 Uganda 4.4 Tanzania 0.3 Burkina Faso 1.3 Ghana 2.0 Mozambique 4.2 Kenya 0.3 Ghana 1.2 Tanzania 1.8 Ethiopia 3.9 Benin 0.2 Benin 1.2 Mali 1.7 Chad 3.6 Chad 0.1 Ethiopia 1.2 Niger 1.4 Malawi 3.3 Cameroon 0.1 Tanzania 1.1 Cameroon 0.9 Guinea 2.3 Burkina Faso 0.1 Chad 0.9 Congo, Dem. Rep. 0.8 Rwanda 2.2 Uganda 0.1 Malawi 0.9 Uganda 0.7 Niger 2.0 Zambia 0.1 Uganda 0.5 Mozambique 0.6 Kenya 2.0 Zimbabwe 0.0 Nigeria 0.5 Kenya 0.6 Ghana 1.7 Guinea 0.0 Côte d'Ivoire 0.5 Lesotho 0.5 Cameroon 1.6 Ghana ­0.1 Rwanda 0.4 Guinea 0.5 Zambia 0.8 Lesotho ­0.1 Lesotho 0.3 Ethiopia 0.5 Mali 0.7 Senegal ­0.1 Cameroon 0.2 Malawi 0.4 Lesotho 0.6 Niger ­0.2 Madagascar 0.2 Zambia 0.3 Namibia 0.4 Nigeria ­0.2 Congo, Dem. Rep. 0.2 Zimbabwe 0.2 Tanzania 0.2 Mali ­0.4 Zambia 0.0 Namibia 0.2 Zimbabwe 0.0 Côte d'Ivoire ­0.5 Guinea 0.0 Senegal ­0.1 Benin ­3.2 Congo, Dem. Rep. ­0.5 Kenya ­0.1 Nigeria ­0.3 Senegal ­3.8 Mozambique ­0.9 Niger ­0.2 Côte d'Ivoire ­0.9 Madagascar ­5.3 Madagascar ­1.1 Mozambique ­0.4 Chad ­1.6 Burkina Faso ­13.1 Ethiopia ­2.2 79 (continued next page) Table 3.5 (continued) 80 Septic tank Improved latrine Traditional latrine Open defecation Rural Senegal 1.7 Madagascar 5.9 Ethiopia 4.3 Niger 2.5 Mali 0.6 Rwanda 4.6 Côte d'Ivoire 3.9 Burkina Faso 1.6 Namibia 0.5 Zimbabwe 1.8 Uganda 3.9 Chad 1.5 Nigeria 0.5 Burkina Faso 1.7 Congo, Dem. Rep. 3.1 Ghana 1.5 Ethiopia 0.3 Lesotho 1.1 Senegal 2.6 Kenya 1.0 Zambia 0.2 Benin 1.0 Malawi 2.5 Benin 0.9 Burkina Faso 0.1 Mali 0.6 Guinea 2.1 Tanzania 0.7 Guinea 0.1 Kenya 0.4 Mali 1.7 Lesotho 0.6 Benin 0.1 Namibia 0.4 Mozambique 1.7 Nigeria 0.5 Chad 0.1 Guinea 0.3 Nigeria 1.3 Namibia 0.5 Côte d'Ivoire 0.1 Tanzania 0.2 Zambia 1.1 Uganda 0.4 Malawi 0.0 Zambia 0.2 Zimbabwe 0.9 Zambia 0.3 Uganda 0.0 Uganda 0.1 Tanzania 0.6 Congo, Dem. Rep. 0.2 Niger 0.0 Malawi 0.1 Cameroon 0.6 Rwanda 0.1 Kenya 0.0 Ethiopia 0.1 Ghana 0.5 Malawi ­0.2 Mozambique 0.0 Congo, Dem. Rep. 0.0 Kenya 0.5 Cameroon ­0.2 Rwanda 0.0 Niger ­0.1 Chad 0.4 Côte d'Ivoire ­0.4 Tanzania 0.0 Chad ­0.1 Benin 0.4 Mozambique ­0.8 Congo, Dem. Rep. 0.0 Mozambique ­0.1 Niger 0.3 Mali ­0.8 Madagascar 0.0 Cameroon ­0.1 Namibia 0.1 Guinea ­0.9 Ghana 0.0 Ghana ­0.8 Burkina Faso ­0.3 Madagascar ­1.0 Zimbabwe ­0.1 Nigeria ­0.9 Lesotho ­0.5 Senegal ­1.0 Lesotho ­0.1 Côte d'Ivoire ­1.3 Rwanda ­1.3 Zimbabwe ­1.5 Cameroon ­0.1 Senegal ­2.1 Madagascar ­3.1 Ethiopia ­2.8 Source: Morella, Foster, and Banerjee 2008. Access to Safe Sanitation: The Millennium Challenge 81 References Banerjee, S., Q. Wodon, A. Diallo, N. Pushak, H. Uddin, C. Tsimpo, and V. Foster. 2008. "Access, Affordability and Alternatives: Modern Infrastructure Services in Sub-Saharan Africa." AICD Background Paper 2, World Bank, Washington, DC. Morella, E., V. Foster, and S. Banerjee. 2008. "Climbing the Ladder: The State of Sanitation in Sub-Saharan Africa." AICD Background Paper 13, World Bank, Washington, DC. Water and Sanitation Program. 2008. "Can Africa Afford to Miss the Sanitation MDG Target? A Review of the Sanitation and Hygiene Status in 32 Countries." World Bank, Washington, DC. CHAPTER 4 Improving the Organization of the Water and Sanitation Sectors Many African governments have reformed their water and sanitation sys- tems to provide quality services for their citizens. The sector reforms are critical in creating the necessary institutional structure for improved ser- vice delivery, but, although costs are predominantly paid up front, it takes time to reap the benefits, and costs are sometimes not shared equitably among the various stakeholders. Most African countries are taking gradual steps, cautiously weighing the benefits and costs based on their socioeconomic conditions. Governments have approached the reform process in various ways, but because most of the documentation of these processes is anecdotal rather than systematic, it is difficult to assess their impact or to replicate successful programs. Collecting this kind of data in Sub-Saharan Africa is challenging, and the situation is made worse by the relatively limited history of the monitoring and evaluation efforts related to the Millennium Development Goals (MDGs), as well as by the broader context of weak institutional capacity. In this chapter, the sector organization and market structure are assessed in four distinct water and sanitation spaces with a focus on devel- oping succinct indexes on the institutional development: urban piped water, standposts and other informal services in the unconnected market, rural water, and sanitation. The indexes are a standardized survey-based 83 84 Africa's Water and Sanitation Infrastructure methodology that employs categorical values (0s and 1s), and the ques- tions in the Africa Infrastructure Country Diagnostic (AICD) Water Supply and Sanitation (WSS) Survey (modules 1 and 3) require an implicit judgment of what is commonly accepted as good practice in other developing countries. The 1s are added to create a composite index for each country. These indexes from the questionnaire responses allow the ranking of institutional maturity and where the country stands at this point in time. It is important to note that a 100 percent score does not imply that there is no scope for improvement. The Heterogeneity of the Urban Water Market No consistent set of institutional arrangements is found across Sub- Saharan Africa. Institutional structures range from national-level utilities responsible for countrywide coverage to those with limited jurisdictions. Generally, the central government is responsible for managing the urban water sector, but several providers, including municipal agencies, public- private partnerships, and corporate utilities, also deliver services. Some utilities are responsible for water, sanitation, and even energy, whereas others handle only water distribution. Generally, water utilities are dedicated to providing water and, in some cases, wastewater facilities. Half of the countries have utilities that jointly provide water and waste- water services (figure 4.1). Only ELECTRA in Cape Verde, Société Tchadienne d'Eau et d'Électricité (STEE) in Chad, Jiro sy Rano Malagasy Figure 4.1 Range of Institutional Arrangements in Water Service Provision 100 80 % countries 60 40 20 0 water and water and urban and rural decentralized vertically energy sanitation combined integrated combined combined yes no Source: Banerjee, Skilling, and others 2008a. Improving the Organization of the Water and Sanitation Sectors 85 (JIRAMA) in Madagascar, and ELECTROGAZ in Rwanda provide both water and energy. Few countries in Africa have unbundled bulk water generation and distribution facilities. Most utilities primarily cover urban areas. In Benin, Kenya, Rwanda, South Africa, and Tanzania, utilities pro- vide services to both urban and rural dwellers. The urban water scorecard is a snapshot of three key institutional dimensions: broad sectoral policy reforms, amount and quality of regula- tion, and enterprise governance. It is composed of three indexes: the urban reform index, the regulation index, and the state-owned enterprise (SOE) governance index. Table 4.1 shows the components of these indexes. First, we define reform parameters as the implementation of sec- toral legislation, restructuring of enterprises, and introduction of policy oversight and private sector participation. Second, autonomous, transpar- ent, and accountable regulatory agencies and regulatory tools (tariff methodology) should be established to monitor quality. Third, to prop- erly maintain facilities, SOEs should encourage shareholder participation, create greater board and management autonomy, and improve account- ing and disclosure mechanisms. They should also consider various forms of management, including outsourcing to the private sector. Note that reform and regulation are country-level indicators, but governance is measured at the enterprise level. Urban Water Reforms across Countries The urban water sector reform was evaluated based on four attributes: legislation, restructuring, policy oversight, and private sector involvement (Vagliasindi and Nellis 2009). At the country level, each subindex is expressed as a percentage of positive responses to the binary questions to the total number of indicators. The urban reform index is an average of these four subindexes; each subindex carries the same weighting. Most African countries have undertaken at least one key reform step. One way to establish a transparent framework for service provision is to outline a water policy that includes the government's sector goals and institutional commitments. In most countries, governments have recently begun the reform process; only eight countries have sector legislation more than five years old. Côte d'Ivoire passed a water law in 1973, but most countries implemented water legislation only in the past decade. As of 2005, all but five countries have established water policies, and two of those countries are in the process of drafting water policies. The most common reform steps are corporatization and the passing of a private sector participation law. However, the passing of a law does not 86 Africa's Water and Sanitation Infrastructure Table 4.1 Urban Reform, Regulation, and the SOE Governance Index Reform Internal governance Legislation Ownership and shareholder quality Existence of de jure reform Concentration of ownership Implementation of reform Corporatization/limited liability Restructuring Rate of return and dividend policy Unbundling/separation of business lines Managerial and board autonomy SOE corporatization Autonomy in hiring/firing/wages/ Existence of regulatory body production/sales Policy oversight Size of the board Oversight of regulation monitoring Presence of independence directors outside the ministry Accounting, disclosure, and performance Dispute arbitration outside the ministry monitoring Tariff approval outside the ministry Publication of annual report Investment plan outside the ministry International Financial Reporting Standards/ Technical standard outside the ministry external audits/independent audit Private sector involvement Audit publication Private de jure/de facto Remuneration of noncommercial activity Private sector management/ Performance contracts/with incentives investment ownership Penalties for poor performance Absence of distressed/renegotiation/ Monitoring/third party monitoring renationalization External governance Regulation Labor market discipline Autonomy Restriction to dismiss employees Formal autonomy on hiring/firing Wages compared with private sector Financial autonomy (partial/full) Benefits compared with private sector Managerial autonomy (partial/full) Capital market discipline Multisectoral agency/commissioners No exemption from taxation Transparency Access to debt compared with Publication of decisions via report/ private sector Internet/public hearing No state guarantees Accountability Public listing Existence of appeal Outsourcing Independence of appeal (partial/full) Billing and collection Tools Meter reading Existence of tariff methodology/ Human resources tariff indexation Information technology Existence of regulatory review Length of regulatory review Source: Vagliasindi and Nellis 2009. Note: SOE = state-owned enterprise. Improving the Organization of the Water and Sanitation Sectors 87 guarantee private sector participation. Although 83 percent of countries have legalized private participation, only 63 percent have been able to attract some kind of private participation in any of the three largest util- ities in their respective countries. Private providers have entered into management contracts in only half of the countries and have invested in water sectors in only 5 percent of cases. Leases have been used widely, and management contracts are the second most common form of private par- ticipation. The cancelation rate of private sector contracts for water supply has been much higher in countries in Africa than in other developing coun- tries. Approximately 29 percent of private contracts for water supply have been prematurely terminated. As a result, just a handful of private opera- tors are still active: one each in Cameroon, Cape Verde, Côte d'Ivoire, Gabon, Ghana, Mozambique, Niger, and Senegal, and four in South Africa. The private sector is disproportionately more involved in the West African francophone countries (Côte d'Ivoire, Guinea, Niger, and Senegal), with some exceptions (Mozambique and Uganda). Senegal's successful private sector experience is presented in box 4.1. Another dis- tinctive feature of the African experience has been the use of concessions for joint power and water utilities, as in Gabon and Mali. Only a single divestiture has occurred: the 1999 sale of 51 percent of equity in the water company in Cape Verde. Policy oversight is relatively well defined in Africa. In at least half of the countries studied, functions such as tariff approval, investment plans, tech- nical standards, regulation monitoring, and dispute arbitration are clearly allocated to bodies other than the line ministries, such as special entities within the ministries, interministerial committees, or regulators. Oversight of economic regulation and tariff setting by bodies other than the line min- istries exists in 78 and 65 percent of the countries, respectively. Progress in restructuring has been relatively slow. Only five countries-- Burkina Faso, Namibia, Niger, South Africa, and Uganda--have separated bulk-water production from the distribution function. In the other coun- tries, the functions are performed in tandem, by the same utility. Niger has made the most progress and reports a score of more than 80 percent on the restructuring subindex. In 2000, the water company Société Nationale des Eaux (SNE) in Niger was separated into the asset-holding company, Société de Patrimoine des Eaux du Niger (SPEN), and a private operator, Société de Exploitation des Eaux du Niger (SEEN), responsible for production, transmission, and distribution in the urban areas (World Bank 2007). 88 Africa's Water and Sanitation Infrastructure Box 4.1 Senegal's Successful Experience with Private Sector Participation The Senegal experience under the affermage is characterized by two remarkable results: first, an impressive expansion of access, and second, a large increase in op- erational efficiency that mainly originated from a reduction of nonrevenue water (NRW). The first result was mainly related to a massive subsidized connection pro- gram sponsored by donors and, in part, to the cash-flow surplus generated by the private operators. In particular, the social connection program, implemented with donor support, provided about 129,000 connections (75 percent of all new connections installed) benefiting poor households living in targeted neighborhoods. A portion of the new connections, however, ended up discon- nected, despite tariffs declining in real terms up to 2006 and the social tariff corresponding to a consumption of six cubic meters per month--mostly applying to poor households. The second result was strictly related to contract innovations geared toward increasing the operator's incentives to perform efficiently. In particular, the contract included targets for NRW reduction and bill collection backed by financial penalties for noncompliance. These targets were then applied to a notional sales volume based on the amount of water actually produced, which was used to determine the operator's remuneration in lieu of the actual water sold. Whenever the operator fell short of the NRW and bill collection targets, the notional sales volume would be lower than the actual sales, penalizing the operator. Another innovation in Senegal's public-private partnership was the responsi- bility of the private operator to finance part of the network's rehabilitation using cash flow. This approach provided the operator with more flexibility to identify and reduce water losses, lessening its dependency on the public asset-holding company. The impact of these innovations on efficiency has been remarkable, making Senegal's affermage a prominent example of private participation in Africa. Today, Senegal can report a level of NRW comparable to the best water utilities in West- ern Europe. These results also confirm that operational efficiency is perhaps the area in which a private operator can make the most positive and consistent impact. Source: Adapted from Marin 2009. Improving the Organization of the Water and Sanitation Sectors 89 Most countries have achieved 40 to 80 percent in the urban reform index. A majority of countries score well on certain subcomponents, but not on others. For instance, Benin scores very high on legislation and pol- icy oversight but very low on restructuring and private sector involve- ment, whereas Rwanda scores high on restructuring and private sector involvement but low on policy oversight and legislation. Côte d'Ivoire, Kenya, Mozambique, Sudan, Tanzania, and Uganda scored more than 50 percent in each of the subindex scores, suggesting a balanced approach to the reform process (figure 4.2). Two Distinct Approaches to Sector Regulation The regulation index is created using four essential attributes of what is conventionally considered a good regulatory framework: autonomy, accountability, transparency, and tools (Vagliasindi and Nellis 2009). The index is an average of these four subindexes and presents a picture of the maturity and depth of the regulatory framework. Anglophone and francophone countries have taken two distinct approaches to sector regulation. About half of the countries (mainly anglo- phone) have established regulatory agencies for the water sector, although a significant number of these do not have private sector participation. Conversely, several francophone countries with private participation have adopted regulatory frameworks without establishing an independent reg- ulatory agency. These approaches appear to be equally effective; in both cases, the established regulatory frameworks typically meet only about half of the corresponding good practice criteria. Line ministries (or subentities), such as ministries of finance/economy and health/environment, continue to play a strong role in the regulation of water services. Parliaments, state water corporations, or asset-holding companies also help to set tariffs or approve investment plans (table 4.2). In some cases, the allocation of regulatory responsibilities is efficient. For instance, monitoring water quality requires different skill sets than those needed to review tariff adjustment proposals. In other cases, the fragmen- tation might create inefficiencies in the sector and a lack of depth in reg- ulatory capacity. The regulatory entities also have a designated responsibility for monitoring and enforcing the license/charter provisions as well as setting customer service regulations. The gaps in water regula- tion fall more within the area of customer service and quality standards. Half of the countries studied have set up regulatory agencies to govern the sector and bring it in the purview of formal rules on tariff and service standards. In the 11 countries with distinct economic regulatory bodies, 90 Africa's Water and Sanitation Infrastructure Figure 4.2 Country Ranking and Prevalence of Key Reform Activities a. Country Cape Verde Mozambique Sudan Niger Côte d'Ivoire Kenya Ghana Madagascar Zambia Uganda Senegal Tanzania South Africa Burkina Faso Nigeria Benin Lesotho Rwanda Ethiopia Namibia Congo, Dem. Rep. Malawi Chad 0 20 40 60 80 100 reform index score (%) b. Prevalence SOE corporatization reform under way recent reform legislation arm's-length oversight PSP in sector management regulatory body PSP in sector investment private sector ownership 0 20 40 60 80 100 countries (%) Source: Banerjee, Skilling, and others 2008a. Note: PSP = private sector participation. Table 4.2 Regulatory Roles in the Urban Water Sector (percent) Entity Line within Regulatory Interministerial Unregulated Role ministry ministry body committee Other or nobody Granting licenses and/or assigning service obligations 57 22 13 9 0 0 Approving investment plans 52 13 13 4 17 0 Establishing technical standards and minimum service levels 40 24 20 8 4 4 Arbitrating in a dispute 36 12 20 12 16 4 Approving tariffs 35 13 22 0 30 0 Setting water quality standards 27 18 23 9 18 5 Monitoring and enforcing compliance with economic rregulation 26 17 30 9 13 4 Providing customer service regulations 26 13 26 9 17 9 Monitoring water quality 26 22 13 9 26 4 Proposing/advising on tariffs 13 25 13 17 33 0 Source: Banerjee, Skilling, and others 2008a. Note: Rows may not add to 100 because roles may be performed by more than one institution. 91 92 Africa's Water and Sanitation Infrastructure 10 were created between 1995 and 2003 (figure 4.3). In Côte d'Ivoire, the regulatory agency, Direction de l'Hydrolique was set up in 1973­74. Of the 11 stand-alone regulators, five have jurisdiction over multiple sec- tors, and the rest are responsible for only WSS activities. The nascent reg- ulators face the challenge of establishing a track record of sound decision making and acquiring competent staff. Most countries are adequately equipped with regulatory tools. Regulatory institutions in a majority of African countries appear to have established a tariff methodology to conduct periodic reviews. Madagascar is the only country that does not have an established set of regulatory tools to manage tariffs. The tariffs in Sub-Saharan Africa are largely regulated-- to the degree that proposals are made and approved. It is sometimes unclear how tariff increases are determined and why they are increased. Most countries use the price cap methodology of adjusting tariffs as opposed to other forms, but some countries raised tariffs based on "rea- sonableness" or to reflect actual costs. Although some countries perform periodic tariff adjustments, few index tariffs on an annual basis. In the 12 countries with periodic tariff reviews, the time between reviews ranges from one to five years. The annual periodic reviews might, in fact, be more comparable to annual indexation. The regulatory agencies are likely to be headed by boards. Only in Côte d'Ivoire and Lesotho are regulators headed by individuals. In all countries, except for Mozambique and Rwanda, the president or the line minister has the authority to appoint the head or commissioner of the regulatory agencies. Clearly, the president and the line ministry play strong roles in the governance of the regulator, and the judicial and leg- islative branches of government play more limited roles. The term limits Figure 4.3 Year of Establishment of Regulatory Agencies Sudan Kenya Lesotho Cape Verde Mozambique 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 Ghana Rwanda Côte d'Ivoire Niger Zambia Tanzania Source: Banerjee, Skilling, and others 2008a. Improving the Organization of the Water and Sanitation Sectors 93 for the head or commissioner vary between three and six years, with an average of 3.3 years. The heads of these institutions can be reappointed, except in Niger, where they serve a single term. Some regulatory agencies have achieved partial financial autonomy. The agencies are most commonly funded by sector levies or license fees, or by the central government. Cape Verde, Mozambique, Niger, and Rwanda use sector levies or license fees to fund the regulatory agencies. Côte d'Ivoire and Lesotho rely completely on the central government for funding, whereas donors play a substantial role in funding the regulators in Ghana, Tanzania, and, to a lesser extent, Sudan (figure 4.4). Almost all countries use a standardized format to compile regulatory reports. Regulatory entities are less likely, however, to share their findings and decisions with the general public. In some cases, there is no mecha- nism to share decisions, but when decisions are made public, this usually occurs in the form of published reports (as in 81 percent of the countries studied). Public hearings are infrequent and held in only 50 percent of the countries. Similarly, hearings are rarely published on the Internet. Consumer participation in the regulatory process is relatively limited. Where consumers have a role in the actual regulatory process, they are most often part of the appeals process rather than reviewers of regulatory proposals or board representatives. A social accountability index includ- ing four indicators1 represents consumer influence in the regulatory process. Burkina Faso, Ghana, Malawi, Tanzania, and Zambia have the most socially accountable regulatory framework. In these countries, consumer Figure 4.4 Understanding Performance in Regulatory Autonomy 100 80 % countries 60 40 20 0 he hire he fire m l m l m l pa ctio ral m l d no ria no cia no ria no cia ar di to to ge to ge bo to an to an ad ad y y y n y o o ris ec yt au ana au a yt au in au l fin by an ju tis m ll f m ul ll m m ed no a fu no rti m al ad to to fu rti au he au pa Source: Banerjee, Skilling, and others 2008a. 94 Africa's Water and Sanitation Infrastructure representation exists in the regulatory body; consumers have the right to comment on draft regulations, review tariff proposals, and appeal regula- tory decisions. Consumer representation is even less frequent within the regulatory body itself. Only in Burkina Faso, Ghana, Namibia, Tanzania, and Zambia do consumers have representation within the regulatory agency (figure 4.5). Only Cape Verde, Kenya, Niger, Senegal, Tanzania, and Uganda have scored higher than 60 percent on the regulation index. A majority of countries have poor regulatory independence across all sectors, demon- strating that the standard model has limited relevance in Africa. Regulatory attributes can be identified that very few countries have adopted, such as formal autonomy to hire and fire, full managerial auton- omy, and full independence to appeal. The countries have neither the authority to hire or fire head commissioners, nor do they allow full inde- pendence to appeal regulatory decisions (figure 4.6). Water Utilities: Halfway toward Good Practice Criteria for Enterprise Governance The SOE governance index is used to determine whether SOEs are being governed using sufficiently commercial principles. Several aspects of SOE management are examined, including ownership and shareholder quality; managerial and board autonomy; accounting, disclosure, and performance monitoring; outsourcing; labor market discipline; and capital market dis- cipline. Using this scoring system, we can see which utilities in Sub- Saharan Africa have adopted policies of good governance and commercial orientation. The goal in governance reforms has been to move toward corporatiza- tion of SOEs, decentralize responsibilities to lower levels of government, and improve the governance of SOEs by adopting modern management methods. In 83 percent of the countries, at least one water utility has been corporatized, thereby laying the foundation for more commercial man- agement. Close to half of the countries sampled have decentralized their water utilities over the past decade, thereby making local communities more responsible for utility management. Lesotho and Zambia began their decentralization processes in the early 1990s, and the rest of the countries decentralized in the past decade. All of the francophone coun- tries studied still have centralized water utilities.2 About 52 percent of the sample utilities are corporatized entities, meaning that the public sector service provider functions as a private Improving the Organization of the Water and Sanitation Sectors 95 Figure 4.5 Prevalence and Key Attributes of the Social Accountability Index a. By country Zambia Tanzania Nigeria Malawi Ghana Burkina Faso Cape Verde Sudan South Africa Senegal Namibia Niger Mozambique Lesotho Congo, Dem. Rep. Rwanda Madagascar Kenya Ethiopia Côte d'Ivoire Benin 0 20 40 60 80 100 social accountability index score (%) b. By key attribute consumer associations have membership in the regulatory body consumers have right to comment draft regulation consumers have right to review tariff proposal consumers have right to appeal regulatory decisions 0 10 20 30 40 50 60 countries (%) Source: Banerjee, Skilling, and others 2008a. 96 Africa's Water and Sanitation Infrastructure Figure 4.6 Country Ranking and Prevalence of Key Attributes of Regulation a. Country Cape Verde Tanzania Uganda Niger Senegal Kenya Zambia Mozambique Ghana Burkina Faso Sudan Rwanda Côte d'Ivoire Congo, Dem. Rep. Nigeria Malawi Chad Ethiopia Namibia Lesotho Benin Madagascar South Africa 0 20 40 60 80 100 regulation index score (%) b. Key attribute transparency regulatory tools accountability autonomy 0 20 40 60 80 100 countries (%) Source: Banerjee, Skilling, and others 2008a. Improving the Organization of the Water and Sanitation Sectors 97 company in terms of efficiency, productivity, and financial sustainability (figure 4.7, panel a). The public sector provider does this through imple- menting some or all of a series of changes, including establishment of a distinct legal identity; segregation of the company's assets, finances, and operations from other government operations; and development of a commercial orientation and managerial independence, while remaining accountable to the government or electorate. Although other utilities are not corporatized, they could be better governed through the adoption of some or all of these corporate practices. The heart of corporate gover- nance is to protect and enhance the long-term value of the company for shareholders (government and other) by increasing sales, controlling costs, and increasing revenue. Nearly half of the African water utilities are SOEs, the majority owned by the central government; others are owned at the state or municipal level. Together, 92 percent are state owned, with ownership varying at dif- ferent levels of government (figure 4.7, panel b). In a few countries, such as Kenya, Namibia, South Africa, and Zambia, where water service deliv- ery has been decentralized to the local level, utilities are majority owned by municipalities. Namibia still provides service through municipal Figure 4.7 Legal Status and Ownership Structure of Water Utilities a. Legal status b. Ownership structure 7% 2% 8% 11% 27% 52% 49% 28% 16% corporatized SOE 100% central government uncorporatized SOE 100% state government limited liability company with shares 100% municipal agency municipal agency mixed holdings traded company Source: Banerjee, Skilling, and others 2008a. Note: SOE = state-owned enterprise. 98 Africa's Water and Sanitation Infrastructure departments, and only the utilities engaged in active public private part- nerships, as in Cape Verde, Côte d'Ivoire, Niger, and Senegal, have diver- sified shareholding. The use of external financial and independent audits is common. Similarly, the management or board determines wages and bonuses in the majority of entities. Utilities perform poorly on indicators such as public listings, outsourcing functions, and dividend payments. Société de Distribution d'Eau de Côte d'Ivoire (SODECI) in Côte d'Ivoire is the only water utility that is listed on a stock exchange; its shares are pub- licly traded. Similarly, only 27 percent of the utilities are required to pay dividends to their shareholders. About 84 percent of the entities have boards of directors, though few are well represented or benefit from the presence of independent direc- tors. Only half the entities have a board with more than five members, and only 40 percent of the entities--notably in Kenya, Tanzania, Uganda, and Zambia--have at least one independent director on the board. Sixty percent have government-appointed boards (figure 4.8). Obviously, the owners' interests are well represented on the boards, and the managerial decision-making process is heavily influenced by politics. Only Société de Exploitation des Eaux du Niger (SEEN), in Niger, has a representative board appointed by shareholders, with independent directors. For instance, in the National Water and Sewerage Company (NWSC) in Uganda, the Ministry of Water, Lands, and Environment appoints the board of directors; in the Office Nationale des Eaux et d'Assainissement Figure 4.8 Performance in Managerial Autonomy 100 80 % utilities 60 40 20 0 ct t oa ers be ard em ale t tio es rin t in t ag t re en s s en hi en fir en w en uc id s es t b ld em o di nd an ide em es em es em es em or rd rs n g g od ec m +b in ho e pr nt d c g cid ag cid ag cid ag ep po re de ana 5 de an de an de an ap sha d e in m m m m ag m Source: Banerjee, Skilling, and others 2008a. Improving the Organization of the Water and Sanitation Sectors 99 (ONEA), the board is appointed by the Council of Ministers (Baietti, Kingdom, and Van Ginneken 2006). Half of the entities have performance contracts with defined and mon- itorable targets. Management through such contracts takes a systematic approach to performance improvement through an "ongoing process of establishing strategic performance objectives; measuring performance; collecting, analyzing, reviewing, and reporting performance data; and using that data to drive performance improvement" (PA Consulting 2007). All entities in South Africa, Tanzania, Uganda, and Zambia use these con- tracts. The NWSC uses annual and multiyear performance contracts (Baietti, Kingdom, and Van Ginneken 2006). Sixty-five percent of the firms use third-party monitoring, which demonstrates a commitment to enhancing external accountability for results. The extent to which these performance contracts are implemented depends on how the internal incentive mechanisms are established. More than half of the utilities have performance-based management systems, and 39 percent penalize for poor performance. In about 57 percent of the utilities, staff members are given periodic performance reviews (figure 4.9). Outsourcing is relatively new and still not widespread. It allows an entity to focus on its core business and potentially lower costs. Utilities in Mozambique and Khartoum, Sudan, are the only utilities that report out- sourcing billing and collection, meter reading, human resources, and information technology. In fact, 88 percent of the utilities score less than 40 percent on the outsourcing subindex. Outsourcing operating expenses can be quite expensive. For instance, outsourcing as a share of operating Figure 4.9 Performance Monitoring 100 80 % utilities 60 40 20 0 pe pa ial pe alti ce pe tra ce rn tiv ce pe aud cal i e d t e s pl in y nd IFRS h it di l au cia ish rt m itor art au itor anc it he ud es rc n an n n te n n rfo id rfo es rfo ct al es rfo it t p ng m hird d ia g t po w ni an co rma ex ince rma p lis ta iti me ch pe rm on m e pu re on - fin ub nc en m r te tiv m al bl al ac co nu rn di pe un an te co de ex in Source: Banerjee, Skilling, and others 2008a. Note: IFRS = International Financial Reporting Standards. 100 Africa's Water and Sanitation Infrastructure expenses of the NWSC in Uganda is in the range of 30 to 40 percent (Baietti, Kingdom, and Van Ginneken 2006). SODECI scores the highest on capital market discipline, which relates to the commercial nature of the utility. Águas de Moçambique (ADeM) in Maputo, SEEN in Niger, FCT in Nigeria, ELECTROGAZ in Rwanda, the South African utilities, and Lusaka Water and Sewerage Company (LWSC) in Zambia also score high on capital market discipline. About 25 percent of the utilities in Africa adhere to the highest levels of labor mar- ket discipline, which relates to the ability to hire and fire workers and to set wages and benefits with regard to the private sector. A majority of the entities score between 40 and 80 percent on the SOE governance index. Africa's state-owned water utilities typically meet only about half the good practice criteria for enterprise governance. Firms do well on "capital market discipline" and "accounting, disclosure, and performance monitoring" subindexes, with more than 60 percent of the utilities scoring between 40 and 60 percent in each subindex. African water utilities rarely outsource. Most are a long way from achieving man- agerial and board autonomy; less than one-fourth score more than 80 per- cent on this subindex. Interestingly, the correlation between the SOE governance index and the earlier reform and regulation index is very low, which is to say that some countries do much better on SOE governance than on reform and regulation, and vice versa. Countries have made more serious efforts to improve internal processes and corporate governance mechanisms during the past decade than in other infrastructure sectors. A growing number of utilities in countries such as Lesotho, Uganda, and Zambia are using performance contracts, though some do not incorporate the penalties, performance- based remuneration, and third-party monitoring that makes these mech- anisms truly effective. The Mozambican and Zambian utilities have the highest scoring internal governance structures when it comes to meeting the needs of their consumers, regulators, governments, and other stake- holders. The LWSC in Zambia is the best-governed utility in Africa according to the criteria developed in this chapter, scoring 73 percent. Johannesburg, SEEN, and Sénégalaise des Eaux (SDE) have also made substantial progress in governance reforms (figure 4.10). In summary, many African countries have initiated water sector reforms, and two major thrusts to this reform agenda have been seen: pri- vate participation and improvement of internal governance. Some coun- tries, such as Burkina Faso, Kenya, Mozambique, Niger, Senegal, Tanzania, Uganda, and Zambia, are remarkable performers that have progressed at Improving the Organization of the Water and Sanitation Sectors 101 Figure 4.10 Country Ranking and Prevalence of Key Attributes of the SOE Governance a. By country Zambia Mozambique Côte d'Ivoire Burkina Faso Sudan Lesotho Senegal Benin Malawi Niger Kenya Congo, Dem. Rep. Uganda Tanzania Rwanda Ethiopia South Africa Madagascar Chad Ghana Cape Verde Nigeria Namibia 0 20 40 60 80 100 SOE governance index score (%) b. By key attribute or subindex performance monitoring managerial autonomy labor market discipline ownership structure capital market discipline outsourcing 0 20 40 60 80 100 utilities (%) Source: Banerjee, Skilling, and others 2008a. Note: SOE = state-owned enterprise. 102 Africa's Water and Sanitation Infrastructure Figure 4.11 Solid Country Performances 100 80 percent 60 40 20 0 o a e r l a da a ga ge ny ni bi qu as an za ne m Ni aF Ke bi Ug Za n Se am in Ta rk oz Bu M urban reform regulation SOE governance Source: Banerjee, Skilling, and others 2008a. Note: SOE = state-owned enterprise. a steady pace in different areas of urban sector reform (figure 4.11). These countries have developed a formal regulatory structure, a market- oriented and accountable internal governance mechanism, and wide- ranging urban sector reforms. Varied Institutional Models for Nonpiped Services in the Urban Water Market Because utilities and standpost operators do not keep track of the differ- ent types of customers they serve, raw coverage numbers conceal essen- tial parts of the urban water picture. Breaking down consumers by type is very important, for example, when it comes to understanding the price structure of the market, because the standpost operator usually charges the direct consumer and the reseller differently. In periurban areas of Accra, although most water is sold through standposts, 20 percent is resold by cart operators (Sarpong and Ambrampah 2006). Likewise, standpost operators in Khartoum sell most of their water (80 percent) to cart operators, who then resell to households (Elamin and Gadir 2006). Similarly, in Ouagadougou, more than 80 percent of water sold at stand- posts is bought by cart operators and not by individuals (Collignon and Vézina 2000). In Luanda, Angola, most of the water delivered in periur- ban areas, where the majority of the population lives, is carried by water trucks that sell water obtained either from the piped-water system or Improving the Organization of the Water and Sanitation Sectors 103 directly from the main river. The water trucks sell to an estimated 10,000 nonmobile water vendors and households that have built water storage tanks; these households in turn sell water to the rest of the population. In periurban areas of Luanda, 70 percent of residents purchase their water from water vendors (Development Workshop 1995). People in urban areas who do not have access to piped water get their water from a number of different sources. There are "formal" sources, such as standposts and boreholes, as well as an emergence of "informal" sources, such as water vendors and tankers, resellers, and small piped systems. The quality of the water from these suppliers is not monitored in the same way as piped water. Public standpipes can be managed by a number of different parties that retain responsibility for payment, supervision, and maintenance. Two main systems are found: one in which the utility retains control, and the other, in which the utility delegates various functions to third parties and serves primarily as a bulk water supplier.3 In a little more than one-quarter of the 24 largest cities studied in the module for small-scale independent providers of the AICD WSS survey, utility staff manages standpipes using one of three management models (free, prepayment, or managed by a paid utility staff member). In almost three-quarters of the cases, utilities had a contract with a third party (whether a private individual or a com- munity organization), a support institution (local government, community- based organization [CBO], or nongovernmental organization [NGO]) to manage the standpipe (table 4.3).4 Direct Management by Utilities. In the past three decades, a shift has occurred so that standpipes that were once owned and provided to the population free of charge by utilities are now run by either private individ- uals or community groups (figure 4.12). The data indicate that many util- ities viewed the free standpipes as a financial drain. As a result, only five of the sample cities still had free standpipes. With the exception of Madagascar, where less than half of the standpipes provide free water, free public standpipes in countries including Namibia, Lesotho, Nigeria, and South Africa were mostly concentrated within larger piped systems or in cities with sufficient levels of piped coverage to help subsidize the costs. Other cities, except for Kaduna, Nigeria, are moving toward paid stand- pipes or kiosks; Johannesburg, Maseru, and Windhoek are installing pre- paid standpipes, and Antananarivo is installing kiosks. The second model, in which the utility hires a salaried attendant, is an increasingly uncommon practice that is still used in a few countries. This 104 Africa's Water and Sanitation Infrastructure Table 4.3 Standpipe Management (percent) free of Management (by) Ownership Country City charge Private Utility Communitya Utility Cape Verde Praia 0 0 100 0 Lesotho Maseru 100 0 97 3 Madagascar Antananarivo 40 60 0 40 Namibia Windhoek 100 0 100 0 Nigeria Kaduna 96 4 96 0 South Africa Johannesburg 100 0 100 0 Sudan Greater Khartoum 0 0 100 0 Zambia Lusaka 0 5 90 5 Private Benin Cotonou 0 100 0 0 Burkina Faso Ouagadougou 0 100 0 0 Niger Niamey 0 100 0 0 Kenya Nairobi 0 88 0 12 Rwanda Kigali 0 100 0 0 Senegal Dakar 0 85 0 15 Community Ethiopia Addis Ababa 0 0 0 100 Malawi Blantyre 0 -- -- 70 Mozambique Maputo 0 44 0 56 Source: Keener, Luengo, and Banerjee 2009. Note: -- = not available. a. In the community category, we merge the delegated management model with direct contracting with a com- munity group and the delegated management model with institution support as discussed later in this section. Figure 4.12 Utility Direct Management Models free-of-charge standpipes utility staff attendant prepayment systems utility utility utility water salary payment salary payment standpipe operator/ consumer standpipe operator vending machine water payment water payment consumer consumer Lesotho, Madagascar, Namibia, Cape Verde, Sudan, and Lesotho, Namibia, South Nigeria, and South Africa Zambia Africa, and Zambia Source: Keener, Luengo, and Banerjee 2009. Improving the Organization of the Water and Sanitation Sectors 105 model has been rejected in some countries because, typically, limited incentive exists for a wage-earning employee to ensure cost recovery. In Zambia, the utility has tried to improve this model by introducing water commissions. A newer model, in which customers can pay for water at standpipes using electronic systems, is being introduced to reduce management costs and problems with nonpayment and to potentially provide more targeted subsidies as payment tokens can be distributed via existing sys- tems. South Africa currently uses this model, and electronic prepayment cards and vending machines are also being introduced in Lesotho and Namibia. In Zambia, customers use tokens or monthly cards instead of vending machines. These systems allow for tariffs to be set at a unit rate that is lower than the smallest coin (Brocklehurst and Janssens 2004; Kariuki and others 2003) and may allow for lower prices, because they eliminate the middleman. In Lesotho, the water utility and retail outlets sell prepaid cards. In some instances, however, independent "operators" sell tokens, at a higher price, at the standpipes. Although this is more convenient for customers, it is important to have formal outlets to main- tain set prices. Delegated Management Model. In the increasingly common delegation model for public standpipes, utilities either sign a contract directly with a standpipe operator, who pays the standpipe bill (and in some cases main- tains the standpipe), or sign a contract with a support institution. In the support institution model, local officials or members of a water commit- tee then supervise operators. Under this system, the institution pays the utility for each standpipe, based on a bulk water price. Community groups or local officials typically select the standpipe operators, and the process is generally far from transparent and often influenced by local politics (figure 4.13). Over the medium term, the delegation model has not always provided reliable service with timely bill payment to the utility and has been largely ineffective in providing a subsidized or "social" price to the end consumer. The most successful delegation models have been those that are heavily monitored by the utility, or another external body, which in turn has increased costs. Conversely, when utilities delegate most of their critical functions such as management, monitoring, maintenance, and oversight, this can result in higher consumer prices and more frequent breakdowns in service. There are exceptions, particularly in areas with high social capital. 106 Africa's Water and Sanitation Infrastructure Figure 4.13 Delegated Management Models direct contract contract with institution support water utility utility water payment supervision of the service payment standpipe operatora water committee support institutionb water payment stipend payment consumer standpipe operatora water payment consumer Private: Benin, Burkina Faso, Kenya, Malawi, Mozambique, Private: Ethiopia, Malawi, Niger, Rwanda, and Senegal Mozambique, and Tanzania CBO: Lesotho, Madagascar, CBO: Senegal and Zambia Malawi, and Tanzania Source: Keener, Luengo, and Banerjee 2009. Note: CBO = community-based organization. a. Standpipe operator can be a private individual or a CBO. b. Support institution: Local leaders, local authority administrators, or NGOs. Community-Based Management (Local Leaders, Local Authority Administrators, NGOs, and CBOs). Community management works only where there is a true sense of community, and where there is per- sonal security and accepted methods for dealing with those who do not follow regulations. Unlike rural areas, urban neighborhoods share a greater degree of heterogeneity. Local leaders and community organizations play various roles in standpipe management and oversight; in some cases (in parts of Addis Ababa, Blantyre, Dar es Salaam, and Maputo), utilities have put local leaders in charge of operations and maintenance, with the assumption that these leaders will act in their constituents' best interest. In these cases, the performance of the standpost, in terms of pricing, maintenance, timely bill payment, and so forth, is largely dependent on the manage- ment skills and legitimacy of the local leader and the degree of oversight by an external party. Improving the Organization of the Water and Sanitation Sectors 107 Several schemes have put community organizations in charge of man- agement or oversight in an effort to make standpipe/kiosk operators more accountable to their customers. These projects have been somewhat more effective than schemes that simply delegate management to a local leader. This practice is still very limited in the urban and periurban areas of Sub- Saharan countries, and in places where there is not enough social cohe- sion or strong local power structures and no oversight from a supporting institution, the model can also lead to corruption and mismanagement. In Blantyre and Lilongwe, for instance, community-managed kiosks that had been developed with extensive community involvement were taken over by local elites as soon as the mediating NGO left. The effectiveness of schemes involving community organizations varies and depends on the community's social cohesion and management capacity, as well as external monitoring. A Water and Sanitation Program report on the role of small- and medium-size organizations providing water in urban areas (Vézina 2002) stressed the limitations of community- based management models that lacked external monitoring and support: there is a tendency to minimize expenses by limiting the extension of the system, and although in principle these organizations are based on the voluntary participation of community members, to reduce operating and maintenance costs, actual management is often controlled by a small group that may monopolize control of the finances. With such arrange- ments, elite capture remains a problem that requires strong institutional controls and active monitoring. Some more recent models for community involvement use sophisti- cated incentives and monitoring to mitigate corruption. In Blantyre, Malawi, the water users association controls as many as 70 water points each. The utility provides technical assistance, legally registers the associ- ation, and monitors operation of the standpost. The association employs both the kiosk attendants and meter inspectors. The latter check the meter readings; if there is a difference between the inspector's meter reading and the amount of revenue collected, it is subtracted from the attendant's salary. Although the price of this water is 25 percent higher than at other kiosks, residents prefer to use these kiosks because the qual- ity of service is monitored and reliable. (This is not necessarily the case with other neighborhood kiosks.) In Dakar, Senegal, about 15 percent of the public standposts have been built via a partnership between the util- ity and an NGO, ENDA Tiers Monde. ENDA partners with communities and local neighborhood associations (for example, women's groups and 108 Africa's Water and Sanitation Infrastructure self-help groups); the community groups pay 25 percent of the capital costs of a standpipe, which is then built by the utility. The community also selects a standpost operator who collects revenue for the utility, and ENDA helps to create a local water council. Private Management. Although utilities contract out the operation of standpipes to private managers on the premise that this will promote effi- ciency and cost recovery, the results are not always positive. Many utilities in Sub-Saharan cities such as Blantyre, Cotonou, Dakar, Kigali, Nairobi, Niamey, Ouagadougou, and Quelimane have leased their installations and sold bulk water to private operators. The model has two particular weak- nesses: (1) The selection process of standpipe operators, particularly when the municipality is involved, is rarely transparent, and (2) because a pri- vate manager is running the standpipe, the water utility is less involved in collecting water revenue, ensuring good quality service, and maintaining adequate tariff levels. The price and hours of operation are also crucial to the success of this model: In the 1990s in Quelimane, Mozambique, pri- vate standpipe operators were billed according to fixed estimates of water consumption, but the water supply was extremely limited and intermit- tent. Certain standpipe operators found it difficult to generate enough water revenue to pay back the water bill and did not have funds to ade- quately maintain the standpipes (SAWA 1997). Household Resellers Reselling of water by households with private connections is commonly believed to be illegal in Sub-Saharan cities (Boyer 2006; Collignon and Vézina 2000; Kariuki and others 2003), but only 4 out of 15 cities in the study with prevalence of household water resellers explicitly pro- hibit the resale of water by households (table 4.4). Only three cities have legalized household resale and require a permit for this business. Box 4.2 presents a case study of regulated water reselling in Abidjan. In the majority of cases, a confusing legal limbo prevails; household water resellers are neither prohibited nor legalized. Even if regulations are in place prohibiting household water resellers, they are not enforced, as in Dakar or Dar es Salaam. Utilities and government simply do not con- trol and rarely contest this practice, and in the case of Kampala, the practice is encouraged in areas at the end of the network. Detailed case studies that highlight the importance of this source in allowing access where standposts or individual connections have not kept pace point to the serious impact that prohibition of this source would have on poor urban households. Improving the Organization of the Water and Sanitation Sectors 109 Table 4.4 Regulation of Household Water Resellers Country City Prohibited License Benin Cotonou No No Chad N'Djamena No -- Congo, Dem. Rep. Kinshasa No No Côte d'Ivoire Abidjan No Yes Ethiopia Addis Ababa No No Ghana Accra No No Lesotho Maseru -- -- Madagascar Antananarivo No No Malawi Blantyre -- -- Mozambique Maputo No No Nigeria Kaduna Yes n.a. Rwanda Kigali No Yes Senegal Dakar Yes Yes Sudan Greater Khartoum Yes No Tanzania Dar es Salaam Yes No Uganda Kampala No No Zambia Lusaka No No % yes 24 18 Source: Keener, Luengo, and Banerjee 2009. Note: n.a. = not applicable, -- = not available. Box 4.2 Regulation in Water Reseller Market in Abidjan Abidjan is one of the few cities with experience in attempting to regulate this sec- tor, though they also focus on removing illegal connections. Although the results have been disappointing because of a lack of incentives, there is still potential to explore better mechanisms for using this source. In the early 1980s, the utility SODECI and the national government decided to address the increasing growth of household water resellers that tapped into illegal connections to the network. The authorities would provide permits to the household water resellers as long as they converted their connections into formal ones. The expected outcomes were an increase in sales among the poor, a reduction in illegal activity, and an im- provement in revenue collection. The campaign did not provide any incentive to the resellers; they were billed as domestic customers and faced an increasing block tariff. Moreover, the water vendor was required to provide a title deed for the permit and to invest in an extension from the meter to the water point. As a result, only 1 percent of the total resale at the household level is currently con- ducted through legalized resellers. Source: Kariuki and others 2003. 110 Africa's Water and Sanitation Infrastructure Table 4.5 Regulation of Water Tankers Country City Regulated Cape Verde Praia Yes Chad N'Djamena Yes Ethiopia Addis Ababa Yes Ghana Accra No Kenya Nairobi Yes Nigeria Kaduna No Rwanda Kigali No South Africa Johannesburg No Sudan Greater Khartoum Yes Tanzania Dar es Salaam No Uganda Kampala No Source: Keener, Luengo, and Banerjee 2009. Given the coverage gap and the ready distribution system that house- hold resellers provide, a valid question is whether to explore methods to partner with private households to increase coverage. Water Tankers The utility emerges as a minor player in the operation of water tankers. The formal and informal private sectors are the main operators in four out of nine cities with water tanker supply (table 4.5). Many Levels of Government Players in the Rural Water Market Different levels of government play various roles in rural water provision. Box 4.3, for instance, presents the typical issues faced by Cross River State in Nigeria. In about one-third of the countries, the central govern- ment is responsible for rural water supply, and it shares this task with regional/state or local governments in another 27. In Cape Verde, Chad, Madagascar, South Africa, and Uganda, local governments are responsible for water supply. The central government, local government, and NGOs play the great- est roles in most aspects of rural water service provision. Urban utilities and community service providers play the smallest roles, though commu- nity service providers are most involved in the direct provision of service. Although regional governments, rural agencies, and the private sector also contribute to water provision in certain countries, they are generally less involved across the range of countries and tasks (table 4.6). Improving the Organization of the Water and Sanitation Sectors 111 Box 4.3 Issues Constraining Rural Water Supply in Cross River State, Nigeria Cross River State, one of the 36 states in Nigeria, is located in the tropical rain for- est belt of Nigeria. About 75 percent of its population, 3.25 million people, lives in rural areas and is engaged in subsistent farming, and more than 70 percent lives with less than $1 a day. Cross River State is one of the states selected by the World Bank to carry out an assessment of the rural water supply based on public expenditure reviews. This is part of a substantial effort implemented by the World Bank to assess rural water sector performance in West Africa. The review, whose findings are reported here, covers the period from 2002 to 2007. Water supply in the Cross River State is in crisis. Coverage stands at only 25 per- cent in urban areas and 31 percent in semiurban and rural areas. Rural water is mainly supplied through boreholes with hand pumps and wells, 65 percent of which are not functioning. Moreover, no water treatment is provided. Meeting the MDG for water is estimated to require an additional 10,098 bore- holes with hand pumps and 2,525 motorized boreholes to be built across the state by 2015, a daunting task given the current financial, institutional, and tech- nical capacity. Lack of adequate budgetary funding and low disbursement efficiency are major constraints. Rural water captures only 0.5 percent of the state capital budget, and execution ratios average less than 20 percent. Weak institutions and frag- mented responsibilities translate to feeble leadership and rural water falling behind in the political agenda. The sector is under the responsibility of the State Rural Water Supply and Sanitation Agency (RUWATSSA), which remains a section of the Rural Development Agency. Differently from in other states, no dedicated ministry champions for reforms and allocations. More importantly, although a rural water policy does exist nationally, this is not necessarily reflected in state policies, and effective cooperation is not pursued between the national and state govern- ments. Responsibilities are decentralized locally, but RUWATSSA continues to be characterized by a weak and poorly funded mandate and loose connections to the national water sector. Maintenance and rehabilitation of rural water schemes are jeopardized by the lack of skilled staff and the substantial underdevelopment of a local private sec- tor. Technical capacity for routine maintenance remains low; spare parts for bore- holes are difficult to find and very expensive where available. (continued next page) 112 Africa's Water and Sanitation Infrastructure Box 4.3 (continued) Finally, no effective strategy to promote community participation has been put in place, with the result that involvement by local communities in rural water provision remains shallow at best. Absent any sense of ownership, rural commu- nities do not take responsibility in preserving and repairing facilities, and they would not have the capacity to do so without adequate training. Source: Iliyas, Eneh, and Oside 2009. Table 4.6 Stakeholder Involvement in Rural Water Activities (percent) Financing Providing Ensuring Planning Preparing rural Providing technical water projects projects water services assistance quality Central government 41 31 50 8 31 48 Regional government 14 17 13 8 11 16 Local government 24 24 18 24 14 10 Rural water agency 16 14 5 14 11 6 Utility 0 0 0 5 11 16 Community 3 7 8 24 3 3 Private sector 3 7 8 16 19 0 Source: Banerjee, Skilling, and others 2008b. Note: Columns do not add to 100 percent because more than one agency can be responsible for performing the activities. Rural water points are typically managed by the level of government closest to the communities themselves, and, in some cases, the govern- ment and community share the responsibility. In half of the countries, the community is primarily responsible for maintaining the rural water points (figure 4.14). The central government does not play a major role, except in Malawi, where it shares this responsibility with the local government, community, and private groups. The ministry in charge of water supply is by far the most important institution when it comes to collecting data and monitoring the rural water points, as in Cape Verde, where the Improving the Organization of the Water and Sanitation Sectors 113 Figure 4.14 Responsibility for Maintenance and Monitoring of Rural Water Points a. Maintenance b. Performance monitoring and data collection 18% 4% 4% 17% 32% 4% 4% 65% 50% state/local ministry of rural affairs community ministry in charge of water supply mixed utility regulator other none Source: Banerjee, Skilling, and others 2008b. Note: Panel b total is 98 percent because of data unavailability for some countries. Ministry of Rural Affairs performs this role. The regulator (Direction de l'Hydrolique) tracks data in Côte d'Ivoire, and the utilities are responsi- ble for the same in Sudan. About half of the countries reportedly have a rural water agency, but most countries at least have an established policy in place specifically for the rural water sector. A few countries, such as Benin, Burkina Faso, Côte d'Ivoire, Ghana, Lesotho, Mozambique, Namibia, Nigeria, Senegal, and Uganda, have both a rural water policy and a rural water agency to ensure service delivery to rural dwellers. The water points are dispersed across the rural space and are often mapped to monitor their functioning. Of these countries, only Benin, Burkina Faso, and Uganda have a rural water map as well. The rural water index is used to measure each country's progress. This is done using five indicators: existence of a rural water agency, existence of a rural water policy, existence of a map of rural water points, existence of a dedicated budget or rural water fund, and existence of a cost-recovery policy (figure 4.15). Burkina Faso, Côte d'Ivoire, and Uganda score the highest and are the best-performing countries in creating wide-ranging reforms for the rural sector. Though we cannot evaluate the performance 114 Africa's Water and Sanitation Infrastructure Figure 4.15 Country Ranking and Prevalence of Key Attributes for the Rural Reform Index a. By country Uganda Côte d'Ivoire Burkina Faso Tanzania Senegal Nigeria Namibia Mozambique Madagascar Ghana Benin Sudan Rwanda Lesotho Ethiopia Chad Zambia South Africa Malawi Kenya Congo, Dem. Rep. Cape Verde Niger 0 20 40 60 80 100 rural reform index (%) b. By attribute rural water policy dedicated budget/fund cost-recovery policy rural water agency map of rural water points 0 20 40 60 80 100 countries (%) Source: Banerjee, Skilling, and others 2008b. Improving the Organization of the Water and Sanitation Sectors 115 of rural facilities using the existing data, we find that the percentage of rural water points in need of rehabilitation and the rural water index are negatively correlated,5 suggesting a positive association between rural reforms and functioning rural water facilities. WSS services in small towns are often neglected and are often not taken into account in urban and rural water strategies. In many countries, particularly more populous ones, people are concentrated in small towns, and there is a need for an explicit strategy to provide infrastructure ser- vices. For instance, in Côte d'Ivoire, Benin, Madagascar, Nigeria, and Senegal, more than 30 percent of the population lives in small towns. Small towns range in size from 2,000 inhabitants in Benin, Ghana, Madagascar, and Niger to 160,000 in Nigeria. Small towns are covered under a rural or urban strategy in only five countries. Half of the coun- tries surveyed have a specific policy or strategy for provision of small town water services. Even among these countries, only a few, such as Côte d'Ivoire, Ghana, Lesotho, and Uganda, have a specialized agency for small town water services. Many Players with No Clear Accountability in the Sanitation Market The sanitation sector is governed by a complicated institutional frame- work characterized by complexity, a multiplicity of actors, and lack of clear accountability for sector leadership. On-site sanitation operation requires financing, technical assistance, maintenance, emptying (or desludging) of facilities, and regulation. In most countries, central min- istries, national and city-level utilities, local government agencies, house- holds, NGOs, and other institutions share these responsibilities. In most countries, regulation is the only area where there is a clear delegation of responsibility to a single entity (figure 4.16). Institutional arrangements tend to differ sharply across urban and rural environments. In rural areas, communities and households typically man- age sanitation, with oversight from ministries of health. The central gov- ernment is generally responsible for urban sanitation under the oversight of ministries of water, environment, housing, or public health. Municipal agencies or utilities are typically responsible for running and maintaining sanitation operations. About 60 percent of water utilities operated sewerage networks, and a similar proportion had some responsibility for on-site sanitation as well (figure 4.17). Sanitation either can be treated as a separate business, with 116 Africa's Water and Sanitation Infrastructure Figure 4.16 Responsibilities for On-Site Sanitation Functions 90 80 70 60 % countries 50 40 30 20 10 0 financing technical maintenance desludging regulation assistance single multiple Source: Morella, Foster, and Banerjee 2008. Figure 4.17 Urban Utilities' Responsibility over On-Site Sanitation and Wastewater Management a. On-site sanitation b. Wastewater management 41% 44% 59% 58% share of countries in which urban share of water utilities with utilities play a role in on-site sanitation sewerage network share of countries in which urban share of water utilities without utilities do not play a role in on-site sewerage network sanitation Source: Morella, Foster, and Banerjee 2008. dedicated staff, organization, and management, or it can be operated jointly with water; both approaches are equally prevalent. Senegal is the only country with a specialized sanitation utility, Office National de l'Assainissement du Sénégal (ONAS), which reports to the Ministry of Sanitation, which was recently restructured as the Ministry of Urban Improving the Organization of the Water and Sanitation Sectors 117 Affairs, Housing, Urban Water, Public Hygiene and Sanitation. In Burkina Faso, the water utility Office Nationale des Eaux et l'Assainissement (ONEA) has a separate department that is responsible for sanitation. The devolution of provision of sanitation services to subnational gov- ernments has been the most significant reform of the past decade, affect- ing 80 percent of countries surveyed. Decentralization first began in large cities, where it had effectively been the practice because of the compre- hensive role played by many water utilities. Utilities in rural areas are also being decentralized, with responsibilities transferred to small local authorities, many of which have been recently established, as in Benin, Burkina Faso, and Mali. Reform progress can be evaluated using a simple scoring system, called the sanitation index. Countries have worked to establish a comprehensive sanitation framework to move more people away from open defecation. This reform index focuses on on-site sanitation, because a vast majority of Africans depend on this source, and the prevalence of piped sewerage is miniscule in comparison. The sanitation index includes six indicators: existence of a national sanitation policy, existence of a hygiene-promotion program, existence of an accepted definition of sanitation, existence of a specific fund for sanitation, involvement of utilities in on-site sanitation, and clear cost-recovery policies for on-site sanitation. The index is calcu- lated by adding the values of the six indicators, and the countries with any missing data points are dropped to ensure consistency. Together, these six indicators provide a holistic measure of a country's sanitation agenda. Most countries have worked to create an accepted definition of sanita- tion and a hygiene-promotion program to establish a strong sanitation framework. Fifteen countries have also established a national sanitation policy, and seven countries have developed operating cost-recovery poli- cies, known to pay significant dividends. Only eight countries have set up a sanitation fund or a dedicated budget line; in Chad and Ethiopia, these are funded exclusively by donors or a combination of government, sector levies, and donors. Côte d'Ivoire is the only country with a fund financed entirely by sector levies. Burkina Faso, Chad, Kenya, Madagascar, and South Africa stand out for having scored 100 percent on the sanitation index. At the other extreme are countries such as the Democratic Republic of Congo, Lesotho, Nigeria, and Zambia, which are struggling to establish appropriate sanitation systems (figure 4.18). The widespread use of on-site sanitation facilities brings up issues of construction, management, and maintenance of latrines. The AICD WSS survey provides an overview of the practice with respect to latrine 118 Africa's Water and Sanitation Infrastructure Figure 4.18 On-Site Sanitation Index a. Sanitation index b. Existence of attributes South Africa Madagascar existence of an accepted Kenya definition of sanitation Chad Burkina Faso Tanzania existence of hygiene promotion program Côte d'Ivoire Cape Verde Namibia Uganda existence of sanitation policy Senegal Ethiopia Sudan involvement of utilities in on- Rwanda site sanitation Mozambique Malawi Ghana existence of cost-recovery Benin requirement for on-site sanitation Niger Lesotho Congo, Dem. Rep. existence of a specific fund for Nigeria sanitation Zambia 0 20 40 60 80 100 0 50 100 sanitation index (%) countries (%) Source: Morella, Foster, and Banerjee 2008. construction and operation. In most cases, the private sector, households, and/or NGO/CBOs are responsible for the construction of on-site sani- tation. The government rarely finances the construction of sanitation facilities. Latrine emptying is predominantly a private sector function, although in a substantial number of cases the municipality and/or local utility takes primary responsibility. Only nine countries reported having formal regulatory oversight of latrines, and the majority of countries report concerns about proximity of unhygienic latrines to drilled holes, with the potential for cross-contamination (table 4.7). Table 4.7 Management of Latrines Problem with Regulation Level of latrine groundwater Latrine construction Emptying of latrines of latrines regulation contamination Benin Households Local private No n.a. Yes Burkina Faso Government Combination No n.a. No Cape Verde NGO/CBO Municipality No n.a. Chad NGO/CBO No Congo, Dem. Rep. Private sector Local private No n.a. No Côte d'Ivoire Government, households Utility, combination Yes Utility Ethiopia Private sector Municipality No n.a. Yes Ghana Yes Kenya NGO/CBO Combination Yes Central government, utility Lesotho Households Utility Yes Central government No Madagascar Households Local private, combination Yes Municipality No Malawi Government/NGO/ Municipality, local private, Yes Central government , Yes households utility municipality, community Mozambique Households/NGO Other No n.a. Yes Namibia Government/households Municipality Yes Niger Households Local private No n.a. Yes Nigeria Local private No n.a. Yes Rwanda Households Combination No n.a. No Senegal Government/NGO/ Local private Yes Central government Yes households South Africa Government Municipality Yes Municipality Yes Sudan Households Local private Yes Municipality No Tanzania Households Local private No n.a. Yes Uganda Households Combination Yes Municipality Yes Zambia Households Local private No n.a. Yes 119 Source: Morella, Foster, and Banerjee 2008. Note: CBO = community-based organization, n.a. = not applicable, NGO = nongovernmental organization. 120 Africa's Water and Sanitation Infrastructure Notes 1. Indicators include "consumer associations have membership," "consumer asso- ciations have a right to appeal regulatory decision," "consumers have a right to comment on draft regulations," and "consumers have a right to review tariff proposals." 2. There is centralized mode of service delivery in urban centers, but they might be decentralized for small towns and rural areas. 3. In the majority of Sub-Saharan cities, the utility follows one of these two models. Examples exist of kiosks that are both owned and operated by private individuals that use utility water, as in Nairobi and Blantyre (Chirwa and Junge 2007; Oenga and Kuria 2006), or that are owned and operated by com- munity groups, as in Dakar (Brocklehurst and Janssens 2004). These are largely the exceptions, however. 4. In about half of the AICD sample cities, more than one management model was being used, either because one model is in the process of being replaced by another (Lesotho, for example) or because of heterogeneous areas demand- ing different management approaches. 5. The correlation coefficient between the percentage of rural water points in need of rehabilitation and the rural water index is ­0.46. References Baietti, A., W. Kingdom, and M. van Ginneken. 2006. "Characteristics of Well- Performing Public Water Utilities." Water Supply and Sanitation Working Note 9, World Bank, Washington, DC. Banerjee, S., H. Skilling, V. Foster, C. Briceño-Garmendia, E. Morella, and T. Chfadi. 2008a. "Ebbing Water, Surging Deficits: Urban Water Supply in Sub-Saharan Africa." AICD Background Paper 12, World Bank, Washington, DC. ------. 2008b. "State of the Sector Review." AICD Working Paper, World Bank, Washington, DC. Boyer, A. 2006. "Survey of Household Water Resale Activity in Peri-Urban Maputo: Preliminary Discussion of Findings." Water and Sanitation Program, Maputo, Mozambique. Brocklehurst, C., and J. G. Janssens. 2004. "Innovative Contracts, Sound Relationships: Urban Water Sector Reform in Senegal." Water Supply and Sanitation Sector Board Discussion Paper, World Bank, Washington, DC. Chirwa, E., and N. Junge. 2007. "Poverty and Social Impact Analysis--Private Sector Participation in the Distribution and Management of Water Services in Improving the Organization of the Water and Sanitation Sectors 121 Low-Income Areas in the Cities of Blantyre and Lilongwe: Republic of Malawi." Ministry of Economic Planning and Development, Ministry of Irrigation and Water Development, Malawi. Collignon, B., and M. Vézina. 2000. "Independent Water and Sanitation Providers in African Cities: Full Report of a Ten-Country Study." Water and Sanitation Program, Washington, DC. Development Workshop. 1995. "Water Supply and Sanitation in Luanda: Informal Sector Study and Beneficiary Assessment." Africa Urban and Water Department, World Bank, Luanda. Elamin, M., and A. Gadir. 2006. "A Study of Small Water Enterprises in Khartoum." Water, Engineering and Development Centre, Loughborough University, Leicestershire, England. Iliyas, M., D. Eneh, and I. Oside. 2009. "Public Expenditure Review in the Rural Water and Sanitation Sector for Cross River State, Nigeria." Africa Urban and Water Department, World Bank, Washington, DC. Kariuki, M., B. Collignon, B. Taisne, and B. Valfrey. 2003. "Better Water and Sanitation for the Urban Poor: Good Practice from Sub-Saharan Africa." Water Utility Partnership for Capacity Building, Abidjan. Keener, S., M. Luengo, and S. G. Banerjee. 2009. "Provision of Water to the Poor in Africa: Experience with Water Standposts and the Informal Water Sector." AICD Working Paper 13, World Bank, Washington, DC. Marin, P. 2009. "Public Private Partnerships for Urban Water Utilities: A Review of Experiences in Developing Countries." Trends and Policy Options 8, Public-Private Infrastructure Advisory Facility and World Bank, Washington, DC. Morella, E., V. Foster, and S. Banerjee. 2008. "Climbing the Ladder: The State of Sanitation in Sub-Saharan Africa." AICD Background Paper 13, World Bank, Washington, DC. Oenga, I., and D. Kuria. 2006. "A Study of Small Water Enterprises in Nairobi." Water, Engineering, and Development Centre, Loughborough University, Leicestershire, England. PA Consulting Group. 2007. "Corporatization of Bangladesh Power Development Board." Auckland. Sarpong, K., and K. M. Abrampah. 2006. "A Study of Small Water Enterprises in Accra." Water, Engineering, and Development Centre, Loughborough University, Leicestershire, England. SAWA (Strategic Alliance for Integrated Water Management Actions). 1997. "Beneficiary Assessment on Urban Water in Mozambique: Maputo and Quelimane." República de Moçambique, Ministério das Obras Públicas e Habitaçao. Direcçao Nacional de Águas, Maputo. 122 Africa's Water and Sanitation Infrastructure Vagliasindi, M., and J. Nellis 2009. "Evaluating Africa's Experience with Institutional Reforms for the Infrastructure Sectors." AICD Working Paper 23, World Bank, Washington, DC. Vézina, M. 2002. "Water Services in Small Towns in Africa: The Role of Small and Medium-Sized Organizations." Water and Sanitation Program, World Bank, Washington, DC. World Bank. 2007. Private Participation in Infrastructure Database. Washington, DC. http://ppi.worldbank.org. CHAPTER 5 Urban Water Provision: The Story of African Utilities Most countries in Sub-Saharan Africa cannot provide adequate water and other services for their citizens because of low coverage rates. People lack proper services because systems fail, often because not enough was invested to appropriately build and maintain them, and also because of the stress that urbanization places on this existing infrastructure. In the past decade, Africa's population grew at an annual average of 2.5 percent, and the urban and slum population grew at almost double that rate. A well-performing utility provides service to customers who demand it, at a level that meets their needs and at a price that they are able and willing to pay (Tynan and Kingdom 2002). This chapter closely examines the performance of the individual utilities that form the core of service provision in African countries, drawing on the Africa Infrastructure Country Diagnostic (AICD) Water Supply and Sanitation (WSS) Survey. This chapter introduces a measure called "hidden cost," which comprehensively quantifies underperformance or inefficiencies and defines the economic burden. The relationship between hidden cost and institutional indicators demonstrates the contribution of institutional reforms to utility performance and ser- vice delivery. 123 124 Africa's Water and Sanitation Infrastructure Access to Utility Water Utilities in Africa operate in service areas of varying sizes (annex 5.1). They can serve as few as 30,000 people, as in Oshakati, Namibia, or more than 15 million residents, as in the Democratic Republic of Congo, Ghana, and Lagos, Nigeria. The utility in Johannesburg has the highest number of res- idential water connections--more than 1 million. About 40 percent of the utilities in Africa have fewer than 20,000 residential water connections. Although water access trends are typically analyzed based on national coverage statistics from household surveys (see chapter 2), it is also inter- esting to look at the trends that emerge directly from the utility data. These statistics focus solely on access within the utility service area and show how utility water is distributed to different segments of the popu- lation. Utility-based coverage statistics tend to differ from the figures found through household surveys. In general, coverage statistics based on household surveys tend to reveal higher access rates because they include informal and illegal connections. With regard to piped-water service, comparing household survey access rates with utility data is complicated by the fact that some service areas fall outside the national or urban geographic spheres covered by household surveys. For the handful of countries where a reasonable match can be made between geographic areas, the population coverage rates reported by the household surveys are 4 to 16 percent higher than those in the utility coverage data. Moreover, the household surveys show an additional served population that represents 14 to 33 percent of the total population with access (table 5.1). Utilities in some countries also provide service for "off-grid" consumers in addition to servicing formal clients when their service area is bigger than the network area. These off-grid provisions include off-grid bore- holes with networks or water quality checks. In Lusaka and Dar es Salaam, community partnerships manage large off-grid systems. About 98 percent of the population in the utility service areas in the middle-income countries receives utility water, whether through pri- vate piped connections, shared connections with neighbors, or stand- post services. In the low-income countries, however, only 69 percent of residents in the service area are accessing utility water, leaving a sizeable minority that must rely on other sources, such as ground or surface water. The low-income, fragile countries have the maximum connection deficit--only 26 percent are covered by piped-water supply and 56 per- cent by some sort of utility water. These countries also have the highest Urban Water Provision: The Story of African Utilities 125 Table 5.1 Comparison of Coverage Statistics for Water, Based on Utility Data versus Household Surveys (percent) Coverage rate derived from Coverage rate Difference Potential household derived from in coverage rate of surveys utility data rates informality (A) (B) (A­B) (A­B)/(A) SONEB (Benin) 29 25 4 14 SDE (Senegal) 77 66 11 15 ONEA (Burkina Faso) 33 25 8 25 JIRAMA (Madagascar) 17 13 4 25 ELECTROGAZ (Rwanda) 16 11 5 30 WASA (Lesotho) 50 34 16 33 Source: Banerjee, Skilling, and others 2008. Note: JIRAMA = Jiro sy Rano Malagasy, ONEA = Office Nationale des Eaux et d'Assainissement, SDE = Sénégalaise des Eaux, SONEB = Société Nationale des Eaux du Benin, WASA = Water and Sanitation Authority. proportion of people sharing taps with neighbors, confirming a degree of informality not witnessed in other countries. The connection deficit varies drastically among income groups. The middle-income countries have piped-water coverage that is multiple times higher than that of other income groups--twice the low- income, three times the resource-rich, and more than three times the low-income, fragile countries. In middle-income countries, the vast majority of people who access utility water do so through private resi- dential connections. In low-income, fragile countries, however, less than half of those who receive utility water do so via private piped connec- tions; the rest share connections with neighbors or rely on communal modalities such as utility standposts. Few people in the middle-income countries informally share connections, but in the low-income coun- tries, this practice is almost as common as the use of formal utility standposts, albeit with substantial regional variations. The East African Community (EAC) and Economic Community of West African States (ECOWAS) regional groups have the highest number of households dependent on a neighbor's connection (table 5.2). The water-abundant countries have fewer utility-provided connec- tions, and the water-scarce countries not only have more private water connections, but also have better coverage through standposts and from neighbors. Overall, the large utilities are better at providing some sort of 126 Africa's Water and Sanitation Infrastructure Table 5.2 Overview of Access Patterns in the Utility Service Area (percent) Access by Access by Access to private sharing of utility residential neighbors' water by piped-water Access by private some connection standpost connection modality Sub-Saharan Africa 44.3 13.0 21.7 64.0 By income Low-income 42.2 23.2 22.5 68.6 Low-income, fragile 25.6 2.2 41.0 56.0 Resource-rich 30.3 15.8 7.4 48.8 Middle-income 88.0 9.7 0.3 97.8 By regional economic community ECOWAS 38.1 8.8 34.3 68.6 SADC 53.2 11.1 8.1 62.2 CEMAC 24.2 -- -- 65.0 EAC 44.7 26.5 40.4 91.6 COMESA 26.0 18.2 23.7 54.7 By water availabilitya High water scarcity 56.4 16.3 15.2 68.8 Low water scarcity 32.5 8.9 19.6 57.1 By utility sizeb Small 47.0 15.5 20.7 68.6 Large 39.5 13.8 25.6 80.9 Sources: AICD WSS Database; Banerjee, Skilling, and others 2008. Note: CEMAC = Central African Economic and Monetary Community, COMESA = Common Market for Eastern and Southern Africa, EAC = East African Community, ECOWAS = Economic Community of West African States, SADC = Southern African Development Community. -- = not available. a. Water abundance is defined as renewable internal freshwater resources per capita in excess of 3,000 cubic meters. b. Large utilities are defined as those serving more than 100,000 connections. utility water to consumers and manage to serve four out of five residents in their service area. The Pace of Expansion of Utility Water Coverage Although utilities might have substantially different access rates for pri- vate piped-water connections, a key issue is how quickly the coverage gap is being closed. This can be gauged by looking at the average annual growth rate of connections in recent years. It is currently 5 percent; however, that value differs from country to country (figure 5.1, panel a). Urban Water Provision: The Story of African Utilities 127 Figure 5.1 Expansion of Utility Water Coverage a. Annual average rates of expansion for private piped-water connections 20 18 16 number of utilities 14 12 10 8 6 4 2 0 < 0% pa 0­5% pa 5­10% pa 10­15% pa >15% pa b. Coverage against level of connection charge 70 coverage by private tap (%) 60 50 40 30 20 R2 = 0.37 10 0 0 50 100 150 200 250 300 connection fee (US$) Source: Banerjee, Skilling, and others 2008. Five utilities (in the Democratic Republic of Congo, Kenya, and Nigeria) actually report an absolute decline in the number of customers con- nected. In contrast, the 10 fastest-expanding utilities (in Benin, Cape Verde, Ethiopia, Malawi, Uganda, and Zambia) are growing at an aver- age annual rate in excess of 7 percent, a pace that would allow the util- ities to double the number of connections if it were sustained over a decade. In absolute terms, utilities are growing fastest in the largest cities; Cape Town, Johannesburg, and Lagos each add between 30,000 and 50,000 new connections each year. Given Sub-Saharan Africa's 3.5 percent urban demographic growth rate, however, more than one-third 128 Africa's Water and Sanitation Infrastructure of the utilities in the region are simply not expanding rapidly enough to achieve proper coverage. One factor that sometimes hampers growth of connections is cost. The average connection fee for piped-water service, among the 26 utilities able to supply this data point, is $265. In the low-income countries, sig- nificant negative correlation is seen between the connection charge and the coverage of private taps in the utility area (figure 5.1, panel b). Water Production Capacity Varies from Country to Country Utilities can expand coverage only if there is sufficient water production in the service area relative to the resident population. Water production varies widely across the country income groups. Middle-income countries produce around 209 liters per day for each resident in the service area, indicating that enough water would be available to adequately serve the entire population if the distribution networks were expanded. By contrast, utilities in the low-income countries produce only 130 liters per capita per day, just enough for those customers who are already connected to the system. If these utilities were to connect their entire unserved populations to the network, the availability of water would drop to only 66 liters per capita per day, suggesting that these utilities need to invest in both water production capacity and water distribution networks to reach universal coverage. The low-income, fragile countries experience the lowest production, at only 77 liters per capita per day for their con- sumers, which falls to only 36 liters per capita per day if the water is spread to all the residents in the service area. Once again, there is a dif- ference in water production between water-scarce and water-abundant countries, with the latter group serving 176 liters per capita per day com- pared with 125 liters per capita per day for the former (table 5.3). This reflects the higher ability of utilities to produce and serve more water in water-rich countries compared with utilities facing arid environments. Two-Part Tariff Structures for Piped Water Many countries in Africa have adopted a two-part tariff structure that incorporates both fixed and water-use charges. Two-part tariffs are designed so that the fixed part helps to cover production and administra- tive costs (such as billing and meter reading) and the water-use portion covers partial operations and maintenance (O&M) costs. Fixed charges can take two forms--a minimum consumption charge and a monthly Urban Water Provision: The Story of African Utilities 129 Table 5.3 Water Production per Capita in the Utility Service Area Water production per capita Water production per capita in the utility service served by utility in service area area (liters per capita per day) (liters per capita per day) Sub-Saharan Africa 116.4 162.9 By income Low-income 66.0 130.2 Low-income, fragile 35.7 76.5 Resource-rich 140.5 208.8 Middle-income 208.8 233.6 By regional economic community ECOWAS 42.3 96.8 SADC 132.5 184.4 CEMAC 107.4 229.5 EAC 71.3 118.9 COMESA 142.4 183.3 By water availabilitya High water scarcity 81.5 125.4 Low water scarcity 115.3 175.9 By utility sizeb Small 102.5 160.1 Large 106.2 189.4 Sources: AICD WSS Database; Banerjee, Skilling, and others 2008. Note: CEMAC = Central African Economic and Monetary Community, COMESA = Common Market for Eastern and Southern Africa, EAC = East African Community, ECOWAS = Economic Community of West African States, SADC = Southern African Development Community. a. Water abundance is defined as renewable internal freshwater resources per capita in excess of 3,000 cubic meters. b. Large utilities are defined as those serving more than 100,000 connections. fixed fee--and they also allow for the recovery of investment costs with- out distorting price signals. The volumetric tariff, which is based on water use, usually takes the form of increasing block tariffs (IBTs). The IBT has long been a common structure in developing countries, where unit prices in the lower brackets of consumption (in cubic meters per month) tend to be smaller than the prices in higher brackets. Fourteen utilities have designed a two-part tariff, including 13 that enforce a "fixed charge plus IBT." Only the National Water and Sewerage Company (NWSC) in Uganda uses a "fixed charge plus linear tariff" structure. In addition to these utilities, seven have a "minimum consump- tion plus IBT" structure. The remaining 24 utilities use an interesting range of structures: 19 impose an IBT structure and three enforce a lin- ear structure, which means that households pay the same price per unit 130 Africa's Water and Sanitation Infrastructure of consumption. The remaining two utilities have different tariff struc- tures: the Central Region Water Board (CRWB) in Malawi charges a flat fee or fixed charge for the first 32 units of consumption, and the Kisumu Water and Sewerage Company (KIWASCO) in Kenya has a U-shaped structure, in which tariffs decline after the first block and rise again after the third. The block structure can add to the complexity of tariffs. It can range from one (linear) to seven, with the average being just above three blocks. The most common is a three-tiered block structure, used by 16 utilities. Ten water utilities in Africa use a four-block structure. At the high end are utilities such as Drakenstein, in South Africa, which has seven blocks. ELECTROGAZ in Rwanda, as well as in Johannesburg and Tygerberg in South Africa, has a six-block water tariff structure (figure 5.2). Twenty-nine percent of the water utilities in Africa use a monthly fixed fee, which is usually based on pipe size. The lowest fees are for the typical residential pipe size of 15 to 20 millimeters. Fixed fees can also be determined based on consumption. This charge is meant to cover the fixed part of the O&M cost. Fifteen percent of the utilities levy this charge, and, in all cases, the fee includes consumption of, at most, 10 cubic meters (m3). The size of the first block varies. In most countries, the first block is usu- ally below 10 m3; only 20 percent have a first block higher than 10 m3. Only 36 percent have a first block of less than 6 m3 per month, which is consid- ered almost subsistence consumption. At the other extreme are utilities with a large consumption spread in the first block. The size of the last block also reveals interesting patterns. The last block can start from 5 m3, as it does in the Société Nationale des Eaux du Benin (SONEB) in Benin or the Dar es Salaam Water and Sewerage Company (DAWASCO) in Tanzania. It can also start at 1,000 m3, as it does in Drakenstein, South Africa, or Katsina, Nigeria. In 64 percent of the utilities, the starting point of the last block is less than or equal to 50 m3. Developing countries have often used the price of a first block as a social tariff, or lifeline, so that the poor can get at least a minimum quan- tity of safe water at a subsidized price. In numerous countries with a min- imum consumption charge, such as Côte d'Ivoire, Malawi (Blantyre Water Board), and Mozambique, the block structure begins from block two, and the price of block one is therefore zero. The price of the last block is often set with cost recovery and water conservation in mind. In about one-third of utilities, the tariffs are set higher than $0.8/m3. The fixed charges, which are expected to be paid every month, irrespective of consumption, are usually less than $4. Of the 44 percent of the utilities Urban Water Provision: The Story of African Utilities 131 Figure 5.2 Variations in Tariff Structures a. by number of blocks 18 16 16 14 number of utilities 12 10 10 8 6 6 5 4 3 3 2 1 0 number of blocks b. by type 20 19 18 16 number of utilities 14 13 12 10 8 7 6 4 3 2 1 1 1 0 number of blocks minimum consumption + IBT fixed charge + IBT IBT fixed charge + linear fixed charge linear U-shaped Source: Banerjee, Foster, and others 2008. Note: IBT = increasing block tariff. that enforce a fixed-fee or minimum-consumption charge, about half are set between $2 and $4 (figure 5.3). Sewerage Charges Linked to Water Bills Sewerage payment structures vary and can be calculated either as a sur- charge percentage on the water bill or by using an independent block or fixed tariff structure. In more than half of utilities, the sanitation charge 132 Africa's Water and Sanitation Infrastructure Figure 5.3 Utility Prices and Charges a. Price of first block b. Price of nth block c. Fixed charges (US$/m3) (US$/m3) (US$) 20 12 15 number of utilities number of utilities 10 number of utilities 15 8 10 10 6 5 4 5 2 0 0 0 price price charge < 0.2 < 0.4 <1 > 0.2 and < 0.3 > 0.4 and < 0.6 > 1 and < 2 > 0.3 and < 0.4 > 0.6 and < 0.8 > 2 and < 4 > 0.4 > 0.8 >4 Source: Banerjee, Foster, and others 2008. is levied as part of the water bill. That charge ranges from 30 percent in Zambia to 85 percent in Lesotho, with an average of 53 percent. Six African utilities use the block-tariff structure for sewerage, with the blocks varying between one and five. Walvis Bay in Namibia stands out because of its use of a decreasing block tariff, in which prices decline with rising consumption. KIWASCO in Kenya is the only utility that reports levying a separate connection fee of $90 specifically for sewer service (table 5.4). Burkina Faso has taken an innovative approach by levying a sanitation tax as a surcharge on the water bill, which is then used to subsidize access to on-site sanitation facilities in Ouagadougou (box 5.1). Modest Water Consumption by End Users Demand management can be reliably assessed only for those water utili- ties with good metering coverage, as they would therefore be expected to have relatively meaningful estimates of water consumption and nonrev- enue water (NRW). There are four categories of sample utilities. The first category comprises 15 utilities that do not report meter coverage. The second category comprises 26 utilities (mainly in Ethiopia, Nigeria, and Zambia) with low meter coverage (less than 50 percent of residential connections), averaging 19 percent for the group. The third category com- prises 11 utilities (mainly in South Africa and Tanzania) with moderate Table 5.4 Structure and Level of Wastewater Tariffs Connection Fixed Price of Type of fee charge Number Size of Size of first block Price of Utility Country tariff (US$) (US$) of blocks first block nth block (US$) nth block ONEA Burkina Faso Flat 0 0 1 0+ 0.04 0.04 AWSA Ethiopia Flat 0 0 1 7.1+ 0.07 0.07 NWASCO Kenya IBT 0 0 4 0­10 60+ 0.13 0.21 KIWASCO Kenya IBT 90 0 5 0­10 60+ 0.21 0.42 Walvis Bay Namibia DBT 0 2.69 4 0­15 85+ 0.34 0.02 ONAS Senegal IBT 0 0 3 0­20 40+ 0.02 0.13 Source: Banerjee, Foster, and others 2008. Note: AWSA = Addis Ababa Water Services Authority, DBT = direct block tariff, IBT = increasing block tariff, KIWASCO = Kisumu Water and Sewerage Company, NWASCO = Nairobi Water and Sanitation Company, ONAS = Office National de l'Assainissement du Sénégal, ONEA = Office Nationale des Eaux et d'Assainissement. 133 134 Africa's Water and Sanitation Infrastructure Box 5.1 Burkina Faso's Sanitation Tax The on-site sanitation problems in Ouagadougou are specifically addressed in the Sanitation Strategic Plan's implementation by the national public utility in charge of water supply and sanitation. A sanitation marketing approach has enhanced construction services offered to households by small providers and stimulated household demand for improved sanitation facilities. Approximately 700 masons and social workers have been trained since the beginning of the program. Burkina Faso's national utility offers to provide part of the material for free to households--equivalent to about a 30 percent subsidy with the rest financed by the households. The subsidy is financed by the utility through a small sanitation tax on the water bill. This example shows that on-site sanitation corresponds to a strong demand from urban dwellers, with more than 60,000 pieces of sanitation equipment sub- sidized so far--latrines as well as gray-water-removal systems. It also demon- strates the importance of a local financing mechanism. Donors have contributed to the mechanism, but only modestly. Most of the funds come from the tax on the water bill. Source: Reproduced from Water and Sanitation Program 2008. meter coverage (50 to 70 percent of residential connections), averaging 58 percent for the group. The fourth and final category comprises an additional 32 utilities (mainly in Burkina Faso, Cape Verde, Côte d'Ivoire, Ethiopia, Mozambique, Lesotho, Namibia, Niger, Rwanda, Senegal, and Uganda)1 with high meter coverage (greater than 70 percent of residen- tial connections), averaging 95 percent for the group. This section focuses only on the last three groups. Although water consumption measurements are not necessarily very accurate, evidence from the African utilities reviewed suggests that end- user water consumption is quite modest. The overall average consumption is 80 liters per capita per day, ranging from 189 liters per capita per day in the middle-income countries to 37 liters per capita per day in the low- income, fragile countries. Among the regional economic communities, con- sumption is particularly low in the EAC (at 42 liters per capita per day) compared with the Southern African Development Community (SADC) Urban Water Provision: The Story of African Utilities 135 and ECOWAS (at 77 to 86 liters per capita per day). In some countries, the actual consumption per capita might be lower because of widespread reselling, particularly in periurban areas with intermittent supply. Pricing is the main way that utilities can manage demand and requires a proper metering system to support volumetric charging and the application of metered tariffs to provide an adequate cost signal to customers. The overall reported rate of water metering in sample African countries whose utilities report medium to large metering ratios stands at 85 percent. Interestingly, the low-income, fragile coun- tries report a 100 percent metering ratio compared with only 68 percent in the middle-income countries, suggesting that rebuilding after a con- flict has involved a more formal release of connections with individual household meters. The average revenue per cubic meter of water billed ranges from around $0.40 in low-income countries to more than $1.10 in middle-income countries. The tariffs in water-abundant countries are two-thirds of those found in water-scarce countries. Within the regional economic communities, the ECOWAS has the highest average revenue, at $0.6 per cubic meter, compared with only $0.3 to $0.50 elsewhere in Africa. Many of the francophone countries of West Africa are in the CFA franc region, where prices tend to be systematically higher (table 5.5). Although this revenue is typically not sufficient to cover full capital costs, these costs are nonetheless quite high com- pared with those in other developing regions. Overall, evidence shows that significant price signals are getting through to a substantial share of the customer base. A fairly strong negative correlation is found between metering lev- els and average residential water consumption in utilities with a meter- ing level of about 50 percent of residential connections. Essentially, these utilities fall into two groups: Those with metering ratios of 50 to 70 percent tend to have average water consumption of about 188 liters per capita per day, and those with metering ratios of 90 to 100 percent tend to have average water consumption of about 50 liters per capita per day. Surprisingly, consumption and price are positively correlated as tariff rates are near cost recovery at high consumption levels. Utility clients pay a substantially higher price per unit of consumption, particularly high-volume nonresidential consumers. Thus, no strong evidence is evi- dent of wasteful overuse of water in Africa, and the relatively modest levels of consumption would not be further reduced by more aggressive use of demand management tools. 136 Africa's Water and Sanitation Infrastructure Table 5.5 Indicators of Demand Management Calculated across Utilities with Metering Ratios above 50 Percent Water consumption Revenue per per capita cubic meter served of water (liters per Metering consumed Nonrevenue capita per day) ratio (%) (US$/m3) water (%) Sub-Saharan Africa 79.5 85.4 0.5 30.1 By income Low-income 64.1 86.7 0.4 31.3 Low-income, fragile 36.9 100.0 0.6 24.8 Resource-rich -- 91.3 0.7 34.3 Middle-income 188.8 68.0 1.1 21.7 By regional economic community ECOWAS 77.0 101.4 0.6 22.1 SADC 85.8 82.1 0.5 30.0 CEMAC -- -- -- -- EAC 41.6 78.9 0.3 28.8 COMESA 60.0 90.1 0.5 35.8 By water availabilitya High water scarcity 102.2 81.0 0.6 30.2 Low water scarcity 68.6 87.2 0.4 30.0 By utility sizeb Small 64.6 85.4 0.4 30.8 Large 133.6 85.5 0.8 27.0 Sources: AICD WSS Database; Banerjee, Skilling, and others 2008. Note: CEMAC = Central African Economic and Monetary Community, COMESA = Common Market for Eastern and Southern Africa, EAC = East African Community, ECOWAS = Economic Community of West African States, SADC = Southern African Development Community. -- = not available. a. Water abundance is defined as renewable internal freshwater resources per capita in excess of 3,000 cubic meters. b. Large utilities are defined as those serving more than 100,000 connections. Substantial Water Losses in Distribution System Although end-user water use is modest, a substantial volume of water is lost during the distribution process. The average level of NRW in the sample is close to 30 percent, well above good practice levels (below 23 percent) for developing countries (Tynan and Kingdom 2002) (figure 5.4). The middle- income countries have the lowest nonrevenue losses, followed by the low- income, fragile countries. This good performance can be attributed to different factors--in the middle-income countries, it is due to superior technical and management performance, and for the low-income, fragile countries, it is due to relatively new systems constructed as part of the Urban Water Provision: The Story of African Utilities 137 Figure 5.4 Frequency Distribution of Nonrevenue Water 35 29 30 25 24 22 % utilities 20 15 12 10 9 5 4 0 less than 20­30 30­40 40­50 50­60 more than 20 60 NRW (%) Sources: AICD WSS Database; Banerjee, Skilling, and others 2008. Note: NRW = nonrevenue water. Figure 5.5 Cross-Plots between NRW and Other Variables a. Against access b. Against metering access to private residential water 90 90 80 80 70 70 connections (%) 60 60 NRW (%) 50 50 40 40 30 30 20 20 10 10 R² = 0.363 R²= 0.060 0 0 0 50 100 150 20 70 120 NRW (%) metering ratio (%) Sources: AICD WSS Database; Banerjee, Skilling, and others 2008. Note: NRW = nonrevenue water. rebuilding effort. Within the regional economic communities, the range is capped between 22 and 36 percent. Nonrevenue water measures include both technical and nontechnical losses. Experience in Asia suggests that NRW tends to be inversely pro- portional to access rates, because lower rates of access invite higher rates of informal and clandestine use, by both households and small-scale providers (McIntosh 2003). This relationship clearly holds for the African utilities, where there is a negative correlation of close to 33 percent between access rates and NRW (figure 5.5). 138 Africa's Water and Sanitation Infrastructure In principle, higher metering rates should help to reduce NRW by enabling utilities to pinpoint the location of losses on the network, but no evidence of such a relationship was found in the sample of African utili- ties. In fact, among utilities claiming 100 percent meter coverage, the level of NRW ranges between 20 and 47 percent. Moreover, utilities reporting moderate levels of meter coverage have an almost identical range for NRW. This suggests that utilities are not using metering effec- tively to control NRW. Difference in Quality of Service among Country Groups It is difficult to properly evaluate some of the services provided by African utilities. The only way to evaluate water quality is to look at the percentage of samples, taken from a water treatment plant, that pass the chlorine test. This indicates the effectiveness of the treatment process but says nothing about the quality of water received at the tap. The scores show a substantial difference in performance between utilities in middle- income countries, which score close to 100 percent on this variable, and those in low-income, fragile countries, which score only 75 percent. Among the regional economic communities, the Central African Economic and Monetary Community (CEMAC) is at the lower end of this indicator, compared to the EAC and SADC, which report a more than 90 percent success rate. On average, utilities for the sample group provide just under 20 hours of continuous service per day. However, low-income, fragile, and resource-rich countries provide, on average, five to six hours less service per day than middle-income countries. The countries with high water scarcity offer longer hours of service compared with water- abundant countries. Finally, the "complaints lodged by customers" indicator provides some- what nebulous information, because low levels of complaints could indicate either good service or a poor system for recording complaints. Overall, the indicators show much higher levels of complaints in low- income countries, where more than 200 complaints were lodged in the preceding year. The middle-income countries, on the other hand, recorded only 26 complaints per 1,000 connections. Among the regional economic communities, the number ranges from 50 to 183 complaints per 1,000 connections. The rate also varies widely among high- and low- water-scarcity countries, where the latter reported more than twice the number of complaints per 1,000 connections (table 5.6). Urban Water Provision: The Story of African Utilities 139 Table 5.6 Indicators of Service Quality Water and wastewater consumer complaints per connection Percentage of Continuous (number per samples passing water service 1,000 residential chlorine test (%) (hours per day) connections) Sub-Saharan Africa 87.9 19.6 78.4 By income Low-income 92.8 19.0 211.0 Low-income, fragile 75.3 18.2 -- Resource-rich 78.1 18.4 41.9 Middle-income 97.2 24.0 25.6 By regional economic community ECOWAS 88.4 22.8 183.0 SADC 90.1 17.2 50.0 CEMAC 68.0 19.5 -- EAC 94.8 16.0 119.7 COMESA 85.5 15.5 69.5 By water availabilitya High water scarcity 86.5 22.2 27.0 Low water scarcity 88.9 17.8 60.7 By utility sizeb Small 89.4 17.6 94.4 Large 80.8 19.6 77.5 Sources: AICD WSS Database; Banerjee, Skilling, and others 2008. Note: CEMAC = Central African Economic and Monetary Community, COMESA = Common Market for Eastern and Southern Africa, EAC = East African Community, ECOWAS = Economic Community of West African States, SADC = Southern African Development Community. -- = not available. a. Water abundance is defined as renewable internal freshwater resources per capita in excess of 3,000 cubic meters. b. Large utilities are defined as those serving more than 100,000 connections. Technical Efficiency and Effective Management of Operations Labor productivity, pipe water breaks, and operating cost are the three indi- cators used to evaluate the technical operations of the utilities (table 5.7). State-owned enterprises (SOEs) can be social buffers to (very inefficiently) transfer rents or resources to the population. Labor productivity rates can be hard to compare because of differing reliance on contractors. Nevertheless, a frequently used international benchmark for labor produc- tivity is 2 employees per 1,000 connections, which has been modified to 5 employees per 1,000 connections for developing countries (Tynan and Kingdom 2002). Overall, African utilities in the sample report an average 140 Africa's Water and Sanitation Infrastructure Table 5.7 Indicators of Operational Efficiency Employees Water pipe per 1,000 water breaks per year Operating connections per km of water cost per cubic (number/1,000 network (number meter of water connections) per year/km) consumed (US$/m3) Sub-Saharan Africa 5.6 8.0 1.2 By income Low-income 9.1 6.6 0.7 Low-income, fragile 11.1 7.9 0.7 Resource-rich 10.0 14.1 0.3 Middle-income 2.9 7.2 1.5 By regional economic community ECOWAS 5.2 3.6 0.7 SADC 5.0 7.3 1.3 CEMAC 6.3 58.0 0.5 EAC 11.0 5.5 0.5 COMESA 14.7 9.7 0.5 By water availabilitya High water scarcity 4.3 5.7 1.3 Low water scarcity 7.1 9.3 0.5 By utility sizeb Small 14.0 7.5 0.6 Large 6.3 13.7 0.7 Sources: AICD WSS Database; Banerjee, Skilling, and others 2008. Note: CEMAC = Central African Economic and Monetary Community, COMESA = Common Market for Eastern and Southern Africa, EAC = East African Community, ECOWAS = Economic Community of West African States, SADC = Southern African Development Community. a. Water abundance is defined as renewable internal freshwater resources per capita in excess of 3,000 cubic meters. b. Large utilities are defined as those serving more than 100,000 connections. of about 5.6 employees per 1,000 connections, which is right around the developing country benchmark cited above. The variation among the income groups is wide, ranging from 11 employees per 1,000 connections in the low-income, fragile countries to just about 3 employees per 1,000 connections in the middle-income countries. A commonly used international benchmark for average operating costs of water utilities is around $0.40 per cubic meter (Global Water Intelligence 2004). The costs reported by the African utilities are substan- tially higher, ranging from $0.30 per cubic meter in resource-rich countries to $1.50 per cubic meter in middle-income countries. The latter result is due to the high cost of water in Namibia and South Africa. Even within the regional economic communities, the average operating cost ranges from $1.30 per cubic meter in the SADC (which includes Namibia and South Africa) compared with $0.50 to $0.70 per cubic meter in other Urban Water Provision: The Story of African Utilities 141 regional blocks. As operation costs depend largely on water availability, the difference in costs between water-scarce countries and water-abundant countries is stark: The former have average operating costs almost three times that of the latter. The rate of bursts per kilometer of water main provides some indica- tion of the condition of the underlying infrastructure, and hence the extent to which it is being adequately operated and maintained. The resource-rich countries report the highest rate of bursts, at 14 per year per kilometer, compared with only 6.6 in low-income countries. The utilities in the CEMAC regional community report a significantly higher number of bursts compared with the other regional blocks.2 Three indicators are used to evaluate the primary components of oper- ating costs: labor costs, energy costs, and service contracts (table 5.8). Table 5.8 Utility Cost Structures (percent) Share of Share of Share of labor costs in energy costs service contracts operating in operating in operating expenses expenses expenses Sub-Saharan Africa 21.4 12.0 11.3 By income Low-income 28.3 14.9 26.3 Low-income, fragile 24.5 11.8 4.0 Resource-rich 33.9 29.7 12.5 Middle-income 15.9 1.6 6.6 By regional economic community CEMAC 34.5 -- -- COMESA 34.9 20.8 4.4 EAC 32.9 14.0 10.5 ECOWAS 22.1 14.8 23.6 SADC 19.1 7.5 6.8 By water availabilitya High water scarcity 18.7 10.8 8.2 Low water scarcity 25.5 10.0 6.8 By utility sizeb Small 33.4 19.7 15.1 Large 29.1 16.2 20.6 Sources: AICD WSS Database; Banerjee, Skilling, and others 2008. Note: CEMAC = Central African Economic and Monetary Community, COMESA = Common Market for Eastern and Southern Africa, EAC = East African Community, ECOWAS = Economic Community of West African States, SADC = Southern African Development Community. -- = not available. a. Water abundance is defined as renewable internal freshwater resources per capita in excess of 3,000 cubic meters. b. Large utilities are defined as those serving more than 100,000 connections. 142 Africa's Water and Sanitation Infrastructure Overall, African utilities allocate just more than 21 percent of their operating expenses to labor and just about 12 percent to energy. The structure of operating expenses differs substantially across the different groups. The share of labor and energy is lowest in the middle-income countries. In particular, utilities in the low-income and resource-rich countries allocate almost twice as high a share of their operating expenses to labor and multiple times to energy compared to the middle- income countries. The share of service contracts is lowest in the low- income, fragile, and middle-income countries. One simple explanation for this is that in both Namibia and South Africa, water-distribution util- ities are not involved in production, but instead purchase their water from bulk suppliers. So, although they spend a significant amount of operating expenses on bulk water purchase, their direct labor and energy costs are correspondingly reduced. Utilities in the ECOWAS allocate more than 23 percent of their operating costs on service contracts--more than utilities in the other regions. This may also explain why they have a correspondingly lower labor share than those in other regional blocks. Financial Efficiency and the Alignment of Operations and Finances Five indicators are used to evaluate the financial performance of the utilities: collection efficiency, operating cost ratio, debt-service ratio, value of gross fixed assets per connection, and average operating rev- enue (table 5.9). A well-performing utility is one that maintains its assets and uses them efficiently. This minimizes the need for new investments and reduces capital costs. The average operating ratio of African utilities shows that operating costs are barely covered and fall short of what is needed to recoup capital expenditures. This ratio is below the benchmark level of 1.3 for develop- ing countries identified by Tynan and Kingdom (2002). Paradoxically, the operating ratio reported for middle-income countries is below unity exhibited by low-income and resource-rich countries. One reason for this may be the exceptionally high operating costs (in excess of $1 per cubic meter) that are reported by utilities in middle-income countries. All the regional economic communities, except the SADC, which includes the middle-income countries of Namibia and South Africa, meet operational cost coverage. The economies of scale of large utilities are evident in the very high operating cost coverage at 3.4, which is three times that of the small utilities. Urban Water Provision: The Story of African Utilities 143 Table 5.9 Utility Financial Ratios Value of Operating gross fixed Average Collection cost assets per operating efficiency coverage Debt-service connection revenue (%) (ratio) ratio (US$) (US$/m3) Sub-Saharan Africa 92.2 0.9 11.1 490.2 0.9 By income Low-income 95.7 1.0 11.4 999.4 0.5 Low-income, fragile 96.9 0.8 20.4 558.7 0.5 Resource-rich 72.4 1.0 157.4 752.4 0.3 Middle-income 99.8 0.8 3.6 358.3 1.2 By regional economic community ECOWAS 105.4 1.0 16.0 934.1 0.8 SADC 86.8 0.8 4.7 385.7 1.0 CEMAC 91.0 1.1 157.4 1,112.1 0.4 EAC 97.5 1.0 21.8 353.8 0.3 COMESA 76.6 1.0 14.3 388.6 0.4 By water availabilitya High water scarcity 83.9 0.9 7.4 372.9 1.1 Low water scarcity 76.3 0.9 8.2 426.5 0.5 By utility sizeb Small 87.1 1.1 36.7 930.1 0.5 Large 91.4 3.4 15.3 1491.0 0.6 Sources: AICD WSS Database; Banerjee, Skilling, and others 2008. Note: CEMAC = Central African Economic and Monetary Community, COMESA = Common Market for Eastern and Southern Africa, EAC = East African Community, ECOWAS = Economic Community of West African States, SADC = Southern African Development Community. a. Water abundance is defined as renewable internal freshwater resources per capita in excess of 3,000 cubic meters. b. Large utilities are defined as those serving more than 100,000 connections. More utilities are able to cover operating costs at extremely low or extremely high levels of consumption than at average levels. More than 50 percent of the utilities recoup the operating cost at consumption lev- els of 4 m3 or 40 m3. Capital cost recovery,3 however, is close to impossi- ble in the African context. The highest number of utilities accomplish capital cost recovery at a subsistence consumption level of 4 m3, which has significant implications for equity. The degree of cost recovery is the lowest at an average consumption level of 10 m3. Households at the low and high ends of consumption are contributing more to cost recovery than the average consumer (figure 5.6). The average revenue per unit of water sold is $0.9, primarily because of relatively higher tariffs in the middle-income countries. The revenue in 144 Africa's Water and Sanitation Infrastructure Figure 5.6 Effective Tariffs at Various Consumption Levels a. Operating cost recovery 100 90 80 70 60 % utilities 50 40 30 20 10 0 4 m3 10 m3 40 m3 consumption b. Capital cost recovery 100 90 80 70 60 % utilities 50 40 30 20 10 0 4 m3 10 m3 40 m3 consumption no yes Source: Banerjee, Skilling, and others 2008. the middle-income countries is three times that of the low-income and low-income, fragile countries, and four times that of the resource-rich countries. Among the regional economic communities, the SADC reports an average revenue of about $1, which is significantly higher than anywhere else on the continent. Water is priced higher in water-scarce Urban Water Provision: The Story of African Utilities 145 countries than in water-abundant countries, suggesting that price signals are aligned with scarcity. Because of inconsistent accounting standards, data on asset values can paint only a broad picture. Replacement cost accounting is not widely practiced, so reported values likely reflect historic costs of investment. The average value of gross fixed assets per water connection is $490. The low-income countries report an average gross fixed value that is three times higher than the value in middle-income countries, primarily because the latter group has a significantly higher number of connections. We have few solid data points on utility debt. It appears that most util- ities do not list long-term debt on their balance sheets. Most utilities are not creditworthy and do not carry their own debt. The central govern- ment borrows the money, and the utilities are simply the recipients of the capital grants. As a result, the derived debt-service ratios indicate that lev- els of debt are so minimal that utilities can easily cover them through their operating revenue. The African utilities surveyed report collection ratios of more than 92 percent, on average. Resource-rich countries have the lowest levels of collection. In the regional economic communities, the collection ratio ranges from 76 percent in the Common Market for Eastern and Southern Africa (COMESA) to 105 percent in the ECOWAS, which may simply reflect a drive to collect arrears from earlier periods. Government entities are some of the most important consumers for water utilities. For instance, 42 percent of the total billings for the Régie de Production et de Distribution d'Eau (REGIDESO), in the Democratic Republic of Congo, are for government entities. Government agencies are responsible for 20 to 30 percent of total billings for the Office Nationale des Eaux et d'Assainissement (ONEA), Société de Distribution d'Eau de Côte d'Ivoire (SODECI), Lilongwe Water Board, and Nikana Water and Sewerage Company (NWSC). These agencies, however, can be the worst offenders in paying bills as well. Though no direct data are avail- able on government arrears, it is worth noting that the highest collection period--in REGIDESO--lasts about 2,000 days. The collection-efficiency ratios reported by the utilities are very high, relative to their experience. We, therefore, carried out a number of cross-checks on the data. First, using household survey data it is pos- sible to calculate the percentage of households with water service that do not report paying a utility bill. This provides a first-order estimate of the extent of undercollection from the residential sector, though the numbers will make the phenomenon seem greater than it really is 146 Africa's Water and Sanitation Infrastructure Figure 5.7 Reported versus Implicit Collection Ratios 80 number of utilities responding (%) 70 60 50 40 30 20 10 0 < 50 50­70 70­90 > 90 collection ratio (%) reported implicit from tariffs derived from surveys Sources: AICD WSS Database; Banerjee, Skilling, and others 2008. because they do not distinguish between formal connections that do not pay for service and informal connections that are not billed. Second, it is possible to compare the average revenue that the utility collects per cubic meter with the average tariff charged based on the tariff schedule. This shows which revenue falls short of the tariffs that have been charged. Figure 5.7 compares the distribution for these three measures of collection efficiency. Whereas the vast majority of utilities report col- lection ratios above 90 percent, almost half of the utilities present implicit collection rates below 70 percent, and more than half of the utilities collect tariff revenue from fewer than 50 percent of their cus- tomers, according to household surveys. The High Cost of Inefficiencies in Operations and Pricing The inefficiency of the service providers and considerable mispricing in the water sector adversely affects optimal resource allocation and the financial sustainability of the sector. One way of presenting a global measure of utility inefficiency is to quantify the dollar cost of observ- able operational inefficiencies. This concept, the "hidden cost," is a measure of wastefulness and ineptitude. Hidden cost indicates the cost of inefficient production and partially quantifies opaque transfers from producers to consumers (Mackenzie and Stella 1996). Hidden cost also provides distorted incentives to the utilities and consumers, leading to Urban Water Provision: The Story of African Utilities 147 overconsumption and wasting of scarce resources (Briceño-Garmendia, Smits, and Foster 2008). Even without explicitly revealing itself in the budget, it affects the macroeconomic stability and underreports the size of the public sector. The hidden cost estimates the financial losses associated with four components--undercollected revenue, distribution losses, underpricing, and overstaffing--and expresses these losses as a percentage of the utili- ties' overall turnover. These inefficiencies can be quantified by comparing the revenue available to the utility with the revenue available to an ideal utility that is able to charge cost-recovery tariffs, collect all of its revenue, minimize distribution losses, and employ an ideal number of workers per connection (box 5.2). Box 5.2 Methodology for Estimation of Hidden Cost The current profile of the utilities on these four indicators is measured against the ideal scenario, which includes the following: Nonrevenue water. An internationally accepted benchmark of 20 percent NRW is employed. Cost-recovery tariff. A capital premium of $0.40/m3 (Global Water Intelligence 2004) is added to the O&M cost (available from the AICD WSS Database) to arrive at the cost-recovery tariff. The collection ratio. This is instituted as 100 percent. Overstaffing. Two hundred connections per employee is an accepted bench- mark. This estimate is taken from two sources: (a) The estimate--averaging more than 302 utilities from developing coun- tries, excluding Sub-Saharan Africa--taken from the database amassed by Gassner, Popov, and Pushak (2008, http://www.ppiaf.org/documents/trends_and _policy/PSP_water_electricity.pdf ) is 230 connections per employee. (b) An analysis of data from 246 water utilities (including 123 utilities from 44 developing countries) proposed a benchmarking target of 5 or fewer staff per 1,000 connections for developing-country water utilities (that is, 200 connections per employee). This target was based on the levels of productivity actually being achieved by the top quartile of developing-country utilities within the data- base. By contrast, many developing-country utilities reported more than 20 staff per 1,000 connections (Tynan and Kingdom 2002). Source: Briceño-Garmendia, Smits, and Foster 2008. 148 Africa's Water and Sanitation Infrastructure The hidden costs constitute 145 percent of the total billings in water utilities in Africa. The utilities that report the lowest hidden cost (as a share of total billings) are Plateau in Nigeria and Togolaise des Eaux in Togo. The highest is Upper Nile Water Corporation in Sudan, which loses 1,700 percent of its revenue to operational and pricing inefficiencies (figure 5.8). The hidden costs, comprising underpricing and operational ineffi- ciencies, amount to 0.4 percent of gross domestic product (GDP) (table 5.10). On average, the contribution of the two components is similar. Underpricing costs Africa 0.2 percent of GDP or $1.5 billion annually. In other words, revising tariffs to make them equal to historic recovery unit average costs, which would enable all African water util- ities to recover capital costs as well, would increase the potential for efficiency gains to $1.5 billion a year. In GDP terms, the countries that are most affected by the pricing inefficiency are low-income fragile states, where it accounts for 0.9 percent of their GDP (or $0.4 billion per year). On the other hand, under-recovery of tariffs weighs the least on GDP for utilities in resource-rich countries (0.1 percent of GDP or $0.2 billion per year). Three types of operational inefficiencies account for 0.2 percent of GDP on average, or $1.3 billion per year: distributional losses, undercol- lection of bills, and overstaffing or labor inefficiencies. First, utilities incur substantial losses on their water distribution networks. Poor network maintenance (which leads to physical leakage) and poor network man- agement (which leads to clandestine connections and various forms of theft) each partially explains these distribution losses. Distribution losses amount to $0.4 billion a year (0.07 percent of GDP). African water util- ities typically lose 35 percent of their water in distribution losses, nearly twice the 20 percent benchmark. Second, water utilities face serious problems in collecting their bills: Undercollection of bills costs almost $0.5 billion a year (0.07 percent of GDP). African water utilities man- age to collect about 90 percent of the bills owed to them by their cus- tomers, short of a best practice benchmark of close to 100 percent. Third, SOEs may retain more employees than are strictly necessary to discharge their functions, often because of political pressure to provide jobs for members of certain interest groups. Overstaffing is estimated to cost utilities at least $0.4 billion a year, or 0.06 percent of GDP. African water utilities have overstaffing ratios of 24 percent over developing-country benchmarks, and a typical utility has approximately 5.6 employees per Urban Water Provision: The Story of African Utilities 149 Figure 5.8 Utility Inefficiencies as Percentage of Total Utility Revenue Upper Nile Water Corporation Kaduna FCT SODECI AdeM Maputo BWB DIRE DAWA ELECTROGAZ AdeM Pemba Khartoum Water Corporation AdeM Nampula KIWASCO REGIDESO LWSC AWSA Drakenstein Municipality AdeM Beira LWB AdeM Quelimane NWSC NWASCO ADAMA ONEA SPEN SDE MSNE South Darfur Water Corporation SONEB MWSC ELECTRA WUC TdE Plateau 0 500 1,000 1,500 2,000 total billings (%) unaccounted losses underpricing collection inefficiencies labor inefficiencies Source: AICD WSS Database; Banerjee, Skilling, and others 2008. Note: ADAMA = Nazareth Water Company; AWSA = Addis Ababa Water Services Authority; BWB = Blantyre Water Board; FCT = Federal Capital Territory Water Board; KIWASCO = Kisumu Water and Sewerage Company; LWB = Lilongwe Water Board; LWSC = Lusaka Water and Sewerage Company; MSNE = Mauritania Société Nationale d'Eau et d'Electricité; MWSC = Mombasa Water and Sewerage Company; NWASCO = Nairobi Water and Sanitation Company; NWSC = National Water and Sewerage Company, Uganda; ONEA = Office Nationale des Eaux et d'Assainissement; REGIDESO = Régie de Production et de Distribution d'Eau; SDE = Sénégalaise des Eaux; SODECI = Société de Distribution d'Eau de Côte d'Ivoire; SONEB = Société Nationale des Eaux du Benin; SPEN = Société de Patrimoine des Eaux du Niger; TdE = Togolaise des Eaux; WUC = Water Utilities Corporation, Botswana. 150 Table 5.10 Hidden Cost of Inefficiencies GDP share (%) US$ million per year Operational inefficiencies Operational inefficiencies Total Tariff Total Tariff Labor operational cost Labor operational cost inefficiencies Losses Undercollection inefficiencies recovery Total inefficiencies Losses Undercollection inefficiencies recovery Total Sub-Saharan Africa 0.06 0.07 0.07 0.2 0.23 0.43 375 425 458 1,259 1,450 2,709 Low-income, fragile countries 0.04 0.17 0.06 0.28 0.93 1.21 17 65 25 106 358 464 Low-income, nonfragile countries 0.08 0.1 0.06 0.24 0.35 0.59 87 111 67 265 381 646 Middle-income countries 0.03 0.06 0.1 0.18 0.2 0.38 68 150 274 492 537 1,029 Resource-rich countries -- 0.05 0.03 0.08 0.1 0.18 -- 103 69 172 214 386 Source: Briceño-Garmendia, Smits, and Foster 2008. Note: -- = not available. Urban Water Provision: The Story of African Utilities 151 1,000 connections though the developing-country benchmark is only 2 employees per 1,000 connections. In some cases, there are 42 employees per 1,000 connections. These results for labor inefficiencies underscore the importance of strengthening external governance mechanisms that can impose discipline on the behavior of SOEs. Overstaffing partially explains why in African countries with a publicly owned operator the share of spending allocated to capital spending frequently remains below 25 percent despite increasing spending needs. Utilities in low-income, nonfragile countries present the highest labor inefficiencies among the four-country group (0.08 percent of their GDP). These inefficiencies can be attributable to the fact that African SOEs are characterized by low investment and high operating inefficiency. Water SOEs account for 40 percent of total public expenditures (central government and nonfinancial enterprises). Despite their large resource base, they invest comparatively little (on average) only 18 percent of the government water resource envelope. As a result, governments are typi- cally required to step in to assume most SOE investment responsibilities, which are confined to undertaking daily O&M. Most SOEs operate at arm's length from the central government and fail in practice to meet cri- teria for sound commercial management. When these enterprises run into financial difficulties, the central government--as the main stakeholder-- acts as the lender of last resort, absorbs debts, and assumes by default the financial, political, regulatory, and mismanagement risks. Lumpy capital- izations and debt swaps that cover the cumulative cost of operational inefficiencies are frequent events in the African utility sector, which have the potential to create a moral hazard that would perpetuate operational inefficiencies if proactive reforms are not undertaken. Undermaintenance is another source of inefficiencies in African WSS utilities, although this has not been quantified given the scarce data for the sector. The underinvestment in O&M can greatly affect continuity of service, level of technical and commercial losses, and adequate capacity and functioning of treatment, transmission, and distribution systems. The lack of institutional capacity and regulation, the absence of fiscal disci- pline and availability of resources, and the persistence of civil conflict in Africa during the past two decades have left WSS facilities neglected by inadequate O&M, which eventually increases the spending needs for rehabilitation and construction of new assets. Operating inefficiencies have been impeding expansion. Inefficiencies not only drain the public purse but also seriously undermine the perform- ance of utilities. One casualty of insufficient revenue is maintenance. 152 Africa's Water and Sanitation Infrastructure The rate of bursts per kilometer of water mains reflects the condition of the underlying infrastructure, and the extent to which it is being adequately operated and maintained. Among African utilities, huge variation is seen between low- and middle-income countries, with bursts ranging from five per kilometer in the latter to just more than one per kilometer in the former. Utility managers often have to choose between paying salaries, buying fuel, or purchasing spare parts. Often they have to cannibalize parts from other working equipment. The investment program is another major casualty. Service expansion-- measured as the percentage of residents in the utility service area that gains access to either piped water or standposts per year--is signifi- cantly higher for more efficient utilities. In particular, utilities with low hidden costs have an average annual increase in coverage of more than 3 percent, essentially twice as much as the annual increase of utilities with high hidden costs (figure 5.9). Overstaffing also seems to hinder expansion. For similar reasons, more efficient utilities deliver better quality water. Utilities with lower rates of employees per connection manage to have on average 85 percent of water supplied with adequate chlorine, compared with 75 percent of the rest of the utilities. Conversely, utilities with higher hidden costs tend to deliver slightly higher quality water. Figure 5.9 Utility Efficiency Affects Access Expansion and Water Quality a. Access expansion b. Water quality Average annualized increase in 4.0 90 access to improved water (%) 3.5 85 chlorine test (%) 3.0 sample passing 2.5 80 2.0 1.5 75 1.0 70 0.5 0 65 employees per hidden cost employees per hidden costs connection connection worse than average better than average better than average worse than average Source: Banerjee, Skilling, and others 2008. Urban Water Provision: The Story of African Utilities 153 The Role of Institutions in Improving Performance Good institutional frameworks help to lower the inefficiency of utilities, and institutional reform is key to improving performance. Utilities that have decentralized or adopted private sector management have substan- tially lower hidden costs than those that have not. Unbundling also has a significant effect, but unbundling is rare in Africa and exclusively con- centrated in middle-income countries, whose superior performance can be explained for many other reasons. Conversely, higher levels of regula- tion and governance, as well as corporatization, are associated with lower efficiency in the form of higher hidden costs (figure 5.10). The reform agenda has had two major thrusts: increasing private par- ticipation and improving governance from within. Private sector participation has helped to improve utility performance, with Senegal being particularly noteworthy. Management contracts, being relatively short-term instruments, have had a material effect on improv- ing revenue collection and service continuity. However, they have not had much of an impact on more intractable issues, such as unaccounted for water and access. Lease contracts (and the associated public-funded investments) have drastically improved access and boosted operational Figure 5.10 Hidden Costs and Institutions 200 hidden costs (% total billings) 150 100 50 0 n en r g ce n n em cto io io io lin an t at at at nd ag se rn liz l tiz gu bu an te ve ra ra re m riva un go nt po gh ce or p gh hi de Ec hi SO yes no Source: Banerjee, Skilling, and others 2008. Note: SOE = state-owned enterprise. 154 Africa's Water and Sanitation Infrastructure efficiency. With the exception of Côte d'Ivoire, however, the associated investments have been publicly financed. The lease contracts in Guinea and Maputo have been affected by a lack of coordination between the private contractor and the government, which has stalled progress in key areas, such as unaccounted-for water. Overall, private sector contracts accounted for almost 20 percent of the increase of household connections in the region, twice the amount that would be expected, given their mar- ket share of only 9 percent (table 5.11). However, half of these gains were made in Côte d'Ivoire alone (which has been adversely affected since the onset of civil war in 2002). About half of the countries (mainly anglophone) have established distinct regulatory agencies for the water sector, although a significant number of these have not adopted private sector participation. Conversely, numerous francophone countries with private participa- tion have adopted regulatory frameworks contractually, without establishing an independent regulatory agency. There does not appear to be any evidence supporting the superiority of any one of these two approaches. Even where explicit regulatory frameworks have been established, these typically meet only about half of the corresponding good practice criteria. However, evidence on the links between intro- ducing an independent regulator and improving performance is negli- gible for the water sector. Similarly, no conclusive evidence is seen for the superiority of regulation by contract over the traditional form of regulation by agency (Vagliasindi and Nellis 2009). Of governance reforms that appear to be the most important drivers of higher performance, two appear especially promising: performance contracts with incentives and independent external audits. Uganda has enjoyed success using a performance contract in its water company, pro- viding the utility with incentives for good performance and producing greater accountability (box 5.3). The introduction of independent audits has also positively affected efficiency. Table 5.11 Overview of Impact of Private Sector Participation on Utility Performance Unit change in performance before and after private participation Household Improved Service Unaccounted-for Collection Labor Contract connections water continuity water ratio productivity Gabon Concession contract +20 ­8 Mali +15 +29 ­14 Côte d'Ivoire Lease contract or +19 +22 +2.6 affermage Guinea +27 0 Maputo +2 +10 ­1 +24 Niger +9 +3 ­5 +3.2 Senegal +18 +17 ­15 +2.8 Johannesburg Management contract 0 +10 Kampala +6 ­2 +12 Zambia +5 ­28 +19 Source: Adapted from Marin 2009. Note: Blank cells denote missing data; household connections and improved water are measured as additional percentage points of households with access; service continuity is measured as additional hours per day of service; unaccounted-for water is measured as reduced percentage points of losses; collection ratio is measured as additional percentage points of collection; and labor productivity is measured as additional thousands of connections served per employee. 155 156 Africa's Water and Sanitation Infrastructure Box 5.3 Uganda's Successful Case of State-Owned Enterprise Reform The National Water and Sewerage Corporation (NWSC) is an autonomous public corporation, wholly owned by the government of Uganda, that is responsible for water and sanitation services in 23 towns with a population of 2.2 million, 75 percent of the population in Uganda's large urban centers. Large inefficiencies before 1998, including poor service quality, very low staff productivity, and high operating expenses, with the collection rate at only 60 percent and a monthly cash deficit of $300,000, posed an urgent need to revamp operations. Turnaround strategies culminated in establishing area performance con- tracts between a NWSC head office, which performs contract oversight and capital investment, as well as regulation of tariffs, rates, and charges, and the area managers, acting as operators and, therefore, responsible for management, operation and maintenance services, revenue collection, and rehabilitation and extension of networks. The objective was to enhance each area's performance by empowering managers and making them accountable for results. A comprehen- sive system of more focused and customer-oriented targets was designed. Typical performance indicators included working ratio, cash operating margin, nonrev- enue water, collection efficiency, and connection ratio. Performance evaluation looked at both processes and outputs and was conducted through regular as well as unannounced visits. Incentives were both financial (including penalties for per- formances below targets) and nonfinancial (including trophies for best perform- ing areas/departments and publication of monthly, quarterly, and annual best as well as worst performances). In fiscal 2003­04, the Area Performance Contracts were changed into Inter- nally Delegated Area Management Contracts (IDAMCs), aimed at giving more autonomy to operating teams and based on clearer roles, better incentive plans, and a larger risk apportioned to operating teams. The IDAMC framework was (continued next page) Urban Water Provision: The Story of African Utilities 157 Box 5.3 (continued) later consolidated by the use of competitive bidding as a basis for awarding contracts to the operating units. A review of 10 years of NWSC operations shows that gains in operational and financial efficiency and service expansion have been substantial and impressive relative to the performance of the NWSC's peers in Africa. NWSC Efficiency Gains Year Performance indicator 1998 2008 Service coverage 48% 72% Total connections 50,826 202,559 New connections per year 3,317 25,000 Metered connections 37,217 201,839 Staff per 1,000 connections 36 7 Collection efficiency 60% 92% NRW 60% 32.50% Proportion of metered accounts 65% 99.60% Annual turnover (billion U Sh) 21 84 Profit (after dep.) (billion U Sh) ­2.0 +3.8 Source: Muhairwe 2009. Note: NRW = nonrevenue water, U Sh = Ugandan shilling. Key success factors are indentified in the empowerment of staff, devolution of power from the center to regional operations, increased customer focus, as well as adoption of private sector­like management practices, including performance-based pay, the "customer pays for good service" principle, and so on. Also, the emphasis on planning, systematic oversight and monitoring, infor- mation sharing through benchmarking, and continuously challenging manage- ment teams with new and clear performance targets have created a strong system of checks and balances and powerfully triggered involvement, engagement, and a sense of pride on the side of the staff, beyond what simple financial incentives may obtain. Sources: Adapted from Muhairwe 2009; National Water and Sewerage Corporation n.d. 158 Africa's Water and Sanitation Infrastructure Annex 5.1 Utilities in the AICD WSS Database Population Coverage in service of service Sewerage No. Country Utility area area network 1 Benin SONEB 2,900,000 National No 2 Burkina Faso ONEA 2,779,875 National Yes 3 Cameroon SNEC -- Yes 4 Cape Verde ELECTRA 231,882 National Yes 5 Chad STEE -- National No 6 Congo, Dem. Rep. REGIDESO 18,000,000 National No 7 Côte d'Ivoire SODECI 8,892,850 National Yes 8 Ethiopia ADAMA 218,111 Urban No 9 Ethiopia AWSA 2,887,000 Urban Yes 10 Ethiopia Dire Dawa 284,000 Urban Yes 11 Ghana GWC 17,199,942 National Yes 12 Kenya KIWASCO 465,613 Urban Yes 13 Kenya MWSC 826,000 Urban No 14 Kenya NWASCO 2,496,000 Urban Yes 15 Lesotho WASA 540,500 National Yes 16 Madagascar JIRAMA 4,885,250 National Yes 17 Malawi BWB 833,418 Urban No 18 Malawi CRWB 288,705 Urban No 19 Malawi LWB 634,447 Urban Yes 20 Mozambique AdeM Beira 580,258 Urban No 21 Mozambique AdeM Maputo 1,778,629 Urban No 22 Mozambique AdeM Nampula 385,809 Urban No 23 Mozambique AdeM Pemba 131,980 Urban No 24 Mozambique AdeM Quelimane 288,887 Urban No 25 Namibia Oshakati Municipality 31,432 Urban Yes 26 Namibia Walvis Bay Municipality 54,025 Urban Yes 27 Namibia Windhoek Municipality 300,000 Urban Yes 28 Niger SEEN/SPEN 2,240,689 National Yes 29 Nigeria Borno -- Urban No 30 Nigeria FCT 6,000,000 Urban No 31 Nigeria Kaduna 3,126,000 Urban No 32 Nigeria Katsina 2,845,920 Urban No 33 Nigeria Lagos 15,367,417 Urban No 34 Nigeria Plateau 1,334,000 Urban No 35 Rwanda ELECTROGAZ 2,010,000 National No 36 Senegal SDE/ONAS 7,808,142 National Yes 37 South Africa Cape Town Metro 3,241,000 Urban Yes 38 South Africa Drakenstein Municipality 213,900 Urban Yes 39 South Africa eThekwini (Durban) 3,375,000 Urban Yes (continued next page) Urban Water Provision: The Story of African Utilities 159 Population Coverage in service of service Sewer No. Country Utility area area network 40 South Africa Johannesburga 3,753,900 Urban Yes 41 Sudan Khartoum Water 7,602,000 Urban Yes Corporation 42 Sudan South Darfur Corporation 2,051,000 Urban No 43 Sudan Upper Nile Water 250,000 Urban No Corporation 44 Tanzania DAWASCO -- Urban Yes 45 Tanzania DUWS 279,000 Urban Yes 46 Tanzania MWSA 458,493 Urban Yes 47 Uganda NWSC 2,284,000 National Yes 48 Zambia LWSC 1,564,986 Urban Yes 49 Zambia NWSC 990,806 Urban Yes 50 Zambia SWSC 294,000 Urban Yes Source: Banerjee, Skilling, and others 2008a. Note: ADAMA = Nazareth Water Company; AWSA = Addis Ababa Water Services Authority; BWB = Blantyre Water Board; CRWB = Central Region Water Board; DAWASCO = Dar es Salaam Water and Sewerage Company; DUWS = Dodoma Urban Water and Sewerage Authority; FCT = Federal Capital Territory Water Board; GWC = Ghana Water Company, JIRAMA = Jiro sy Rano Malagasy; KIWASCO = Kisumu Water and Sewerage Company; LWB = Lilongwe Water Board; LWSC = Lusaka Water and Sewerage Company; MWSA = Mwanza Water and Sewerage Authority; MWSC = Mombasa Water and Sewerage Company; NWASCO = Nairobi Water and Sanitation Company; NWSC = National Water and Sewerage Company, Uganda; NWSC = National Water and Sewerage Company, Zambia; ONAS = Office National de l'Assainissement du Sénégal; ONEA = Office Nationale des Eaux et d'Assainissement; REGIDESO = Régie de Production et de Distribution d'Eau; SDE = Sénégalaise des Eaux; SEEN = Société de Exploitation des Eaux du Niger; SNEC = Société National des Eaux du Cameroon; SODECI = Société de Distribution d'Eau de Côte d'Ivoire; SONEB = Société Nationale des Eaux du Benin; SPEN = Société de Patrimoine des Eaux du Niger; STEE = Société Tchadienne d'Eau et d'Electricité; SWSC = Southern Water and Sewerage Company; WASA = Water and Sanitation Authority. -- = data not available. Notes 1. Francophone countries have a much stronger metering tradition, which reflects different traditions in France and England. 2. This number reflects the value for SNDE (Société Nationale de Distribution d'Eau, the Republic of Congo). 3. Capital cost = O&M cost + capital premium of $0.4/m3. The capital premium is based on internationally used benchmarks computed by Global Water Intelligence (2004). References Banerjee, S., V. Foster, Y. Ying, H. Skilling, and Q. Wodon. 2008. "Cost Recovery, Equity, and Efficiency in Water Tariffs: Evidence from African Utilities." AICD Working Paper 7, World Bank, Washington, DC. 160 Africa's Water and Sanitation Infrastructure Banerjee, S., H. Skilling, V. Foster, C. Briceño-Garmendia, E. Morella, and T. Chfadi. 2008. "Ebbing Water, Surging Deficits: Urban Water Supply in Sub-Saharan Africa." Background Paper 12, Africa Infrastructure Country Diagnostic, World Bank, Washington, DC. Briceño-Garmendia, C., K. Smits, and V. Foster. 2008. "Financing Public Infrastructure in Sub-Saharan Africa: Patterns and Emerging Issues." AICD Background Paper 15, World Bank, Washington, DC. Global Water Intelligence. 2004. "Tariffs: Half-Way There." Global Water Intelligence, Oxford, UK. Mackenzie, G. A., and P. Stella. 1996. "Quasi-Fiscal Operations of Public Financial Institutions." Occasional Paper 142, International Monetary Fund, Washington, DC. Marin, P. 2009. "Public-Private Partnerships for Urban Water Utilities: A Review of Experiences in Developing Countries." Trends and Policy Options 8. PPIAF and World Bank, Washington, DC. McIntosh, A. C. 2003. "Asian Water Supply: Reaching the Poor." Asian Development Bank and International Water Association, Manila and London. Muhairwe, W. T. 2009. "Fostering Improved Performance through Internal Contractualisation." Paper presented at World Bank Water Week, Washington, DC, February 17­20. National Water and Sewerage Corporation. n.d. NWSC Corporate Plan 2006­09, Kampala, Uganda. http://www.nwsc.co.ug/affairs02.php?cat=corporate_plan. Tynan, N., and W. Kingdom. 2002. "A Water Scorecard: Setting Performance Targets for Water Utilities." Public Policy for the Private Sector Note 242, World Bank, Washington, DC. Vagliasindi, M., and J. Nellis 2009. "Evaluating Africa's Experience with Institutional Reforms for the Infrastructure Sectors." AICD Working Paper 23, World Bank, Washington, DC. Water and Sanitation Program. 2008. "Can Africa Afford to Miss the Sanitation MDG Target? A Review of the Sanitation and Hygiene Status in 32 Countries." World Bank, Washington, DC. CHAPTER 6 Cost Recovery, Affordability, and Subsidies The need to provide Africans with safe drinking water is immense and immediate. As a poor continent, however, Africa lacks the level of house- hold and government funds required to significantly expand water networks and improve service quality. In the best-case scenario, its gov- ernments could set tariffs at cost-recovery levels so that water service providers could justify investments in expanded networks, but a signifi- cant share of the existing and potential consumer base cannot afford to pay at that rate. This chapter uses household survey data in the Africa Infrastructure Country Diagnostic (AICD) to examine Africa's ability to pay for water services and implement operating and capital cost-recovery tariffs. It eval- uates the targeting and actual performance of existing tariffs' subsidy mechanisms and considers alternative systems with potentially better outcomes. Average Monthly Spending on Water Most African households live on very modest budgets. The average African household survives on not more than $180 per month; urban household budgets are about $100 per month higher than those of rural households. Household budgets range from $60 per month in the lowest 161 162 Africa's Water and Sanitation Infrastructure Figure 6.1 Spending on Water Services a. Frequency distribution of overall b. Spending on water services household budget share devoted to (in 2002 US$) water services 80 7 water spending per month 70 6 60 5 % countries (2002 US$) 50 4 40 3 30 20 2 10 1 0 0 spending on water services ur l n l Q1 Q2 Q3 Q4 Q5 na ra ba ru tio as a share of household budget na households below 2.5% 2.6­5.0% 5.1­10.0% Source: Banerjee, Wodon, and others 2008. Note: Q = quintile. quintile to no more than $400 per month in the highest income quintile except in middle-income countries, where the richest quintile has a monthly budget of $200 to $1,300 (table 6.1). On average, Africans spend more than half their household budget on food. Monthly spending on water averages $4, or 2 percent of household budgets, and rarely exceeds 3 percent. Only in Cameroon, Mauritania, and Rwanda are water expenses more than 5 percent of the household budget. Spending on water services increases with rising income levels: The top 20 percent of African households pay $6 per month (2 percent of income), primarily because they are disproportionately connected to formal water networks (figure 6.1). Wide Price Variations among Service Providers in the Urban Water Market The price of water in the unserved market is substantially higher than the price utilities charge for household connections. Utilities supply piped water delivered through public standposts in addition to piped connec- tions to houses and yards. Prices at public standposts are usually subsi- dized so that low-income households in periurban areas can benefit from improved water supply. The important policy questions are whether this practice realizes the objective of providing affordable water to public standpost users and the extent of cross-subsidy between the low-volume Table 6.1 Monthly Household Budget Total household budget Food expenditure as a share of total household (2002 US$) budget (%) National Rural Urban Q1 Q2 Q3 Q4 Q5 National Rural Urban Q1 Q2 Q3 Q4 Q5 Overall 177 130 241 59 97 128 169 340 55 61 48 63 64 63 60 48 Low-income countries 139 109 208 53 80 103 135 258 59 64 50 67 68 66 64 52 Middle-income countries 300 199 350 79 155 181 282 609 45 54 42 51 55 52 50 38 Source: Banerjee, Wodon, and others 2008. Note: Q = quintile. 163 164 Africa's Water and Sanitation Infrastructure Figure 6.2 Comparison of Official and Retail Standpost and Smalla Piped Consumer Prices 60 50 40 utilities (%) 30 20 10 0 Ratio of unofficial to official standpost price Ratio of official piped-water price at 4 m3 to official standpost price <1 > 1 and < 2 > 2 and < 5 >5 Source: Banerjee, Foster, and others 2008. Note: The Democratic Republic of Congo is not included in the graph because the formal standpost price is almost negligible. Figure based on information available for 12 utilities. a. Refers to minimum consumption level of 4 m3. consumers at public standposts and those who have household connec- tions to piped water. The average official price is $0.63/cubic meter (m3) at public standposts and $0.55/m3 for small consumers of household connections to piped water. Standpost consumers are paying more to approximately half the utilities. For the rest, the evidence suggests that consumers whose house- holds are connected to piped water are cross-subsidizing standpost con- sumers at the same level of consumption (figure 6.2). This would be extremely inequitable if the standpost and low-volume piped-water con- sumers were in similar income strata. The official standpost tariff may not, however, be what consumers really pay. Operators and middlemen come between the utility and con- sumers. The result is a highly dynamic market in which, except in Ouagadougou, informal retail prices are much higher than the official standpost tariffs. For half the utilities, the informal standpost price is between two and five times the formal standpost price. This is true of dense periurban areas with shortages of households connected to piped water and a significant dependence on public standposts (box 6.1). For instance, in Antananarivo, Lusaka, and Cotonou, retail prices are more than five times higher than official tariffs. In the largest African cities, alternatives to piped water supply are priced from 1.3 times as high for small piped-water networks to 10 to 20 Cost Recovery, Affordability, and Subsidies 165 Box 6.1 Piped Water Delivered through Public Standposts in Kigali, Rwanda The water production capacity of ELECTROGAZ, the main utility in Kigali, is inad- equate to meet network demand. The lack of bulk supply causes rolling outages throughout the city and often forces residents with private connections to seek water at public sources, such as public standposts. The financial stability of Kigali public standposts can be estimated from the tariff paid by standpost operators (RF 240, $0.42 per cubic meter), the total cost of production by ELECTROGAZ (RF 205), the rate of unaccounted-for water in distri- bution and selling (35 percent and 5 percent, respectively), and the volume and price of water sold at the public standposts. Three operators selling 100 jerricans each per day at RF 10, 20, and 30 per jerrican would earn estimated monthly net incomes of $314, $949, and $1,584 (the 2008 gross domestic product per capita was $370). The combination of a low tariff and a 35 percent rate of unaccounted- for water in distribution creates losses for the utility. Of the roughly 240 public standposts in Kigali, an estimated 193 (80 percent) were operating in December 2008. Utility officials estimate that 60,000 people use piped water delivered through public standposts, though this figure includes consumers who use them only when their primary source is unavailable. Based on total water volume recorded at meters, public standposts could supply only 48,500 people with 20 liters daily. That figure is equal to the upper segment of the population that depends primarily on public standposts (about 6 percent of the city's population). The utility's limited production capacity has affected both the level of peak de- mand at public standposts and the cost of production. Observations and inter- views with consumers indicate that prices have often been higher in areas when and where water service has been cut--and lower after periods of precipitation that increase the availability of other supply options, such as rainwater and natu- ral springs. Source: Keener and others (forthcoming). times as high for mobile distributors (table 6.2). The lower prices are paid by small utility consumers, and the higher prices are paid by unserved consumers of alternatives. They do not benefit from utility service and must pay significantly more. Moreover, the prices charged by each water provider in the informal sector also show a higher variation than those Table 6.2 Prices by Alternate Water Service Provider Household Small piped Household 166 connection network Standpipe reseller Water tanker Water vendor Country Largest city (US$/m3) (US$/m3) (US$/m3) (US$/m3) (US$/m3) (US$/m3) Benin Cotonou 0.41 n.a. 1.91 1.91 n.a. n.a. Burkina Faso Ouagadougou 0.90 n.a. 0.48 n.a. n.a. 1.67 Ethiopia Addis Ababa 0.19 n.a. 0.87 1.44 3.85 -- Mozambique Maputo 0.96 0.98 0.98 0.98 n.a. -- Niger Niamey 0.52 n.a. 0.48 n.a. n.a. 1.79 Nigeria Kaduna 0.17 n.a. -- -- 3.43 5.71 Rwanda Kigali 0.44 n.a. 1.79 1.79 4.48 n.a. Senegal Dakar 0.37 n.a. 1.53 -- n.a. 2.29 South Africa Johannesburg 0.05 n.a. n.a. n.a. -- -- Congo, Dem. Rep. Kinshasa 0.05 2.11 1.02 1.01 n.a. n.a. Ghana Accra 0.52 n.a. 5.51 1.53 5.46 6.89 Kenya Nairobi 0.18 0.60 1.73 n.a. 3.74 3.47 Lesotho Maseru 0.40 n.a. 2.58 -- -- -- Malawi Blantyre 0.12 n.a. 1.16 3.38 n.a. n.a. Namibia Windhoek 1.45 n.a. n.a. n.a. n.a. n.a. Sudan Great Khartoum 0.37 n.a. 1.15 -- 4.32 3.00 Zambia Lusaka 0.56 n.a. 1.67 -- n.a. 3.00 Cape Verde Praia 2.67 n.a. 9.44 n.a. 9.67 11.38 Chad N'Djamena 0.22 -- -- -- n.a. -- Côte d'Ivoire Abidjan 0.04 -- 0.93 1.82 n.a. 3.35 Madagascar Antananarivo 0.11 0.47 1.24 -- n.a. 2.33 Tanzania Dar es Salaam 0.39 -- 0.87 0.98 2.40 2.56 Uganda Kampala 0.25 n.a. 1.40 1.40 -- 4.50 Average 0.49 1.04 1.93 1.63 4.67 4.00 Median 0.37 0.79 1.24 1.49 4.08 3.00 Minimum 0.04 0.47 0.48 0.98 2.40 1.67 Maximum 2.67 2.11 9.44 3.38 9.67 11.38 Source: Keener, Luengo, and Banerjee 2009. Note: n.a = not applicable, -- = not available. Cost Recovery, Affordability, and Subsidies 167 offered by the utilities to connected households; this further underscores the volatility and inequity in the market structure. Households with private connections or yard taps face water prices significantly lower than those dependent on piped water delivered through public standpipes and the informal market. The prices for each water provider in the informal sector also show higher variability than those offered by the utilities to connected households. This applies to alternative providers in different cities (the standard deviation of the prices for each informal water service is 1.3 to 5 times higher than for the household connection), as well as for different neighborhoods within the same city (figure 6.3). Cape Verde's prices for formal and informal water services are highest because of the specifications of its water pro- duction system. When formal household connections to piped water are not available or the retail public standpipe price varies from the official price, utilities lose potential revenue from unserved or underserved customers. For the cities studied, the ratio between informal to formal standpipe prices goes from 0.9 in Ouagadougou, to 20.4 in Kinshasa, with a median ratio of 3. High retail prices and the size of the population coverage by standpipes Figure 6.3 Price by Water Service Provider 12 10 8 US$/m3 6 4 2 0 HH small piped standpipe water water connection network HH reseller or kiosk vendor tanker average 0.39 1.04 1.63 1.52 3.38 4.67 max 1.45 2.11 3.38 5.51 6.89 9.67 min 0.04 0.47 0.98 0.48 1.67 2.40 Source: Keener, Luengo, and Banerjee 2009. Note: The average prices are presented. Cape Verde is excluded from this graph because it is an outlier. HH = household. 168 Africa's Water and Sanitation Infrastructure combine to create an economic environment in which estimates of the total gross profit1 captured by standpipe operators ranged from $15,477 in Khartoum to almost $10 million in Lusaka.2 These amounts can represent a significant percentage of formal utilities' revenue: in Maputo, 12 percent; Addis Ababa, 44 percent; and Lusaka, 120 percent. Thus, although standpipes are already heavily subsidized by utilities, none of this subsidy reaches the final consumers. Two-Part Tariffs and the Small Consumer The tariff at an average consumption level of 10 m3 is about $0.49/m3 in Africa. However, tariffs at ELECTRA, in Cape Verde, exceed $3 for that consumption level because of the expense of desalination, which raises the cost of water production. If Cape Verde is excluded from the conti- nental figure, the average tariff is $0.43/m3. The tariff levels in Africa are comparable to the average in Latin America and the Caribbean, which at $0.41/m3 at an average consumption of 15 m3 is higher than other regions in the world, such as East Asia, Eastern Europe, and the Middle East. South Asian water tariffs are the world's lowest, with an observed average tariff of only $0.09/m3 (table 6.3). The implementation of the increasing block tariff (IBT) structure is based on the implicit assumption that small consumers are poor and large consumers will cross-subsidize the small ones. To investigate whether small consumers pay lower prices than large consumers, the water price Table 6.3 Comparison of Water Tariffs in Africa and Other Global Regions at Various Levels of Consumption ($/m3) Consumption level 4 m3 10 m3 15 m3 40 m3 Average 0.55 0.49 0.52 0.65 Median 0.41 0.38 0.40 0.51 Comparable tariffs (average consumption = 15 m3) Organisation for Economic Co-operation and Development 1.04 Latin America and the Caribbean 0.41 Middle East and North Africa 0.37 East Asia and Pacific 0.25 Europe and Central Asia 0.13 South Asia 0.09 Sources: Banerjee, Foster, and others 2008; Foster and Yepes 2006. Cost Recovery, Affordability, and Subsidies 169 per cubic meter for three consumption levels--4, 10, and 40 m3/month-- is calculated. The effective price sharply declines at the average con- sumption of 10 m3, and then rises again. The price at the subsistence consumption rate of 4 m3 is roughly comparable to the price at 20 m3 of consumption (figure 6.4). The two-part IBT tariff structure can fail to favor small consumers for two reasons. First, the fixed-fee and minimum-consumption charges place an enormous burden on low-volume consumers. This is the part of the water bill the households cannot control regardless of their level of consumption. Komives and others (2005) compare the average price per m3 of IBT, IBT with fixed-fee, and IBT with fixed-fee and minimum- consumption charges. They find that low-volume consumers under the two-tariff regimes bear the burden of higher prices. Small consumers pay the lowest prices in only a few countries in Africa. Among the 45 utilities in the sample, the effective price increases with rising consumption in 27 utilities. In the majority of utilities, high-end consumers pay more than low-end or average consumers. Inequity is more prevalent, however, at the lower end of consumption, among households consuming 4 to 10 m3/month. In 16 utilities, the effective tariffs of small consumers are higher than those of average consumers. This difference is pronounced in the case of five utilities in Mozambique. Because these utilities have a minimum threshold of 10 m3, the small consumer whose water intake is about 4 m3 pays on average about $0.57 more than those consuming 10 m3 and about $0.40 more than those consuming 40 m3 (figure 6.5). Figure 6.4 Average Water Tariffs for Africa at Different Consumption Levels 0.65 0.60 0.55 0.50 US$/m3 0.45 0.40 0.35 0.30 0.25 0.20 4 m3 10 m3 15 m3 20 m3 30 m3 40 m3 50 m3 consumption per month average median Source: Banerjee, Foster, and others 2008. Note: Cape Verde is not included in this graph because it is an outlier. 170 Africa's Water and Sanitation Infrastructure Figure 6.5 Utilities Charging Higher Effective Prices to Small Consumers 80 61 60 % utilities 40 36 31 33 32 20 7 0 ratio of price at 4 m3 ratio of price at 4 m3 to price at 10 m3 to price at 40 m3 greater than 1 equal to 1 less than 1 Source: Banerjee, Foster, and others 2008. Second, the arrangement of the block's size and price is important, particularly that of the first block. If the first block is wide, it allows leak- age of the implicit subsidy to the nonpoor and leads to a higher price per m3 for the low-volume consumers in the band. Fixed and minimum consumption charges have a significant impact on the unit price paid by small consumers. With a fixed charge, small consumers usually have to pay a higher price per unit than large consumers. For utilities that impose a fixed-fee or minimum consumption charge, the average price at 4 m3 is $0.64/m3, as opposed to $0.47/m3 for those who do not. The size of the first block can also impact the price paid by small consumers. Generally speaking, the larger the size of the first block in an IBT structure, the higher the probability that subsidies for the low price of the first block will leak to large consumers. Of the 45 utilities in the sample, only nine have a tariff design with a first block that rises above 10 m3 (the rest have a flat or linear structure). This effect, though important, is overwhelmed by the fixed-fee and minimum consumption charges, which can erase the block-tariff structure's positive impact on small consumers. The subsidy to the low block under the current IBT structure does not benefit small consumers (usually the poor) exclusively; instead, a large amount of the subsidy leaks to large consumers (usually the nonpoor). Further, the fixed and minimum consumption charges and the large size of the low blocks often cause small consumers to pay higher effective prices per unit than large consumers. Cost Recovery, Affordability, and Subsidies 171 Paying for Water: How Common? The discussion so far has focused on formal utility customers who report paying a utility bill. But to focus only on this category of users is to miss a substantial part of the African story. Household surveys provide unique insights into two other key categories of consumers. First, there are those who do not have their own household connection to piped water but nonetheless register expenditure because they are accessing the network through some secondary source, usually a neighbor's tap. Second, there are those who do have a household connection to piped water but do not register any expenditure, whether because they are in arrears or because the connection itself is clandestine. About 61 percent of the African population is not connected to and does not pay for formal water services (figure 6.6). The traditional cus- tomers who connect and pay are actually a minority of those who use the service; the population that connects but does not pay is almost as large as the percentage that connects and pays. Moreover, for access to household connections to piped water, the population that is uncon- nected but nevertheless pays to obtain the service through secondary sources is slightly higher than the one that connects and pays for pro- prietary service. Figure 6.6 Connection and Payment, by Consumer Categories 70 60 50 % consumers 40 30 20 10 0 category connected and pay connected and do not pay unconnected and pay unconnected and do not pay Source: Banerjee, Wodon, and others 2008. 172 Africa's Water and Sanitation Infrastructure Overall, an estimated 12 percent of those who have household con- nections to piped water do not appear to be paying for them in any given month. Nonpayment rates in excess of 65 percent can be found in 30 percent of customers with household connections to piped water (figure 6.7, panel a). The extent to which nonpayment is higher among the poorest can be seen as an indicator that households are facing affordability problems. In the first quintile, the nonpayment ratio amounts to approximately 63 percent of households, and this ratio declines steadily to 26 percent of households in the fifth quintile (figure 6.7, panel b). This pattern indi- cates that nonpayment, to some extent, does represent an affordability issue, given the decline as household budgets rise across the distribution. Nevertheless, the existence of a significant nonpayment rate, even among Figure 6.7 Nonpayment Rates of Water Services a. Nonpayment rate 50 40 % countries 30 20 10 0 piped water up to15% 15­35% 35­65% over 65% b. Nonpayment rate by quintile 100 80 % not paying 60 40 20 0 Q1 Q2 Q3 Q4 Q5 quintile Source: Banerjee, Wodon, and others 2008. Note: Q = quintile. Cost Recovery, Affordability, and Subsidies 173 the richest quintiles, suggests that problems of payment culture also exist. Moreover, given that the majority of connected households are in the richer quintiles, in absolute terms the largest number of nonpaying cus- tomers also comes from the richer quintiles (even though the nonpay- ment ratio for this group is comparatively low). Recovering Operating Costs: Affordable As utilities move toward commercial entities, it becomes essential to establish demand for services. Affordability is typically measured by the share of infrastructure spending in the total household budget and whether it exceeds a set threshold (Fankhauser and Tepic 2005). There is no absolutely scientific basis for determining the value of such afford- ability thresholds; however, based on experience with actual household expenditure patterns and results of willingness to pay surveys, certain thresholds have come to be widely used by practitioners. The World Health Organization, for example, uses a 5 percent affordability thresh- old for water and sanitation services in developing countries. The evi- dence presented on current expenditure patterns earlier suggests that households spend 2 to 5 percent on water services. In the discussion that follows, 5 percent is used as a reference affordability threshold. To estimate the percentage of African households likely to face afford- ability problems for modern infrastructure services, two elements are needed. First, indicative values of the true cost of infrastructure services are needed as a reference point. The absolute cost of the total monthly bill can be computed based on different assumptions about subsistence household consumption and the tariff applied. For piped-water service, subsistence consumption ranges between 4 m3 per month (based on an absolute minimum consumption of 25 liters per capita per day for a fam- ily of five) and 10 m3 per month (based on a somewhat more comfort- able but still modest level of 60 liters per capita per day for a family of five). The indicative tariff ranges from $0.40/m3 to $0.80/m3, depending on whether the goal is operating or full capital cost recovery. The lower- bound monthly bill is about $2, and the upper-bound monthly bill is about $8 for household connections to piped water (table 6.4). Second, the survey data on budget expenditures are used to estimate what percentage of households would hit the 5 percent affordability thresholds at different levels of absolute expenditure. For example, a house- hold with a monthly budget of $100 would hit the affordability threshold of 5 percent of income once any service cost more than $5 per month. 174 Africa's Water and Sanitation Infrastructure Table 6.4 Reference Points for the True Cost of Infrastructure Services Piped water Reference Lower bound Subsistence household consumption 4 m3 Tariff (operating cost recovery) $/m3 $0.40/m3 Total monthly bill ($) $2 Upper bound Subsistence household consumption 10 m3 Tariff (capital cost recovery) $/m3 $0.80/m3 Total monthly bill ($) $8 Source: Banerjee, Wodon, and others 2008. By pooling all African households across countries and grouping them into a common set of quintiles based on purchasing power parity adjust- ments to their budgets, it is possible to report results for the continent as a whole. Figure 6.8 plots the share of budget required to meet increasing levels of spending on infrastructure services for the average household in each of the continental income quintiles. The average household in the first quintile hits the 5 percent afford- ability threshold at close to $4 per month, which is more than enough to pay for the subsistence minimum consumption of piped water. The aver- age household in the second quintile hits the 5 percent affordability threshold at close to $8 per month and would be able to pay for the upper bound of piped water. Households in the third, fourth, and fifth quintiles do not face any affordability constraints within the range of ser- vice baskets considered here. Very modest consumption baskets priced at levels compatible with operating cost recovery appear to be affordable across the full range of household budgets in Africa. Nevertheless, an estimated 60 percent of the African population cannot afford to pay full cost-recovery tariffs or extend consumption beyond the absolute minimum subsistence level. These continental results mask a great deal of variation across individ- ual countries because almost all the households in the poorer countries may be in the bottom quintile for Africa as a whole, whereas almost all the households in the more affluent countries may be in the uppermost quintile for Africa as a whole. Table 6.5 provides a similar type of analy- sis at the country level to calculate the percentage of households in each country that would fall beyond the 5 percent affordability threshold at any particular absolute monthly cost of service. The countries divide into three groups. At one extreme is group 1, in which a majority of urban households can afford a monthly expenditure of $8, and often considerably more. At the other extreme is group 3, in Figure 6.8 Share of Average Urban Household Budget Required to Purchase Subsistence Amounts of Piped Water, by Continental Income Quintiles 10 9 8 % household budget 7 6 5 4 3 2 1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 cost of subsistence consumption (US$/month) Q1 Q2 Q3 Q4 Q5 Source: Banerjee, Wodon, and others 2008. Note: Q = quintile. 175 176 Africa's Water and Sanitation Infrastructure Table 6.5 Share of Urban Households Whose Utility Bill Would Exceed 5 Percent of the Monthly Household Budget at Various Prices Monthly bill ($) Group 2 4 6 8 10 12 14 16 1 Cape Verde 0 0 0 0 0 0 0 0 Morocco 0 0 0 0 0 0 0 0 Senegal 0 0 0 0 0 0 1 1 South Africa 0 0 0 0 1 1 1 1 Cameroon 0 0 0 0 1 2 7 17 Côte d'Ivoire 0 0 1 2 3 5 7 10 Congo, Rep. 0 0 3 5 12 21 28 35 2 Ghana 0 2 7 11 30 46 55 67 Benin 0 2 4 12 33 45 60 71 Kenya 0 0 5 20 36 62 72 78 Sierra Leone 0 4 16 30 44 54 62 67 São Tomé and 0 2 13 29 46 64 77 81 Principe Burkina Faso 0 4 20 34 47 62 72 78 Zambia 0 4 18 35 50 58 67 76 Nigeria 3 10 23 35 57 78 89 95 Madagascar 0 16 28 47 61 68 78 85 3 Niger 1 11 28 55 70 79 89 93 Tanzania 1 8 25 55 75 89 96 98 Guinea-Bissau 0 6 38 65 81 89 91 93 Uganda 2 17 45 65 82 90 96 97 Burundi 7 29 53 72 82 90 97 100 Malawi 2 32 66 78 87 92 93 94 Congo, Dem. Rep. 9 49 79 91 98 99 100 100 Ethiopia 40 87 95 99 99 99 99 100 Summary Low-income 5.0 18.4 32.4 44.5 59.5 72.3 79.7 84.3 Middle-income 0.0 0.0 0.1 0.2 1.2 1.8 2.9 4.7 All 3.7 13.7 24.2 33.2 44.7 54.3 60.2 64.1 Source: Banerjee, Wodon, and others 2008. which at least 70 percent, and in some cases more than 90 percent, of urban households would be unable to afford a monthly expenditure of $8 for water. All the remaining countries fall into group 2, in which a substantial share of the urban population--between one- and two- thirds--would face difficulties covering an upper-bound monthly expenditure. The High Cost of Connecting to Water and Sanitation Services Network connection costs can prove to be a significant barrier to con- sumer access in Africa. The connection charges vary widely, from about Cost Recovery, Affordability, and Subsidies 177 $6 in the Upper Nile in Sudan to more than $240 in Côte d'Ivoire, Mozambique, and Niger,3 and more than $300 in Drakenstein, eThekwini, and Johannesburg, South Africa. Connection costs can vary even among water utilities in the same country. For instance, Addis Ababa Water Services Authority (AWSA), Nazareth Water Company (ADAMA), and Dire Dawa--three utilities in Ethiopia--charge connection costs of $14, $9, and $43, respectively. A comparison with the gross national income (GNI) per capita suggests that, in some countries, the connection charge is relatively expensive. On average across Africa, the connection charge is 28 percent of the GNI per capita. In middle-income countries such as South Africa and Namibia, though the connection cost is high, it is negligible compared with GNI per capita, but in countries such as Niger, the connec- tion charge is more than 100 percent of the GNI per capita (figure 6.9). Similarly, for sanitation, the capital costs associated with infrastructure facilities can be considered prohibitive when compared with the limited budgets. For instance, standardized unit costs drawn from the Senegal san- itation sector can be employed to estimate the percentage of households' monthly budget that would be absorbed by the upfront investment cost associated with different types of sanitation facilities (table 6.6). The results indicate that although traditional latrines look quite affordable across the income spectrum in Senegal, improved latrines represent more than a month of the household budget even for households in the high- est income group. These findings are borne out by the patterns of access to sanitation already observed across the socioeconomic spectrum. Half of Sub-Saharan African households have invested in traditional latrines in Figure 6.9 Formal Water Connection Cost a. Frequency distribution of connection cost 50 across utilities 45 40 35 % utilities 30 25 20 15 10 5 0 < 20 US$ > 20 US$ and < 100 US$ > 100 US$ and < 200 US$ > 200 US$ (continued next page) 178 Africa's Water and Sanitation Infrastructure Figure 6.9 (continued) b. Comparison of connection cost with GNI per capita SEEN AdeM Quelimane AdeM Pemba AdeM Nampula Adem Maputo ADeM Beira ELECTROGAZ ONEA CRWB FCT WB SONEB NWC South Darfur SODECI Dire Dawa SDE WASA NWC Khartoum KIWASCO NWSC LWSC AWSA Katsina WB Windhoek Johannesburg eThekwini Drakenstein SWSC NWASCO DAWASCO ADAMA DUWS MWSA Tygerberg Kaduna ELECTRA NWC Upper Nile Oshakati 0 10 20 30 40 50 60 70 80 90 100 110 connection cost as % of GNI per capita Source: Banerjee, Foster, and others 2008. Note: ADAMA = Nazareth Water Company; AWSA = Addis Ababa Water Services Authority; CRWB = Central Region Water Board; DAWASCO = Dar es Salaam Water and Sewerage Company; DUWS = Dodoma Urban Water and Sewerage Authority; FCT WB = Federal Capital Territory Water Board; GNI = gross national income; KIWASCO = Kisumu Water and Sewerage Company; LWSC = Lusaka Water and Sewerage Company; MWSA = Mwanza Water and Sewerage Authority; NWASCO = Nairobi Water and Sanitation Company; NWC = National Water Company; NWSC = National Water and Sewerage Company, Uganda; ONEA = Office Nationale des Eaux et d'Assainissement; SDE = Sénégalaise des Eaux; SEEN = Société de Exploitation des Eaux du Niger; SODECI = Société de Distribution d'Eau de Côte d'Ivoire; SONEB = Société Nationale des Eaux du Benin; SWSC = Southern Water and Sewerage Company; WASA = Water and Sanitation Authority; WB = Water Board. Cost Recovery, Affordability, and Subsidies 179 Table 6.6 Cost of Facility as Percentage of Monthly Household Budget in Senegal National Rural Urban Q1 Q2 Q3 Q4 Q5 Total monthly household budget in Senegal (2002 US$) 227 154 315 102 134 166 225 394 Cost of facility as percentage of monthly household budget Septic tank 289 427 209 641 491 396 292 167 Improved latrine 194 286 140 430 330 266 196 112 Traditional latrine 22 32 16 48 37 30 22 13 Source: Morella, Foster, and Banerjee 2008. Note: Q = quintile. the absence of any far-reaching subsidization policy; this corroborates other evidence that investments of this size are affordable across the income spectrum. At the same time, the fact that improved latrines are confined to upper-income groups bears out the high budget shares that families would need to finance an improved latrine. The Cost of Subsidizing Capital and Operating Expenses The affordability of infrastructure services needs to be considered not only at the household level, but also at the level of the public finances of each country. To the extent that households cannot afford to pay cost-recovery tariffs, the move toward universal access will create bur- geoning liabilities for the state, which must bridge the gap between the tariffs the public can afford to pay and the real cost of service provi- sion. This analytical framework also can be used to estimate the aggre- gate value of these subsidies in each country, which helps to assess whether subsidizing services to reach universal coverage is an afford- able strategy at the country level. Once again, no absolute scientific method can determine the affordability threshold at the country level; nevertheless, it is possible to get a sense of when costs reach a level that is manifestly unattainable. A one-time, finite capital subsidy of $200 per unserved household, designed to cover the costs of connection of these households over a 10- year period, will cost approximately 1 percent of the annual African gross domestic product (GDP). An estimated 60 percent of the countries would face costs in excess of 1 percent of GDP. The cost would exceed 2 percent of GDP in Ethiopia, Malawi, the Democratic Republic of Congo, the Republic of Congo, and Sudan. The highest burden on fiscal 180 Africa's Water and Sanitation Infrastructure Figure 6.10 Subsidy Needed to Maintain Affordability of Water Services 45 41 40 35 34 30 28 % countries 25 25 25 20 19 16 15 13 10 5 0 capital subsidy operating subsidy < 0.5% of GDP > 0.5% and < 1.0% of GDP > 1% and < 2% of GDP > 2% of GDP Source: Banerjee, Wodon, and others 2008. resources would be for the Democratic Republic of Congo, which must spend a projected 18 percent of its GDP on household connections to piped water. In more affluent countries, such as Gabon, the cost of this policy would amount to no more than 0.02 percent of the GDP. An indefinite, ongoing operating subsidy of $2 per month to ensure that currently unserved customers can continue to afford service once con- nected places similar strains on the government budget. For 40 percent of the countries, providing a monthly subsidy of $2 for water would amount to spending 1 to 2 percent of GDP. For 16 percent of the countries, it will be more than 2 percent of GDP. The highest burden would be on the Democratic Republic of Congo, followed by Ethiopia, Malawi, Niger, and Sudan, which would have to spend more than 2 percent to maintain a sustainable consumer base for water services. Like the capital subsidy, this operating subsidy would consume 1.1 percent of the African GDP (figure 6.10). Poor Targeting of Utility Subsidies Customers receive substantial subsidies in most African countries, because residential water tariffs tend to be below utility costs. The work- ing assumption is that the price per m3 in the highest bracket of con- sumption in the tariff schedule can be used as a first approximation of Cost Recovery, Affordability, and Subsidies 181 the cost of providing the service. (Actually, the estimates of targeting performance are not very sensitive to that assumption.) As shown by Komives and others (2005), a simple framework can be used not only to analyze the targeting performance of water subsidies in about 20 African countries for which data are available, but also to understand what affects targeting performance through so-called access (who uses water) and subsidy design factors (who benefits from subsidies and by how much among users). The targeting performance indicator used in the analysis, denoted by (omega), is simply the share of the subsidies received by the poor divided by the proportion of the population in poverty. In other words, a value of one for implies that the subsidy distribution among the poor is proportional to their share in the overall population. If the poor account for 30 percent of the population, then a neutral targeting mechanism would allocate 30 percent of the subsidy to the poor. A value (lower) greater than one implies that the subsidy distribution is (regressive) progressive, since the share of benefits allocated to the poor is (lower) larger than their share in the total population. For instance, suppose that 30 percent of the population is poor and obtains 60 percent of the subsidy benefits. In such a case, would equal two, meaning that the poor were receiving twice as much subsidy as the population on average. Utility subsidies tend to be very poorly targeted. As shown in figure 6.11, in none of the countries is the targeting indicator superior to one; it is often well below one. Although comparability issues are found among countries, on average the poor are benefiting only from one- fourth to one-third of what a household randomly selected in the popu- lation would get. The targeting performance indicator can be deconstructed into "access" and "subsidy design" factors4 to allow analysis of why subsidies are targeted as they are. Access factors are those related to the availabil- ity of water service in the area in which a household is located and to the household's decision to connect to the network when service is available. These access factors have a strong influence on targeting performance but are usually difficult to change in the short run. Policy design is more sus- ceptible to subsidy factors, such as tariff structure changes that affect who is targeted to receive the subsidies. Policy design also is affected by rates of subsidization and the quantities of water consumed by the households that benefit from the subsidies. Investigations reveal that most water sub- sidy mechanisms are poorly targeted, essentially because most of the poor 182 Africa's Water and Sanitation Infrastructure Figure 6.11 Overall Targeting Performance () of Utility Subsidies Burkina Faso 0.02 Burundi 0.15 Cameroon 0.30 Cape Verde 0.24 Central African Republic 0.66 Chad 0.26 Congo, Dem. Rep. 0.54 Congo, Rep. 0.43 Côte d'Ivoire 0.28 Gabon 0.64 water Ghana 0.12 Guinea 0.12 Malawi (Blantyre) 0.15 Malawi (Lilongwe) 0.04 Niger 0.27 Nigeria (FCT) 0.36 Nigeria (Kaduna) 0.53 Rwanda 0.01 Senegal 0.77 Togo 0.49 Uganda 0.07 0 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 Omega Source: Banerjee, Wodon, and others 2008. Note: FCT = Federal Capital Territory. lack access to the water network and, therefore, cannot benefit from water subsidies, but also because the existing tariff structures are not designed to target subsidies to the poor. This can be seen clearly in figure 6.12, which deconstructs the value of the targeting indicator into access and subsidy design factors. The curves added to the graphs represent combinations of access and subsidy design factor values that result in the same value for . The closer a country is located to the upper right of the graphs, the better the target- ing performance, because again is the product of the access and sub- sidy design factors. Cost Recovery, Affordability, and Subsidies 183 Figure 6.12 Access Factors and Subsidy Design Factors Affecting Targeting Performance 1.4 1.2 Gabon Nigeria (Kaduna) 1.0 Burkina Faso Togo Congo, Dem. Rep. Senegal subsidy design factors Chad Congo, Rep. Niger Central African Republic 0.8 Ghana Malawi (Blantyre) Cameroon Guinea Côte d'Ivoire Rwanda Burundi 0.6 Cape Verde Nigeria (FCT) Malawi (Lilongwe) Uganda 0.4 0.2 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 access factors Source: Banerjee, Wodon, and others 2008. Note: See text for an explanation of the different curves. The two variables used to compute the access factors are, first, whether a household is located in an area served by the water network, and, sec- ond, whether a household in such an area is actually connected to and get- ting service from the network. The value of the access factors is simply the rate of connection to the network among the poor (which depends on access and uptake when there is access) divided by the rate of the connec- tion in the population as a whole. As expected, the access factors are much lower than one for all countries, simply because on average the poor have much lower connection rates than the population as a whole. Subsidy design factors, which take into account who benefits from subsidies among households connected to the network and how large the subsidies are, make up the second variable affecting the value of the tar- geting parameter. The subsidy design factor represents the ratio of the average benefit from the subsidy among all poor households that are con- nected to the network, divided by the average benefit among all house- holds connected to the network, whether poor or nonpoor. Surprisingly, in many countries the subsidy design factors are also below unity, thereby limiting targeting performance. The main explanation is that although the 184 Africa's Water and Sanitation Infrastructure rate of subsidization of the poor (that is, the discount versus the full cost of providing water for the utility) is often larger than for the population as a whole that is connected to the network, the quantities consumed by the population as a whole tend to be larger than those consumed by the poor, so that the overall subsidy received by the poor is lower on average than that received by the population as a whole. Consumption subsidies for water appear to be poorly targeted in African countries for several reasons. Access factors are important in determining the potential beneficiaries of consumption subsidies. Poor households tend to live in areas without water service, and so it is impos- sible for them to benefit from the subsidies. Even when they live in an area that offers potential access to the network, many among the poor remain unconnected to the networks because they live too far from the water pipes or the cost of connecting to the network and purchasing the equipment required to use water is too high. Good subsidy design mech- anisms would allow countries to compensate for the negative impact of access factors on targeting performance. Unfortunately, the traditional IBT structures that prevail in many countries tend to be poorly targeted. They spread subsidies to all households connected to the network; even those that consume high amounts of water benefit from a subsidy for the part of their consumption that belongs to the lower level blocks of the tariff structure. In addition, the lower blocks often are too high in terms of consumption to target the poor well. Finally, significant differences in unit prices may not be present among the various blocks. Connection Subsidies as a Viable Alternative One possible alternative is to provide connection rather than consump- tion subsidies, assuming that the generation or production capacity is sufficient to expand the network. Figure 6.13 provides the potential tar- geting performance of connection subsidies under the three scenarios. First, we assume that connection subsidies will be distributed in the same way as existing connections. This is a pessimistic assumption from a distributional point of view because it tends to favor better-off house- holds, but it could be realistic if access rates to the network are low. Second, we assume that new connections could be distributed randomly among households that currently are not connected but are located in a neighborhood where connections are available. Third, we assume that new connection subsidies could be randomly distributed among all households that do not currently have access. This is a very optimistic Cost Recovery, Affordability, and Subsidies 185 Figure 6.13 Potential Targeting Performance of Connection Subsidies under Various Scenarios Burkina Faso Burundi Cameroon Cape Verde Central African Republic Chad Congo, Dem. Rep. Congo, Rep. Côte d'Ivoire Gabon water Ghana Guinea Malawi (Blantyre) Malawi (Lilongwe) Niger Nigeria (FCT) Nigeria (Kaduna) Rwanda Senegal Togo Uganda 0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 Scenario 3: Distribution of connection subsidies mirrors distribution of existing connections Scenario 2: Only households with access but no connection receive subsidy Scenario 1: All unconnected households receive subsidy Source: Banerjee, Wodon, and others 2008. Note: FCT = Federal Capital Territory. 186 Africa's Water and Sanitation Infrastructure assumption given that many of these households are not located in neigh- borhoods where access is available. The value of is largest under the assumption that new connections benefit households that are selected randomly from the population with- out access. In all countries, is larger than one under this assumption. Yet, the assumption is not realistic. The second scenario assumes that house- holds that benefit from new connections are selected from unserved households located in areas where there is already access to the network. The values of , although often lower than one, are still much better than those for consumption subsidies. In the third scenario, targeting perform- ance remains poor. Thus, if connection subsidies could be designed to reach the majority of households not connected today but living in areas where service is provided, the targeting performance of those subsidies would be better than that of consumption subsidies. In addition, connec- tion subsidies help to reduce the cost of service for users (compared with street vendors for water, for example) and bring positive externalities in areas such as education and health. Finally, it is often argued that any removal of utility subsidies would be detrimental. Again, the household survey evidence provides an opportu- nity to test this hypothesis. In most countries, water spending represents only a tiny fraction of total consumption for the population as a whole. Among households connected to the network and consuming water, the fraction is much higher, typically 3 to 5 percent. This, in turn, is directly related to the impact of a proportional increase in water tariffs on poverty. For simplicity, relative poverty measures can be used: The poverty line in each country is set at half the mean level of per capita con- sumption. In many countries, the impact of a 50 percent increase in tar- iffs or even of a doubling of the tariffs is truly marginal at the national level, with estimates of the shares of the population living in poverty changing by barely one-tenth of a percentage point. Among households with a connection to the network, the impact is larger, but still fairly lim- ited. There is rarely an increase in the share of households in poverty larger than one or two percentage points, and because the households that benefit from a connection tend not to be poor compared with other households, the increase in poverty starts from a very low base. Thus, in general, it can be said that the impact on poverty of an increase in tariffs is small in most cases. This does not mean that such a poverty impact does not have negative consequences on those hit by it. It does mean, however, that if subsidies were reduced, and the funds were used in a different, more pro-poor way, there would be a potentially sub- stantial gain for poverty reduction. Cost Recovery, Affordability, and Subsidies 187 Annex 6.1 Methodology for Estimating the Annual Gross Profit and the Annual Cross-Subsidy between Household Consumers and Standpipes Captured by Standpipe Operators in a City The following figure shows the prices charged by the utility to the stand- pipe operators (formal or official standpipe price), to a household with a private connection, and by the standpipe operator to the consumers (informal standpipe tariff). We define unitary standpipe operator gross profit, unitary cross-subsidy between consumers with a household connection and standpipe operators in the following way: Unitary standpipe operator gross profit (PG) ($/m3) = informal standpipe price ($/m3) ­ formal standpipe price ($/m3) Unitary cross-subsidy household (HH) connection-standpipe operator (SHH-Stdp) ($/m3) = HH consumer price ($/m3) ­ formal standpipe price ($/m3). water price ($/m3) informal standpipe tariff unitary gross profit household with private connection tariff subsidy unitary cross- formal standpipe tariff Source: Luengo, Keener, and Banerjee 2008.. Because households with private connections are assessed tariffs based on consumption levels, we have to define a common level of con- sumption to compare tariff structures across countries. The reference we use for this in an average consumption level of 60 liters per capita per day for people with a household private connection (Water Utility Partnership 2002). When analyzing the cross-subsidies between small and large consumers, one interesting finding is that the fixed-fee and minimum-consumption charge means an economic burden on low- volume consumers with a household connection. Although the increas- ing block tariff is commonplace in African countries, the two-part 188 Africa's Water and Sanitation Infrastructure tariff structure can fail to lead to a price that favors small consumers (Banerjee, Wodon, and others 2008). Except in a few countries, among those who have a household connection, average consumers (60 liters per capita per day), not small consumers (25 liters per capita per day) pay the lowest price. In that sense, the 60 liters per capita per day ref- erence can help us to define the lower boundary (and a better esti- mate) of the cross-subsidy between consumers with a household connection and standpipe operators. To estimate the annual gross profit of the standpipe operators and the annual cross-subsidy between the consumer with a household connection and the standpipe operator, we use the following formulation: Annual gross profit of standpipe operators ($/year) = PG × U × 365 (days/year) × 1,000 (liters/m3) × P × C, where PG ($/m3): Unitary standpipe operator gross profit U (liters per capita per day): Standpipe unit consumption; based on the AICD data, it is fixed at 25 liters per capita per day P (#): City population C (%): Coverage of the water service by standpipes Annual cross-subsidy between the consumer with a household connection and the standpipe operator ($/year) = SHH-Stdp × U × 365 (days/year) × 1,000 (liters/m3) × P × C, where SHH-Stdp ($/m3): Unitary cross-subsidy between household consumer- standpipe operator U (liters per capita per day): Standpipe unit consumption; based on the AICD data, it is fixed at 25 liters per capita per day P (#): City population C (%): Coverage of the water service by standpipes. Notes 1. Gross profit = revenue from water sales ­ cost of water sales. This calculation does not include operation and maintenance costs, other overhead costs, taxes, and financial costs. 2. See annex 6.1 for the calculation methodology. 3. Based on 26 utilities for which information on connection charges were available. 4. = (access factors) (subsidy design factors). Cost Recovery, Affordability, and Subsidies 189 References Banerjee, S., V. Foster, Y. Ying, H. Skilling, and Q. Wodon. 2008. "Cost Recovery, Equity and Efficiency in Water Tariffs: Evidence from African Utilities." AICD Working Paper 7, World Bank, Washington, DC. Banerjee, S., Q. Wodon, A. Diallo, N. Pushak, H. Uddin, C. Tsimpo, and V. Foster. 2008. "Access, Affordability and Alternatives: Modern Infrastructure Services in Sub-Saharan Africa." AICD Background Paper 2, World Bank, Washington, DC. Foster, V., and T. Yepes. 2006. "Is Cost Recovery a Feasible Objective for Water and Electricity? The Latin American Experience." Policy Research Working Paper 3943, World Bank, Washington, DC. Fankhauser, S., and S. Tepic. 2005. "Can Poor Consumers Pay for Energy and Water? An Affordability Analysis for Transition Countries." Working Paper 92, European Bank for Reconstruction and Development, London. Keener, S., S. G. Banerjee, N. Junge, and G. Revell. Forthcoming. "Informal Water Service Providers and Public Stand Posts in Africa." Africa Post-Conflict and Social Development Department, World Bank, Washington, DC. Keener, S., M. Luengo, and S. G. Banerjee. 2009. "Provision of Water to the Poor in Africa: Experience with Water Standposts and the Informal Water Sector." AICD Working Paper 13, World Bank, Washington, DC. Komives, K., V. Foster, J. Halpern, and Q. Wodon. 2005. "Water, Electricity, and the Poor: Who Benefits from Utility Subsidies?" Water and Sanitation Unit, World Bank, Washington, DC. Morella, E., V. Foster, and S. Banerjee. 2008. "Climbing the Ladder: The State of Sanitation in Sub-Saharan Africa." AICD Background Paper 13, World Bank, Washington, DC. Water Utility Partnership. 2002. "Final Project Summary Report: Service Providers' Performance Indicators and Benchmarking Network Project." Abidjan. CHAPTER 7 Spending Needed to Meet Goals in Water and Sanitation The Millennium Development Goal (MDG) for sustainable access to safe drinking water and improved sanitation presents an enormous financing challenge, particularly to many low-income countries. This chapter focuses on the levels of investments required to meet the water and sanitation MDG, assuming that access patterns remain broadly the same during the period from 2006 to 2015. The analysis presented here takes into account population growth and estimates the investment needed to expand access, rehabilitate existing assets, and ensure adequate maintenance. The Challenge of Expanding Coverage The progress toward the MDG for sustainable access to safe drinking water and basic sanitation has been made mostly in the water space as of 2006. Twenty-six countries are on track to meet the water MDG. At one end stand Niger, Equatorial Guinea, Nigeria, Mozambique, Sierra Leone, and the Democratic Republic of Congo, which show coverage rates of more than 25 percentage points below the MDG targets (figure 7.1). At the other end, five countries had already reached the target as of 2006. Among these, two are middle-income countries: Namibia and South Africa. The rest are low-income countries: Burkina Faso; Malawi, 191 192 Africa's Water and Sanitation Infrastructure Figure 7.1 Water MDG Gap, 2006 Niger Equatorial Guinea Nigeria Mozambique Sierra Leone Congo, Dem. Rep. Madagascar Tanzania Swaziland Angola Rwanda Zambia Benin Togo Liberia Ethiopia Chad Burundi Kenya Central African Republic Sudan Eritrea Lesotho Mauritania Zimbabwe Uganda Senegal Mali Gambia, The Gabon Cameroon Guinea Côte d'Ivoire Botswana Mauritius Ghana South Africa Burkina Faso Malawi Namibia ­20 ­15 ­10 -5 0 5 10 15 20 25 30 35 percentage point difference between MDG target and current access Source: JMP 2006. Spending Needed to Meet Goals in Water and Sanitation 193 where improved water coverage doubled from 1990 to 2006; and Ghana. In the middle, 17 countries are 10 to 25 percentage points away from the target. These include, in decreasing order, Madagascar, Tanzania, Swaziland, Angola, Rwanda, Zambia, Benin, Togo, Liberia, Ethiopia, Chad, Burundi, Kenya, the Central African Republic, Sudan, Eritrea, and Lesotho. Also, Botswana, Cameroon, Côte d'Ivoire, Gabon, The Gambia, Guinea, Mali, Mauritania, Senegal, Uganda, and Zimbabwe are less than 10 percentage points away from the target. Progress is more modest in sanitation: in 29 countries, improved san- itation coverage will have to more than double for the MDG target to be reached. In the sanitation space, at one end stand Eritrea, Sierra Leone, Togo, Niger, Chad, Ghana, Madagascar, Rwanda, Ethiopia, and Burkina Faso, all more than 40 percentage points away from MDG tar- gets. A second group--including Liberia, Guinea, Mauritania, Côte d'Ivoire, Senegal, Tanzania, Nigeria, Uganda, Sudan, Gabon, Burundi, Lesotho, Mozambique, Namibia, Kenya, the Democratic Republic of Congo, Zimbabwe, Benin, Swaziland, Equatorial Guinea, the Central African Republic, Mali, Botswana, and The Gambia--show coverage rates between 20 and 40 percentage points below targets. Only Zambia, South Africa, Cameroon, Malawi, Angola, and Mauritius report coverage rates less than 20 percentage points away from targets (figure 7.2). For countries that have already reached the water MDG, the analysis presented here sets the bar a little higher, assuming that the number of people without access in 2006 (instead of 2000) is halved by 2015. Also, it is assumed that the water and sanitation MDG is reached equally in urban and rural areas. The challenge is particularly severe for rural areas, whereas in some countries urban access is already on or above target. In this case, current urban access is assumed to be maintained in 2015. The water and sanitation MDG targets translate into 764 million water customers and 646 million sanitation customers by 2015 using demo- graphic projections for urban and rural populations, and assuming urban and rural population growth rates are equal to the averages for the past decade. This means that improved water service will need to be extended to an additional 308 million Africans, equal to one-third of the overall population in 2006 (table 7.1). Almost 70 percent of the new customers will be located in rural areas. To reach the sanitation MDG, the popula- tion with improved sanitation will need to more than double. New cus- tomers stand at 409 million people, equal to more than half of the overall population in 2006. Again, almost 70 percent of new customers will be located in rural areas. 194 Africa's Water and Sanitation Infrastructure Figure 7.2 Sanitation MDG Gap, 2006 Eritrea Sierra Leone Togo Niger Chad Ghana Madagascar Rwanda Ethiopia Burkina Faso Liberia Guinea Mauritania Côte d'Ivoire Senegal Tanzania Nigeria Uganda Sudan Gabon Burundi Lesotho Mozambique Namibia Kenya Congo, Dem. Rep. Zimbabwe Benin Swaziland Equatorial Guinea Central African Republic Mali Botswana Gambia, The Zambia South Africa Cameroon Malawi Angola Mauritius 0 5 10 15 20 25 30 35 40 45 50 percentage point difference between MDG target and current access Source: JMP 2006. Middle-income countries are better positioned with respect to the MDG challenge in both absolute and relative terms, given the typically higher starting levels of coverage. They will have to improve water service for 9 million Africans and improve sanitation for 16 million. Nonfragile, low-income countries will face the largest number of new Spending Needed to Meet Goals in Water and Sanitation 195 Table 7.1 Additional Population to Be Served by 2015 (millions of people) Water Sanitation National Urban Rural National Urban Rural Angola 7.0 4.3 2.7 5.3 3.6 1.6 Benin 3.7 1.4 2.4 3.8 0.9 3.0 Botswana 0.3 0.2 0.0 0.6 0.3 0.3 Burkina Faso 6.0 1.5 4.6 8.2 1.1 7.1 Burundi 4.0 0.5 3.5 4.5 0.6 3.9 Cameroon 4.2 3.6 0.6 6.5 4.0 2.5 Cape Verde 0.2 0.1 0.0 0.2 0.1 0.1 Central African Republic 1.1 0.3 0.8 1.4 0.4 1.0 Chad 3.8 1.1 2.8 6.6 1.6 5.0 Congo, Dem. Rep. 26.2 5.3 20.9 26.8 8.5 18.3 Congo, Rep. 1.2 0.6 0.6 2.0 1.3 0.7 Côte d'Ivoire 4.5 4.2 0.3 8.9 3.7 5.1 Equatorial Guinea 0.2 0.1 0.1 0.2 0.1 0.1 Eritrea 1.9 0.4 1.5 3.2 0.7 2.5 Ethiopia 23.3 5.6 17.8 42.5 6.2 36.3 Gabon 0.3 0.3 0.0 0.6 0.5 0.1 Gambia, The 0.6 0.5 0.2 0.8 0.6 0.2 Ghana 7.1 4.0 3.0 12.6 6.6 6.0 Guinea 1.5 0.9 0.6 4.4 1.3 3.1 Kenya 11.9 2.5 9.4 16.8 5.9 10.8 Lesotho 0.4 0.2 0.2 0.8 0.2 0.5 Liberia 1.5 0.9 0.5 2.2 1.1 1.1 Madagascar 8.0 1.7 6.3 10.9 3.2 7.7 Malawi 4.9 1.4 3.5 4.5 1.6 2.9 Mali 3.3 1.7 1.6 5.2 1.6 3.6 Mauritania 0.8 0.3 0.6 1.6 0.4 1.2 Mauritius 0.1 0.0 0.1 0.2 0.1 0.1 Mozambique 9.5 2.7 6.8 9.5 2.9 6.6 Namibia 0.4 0.2 0.1 0.8 0.2 0.6 Niger 7.5 0.8 6.7 8.7 1.0 7.7 Nigeria 69.7 29.0 40.7 72.0 37.7 34.4 Rwanda 4.3 1.3 3.1 6.0 1.5 4.6 Senegal 3.5 1.5 2.0 6.2 1.5 4.8 Sierra Leone 3.3 0.8 2.5 3.8 1.3 2.5 South Africa 7.7 5.9 1.8 13.4 7.8 5.6 Sudan 11.8 6.9 4.9 17.8 7.7 10.1 Swaziland 0.4 0.1 0.3 0.4 0.1 0.3 (continued next page) 196 Africa's Water and Sanitation Infrastructure Table 7.1 (continued) Water Sanitation National Urban Rural National Urban Rural Tanzania 15.4 3.5 11.9 20.4 6.5 13.9 Togo 15.9 1.4 14.5 20.7 2.1 18.6 Uganda 9.3 1.4 7.9 15.6 2.4 13.3 Zambia 3.6 0.7 2.9 3.7 1.2 2.5 Zimbabwe 2.0 1.0 1.0 4.2 1.1 3.0 Resource-rich 101.8 46.5 55.3 114.6 57.7 57.0 Middle-income 9.3 6.7 2.6 16.3 8.7 7.6 Low-income, fragile 62.5 16.2 6.3 81.0 21.5 59.4 Low-income, nonfragile 118.5 31.2 87.4 172.7 43.3 129.4 Sub-Saharan Africa 307.7 92.4 215.3 408.7 130.0 278.7 Sources: JMP 2006; World Development Indicators Database 2006 (http://data.worldbank.org/). water customers in absolute terms: 120 million. Yet, in relative terms, they are better positioned than fragile, low-income countries, which will need to raise by more than 40 percent the number of people with access to improved water. Interestingly, even in relative terms, fragile, low- income countries score second to resource-rich countries, which will need to add more than 100 million Africans to water service, equal to 42 per- cent of their current population. In the sanitation space, nonfragile, low- income countries face the most difficult challenge: they will have to add more than 170 million customers, equal to more than half of their cur- rent population. Fragile, low-income countries follow closely in relative terms, although in absolute terms resource-rich countries will need to add a larger number of customers given their size. The analysis assumes a base scenario in which the share of the popula- tion using high-quality or improved water and sanitation services relative to the overall served population will remain the same in 2015, although in absolute terms more people will enjoy high-quality services because of demographic growth (figure 7.3). As a result, in 2015 private water con- nections will account for no more than one-third of improved water cov- erage, standposts for another third, and wells and boreholes for the remaining 40 percent. Similarly, improved sanitation coverage will still be achieved predominantly through safe traditional latrines (around 60 per- cent); ventilated improved pit latrines will account for one-fifth of Spending Needed to Meet Goals in Water and Sanitation 197 Figure 7.3 Population Split across Water and Sanitation Modalities Given Current and Target Coverage by 2015 under the Base Scenario Assumptions a. Water improved water 2015 21% 23% 31% coverage 2006 18% 18% 23% 0 10 20 30 40 50 60 70 80 % national population piped water public tap/standposts safe wells/boreholes b. Sanitation improved sanitation 2015 2% 11% 11% 39% coverage 2006 1% 6% 6% 18% 0 10 20 30 40 50 60 70 80 % national population sewer connection septic tanks improved latrines safe traditional pit latrine Source: Authors' compilation. improved sanitation coverage, septic tanks for another fifth, and sewer connections for less than 5 percent. The Unit Cost of Service Provision across Countries The unit costs of infrastructure determine the level of spending on service expansion, rehabilitation, and operations and maintenance (O&M). Unit costs vary to a large extent across countries and within regions as a result of density, location, technological innovation, and level of local market development--factors that are almost all exogenous, at least in the short and medium terms.1 Concentration largely reduces investment costs for water and sewerage networks in dense city centers, whereas great distances make it impractical to roll out connection lines into dispersed rural areas. 198 Africa's Water and Sanitation Infrastructure Poor capacity in the local construction sector, lack of skilled construction workers, shortage of materials, and scarce financing reduce the range of available on-site technologies and constrain the development of innova- tions that would ensure higher quality at more affordable prices. Efficiency considerations call for understanding of what level of service can be realis- tically provided to as many people as possible rather than channeling lim- ited public resources into higher-quality services accessible by considerably fewer people. Therefore, the typology of country settings ultimately shapes the trade-off between political objectives and spending constraints. Vast differences in costs are seen across countries and between urban and rural areas. The capital cost per capita of an urban residential water connection can range from $200 in countries where urbanization and concentration have taken off up to $1,000 in countries that are primarily rural. Similarly, the capital cost per capita of an urban standpost connec- tion can range from $60 to $150. The price of a residential water connec- tion in rural areas fluctuates even more across countries. It increases from $700 in countries where rural areas are more densely populated to a price 10 times higher in countries with mostly remote rural spaces. The capital cost per capita of a standpost installed in rural areas fluctuates less, with a range of $100 to $200. Similar differences in network infrastructure prices can be found within countries between urban and rural areas. Table 7.2 reports capital costs per capita of water connections at different urban and rural loca- tions. Locations span from megacities with populations of more than 3 million people and densities of about 5,000 people per square kilome- ter, to rural areas more than six hours of travel time away from the near- est urban center, whose typical densities barely reach 15 people per square kilometer. The capital cost per capita of a private connection increases exponentially from highly dense megacities to remote rural areas. The capital cost per capita of a standpost connection quadruples. The considerable sensitivity to density makes infrastructure networks less affordable in Africa than elsewhere. Africa remains a predominantly rural continent and is therefore low-density. Sixty-six percent of Africans still live in rural areas, and of those 50 percent live in the rural hinterlands and up to 16 percent in remote villages. Also, one-third of the urban pop- ulation--equal to 10 percent of the overall population--lives in peri- urban areas with fewer than 100 people per square kilometer, and a slightly larger share--13 percent of the overall population--lives in cities with a population of more than 1 million and densities of about 3,500 people per square kilometer. Table 7.2 Unit Costs of Water Network Infrastructure Services by Location in the Median Country Large cities >3 2.0­2.99 1.0­1.99 0.5­0.99 0.1­0.49 million million million million million Secondary Rural Deep rural people people people people people cities hinterland area Median density (inhabitants/km2) 5,009 4,083 2,855 2,712 1,373 1,282 38 13 $ per capita Private water connection 232 255 302 309 428 443 1,825 3,156 Standpost connection 66 72 85 87 119 123 268 273 Source: Authors' compilation. Note: Cities are classified by population size with typologies spanning from secondary cities with populations of fewer than 100,000 people to megacities with more than 3 million inhabi- tants. Nonurban areas are classified by distance or travel time to the nearest city. In particular, "rural hinterland" indicates rural locations within six hours' travel time from the closest urban center, and deep rural areas are those more than six hours away. Urban and rural locations are assigned with the median of the densities estimated for each location in 42 Sub-Saharan African countries. 199 200 Africa's Water and Sanitation Infrastructure Unit costs of on-site facilities also vary across countries. The price of a borehole with hand pump is $20 to $90 per capita, and the price of a well with hand pump is $15 to $80 per capita (table 7.3). More advanced technologies, such as boreholes with hand or even electric pumps, are typically used in urban areas, whereas less-expensive technologies are more common in rural areas, where low densities require less capacity. Unit costs of on-site sanitation services range from $39 for a traditional latrine to $60 for an improved latrine to $125 for a septic tank (table 7.4). Sanitation unit costs are adjusted by a construction index factor (box 7.1) to reflect differences across local construction markets and lev- els of technological innovation in the sanitation sector. For sewerage networks, owing to their low prevalence in Africa, a median unit cost based on experience from World Bank operations has Table 7.3 Unit Costs of Wells and Boreholes Borehole with hand pump Well with hand pump ($ per capita) ($ per capita) Benin 50 36 Burkina Faso 36 26 Cameroon 76 58 Cape Verde 50 36 Chad 50 36 Congo, Dem. Rep. 50 36 Côte d'Ivoire 50 36 Ethiopia 50 36 Ghana 22 20 Kenya 50 36 Lesotho 50 36 Madagascar 50 17 Malawi 50 36 Mozambique 50 36 Namibia 50 36 Niger 94 82 Nigeria 50 36 Rwanda 50 36 Senegal 50 36 South Africa 50 36 Sudan 50 36 Tanzania 50 36 Uganda 50 36 Zambia 50 36 Source: World Bank's public expenditures reviews for Cameroon, Côte d'Ivoire, and Niger. Spending Needed to Meet Goals in Water and Sanitation 201 Table 7.4 Unit Costs of On-Site Sanitation Services Septic tank Improved latrine Traditional latrine $ per capita 125 57 39 Source: World Water Assessment Programme 2000, http://www.unesco.org/water/wwap/wwdr/indicators/. Box 7.1 The Construction Index Factor The construction index used in this analysis results from the Basket of Construc- tion Components (BOCC) approach introduced in the 2003 to 2006 round of the International Comparison Program (ICP) to calculate comparable prices in the construction sector. The ICP, the world's largest statistical initiative, produces internationally compa- rable price levels, economic aggregates in real terms, and purchasing power parity estimates. The ICP uses a series of statistical surveys to collect price data for a basket of goods and services. By using estimates of purchasing power parity as conversion factors, the resulting comparisons of gross domestic product allow for measuring the relative social and economic well-being of countries, monitoring the incidence of poverty, tracking progress toward the MDGs, and targeting programs effectively. The launch of BOCC followed the conclusion that lack of comparability of cap- ital goods in different countries had weakened the effectiveness of the past ICP round. In particular, BOCC resulted from the attempt to respond to the following issues: Given the nature of the construction sector and the inherent difficulties in construction price comparisons, what improvements can be made? What basis and level of comparison is appropriate for the sector? How can quality and level- of-service differences among countries be incorporated in these comparisons? The BOCC measures relative prices at the level of the construction compo- nent, which can be thought of as an aggregation of several construction work items. These items include the material put in place, labor and equipment, and any consumables required. The price comparisons are performed using three baskets: residential, nonresidential, and civil works. Each basket is broken down into construction systems. Under each system a set of construction components is identified and defined. The approach was endorsed by the ICP Technical Advi- sory Group as a much simpler price comparison tool than the current practice, and it is expected to reduce resource and expertise requirements in the price col- lection process in the construction sector. Source: Adapted from World Bank, "International Comparison Program 2011," http://www.worldbank.org/ data/icp. 202 Africa's Water and Sanitation Infrastructure been estimated at $400. After adjusting for the construction index factor, the average cost per capita of a sewerage connection is estimated at $440. To Close the MDG Coverage Gap The total spending required for reaching the water and sanitation MDGs is valued at $22.6 billion per year or 3.5 percent of Africa's gross domes- tic product (GDP). Most of the needs come from the water sector, which is estimated to require allocations up to $17 billion per year or 2.7 per- cent of Africa's GDP (table 7.5). The cost of new infrastructure appears to carry the heaviest weight and require allocations up to 1.5 percent of Africa's GDP every year, or 43 percent of overall spending. O&M needs immediately follow and stand at 1.1 percent of Africa's GDP, or 31 percent of overall costs. Rehabilitation of existing assets requires lower yet substantial allocations--up to 0.9 per- cent of Africa's GDP--which accounts for one-fourth of the overall needs. A similar composition can be observed for water spending needs, 42 percent of which are generated by investments in expansion, 25 per- cent by rehabilitation of existing assets, and 33 percent by O&M. The sanitation sector shows a different composition: investments in new infrastructure dominate spending needs and account for more than 40 percent, and rehabilitation and O&M each account for one-fourth. A larger share of spending on water and sanitation is allocated to rural areas because of the large urban-rural divide in access to infrastructure services, quality of service, and asset conditions, which are estimated to account for 59 percent of overall requirements (figure 7.4). In particular, rural areas should absorb up to 63 percent of the overall investments in new infrastructure. Almost the same share of rehabilitation spending should be channeled to rural areas, owing to a much more severe obsoles- cence of rural infrastructure. O&M needs are almost evenly split between urban and rural areas. These distribution patterns do not apply equally to water and sanita- tion. In the water space, more than 60 percent of spending needs origi- nate from rural areas, whether they are investments in new infrastructure, rehabilitation of existing assets, or maintenance. In the sanitation space, 55 percent of overall spending needs originate from urban areas. O&M needs mainly concern urban sanitation assets, yet rural areas account for 57 percent of rehabilitation needs. The composition of spending needs differs between middle- and low- income countries (table 7.6). Low-income countries, whether fragile or Table 7.5 Overall Water and Sanitation Spending Needs Share of GDP (%) $ million/year CAPEX CAPEX Total Expansion Rehabilitation Total CAPEX O&M Total Needs Expansion Rehabilitation Total CAPEX O&M needs Water 1.13 0.68 1.80 0.89 2.69 7,225 4,327 11,553 5,686 17,239 Sanitation 0.41 0.21 0.62 0.22 0.84 2,617 1,352 3,969 1,432 5,401 Total 1.54 0.89 2.42 1.11 3.53 9,843 5,679 15,522 7,118 22,640 Source: Authors' calculations. Note: CAPEX = capital expenditure, GDP = gross domestic product, O&M = operations and maintenance. 203 204 Africa's Water and Sanitation Infrastructure Figure 7.4 Urban-Rural Split of Spending Needs a. Water and sanitation 100 90 80 70 percent 60 50 40 30 20 10 0 n n M l ta io io O& to ns at ilit pa ab ex h re b. Water 100 90 80 70 60 percent 50 40 30 20 10 0 n n M l ta sio io O& to at n ilit pa b ex ha re c. Sanitation 100 90 80 70 60 percent 50 40 30 20 10 0 n n M l ta sio io O& to at n lit pa bi ex ha re rural urban Source: Authors' compilation. Table 7.6 Split of Spending Needs by Category Share of GDP (%) US$ million per year CAPEX CAPEX Total Total New Total spending New Total spending investment Rehabilitation CAPEX O&M needs investment Rehabilitation CAPEX O&M needs Sub-Saharan Africa 1.5 0.9 2.4 1.1 3.5 9,843 5,679 15,522 7,118 22,640 Resource-rich 1.3 0.8 2.1 0.8 2.9 2,864 1,741 4,605 1,759 6,364 Middle-income 0.4 0.4 0.7 0.7 1.5 1,034 951 1,985 1,991 3,976 Low-income, fragile 5.9 2.7 8.5 3.3 11.8 2,208 1,006 3,213 1,223 4,437 Low-income, nonfragile 3.4 1.8 5.1 1.9 7.1 3,714 1,968 5,682 2,128 7,810 Source: Authors' calculations based on access data as of 2006. Note: CAPEX = capital expenditure, GDP = gross domestic product, O&M = operations and maintenance. 205 206 Africa's Water and Sanitation Infrastructure nonfragile, and resource-rich countries show much similarity, with costs divided almost equally among expansion and rehabilitation and mainte- nance. Conversely, middle-income countries focus more on maintenance, which accounts for half the overall needs, and the high coverage rates and relatively lower rehabilitation backlog make infrastructure expansion and rehabilitation less of a priority. The total spending needs range from a maximum of $3.3 billion per year in the case of South Africa to a minimum of $19 million per year in the case of Equatorial Guinea, with a fair number of countries, including Nigeria, Sudan, Kenya, the Democratic Republic of Congo, Ethiopia, and Tanzania, that should spend between $1.0 and $2.3 billion per year to halve the gap of people without access to water and sanitation services by 2015 (figure 7.5). Middle-income countries together report the highest needs, almost $3 billion per year, followed by resource-rich countries, with $1.5 billion per year. Despite the lower size of their economies, low- income countries altogether account for a similar amount, owing to the larger service gap they have to make up for. It should be noted, however, that for some countries, part of the infor- mation required to calculate specific spending components is not avail- able. For these, estimates may be just lower bounds of the actual spending needs. Normalizing needs by the size of the countries' economies reveals that most countries should allocate well over 3 percent of their GDP every year to water and sanitation. As expected, the level of spending required by the MDG varies to a large extent across countries. Three country groups can be identified. The first group represents countries with large spending needs--more than 10 percent of the GDP per year. The second includes countries with medium spending needs--3 to 10 percent of the GDP per year. The third group consists of those with needs less than 3 percent of GDP per year. Among these, Equatorial Guinea stands at the bottom of the distri- bution, with overall needs below 0.3 percent of the GDP. On the oppo- site end, Togo, the Democratic Republic of Congo, and Liberia, show manifestly unaffordable needs that reach more than 20 percent of the GDP per year. The affordability of the MDG challenge appears to correlate strongly to a country's income. Halving the population without access to water and sanitation services by 2015 is estimated to require only 1.5 percent of middle-income countries' GDP per year. Resource-rich countries should invest twice as much annually--3 percent of their Spending Needed to Meet Goals in Water and Sanitation 207 Figure 7.5 Africa's Water and Sanitation Needs by Country a. US$ (billion per year) Sub-Saharan Africa resource-rich middle-income low-income, nonfragile low-income, fragile South Africa Nigeria Sudan Kenya Congo, Dem. Rep. Ethiopia Tanzania Côte d'Ivoire Togo Madagascar Angola Ghana Zambia Cameroon Zimbabwe Mozambique Uganda Senegal Mali Namibia Benin Niger Congo, Rep. Malawi Burkina Faso Sierra Leone Central African Republic Eritrea Chad Botswana Guinea Liberia Mauritius Rwanda Burundi Mauritania Gambia, The Gabon Lesotho Swaziland Cape Verde Equatorial Guinea 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 US$ (billion per year) expansion rehabilitation O&M (continued next page) 208 Africa's Water and Sanitation Infrastructure Figure 7.5 (continued) b. Percentage of GDP Sub-Saharan Africa low-income, fragile low-income, nonfragile resource-rich middle-income Togo Congo, Dem. Rep. Liberia Sierra Leone Gambia, The Eritrea Burundi Zimbabwe Madagascar Central African Republic Kenya Ethiopia Niger Sudan Tanzania Malawi Benin Zambia Mali Mozambique Côte d'Ivoire Namibia Rwanda Mauritania Ghana Uganda Senegal Burkina Faso Lesotho Guinea Congo, Rep. Cameroon Chad Cape Verde Nigeria Angola Mauritius South Africa Botswana Swaziland Gabon Equatorial Guinea 0 10 20 30 40 % of GDP expansion rehabilitation O&M Source: Authors' compilation. Note: O&M = operations and maintenance. Spending Needed to Meet Goals in Water and Sanitation 209 GDP. The bill becomes prohibitively expensive for low-income coun- tries, which are required to allocate at least 7 percent of GDP annu- ally to water and sanitation every year, and especially for fragile states, for which water and sanitation needs reach almost 12 percent of GDP per year. Compared with existing spending on water and sanitation--a topic of discussion for the next chapter--delivering the additional financing needed to meet the water and sanitation MDGs looks comfortably manageable only for middle-income countries and barely manageable for resource-rich countries. Both might be able to afford service expan- sion in tandem with maintaining and even improving service standards. This is not the case for low-income countries, however, particularly fragile states. Realistically, these countries must either accommodate new customers with lower-cost technologies that substantially reduce investment needs and maintenance costs or postpone their achievement of the goals. Annex 7.1 Unit Cost Matrix Model: A Methodology for Estimating Nonstandardized Unit Costs of Network Assets The unit costs matrix model is designed specifically to estimate the cap- ital cost per capita of expanding networks in all relevant infrastructure sectors, including water and sanitation, energy, information and commu- nication technologies, and roads, as a function of density and location. The main value of this model is that it allows estimation of country-specific, as opposed to standardized, unit costs. As such, it provides a tool to assess the affordability and efficiency of networks given a country's typical geog- raphy, urbanization, and density patterns, and to explore the viability of lower-cost technological alternatives. A prerequisite to the analysis is the definition of density-based city categories and rural regions, which ideally compose an urban-rural gra- dient. Cities are classified by population size using data from Henderson (2002)2 so that typologies span from secondary cities with populations fewer than 100,000 people to megacities with more than 3 million inhabitants. Nonurban areas, including rural hinterlands and deep rural regions, are classified by distance or travel time to the near- est city. Densities are attributed to each typology (table 7.1A) using extent layers from the Global Rural-Urban Mapping Project (GRUMP). This makes it possible to convert the distribution of human population Table 7.1A Population Density across Urban and Rural Typologies (Number of People per Square Kilometer) Large urban Secondary cities Rural hinterland Deep rural area 210 >3 2.0­2.99 1.0­1.99 0.5­0.99 0.1­0.49 Between 1 and 6 More than 6 hours' million million million million million < 0.1 million hours' travel time travel time from people people people people people people from nearest city nearest city Benin 4,861 1,446 840 46 13 Burkina Faso 2,108 271 268 43 15 Cameroon 4,897 645 1,640 31 7 Cape Verde 1,248 102 40 Chad 2,854 1,373 200 23 4 Congo, Dem. Rep. 2,571 2,617 1,367 35 13 Côte d'Ivoire 4,743 2,306 1,430 1,142 34 13 Ethiopia 4,724 1,644 1,440 107 25 Ghana 3,159 3,199 413 690 71 29 Kenya 2,461 19,928 1,367 1,682 89 5 Lesotho 1,168 -- 71 35 Madagascar 3,134 1,040 1,692 40 15 Malawi -- -- -- -- -- -- 114 12 Mozambique 5,008 2,318 1,601 26 9 Namibia 534 -- 2 2 Niger 1,573 1,950 1,246 36 2 Nigeria 5,394 4,349 2,806 2,614 1,315 91 50 Rwanda 2,650 -- 309 103 Senegal 8,630 1,903 1,383 33 7 South Africa 1,765 1,076 574 400 22 1 Sudan -- -- -- -- -- 807 22 6 Tanzania 4,083 5,406 2,032 1,672 36 16 Uganda 2,529 3,060 122 37 Zambia 1,307 1,105 850 582 14 8 Sources: Authors' compilation based on GRUMP data and Henderson 2002. Note: Blank cells indicate no cities with that population size. -- = not available. Spending Needed to Meet Goals in Water and Sanitation 211 from national or subnational spatial units (usually administrative units) to a series of geo-referenced quadrilateral grids. In urban areas, where multiple cities of a country fall in the same category, the median den- sity of the category is calculated. It should be noted that density figures are approximate at best owing to the limitations associated with input data. A particular limitation is posed by the paucity of data sets that observe city populations at the same point in time. Henderson (2002) is one of the few, but its data are no more recent than 2000. The analysis disaggregates unit cost structures of network water ser- vices in subcomponents, such as water production and storage, distribu- tion, and connection, and estimates them separately. Although water production, storage costs, and connection costs do not vary by density and location, distribution costs are a function of distance from the water source and concentration of connections. Standard values for key inputs to the analysis, such as water produc- tion capacity per day, storage capacity per connection, and urban and rural water consumption by house and standpost connection, are derived from World Bank water programs in Africa. In addition to these, a few assumptions are made regarding the num- ber of people per standpost--no more than 200--and the normative walking distance to a standpost--1 kilometer maximum. Unit prices of materials and technologies (such as the cost of a well with an electric pump or of a meter of water main and small diameter pipe) and connection costs are derived from a study undertaken as part of the Africa Infrastructure Country Diagnostic that collected evidence on unit costs from water and sanitation programs financed by donors in Africa between 2002 and 2006 (box 7.1A). Box 7.1A Unit Costs of Infrastructure Projects Study The objective of the Unit Costs of Infrastructure Projects Study is to design, gen- erate, and analyze a database of standardized unit costs for different types of commonly financed infrastructure investments in Sub-Saharan Africa over the past decade. Actual unit costs are gathered from recently completed projects by (continued next page) 212 Africa's Water and Sanitation Infrastructure Box 7.1A (continued) using documentation on procured contracts obtained from four development finance institutions. The analysis spans relevant infrastructure sectors, including roads, water and sanitation, and energy. Although the objective was to compile a representative sample of projects, with a target of 150 contracts per sector, practical constraints limited the sample to 115 road contracts, 144 water contracts, and 58 electricity contracts over a shorter period of time--approximately 2002­06. The study focuses on unit output costs--that is, the cost per unit of infra- structure (a water connection, for example) as opposed to the cost per unit of in- put (such as labor costs). Standardized output costs are especially useful for planning purposes and for estimating value for money. The spread of unit cost values is described using the median--not affected by outlier values--as the center point and the interquartile range to explain the distance from the center. Outlier values are excluded in the calculation of the range. Three main challenges emerged from this study and are likely to affect similar exercises of this kind. First, the great variability among collected unit cost figures mainly reflects differences in project design. This is an issue because available in- formation on project design does not easily allow standardizing the infrastructure outputs being compared. Where this information is available, it takes the form of technical specifications that run to hundreds of pages. The variability in the de- sign of the outputs made it necessary to subdivide contracts into ever-smaller categories--something not conducive to making generalized conclusions. Sec- ond, many practical challenges are involved in parsing and compiling informa- tion. Not least of these is the difficulty of obtaining decentralized paper records of projects from donors. Even where electronic databases are maintained, locating and segregating the relevant data remains a complex and time-consuming exer- cise. Third, data collection difficulties normally occur, reducing the sample size and the significance of the comparisons being made. As far as water and sanitation is concerned, the 144 sampled projects include 33 well contracts, 60 distribution main contracts, 14 reservoir contracts, 26 service connection contracts, and 11 public latrine contracts. Data are drawn from only one development institution, and the country coverage is highly skewed, with more than 80 percent of the contracts coming from just five countries: Mozam- bique, Namibia, Nigeria, Tanzania, and Zambia. The water and sanitation unit costs are summarized in the following table. (continued next page) Spending Needed to Meet Goals in Water and Sanitation 213 Box 7.1A (continued) Unit Costs for Water and Sanitation Projects, 2006 US$ Lower Upper Type Unit quartile Median quartile Wells--no pump $/well 5,297 6,341 6,707 Wells--electric pump $/well 14,112 37,492 54,701 Wells--electric and hand pump $/well 11,288 13,959 14,896 Pipe--small diameter $/m 14 26 40 Pipe--midsize diameter $/m 122 144 219 Pipe--mains $/m 358 457 633 Reservoir construction--steel $/kl 437 1,067 2,584 Service connection--yard $/conn 13 24 74 Service connection--standpipe $/conn 177 282 363 Latrines--public $/conn 14,014 19,659 29,662 Source: Adapted from Africon 2008. Note: Italicized rows denote sample sizes large enough to provide reliable unit cost predictions. conn = connection, kl = kiloliter, m = meter. Annex 7.2 Methodology for Quantifying Rehabilitation and O&M Needs Network infrastructure: UCi Ri = k × , 30 UCi = Unit cost per capita of asset I, k = coefficient that takes a value of 5 or 10 depending on country category. Nonnetwork infrastructure: UCi Ri = a × , l UCi = Unit cost per capita of asset I, a = value of the components of asset i to be replaced expressed as percentage of the total cost of I, l = life span of asset I. 214 Africa's Water and Sanitation Infrastructure Values for a and l: a (%) l (years) Water Urban areas 40 10 Rural areas 80 5 Sanitation Septic tank 12.5 10 Improved latrine 12.5 10 Traditional (safe) latrine 100 5 Per capita O&M: O&M = p × UCi , UCi = Unit cost per capita of asset I, p = coefficient that takes a value of 3% for network assets and 1.5% for nonnetwork assets. Notes 1. Based on unit cost matrix model (annex 7.1) designed for this analysis. It esti- mates the capital cost per capita of a network connection at varying levels of density in both urban and rural areas. 2. This is one of the few databases compiling city populations at the same point of time. References Africon. 2008. "Unit Costs of Infrastructure Projects in Sub-Saharan Africa." AICD Background Paper 11, World Bank, Washington, DC. Henderson, J. Vernon. 2002. "World Cities Data." http://www.econ.brown.edu/ faculty/henderson/worldcities.html. JMP (Joint Monitoring Programme). 2006. "Meeting the MDG Drinking Water and Sanitation Target: The Urban and Rural Challenge of the Decade." World Health Organization, Geneva, and United Nations Children's Fund, New York. CHAPTER 8 Bridging the Funding Gap The price tag for many countries to accomplish the Millennium Development Goals (MDGs) for water supply and sanitation (WSS) is prohibitive when compared with current levels of spending. This chapter delves into the levels and composition of spending on WSS, evaluates how much more can be done within Africa's existing resource envelope by alleviating inefficiencies, and finally arrives at the annual funding gap. It further explores the potential for raising additional financing and pol- icy adjustments to reduce the burden of the funding gap. Current Spending on Water and Sanitation Africa is spending a total of $7.9 billion a year to address its WSS needs, which is equivalent to 1.2 percent of Sub-Saharan Africa's gross domes- tic product (GDP). Existing spending on infrastructure in Africa is higher than previously thought when the calculation takes into account budget and off-budget spending--including state-owned enterprises (SOEs) and extrabudgetary funds--as well as external financing, a category that com- prises official development assistance (ODA) from the member states of the Organisation for Economic Co-operation and Development (OECD), financiers from outside the OECD, self-household financing, 215 216 Africa's Water and Sanitation Infrastructure and private participation in infrastructure. Overall, however, these num- bers might be underestimated given the complexity of traced resources allocated to the sector, in particular those coming from nongovernmental organizations and allotted to sanitation or rural water programs, which are not always centrally recorded and hence could not be fully captured in this exercise. In absolute terms, spending levels vary significantly across the coun- try groups (table 8.1): Middle-income countries spend $2.6 billion, fol- lowed by low-income countries ($1.8 billion), and resource-rich countries ($1.7 billion); fragile states spend about $0.5 billion in capi- tal investment and operations and maintenance (O&M). Expressed as a percentage of GDP, infrastructure spending fluctuates widely across dif- ferent country groups; whereas low-income countries and fragile states spend 1.1 percent and 1.7 percent of their GDP, respectively, middle- income countries and resource-rich countries spend 1 percent or less of GDP (1.0 percent and 0.8 percent, respectively). The composition of spending also varies substantially across country groups. Middle-income countries allocate 80 percent of WSS spending to maintenance, likely reflecting the fact that they have already built much of the infrastructure needed. By contrast, all the other country groups allocate at most 30 percent to this item. Therefore, resource-rich coun- tries, low-income countries, and fragile states spend 70 to 90 percent of their budgets on capital investments. Although this reflects their need to build new WSS facilities, a danger exists of neglecting the maintenance needs of the limited network that is available. Table 8.1 Spending by Functional Category, Annualized Average Flows, 2001­05 GDP share (%) US$ (million per year) Total Total Total Total O&M CAPEX spending O&M CAPEX spending Sub-Saharan Africa 0.5 0.7 1.2 3,112 4,778 7,890 Low-income, fragile 0.3 0.8 1.1 128 313 441 Low-income, nonfragile 0.3 1.4 1.7 307 1,533 1,840 Middle-income 0.7 0.2 1.0 1,996 641 2,637 Resource-rich 0.1 0.7 0.8 188 1,564 1,753 Sources: Foster and Briceño-Garmendia 2009; Briceño-Garmendia, Smits, and Foster 2008 for public spending; PPIAF 2008 for private flows; Foster and others 2008 for non-OECD financiers. Note: Aggregate public sector covers general government and nonfinancial enterprises. Figures are extrapola- tions based on the 24-country covered in AICD Phase 1. Total might not add up exactly because of scaling up among country groups and rounding error. CAPEX = capital expenditure, GDP = gross domestic product, O&M = operations and maintenance. Bridging the Funding Gap 217 The explanations for this composition of spending are different in each case. For low-income fragile states, the problem is the limited flow of resources available; this fosters a preference for investing in expansion of access to new customers. Resource-rich countries, in contrast, have a lim- ited propensity to spend on infrastructure. The spending effort is relative to the size of the economy. The diver- gence in WSS spending across countries is also considerable; it ranges from 0.7 percent of GDP in Chad to 3.1 percent in Ethiopia. Particularly important differences are seen in the shares of spending in O&M and cap- ital spending (figure 8.1). Namibia, Ethiopia, Botswana, and South Africa allocate the highest percentages of their GDP to O&M of the existing infrastructure, whereas Chad and Madagascar spend the least in this cat- egory. Surprisingly, Uganda and Senegal, some of the best performers in Sub-Saharan Africa, assign less than 0.05 percent of GDP to O&M. Ethiopia, Benin, Zambia, and Niger dedicate the highest percentages of GDP to capital investment, and South Africa the least, at 0.07 percent of its GDP. Three key players are seen in WSS sector financing: the public sector, donors, and households (table 8.2). In Sub-Saharan Africa, households are important financiers of capital investment (0.3 percent of Sub-Saharan African GDP) and account for $2.1 billion, most of it dedicated to the construction of on-site sanitation facilities, such as latrines. The level of contributions from OECD donors is similar to that of domestic public resources (comprising tax revenue and user charges raised by SOEs), equivalent to 0.2 percent of Sub-Saharan African GDP. The contribution of non-OECD countries is only 0.03 percent of Sub-Saharan African GDP, and that of the private sector is almost nonexistent (close to 0 per- cent of Sub-Saharan African GDP). Financing follows specialization patterns. Across country groups, households' contribution to rehabilitation and construction of new facilities ranges between 0.2 percent (middle-income countries) and 1.4 percent (low-income, nonfragile countries; figure 8.2). The role of ODA is particularly important to low-income, nonfragile countries because it represents on average 0.7 percent of GDP of countries with limited domestic resources but adequate institutional capacity. In resource-rich countries, the public sector plays a significant part in financing the WSS sector (0.3 percent of GDP), but its role in the frag- ile states and middle-income countries is very modest. Non-OECD finance has shown a preference for low-income countries (fragile and nonfragile) and resource-rich countries. 218 Africa's Water and Sanitation Infrastructure Figure 8.1 Water and Sanitation Spending from All Sources as a Percentage of GDP, Annual Averages by Functional Category, 2001­05 Sub-Saharan Africa low-income, nonfragile low-income, fragile middle-income resource-rich Ethiopia Namibia Benin Zambia Niger Lesotho Ghana Rwanda Burkina Faso Tanzania Mozambique Malawi Madagascar Botswana Senegal Mali Cape Verde Uganda Côte d'lvoire Kenya Congo, Rep. Cameroon South Africa Nigeria Chad 0 0.4 0.8 1.2 1.6 2.0 2.4 2.8 3.2 % GDP O&M CAPEX Source: Authors' calculations. Note: CAPEX = capital expenditure, GDP = gross domestic product, O&M = operations and maintenance. Table 8.2 Capital Investments of the Most Important Players, Annualized Average Flows, 2001­05 GDP share (%) US$ (million per year) Public Non-OECD Household Total Public Non-OECD Household Total sector ODA financiers PPI self-finance CAPEX sector ODA financiers PPI self-finance CAPEX Sub-Saharan Africa 0.2 0.2 0.03 0.0 0.3 0.7 1,252 1,227 163 10 2,125 4,778 Low-income, fragile 0.1 0.3 0.05 0.0 0.4 0.8 30 105 20 0 165 313 Low-income, nonfragile 0.2 0.7 0.05 0.0 0.4 1.4 243 783 55 2 451 1,533 Middle-income 0.1 0.0 0.00 0.0 0.1 0.2 324 101 8 2 206 641 Resource-rich 0.3 0.1 0.04 0.0 0.2 0.7 717 238 80 7 522 1,564 Source: Foster and Briceño-Garmendia 2009; Briceño-Garmendia, Smits, and Foster 2008 for public spending; PPIAF 2008 for private flows; and Foster and others 2008 for non-OECD financiers. Note: CAPEX = capital expenditure, GDP = gross domestic product, ODA = official development assistance, OECD = Organisation for Economic Co-operation and Development, PPI = private participation in infrastructure. 219 220 Africa's Water and Sanitation Infrastructure Figure 8.2 Water and Sanitation Capital Investment as a Percentage of GDP, by Funding Source, Annualized Averages for 2001­05 1.60 1.40 percentage of GDP 1.20 1.00 0.80 0.60 0.40 0.20 0 ca tri e tri h tri ile tri ile un om un -ric ri un g un g es es es es Af co , fra co nfra co rce co inc n e e- u ra no m so dl ha co e, re id Sa m -in m co b- w Su -in lo w lo household self-finance PPI non-OECD financiers ODA public sector Source: Briceño-Garmendia, Smits, and Foster 2008; Foster and Briceño-Garmendia 2009. Note: GDP = gross domestic product, ODA = official development assistance, OECD = Organisation for Economic Co-operation and Development, PPI = private participation in infrastructure. Poor Budget Execution by the WSS Sector African governments allocate 0.7 percent of their GDP to support the provision of WSS infrastructure from their central government budgets alone (table 8.3). For Africa, this effort translates to an estimated $180 million a year for an average country. For a perspective on this figure, an investment of $100 million can purchase about 100,000 new household connections to water and sewerage. It runs well short of covering the WSS spending needs presented in chapter 7 of this book. As a percentage of GDP, budget spending on WSS infrastructure is comparable across resource-rich and low-income countries (fragile and nonfragile). In absolute terms, however, middle-income countries have a much larger infrastructure budget, with spending per capita several times higher than in low-income countries because of the much larger value of GDP (table 8.4). Overall, WSS spending is the second-largest infrastruc- ture item in central government accounts, after spending on transport, particularly in the middle-income countries. It ranges from about half of all Bridging the Funding Gap 221 Table 8.3 Annual Budgetary Flows, Annualized Averages, 2001­05 Share of GDP (%) US$ (billion per year) Sub-Saharan Africa 0.7 4.4 Low-income, fragile 0.4 0.2 Low-income, nonfragile 0.5 0.5 Middle-income 0.9 2.5 Resource-rich 0.4 0.9 Sources: Foster and Briceño-Garmendia 2009; Briceño-Garmendia, Smits, and Foster 2008. Note: Annualized averages for 2001­06 weighted by country GDP. Figures are extrapolations based on the 24-country sample covered in the AICD Phase 1. GDP = gross domestic product. central government spending on infrastructure in middle-income coun- tries to 60 percent in low-income countries. In Sub-Saharan Africa, about 40 percent of budgetary spending in water goes to O&M (table 8.1). In middle-income countries the percent- age allocated to O&M is more than 75 percent of the public spending in WSS infrastructure. Resource-rich and low-income (nonfragile) countries spend most of their budgetary resources in capital investments; very little remains for O&M. In low-income countries (fragile), the public spending in O&M is close to 30 percent. On average, in Sub-Saharan Africa governments finance 75 percent of the total budgetary spending, and the utilities contribute the remaining 25 percent (figure 8.3). The distribution of responsibilities among the central government and the utilities varies across the four typologies: In resource-rich countries, most of the public spending is financed by the central government (80 percent), whereas in low- income countries (fragile), 70 percent of the spending in the sector comes from nonfinancial public institutions (equivalent to 0.04 percent of their GDP). In comparison with the central government, nonfinancial public insti- tutions, such as utilities and other service providers, make little infrastruc- ture investment (at most 20 percent of total capital investment) in both absolute and relative terms. This spending pattern reflects government control of some of the main sources of investment finance, be they roy- alty payments (in resource-rich countries) or external development funds (in fragile states and other low-income countries). It also reflects, to some extent, SOEs' limited capability to fund their capital investments through user fees. 222 Table 8.4 Public Infrastructure Spending by Institution in the WSS Sector, 2001­05 Share of GDP (%) US$ (million per year) OPEX CAPEX OPEX CAPEX On-budget Off-budget On-budget Off-budget On-budget Off-budget On-budget Off-budget Sub-Saharan Africa 0.35 0.14 0.17 0.03 2,216 896 1,073 180 Low-Income, fragile 0.04 0.29 0.08 0.00 16 111 30 0 Low-Income, nonfragile 0.15 0.13 0.16 0.06 164 143 176 67 Middle-income 0.62 0.18 0.10 0.02 1,691 494 275 49 Resource-rich 0.03 0.05 0.30 0.02 68 121 663 54 Source: Authors' calculations. Note: CAPEX = capital expenditure, GDP = gross domestic product, OPEX = operating expenditure. Bridging the Funding Gap 223 Table 8.5 Average Budget Variation Ratios for Capital Spending Overall Water supply and infrastructure sanitation sector Sub-Saharan Africa 75 66 Low-income, fragile -- -- Low-income, nonfragile 76 72 Middle-income 78 66 Resource-rich 65 43 Sources: Foster and Briceño-Garmendia 2009, adapted from Briceño-Garmendia, Smits, and Foster 2008. Note: Budget variation ratio is defined as executed budget divided by allocated budget. Based on annualized averages for 2001­06. -- = not available. Figure 8.3 Split Investment Responsibilities between Governments and Public Enterprises 0.9 0.8 0.7 share of GDP, % 0.6 0.5 0.4 0.3 0.2 0.1 0 ou i c a h le le e a le e le ch ric m m ric gi gi gi gi -ri r co co Af Af ra fra fra ra e- ce in in nf nf rc n n e, e, ur e- e- ra ra no no m m so dl dl ha ha s co co e, e, re re id id Sa Sa m m -in -in m m co co b- b- w w Su Su -in -in lo lo w w lo lo OPEX CAPEX off-budget on-budget Sources: Briceño-Garmendia, Smits, and Foster 2008; Foster and Briceño-Garmendia 2009. Note: Based on annualized averages for 2001­06. Averages weighted by country GDP. In Sub-Saharan Africa, SOE spending (off-budget) in O&M accounts at most for 30 percent of total spending on this item. The SOEs are essentially asset administrators. Interestingly, in fragile states almost 90 percent of O&M expenses are financed by nonfinancial public insti- tutions, whereas in middle-income countries, approximately 80 percent of the spending on O&M is in the budget. Inefficiencies within the public expenditure management systems are particularly detrimental because central governments are such major 224 Africa's Water and Sanitation Infrastructure players in capital investment and O&M relative to nonfinancial public institutions. A key issue is that central governments face significant prob- lems in executing their infrastructure capital budgets. African countries are, on average, unable to spend more than one-quarter of their WSS cap- ital budgets. In particular, resource-rich countries executed less than 45 percent of their budgets. The poor timing of project appraisals and late releases of budgeted funds because of procurement problems often pre- vent the use of resources within the budget cycle. Delays affecting in-year fund releases are also associated with poor project preparation, leading to changes in the terms agreed on with contractors in the original contract (deadlines, technical specifications, budgets, costs, and so on). In other cases, cash is reallocated to nondiscretionary spending driven by political or social pressures. Compared with the other infrastructure sectors, the WSS sector is the worst offender of unused budget allocations, in particular in resource-rich countries, where governments are able to spend barely 66 percent of budget allocations (table 8.5). Even after Efficiency Savings, a Persistent Funding Gap Inefficiencies of various kinds total an estimated $2.9 billion a year (0.5 percent of GDP; table 8.6). In absolute terms, the gains can be maximized for higher-income countries so that they contribute about 0.4 percent of GDP. In relative terms, the low-income fragile countries can leverage the most from exploiting the efficiency gains, amounting to 1.2 percent of GDP. Three opportunities can be identified for efficiency gains. First, raising user charges closer to cost-recovery levels would provide more efficient price signals and help capture lost revenue of about $1.5 billion per year. Second, reducing utilities' operating inefficiencies would prevent waste of significant resources, support healthier utilities, and improve service quality, leading to savings of about $1.3 billion per year. Third, improv- ing budget-execution rates would increase the potential of fully using resources allocated to public investment by about $0.2 billion per year. If the bottlenecks in capital execution could be resolved, countries could, on average, increase their capital spending by 4 percent without any increase in current budget allocations. For middle-income countries, an additional potential efficiency gain comes from reallocating $0.3 billion of existing spending to those subsectors in greatest need. This tactic would generate the highest economic returns, which would increase the Table 8.6 Potential Gains from Greater Efficiency GDP share (%) US$ (million per year) Operational inefficiencies Operational inefficiencies Total Tariff Total Tariff Labor Under- operational Capital cost Labor Under- operational Capital cost inefficiencies Losses collection inefficiencies execution recovery Total inefficiencies Losses collection inefficiencies execution recovery Total Sub-Saharan Africa 0.06 0.07 0.07 0.20 0.03 0.23 0.45 375 425 458 1,259 168 1,450 2,877 Low-income, fragile 0.04 0.17 0.06 0.28 0.02 0.93 1.23 17 65 25 106 6 358 471 Low-income, nonfragile 0.08 0.10 0.06 0.24 0.03 0.35 0.62 87 111 67 265 39 381 685 Middle-income 0.03 0.06 0.10 0.18 0.00 0.20 0.38 68 150 274 492 8 537 1,037 Resource-rich -- 0.05 0.03 0.08 0.06 0.10 0.23 -- 103 69 172 137 214 522 Source: AICD, adapted from Briceño-Garmendia, Smits, and Foster 2008. Note: Based on annualized averages for 2001­06. Averages weighted by country GDP. Figures are extrapolations based on the 24-country sample covered in AICD Phase 1, and they are lower bounds because inefficiencies might be higher as reported in the table due to data constraints. Totals may not add exactly because of rounding errors. -- = not available, GDP = gross domestic product. 225 226 Africa's Water and Sanitation Infrastructure impact of the current budget envelope on covering needs and raise the value for money of public funds. At the country level, Madagascar and Mali have the highest potential gains as a percentage of GDP (1.5 percent), results that would stem in particular from tackling the underpricing of tariffs (around 80 percent of the total gains; figure 8.4). Malawi is close to Madagascar and Mali in its level of inefficiencies (almost 0.9 percent of GDP), but the gains from resolving tariffs below cost-recovery levels account for around 40 percent of the total gains, whereas the operational inefficiencies account for about 50 percent of the potential gains. Nigeria has the lowest potential effi- ciency gains as a percentage of GDP (0.14 percent). Even if all the efficiency gains are internalized, a funding gap remains. Existing spending and potential efficiency gains can be calculated from estimated spending needs to gauge the extent of the financial shortfall. Africa would still face an annual funding gap of $11.9 billion a year, or 1.8 percent of GDP, to meet the MDG for WSS (figure 8.5). The smallest funding gap is found in middle-income countries where the highest inefficiencies are present. After tackling the ineffi- ciencies, middle-income countries would have a negligible funding gap of $0.3 billion. In fact, for these countries, potential exists for realloca- tion of resources of $0.2 billion, which can be swung from O&M to capital expenditure or transferred to some other infrastructure sector. The largest funding gap remains in low-income countries (nonfragile), representing about half of the total funding gap for Sub-Saharan Africa ($5.3 billion; table 8.7). The net annual funding gap represents 9.4 percent of the GDP of frag- ile states and less than 0.1 percent of the GDP of middle-income coun- tries. The gap between the low-income (nonfragile) and resource-rich countries is 4.8 percent and 1.8 percent of GDP, respectively (table 8.8). Although the infrastructure funding gap is primarily for capital invest- ment ($8.6 billion), a shortfall of almost one-fourth also exists for O&M (table 8.8). In the aggregate, Africa needs to increase water infrastructure capital investment by 1.3 percent of GDP; low-income, nonfragile coun- tries need to invest an additional 3.3 percent, and fragile states an addi- tional 6.8 percent. The shares of GDP for middle-income countries and resource-rich countries are below the African share of GDP (0.1 percent and 1.2 percent, respectively). The remainder of the infrastructure fund- ing gap ($3.2 billion) relates to O&M needs and is approximately evenly distributed across fragile states, low-income countries, and resource-rich countries. Middle-income countries do not face an O&M funding gap. Bridging the Funding Gap 227 Figure 8.4 Potential Efficiency Gains from Different Sources Sub-Saharan Africa low-income, fragile low-income, nonfragile middle-income resource-rich Madagascar Mali Malawi Zambia Côte d'lvoire Senegal Mozambique Tanzania Lesotho Kenya Cape Verde Congo, Rep. Niger Rwanda Sudan South Africa Burkina Faso Namibia Ethiopia Botswana Gabon Benin Nigeria 0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 % GDP operational inefficiencies undercapital execution underpricing Source: Authors' compilation. Note: GDP = gross domestic product. Closing the $11.8 billion WSS infrastructure funding gap depends in part on raising additional funds, but it may also require taking more time to attain targets or using lower-cost technologies, such as standposts and traditional latrines. 228 Africa's Water and Sanitation Infrastructure Figure 8.5 Water Infrastructure Funding Gap a. Percentage of GDP Sub-Saharan Africa 1.8 low-income, fragile 9.4 low-income, nonfragile 4.8 resource-rich 1.8 middle-income 0.1 Madagascar 9.2 Kenya 9.0 Ethiopia 6.4 Niger 5.6 Tanzania 5.3 Malawi 5.1 Benin 4.1 Zambia 3.5 Mozambique 3.4 Mali 3.1 Côte d'Ivoire 3.0 Uganda 2.6 Ghana 2.5 Rwanda 2.4 Congo, Rep. 2.3 Senegal 1.9 Namibia 1.9 Burkina Faso 1.8 Cameroon 1.8 Chad 1.8 Lesotho 1.3 Nigeria 1.3 Cape Verde 0.7 South Africa 0.2 0 5.0 10.0 % GDP (continued next page) Bridging the Funding Gap 229 Figure 8.5 (continued) b. US$ (million per year) Sub-Saharan Africa 11,873 low-income, fragile 3,620 low-income, nonfragile 5,285 resource-rich 4,089 middle-income 312 Kenya 1,678 Nigeria 1,414 Ethiopia 792 Tanzania 749 South Africa 604 Côte d'Ivoire 486 Madagascar 466 Cameroon 305 Ghana 269 Zambia 255 Uganda 228 Mozambique 226 Niger 186 Benin 177 Senegal 168 Mali 165 Malawi 144 Congo, Rep. 143 Namibia 119 Chad 105 Burkina Faso 100 Rwanda 58 Lesotho 19 Cape Verde 7 0 5,000 10,000 15,000 US$ (million per year) Source: Briceño-Garmendia, Smits, and Foster 2008. Note: GDP = gross domestic product. 230 Table 8.7 Funding Gap (US$ million per year) Gain from Sources of inefficiency Spending eliminating Underexecution Operating Funding gap Total needs traced to needs inefficiencies of budget inefficiencies Underpricing or surplus Sub-Saharan Africa ­22,640 7,890 2,877 168 1,259 1,450 ­11,873 Low-income, fragile ­4,531 441 471 6 106 358 ­3,620 Low-income, nonfragile ­7,810 1,840 685 39 265 381 ­5,285 Middle-income ­3,987 2,637 1,037 8 492 537 ­312 Resource-rich ­6,364 1,753 522 137 172 214 ­4,089 Source: Briceño-Garmendia, Smits, and Foster 2008. Bridging the Funding Gap 231 Table 8.8 Size and Composition of the Annual Funding Gap by O&M and Capital Expenditure Share of GDP (%) US$ (million per year) Total Total CAPEX funding CAPEX funding gap O&M gap gap gap O&M gap gap Sub-Saharan Africa 1.3 0.5 1.8 8,648 3,225 11,873 Low-income, fragile 6.8 2.6 9.4 2,627 993 3,620 Low-income, nonfragile 3.3 1.5 4.8 3,673 1,612 5,285 Middle-income 0.1 0 0.1 312 -- 312 Resource-rich 1.2 0.6 1.8 2,696 1,393 4,089 Source: Briceño-Garmendia, Smits, and Foster 2008. Note: CAPEX = capital expenditure. Limited Scope for Raising Additional Finance Limited financing sources are available, and the global financial crisis is likely to affect some of them adversely. Domestic public finance is one of the main sources of funding today, but it presents little scope for an increase, except possibly in countries that enjoy natural resource wind- falls. Another point to consider is that household finance is one of the most important sources of funding today for capital investments in African infrastructure (0.3 percent of GDP; see table 8.2), mainly of san- itation facilities. It is very likely that this source will be affected by the financial crisis, but given that other forms of private participation have not been very important in the WSS sector, no concern is found on that score about the negative impacts of the downturn on global markets. In addition, ODA is an important player in financing capital investments in the sector (0.2 percent of GDP; see table 8.2). It has grown substantially in recent years, in line with political pledges, but this assistance could slow down if countercyclical assistance were put in place. Finally, local capital markets have so far contributed little to WSS sector finance outside South Africa, and there is not much expectation that they could eventu- ally assume greater role in some of the region's larger economies. Little Scope for Domestic Finance A key question is the extent to which countries may be willing to allocate additional fiscal resources to the WSS sectors. In the run-up to the cur- rent financial crisis, the fiscal situation in Sub-Saharan Africa was favor- able. Rapid economic growth averaged 4 percent a year from 2001 to 2005, which translated to increased domestic fiscal revenue of just over 232 Africa's Water and Sanitation Infrastructure 3 percent of GDP on average. In resource-rich countries, burgeoning resource royalties added 7.7 percent of GDP to the public budget. In low- income countries, substantial debt relief increased external grants by almost 2 percent of GDP. Surprisingly little additional resources were available during the recent growth surge allocated to infrastructure (table 8.9). The most extreme case is that of the resource-rich countries, particularly Nigeria. Huge debt repayments more than fully absorbed the fiscal windfalls in these coun- tries. As a result, budgetary spending actually contracted by 3.7 percent of GDP. Infrastructure investment bore much of that contraction and fell by almost 1.5 percent of GDP. In middle-income countries, budgetary spending increased by almost 4.1 percent of GDP, but the effect on infra- structure spending was almost negligible and the additional resources went primarily to current social sector spending. Only in the low-income countries did the overall increases in budgetary expenditure have some effect on infrastructure spending. Even there, however, the effect was fairly modest and confined to capital spending. The low-income countries (nonfragile) have allocated 30 percent of the budgetary increase to infra- structure investments. The fragile states, despite seeing their overall bud- getary expenditures increase by about 3.9 percent of GDP, have allocated only 6 percent of the increase to infrastructure. Compared with other developing regions, public financing capabilities in Sub-Saharan Africa are characterized by weak tax revenue collection. Domestic revenue generation of around 23 percent of GDP trails averages for other developing countries and is lowest for low-income countries (less than 15 percent of GDP a year). Despite the high growth rates in the past decade, domestically raised revenue grew by less than 1.2 percent of GDP, Table 8.9 Net Change in Central Government Budgets, by Economic Use (% of GDP) Sub- Low- Low- Saharan Middle- Resource- income, income, Use Africa income rich nonfragile fragile Net expenditure budget 1.89 4.08 ­3.73 1.69 3.85 Current infrastructure spending as a share of expenditures 0 0.02 0.03 0 0.09 Capital infrastructure spending as a share of expenditures ­0.14 0.04 ­1.46 0.54 0.22 Sources: Foster and Briceño-Garmendia, 2009, adapted from Briceño-Garmendia, Smits, and Foster 2008. Note: Based on annualized averages for 2001­06. Averages weighted by country GDP. Totals are extrapolations based on the 24-country sample as covered in AICD Phase 1. GDP = gross domestic product. Bridging the Funding Gap 233 suggesting increasing domestic revenue from current levels would require undertaking challenging institutional reforms to increase the effectiveness of revenue collection and broaden the tax base. Without such reforms, domestic revenue generation will remain weak. The borrowing capacity from domestic and external sources is also limited. Domestic borrowing is often very expensive, with interest rates far exceeding those on concessional external loans. Particularly for the poorest countries, the scarcity of private domestic savings means that public domestic borrowing tends to precipitate sharp increases in interest rates, building a vicious circle. For many Sub-Saharan African countries, the share of debt service to GDP is more than 6 percent. The global financial crisis can be expected to reduce fiscal receipts because of lower taxes, royalties, and user charge taxes--Africa is not exempt from its impact. Growth projections for the coming years have been revised downward from 5.1 percent to 3.5 percent, which will reduce tax revenue and likely depress the demand and willingness to pay for infrastructure services. Commodity prices have fallen to levels of the early 2000s. The effect on royalty revenue, however, will depend on the savings regime in each country. Various oil producers have been saving royalty revenue in excess of $60 a barrel, so the current downturn will affect sav- ings accounts more than budgets. Overall, the adverse situation created by the global financial crisis will put substantial pressure on public sector budgets. In addition, many African countries are devaluing their currency, reducing the purchasing power of domestic resources. According to recent global experience, fiscal adjustment episodes tend to fall disproportionately on public investment in general and WSS infra- structure in particular. Experience from earlier crises in East Asia and Latin America indicates that infrastructure spending is especially vulner- able to budget cutbacks during crisis periods. Based on averages for eight Latin American countries, cuts in infrastructure investment amounted to about 40 percent of the observed fiscal adjustment between the early 1980s and late 1990s (Calderón and Servén 2004). This reduction was remarkable because public infrastructure investment already represented less than 25 percent of overall public expenditure in Latin American countries. These infrastructure investment cuts were later identified as the underlying problem holding back economic growth in the whole region during the 2000s. Similar patterns were observed in East Asia dur- ing the financial crisis of the mid-1990s. For example, Indonesia's total public investment in infrastructure dropped from 6 to 7 percent of GDP during the period from 1995 to 1997 to 2 percent in 2000. Given recent 234 Africa's Water and Sanitation Infrastructure spending patterns, there is every reason to expect that in Africa changes in the overall budget envelope will affect infrastructure investment in a similar pro-cyclical manner. Self- or Household Finance Self- or household finance has been the main source of external financing in the sanitation sector, representing almost half of the total spending in capital investment ($2.1 billion, or 0.3 percent of the African GDP; see table 8.2). These figures are largely driven by private investments at the household level in people's own sanitation facilities. Households in Africa's resource-rich countries and low-income countries (nonfragile) invest the largest volume of funds in absolute terms (almost half of the total household or self-finance for the region). Households in fragile states invest the least in absolute terms (less than $0.2 billion). Nigeria has the highest amount of the total household investment in sanitation (equivalent to $295 million), which accounts for 0.3 percent of its GDP. The financial crisis is likely to affect households' willingness to invest in new WSS facilities or improvements given the potential reductions in household income, although it is hard to make exact predictions. Official Development Assistance Commitments of ODA from OECD donors to water infrastructure in Sub-Saharan Africa have increased from $0.8 to $1.2 billion between 1995 and 2007. Across countries, The Gambia captures the highest level of ODA commitments as a percentage of GDP (1.5 percent of GDP), fol- lowed by Benin and Burundi (1.4 percent of GDP) (figure 8.6). In absolute terms, Tanzania and Nigeria receive the largest shares of ODA commitments in the region (around 20 percent of this source). A significant lag occurs between ODA commitments and their dis- bursement, which suggests that disbursements should continue to increase in the coming years. The commitments just reported are signif- icantly higher than the estimated ODA disbursements of $1.2 billion, or 0.2 percent of Sub-Saharan African GDP (see table 8.2). This gap reflects delays typically associated with project implementation. Because ODA is channeled through the government budget, the execution of funds faces some of the same problems affecting domestically financed public investment, including procurement delays and the capacity of low-income countries to execute funds. Divergences between donor and country financial systems, as well as unpredictability in the release of funds, may further hinder the disbursement of donor resources. Bearing this in mind, as long as all commitments up to 2007 are fully honored, Bridging the Funding Gap 235 Figure 8.6 Aid Commitments for Water Supply and Sanitation as a Percentage of GDP, 2001­05 Gambia, The 1.5 Burundi 1.4 Benin 1.4 Tanzania 1.0 Lesotho 1.0 Ghana 0.9 Cape Verde 0.9 Mozambique 0.8 Ethiopia 0.7 Liberia 0.7 Rwanda 0.7 Burkina Faso 0.7 Congo, Dem. Rep. 0.7 Niger 0.7 Sierra Leone 0.6 Mali 0.6 Eritrea 0.6 Zambia 0.6 Mauritania 0.6 Uganda 0.5 Malawi 0.5 Kenya 0.5 Chad 0.5 Guinea-Bissau 0.5 Senegal 0.4 Guinea 0.4 Madagascar 0.4 Comoros 0.3 Mauritius 0.1 São Tomé and Principe 0.1 Cameroon 0.1 Gabon 0.1 Central African Republic 0.1 Seychelles 0.1 Nigeria 0.1 Namibia 0.1 Togo 0.1 Angola 0.1 Swaziland < 0.1 Sudan < 0.1 Zimbabwe < 0.1 South Africa < 0.1 Equatorial Guinea < 0.1 Botswana < 0.1 Côte d'Ivoire < 0.1 Congo, Rep. < 0.1 0 0.4 0.8 1.2 1.6 % GDP Source: Briceño-Garmendia, Smits, and Foster 2008. Note: GDP = gross domestic product. 236 Africa's Water and Sanitation Infrastructure ODA disbursements could be expected to rise significantly over the next few years (IMF 2009; World Economic Outlook 2008). ODA was set to increase further before the crisis, but prospects no longer look as promising. The three multilateral agencies--the African Development Bank, the European Commission, and the World Bank-- secured record replenishments for their concessional funding windows for the three to four years beginning in 2008. In principle, funding alloca- tions to African WSS sectors totaling $1.2 billion a year (see table 8.2) could come from the multilateral agencies alone in the near future. In practice, however, the crisis may divert multilateral resources from infra- structure projects toward emergency fiscal support. Bilateral support, based on annual budget determinations, may be more sensitive to the fiscal squeeze in the OECD countries, and some decline can be anticipated. Historical trends suggest that ODA has tended to be pro-cyclical rather than countercyclical (IMF 2009; ODI 2009; UBS Investment Research 2008; World Economic Outlook 2008; and references cited therein). Other Financial Players: The Private Sector and Non-OECD Financiers Most of the private investment commitments in the water sector ($10.7 million) come from foreign participation and predominantly target Sudan ($6.7 million, or 65 percent of the total). This number may be signifi- cantly underestimated, however, given the difficulty in capturing data on private investment coming through nongovernmental organizations and foundations when the resources do not enter the public treasuries. Private capital flows, in particular, are likely to be affected by the global financial crisis. In the aftermath of the Asian financial crisis, pri- vate participation in developing countries fell by about one-half over a period of five years, following its peak in 1997. Existing transactions are also coming under stress as they encounter difficulties refinancing short- and medium-term debt. Given the very limited volume of private participation in the WSS sector in Africa, however, the downside risk is also limited. Non-OECD countries financed less than $0.2 billion worth of African WSS infrastructure annually between 2001 and 2005 or 0.03 percent of GDP (see table 8.2). Non-OECD financiers have been active primarily in resource-rich countries, mainly oil-exporting countries (Angola, Nigeria, and Sudan), which receive half of the total resources coming from non-OECD financiers ($80 million). Just over one-third of the resources, or $55 million, has gone to low-income countries (nonfragile). Bridging the Funding Gap 237 These financiers' contribution to middle-income countries is very small ($8 million per year). China's official economic assistance for infrastructure project quadru- pled between 2001 and 2005 and reached more than 35 Sub-Saharan African countries. Most of the inflows went to resource-rich countries; in some cases, they made use of barter arrangements under the "Angola mode."1 The WSS sector accounts for a relatively small share of China's assistance (2 percent, or $0.14 billion) when compared with the commit- ments made to other sectors. Most of the projects are focused on meet- ing immediate social needs directly related to water supply, such as smaller dams in Cape Verde and Mozambique. How the current economic downturn will affect non-OECD finance is difficult to predict because of the relatively recent nature of these capital inflows. Local Sources of Finance Local capital markets are a major source of WSS sector infrastructure finance in southern Africa and the resource-rich countries, but not yet elsewhere (table 8.10); they account for about $3 billion. Local infra- structure finance consists primarily of commercial bank lending, some Table 8.10 Outstanding Financing Stock for Water and Sanitation Infrastructure, as of 2006 Share of Share of total infra- Outstanding financing Bank Corporate Equity total structure for infrastructure loans bonds issues Total stock (%) stock (%) Resource-rich 1,119 -- 2 1,121 37 43 Low-income, nonfragile 350 -- -- 350 11 5 Low-income, fragile 69 -- 11 80 3 17 Middle-income (excluding South Africa) 103 -- -- 103 3 19 South Africa 1,264 -- 130 1,393 46 2 Total 2,905 -- 142 3,047 100 4 Share of total stock (%) 95 -- 5 100 Share of total infrastructure stock (%) 4 -- 0 4 Source: Adapted from Irving and Manroth 2009. Note: Bank loans combine transport, communication, energy, and water for the Democratic Republic of Congo, Ghana, Lesotho, and Zambia. Bank loans combine electricity, water, and gas/public utilities for Benin, Burkina Faso, Cape Verde, Côte d'Ivoire, Ethiopia, Malawi, Namibia, Niger, Nigeria, Rwanda, Senegal, South Africa, Uganda, and Zambia. -- = not available. 238 Africa's Water and Sanitation Infrastructure corporate bond and stock exchange issues, and a nascent entry of institu- tional investors. These markets remain underdeveloped, shallow, and small, in particular for financing the WSS sector in fragile states. Long- term financing with maturities commensurate with infrastructure proj- ects is scarce.2 The capacity of local banking systems remains too small and constrained by structural impediments to finance infrastructure. Most countries' banks have significant asset-liability maturity mismatches for infrastructure financing. Bank deposits and other liabilities still have largely short-term tenors. More potential may exist for syndicated lend- ing with local bank participation, though the increase in new loans over the 2000­06 period occurred in a favorable external financing environ- ment. The African banking system did not feel the effects of the global financial crisis at first, but the crisis slowly but surely affected financial systems around the region and added to the already enormous challenge of developing local financial markets. Costs of Capital from Different Sources The various sources of infrastructure finance differ greatly in their associ- ated costs of capital (figure 8.7). For public funds, raising taxes is not a costless exercise. Each dollar raised and spent by a Sub-Saharan African government has a social value premium (or marginal cost of public funds) of almost 20 percent. That premium captures the incidence of that tax on Figure 8.7 Costs of Capital by Funding Source, 2001­05 public funds 1.17 India 0.91 China 0.87 Arabs 0.65 ODA 0.51 IDA 0.33 grants 0 0 0.2 0.4 0.6 0.8 1 1.2 1.4 cost of raising US$1 of financing, US$ Sources: Average marginal cost of public funds as estimated by Warlters and Auriol 2005; cost of equity for private sector as in Estache and Pinglo 2004 and Sirtaine and others 2005; authors' calculations. Note: IDA = International Development Association, ODA = official development assistance. Bridging the Funding Gap 239 the society's welfare (caused by changes in consumption patterns and administrative costs, among other things).3 To allow comparisons across financing sources, this study standardized the financial terms as the pres- ent value of a dollar raised through each of the different sources. In doing so, it recognized that all loans must ultimately be repaid with tax dollars, each of which attracts the 20 percent cost premium. Wide variation exists in lending terms. The most concessional International Development Association (IDA) loans charge zero interest (0.75 percent service charge) with 10 years of grace. India, China, and the Gulf States and Arab funds charge 4 percent, 3.6 percent, and 1.5 percent interest, respectively, and grant four years of grace. The cost of non-OECD finance is somewhere between that of public funds and ODA. The subsidy factor for Indian and Chinese funds is about 25 percent, and for the Arab funds, 50 percent. Official development assistance typically provides a subsidy factor of 60 percent, rising to 75 percent for IDA resources. In addition to the cost of capital, the different sources of finance differ in the transaction costs associated with their use, which may offset or accentuate some of the differences. Promising Ways to Increase Funds Given this setting, what are the best ways to increase availability of funds for water infrastructure development? The place to start is clearly to get the most from existing budget envelopes, which can provide up to $2.9 billion a year of additional resources internally if inefficiencies are tackled, equivalent to one-fourth of the total funding gap. For middle-income countries, reducing inefficiencies would imply not only completely closing the funding gap, but also achieving total positive net savings. In particular, for Botswana this would in and of itself be enough to close the funding gap. In the case of resource-rich and low-income countries (fragile and nonfragile), reducing inefficiencies would con- tribute to reducing the gap by more than 10 percent. Beyond that, a substantial funding gap still remains. Before the finan- cial crisis, the prospects for reducing--if not closing--this gap were rea- sonably good. Resource royalties were at record highs, and all sources of external finance were buoyant and promising further growth. With the onset of the global financial crisis, that situation has changed significantly and in ways that are not yet entirely foreseeable. The possibility exists across the board that all sources of WSS finance in Africa may fall rather than increase, further widening the funding gap. Only resource-rich 240 Africa's Water and Sanitation Infrastructure countries have the potential to use natural resource savings accounts to provide a source of financing for infrastructure, but only if macroeco- nomic conditions allow. The international community's agreement on a major stimulus package for Africa, with a focus on infrastructure as part of the effort to rekindle economic growth and safeguard employment, is one of the few options for reversing the overall situation. Other Ways to Reach the MDG Except for middle-income countries, all other countries face a substan- tial funding gap even if all the existing sources of funds--including efficiency gains--are tapped. What other options do these countries have? Realistically, they need to either defer the attainment of the infrastructure targets proposed here or try to achieve them by using lower-cost technologies. Taking More Time Extending the time horizon for the achievement of these goals could make the targets more affordable. What if countries delay the upper bound of MDG attainment by 5 to 10 years without increasing existing resource envelopes? One caveat to this analysis must be taken into account. The spending needs presented in chapter 7 are based on nonstandardized unit costs. Many of the variables are exogenous at least in the short run, which should guarantee that assumptions made on these variables remain valid if the analysis time horizon is not extended too long into the future, but this may not be the case for density. Africa is urbanizing rapidly and is expected to become predominantly urban by 2020. Although urban sprawl is more common in Africa than elsewhere, it is likely that 15 years from now average urban densities will be greater than those assumed here. More important, the density data set used in this analysis dates to 2000, because it is one of the very few available that observed city pop- ulations and densities at the same point in time. As density increases, net- work infrastructure unit costs decrease, as does the investment required on new infrastructure. Therefore, the results presented here should be taken as an upper bound of the overall spending needs that countries would face to reach the MDG by either 2020 or 2025. As the time horizon to achieve the water and sanitation MDG is extended by five years, annual spending needs for Africa as a whole decrease from 3.5 percent to 3.2 percent of GDP per year (figure 8.8). Bridging the Funding Gap 241 Figure 8.8 Spending Needs by Country Type under Different Time Horizons (base scenario, % of GDP) 14 12 10 8 % GDP 6 4 2 0 2015 2020 2025 Sub-Saharan Africa resource-rich middle-income low-income, fragile low-income, nonfragile Source: Authors' calculations. Note: GDP = gross domestic product. The same trend is observed across all income groups. Savings are not out- standing--annual needs decrease by less than 10 percent--but are still beneficial, especially for low-income, fragile countries. If the time horizon is extended by an additional five years, to 2025, the decreasing trend continues, but at a falling rate. The overall annual burden for Africa lessens only marginally, as does the burden for the resource-rich and low-income, nonfragile countries. Fragile countries would save the most, yet just 1 percent of GDP per year, with overall needs decreasing from 12 percent to 11 percent of GDP. Middle-income countries would also save if they could take 10 more years to reach tar- gets, yet they would be better off by postponing the achievement of the water and sanitation MDG by 5 rather than 10 more years. Zeroing in on the composition of needs reveals that if countries are allowed to take more time, the annual spending to be allocated to new infrastructure decreases (figure 8.9, panel a). Although the overall num- ber of new customers to be served rises with time as the population grows, fewer new customers per year would need to be accommodated. Similarly, as the time horizon is extended, annual rehabilitation costs decrease (figure 8.9, panel b). In fact, this analysis considers only the 242 Africa's Water and Sanitation Infrastructure Figure 8.9 Annual Spending Needs over Different Time Horizons, by Country Type a. Expansion needs 3 2 2 % GDP 1 1 0 2015 2020 2025 b. Rehabilitation needs 3 3 2 % GDP 2 1 1 0 2015 2020 2025 c. O&M needs 3 3 2 % GDP 2 1 1 0 2015 2020 2025 Sub-Saharan Africa resource-rich middle-income low-income, fragile low-income, nonfragile Source: Authors' calculations. Bridging the Funding Gap 243 rehabilitation backlog related to existing assets and assumes that no signif- icant rehabilitation will need to occur on new assets, supposing that these are adequately maintained. It may be argued that some on-site facilities' life span is less than 20 years, which implies that some rehabilitation should take place at some point between 2006 and 2025. The assump- tions made here, however, mainly reflect the fact that the largest rehabil- itation needs originate from assets--markedly network assets--with a life span more than 20 years. Conversely, O&M needs increase as the time horizon is extended (figure 8.9, panel c). Overall, countries will need to add a larger number of new customers, and therefore, more assets will need to be maintained. Sub-Saharan Africa would be able to reach the MDG by 2027 if it were to tackle its utilities inefficiencies given the current levels of spend- ing. If resource-rich and low-income countries (nonfragile) spread the spending needs over 26 to 32 years rather than 10 years, they could achieve the proposed targets within the existing spending envelopes. Fragile states would need more than 57 years to achieve the MDG tar- gets if currents levels of spending were not changed (table 8.11). Middle- income countries could achieve the MDG's target in 21 years given the current level of spending, but this conclusion assumes they have first fully captured efficiency gains. Without such efficiency gains, the targets could not be met even over 30 years without increasing spending above current levels. Using Lower-Cost Technologies Using alternative lower-cost technologies to provide water and sanitation services to new customers appears to respond to both affordability and efficiency considerations. A direct water connection is regarded as the modality at the top of the water ladder for safety and time-saving reasons; similarly, septic tanks are more likely to deliver health benefits than are improved or traditional latrines. For these reasons, higher-level services Table 8.11 Time Needed to Meet the MDG Targets with Today's Budget Envelopes Sub- Low- Low- Years to reach MDG target Saharan Middle- Resource- income, income, (counting from 2006) Africa income rich nonfragile fragile Existing spending plus efficiency gains 21 10 26 32 57 Existing spending only 33 28 94 69 104 Source: Authors' calculations. Note: MDG = Millennium Development Goal. 244 Africa's Water and Sanitation Infrastructure attract the attention of policy makers, but they come at a substantially higher cost. Unit cost analysis reveals that the cost of network expansion is highly sensitive to density. This is especially detrimental to African loca- tions, generally less dense than their counterparts in other regions. When the ultimate goal is expanding access and economies of scale are not pos- sible, nonnetwork, lower-cost technological alternatives might offer a much more efficient solution. Moreover, in some cases--and markedly within the range of on-site sanitation alternatives once the basic level of sanitary protection is reached--higher costs are associated with diminish- ing returns in terms of safety and health. This analysis assumes a pragmatic scenario (as opposed to the base scenario assumed in chapter 7) assuming that all new customers are served with lower-cost alternatives: standposts and improved latrines in urban areas and protected wells and boreholes and traditional latrines in rural areas. The overall spending needs in this scenario are presented in figure 8.10. In 2015, the share of the population enjoying higher-quality serv- ices, such as a direct water connection, a sewer connection, or a septic tank, will be lower than it is today. Under this assumption, however, overall spending needs for Africa drop from 3.5 percent to 2.3 percent of GDP per year (table 8.12). Under the pragmatic scenario, the bill looks substantially more manage- able across all countries, with the majority of them required to allocate no more than 5 percent of GDP annually to water and sanitation. Equatorial Guinea, Gabon, Swaziland, South Africa, Botswana, Angola, Mauritius, Cape Verde, Nigeria, Democratic Republic of Congo, Cameroon, Chad, Senegal, Lesotho, and Burkina Faso report, in increasing order, needs below 3 percent. Zimbabwe, The Gambia, Liberia, and Togo still stand as outliers with spending needs over 10 percent of their GDP. Overall spending needs would become substantially less prohibitive for low-income countries were the pragmatic scenario to be adopted. Compared with the base scenario, the use of lower-cost technologies makes needs drop considerably for fragile states, from 12 percent to 7 per- cent of GDP, and to a lower extent for low-income, nonfragile countries, from 7 percent to 4 percent of GDP. Nevertheless, the bill is still high, especially for fragile states, and largely in excess of current spending (figure 8.11). Conversely, under the pragmatic scenario, middle-income countries would need to allocate only 1 percent of their GDP to water and sanitation, and resource-rich countries, 2 percent. Although a pragmatic scenario better matches the capacity of most low-income countries, a high-end scenario based on the use of top-quality Bridging the Funding Gap 245 Figure 8.10 Spending Needs by Country under Different Level-of- Service Assumptions Sub-Saharan Africa low-income, fragile resource-rich Togo Sierra Leone Eritrea Gambia, The Zimbabwe Malawi Uganda Tanzania Benin Mali Zambia Mauritania Mozambique Congo Senegal Côte d' Ivoire Chad Mauritius Angola South Africa Equatorial Guinea 0 10 20 30 40 % GDP base pragmatic Source: Authors' calculations. 246 Table 8.12 Spending Needs to Meet the MDG Targets under Different Level-of-Service Scenarios Share of GDP (%) US$ (million per year) Pragmatic Base Pragmatic Base Water Sanitation Total Water Sanitation Total Water Sanitation Total Water Sanitation Total Sub-Saharan Africa 1.6 0.7 2.3 2.7 0.8 3.5 10,392 4,688 15,080 17,239 5,401 22,640 Resource-rich 1.3 0.7 2.1 2.1 0.7 2.9 2,963 1,636 4,599 4,718 1,646 6,364 Middle-income 0.7 0.3 1.1 1.0 0.5 1.5 2,023 862 2,885 2,733 1,243 3,976 Low-income, fragile 4.8 2.3 7.1 8.9 2.9 11.8 1,822 857 2,679 3,337 1,099 4,437 Low-income, nonfragile 3.2 1.2 4.4 5.8 1.3 7.1 3,560 1,322 4,881 6,410 1,400 7,810 Source: Authors' calculations. Note: In the base scenario, it is assumed that the relative prevalence of WSS supply modalities will remain constant between 2006 and 2015. In the pragmatic scenario, access to new customers is granted using low-cost technologies, which provide improved safe drinking water and sanitation. GDP = gross domestic product, MDG = Millennium Development Goal. Bridging the Funding Gap 247 Figure 8.11 Overall Spending Needs by Country Groups under Different Service Assumptions 20 18 16 14 current spending = 1.23 percent % GDP per year of GDP p.a. 12 10 8 6 4 2 0 a e ch le le ric m gi gi -ri co Af ra fra ce in nf n ur e, e- ra no m so dl ha co e, re id Sa m -in m b- co w Su -in lo w lo pragmatic base high-end Source: Authors' calculations. Note: GDP = gross domestic product, p.a. = per annum (every year). technologies could be considered for middle-income and resource-rich countries. If a high-end scenario were to be adopted, the bill would remain below 3 percent of GDP not only for middle-income countries such as South Africa, Botswana, and Swaziland, but also for low-income countries such as Equatorial Guinea, Gabon, and Angola. Resource-rich countries, however, would end up allocating more than 5 percent of their GDP to water and sanitation. Also, the high-end scenario does not appear feasible for low-income countries, fragile and nonfragile alike, which would be required to allocate more than 15 percent of GDP annually to water and sanitation. The availability of alternative lower-cost technologies has the potential to reduce the funding gap by more than 60 percent (table 8.13). This implies a reduced cost of meeting the MDG by almost 5 percent of GDP for fragile states, which represents reductions in the funding gap of more than 50 percent. Similarly, if the low-income countries (nonfragile) adopted a pragmatic scenario rather than a base scenario, savings would account for as much as 38 percent, or 2.7 percent of GDP, leading to 248 Africa's Water and Sanitation Infrastructure Table 8.13 Funding Gaps under Base and Pragmatic Scenarios Share of US$ (million Reduction GDP (%) per year) of the Base Pragmatic Base Pragmatic funding scenario scenario Savings scenario scenario Savings gap (%) Sub-Saharan Africa 1.8 0.7 1.2 11,873 4,313 7,560 64 Resource-rich 1.8 1.0 0.8 4,089 2,324 1,765 43 Middle-income 0.1 ­0.2 0.4 312 ­779 1,091 350 Low-income, fragile 9.4 4.8 4.7 3,620 1,862 1,758 49 Low-income, nonfragile 4.8 2.1 2.7 5,285 2,356 2,929 55 Source: Authors' calculations. Note: GDP = gross domestic product. reductions in the funding gap of more than 55 percent compared with the base scenario. For the middle-income countries, the savings would be at 0.4 percent of GDP, chiefly because these countries have high rates of access to network services whose expansion has a reduced marginal cost, which would lead the funding gap to disappear. Notes 1. Essentially, the Angola mode was devised to enable African nations to pay for infrastructure with natural resources. In a single transaction, China bundles development-type assistance with commercial-type trade finance. A Chinese resource company makes repayments in exchange for oil or mineral rights. The China Export-Import Bank acts as a broker, receives money for the sale, and pays the contractor for providing the infrastructure. This arrangement safeguards against currency inconvertibility, political instability, and expro- priation. 2. Because South Africa's financial markets are so much more developed than those of the other 23 focus countries, this section excludes South Africa. 3. The marginal cost of public funds measures the "change in welfare associated with raising an additional unit of tax revenue" (Warlters and Auriol 2005, 2). References Briceño-Garmendia, C., K. Smits, and V. Foster. 2008. "Financing Public Infrastructure in Sub-Saharan Africa: Patterns and Emerging Issues." AICD Background Paper 15, World Bank, Washington, DC. Bridging the Funding Gap 249 Calderón, César, and Luis Servén. 2004. "Trends in Infrastructure in Latin America, 1980-2001." Policy Research Working Paper 3401, World Bank, Washington, DC. Estache, Antonio, and Maria Elena Pinglo. 2004. "Are Returns to Private Infrastructure in Developing Countries Consistent with Risks since the Asian Crisis?" Policy Research Working Paper 3373, World Bank, Washington, DC. Foster, Vivien, and Cecilia Briceño-Garmendia, eds. 2009. Africa's Infrastructure: A Time for Transformation. Paris and Washington, DC: Agence Française de Développement and World Bank. Foster, Vivien, William Butterfield, Chuan Chen, and Nataliya Pushak. 2008. "Building Bridges: China's Growing Role as Infrastructure Financier for Sub- Saharan Africa." Trends and Policy Options 5. Public-Private Infrastructure Advisory Facility, World Bank, Washington, DC. IMF (International Monetary Fund). 2009. "The State of Public Finances: Outlook and Medium-Term Policies after the 2008 Crisis." IMF, Washington, DC. Irving, Jacqueline, and Astrid Manroth. 2009. "Local Sources of Financing for Infrastructure in Africa: A Cross-Country Analysis." Policy Research Working Paper 4878, World Bank, Washington, DC. ODI (Overseas Development Institute). 2009. A Development Charter for the G-20. London: ODI. PPIAF (Public-Private Infrastructure Advisory Facility). 2008. Private Participation in Infrastructure Project Database. http://ppi.worldbank.org/. Sirtaine, Sophie, Maria Elena Pinglo, J. Luis Guasch, and Vivien Foster. 2005. "How Profitable Are Infrastructure Concessions in Latin America? Empirical Evidence and Regulatory Implications." Trends and Policy Options 2, PPIAF, World Bank, Washington, DC. Warlters, Michael, and Emmanuelle Auriol. 2005. "The Marginal Cost of Public Funds in Africa." Policy Research Working Paper 3679, World Bank, Washington, DC. World Economic Outlook. 2008. "Estimating the Size of the European Stimulus Packages for 2009." International Monetary Fund, Washington, DC. UBS Investment Research. 2008. "Global Economic Perspectives: The Global Impact of Fiscal Policy." UBS Investment Research, London. CHAPTER 9 Policy Options for the Water and Sanitation Sectors Policy Options for the Water Sector The analyses of the water sector performed by the Africa Infrastructure Country Diagnostic reveal the following key areas for policy attention. The institutional reform agenda remains as relevant as before, even if its focus has shifted toward a more pluralistic view of public and private sector roles. The reform agenda also needs to move beyond utilities to encompass relevant government agencies and the whole public expen- diture framework that underpins, and too often hinders, sector invest- ment programs. Room for improvement can be found in cost recovery so that scarce subsidy resources are redirected to provide access among the poorest. For meeting the needs of the majority of people who do not enjoy access to household connections to piped water, greater thought needs to be given to making standposts effective sources of urban water supply and optimizing the use of small-scale independent providers. The burgeoning use of wells and boreholes for supply in urban areas demands policy mak- ers' attention; they must improve their understanding of this trend to develop suitable regulatory tools. However, Africa remains a predominantly rural continent with a pop- ulation of approximately 400 million people excluded from any form of 251 252 Africa's Water and Sanitation Infrastructure utility-provided water. This segment of the continent's population often depends on unsafe supply sources, such as surface water, in addition to wells and boreholes. The central challenge is to reduce reliance on surface water through a sustainable network of water access points, most typi- cally boreholes. Inadequate maintenance of rural water systems reflects both institu- tional weaknesses and inappropriate technology choices. In addition to weak institutional capacity, undermaintenance is worsened by inadequate attention to technology choices, low pump density, restrictive mainte- nance systems, and lack of a supply chain to adequately maintain com- plex machinery. This sector needs specialized attention through either improvements to the supply chain of spare parts for rural water points, development of community-designed and -maintained small systems, or better execution of rural water funds. The Importance of Continued Institutional Reforms Institutional reforms are key to improving water sector performance. Countries pursuing institutional reforms create more efficient and effec- tive sector institutions and promote more rapid expansion of higher quality services. The potential dividend is large, because addressing util- ity inefficiencies alone could make a substantial contribution to closing the sector funding gap in many countries. Although the majority of African countries have embarked on the sec- tor reform agenda, few have completed it. The experience of those coun- tries that are farthest ahead provides some guidance for the region. A strong correlation is found between aggressive pursuit of institutional reforms and progress toward achieving the targets of the Millennium Development Goals. The countries that have been most successful in bringing the rural population out of surface water are, without exception, among the most aggressive reformers in Africa. Benin, Côte d'Ivoire, Mozambique, Namibia, Nigeria, Senegal, and Uganda are outstanding performers in reducing the share of population consuming surface water and rank highest in rural reform. Conversely, the Democratic Republic of Congo, Kenya, Malawi, Niger, and Zambia increasingly rely on surface water and score very low on the rural reform index. Burkina Faso and Tanzania perform poorly on access expansion, which is surprising given their strong track record on institutional reforms. For moderate reformers, the results can go either way. The degree of reform also affects how adequately rural water points are maintained. The percentage of rural water points needing rehabilitation Policy Options for the Water and Sanitation Sectors 253 tends to be lower for countries with more advanced rural reform processes. Thus, Benin and Uganda score high on sector reform and in maintaining rural water points. The opposite is true for the Democratic Republic of Congo and Malawi. In rural areas, a few critical interventions can make a difference. Establishing a clear sector policy, creating a strong central capability for sector financing and project implementation, moving to greater cost recovery, and developing a system to monitor the condition of rural water points are all measures that, when implemented as a package, can boost performance. The governments can also take a leading role in ini- tial supply-chain management and donor coordination until the private sector is capable of taking over. In urban areas, the story is more complex. The traditional reform agenda of the 1990s has not fully proven its complete relevance to the sector. In addition, unlike in rural areas, no clear evidence is seen in urban areas that regulation has made a positive contribution to sector perfor- mance across the board. Certain types of institutional reforms hold the key to improving util- ity performance. Good institutional frameworks pay off in improving utilities' efficiency. Utilities that have decentralized or adopted private sector management reveal substantially lower hidden costs than those that have not. A large effect is also associated with unbundling; however, unbundling is uncommon in Africa and is concentrated exclusively in middle-income countries whose superior performance can be explained for many other reasons. The new reform agenda for water retains a role for private participa- tion. Private sector participation, although controversial in implementa- tion, has in many cases been a useful tool for improving operational performance and efficiency. Expectations that the private sector would finance new infrastructure for water utilities have not been met; negligi- ble private capital flows are dwarfed by public and donor finance. Despite this, the private sector has contributed to expanding access, though typi- cally with public funding. Lease contracts may be the form of private participation best suited to African water utilities. They have provided greater scope for operational improvements by transferring more responsibility to the private sector than in a management contract. In contrast to concessions, lease contracts are recognized explicitly as requiring publicly funded investments, even in cases in which the private sector can help execute them. A key lesson from Africa's experience with lease contracts is that it is difficult to 254 Africa's Water and Sanitation Infrastructure achieve seamless coordination on investment plans between the contrac- tor and the public holding company. Incorporating clear contractual incentives for efficiency improvements--for example, by basing the con- tractor's revenue on ideal rather than actual performance parameters--is important. The new agenda places greater emphasis on broader reforms to gov- erning state-owned enterprises. Given the limited scope of private partic- ipation, state-owned utilities remain center stage. Without addressing the typical deficiencies that afflict such enterprises--including numerous and conflicting objectives, political interference, and lack of transparency--it will be difficult for the sector to exit low-level equilibrium. Three key areas for attention are internal process improvements, increased manage- rial autonomy, and more stringent performance monitoring. It is essential to incorporate measures to streamline corporate processes such as pro- curement, financial management, and performance management to strengthen commercial principles and accountability mechanisms. Measures to broaden the board of directors, increase use of external audit and independent audit of accounts, and incorporate independent directors from beyond the public sector would help to depoliticize decision mak- ing and consolidate the arm's-length relationship. Adopting performance- based monitoring arrangements that mimic private sector contracts is also of interest, but only to the extent that they create credible incentives by incorporating meaningful rewards and penalties at the personal and cor- porate levels and are subject to third-party monitoring. The Benefits of More Effective Public Expenditure The bulk of investment in the water sector is made by relevant govern- ment agencies through the budgetary process, often with external sup- port. The existing patterns of spending clearly show that although utilities are instrumental in delivering services, the general government--using either domestic or external capital--continues to make most of the investment decisions. For this reason, a solid public investment appraisal system and strong public spending management are prerequisites for improving both urban and rural water supply. Major bottlenecks hold back the disbursement of public investment funds. Capital budget execution ratios for public investment in water are fairly low, 75 percent on average. In many instances, the binding constraint is not availability of budgetary resources, but rather the capacity to dis- burse them in a timely fashion. In Tanzania, there were steep increases in budget allocations to the sector following its identification as a priority in Policy Options for the Water and Sanitation Sectors 255 the country's poverty reduction strategy. Disbursements increased at a much slower pace, in contrast, and no immediately discernible impact on access is seen. The budgeting process needs to move toward a medium-term frame- work and make stronger links between sector objectives and resource allocations. This needs to be underpinned by clear sector plans that detail specific activities and their associated costs. It is essential that mainte- nance needs be incorporated into medium-term sector planning tools to prevent asset rehabilitation. Administrative processes that delay the release of budgeted funds also need to be overhauled. At the same time, procedures for procurement, disbursement, financial management, and accountability should be modernized and streamlined. Donor resources are best channeled programmatically as budgetary support or through sectorwide projects. Given the sector's high depen- dence on external funds, a solid public expenditure management system for African countries also requires that donors improve the predictability of their support and make progress on streamlining and harmonizing administrative procedures. In that sense, it is preferable to focus on multi- donor initiatives that pool funds to provide general budgetary support for a sectorwide program of interventions. Technical assistance to the sector should include support to relevant government agencies for project identification and appraisal. Technical assistance to the sector has traditionally been understood as improving management practices of utilities. However, an equally important role is available for technical assistance to support government agencies in improving the framework for identifying, appraising, prioritizing, plan- ning, and procuring investment projects. Donors can support countries in the development of good project identification and appraisal tools that systematically consider the technological alternatives for expanding access and that weigh the importance of spending on maintenance and rehabilitation against new investment. Institutional Models to Connect the Unconnected The role of standposts in urban water supply has the potential to expand to serve safe water to a larger number of consumers. Most countries' gov- ernments and utilities continue to focus attention on expansion of piped- water connections, but rapid urbanization and the utilities' weak financial position make this a questionable strategy to pursue so single-mindedly. Standpost use is very limited in the African urban water scene, is expand- ing relatively slowly, and remains concentrated among the more affluent 256 Africa's Water and Sanitation Infrastructure segments of the population. Simple simulations suggest that the rate of service expansion could double if utilities shifted their investment bud- gets from piped-water connections to standposts. As long as urban house- holds are inconvenienced by higher payments and longer water collection times, however, standposts will not necessarily be a superior solution, even if they are a cheaper alternative to private piped-water connections. In low-income countries, resale of water by neighbors through informal standpost arrangements is almost as prevalent as formal standposts. The explanation of this paradox may lie in the problematic institu- tional arrangements associated with standposts in African cities. Utilities charge little or nothing for standpost water, and standpost revenue con- stitutes a negligible portion of the revenue base. This means that utilities lack a financial incentive to expand the service. Standpost operators, where they exist, often charge substantial markups that make the service prohibitively expensive and may generate significant revenue that is never captured by the utility. Quality of service provided by standposts can be very low because of both high rates of malfunction and the large numbers of people expected to rely on each one. The solution to this conundrum is not yet clear, but it will require intensive experimentation with alternative network designs and institu- tional setups. Standposts cover a wide range of communal arrangements or delegated management models, some of which may be less promising than others. One option would be to increase the density of standposts to increase competition, with immediate impact on water supply quality and price. Yard taps can provide communal access to four or five contigu- ous households; this option lowers costs but only partially addresses the problem of maintenance and management. Whatever the approach, an important component of the solution will be to ensure fairer distribution of revenue among utilities and standpost operators or other secondary water retailers. The experiences of the handful of low-income countries that have achieved more than 20 percent urban coverage of standposts-- notably Côte d'Ivoire, Rwanda, and Senegal--deserve study. The popularity of the household resale option could also be exploited by making it an explicit part of the utility's rollout strategy. Household resale of water through yard taps appears in wide use in many African cities. Survey evidence highlights a variety of reasons why residents may find this approach preferable to official standposts. Neighbors can offer more convenient opening hours, better water pressure levels, and more convenient proximity, which reduces the time needed to collect the water. In addition, they offer more flexible payment mechanisms than Policy Options for the Water and Sanitation Sectors 257 either public standposts or a private connection. It is therefore advisable to give increasing recognition to this water supply modality, to remove any legal barriers to its implementation, and to consider making these household- based water retail enterprises an integral component of the utilities' expansion plans. Ultimately, investing in utility production and distribution of water is the best policy for maintaining low-cost alternatives. Within cities, the formal and informal water markets are strongly connected, which influ- ences the final price offered to the consumer. The greater the disruption within the formal piped-water system, the higher the price in the infor- mal sector relative to the formal one. Increasing water production capac- ity and improving the efficacy of the distribution network can have a significant impact on the welfare of the unconnected as well as the con- nected, because it drives down the premium on alternative sources of water supply. Accompanying Cost Recovery with Careful Social Policies Underpricing is debilitating the water sector and slowing coverage expan- sion without contributing much to equity objectives. Underpricing water is contributing to utilities' financial weakness, slowing access expansion, and restraining quality of service. Because utility customers are drawn from the upper end of the income distribution, the result is a highly regressive incidence of subsidies to the sector. A large (and generally poor) segment of the urban population is paying multiples of these prices to access utility water indirectly, and in many cases more than the utility cost-recovery price. Countries need to make progress toward further cost recovery while considering the economic circumstances of their populations. Although African households' purchasing power is quite limited, analysis confirms that operating cost recovery is a perfectly feasible objective for just about all African countries. Tariffs that recover full capital costs also look to be affordable for the richest 40 percent of the population in low-income countries, of which 10 percent already has access to piped water. There is thus little economic justification for the subsidies that exist today. Countries would be better served by recovering full costs from their existing customer base and using the resulting cash flow to accelerate access expansion in poor neighborhoods. In the longer term, however, as access to piped water increases, low-income countries will need social tar- iffs that provide water priced at operating cost-recovery levels for a min- imum level of consumption to the substantial share of their population 258 Africa's Water and Sanitation Infrastructure that cannot afford full capital cost-recovery tariffs. The key is verifying water tariffs' affordability with reference to household budgets, rather than simply assuming that they will be unaffordable. Government entities need to become better customers. Government entities can easily capture 20 to 30 percent of total billings. They can be the worst offenders in paying bills, as well, with a significant lag in pay- ment time. Often, they repay a large chunk of arrears with little indica- tion of future payment schedules. This hampers efforts to sustain a robust payment culture and to improve utilities' investment-planning base. A need is also present to rethink the design of water utility tariff structures. Most African utilities are increasing block tariffs to make water tariffs more equitable, but in reality, half of the utilities using this strategy incorporate fixed charges or minimum consumption thresholds that inflate the costs of water for poor households with modest levels of consumption. Compounding this counterproductive result is the fact that a significant share of utilities with increasing block tariffs also have very high subsistence blocks (in excess of 10 cubic meters); as a result, they end up providing subsidized water to the vast majority of con- sumers, rather than a targeted group of low-volume users. Connection charges should be kept as low as possible, and subsidies could be reoriented toward connections. The majority of African water utilities levy piped-water connection charges in excess of $100, an insur- mountable barrier for low-income households. Utilities intent on achiev- ing universal access should explore ways to radically reduce connection charges to levels that are more in line with household affordability. Several alternative means of recovering connection charges are available, including offering payment plans that spread them out over time or sharing connection costs across the whole customer base through the general tariff. Connection costs may also be more suited to public sub- sidy than water-usage tariffs. They have the advantage of being one-time payments linked to a concrete and monitorable action that addresses a real affordability constraint. Simulations suggest that connection subsi- dies can be much more pro-poor than general subsidies to the water tar- iff, particularly if simple targeting mechanisms are used. Toward a Better Understanding of Groundwater in Urban Water Supply Groundwater, sourced from water wells (boreholes and dug wells), now supplies one-fourth of urban dwellers and is by far the fastest-growing source of improved water supply in African cities. Although wells and Policy Options for the Water and Sanitation Sectors 259 boreholes have long been a dominant source of improved water in rural areas, they have also become an increasingly important source of water supply in almost all urban areas. This is true in more than just those cities (such as Abidjan and Lusaka) where groundwater has long been a major source of utility supply. With utility coverage rates falling in urban Africa, groundwater has essentially stepped into the breach, with the fairly rapid growth of boreholes showing the appetite for lower-cost solutions. Investments in boreholes provide the opportunity to reach a wider demo- graphic with relatively modest resources. One in four urban Africans relies on wells and boreholes for improved supply, a figure that rises to one in two urban Africans in the low-income countries. In Burkina Faso, Malawi, Mali, Mozambique, Uganda, and Zimbabwe, the share rises as high as three in four. In Malawi, Nigeria, and Rwanda, reliance on urban wells and boreholes is increasing particularly rapidly; more than 3 percent of the population gains access to this water source each year. Too little is known about the physical, institutional, and financial characteristics of groundwater use. Household surveys provide a good picture of overall reliance but leave many questions unanswered. The prevalence of simple, shallow, hand-dug wells relative to professionally drilled boreholes is unknown, and so then is the extent to which ground- water supplies are adequately protected from direct wellhead contamina- tion. The institutional arrangements associated with groundwater supplies are also unclear, particularly in terms of the extent to which they consti- tute stopgap services provided by municipalities as opposed to private or communal self-supply initiatives. Depending on the conditions and arrangements, the capital costs of such wells could be anywhere between $5,000 and $25,000 (or $10 to $20 per capita). Extensive decentralized and uncoordinated in situ groundwater use in the urban environment raises risks of contamination by in situ sanitation. In addition to growing groundwater reliance, African cities are character- ized by heavy use of low-grade in situ sanitation, mainly in the form of traditional latrines. Deployment of latrine sanitation at excessive popula- tion densities or lack of proper latrine operation can lead to increasing groundwater contamination that can affect the entire urban aquifer pro- viding the groundwater supplies. Furthermore, extensive unregulated use of groundwater by private actors may prevent the most rational and efficient exploitation of the resource for public water supply. In particular, it prevents cities from reaching economies of scale in groundwater exploitation and from fol- lowing the principle of conjunctive surface and groundwater use that 260 Africa's Water and Sanitation Infrastructure allows groundwater to play its natural role as a backup supply in times of drought. An urgent need exists to develop an improved understanding of the benefits and risks of groundwater use in fast-growing Africa cities and towns, as well as how this varies with the hydrogeological setting. This should begin with a city-level appraisal of the quantity and quality of available urban groundwater resources; the drivers, dynamics and patterns of usage; and an assessment of the vulnerability of urban aquifers to pol- lution from the land surface. Creating a groundwater-monitoring frame- work and the promulgation of appropriate construction and operation protocols for wells and in situ sanitation facilities (mainly latrines) would help to safeguard groundwater quality, but should be accompanied by guidelines for safe use of groundwater sources. Appropriate governance arrangements also need to be put in place, recognizing the broad reach of groundwater resources, and must involve water utilities, public health authorities, and municipal agencies. They should also provide a suitable channel for public consultation. Policy Options for the Sanitation Sector The analyses of the sanitation sector performed by the Africa Infrastructure Country Diagnostic reveal the following key areas for policy attention. The ultimate objective should be to provide universal access by expanding service and reducing open defecation as much as possible. Policy makers are often tempted to concentrate infrastructure enhance- ment efforts on the higher rungs of the sanitation ladder, a strategy that often runs counter to the needs of the majority of the population. For example, officials may channel limited public resources into sewerage networks that serve only a few people and fail to address the more urgent need to significantly reduce the incidence of open defecation. Policy deci- sions and infrastructure programs achieve the greatest public health gains when they take local access patterns into account. Those programs then can be augmented with low-cost initiatives to leverage household spend- ing for latrine construction. Public spending should target helping people on the lowest rungs to move up the ladder. More expensive options should be left to households with the resources to take them up. Complexity, a multiplicity of actors, and lack of accountability for sec- tor leadership are the three features of the institutional framework gov- erning the sanitation sector emerging from an institutional survey of line ministries, sector institutions, and water utilities. Unlike water, many parts Policy Options for the Water and Sanitation Sectors 261 of the supply chain for sanitation--hygiene promotion, latrine construc- tion, and latrine emptying, for example--are in the hands of different public and private players, which prevents one agency from championing the sector and contributes to sanitation's falling between the cracks. The recent trend toward government decentralization has complicated the capture of adequate public resources for sanitation and allocated respon- sibilities to entities that lack technical capacity. Fifteen countries have adopted formal national sanitation policies, and most countries have an accepted definition of sanitation and a hygiene promotion program. Only seven countries, however, have policies that include cost recovery, and only eight countries have a sanitation fund or a dedicated budget line (in some cases funded exclusively by donors, as in Chad and Ethiopia, or by a combination of the government, sector levies, and donors). Côte d'Ivoire has the only fund financed entirely by sector levies. Sanitation challenges vary both across and within African countries, and solutions must be tailored to individual national or regional needs. Open defecation rates remain high in some African countries, especially in rural areas. Countries often pursue solutions such as construction of traditional latrines or septic tanks that reach a small share of the popula- tion, predominantly wealthier urban residents. The policy options for each issue are presented as separate cases in this summary, and countries may need to use different combinations of these approaches to meet their national and regional challenges. The first option is to stimulate demand for sanitation and behavior change where open defecation prevails. The second is to ensure adequate supply before addressing demand in settings dominated by traditional latrines. Finally, the third is to expand access to improved sanitation across larger shares of the population, which in high- density settlements requires making sewerage systems more affordable. Stimulating Demand for Sanitation and Changing Behavior Where Open Defecation Prevails Unlike other infrastructure services, demand for sanitation cannot be assumed. Populations accustomed to open defecation may require a sub- stantial change in cultural values and behavior to use a fixed-point facil- ity. Without such change, people may not use latrines or may use them in a way that undermines the potential health benefits. A study in South India showed that when a large public investment was made in latrine construction but neglected to address the need for accompanying hygiene education, only 37 percent of men used the facilities despite 100 percent coverage (WSP-SA 2002). Hygiene education is a critical component of 262 Africa's Water and Sanitation Infrastructure addressing any sanitation challenge that a country faces. Safe disposal of feces and hand washing with soap protect health in all sanitation settings. Promoting hygiene can start a virtuous cycle that builds demand for bet- ter sanitation, raises awareness of the benefits of sanitation, and estab- lishes codes of conduct and new life standards. Incorrect use of latrines can dramatically reduce or even reverse their health benefits. A facility is sanitary and safe because of the technology and material used and because of good practices and behaviors, such as keeping the facility contained and clean. An improved latrine that is not used and emptied correctly still poses high risks of environmental con- tamination and disease. It thus makes little sense to roll out a physical investment program without an accompanying promotion of hygiene and adequate ways of emptying the latrines on a regular basis. Effective hygiene promotion alone may stimulate self-financed household invest- ment in better facilities. Too often these "soft" aspects of sanitation are overlooked in favor of the "hard" aspects, such as installing and upgrading infrastructure. Changing behavior requires sustained communication and public edu- cation at the community level. It is important to understand people's motivations with regard to hygiene and sanitation. Health is one, but not necessarily the first: Convenience, dignity, and social status may be regarded as more important. For communitywide involvement, it is essential to adapt hygiene and sanitation promotion programs to cultural and institutional norms and then market them in language that demon- strates awareness of and respect for those customs. A successful example is the Regional Health Bureau's Sanitation Advocacy Campaign launched in 2003 in southern Ethiopia, which increased latrine coverage from 13 percent of the population to 78 percent in just two years. Encouraging peer pressure can also help. Once a community recognizes certain behav- iors as desirable, there is pressure to conform. Social institution and lead- ers then begin to contribute, and compliance with the new standards becomes tied to one's social status. Ensuring Adequate Supply before Addressing Demand in Settings Dominated by Traditional Latrines Where traditional latrines prevail, the problem becomes how to upgrade them to more hygienic facilities to achieve the full health benefits of fixed-point defecation. Countries in which traditional latrines are widely used have already overcome the behavioral challenge of moving people out of open defecation. The problem is rather of improving facilities. Policy Options for the Water and Sanitation Sectors 263 The debate centers on whether the main impediment to upgrading latrines comes from the supply side or the demand side. From the demand side, low coverage of improved latrines can be explained by low household incomes and high capital costs. In addi- tion, poor dwellers in urban slum settings often do not own their land or house and so have fewer incentives to invest in improving their liv- ing conditions. Although traditional latrines are more affordable across all income lev- els, improved latrines often remain a luxury. The fact that half of African households have invested in traditional latrines in the absence of any sub- sidy suggests that large investment costs are affordable across the income spectrum. Yet improved latrines are a luxury good limited to the wealth- iest households. To address the affordability problem, public policy will likely have to incorporate a public subsidy for incremental capital costs associated with building improved facilities. A subsidy may have drawbacks, however, including distorted demand and markets. Subsidies can reduce the demand of households with the ability to pay, and suggesting a standard facility may encourage poor households to feel entitled to such a facility regardless of whether it is the most appropriate for their circumstances and geographic location. Widespread adoption of a standard could also discourage innovations that may lower costs. From the supply side, low incidence of improved latrines can be explained by poor knowledge in the construction sector about required designs, lack of skilled construction workers, and shortage of materials. Access patterns already provide some clues that supply-side issues are a real constraint in Africa. First, the prevalence of improved latrines is low, even in middle-income countries, except in a handful of cases. Second, traditional latrines are used by 40 to 50 percent of the population, even among the highest-income groups, who may be able to pay for more advanced facilities. Supply bottlenecks should be tackled first. Otherwise, subsidy resources may be wasted on households that could have financed the facilities on their own. Allowing the local market to develop also provides space for innovation that can lower the cost of improved latrines. Technological innovation is needed to secure greater health benefits with cheaper variants that are tailored to a locality's circumstances. Thus, an important starting point is a more nuanced understanding of the facilities covered by the term traditional latrine and the best practices for their use and maintenance. 264 Africa's Water and Sanitation Infrastructure The supply problem is compounded by a weak private sector domi- nated by small entrepreneurs at the local level. Constructing latrines demands skills that are not widely available, and small enterprises often do not have the resources to develop new skills or adopt new technologies. Policies need to address supply-side limitations. Government support is best channeled toward conducting research, developing products, mar- keting latrines, and opening supply channels for key inputs. Training small service providers and providing access to credit can also help. The National Sanitation Program in Lesotho, established 20 years ago, is ded- icated to sanitation promotion and private sector training. Households directly employ private latrine builders trained under the program, which has increased national sanitation coverage from 20 percent of the popu- lation to 53 percent. Making Sewerage More Affordable in High-Density Settlements In much of Africa, on-site sanitation is the most cost-effective and only practical way to secure the health benefits of the hygienic disposal of feces. On-site sanitation also has its limits, however. Water consumption rises with urban population growth, which creates the challenge of safely returning large volumes of wastewater. In addition, increased urban pop- ulation density constrains the use of latrines (particularly the simpler types), which require rotation of sites and, therefore, a greater area of land than may be available. At high population densities, sewerage systems are both more suitable and more cost effective. It is critical to reduce the cost of sewerage networks through technolog- ical innovation. Although annual population growth in Africa averages 2.5 percent, the urban population is growing at 3.9 percent. By 2020, nearly 60 percent of the African population will be in urban areas, and within 20 years the population of most African cities will have doubled. Africa's burgeoning cities will need to develop more extensive sewerage networks to deal with this influx of people. The statistics on affordabil- ity suggest that waterborne sewerage is far beyond the reach of all but the most affluent households, and the public subsidies to support such sewerage networks are equally unaffordable. Condominial sewerage systems, a lower-cost alternative developed in Latin America, could be explored in Africa. These low-cost secondary pipe networks are built upstream of the main sewerage networks at the residents' initiative. The public collection network just touches each housing block (or condominium) instead of surrounding it. Decentralized microsystems of collection, treatment, and disposal can also replace the Policy Options for the Water and Sanitation Sectors 265 conventional centralized treatment system. Construction costs are reduced by using small-diameter pipes, with work partially carried out by residents. Experiences in Latin America reveal savings of up to 65 per- cent. Pilot condominial systems are being implemented in several African countries, most notably in the periurban areas of Dakar, Senegal. The Dakar system was expected to furnish 60,000 households (270,000 peo- ple) with on-site sanitation and to support 160 condominial schemes serving 130,000 by 2009. Addressing Several Common Challenges for All Countries Several common challenges cut across all sanitation settings: securing fis- cal space for sanitation expenditures, coordinating the numerous players in the sector, and developing a more refined approach to measuring progress. Securing Fiscal Space. The unglamorous nature of sanitation puts it at a disadvantage in the competition for fiscal resources. Government decen- tralization and poor accounting for sector expenditures make it hard to understand the exact amount of public resources allocated. It is estimated that fewer than half of the countries reported any spending on sanitation, and those that did averaged no more than 0.23 percent of gross domestic product, including both investment and operation and maintenance. At the 2008 African Conference on Sanitation and Hygiene in Durban, South Africa, governments committed to raising public expendi- ture on sanitation to 0.5 percent of gross domestic product by 2010. This would require spending close to the levels needed to reach the target spelled out in the Millennium Development Goals (MDGs), but reach- ing the target will still be difficult because of the need to make up for lag- ging past performance. Better accounting of public expenditure on sanitation will also be needed to monitor progress toward the target. Although governments are called upon to provide more resources, innovative financing approaches that help providers and operators are also needed. Cost recovery has proven to be a limited incentive because the only tariffs in sanitation are on wastewater and apply only to the minority of the population served by waterborne sewerage. Moreover, most African utilities are responsible for providing waste- water services in addition to water supply, which makes it likely that water pays for sanitation. Burkina Faso has taken an innovative approach by levying a sanitation tax as a surcharge on the water bill; funds col- lected are then used to subsidize access to improved sanitation facili- ties in Ouagadougou. 266 Africa's Water and Sanitation Infrastructure Needed--A Champion for the Sanitation Sector. Given that on-site san- itation, as opposed to waterborne sewerage, will likely continue to domi- nate sanitation in Africa, households rather than government will remain center stage. Even so, the government's role in promoting demand and addressing supply bottlenecks remains. Even within the public sector, dis- persion and duplication of sanitation functions too often prevent one entity from leading, and as a result sanitation issues fail to be addressed by any agency. A key policy issue is therefore to identify and empower a clear sanita- tion champion within the public sector. Senegal demonstrates its decision to take sanitation seriously by creating a dedicated sanitation utility. Senegal was also the first country to establish a government body at the national level--the Ministry of Urban Affairs, Housing, Urban Water, Public Hygiene and Sanitation (originally the Ministry for Prevention, Public Hygiene and Sanitation)--to coordinate sector activity. Although it may not always be necessary to create a ministry in the central govern- ment, Senegal provides an important lesson in singling out one entity with a clear mandate to lead. Measuring Progress. Although the Joint Monitoring Programme (JMP) has made strides in monitoring progress toward the MDG target for sanitation, a commensurate effort has not been made to create detailed and frequent country-level monitoring and evaluation systems critical to guiding policy interventions. Most countries have no evalua- tion system, and the countries that are developing such a system have found it is not possible to provide a clear picture of the sector. In any event, monitoring and evaluation systems rarely measure the impact of improved sanitation on health, which is clearly relevant to demonstrate the first-order benefits. At the country level, better monitoring and evaluation systems could be built by ensuring more coordination at the ministerial level--for instance, between the ministry in charge of sanitation and the ministry in charge of health. A larger role should be played at the local level, espe- cially by the decentralized technical departments, in collecting data and monitoring progress. This would require more capacity and resources from the central government. A limitation of the JMP's framework is the classification of traditional latrines, which will continue to dominate African sanitation. Traditional latrines include a heterogeneous collection of installations, some of which can be regarded as improved sanitation. Unfortunately, the JMP's Policy Options for the Water and Sanitation Sectors 267 household survey instruments, which track progress toward achieving the MDG target, cannot distinguish among the differing quality of instal- lations within the latrine category. As a result, the data on progress in sanitation in Africa are least clear precisely where most of the progress is taking place. The precision of household survey instruments should be improved in this respect. It may also be relevant to track the intermedi- ate goal of increasing the share of households making use of some kind of sanitation facility, even if it is an unimproved latrine. Reference WSP-SA (Water and Sanitation Program­South Asia). 2002. "Strategic Sanitation Planning: Lessons from Bharatpur, Rajasthan." WSP-SA, New Delhi. APPENDIX 1 Access to Water Supply and Sanitation Facilities 269 270 Table A1.1 Piped Water (percentage of population) Time period (national) Location Expenditure quintile Country Early 1990s Late 1990s Early 2000s Rural Urban Q1 Q2 Q3 Q4 Q5 Benin -- 23.15 28.74 10.91 60.37 0.22 8.02 10.23 36.65 88.63 Burkina Faso 5.64 3.62 5.89 0.06 32.98 0.00 0.00 0.00 0.40 34.06 Cameroon 12.07 11.34 12.95 2.20 24.23 0.00 0.37 4.34 10.84 49.27 Central African Republic 2.65 -- -- 0.00 6.24 0.00 0.00 0.00 0.00 13.26 Chad -- 3.36 4.45 0.00 21.71 0.00 0.00 0.00 0.02 22.27 Comoros -- 22.67 -- 15.06 42.52 0.00 38.42 13.78 20.67 46.21 Congo, Dem. Rep. 21.00 -- 15.03 0.35 40.45 0.00 0.00 0.00 6.88 58.84 Congo, Rep. -- -- 25.81 2.99 46.21 0.00 0.27 5.47 33.77 89.77 Côte d'Ivoire 23.98 27.93 -- 6.73 64.58 0.00 1.73 2.62 38.23 97.63 Ethiopia -- 4.21 5.98 0.21 48.45 0.00 0.00 0.00 0.00 29.94 Gabon -- 43.03 -- 8.84 55.06 0.07 6.90 30.08 78.44 99.76 Ghana 13.65 15.38 15.08 1.66 33.91 0.56 2.24 2.12 10.72 60.14 Guinea -- 9.62 9.13 1.22 28.06 0.00 0.00 0.44 1.56 43.77 Kenya 16.04 19.54 17.94 10.04 49.67 0.01 1.20 4.35 21.91 62.36 Lesotho -- 11.03 10.74 2.13 50.44 0.00 0.37 0.26 2.81 50.28 Madagascar 5.29 5.90 5.30 2.03 17.20 0.19 0.00 0.00 2.01 24.27 Malawi 6.11 7.74 6.49 1.68 32.04 0.00 0.81 0.42 1.12 30.14 Mali -- 5.66 9.06 1.86 29.25 0.24 1.27 1.25 4.48 38.34 Mauritania -- -- 17.41 9.84 27.51 0.00 0.00 5.31 25.29 56.57 Mozambique -- 6.55 6.86 0.33 19.72 0.00 0.00 0.00 0.04 34.43 Namibia 30.53 37.29 -- 16.48 79.30 0.00 1.76 16.59 68.35 99.82 Niger 5.39 6.09 -- 0.20 31.29 0.00 0.00 0.03 4.58 26.04 Nigeria 10.58 10.28 6.88 2.49 15.49 0.13 1.43 3.83 11.45 17.60 Rwanda 1.77 6.28 2.95 0.59 15.97 0.00 0.00 0.30 1.26 13.33 Senegal 26.60 31.10 43.36 17.68 76.76 0.92 9.08 36.24 74.59 96.48 South Africa -- 59.18 -- 24.99 87.72 3.05 24.53 71.70 97.15 99.56 Sudan -- 21.12 -- 9.73 37.44 0.02 0.22 5.39 44.57 77.45 Tanzania 10.23 13.78 7.36 2.86 21.87 0.00 0.00 0.20 6.92 29.72 Togo -- 17.75 -- 3.11 51.30 0.99 2.22 4.98 17.28 63.35 Uganda 1.80 -- 1.99 0.15 14.39 0.00 0.00 0.13 0.04 9.89 Zambia 31.41 21.03 18.32 2.73 46.43 0.00 0.12 0.28 14.67 76.75 Zimbabwe 26.68 32.75 -- 4.43 93.04 0.00 11.42 6.64 49.01 98.57 Country typology Resourch-rich 14.65 15.10 12.01 3.18 23.52 0.08 1.02 3.95 17.34 35.39 Middle-income 63.26 56.46 52.07 22.90 87.00 2.80 22.64 66.60 92.24 97.63 Fragile states 27.01 24.30 26.09 2.85 46.56 0.05 2.08 1.65 17.28 67.45 Nonfragile, low-income 8.35 8.18 10.54 2.73 36.05 0.10 0.91 2.42 8.52 37.89 Level of urbanization High 24.24 22.91 21.84 7.01 38.76 0.69 5.83 17.18 31.99 49.01 Medium 24.56 23.41 21.77 3.75 41.28 0.05 1.60 2.64 20.12 64.39 Low 6.42 5.94 8.12 2.10 32.70 0.03 0.26 0.70 4.56 31.41 Level of water scarcity High 18.41 16.61 16.66 4.22 38.74 0.38 3.52 9.38 20.88 42.51 Low 15.95 16.63 16.37 3.24 37.81 0.08 0.90 3.08 13.60 52.44 Overall 17.60 16.62 16.57 3.92 38.39 0.28 2.65 7.28 18.46 45.81 Source: Banerjee, Wodon, and others 2008. Note: Location and expenditure quintile data refer to latest available year. -- = not available. 271 272 Table A1.2 Standposts (percentage of population) Time period (national) Location Expenditure quintile Country Early 1990s Late 1990s Early 2000s Rural Urban Q1 Q2 Q3 Q4 Q5 Benin -- 5.49 14.12 18.74 5.94 13.36 18.07 20.17 16.18 2.83 Burkina Faso 10.16 8.29 12.74 4.03 53.21 0.00 0.00 4.76 16.92 46.85 Cameroon 22.34 23.39 26.01 9.56 43.26 0.00 17.03 28.44 49.15 35.50 Central African Republic 19.08 -- -- 1.55 42.85 0.00 2.25 10.46 35.97 46.72 Chad -- 5.67 6.62 2.38 23.02 0.00 1.26 7.02 6.52 18.37 Comoros -- 26.66 -- 25.31 30.19 69.10 7.95 25.62 15.37 6.75 Congo, Dem. Rep. 7.09 -- 12.11 4.92 24.55 0.00 2.59 7.77 23.52 24.34 Congo, Rep. -- -- 23.49 5.34 39.71 3.86 11.29 45.58 50.46 6.20 Côte d'Ivoire 21.44 23.12 -- 27.61 15.36 23.62 32.62 29.45 29.32 0.50 Ethiopia -- 11.62 15.82 12.42 40.86 0.00 4.24 17.72 20.07 37.27 Gabon -- 30.35 -- 10.93 37.19 13.50 54.94 63.39 19.84 0.00 Ghana 18.65 21.18 20.48 7.89 38.14 2.48 21.59 17.06 34.99 26.99 Guinea -- 10.94 12.75 2.16 38.09 0.00 0.00 4.60 16.57 42.83 Kenya 11.15 9.41 9.44 6.70 20.45 3.08 5.16 10.21 15.26 13.53 Lesotho -- 51.68 48.04 50.15 38.35 55.03 42.97 56.23 54.95 31.16 Madagascar 11.71 11.21 17.95 10.11 46.51 0.00 0.46 5.57 33.90 49.89 Malawi 19.54 15.77 12.71 7.00 43.04 0.00 0.14 6.24 22.83 34.37 Mali -- 11.14 20.30 16.25 31.69 0.66 11.26 27.47 28.28 33.95 Mauritania -- -- 14.88 8.20 23.79 0.00 23.93 19.05 20.62 11.51 Mozambique -- 17.76 17.62 4.80 42.86 0.00 2.65 6.01 29.34 51.39 Namibia 19.21 20.80 -- 21.85 18.68 13.48 29.86 38.16 22.51 0.15 Niger 11.31 12.58 -- 6.76 37.48 0.00 0.00 0.00 32.26 31.00 Nigeria 13.12 13.52 9.38 5.56 16.91 3.66 6.71 8.76 13.05 14.77 Rwanda 20.92 29.40 24.71 21.83 40.60 0.00 3.45 50.51 22.74 47.55 Senegal 17.94 16.51 18.12 23.05 11.72 19.00 32.59 23.48 13.59 1.82 South Africa -- 19.25 -- 30.16 10.15 30.74 47.69 17.36 0.42 0.00 Sudan -- 7.96 -- 5.27 11.81 7.60 9.95 8.22 7.76 5.50 Tanzania 20.45 20.39 25.20 18.97 45.31 12.08 14.94 19.79 36.86 42.49 Togo -- 17.64 -- 15.25 23.10 8.10 11.46 20.16 27.63 20.90 Uganda 4.17 -- 7.23 1.24 47.49 0.38 0.20 0.85 2.68 32.34 Zambia 17.92 15.93 15.61 4.20 36.20 0.04 2.50 13.37 45.22 16.89 Zimbabwe 8.01 7.21 -- 7.90 5.73 0.58 8.71 10.47 15.56 0.70 Country typology Resourch-rich 13.59 14.97 12.56 5.51 19.82 3.71 7.97 11.49 17.33 14.93 Middle-income 23.20 20.56 18.92 31.03 10.75 30.98 46.77 19.76 3.48 1.23 Fragile states 17.45 17.01 18.37 10.55 23.07 4.82 8.58 12.24 23.41 19.65 Nonfragile, low-income 13.11 12.62 16.17 10.76 36.23 3.11 7.57 14.27 23.13 33.66 Level of urbanization High 17.16 17.09 16.03 12.33 19.68 10.13 19.36 15.43 17.06 12.81 Medium 14.97 15.72 15.03 6.60 22.33 3.00 5.88 9.57 21.39 21.33 Low 12.19 11.55 15.59 10.71 38.87 2.65 5.15 14.04 21.95 34.90 Level of water scarcity High 14.12 13.45 14.00 9.30 20.16 5.33 10.42 11.94 14.51 19.97 Low 16.02 17.07 18.88 13.10 31.34 5.88 10.97 16.60 31.03 30.47 Overall 14.75 14.66 15.62 10.46 24.35 5.51 10.60 13.49 20.00 23.46 Source: Banerjee, Wodon, and others 2008. Note: Location and expenditure quintile data refer to latest available year. -- = not available. 273 274 Table A1.3 Wells/Boreholes (percentage of population) Time period (national) Location Expenditure quintile Country Early 1990s Late 1990s Early 2000s Rural Urban Q1 Q2 Q3 Q4 Q5 Benin -- 54.71 44.93 54.40 28.15 49.70 60.09 62.41 44.20 8.23 Burkina Faso 78.58 82.30 67.66 79.45 12.89 65.35 83.06 84.16 77.14 18.61 Cameroon 28.14 26.68 32.13 48.15 15.34 46.57 55.87 33.57 19.61 5.02 Central African Republic 38.48 -- -- 40.98 35.09 31.40 34.63 53.34 42.61 30.42 Chad -- 72.07 65.99 74.34 33.64 51.72 76.17 76.83 85.42 40.16 Comoros -- 46.10 -- 54.50 24.19 19.09 49.72 57.46 62.13 46.00 Congo, Dem. Rep. 49.04 -- 8.89 7.54 11.21 3.99 12.72 14.66 11.22 3.58 Congo, Rep. -- -- 15.35 25.46 6.31 17.75 27.05 21.40 8.45 2.09 Côte d'Ivoire 41.33 41.29 -- 53.58 20.03 55.75 55.24 63.86 29.70 1.87 Ethiopia -- 6.07 9.88 10.96 1.97 5.46 13.53 11.97 10.89 7.55 Gabon -- 8.35 -- 23.56 3.00 22.39 14.82 3.41 1.12 0.00 Ghana 32.51 35.33 42.10 57.41 20.61 51.25 55.55 49.40 46.18 7.95 Guinea -- 46.52 50.48 58.96 30.17 30.48 70.84 66.88 71.50 12.86 Kenya 24.96 21.98 21.64 23.54 14.04 11.78 24.99 29.60 25.52 16.48 Lesotho -- 14.97 33.33 38.40 9.97 43.52 42.64 31.93 33.42 15.00 Madagascar 15.67 23.06 21.72 22.42 19.18 18.98 17.90 18.65 28.44 24.63 Malawi 58.92 65.91 69.02 77.59 23.57 83.74 82.86 76.69 68.59 33.19 Mali -- 78.60 65.11 75.20 36.81 94.67 79.64 62.18 61.92 26.80 Mauritania -- -- 45.25 67.77 15.25 95.76 45.07 53.49 24.31 6.06 Mozambique -- 46.34 59.38 72.83 32.93 68.28 78.48 76.80 59.89 13.03 Namibia 27.26 31.57 -- 47.14 0.13 66.50 51.99 34.72 4.41 0.03 Niger 75.17 71.95 -- 86.85 8.18 99.00 95.15 92.01 52.15 20.31 Nigeria 31.64 44.17 53.71 56.77 47.71 59.03 57.79 52.13 48.35 51.24 Rwanda 1.53 9.77 20.21 20.68 17.65 6.16 37.13 17.96 25.34 14.96 Senegal 51.44 49.22 35.61 55.57 9.66 77.15 55.62 36.32 8.18 0.44 South Africa -- 4.07 -- 8.55 0.33 8.92 7.26 3.17 0.93 0.06 Sudan -- 45.29 -- 51.44 36.48 74.44 58.07 40.75 26.57 11.55 Tanzania 31.03 40.97 41.04 47.81 19.20 57.02 54.11 44.72 34.18 15.06 Togo -- 38.49 -- 45.05 23.46 39.69 50.56 48.81 38.48 15.05 Uganda 40.48 -- 68.23 73.25 34.52 65.55 68.72 72.30 82.22 52.11 Zambia 24.24 45.32 46.87 64.22 15.57 69.37 63.12 60.14 35.70 5.89 Zimbabwe 54.50 52.23 -- 76.22 1.17 80.47 67.62 76.35 34.35 0.73 Country typology Resourch-rich 36.70 47.72 48.97 56.70 39.35 59.74 57.96 49.47 42.44 36.71 Middle-income 5.56 5.58 5.97 12.93 0.46 12.67 10.52 5.61 2.35 0.64 Fragile states 61.45 42.44 52.96 41.51 13.71 27.65 35.18 38.96 25.74 5.59 Nonfragile, low-income 34.37 33.55 38.33 43.36 18.14 42.68 46.63 43.21 37.69 17.20 Level of urbanization High 27.89 34.16 39.42 48.52 28.75 49.00 47.52 41.43 33.62 28.06 Medium 57.03 49.49 51.84 50.13 21.07 44.17 46.67 43.96 31.05 8.25 Low 33.28 32.04 36.81 40.92 16.75 38.02 43.76 40.98 38.37 20.28 Level of water scarcity High 34.97 36.31 41.12 44.33 28.17 44.09 46.22 42.57 37.18 25.25 Low 41.48 38.84 42.16 46.36 16.66 42.52 45.12 40.42 30.19 10.36 Overall 37.12 37.15 41.46 44.96 23.85 43.57 45.85 41.86 34.86 20.30 Source: Banerjee, Wodon, and others 2008. Note: Location and expenditure quintile data refer to latest available year. -- = not available. 275 276 Table A1.4 Surface Water (percentage of population) Time period (national) Location Expenditure quintile Country Early 1990s Late 1990s Early 2000s Rural Urban Q1 Q2 Q3 Q4 Q5 Benin -- 16.51 12.13 15.87 5.50 36.60 13.82 7.04 2.90 0.28 Burkina Faso 3.91 4.87 13.42 16.21 0.50 34.63 16.51 10.75 5.33 0.04 Cameroon 35.05 37.14 27.65 39.96 14.75 53.22 26.74 33.03 18.63 6.53 Central African Republic 39.14 -- -- 57.42 14.35 68.60 63.08 36.00 20.29 7.69 Chad -- 11.97 18.48 22.35 3.50 48.24 21.57 13.11 6.67 2.34 Comoros -- 2.88 -- 3.49 1.27 9.72 1.33 1.34 0.40 0.47 Congo, Dem. Rep. 22.42 -- 62.29 85.76 21.67 96.01 84.68 73.14 54.95 12.34 Congo, Rep. -- -- 30.28 59.11 4.52 75.88 54.68 16.78 3.76 0.13 Côte d'Ivoire 12.90 7.44 -- 11.75 0.00 19.62 10.41 4.06 2.70 0.00 Ethiopia -- 78.03 67.94 76.11 7.88 94.49 81.62 69.96 68.68 24.71 Gabon -- 18.02 -- 56.35 4.53 63.97 23.20 2.79 0.14 0.00 Ghana 34.37 27.35 20.12 32.36 2.95 45.35 19.96 29.72 3.96 0.60 Guinea -- 31.69 27.48 37.50 3.53 69.47 28.91 27.89 10.28 0.35 Kenya 44.66 47.00 46.35 56.34 6.20 84.21 64.54 51.83 29.15 1.66 Lesotho -- 20.44 7.67 9.15 0.87 1.37 13.73 11.39 8.42 3.46 Madagascar 65.27 59.25 55.02 65.44 17.07 80.83 81.63 75.78 35.63 1.17 Malawi 15.30 10.49 11.67 13.61 1.34 16.14 16.13 16.31 7.46 2.29 Mali -- 4.27 5.08 6.28 1.72 4.29 7.62 8.34 4.59 0.54 Mauritania -- -- 4.65 7.61 0.71 4.18 12.85 4.29 1.58 0.58 Mozambique -- 28.75 14.98 20.96 3.20 31.64 17.68 14.75 9.15 0.56 Namibia 21.00 8.30 -- 12.42 0.00 17.11 14.42 7.06 2.89 0.00 Niger 3.28 2.60 -- 3.16 0.19 1.00 2.86 4.66 3.16 1.54 Nigeria 42.02 27.77 23.18 30.76 8.28 33.39 29.96 28.19 16.14 8.15 Rwanda 75.41 54.07 51.43 56.29 24.58 93.72 58.60 29.86 50.19 23.42 Senegal 2.64 1.78 0.98 1.72 0.02 1.81 1.28 1.38 0.44 0.00 South Africa -- 14.40 -- 31.30 0.29 53.31 14.48 3.95 0.23 0.00 Sudan -- 19.84 -- 28.50 7.44 14.84 27.16 35.38 13.14 2.70 Tanzania 34.33 24.27 24.39 29.76 7.08 30.91 30.91 34.73 20.58 4.76 Togo -- 25.50 -- 35.91 1.66 50.64 34.95 25.29 15.90 0.46 Uganda 53.08 -- 21.59 24.47 2.19 33.70 30.41 25.28 14.35 4.07 Zambia 26.04 16.56 18.99 28.72 1.44 30.52 34.25 25.95 4.17 0.00 Zimbabwe 10.61 7.56 -- 11.11 0.00 18.95 11.49 6.19 0.95 0.00 Country typology Resourch-rich 35.61 27.41 23.68 30.94 8.15 33.52 29.54 28.54 14.39 6.19 Middle-income 18.16 14.39 13.02 28.63 0.30 49.76 14.45 4.37 0.66 0.14 Fragile states 30.77 31.97 45.62 44.38 15.16 67.26 53.98 44.72 31.66 6.75 Nonfragile, low-income 51.41 34.55 37.42 42.17 5.66 53.87 43.78 38.65 28.36 7.93 Level of urbanization High 31.56 22.96 20.38 28.62 5.69 37.30 23.64 20.91 10.34 4.75 Medium 34.24 31.45 37.93 38.07 12.42 52.03 44.59 39.73 24.30 4.96 Low 53.87 37.05 41.25 45.33 7.07 59.11 49.87 43.06 33.22 9.52 Level of water scarcity High 45.14 31.65 32.56 39.84 5.78 48.35 36.98 32.04 22.15 7.69 Low 31.50 27.84 32.91 36.55 12.16 51.26 42.65 38.04 23.49 4.60 Overall 40.64 30.38 32.68 38.83 8.17 49.32 38.86 34.04 22.60 6.66 Source: Banerjee, Wodon, and others 2008. Note: Location and expenditure quintile data refer to latest available year. -- = not available. 277 278 Table A1.5 Septic Tank (percentage of population) Time period (national) Location Expenditure quintile Country Early 1990s Late 1990s Early 2000s Rural Urban Q1 Q2 Q3 Q4 Q5 Benin -- 0.00 2.39 0.35 6.00 0.00 0.00 0.00 0.63 11.30 Burkina Faso 0.89 0.58 1.86 0.49 8.22 0.00 0.02 0.00 1.42 9.02 Cameroon 6.56 6.41 8.07 0.73 15.76 0.00 0.01 0.35 1.74 38.32 Central African Republic 1.11 -- -- 0.11 2.48 0.00 0.00 0.00 0.31 5.26 Chad -- 0.24 1.83 0.46 7.13 0.00 0.00 0.00 1.19 7.97 Comoros -- 2.93 -- 1.16 7.55 0.00 0.08 0.00 0.73 14.14 Congo, Dem. Rep. 1.56 -- 1.42 0.02 3.83 0.00 0.00 0.00 0.00 6.07 Congo, Rep. -- -- 5.33 0.35 9.78 0.00 0.00 0.03 2.49 24.21 Côte d'Ivoire 14.03 12.45 -- 2.26 30.07 0.00 0.00 0.00 2.15 60.26 Ethiopia -- 0.34 2.13 1.34 7.99 0.00 0.00 1.16 3.60 5.93 Gabon -- 24.50 -- 4.45 31.56 0.09 0.91 4.51 21.69 95.49 Ghana 5.94 7.57 10.28 1.52 22.56 0.27 2.09 1.41 4.42 43.35 Guinea -- 2.65 2.62 0.58 7.51 0.00 0.00 0.00 1.11 12.04 Kenya 7.99 9.75 8.97 1.48 39.06 0.00 0.00 0.25 2.06 42.64 Lesotho -- 2.11 1.61 0.15 8.34 0.00 0.00 0.10 0.47 7.50 Madagascar 2.54 2.26 1.88 0.50 6.89 0.00 0.00 0.04 1.49 7.85 Malawi 2.62 3.30 3.58 0.89 17.87 0.00 0.81 0.08 0.67 16.37 Mali -- 1.12 6.05 3.01 14.56 0.00 0.03 6.86 5.54 17.93 Mauritania -- -- 1.77 0.05 4.06 0.00 0.00 0.00 0.44 8.41 Mozambique -- 3.22 2.88 0.21 8.12 0.00 0.00 0.00 0.02 14.42 Namibia 26.65 30.56 -- 6.80 78.54 0.00 0.00 1.84 51.56 99.49 Niger 1.25 1.05 -- 0.23 4.58 0.00 0.00 0.00 2.42 2.89 Nigeria 8.46 11.90 13.12 5.65 27.80 0.07 0.46 1.43 9.73 54.14 Rwanda 1.05 1.47 1.16 0.24 6.27 0.00 0.00 0.26 0.27 5.34 Senegal 10.62 9.07 36.04 14.15 64.51 1.10 7.46 37.24 56.93 77.81 South Africa -- 46.37 -- 5.84 80.21 0.13 3.75 35.50 92.93 99.62 Sudan -- 6.42 -- 1.12 14.02 0.21 0.12 1.42 6.59 31.30 Tanzania 1.41 1.66 2.75 0.47 10.12 0.00 0.00 0.00 0.93 12.84 Togo -- 0.00 -- 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Uganda 1.59 -- 1.73 0.40 10.67 0.00 0.15 0.28 0.90 7.37 Zambia 27.13 20.69 18.09 2.11 46.92 0.00 0.12 1.07 13.65 75.85 Zimbabwe 26.25 31.45 -- 1.54 95.12 0.00 11.32 5.92 42.64 99.21 Country typology Resourch-rich 8.94 12.41 11.18 4.08 23.63 0.08 0.32 1.25 8.32 47.82 Middle-income 49.37 44.02 40.78 5.51 79.18 0.12 3.44 32.71 87.57 95.99 Fragile states 7.13 7.21 6.65 0.85 14.48 0.00 1.43 0.75 5.87 27.15 Nonfragile, low-income 2.68 2.20 4.89 1.33 17.44 0.06 0.48 2.13 4.26 17.93 Level of urbanization High 17.16 17.53 18.77 5.07 38.09 0.14 1.44 8.89 24.80 60.81 Medium 7.42 9.27 6.42 0.80 14.26 0.04 0.95 0.88 6.57 26.57 Low 1.91 1.35 3.20 0.95 13.35 0.00 0.06 0.73 2.14 13.16 Level of water scarcity High 10.71 10.76 11.39 2.45 31.93 0.06 0.97 4.43 14.73 38.70 Low 5.53 6.59 6.49 1.64 14.45 0.06 0.42 2.52 5.36 24.75 Overall 9.00 9.37 9.77 2.20 25.37 0.06 0.78 3.80 11.61 34.06 Source: Banerjee, Wodon, and others 2008. Note: Location and expenditure quintile data refer to latest available year. -- = not available. 279 280 Table A1.6 Improved Latrine (percentage of population) Time period (national) Location Expenditure quintile Country Early 1990s Late 1990s Early 2000s Rural Urban Q1 Q2 Q3 Q4 Q5 Benin -- 1.46 13.87 5.15 29.35 0.00 1.06 3.67 12.78 51.88 Burkina Faso 0.71 0.25 17.92 6.81 69.55 0.09 0.83 3.06 19.84 74.35 Cameroon 0.00 23.54 26.98 13.27 41.36 0.02 0.96 24.57 63.73 45.65 Central African Republic 13.28 -- -- 18.43 6.31 16.06 27.59 13.42 6.73 2.61 Chad -- 7.51 2.74 0.23 12.46 0.00 0.00 0.00 0.45 13.26 Comoros -- 20.71 -- 15.33 34.74 0.00 0.32 18.06 24.41 61.82 Congo, Dem. Rep. 10.77 -- 9.77 0.35 26.07 0.00 0.00 0.00 1.81 40.42 Congo, Rep. -- -- 15.07 4.11 24.87 0.22 0.62 7.74 18.88 48.03 Côte d'Ivoire 22.48 13.30 -- 7.73 22.92 0.00 1.21 10.59 28.38 26.92 Ethiopia -- 0.30 0.89 0.48 3.88 0.00 0.00 0.06 0.51 3.85 Gabon -- 22.09 -- 8.40 26.91 1.89 12.71 38.22 53.28 4.27 Ghana 13.19 21.84 22.63 10.93 39.04 0.92 22.06 17.24 38.59 35.18 Guinea -- 0.00 2.06 1.65 3.02 0.00 0.55 3.77 2.23 3.77 Kenya 5.57 6.19 7.96 7.12 11.33 0.00 0.21 5.92 11.88 21.82 Lesotho -- 18.01 20.78 16.93 38.49 0.00 0.76 25.40 34.27 43.83 Madagascar 30.79 4.40 49.01 44.15 66.71 0.24 20.36 66.40 76.58 82.34 Malawi 0.67 0.64 1.20 0.96 2.44 0.00 0.00 0.07 0.01 5.90 Mali -- 7.77 10.79 7.08 21.18 0.04 0.62 7.80 17.32 28.30 Mauritania -- -- 3.82 0.25 8.59 0.00 0.00 0.30 3.27 15.57 Mozambique -- 0.88 1.81 0.07 5.26 0.00 0.00 0.00 0.02 9.09 Namibia 0.40 2.74 -- 3.14 1.95 0.00 0.00 6.72 7.00 0.01 Niger 12.24 12.14 -- 2.14 54.93 0.09 0.06 0.29 2.24 58.42 Nigeria 0.00 6.31 2.89 1.70 5.22 0.00 0.41 2.05 5.22 6.78 Rwanda 0.00 8.16 29.32 25.87 48.42 0.00 0.40 4.07 75.40 66.82 Senegal 21.97 23.08 10.10 10.78 9.21 6.48 16.32 10.02 10.23 7.43 South Africa -- 0.00 -- 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Sudan -- 0.00 -- 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Tanzania 1.32 0.93 3.69 1.06 12.17 0.00 0.00 0.09 1.25 17.11 Togo -- 18.01 -- 11.55 32.82 0.00 3.94 6.72 31.79 47.69 Uganda 1.65 -- 2.39 1.43 8.86 0.00 0.08 0.34 2.30 9.31 Zambia 1.39 0.38 1.56 1.25 2.12 0.00 0.25 1.19 4.43 1.93 Zimbabwe 21.18 24.96 -- 35.73 2.03 6.54 23.27 57.63 36.06 0.36 Country typology Resourch-rich 1.63 6.63 6.39 2.26 8.04 0.02 0.44 3.71 9.35 9.62 Middle-income 0.94 0.93 1.43 1.36 0.58 0.00 0.03 1.28 1.64 1.73 Fragile states 20.28 17.07 16.23 10.34 22.22 1.38 4.36 10.33 12.43 28.80 Nonfragile, low-income 5.31 4.36 9.90 6.16 21.72 0.34 3.51 6.89 13.55 24.26 Level of urbanization High 4.23 9.05 8.39 3.96 11.53 0.41 3.17 5.76 13.41 12.86 Medium 10.80 9.82 10.89 5.88 14.21 0.88 2.72 5.81 5.94 19.27 Low 5.20 2.63 8.87 6.06 21.11 0.03 1.58 6.44 12.42 24.28 Level of water scarcity High 4.49 6.16 7.57 4.06 8.86 0.24 1.99 3.90 8.38 13.54 Low 9.73 8.12 12.45 8.38 23.38 0.64 3.36 10.33 17.01 29.49 Overall 6.22 6.81 9.19 5.38 14.31 0.37 2.44 6.04 11.24 18.84 Source: Banerjee, Wodon, and others 2008. Note: Location and expenditure quintile data refer to latest available year. -- = not available. 281 282 Table A1.7 Traditional Latrine (percentage of population) Time period (national) Location Expenditure quintile Country Early 1990s Late 1990s Early 2000s Rural Urban Q1 Q2 Q3 Q4 Q5 Benin -- 24.07 15.25 9.16 26.04 0.30 2.29 9.21 37.11 27.35 Burkina Faso 26.13 22.13 10.03 9.18 13.96 0.16 0.20 12.33 25.27 10.34 Cameroon 45.21 59.89 57.61 73.20 41.27 84.44 84.10 70.54 33.53 15.36 Central African Republic 59.45 -- -- 40.24 85.51 0.00 41.29 75.50 89.32 91.27 Chad -- 20.52 23.62 13.25 63.82 0.00 0.26 1.72 47.00 70.07 Comoros -- 75.07 -- 82.30 56.23 99.75 96.53 80.67 72.63 24.04 Congo, Dem. Rep. 71.25 -- 76.08 81.75 66.27 81.80 76.96 86.03 89.81 51.22 Congo, Rep. -- -- 69.76 78.41 62.03 79.99 80.23 87.10 74.40 26.93 Côte d'Ivoire 21.13 38.62 -- 35.60 43.84 3.10 29.64 84.56 65.50 12.81 Ethiopia -- 16.90 34.70 29.00 76.55 0.00 12.07 55.72 32.81 73.47 Gabon -- 50.91 -- 82.93 39.63 92.59 82.89 55.12 23.79 0.06 Ghana 47.22 40.82 40.53 50.40 26.69 88.07 52.26 25.75 24.96 11.33 Guinea -- 61.09 67.44 59.32 86.86 4.41 88.32 68.51 92.93 83.64 Kenya 67.74 67.61 64.32 69.33 44.21 86.73 81.20 59.62 61.48 32.40 Lesotho -- 36.16 32.85 30.13 45.44 0.00 13.27 48.62 55.55 47.39 Madagascar 5.63 32.78 2.54 1.53 6.21 0.00 0.01 0.38 3.85 8.47 Malawi 71.63 79.46 80.67 81.84 74.46 98.52 88.26 67.01 73.84 75.57 Mali -- 63.93 62.06 61.92 62.47 77.09 71.83 47.43 61.16 52.70 Mauritania -- -- 44.35 28.27 65.77 0.00 12.51 57.66 76.79 75.26 Mozambique -- 34.78 48.01 37.91 67.88 0.00 10.27 77.30 84.02 73.05 Namibia 7.42 7.50 -- 8.97 4.54 0.06 0.12 22.48 14.61 0.28 Niger 3.77 6.92 -- 3.58 21.23 0.00 0.02 0.55 9.92 24.38 Nigeria 61.61 53.97 59.40 60.05 58.13 75.18 75.03 61.96 60.98 23.59 Rwanda 92.95 87.22 66.13 70.40 42.50 100.00 97.34 87.83 20.05 25.21 Senegal 28.74 35.10 31.28 38.42 21.99 56.40 38.63 24.03 24.42 12.81 South Africa -- 34.08 -- 63.87 9.20 47.60 72.02 47.11 3.60 0.00 Sudan -- 48.91 -- 41.49 59.55 19.34 36.79 57.45 77.46 65.73 Tanzania 82.38 84.98 79.22 80.46 75.23 87.34 74.31 80.39 85.50 68.46 Togo -- 14.92 -- 6.03 35.28 2.16 4.93 5.97 13.03 48.56 Uganda 78.74 -- 80.25 80.60 77.89 86.46 83.07 67.65 82.85 81.14 Zambia 42.10 51.32 53.09 56.66 46.65 14.48 70.83 79.94 78.82 21.28 Zimbabwe 13.48 14.89 -- 20.78 2.33 12.00 20.61 25.24 15.86 0.33 Country typology Resourch-rich 57.05 57.19 54.77 55.73 56.54 60.07 65.88 60.52 61.85 31.82 Middle-income 37.23 33.01 30.36 58.24 9.58 43.76 66.72 46.14 6.11 1.88 Fragile states 56.86 54.17 65.26 50.51 58.92 45.73 58.27 71.96 72.68 42.58 Nonfragile, low-income 44.46 40.27 50.10 47.60 50.17 45.72 42.97 51.13 48.97 50.87 Level of urbanization High 50.93 47.82 51.35 57.35 42.05 65.65 67.68 56.80 44.76 17.13 Medium 55.17 56.15 55.06 46.10 58.91 34.01 48.32 65.14 73.88 51.64 Low 44.98 40.51 51.81 48.04 57.12 44.91 44.57 51.84 49.50 55.38 Level of water scarcity High 47.48 40.38 47.88 48.31 46.80 49.14 52.45 51.29 46.32 38.33 Low 54.30 60.64 61.62 55.68 56.86 51.80 57.28 68.03 68.06 44.19 Overall 49.73 47.12 52.44 50.56 50.57 50.02 54.05 56.85 53.55 40.27 Source: Banerjee, Wodon, and others 2008. Note: Location and expenditure quintile data refer to latest available year. -- = not available. 283 284 Table A1.8 Open Defecation (percentage of population) Time period (national) Location Expenditure quintile Country Early 1990s Late 1990s Early 2000s Rural Urban Q1 Q2 Q3 Q4 Q5 Benin -- 73.75 67.67 84.71 37.47 99.48 95.48 86.75 47.92 8.67 Burkina Faso 71.64 76.56 69.96 83.30 8.00 99.64 98.89 84.29 53.07 6.02 Cameroon 12.55 9.82 7.17 12.63 1.44 15.35 14.56 4.35 0.95 0.61 Central African Republic 25.91 -- -- 41.13 5.26 83.90 30.97 10.97 3.21 0.39 Chad -- 71.48 71.61 85.83 16.47 99.89 99.55 98.00 51.36 8.23 Comoros -- 0.27 -- 0.37 0.00 0.00 0.73 0.30 0.40 0.00 Congo, Dem. Rep. 16.41 -- 12.24 17.40 3.29 17.74 22.30 13.53 8.32 1.54 Congo, Rep. -- -- 9.53 16.65 3.16 19.21 18.73 4.75 4.15 0.77 Côte d'Ivoire 42.10 35.41 -- 54.21 2.92 96.90 68.55 4.85 3.51 0.00 Ethiopia -- 82.45 62.20 69.11 11.36 100.00 87.83 42.98 63.07 16.50 Gabon -- 2.09 -- 3.75 1.51 5.11 2.77 1.78 0.72 0.07 Ghana 26.52 24.00 24.57 36.90 7.28 10.74 23.46 55.11 28.63 4.12 Guinea -- 34.39 27.60 38.14 2.40 95.58 10.99 26.64 3.64 0.51 Kenya 17.79 15.86 18.29 21.84 4.02 13.27 18.33 33.97 23.87 2.03 Lesotho -- 40.41 44.65 52.67 7.65 100.00 85.70 25.70 9.63 1.25 Madagascar 61.01 60.54 46.57 53.82 20.18 99.76 79.63 33.18 18.08 1.32 Malawi 24.98 16.52 14.46 16.23 5.06 1.48 10.94 32.79 25.21 2.02 Mali -- 26.70 20.92 27.73 1.79 22.77 27.28 37.38 15.96 1.04 Mauritania -- -- 49.30 70.44 21.13 99.94 86.54 41.28 17.82 0.40 Mozambique -- 60.27 46.73 61.34 17.98 100.00 88.82 21.53 15.50 3.04 Namibia 63.56 56.60 -- 79.08 11.23 99.33 98.96 65.13 19.49 0.00 Niger 82.30 79.46 -- 93.78 18.22 99.86 99.53 99.01 84.75 13.46 Nigeria 29.38 26.02 24.52 32.53 8.79 24.72 24.07 34.39 23.98 15.46 Rwanda 5.95 2.96 3.32 3.45 2.60 0.00 2.23 7.68 4.22 2.55 Senegal 38.39 32.44 22.01 36.12 3.67 35.43 37.26 28.47 7.50 1.19 South Africa -- 12.78 -- 26.14 1.62 50.13 11.32 2.29 0.14 0.00 Sudan -- 42.65 -- 55.92 23.63 78.42 60.50 38.78 14.05 2.13 Tanzania 13.98 12.38 14.29 17.96 2.44 12.54 25.63 19.51 12.31 1.56 Togo -- 64.09 -- 79.12 29.68 94.81 86.07 85.05 51.83 2.54 Uganda 17.41 -- 14.78 16.64 2.31 13.26 16.33 30.22 12.26 1.78 Zambia 29.11 27.01 27.04 39.94 3.78 85.52 28.75 17.64 3.07 0.13 Zimbabwe 38.71 28.41 -- 41.61 0.32 81.46 44.46 10.64 5.00 0.00 Country typology Resourch-rich 33.25 31.72 27.56 37.67 11.11 39.44 32.86 33.98 20.10 10.51 Middle-income 17.71 15.79 14.25 31.17 1.95 54.14 17.89 5.82 1.31 0.05 Fragile states 51.61 42.71 38.56 37.74 3.87 52.49 35.12 16.44 8.64 0.99 Nonfragile, low-income 58.44 42.73 40.33 44.67 9.67 53.81 52.81 39.49 32.59 6.18 Level of urbanization High 26.50 23.25 21.44 32.86 6.05 33.36 25.34 25.75 16.00 8.55 Medium 57.13 49.19 42.36 46.50 11.43 64.36 46.83 27.11 12.80 1.89 Low 59.19 43.36 41.31 44.74 8.09 55.00 53.64 40.72 35.61 6.85 Level of water scarcity High 53.20 41.64 38.94 44.62 10.21 50.13 42.97 38.47 29.63 8.82 Low 32.58 28.48 24.58 34.06 4.91 47.30 38.53 18.75 9.35 1.25 Overall 46.39 37.26 34.18 41.39 8.22 49.19 41.50 31.92 22.89 6.30 Source: Banerjee, Wodon, and others 2008. Note: Location and expenditure quintile data refer to latest available year. -- = not available. 285 286 Africa's Water and Sanitation Infrastructure Table A1.9 Annualized Change in Water Access: National (percentage of population) Technology Country Piped water Standpost Well/borehole Surface water Benin 1.78 1.88 ­0.39 ­0.40 Burkina Faso 0.69 1.40 ­0.77 2.31 Cameroon 0.45 0.81 1.33 ­0.99 Chad 0.23 0.26 0.73 1.20 Congo, Dem. Rep. ­0.16 1.12 ­4.75 7.53 Côte d'Ivoire 1.47 0.94 1.16 ­0.72 Ethiopia 0.44 1.09 0.89 ­0.36 Ghana 0.26 0.30 2.09 ­0.88 Guinea 0.08 0.49 1.45 ­0.16 Kenya 0.09 0.20 0.40 0.86 Lesotho 0.00 ­0.47 3.75 ­2.45 Madagascar 0.03 1.18 0.25 0.53 Malawi ­0.09 ­0.31 2.69 0.60 Mali 0.83 2.14 ­0.57 0.28 Mozambique 0.16 0.28 2.95 ­1.81 Namibia 1.43 0.57 1.06 ­1.19 Niger 0.27 0.52 1.55 ­0.02 Nigeria ­0.57 ­0.66 3.60 ­0.39 Rwanda ­0.39 0.33 2.51 1.81 Senegal 1.98 0.44 ­0.99 ­0.07 Tanzania ­1.01 1.36 0.82 0.50 Uganda 0.12 0.71 6.53 ­2.75 Zambia ­0.13 0.19 0.95 0.66 Zimbabwe 1.69 ­0.02 0.52 ­0.42 Country typology Resourch-rich ­0.40 ­0.41 3.03 ­0.28 Middle-income 0.75 0.07 2.34 ­1.79 Fragile states 0.42 0.87 ­2.32 4.15 Nonfragile, low-income 0.21 0.84 1.42 ­0.15 Level of urbanization High ­0.06 ­0.22 2.75 ­0.50 Medium 0.28 0.76 ­1.48 3.17 Low 0.08 0.88 1.41 0.08 Level of water scarcity High 0.05 0.18 2.34 ­0.24 Low 0.12 0.98 ­0.53 1.80 Overall 0.07 0.48 1.27 0.52 Source: Banerjee, Wodon, and others 2008. Access to Water Supply and Sanitation Facilities 287 Table A1.10 Annualized Change in Water Access: Urban (percentage of population) Technology Country Piped water Standpost Well/borehole Surface water Benin 3.58 0.80 ­1.09 0.30 Burkina Faso 3.40 4.00 ­1.01 0.13 Cameroon ­0.01 1.34 0.90 0.72 Chad 1.56 2.07 ­0.34 0.27 Congo, Dem. Rep. ­0.31 2.56 ­0.67 3.43 Côte d'Ivoire 3.81 ­0.52 0.36 ­0.04 Ethiopia 4.77 ­0.27 ­0.23 ­1.08 Ghana ­0.18 0.60 2.65 ­0.21 Guinea 0.47 2.10 0.10 0.16 Kenya 0.03 ­0.12 1.49 0.35 Lesotho 2.69 ­0.47 0.63 ­0.66 Madagascar 0.39 2.03 0.33 ­0.41 Malawi ­0.64 3.01 3.10 0.10 Mali 3.00 1.25 ­0.37 0.28 Mozambique ­0.12 0.80 2.31 0.39 Namibia 1.75 1.15 ­0.21 ­0.08 Niger 1.49 0.93 0.28 ­0.12 Nigeria ­1.37 ­0.63 3.99 1.06 Rwanda ­0.66 3.67 3.03 3.15 Senegal 2.28 ­0.42 ­0.25 ­0.03 Tanzania ­3.50 3.91 1.37 0.70 Uganda 0.85 4.67 2.01 ­1.98 Zambia ­0.05 1.09 0.42 0.01 Zimbabwe 2.93 0.46 ­0.19 0.00 Country typology Resourch-rich ­0.98 ­0.17 3.19 0.91 Middle-income 2.19 0.38 0.19 ­0.36 Fragile states 0.99 1.65 ­0.33 1.99 Nonfragile, low-income 1.20 1.53 0.94 ­0.20 Level of urbanization High ­0.44 ­0.31 3.01 0.72 Medium 0.47 1.72 0.02 1.70 Low 1.32 1.77 0.76 ­0.25 Level of water scarcity High 0.86 0.56 1.95 0.13 Low 0.01 1.87 0.40 1.08 Overall 0.54 1.05 1.37 0.49 Source: Banerjee, Wodon, and others 2008. 288 Africa's Water and Sanitation Infrastructure Table A1.11 Annualized Change in Water Access: Rural (percentage of population) Technology Country Piped water Standpost Well/borehole Surface water Benin 1.04 2.46 ­0.16 ­0.93 Burkina Faso ­0.01 0.63 ­0.37 2.81 Cameroon ­0.10 ­0.77 2.46 ­1.40 Chad ­0.03 ­0.17 0.74 1.34 Congo, Dem. Rep. 0.04 0.31 ­7.09 9.89 Côte d'Ivoire 0.44 1.77 1.37 ­1.28 Ethiopia 0.04 1.55 1.02 ­0.76 Ghana ­0.20 ­0.55 2.33 ­0.80 Guinea 0.01 ­0.10 1.93 ­0.45 Kenya 0.05 0.26 0.14 1.05 Lesotho 0.04 ­0.66 4.25 ­3.19 Madagascar 0.01 1.14 0.20 0.50 Malawi 0.03 ­0.93 2.58 0.69 Mali 0.25 2.77 ­1.18 0.22 Mozambique ­0.07 ­0.58 3.73 ­2.38 Namibia 1.18 0.30 1.75 ­1.69 Niger ­0.08 0.35 2.05 0.01 Nigeria ­0.30 ­0.82 3.09 ­1.24 Rwanda ­0.04 ­0.18 2.35 1.10 Senegal 1.25 1.09 ­1.07 ­0.09 Tanzania ­0.18 0.59 0.60 0.41 Uganda 0.01 0.12 7.20 ­2.86 Zambia 0.19 ­0.11 0.85 0.83 Zimbabwe 0.27 ­0.20 1.47 ­0.49 Country typology Resourch-rich ­0.23 ­0.73 2.74 ­0.97 Middle-income 0.64 ­0.16 2.94 ­2.40 Fragile states 0.15 0.47 ­3.47 5.38 Nonfragile, low-income 0.05 0.68 1.53 ­0.29 Level of urbanization High ­0.11 ­0.44 2.55 ­1.14 Medium 0.15 0.18 ­2.29 4.13 Low 0.00 0.76 1.46 ­0.04 Level of water scarcity High ­0.08 0.05 2.29 ­0.62 Low 0.11 0.53 ­1.03 2.24 Overall ­0.01 0.23 1.05 0.45 Source: Banerjee, Wodon, and others 2008. Access to Water Supply and Sanitation Facilities 289 Table A1.12 Annualized Change in Sanitation Access: National (percentage of population) Technology Improved Traditional Open Country Septic tank latrine latrine defecation Benin 0.48 2.53 ­1.08 0.90 Burkina Faso 0.34 4.43 ­2.25 1.04 Cameroon 0.38 0.95 0.57 ­0.29 Chad 0.23 ­0.52 0.90 1.60 Congo, Dem. Rep. 0.04 0.26 3.63 ­0.05 Côte d'Ivoire 0.08 ­1.20 4.10 ­0.14 Ethiopia 0.37 0.12 3.92 ­2.30 Ghana 0.70 0.61 0.79 0.61 Guinea 0.04 0.34 2.09 ­0.55 Kenya 0.05 0.48 0.77 0.82 Lesotho ­0.09 0.64 ­0.48 1.05 Madagascar ­0.01 6.46 ­3.69 ­0.84 Malawi 0.17 0.16 2.61 ­0.04 Mali 1.02 0.81 1.36 ­0.43 Mozambique 0.00 0.17 2.79 ­1.25 Namibia 1.00 0.30 0.15 0.35 Niger 0.00 0.32 0.63 1.81 Nigeria 0.63 ­0.68 2.84 0.34 Rwanda 0.00 4.59 ­0.44 0.20 Senegal 3.50 ­1.29 0.03 ­0.84 Tanzania 0.25 0.57 0.52 0.63 Uganda 0.10 0.20 3.96 0.38 Zambia ­0.12 0.20 1.08 0.42 Zimbabwe 1.51 1.13 0.52 ­1.37 Country typology Resourch-rich 0.53 ­0.45 2.39 0.35 Middle-income 0.48 0.46 ­0.15 0.68 Fragile states 0.25 0.11 3.14 ­0.29 Nonfragile, low-income 0.38 0.99 1.47 ­0.31 Level of urbanization High 0.73 ­0.49 2.38 0.21 Medium 0.22 0.50 2.40 ­0.31 Low 0.24 1.08 1.53 ­0.25 Level of water scarcity High 0.44 0.25 2.24 ­0.06 Low 0.34 0.71 1.59 ­0.18 Overall 0.40 0.42 2.00 ­0.11 Source: Banerjee, Wodon, and others 2008. Note: VIP = ventilated improved pit. 290 Africa's Water and Sanitation Infrastructure Table A1.13 Annualized Change in Sanitation Access: Urban (percentage of population) Technology Improved Traditional Open Country Septic tank latrine latrine defecation Benin 1.20 5.30 ­3.19 0.24 Burkina Faso 1.32 17.17 ­13.07 0.15 Cameroon 0.24 0.92 1.62 0.15 Chad 0.90 ­1.57 3.58 0.15 Congo, Dem. Rep. 0.15 0.76 4.70 ­0.48 Côte d'Ivoire 0.48 ­0.90 4.46 ­0.48 Ethiopia 1.20 0.46 3.88 ­2.23 Ghana 1.24 1.99 1.74 ­0.09 Guinea 0.04 0.50 2.27 ­0.03 Kenya ­0.07 0.60 1.97 0.27 Lesotho 0.28 0.53 0.64 ­0.12 Madagascar 0.17 8.51 ­5.30 ­1.15 Malawi 0.87 0.41 3.27 0.98 Mali 2.32 1.69 0.66 ­0.41 Mozambique ­0.43 0.62 4.22 ­0.90 Namibia 1.75 0.19 0.36 0.37 Niger ­0.17 1.39 2.04 ­0.19 Nigeria 0.53 ­0.30 5.14 ­0.20 Rwanda 0.39 6.15 2.24 0.40 Senegal 5.65 ­0.05 ­3.82 ­0.14 Tanzania 1.12 1.82 0.17 0.28 Uganda 0.54 0.71 4.37 0.14 Zambia 0.04 0.29 0.81 0.07 Zimbabwe 2.99 0.24 0.00 ­0.02 Country typology Resourch-rich 0.49 ­0.21 4.41 ­0.13 Middle-income 1.06 0.35 0.49 0.14 Fragile states 0.59 0.35 3.78 ­0.37 Nonfragile, low-income 0.95 2.30 1.14 ­0.53 Level of urbanization High 0.88 0.01 3.89 ­0.18 Medium 0.45 0.91 2.93 ­0.34 Low 0.81 2.31 1.29 ­0.55 Level of water scarcity High 0.77 1.12 3.02 ­0.42 Low 0.73 1.43 1.67 ­0.31 Overall 0.76 1.24 2.52 ­0.38 Source: Banerjee, Wodon, and others 2008. Access to Water Supply and Sanitation Facilities 291 Table A1.14 Annualized Change in Sanitation Access: Rural (percentage of population) Technology Improved Traditional Open Country Septic tank latrine latrine defecation Benin 0.07 0.98 0.37 0.93 Burkina Faso 0.11 1.68 ­0.27 1.61 Cameroon ­0.09 ­0.13 0.59 ­0.24 Chad 0.07 ­0.06 0.44 1.49 Congo, Dem. Rep. ­0.02 0.03 3.08 0.21 Côte d'Ivoire 0.06 ­1.27 3.92 ­0.40 Ethiopia 0.27 0.09 4.30 ­2.76 Ghana ­0.04 ­0.79 0.54 1.46 Guinea 0.07 0.28 2.11 ­0.93 Kenya 0.02 0.45 0.50 0.97 Lesotho ­0.05 1.11 ­0.53 0.60 Madagascar ­0.03 5.92 ­3.13 ­0.97 Malawi 0.04 0.11 2.47 ­0.24 Mali 0.59 0.63 1.72 ­0.81 Mozambique 0.01 ­0.08 1.69 ­0.80 Namibia 0.53 0.36 0.06 0.46 Niger 0.03 ­0.06 0.28 2.46 Nigeria 0.52 ­0.93 1.29 0.47 Rwanda 0.00 4.55 ­1.35 0.14 Senegal 1.67 ­2.08 2.64 ­1.00 Tanzania ­0.01 0.19 0.64 0.71 Uganda 0.03 0.13 3.90 0.42 Zambia 0.17 0.15 1.15 0.32 Zimbabwe ­0.05 1.79 0.90 ­1.55 Country typology Resourch-rich 0.41 ­0.73 1.17 0.45 Middle-income 0.25 0.72 ­0.22 0.52 Fragile states 0.00 0.05 2.85 ­0.25 Nonfragile, low-income 0.16 0.57 1.66 ­0.28 Level of urbanization High 0.43 ­0.95 1.47 0.36 Medium 0.02 0.31 2.12 ­0.17 Low 0.12 0.82 1.69 ­0.32 Level of water scarcity High 0.25 ­0.05 1.78 ­0.01 Low 0.13 0.38 1.57 ­0.14 Overall 0.20 0.11 1.70 ­0.06 Source: Banerjee, Wodon, and others 2008. 292 Africa's Water and Sanitation Infrastructure Reference Banerjee, S., Q. Wodon, A. Diallo, T. Pushak, H. Uddin, C. Tsimpo, and V. Foster. 2008. "Access, Affordability and Alternatives: Modern Infrastructure Services in Sub-Saharan Africa." AICD Background Paper 2, Africa Infrastructure Country Diagnostic, World Bank, Washington, D.C. APPENDIX 2 Institutions in the Water and Sanitation Sector 293 294 Table A2.1 Specification of Urban Water Reform Index Subindex Indicator Definition Legislation 1. Existence of reform 0 = No reform of the water services delivery; 1 = reform of the water services delivery 2. Legal reform 0 = No new sector legislation passed within the past 10 years; 1 = new sector legislation passed in the past 10 years Restructuring 3. Unbundling 0 = Same entity responsible for bulk water production and distribution in urban areas; 1 = different entities responsible for bulk water production and distribution in urban areas 4. Separation of business lines 0 = No separation of water and wastewater services from provision in urban area; 1 = separation of water and wastewater provisions in urban areas 5. SOE corporatization 0 = No state-owned water utility corporatized in urban area; 1 = at least one state-owned water utility corporatized 6. Existence of regulatory body 0 = No autonomous regulatory body; 1 = autonomous regulatory body Policy oversight 7. Tariff approval oversight 0 = Oversight on tariff by line ministry; 1 = oversight on tariff by a special entity within the ministry, an interministerial committee, or the regulator 8. Investment plan oversight 0 = Oversight on investment plan by line ministry; 1 = oversight on investment plan by a special entity within the ministry, an interministerial committee, or the regulator 9. Technical standard oversight 0 = Oversight on technical standards by line ministry; 1 = oversight on technical standards by a special entity within the ministry, an interministerial committee, or the regulator 10. Regulation monitoring oversight 0 = Oversight on compliance with economic regulation by line ministry; 1 = oversight on compliance with economic regulation by a special entity within the ministry, an interministerial committee, or the regulator 11. Dispute arbitration oversight 0 = Oversight on dispute arbitration by line ministry; 1 = oversight on dispute arbitration by a special entity within the ministry, an interministerial committee, or the regulator Private sector involvement 12. Private de jure 0 = Private participation forbidden by law; 1 = private participation allowed by law 13. Private de facto 0 = No private participation in the three largest utilities; 1 = at least a form of private participation in the three largest utilities 14. Private sector management 0 = No private sector involvement or service and works contracts only; 1 = management contract, affermage, lease, concession 15. Private sector investment 0 = No private sector involvement, service and works contracts, management contract, affermage, lease; 1 = concession 16. Absence of distressed private 0 = Canceled, distressed private sector participation; 1 = operational, concluded sector participation and not renewed private sector participation 17. Absence of renationalization 0 = Canceled; 1 = distressed, operational, concluded and not renewed private sec- tor participation 18. Private ownership 0 = Concession, management, lease contract; 1 = greenfield/divestiture Source: Banerjee, Wodon, and others 2008. Note: SOE = state-owned enterprise. 295 Table A2.2 Urban Water Reform Index 296 Legislation Restructuring Attribute 1. Existence of 4. Separation of 5. SOE 6. Existence of Country reform 2. Legal reform 3. Unbundling business lines corporatization regulatory body Benin 1 1 0 0 1 0 Burkina Faso 1 1 1 0 1 0 Cape Verde 1 1 0 0 1 1 Chad 0 0 0 0 1 0 Congo, Dem. Rep. 0 0 0 1 0 1 Côte d'Ivoire 1 0 0 1 1 1 Ethiopia 1 1 0 0 0 0 Ghana 1 1 0 1 1 1 Kenya 1 1 0 0 1 1 Lesotho 1 0 0 0 1 1 Madagascar 1 1 0 1 0 0 Malawi 1 0 0 1 1 0 Mozambique 1 1 0 1 1 1 Namibia 1 1 1 0 0 0 Niger 1 1 1 1 1 1 Nigeria 0 0 0 1 1 0 Rwanda 0 0 0 1 1 1 Senegal 1 1 0 1 1 0 South Africa 1 1 1 0 1 0 Sudan 1 1 0 1 1 1 Tanzania 1 1 0 1 1 1 Uganda 1 1 1 0 1 0 Zambia 1 1 0 0 1 1 Subindex Legislation Restructuring Countries sharing attribute (%) 83 70 22 52 83 52 Table A2.2 Urban Water Reform Index (continued) Policy oversight Attribute 7. Tariff approval 8. Investment plan 9. Technical standard 10. Regulation 11. Dispute arbitration Country oversight oversight oversight monitoring oversight oversight Benin 0 1 1 1 1 Burkina Faso 1 1 1 0 0 Cape Verde 1 1 1 1 1 Chad 1 0 0 1 0 Congo, Dem. Rep. 1 0 0 1 1 Côte d'Ivoire 1 1 1 1 1 Ethiopia 1 0 0 1 0 Ghana 1 0 0 0 1 Kenya 0 1 1 1 1 Lesotho 1 1 1 1 1 Madagascar 1 1 1 1 1 Malawi 0 0 0 0 1 Mozambique 1 0 0 1 1 Namibia 0 0 0 0 0 Niger 0 0 0 1 1 Nigeria 1 1 1 1 1 Rwanda 0 0 1 1 0 Senegal 0 1 1 1 0 South Africa 1 0 0 0 0 Sudan 1 1 0 1 1 Tanzania 0 0 1 1 1 Uganda 1 0 1 1 0 Zambia 1 1 1 1 1 Subindex Policy oversight Countries sharing 297 attribute (%) 65 48 57 78 65 (continued next page) 298 Table A2.2 Urban Water Reform Index (continued) Private sector involvement 16. Absence of 14. Private 15. Private distressed 17. Absence of Urban water Attribute 12. Private 13. Private sector sector reform index private sector renationaliza- 18. Private Country de jure de facto management investment participation tion ownership (%) Benin 0 0 0 0 51 Burkina Faso 1 0 0 0 58 Cape Verde 1 Chad 0 0 0 0 68 Congo, Dem. Rep. 0 1 1 0 0 76 Côte d'Ivoire 1 1 1 0 1 16 Ethiopia 1 0 0 0 38 Ghana 1 1 1 0 1 1 0 40 Kenya 1 1 1 0 1 1 0 74 Lesotho 0 0 0 0 78 Madagascar 1 1 0 0 1 50 Malawi 1 0 0 0 71 Mozambique 1 1 1 1 1 1 0 35 Namibia 1 0 0 0 1 0 84 Niger 1 1 1 0 1 1 0 36 Nigeria 1 1 0 0 1 80 Rwanda 1 1 1 0 0 53 Senegal 1 1 1 0 1 1 0 44 South Africa 1 1 1 0 1 1 0 73 Sudan 1 1 0 0 1 63 Tanzania 1 1 1 0 0 0 0 79 Uganda 1 1 1 0 1 1 0 74 Zambia 1 0 0 0 1 0 73 Subindex Private sector involvement Countries sharing attribute (%) 83 64 50 5 79 90 0 Source: Banerjee, Wodon, and others 2008. Note: SOE = state-owned enterprise. Blank cells: not applicable. 299 300 Table A2.3 Specification of Regulation Index Subindex Indicator Definition Autonomy 1. Formal autonomy: hire 0 = Appointment by government/line ministry; 1 = otherwise 2. Formal autonomy: fire 0 = Firing by government/line ministry; 1 = otherwise 3. Partial financial autonomy/operating 0 = Budget fully funded by government; 1 = at least a portion of budget funded through budget: central government fees and/or donors 4. Full financial autonomy/operating 0 = At least a portion of budget funded through government and/or donors; 1 = budget budget: sector levies fully funded through fees 5. Partial managerial autonomy/vetoing 0 = Veto decision by government/line ministry/others; 1 = no veto decision instance 6. Full managerial autonomy/vetoing 0 = Veto decision by government/line ministry/others; 1 = no veto decision instance 7. Multisectoral 0 = Sector specific regulator; 1 = multisectoral regulator 8. Commissioner 0 = Individual; 1 = board of commissioners Transparency 9. Publicity of decisions: reports only 0 = Regulatory decisions not publicly available; 1 = regulatory decisions publicly available through reports 10. Publicity of decisions: Internet only 0 = Regulatory decisions not publicly available or available only through reports; 1 = regulatory decisions publicly available through Internet 11. Publicity of decisions: public hearing 0 = Regulatory decisions not publicly available or available only through reports/Internet; only 1 = regulatory decisions publicly available through public hearings Accountability 12. Appeal 0 = No right to appeal regulatory decisions; 1 = right to appeal regulatory decision 13. Partial independence of appeal 0 = Appeal to government/line ministries; 1 = appeal to bodies other than government/line ministries 14. Full independence of appeal 0 = No recourse to independent arbitration; 1 = possibility to appeal to independent arbitration Tools 15. Tariff methodology 0 = No tariff methodology; 1 = some tariff methodology 16. Tariff indexation 0 = No tariff indexation 1 = some tariff indexation 17. Regulatory review 0 = No tariff review; 1 = periodic tariff review 18. Length of regulatory review 0 = No tariff review or review lower than every 3 years; 1 = multiyear tariff review (greater than or equal to 3) Source: Banerjee, Wodon, and others 2008. 301 302 Table A2.4 Regulation Index Attribute of regulatory agencies Autonomy 3. Partial financial 4. Full autonomy/ financial 5. Partial 6. Full Attribute of operating autonomy/ managerial managerial regulatory 1. Formal 2. Formal budget: operating autonomy/ autonomy/ agencies autonomy: autonomy: central budget: vetoing vetoing Country hire fire government sector levies instance instance 7. Multisectoral 8. Commissioner Benin 0 0 0 0 0 0 0 0 Burkina Faso 0 0 0 0 0 0 0 0 Cape Verde 0 0 1 1 1 0 1 1 Chad 0 0 0 0 0 0 0 0 Congo, Dem. Rep. 0 0 0 1 1 1 1 Côte d'Ivoire 0 0 0 0 0 0 0 0 Ethiopia 0 0 0 1 0 0 Ghana 0 0 0 0 1 1 Kenya 0 0 1 0 0 0 0 1 Lesotho 0 0 0 0 0 0 0 0 Madagascar 0 0 0 0 0 0 0 0 Malawi 0 0 0 0 0 0 0 0 Mozambique 0 0 1 1 1 0 0 1 Namibia 0 0 0 0 0 0 0 0 Niger 0 0 1 1 1 0 1 1 Nigeria 0 0 0 0 0 0 0 0 Rwanda 0 0 1 1 0 0 1 1 Senegal 0 0 0 0 0 0 0 0 South Africa 0 0 0 0 0 0 0 0 Sudan 0 0 1 0 0 0 0 1 Tanzania 0 0 1 0 0 0 1 1 Uganda 0 0 0 0 0 0 0 0 Zambia 0 0 1 0 1 0 0 1 Subindex Autonomy Countries sharing attribute (%) 0 0 36 19 26 4 27 43 (continued next page) 303 304 Table A2.4 Regulation Index (continued) Attribute of regulatory agencies Transparency Accountability Tools 11. 9. 10. Publicity Publicity Publicity of Attribute of of of decisions: 13. 14. 18. regulatory decisions: decisions: public Partial Full 15. 16. 17. Length of agencies reports Internet hearing 12. independence independence Tariff Tariff Regulatory regulatory Regulation Country only only only Appeal of appeal of appeal methodology indexation review review index (%) Benin 0 0 1 1 0 1 0 0 25 Burkina Faso 1 1 1 1 0 0 1 0 1 1 52 Cape Verde 1 1 1 1 1 0 1 0 1 1 76 Chad 0 0 0 1 0 0 1 33 Congo, Dem. Rep. 0 0 0 1 0 0 1 0 1 0 44 Côte d'Ivoire 1 0 0 1 1 0 1 0 1 1 35 Ethiopia 1 0 1 0 1 0 0 29 Ghana 1 1 1 0 1 0 0 0 1 1 54 Kenya 1 1 1 1 1 0 0 1 1 0 60 Lesotho 1 0 0 0 0 1 1 1 27 Madagascar 0 0 0 1 1 0 0 0 0 17 Malawi 1 0 1 0 0 1 25 Mozambique 1 0 0 1 1 0 0 1 1 1 56 Namibia 1 0 1 0 0 21 Niger 1 1 0 1 1 0 1 0 1 0 61 Nigeria 1 1 0 17 Rwanda 1 1 1 0 1 0 0 46 Senegal 1 1 1 1 1 0 1 1 1 0 60 South Africa 0 0 0 1 0 0 0 0 1 0 15 Sudan 1 0 0 1 0 0 1 1 1 48 Tanzania 1 1 1 1 1 0 1 1 0 68 Uganda 1 1 1 1 1 1 50 Zambia 1 1 1 1 1 0 0 0 1 59 Subindex Transparency Accountability Tools Countries sharing attribute (%) 81 48 50 79 69 0 60 41 70 54 Source: Banerjee, Wodon, and others 2008. 305 Table A2.5 Specification of SOE Governance Index 306 Subindex Indicator Definition Ownership and 1. Concentration of ownership 0 = Ownership diversified; 1 = 100% owned by one state body (central shareholder government or municipal government) quality 2. Corporatization 0 = Noncorporatized (uncorporatized state owned enterprise); 1 = corporatized 3. Limited liability 0 = Nonlimited liability; 1 = limited liability company 4. Rate of return policy 0 = No requirement to earn a rate of return; 1 = requirement to earn a rate of return 5. Dividend policy 0 = No requirement to pay dividends; 1 = requirement to pay dividends Managerial and 6. Hiring 0 = Either manager or board has not the most decisive influence on hiring board decisions; 1 = either manager or board has the most decisive influence on autonomy hiring decisions 7. Laying off 0 = Either manager or board has not the most decisive influence on firing decisions; 1 = either manager or board has the most decisive influence on firing decisions 8. Wages 0 = Either manager or board has not the most decisive influence on setting wages/bonuses; 1 = either manager or board has the most decisive influence on setting wages/bonuses 9. Production 0 = Either manager or board has not the most decisive influence on how much to produce; 1 = either manager or board has the most decisive influence on how much to produce 10. Sales 0 = Either manager or board has not the most decisive influence on what to sell; 1 = either manager or board has the most decisive influence on what to sell 11. Size of board 0 = Number of members of board lower than a given threshold (< 5); 1 = number of members of board greater than a given threshold (> 5) 12. Selection of board members 0 = Board members appointed only by government; 1 = board members appointed by shareholders (either group of shareholder; all shareholder; other) 13. Presence of independent directors 0 = No independent directors in the board; 1 = at least one independent director in the board Accounting and 14. Publication of annual reports 0 = Annual reports not publicly available; 1 = annual reports publicly available disclosure, 15. International Financial Reporting Standards 0 = IFRSs not applied; 1 = compliance to IFRSs performance (IFRSs) monitoring 16a. External audits/existence of financial 0 = No operational or financial audit; 1 = at least some form of external audit external audit 16b. External audits/existence of operational 0 = No operational or financial audit; 1 = at least some form of external audit external audit 17. Independent audit of accounts 0 = No independent audit of accounts; 1 = independent audit of accounts 18. Audit publication 0 = Audit not publicly available; 1 = audit not publicly available 19. Remuneration for noncommercial activities 0 = No remuneration of noncommercial activities; 1 = remuneration of noncommercial activities 20. Performance contracts 0 = No performance contracts; 1 = existence of performance contract 21. Performance contracts with performance- 0 = Performance-based incentive systems; 1 = existence of performance-based based incentive systems incentive systems 22. Penalties for poor performance 0 = No penalties for poor performance; 1 = penalties for poor performance 23. Monitoring 0 = No periodic monitoring of performance; 1 = periodic monitoring of performance (at least semiannual) 24. Third-party monitoring 0 = No monitoring of performance by third party (private sector auditor); 1 = monitoring of performance by third party (continued next page) 307 308 Table A2.5 (continued) Subindex Indicator Definition Outsourcing 25. Billing and collection 0 = No billing and collection outsourcing; 1 = billing and collection outsourcing 26. Meter reading 0 = No meter reading outsourcing; 1 = meter reading outsourcing 27. Human resources (HR) 0 = No HR outsourcing; 1 = HR outsourcing 28. Information technology (IT) 0 = No IT outsourcing; 1 = IT outsourcing Labor market 29. Restrictions to dismiss employees 0 = Restrictions to dismiss employees only within public service guidelines; discipline 1 = restrictions to dismiss employees according to corporate law or contract 30. Wages: compared with private sector 0 = Wages compared with public sector; 1 = wages compared with private sector (or between public and private sectors) 31. Benefits: versus private sector 0 = Benefits compared with public sector; 1 = benefits compared with private sector (or between public and private sectors) Capital market 32. No exemption from taxation 0 = Exemption from taxation; 1 = no exemption from taxation discipline 33. Access to debt: versus private sector 0 = Access to debt below the market rate; 1 = access to debt equal or above the market rate 34. No state guarantees 0 = At least one state guarantee; 1 = no state guarantee 35. Public listing 0 = No public listing; 1 = public listing Source: Banerjee, Wodon, and others 2008. Note: SOE = state-owned enterprise. Table A2.6 SOE Governance Index Attributes of SOEs Ownership and shareholder quality Managerial and board autonomy 4. 13. Rate 11. 12. Presence 1. 3. of 5. 7. Size Selection of Concentration 2. Limited return Dividend 6. Laying 8. 9. 10. of of board independent Country Utility of ownership Corporatization liability policy policy Hiring off Wages Production Sales board members directors Benin SONEB 1 1 0 0 0 1 1 1 1 1 1 0 0 Burkina Faso ONEA 1 1 0 1 1 1 1 1 1 1 1 0 0 Cape Verde ELECTRA 0 1 0 0 0 1 1 1 0 0 1 1 0 Chad STEE 1 1 0 0 0 1 1 1 1 1 0 1 0 Congo, Dem. Rep. REGIDESO 1 0 0 1 1 1 1 1 1 1 1 0 0 Côte d'Ivoire SODECI 0 1 0 1 1 1 1 0 0 0 Ethiopia ADAMA 1 0 0 1 0 1 1 1 1 1 0 0 0 AWSA 1 0 0 1 0 1 1 1 1 0 0 0 0 Dire Dawa 1 0 0 1 0 0 0 0 1 1 1 0 0 Ghana GWC 1 1 1 0 0 0 0 1 0 1 1 0 1 Kenya KIWASCO 1 1 1 1 0 0 1 0 0 0 0 1 1 MWSC 1 0 0 0 0 1 1 1 1 1 0 1 1 NWASCO 1 1 0 1 0 1 1 1 1 1 0 1 1 Lesotho WASA 1 1 0 1 0 1 1 1 1 1 1 0 1 Madagascar JIRAMA 1 0 0 0 0 1 1 1 1 0 1 1 0 Malawi BWB 1 1 0 1 1 0 0 1 1 1 0 0 0 CRWB 1 1 0 1 1 1 1 1 1 1 0 0 0 LWB 1 1 0 1 1 0 0 0 1 1 1 0 0 Mozambique AdeM Beira 1 1 0 0 0 1 1 1 1 1 0 AdeM Maputo 1 0 0 1 1 1 1 1 1 1 1 1 0 AdeM Nampula 1 1 0 0 0 1 1 1 1 1 0 AdeM Pemba 1 1 0 0 0 1 1 1 1 1 0 (continued next page) 309 310 Table A2.6 SOE Governance Index (continued) Attributes of SOEs Ownership and shareholder quality Managerial and board autonomy 4. 13. Rate 11. 12. Presence 1. 3. of 5. 7. Size Selection of Concentration 2. Limited return Dividend 6. Laying 8. 9. 10. of of board independent Country Utility of ownership Corporatization liability policy policy Hiring off Wages Production Sales board members directors AdeM Quelimane 1 1 0 0 0 1 1 1 1 1 0 Namibia Oshakati Municipality 1 0 0 0 0 0 0 Walvis Bay Municipality 1 0 0 1 1 1 1 1 Windhoek Municipality 1 0 0 0 0 0 1 0 Niger SEEN 0 1 1 1 1 1 0 0 1 1 1 SPEN 1 1 0 0 0 1 1 1 0 0 1 0 0 Nigeria Borno 0 1 0 0 0 1 1 1 0 0 1 FCT 1 0 0 0 0 0 0 0 1 1 Kaduna 0 1 0 0 0 0 1 1 0 0 0 1 0 Katsina 0 1 0 0 0 0 1 1 0 0 0 1 0 Lagos 0 1 0 0 0 0 1 1 0 0 1 Plateau 0 1 0 0 0 0 1 1 0 0 1 Rwanda ELECTROGAZ 1 0 0 1 1 1 1 1 1 1 1 0 0 Senegal ONAS 1 0 0 0 0 1 1 1 0 0 1 1 SDE 0 1 1 1 0 1 0 0 0 1 0 South Africa Cape Town Metro 1 0 0 0 0 0 0 1 0 0 Drakenstein Municipality 1 0 0 0 0 0 0 1 0 0 eThekwini Metro (Durban) 1 0 0 0 0 0 0 1 0 0 Joburg 1 1 1 0 0 0 0 1 0 0 1 0 Sudan Khartoum Water Corporation 0 1 0 1 0 1 0 1 1 1 0 0 0 South Darfur Water Corporation 0 1 0 1 1 1 1 1 1 1 0 0 1 Upper Nile Water Corporation 0 0 0 0 0 0 0 0 1 1 1 1 0 Tanzania DAWASCO 1 1 0 1 0 1 1 1 1 1 1 0 1 DUWS 1 1 0 0 0 1 1 1 1 0 0 0 1 MWSA 1 0 0 1 0 1 1 1 1 1 0 0 0 Uganda SONEB 1 1 0 1 0 1 1 1 1 1 0 0 1 Zambia LWSC 1 1 1 1 1 1 1 1 1 1 1 0 1 NWSC 1 0 0 1 1 1 1 1 1 1 0 1 1 SWSC 1 1 1 1 0 0 1 0 0 0 1 1 Subindex Ownership and shareholder quality Managerial and board autonomy % of utilities sharing attribute 76 65 11 45 27 65 71 86 64 57 49 40 40 (continued next page) 311 Table A2.6 SOE Governance Index (continued) 312 Attributes of SOEs Accounting and disclosure, performance monitoring 16a. 16b. 21. External External Performance audits/ audits/ contracts existence existence 19. with 22. 14. of of 17. Remuneration performance- Penalties Publication financial operational Independent 18. for 20. based for 24. of annual 15. external external audit of Audit noncommercial Performance incentive poor 23. Third-party Country Utility reports IFRSs audit audit accounts publication activities contracts systems performance Monitoring monitoring Benin SONEB 1 1 1 1 1 1 0 0 0 0 1 1 Burkina Faso ONEA 1 1 1 1 1 0 0 1 0 1 1 0 Cape Verde ELECTRA 1 0 1 1 1 1 0 0 0 0 1 0 Chad STEE 0 0 1 1 1 0 0 1 0 1 0 1 Congo, Dem. Rep. REGIDESO 1 1 1 0 1 1 1 1 0 0 0 0 Côte d'Ivoire SODECI 0 1 1 1 1 1 1 0 1 Ethiopia ADAMA 0 0 1 0 0 0 0 1 1 1 4 AWSA 0 1 1 1 1 1 1 0 1 1 1 0 Dire Dawa 0 0 0 0 0 0 0 0 0 1 1 Ghana GWC 1 1 1 1 1 0 1 0 1 1 0 Kenya KIWASCO 1 1 1 1 1 1 0 1 1 1 1 1 MWSC 0 0 1 1 1 1 0 0 0 0 1 1 NWASCO 1 1 1 1 1 1 0 0 0 0 1 1 Lesotho WASA 1 1 1 1 1 1 0 0 0 1 0 Madagascar JIRAMA 0 1 1 0 0 1 0 0 0 0 1 Malawi BWB 1 1 1 1 1 1 0 1 1 1 1 1 CRWB 1 0 1 0 1 1 0 0 0 1 1 1 LWB 1 1 1 0 1 1 0 0 1 1 1 1 Mozambique AdeM Beira 1 1 1 1 1 1 0 0 1 0 1 1 AdeM Maputo 1 0 1 1 1 0 1 1 0 0 0 1 AdeM Nampula 1 1 1 1 1 1 0 0 1 0 1 1 AdeM Pemba 1 1 1 1 1 1 0 0 1 0 1 1 AdeM Quelimane 1 1 1 1 1 1 0 0 1 0 1 1 Namibia Oshakati Municipality 0 0 1 0 1 1 0 0 0 0 1 Walvis Bay Municipality 1 0 1 0 1 1 0 1 0 0 1 Windhoek Municipality 1 0 1 1 1 1 0 0 1 1 1 1 Niger SEEN 1 1 1 1 1 0 0 1 1 0 0 0 SPEN 1 1 1 1 1 0 0 1 1 0 0 1 Nigeria Borno 0 1 1 0 1 0 0 0 0 0 0 1 FCT 1 1 1 0 1 0 0 0 0 1 1 1 Kaduna 0 1 1 0 1 0 0 0 0 0 0 1 Katsina 0 1 1 0 1 0 0 0 0 0 0 1 Lagos 0 1 1 0 1 0 0 0 0 0 0 1 Plateau 0 1 1 0 1 0 0 0 0 0 0 1 Rwanda ELECTROGAZ 0 0 1 1 0 0 0 1 0 0 0 1 Senegal ONAS 0 1 1 1 1 1 0 1 0 1 0 SDE 0 1 1 1 1 0 1 1 1 1 1 1 South Africa Cape Town Metro 1 0 1 0 1 1 1 1 1 0 0 0 Drakenstein Municipality 1 0 1 0 1 1 1 1 1 0 0 0 eThekwini Metro (Durban) 1 0 1 0 1 1 1 1 1 0 0 0 Joburg 1 1 1 0 1 1 1 1 1 1 1 Sudan Khartoum Water Corporation 1 0 1 0 1 0 1 1 1 0 0 0 313 (continued next page) 314 Table A2.6 SOE Governance Index (continued) Attributes of SOEs Accounting and disclosure, performance monitoring 16a. 16b. 21. External External Performance audits/ audits/ contracts existence existence 19. with 22. 14. of of 17. Remuneration performance- Penalties Publication financial operational Independent 18. for 20. based for 24. of annual 15. external external audit of Audit noncommercial Performance incentive poor 23. Third-party Country Utility reports IFRSs audit audit accounts publication activities contracts systems performance Monitoring monitoring South Darfur Water Corporation 0 0 1 0 0 1 0 1 0 0 0 Upper Nile Water Corporation 0 0 0 0 0 0 0 0 0 0 0 Tanzania DAWASCO 1 1 1 1 1 1 1 1 1 1 1 0 DUWS 1 1 1 1 1 0 1 1 1 1 1 1 MWSA 1 1 1 1 1 1 1 1 1 1 0 1 Uganda SONEB 1 1 1 1 1 1 1 1 1 1 1 0 Zambia LWSC 1 1 1 1 1 1 0 1 0 0 1 1 NWSC 1 1 1 1 1 0 0 1 1 0 1 1 SWSC 1 1 1 1 1 1 0 0 1 1 1 0 Subindex Accounting and disclosure, performance monitoring % of utilities sharing attribute 65 67 96 57 88 64 32 49 51 39 57 65 Table A2.6 SOE Governance Index (continued) Attributes of SOEs Outsourcing Labor market discipline Capital market discipline 33. Access 29. 30. 31. 32. to 25. Restrictions Wages: Benefits: No debt: 34. Billing 26. 27. 28. to compared versus exemption versus No 35. SOE and Meter Human Information dismiss with private private from private state Public governance Country Utility collection reading resources technology employees sectror sector taxation sector guarantees listing index Benin SONEB 0 0 0 0 1 1 1 0 1 1 0 55 Burkina Faso ONEA 0 0 0 0 1 1 1 1 0 0 0 58 Cape Verde ELECTRA 1 0 0 0 0 1 1 0 0 0 0 37 Chad STEE 0 0 0 0 0 1 1 0 1 0 44 Congo, Dem. Rep. REGIDESO 0 1 1 0 0 0 0 52 Côte d'Ivoire SODECI 0 0 0 0 1 1 1 1 1 63 Ethiopia ADAMA 0 0 0 0 1 1 0 45 AWSA 1 0 0 0 1 1 0 50 Dire Dawa 0 0 0 0 1 1 0 29 Ghana GWC 0 0 0 0 0 1 1 0 0 0 0 42 Kenya KIWASCO 0 0 0 0 1 1 0 0 1 0 56 MWSC 0 0 0 0 0 1 1 0 0 0 0 37 NWASCO 0 0 0 0 1 1 0 0 1 0 57 Lesotho WASA 0 0 0 0 1 1 1 1 0 0 0 56 Madagascar JIRAMA 0 0 1 0 0 1 1 1 1 0 0 46 Malawi BWB 0 0 0 0 0 1 1 1 1 0 0 54 CRWB 0 1 1 0 0 0 0 54 LWB 0 0 0 0 1 1 1 1 1 0 0 57 (continued next page) 315 Table A2.6 SOE Governance Index (continued) 316 Attributes of SOEs Outsourcing Labor market discipline Capital market discipline 33. Access 29. 30. 31. 32. to 25. Restrictions Wages: Benefits: No debt: 34. Billing 26. 27. 28. to compared versus exemption versus No 35. SOE and Meter Human Information dismiss with private private from private state Public governance Country Utility collection reading resources technology employees sectror sector taxation sector guarantees listing index Mozambique AdeM Beira 1 1 1 1 0 1 1 1 1 0 0 69 AdeM Maputo 0 0 0 1 1 1 1 1 1 1 0 68 AdeM Nampula 1 1 1 1 0 1 1 1 1 0 0 69 AdeM Pemba 1 1 1 1 0 1 1 1 1 0 0 69 AdeM Quelimane 1 1 1 1 0 1 1 1 1 0 0 69 Namibia Oshakati Municipality 0 0 0 0 0 0 0 1 0 20 Walvis Bay Municipality 0 0 0 1 0 0 0 0 1 1 0 44 Windhoek Municipality 0 0 0 0 0 0 1 1 0 0 0 31 Niger SEEN 0 0 0 1 0 1 1 1 1 1 0 61 SPEN 0 0 0 1 0 1 1 1 0 0 0 46 Nigeria Borno 0 0 0 0 0 0 1 1 1 0 0 34 FCT 0 0 0 1 1 1 0 39 Kaduna 1 0 0 0 0 0 1 1 1 0 0 33 Katsina 0 0 0 0 0 0 1 1 1 0 0 29 Lagos 1 0 0 0 0 1 1 1 1 0 0 41 Plateau 0 0 0 0 0 0 1 1 1 0 0 31 Rwanda ELECTROGAZ 0 0 0 1 0 1 0 1 1 1 0 50 Senegal ONAS 0 0 0 0 1 1 1 0 1 0 51 SDE 0 0 0 0 1 1 1 1 1 1 0 60 South Africa Cape Town Metro 0 0 0 1 1 1 0 35 Drakenstein Municipality 0 0 0 1 1 1 0 35 eThekwini Metro (Durban) 0 0 0 1 1 1 0 35 Joburg 1 1 0 1 0 1 1 1 1 1 0 66 Sudan Khartoum Water Corporation 1 1 1 1 0 1 1 1 1 0 0 59 South Darfur Water Corporation 0 1 1 1 1 0 0 56 Upper Nile Water Corporation 0 0 0 1 0 0 17 Tanzania DAWASCO 0 0 1 0 0 1 0 0 1 0 0 54 DUWS 0 0 0 0 0 1 1 0 1 1 0 52 MWSA 0 0 0 0 0 0 0 0 1 0 0 37 Uganda SONEB 0 0 1 0 1 0 0 1 0 0 0 52 Zambia LWSC 0 0 0 0 1 1 1 1 1 1 0 73 NWSC 1 1 1 1 1 0 0 75 SWSC 0 0 0 0 1 1 1 0 0 0 52 Subindex Outsourcing Labor market discipline Capital market discipline % of utilities sharing attribute 23 15 21 28 25 67 73 67 70 42 2 Source: Banerjee, Wodon, and others 2008. Note: IFRSs = International Financial Reporting Standards; SOE = state-owned enterprise. 317 318 Africa's Water and Sanitation Infrastructure Table A2.7 Specification of Rural Water Reform Index Specification Definition Rural water agency Is there a specialized rural water agency? Yes = 1, no = 0 Rural water policy Is there a specific policy or strategy for the rural Yes = 1; no = 0 water sector? Map of rural water points Is there a current map of the rural water points? Yes = 1; no = 0 Dedicated budget/fund Is there funding available to specifically support Yes = 1; no = 0 rural water services? Cost-recovery policy Is there a cost-recovery policy for rural water Yes = 1; no = 0 services? Source: Banerjee, Wodon, and others 2008. Table A2.8 Rural Water Reform Index Rural Map of Water Rural Rural rural Dedicated Cost- Reform water water water budget/ recovery Index agency policy points fund policy (%) Benin 1 1 1 0 1 67 Burkina Faso 1 1 1 1 1 83 Cape Verde 0 1 1 0 0 33 Chad 0 1 0 1 1 50 Congo, Dem. Rep. 1 0 0 1 0 33 Côte d'Ivoire 1 1 1 1 1 83 Ethiopia 0 1 0 1 1 50 Ghana 1 1 0 1 1 67 Kenya 0 0 0 1 1 33 Lesotho 1 1 0 1 0 50 Madagascar 0 1 1 1 1 67 Malawi 0 1 0 1 0 33 Mozambique 1 1 0 1 1 67 Namibia 1 1 0 1 1 67 Niger 0 1 0 0 0 17 Nigeria 1 1 0 1 1 67 Rwanda 0 1 0 1 1 50 Senegal 1 1 0 1 1 67 South Africa 0 0 0 1 1 33 Sudan 0 1 0 1 1 50 Tanzania 0 1 1 1 1 67 Uganda 1 1 1 1 1 83 Zambia 0 1 0 1 0 33 % of countries sharing attribute 46 83 29 83 71 Source: Banerjee, Wodon, and others 2008. Institutions in the Water and Sanitation Sector 319 Table A2.9 Specification of On-Site Sanitation Index Indicator Definition Existence of an accepted 1 = Existence of accepted definition; 0 = No accepted definition of sanitation definition Existence of sanitation policy 1 = Existence of policy/strategy; 0 = No policy/strategy Existence of hygiene 1 = Existence of hygiene promotion program by the promotion program government; 0 = No hygiene promotion program by the government Households responsible for 1 = Households responsible for financing sanitation invest- investment finance ments; 0 = Otherwise Government/private sector/ 1 = Either municipal government/private sector/water utility/NGO/CBO utility, NGO/CBO responsible for technical assistance; responsible for technical 0 = Otherwise assistance Government/private sector/ 1 = Either municipal government/private sector/water utility responsible utility responsible for desludging; 0 = Otherwise for desludging Government responsible 1 = Either central/local/municipal government responsible for regulation for regulation; 0 = Otherwise Existence of cost-recovery 1 = Requirement for cost recovery; 0 = No requirement requirement for on-site for cost recovery sanitation Source: Morella, Foster, and Banerjee 2008. Note: CBO = community-based organization; NGO = nongovernmental organization. 320 Table A2.10 On-Site Sanitation Index Existence of Existence of Existence of Involvement Existence of cost-recovery an accepted Existence hygiene of utilities in a specific requirement On-site definition of of sanitation promotion on-site fund for for on-site sanitation Country sanitation policy program sanitation sanitation sanitation index (%) Zambia 0 0 0 0 0 0 0 Nigeria 1 0 0 0 0 20 Congo, Dem. Rep. 1 0 0 0 1 0 33 Lesotho 0 1 1 0 0 0 33 Niger 1 0 0 1 0 0 33 Benin 1 0 1 1 0 0 50 Ghana 1 1 0 0 50 Malawi 1 0 1 0 0 1 50 Mozambique 1 1 1 0 0 0 50 Rwanda 1 1 1 0 0 0 50 Sudan 1 0 1 1 0 0 50 Ethiopia 1 0 1 1 1 0 67 Senegal 1 1 1 1 0 0 67 Uganda 1 1 1 1 0 0 67 Namibia 1 1 1 1 0 80 Cape Verde 1 1 1 1 0 1 83 Côte d'Ivoire 1 1 1 0 1 1 83 Tanzania 1 1 1 1 1 0 83 Burkina Faso 1 1 1 1 1 1 100 Chad 1 1 1 100 Kenya 1 1 1 1 1 1 100 Madagascar 1 1 1 1 1 1 100 South Africa 1 1 1 1 1 1 100 % of countries sharing attribute 91 65 82 59 36 37 Source: Morella, Foster, and Banerjee 2008. 321 322 Africa's Water and Sanitation Infrastructure Reference Banerjee, S., Q. Wodon, A. Diallo, T. Pushak, H. Uddin, C. Tsimpo, and V. Foster. 2008. "Access, Affordability and Alternatives: Modern Infrastructure Services in Sub-Saharan Africa." AICD Background Paper 2, Africa Infrastructure Country Diagnostic, World Bank, Washington, D.C. Morella, E., V. Foster, and S. Banerjee. 2008. "Climbing the Ladder: The State of Sanitation in Sub-Saharan Africa." AICD Background Paper 13. World Bank, Washington, DC. APPENDIX 3 Performance Indicators of Selected Water Utilities 323 324 Table A3.1 Access to Utility Water Access, private residential water connection (% of population) Country Utility 1995­99 2000 2001 2002 2003 2004 2005 Benin SONEB 24.6 25.0 Botswana DWA 35.0 WUC Burkina Faso ONEA 15.4 16.3 21.9 22.5 22.8 23.2 24.8 Cameroon SNEC Cape Verde ELECTRA 34.3 36.7 37.9 42.2 45.2 46.3 Chad STEE Congo, Dem. Rep. REGIDESO 19.0 Congo, Rep. SDNE 22.8 24.2 Côte d'Ivoire SODECI 28.5 30.2 29.7 29.4 30.1 30.2 29.9 Ethiopia ADAMA 28.2 30.9 32.3 AWSA Dire Dawa 17.8 18.9 Gabon SEEG Ghana GWC 9.0 9.1 8.5 9.0 8.8 Kenya KIWASCO 11.4 10.0 MWSC 32.5 36.3 33.7 34.2 34.5 NWASCO 37.4 50.9 Lesotho WASA 33.7 Liberia LWSR 2.7 3.0 Madagascar JIRAMA 11.6 11.4 11.7 11.7 12.2 12.6 12.7 Malawi BWB 23.0 22.9 22.5 22.4 25.0 24.0 25.3 CRWB 17.8 18.0 17.6 17.9 18.8 20.4 LWB 33.2 32.1 35.6 Mali EDM 18.6 20.1 24.7 25.6 26.3 27.2 Mauritania MSNE 59.1 63.0 Mozambique AdeM Beira 11.5 11.6 11.4 10.7 Adem Maputo 23.5 23.8 25.2 25.5 Adem Nampula 7.3 8.1 7.7 8.1 AdeM Pemba 14.5 14.6 15.1 15.8 AdeM Quelimane 4.9 4.8 4.2 4.7 Namibia Oshakati Municipality 49.3 Walvis Bay Municipality 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Windhoek Municipality 79.4 78.1 76.6 74.9 74.3 73.1 Niger SPEN/SEEN Nigeria Borno FCT 10.0 Kaduna 50.0 51.4 50.8 50.8 50.8 48.2 Katsina Lagos 3.0 3.0 2.9 3.0 Plateau Rwanda ELECTROGAZ 11.2 11.2 11.2 11.2 11.2 11.2 11.2 Senegal SDE 56.9 57.9 60.8 61.5 63.7 65.8 Seychelles PUC 96.5 96.9 South Africa Cape Town Metro 85.6 86.6 89.0 90.6 92.4 Drakenstein Municipality 87.0 87.9 88.7 89.5 90.3 eThekwini Metro (Durban) 94.6 94.9 95.2 95.5 90.1 Joburg 85.2 86.1 86.9 87.6 88.4 Sudan Khartoum Water Corporation 26.8 South Darfur Water Corporation 10.5 Upper Nile Water Corporation 38.4 325 (continued next page) 326 Table A3.1 (continued) Access, private residential water connection (% of population) Country Utility 1995­99 2000 2001 2002 2003 2004 2005 Tanzania Arusha Babati Bukoba DAWASCO DUWS 33.9 Iringa Kigoma Lindi Mbeya Morogoro Moshi Mtwara Musoma MWSA 16.5 21.1 20.1 Shinyanga Singida Songea Sumbawanga Tabora Tanga Togo TdE 38.9 39.9 40.9 40.7 40.5 Uganda NWSC 15.1 16.0 17.1 21.4 22.8 27.1 Zambia AHC-MMS CHWSC CWSC KWSC LukangaWSC LWSC 16.3 22.1 20.5 24.1 22.3 18.5 23.4 MulongaWSC NorthWesternWSC NWSC 55.1 60.1 SWSC 36.5 36.1 40.6 53.2 58.9 WesternWSC Country typology Resourch-rich 30.4 Middle-income 76.0 Fragile states 17.3 Nonfragile, low-income 25.5 Level of water scarcity High 37.4 Low 36.1 Size of the utility Small 35.5 Large 37.5 Overall 36.8 Source: Banerjee, Skilling, and others 2008. 327 328 Table A3.2 Distribution Infrastructure Kilometers of water mains per 1,000 population Kilometers of water mains per 1,000 water connections (km/1,000 capita) (km/1,000 connections) Country Utility 1995­99 2000 2001 2002 2003 2004 2005 1995­99 2000 2001 2002 2003 2004 2005 Benin SONEB 1.4 1.4 Botswana DWA WUC Burkina Faso ONEA 0.9 0.9 1.1 1.1 1.1 1.2 1.2 36.2 36.1 32.1 32.6 32.5 33.5 32.7 Cameroon SNEC Cape Verde ELECTRA 1.6 1.6 1.5 1.5 1.4 1.4 Chad STEE Congo, Dem. Rep. REDIGESO 0.7 50.1 Congo, Rep. SDNE Côte d'Ivoire SODECI 1.5 1.5 1.4 1.4 1.4 1.4 1.3 21.9 Ethiopia ADAMA 0.7 9.6 AWSA 0.7 0.7 0.7 0.7 0.7 0.7 0.7 8.4 Dire Dawa Gabon SEEG 2.5 2.4 2.5 2.6 2.5 2.4 2.3 18.1 17.4 17.9 18.7 18.0 17.5 16.8 Ghana GWC 0.3 0.4 0.4 0.4 0.5 Kenya KIWASCO 0.2 0.2 13.7 14.7 MWSC NWASCO 1.0 1.0 11.3 10.6 Lesotho WASA Liberia LWSR 0.1 0.1 25.0 22.5 Madagascar JIRAMA 0.6 0.6 0.6 0.6 0.6 0.6 0.6 24.0 24.3 23.7 23.7 22.9 22.3 22.1 Malawi BWB 1.5 1.4 1.4 1.3 1.3 1.3 1.2 24.7 24.3 24.0 23.3 23.8 24.3 22.6 CRWB 3.7 3.6 3.6 4.8 5.1 5.6 115.9 112.9 113.7 151.6 155.8 160.5 LWB 1.8 1.8 1.7 1.7 1.7 1.6 1.6 45.7 43.8 Mali EDM Mauritania MSNE Mozambique AdeM Beira 0.1 0.1 0.1 0.1 4.0 3.8 3.8 3.9 Adem Maputo 0.5 0.5 0.4 0.4 Adem Nampula 0.2 0.2 0.2 0.2 11.2 10.1 11.2 10.9 AdeM Pemba 0.3 0.3 0.3 0.3 10.9 10.5 9.8 9.1 AdeM Quelimane 0.2 0.2 0.2 0.2 19.6 19.0 20.8 17.6 Namibia Oshakati Municipality 0.4 Walvis Bay Municipality 7.0 6.7 6.4 6.2 6.0 5.8 5.5 36.8 35.6 30.3 29.6 28.2 28.3 27.1 Windhoek Municipality 4.5 29.9 Niger SPEN/SEEN 1.0 1.0 1.0 1.0 1.0 1.1 1.1 30.8 30.6 28.3 29.7 29.8 Nigeria Borno 0.2 0.2 0.2 0.2 7.3 7.3 7.2 6.5 FCT Kaduna 0.7 0.7 0.7 0.7 0.7 23.5 24.3 Katsina 0.3 0.2 0.2 0.2 16.6 18.0 Lagos 0.2 0.2 0.1 0.1 16.5 16.3 16.0 14.5 Plateau 1.3 1.1 1.1 61.9 62.8 67.9 Rwanda ELECTROGAZ 1.3 1.3 1.4 1.2 1.2 1.1 1.2 60.4 Senegal SDE 1.5 1.5 1.5 1.5 1.5 1.5 24.3 23.6 22.8 22.3 21.8 20.8 Seychelles PUC 3.9 3.9 14.7 14.3 South Africa Cape Town Metro Drakenstein Municipality 3.0 3.0 2.9 2.8 2.8 13.4 13.0 12.7 12.3 12.0 eThekwini Metro (Durban) 4.0 3.9 3.9 3.8 3.7 17.6 17.2 Joburg Sudan Khartoum Water Corporation 0.4 10.6 South Darfur Water Corporation 0.1 12.1 Upper Nile Water Corporation 0.2 3.2 329 (continued next page) 330 Table A3.2 (continued) Kilometers of water mains per 1,000 population Kilometers of water mains per 1,000 water connections (km/1,000 capita) (km/1,000 connections) Country Utility 1995­99 2000 2001 2002 2003 2004 2005 1995­99 2000 2001 2002 2003 2004 2005 Tanzania Arusha 10.9 Babati 17.3 Bukoba 17.5 DAWASCO DUWS 0.7 4.7 Iringa 16.1 Kigoma 1.8 1.2 25.2 25.7 Lindi 43.8 Mbeya 16.7 Morogoro 15.8 Moshi 20.5 Mtwara 38.6 Musoma 18.0 MWSA 0.6 0.6 0.6 15.8 15.1 16.1 Shinyanga 30.2 Singida 25.9 Songea 29.7 Sumbawanga 35.9 Tabora 25.6 Tanga 22.4 Togo TdE 1.9 1.8 1.8 1.8 1.8 40.7 41.1 39.9 39.6 39.1 Uganda NWSC 0.9 0.8 0.9 1.0 1.2 1.3 28.3 27.2 25.1 23.9 24.5 26.4 22.6 Zambia AHC-MMS CHWSC CWSC KWSC LukangaWSC LWSC 1.5 1.9 1.9 2.1 1.8 1.4 1.5 67.8 62.6 58.7 54.3 46.2 MulongaWSC NorthWestern WSC NWSC 1.7 1.7 24.8 23.1 SWSC 1.4 1.4 1.4 18.7 17.0 WesternWSC Country typology Resourch-rich 0.9 23.0 Middle-income 3.2 20.1 Fragile states 0.7 31.5 Nonfragile, low-income 1.0 25.2 Level of water scarcity High 1.2 29.6 Low 1.4 22.1 Size of the utility Small 1.4 25.9 Large 1.2 19.5 Overall 1.3 24.7 (continued next page) 331 332 Table A3.2 (continued) Metering ratio (%) Nonrevenue water (%) Country Utility 1995­99 2000 2001 2002 2003 2004 2005 1995­99 2000 2001 2002 2003 2004 2005 Benin SONEB 25.6 23.8 Botswana DWA 23.9 21.4 23.4 24.4 24.4 26.1 27.9 WUC 14.1 12.6 Burkina Faso ONEA 97.4 97.5 97.9 97.9 98.0 98.1 98.2 19.5 17.1 15.9 14.0 15.2 17.0 18.3 Cameroon SNEC 37.0 Cape Verde ELECTRA 26.4 23.5 28.4 29.7 30.3 31.2 Chad STEE Congo, Dem. Rep. REDIGESO 28.6 45.2 38.8 36.4 38.5 44.2 37.6 40.7 Congo, Rep. SDNE 18.6 18.7 17.9 19.3 19.5 21.0 27.7 27.7 27.8 27.7 27.8 27.7 Côte d'Ivoire SODECI 100.0 14.2 17.7 17.5 18.8 20.2 21.7 21.7 Ethiopia ADAMA 90.2 90.1 42.7 AWSA 24.0 32.2 34.3 30.0 26.9 33.6 36.8 DIRE DAWA 21.6 Gabon SEEG 13.3 15.8 14.9 15.5 16.7 16.3 17.6 Ghana GWC 52.0 58.0 57.0 53.0 Kenya KIWASCO 48.6 58.2 68.3 71.4 MWSC 52.6 41.8 40.7 34.8 38.3 NWASCO 40.0 37.8 Lesotho WASA 28.1 27.9 27.8 Liberia LWSR 52.5 65.1 7.0 28.8 Madagascar JIRAMA 97.4 97.2 97.1 97.1 97.1 97.1 97.1 31.2 32.5 32.4 35.9 36.0 32.8 33.5 Malawi BWB 42.7 32.9 22.6 34.3 35.6 34.0 45.8 43.6 47.0 51.1 CRWB 28.6 26.2 26.3 28.5 18.7 16.7 LWB 98.1 39.6 32.9 39.2 16.8 17.1 16.6 22.1 Mali EDM 96.0 96.0 96.0 96.0 36.7 32.1 29.8 26.7 Mauritania MSNE 100.0 99.9 30.4 32.0 Mozambique AdeM Beira 68.0 99.2 98.5 99.9 52.1 54.2 53.1 60.1 Adem Maputo 100.3 99.8 99.3 98.2 57.8 62.4 54.4 62.1 Adem Nampula 100.0 100.0 99.8 100.0 27.2 43.2 45.1 44.1 AdeM Pemba 100.8 102.6 97.7 99.1 50.9 52.9 51.2 45.0 AdeM Quelimane 108.1 100.7 113.7 100.0 26.5 26.3 36.8 35.2 Namibia Oshakati Municipality 12.4 28.9 24.5 34.7 28.2 20.8 Walvis Bay Municipality 100.0 100.0 100.0 100.0 100.0 100.0 100.0 16.4 25.8 27.1 18.1 11.5 10.7 16.0 Windhoek Municipality 19.8 18.1 18.4 20.2 10.5 13.8 Niger SPEN/SEEN 96.3 96.2 96.9 97.1 96.8 15.8 20.9 22.6 17.2 17.4 16.9 18.8 Nigeria Borno FCT 23.6 80.0 Kaduna 7.7 16.1 39.1 38.3 51.1 68.4 58.0 21.2 Katsina 3.2 6.5 30.0 29.0 56.5 14.4 Lagos 67.1 66.7 60.4 56.5 Plateau 5.8 7.2 23.6 27.6 33.2 33.3 23.5 Rwanda ELECTROGAZ 98.7 44.6 54.2 48.9 44.6 50.6 43.8 38.3 Senegal SDE 111.6 116.2 112.9 115.9 116.6 117.3 25.6 22.3 21.5 20.1 19.9 20.1 Seychelles PUC 46.1 45.0 16.7 20.3 South Africa Cape Town Metro 60.3 10.0 20.2 36.5 16.4 18.0 Drakenstein Municipality 60.7 12.9 12.9 12.3 14.3 11.6 eThekwini Metro (Durban) 57.1 66.4 30.1 30.9 31.2 29.1 32.1 Joburg 52.4 39.4 43.7 39.3 32.8 30.9 Sudan Khartoum Water Corporation 40.0 South Darfur Water Corporation 48.9 Upper Nile Water Corporation 29.0 Tanzania Arusha 90.1 100.0 34.0 34.6 333 (continued next page) 334 Table A3.2 (continued) Metering ratio (%) Nonrevenue water (%) Country Utility 1995­99 2000 2001 2002 2003 2004 2005 1995­99 2000 2001 2002 2003 2004 2005 Babati 22.3 50.6 Bukoba 63.9 62.0 60.0 60.0 DAWASCO DUWS 27.9 31.0 31.0 Iringa 58.5 74.4 54.0 53.0 Kigoma 36.1 34.8 40.0 49.0 Lindi 11.1 18.4 86.0 75.0 Mbeya 35.5 38.7 37.0 43.0 Morogoro 64.9 91.8 39.0 37.5 Moshi 100.0 82.8 60.0 33.0 Mtwara 66.8 74.7 52.0 43.0 Musoma 23.7 42.2 66.0 63.0 MWSA 104.5 100.0 100.0 57.0 50.0 48.9 Shinyanga 22.8 45.9 46.0 39.0 Singida 33.8 36.9 56.0 49.8 Songea 13.3 32.9 31.0 34.0 Sumbawanga 33.8 38.4 50.0 48.0 Tabora 59.4 68.7 28.0 28.0 Tanga 100.0 99.5 34.0 34.0 Togo TdE 100.0 100.0 100.0 100.0 100.0 28.4 35.4 26.5 24.3 28.0 Uganda NWSC 79.2 83.1 84.3 89.1 91.9 93.6 94.5 43.5 42.6 40.4 39.2 38.2 34.5 Zambia AHC-MMS 15.0 32.0 CHWSC 60.0 CWSC 98.0 29.0 KWSC 7.0 57.0 LukangaWSC LWSC 32.6 33.6 38.6 38.1 33.3 58.7 57.8 59.5 55.7 56.0 MulongaWSC 16.0 61.0 NorthWestern WSC 86.0 45.0 NWSC 41.6 55.0 50.0 36.6 36.8 SWSC 73.0 51.8 49.4 50.5 56.0 56.0 WesternWSC 17.0 44.0 Country typology Resourch-rich 33.5 41.5 Middle-income 64.1 21.9 Fragile states 64.5 30.4 Nonfragile, low-income 73.8 40.1 Level of water scarcity High 60.5 33.3 Low 64.4 39.6 Size of the utility Small 61.2 39.4 Large 68.5 30.7 Overall 63.3 37.3 Source: Banerjee, Skilling, and others 2008. 335 336 Table A3.3 Treatment Samples passing chlorine test (%) Country Utility 1995­99 2000 2001 2002 2003 2004 2005 Benin SONEB Botswana DWA WUC Burkina Faso ONEA 99.0 99.0 99.0 99.0 99.0 Cameroon SNEC Cape Verde ELECTRA Chad STEE Congo, Dem. REDIGESO Rep. 36.0 39.0 36.0 46.0 32.0 36.0 Congo, Rep. SDNE 68.0 Côte d'Ivoire SODECI 90.0 90.0 90.0 90.0 90.0 90.0 90.0 Ethiopia ADAMA 100.0 AWSA DIRE DAWA Gabon SEEG Ghana GWC 85.5 91.1 80.5 Kenya KIWASCO 99.0 99.0 MWSC NWASCO 84.0 84.0 Lesotho WASA Liberia LWSR Madagascar JIRAMA Malawi BWB 99.9 99.8 97.8 99.8 99.9 99.9 99.8 CRWB 90.0 91.0 89.0 87.0 90.0 93.0 LWB 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Mali EDM 95.6 97.1 0.0 Mauritania MSNE Mozambique AdeM Beira 100.0 100.0 100.0 83.0 Adem Maputo 100.0 100.0 83.5 99.1 Adem Nampula 100.0 100.0 62.6 76.6 AdeM Pemba 71.1 71.0 100.0 100.0 AdeM Quelimane 100.0 100.0 100.0 100.0 Namibia Oshakati Municipality 95.0 95.0 95.0 95.0 Walvis Bay Municipality 99.0 99.0 99.0 99.0 99.0 99.0 99.0 Windhoek Municipality 99.9 Niger SPEN/SEEN Nigeria Borno FCT 100.0 Kaduna 50.0 Katsina Lagos Plateau Rwanda ELECTROGAZ 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Senegal SDE 98.6 96.6 99.3 98.6 95.1 Seychelles PUC South Africa Cape Town Metro Drakenstein Municipality eThekwini Metro (Durban) Joburg Sudan Khartoum Water Corporation 100.0 South Darfur Water Corporation 70.0 Upper Nile Water Corporation 40.0 337 (continued next page) 338 Table A3.3 (continued) Samples passing chlorine test (%) Country Utility 1995­99 2000 2001 2002 2003 2004 2005 Tanzania Arusha Babati Bukoba DAWASCO DUWS Iringa Kigoma Lindi Mbeya Morogoro Moshi Mtwara Musoma MWSA 93.0 95.0 98.0 Shinyanga Singida Songea Sumbawanga Tabora Tanga Togo TdE 99.8 99.9 99.9 99.8 99.8 Uganda NWSC Zambia AHC-MMS CHWSC CWSC KWSC LukangaWSC LWSC 95.0 80.0 74.0 83.0 81.0 MulongaWSC NorthWesternWSC NWSC 100.0 100.0 100.0 97.5 99.2 SWSC 96.0 98.0 95.0 WesternWSC Country typology Resourch-rich 78.1 Middle-income 98.0 Fragile states 63.0 Nonfragile, low-income 88.7 Level of water scarcity High 85.9 Low 84.1 Size of the utility Small 90.3 Large 65.5 Overall 84.8 Source: Banerjee, Skilling, and others 2008. 339 340 Table A3.4 Staffing Collection ratio (% of connections billed) Employees per 1,000 water connections (number/1,000 connections) Country Utility 1995­99 2000 2001 2002 2003 2004 2005 1995­99 2000 2001 2002 2003 2004 2005 Benin SONEB 93.3 115.9 Botswana DWA WUC Burkina Faso ONEA 100.0 100.0 100.0 100.0 100.0 100.0 100.0 9.7 9.7 7.7 7.0 6.8 7.0 6.2 Cameroon SNEC Cape Verde ELECTRA 95.9 84.2 94.4 98.3 Chad STEE Congo, Dem. Rep. REDIGESO 32.6 36.1 36.6 43.7 52.2 70.0 18.0 Congo, Rep. SDNE 79.0 0.0 83.0 81.0 83.0 88.0 5.3 4.9 7.9 7.7 7.2 7.0 Côte d'Ivoire SODECI 171.0 158.1 158.1 128.1 143.2 136.3 2.8 Ethiopia ADAMA 139.2 152.5 140.1 AWSA 83.6 85.0 70.6 71.2 83.8 Dire Dawa 17.7 16.1 Gabon SEEG 8.2 7.8 7.4 7.0 6.5 5.9 5.5 Ghana GWC 77.0 74.0 75.0 75.0 75.0 Kenya KIWASCO 77.3 96.7 20.8 20.7 MWSC 111.2 90.3 74.2 101.8 NWASCO 73.9 91.8 10.0 9.0 Lesotho WASA Liberia LWSR 57.0 63.0 12.6 16.4 Madagascar JIRAMA 0.8 Malawi BWB 100.0 100.0 100.0 15.2 15.0 13.7 Malawi CRWB 40.8 39.3 42.0 46.4 43.3 41.3 Malawi LWB 20.6 18.4 Mali EDM 96.0 96.0 96.0 96.0 7.7 5.9 6.0 5.7 5.3 Mauritania MSNE 95.0 104.6 22.8 21.8 Mozambique AdeM Beira 100.0 100.0 100.0 100.0 23.4 21.0 19.0 19.2 Adem Maputo 100.0 100.0 100.0 100.0 8.5 8.3 7.2 6.6 Adem Nampula 100.0 100.0 100.0 100.0 19.2 17.4 14.9 14.9 AdeM Pemba 100.0 100.0 100.0 100.0 24.1 24.2 21.9 21.4 AdeM Quelimane 100.0 100.0 100.0 100.3 25.7 24.3 26.0 23.3 Namibia Oshakati Municipality Walvis Bay Municipality 11.6 11.0 9.6 9.3 8.4 8.4 7.1 Windhoek Municipality 3.4 Niger SPEN/SEEN 79.6 88.7 93.6 92.1 87.6 9.0 8.4 7.3 6.6 6.8 Nigeria Borno 18.5 18.3 18.0 16.7 FCT 20.0 31.2 Kaduna 23.5 23.6 Katsina 9.5 14.4 Lagos 11.0 10.4 9.9 8.7 Plateau 31.8 30.0 28.3 Rwanda ELECTROGAZ 120.0 121.0 116.4 74.5 38.6 Senegal SDE 88.8 89.7 89.0 89.4 4.4 4.1 3.8 3.8 3.7 3.5 Seychelles PUC 98.7 100.0 19.7 19.2 South Africa Cape Town Metro 3.2 2.9 Drakenstein Municipality 100.0 4.1 4.0 2.8 2.8 eThekwini Metro (Durban) 4.2 4.0 (continued next page) 341 342 Table A3.4 (continued) Collection ratio (% of connections billed) Employees per 1,000 water connections (number/1,000 connections) Country Utility 1995­99 2000 2001 2002 2003 2004 2005 1995­99 2000 2001 2002 2003 2004 2005 Joburg 2.1 Sudan Khartoum Water Corporation 62.5 10.9 South Darfur Water Corporation 49.3 19.2 Upper Nile Water Corporation 8.3 9.4 Tanzania Arusha Babati Bukoba DAWASCO DUWS 98.1 105.4 3.8 Iringa Kigoma Lindi Mbeya Morogoro Moshi Mtwara Musoma MWSA 90.5 97.2 94.8 14.0 11.8 11.4 Shinyanga Singida Songea Sumbawanga Tabora Tanga Togo TdE 97.4 63.5 87.1 72.1 54.5 15.4 15.5 14.6 13.7 12.3 Uganda NWSC 100.8 110.5 103.2 109.9 101.6 100.2 99.7 26.2 20.2 16.2 11.5 10.6 9.5 8.6 Zambia AHC-MMS 82.0 CHWSC 76.0 CWSC 81.0 KWSC 65.0 LukangaWSC LWSC 66.5 80.2 80.0 77.0 14.5 13.1 12.9 12.1 10.5 MulongaWSC 58.0 North WesternWSC 94.0 NWSC 81.0 9.9 9.5 SWSC 57.0 11.7 10.6 WesternWSC 76.0 Country typology Resourch-rich 65.0 14.5 Middle-income 99.4 5.9 Fragile states 89.8 12.4 Nonfragile, low-income 98.5 14.8 Level of water scarcity High 83.0 16.0 Low 88.0 10.3 Size of the utility Small 84.1 16.3 Large 91.9 6.6 Overall 86.0 13.1 Source: Banerjee, Skilling, and others 2008. 343 344 Table A3.5 Financial Performance Debt service ratio (ratio) Operating cost coverage (ratio) Country Utility 1995­99 2000 2001 2002 2003 2004 2005 1995­99 2000 2001 2002 2003 2004 2005 Benin SONEB 22.1 44.2 1.0 1.2 Botswana DWA WUC Burkina Faso ONEA 17.5 17.2 8.5 9.5 12.1 5.8 5.1 1.0 1.1 1.1 1.1 0.9 Cameroon SNEC 0.8 Cape Verde ELECTRA Chad STEE 113.2 19.6 34.5 20.0 35.3 3.0 3.7 2.6 3.0 4.2 Congo, Dem. Rep. REDIGESO 1.7 2.0 2.6 4.2 7.2 11.2 0.1 0.1 0.3 0.4 0.3 0.6 Congo, Rep. SDNE 178.9 172.6 194.6 0.0 1.0 0.9 0.8 0.7 Côte d'Ivoire SODECI 1.1 1.0 1.0 1.0 1.0 1.0 Ethiopia ADAMA 1.9 1.8 1.1 AWSA 1.1 0.5 0.5 1.0 0.8 1.2 1.0 Dire Dawa Gabon SEEG 1.0 1.1 1.1 1.1 1.2 1.1 1.0 Ghana GWC Kenya KIWASCO 1.0 1.0 MWSC 2.5 1.4 1.4 NWASCO 218.3 51.9 1.4 2.6 Lesotho WASA 9.2 11.9 38.5 1.3 1.2 1.2 Liberia LWSR 182.8 30.1 1.3 1.0 Madagascar JIRAMA Malawi BWB 9.8 7.6 9.3 1.0 1.0 1.0 CRWB LWB Mali EDM 0.6 Mauritania MSNE 41.5 36.1 1.0 1.2 Mozambique AdeM Beira 1.0 1.2 1.4 1.3 Adem Maputo 0.8 0.6 1.0 0.8 Adem Nampula 1.0 1.1 1.3 1.5 AdeM Pemba 0.4 1.0 1.4 0.8 AdeM Quilimane 0.6 1.3 1.3 1.3 Namibia Oshakati Municipality 0.8 1.1 1.3 Walvis Bay Municipality Windhoek Municipality 5.2 5.6 5.6 5.3 5.9 5.2 1.1 1.1 1.1 1.3 1.1 0.9 Niger SPEN/SEEN 5472.1 69.3 26.7 12.7 12.6 1.6 0.8 0.8 1.0 1.0 Nigeria Borno FCT Kaduna Katsina 0.9 0.5 0.6 1.1 Lagos Plateau Rwanda ELECTROGAZ 1.5 3.5 3.7 3.4 1.7 0.8 Senegal SDE 4.2 3.9 3.5 0.8 0.9 0.9 0.9 1.0 1.0 Seychelles PUC 0.5 0.5 South Africa Cape Town Metro 4.9 4.6 3.8 3.8 1.0 1.0 1.1 0.9 0.9 Drakenstein Municipality 5.2 6.9 1.6 1.9 1.3 eThekwini Metro (Durban) 2.3 2.6 2.5 0.7 0.7 0.7 (continued next page) 345 346 Table A3.5 (continued) Debt service ratio (ratio) Operating cost coverage (ratio) Country Utility 1995­99 2000 2001 2002 2003 2004 2005 1995­99 2000 2001 2002 2003 2004 2005 Joburg Sudan Khartoum Water Corporation 0.9 South Darfur Water Corporation 1.0 Upper Nile Water Corporation 0.0 Tanzania Arusha 0.9 Babati 0.6 Bukoba 0.8 DAWASCO DUWS 1.0 0.9 Iringa 0.8 Kigoma 0.6 Lindi 0.4 Mbeya 1.0 Morogoro 0.7 Moshi 0.7 Mtwara 0.9 Musoma 0.6 MWSA 1.1 1.4 0.9 Shinyanga 0.5 Singida 1.4 Songea 0.4 Sumbawanga 0.9 Tabora 1.2 Tanga 1.2 Togo TdE 59.6 51.0 36.9 43.9 83.0 0.7 1.3 0.8 1.3 0.7 Uganda NWSC 4.4 3.5 3.0 2.9 4.8 4.7 0.9 1.1 1.1 1.1 1.1 1.2 1.2 Zambia AHC-MMS 0.9 CHWSC 0.5 CWSC 1.1 KWSC 1.5 LukangaWSC LWSC 0.8 0.8 0.9 1.0 MulongaWSC 1.0 NorthWesternWSC 0.7 NWSC 1.0 SWSC 1.1 WesternWSC 0.9 .Country typology Resourch-rich 115.0 1.1 Middle-income 15.4 1.0 Fragile states 20.7 0.9 Nonfragile, low-income 19.0 1.0 Level of water scarcity High 22.2 1.2 Low 41.2 0.9 Size of the utility Small 17.6 0.9 Large 51.5 1.0 Overall 30.5 1.0 Source: Banerjee, Skilling, and others 2008 347 348 Africa's Water and Sanitation Infrastructure Reference Banerjee, S., H. Skilling, V. Foster, C. Briceño-Garmendia, E. Morella, and T. Chfadi. 2008. "Ebbing Water, Surging Deficits: Urban Water Supply in Sub- Saharan Africa." AICD Background Paper 12. World Bank, Washington, DC. APPENDIX 4 Tariffs 349 Table A4.1 Structure of Domestic Tariffs 350 Metering Minimum Size of Size of Price of Price of Type of ratio consumption Fixed Number first block nth block first block nth block Country Utility tariff (%) (m3) charge of blocks (m3) (m3) ($) ($) Benin SONEB IBT 89.1 0 No 2 5 5+ 0.41 0.85 Botswana WUC IBT n.a. 0 Yes 4 10 25+ 0.43 1.61 Burkina Faso ONEA IBT 98.2 0 Yes 3 6 30+ 0.39 2.13 Cape Verde ELECTRA IBT 91.2 0 Yes 5 7 20+ 0 1.2 Chad STEE IBT n.a. 0 No 3 8 300+ 2.67 4.67 Congo, Dem. Rep. REGIDESO IBT 28.2 0 No 4 10 40+ 0.05 0.12 Congo, Rep. SDNE IBT 17.3 0 No 3 25 65+ 0.2 0.3 Côte d'Ivoire SODESI IBT 100 9 No 3 7 20+ 0.19 0.42 Ethiopia AWSA IBT n.a. 0 No 4 5 30+ 0.26 0.44 ADAMA IBT 90.1 0 No 4 5 50+ 0.14 0.34 Dire Dawa IBT n.a. 0 No 2 20 20+ 0.52 0.73 Ghana GWC IBT n.a. 0 No 4 10 60+ 0.18 0.52 Kenya NWASCO IBT n.a. 0 No 5 10 60+ 0.6 0.6 KIWASCO U-shaped 58.2 0 Yes 4 5 24+ 0.29 1.18 Lesotho WASA IBT 98.2 0 Yes 2 10 10+ 0.03 0.08 Madagascar JIRAMA IBT 97.1 0 Yes 1 10 30+ 0.3 0.61 Malawi BWB IBT 22.6 5 No 0 0 0 0 0 CRWB Flat n.a. 0 No 4 15 85+ 0.71 3.48 LWB IBT 98.1 0 Yes 3 4 40+ 0 0.52 Mali EDM IBT 96 0 No 3 20 61+ 0.2 1.09 Mozambique AdeM Beira IBT 99.9 10 Yes 3 9 30+ 0 0.66 Adem Maputo IBT 98.2 10 Yes 3 9 30+ 0 0.71 Adem Nampula IBT 100 10 Yes 3 9 30+ 0 0.58 AdeM Pemba IBT 99.1 10 Yes 3 9 30+ 0 0.57 AdeM Quelimane IBT 100 10 Yes 3 9 30+ 0 0.57 Namibia Oshakati IBT 96.5 0 Yes 3 10 40+ 0.26 0.92 Walvis Bay IBT 100 0 Yes 3 6 45+ 0.8 2.46 Windhoek IBT n.a. 0 Yes 4 6 40+ 1.01 1.94 Niger SEEN IBT 96.8 0 No 1 0 0 0.39 0.39 Nigeria FCT Linear 23.6 0 No 2 30 30+ 0.16 0.19 Kaduna IBT 16.1 0 No 3 30 1,000+ 0.19 0.28 Katsina IBT 6.5 0 No 6 5 500+ 0.44 1.09 Rwanda ELECTROGAZ IBT 98.7 0 No 4 20 60+ 0.37 0.73 Senegal SDE IBT 117.3 0 No 2 15 15+ 0.22 0.47 South Africa Drakenstein IBT 60.7 0 Yes 7 6 1,000+ 0 1.86 eThekwini IBT 66.4 0 Yes 3 6 30+ 0 1.77 Tygerberg IBT 60.3 0 Yes 6 6 50+ 0 1.86 Johannesburg IBT 52.4 0 No 6 6 40+ 0 1.4 Sudan Khartoum Water Corporation IBT n.a. 0 No 1 0 0 0.64 0.64 South Darfur Water Corporation Linear n.a. 0 No 1 0 0 0.59 0.59 Upper Nile Water Corporation Linear 0 0 No 3 20 40+ 0.37 1.46 Tanzania DAWASCO IBT 70.5 0 No 2 5 5+ 0.39 0.52 DUWS IBT 27.9 10 Yes 3 14 25+ 0 0.51 MWSA IBT 100 0 Yes 3 24 75+ 0.24 0.28 NWSC Linear 94.5 0 Yes 1 0 0 0.65 0.65 Zambia LWSC IBT 33.3 0 Yes 5 6 170+ 0.25 0.55 NWSC IBT n.a. 0 No 4 6 50+ 0.25 0.37 SWSC IBT n.a. 6 No 4 10 50+ 0.3 0.47 Simple average 0.31 0.95 By utility size Small 0.29 0.92 351 Large 0.2 0.66 Source: Banerjee, Foster, and others 2008. Note: IBT = increasing block tariff. 352 Table A4.2 Domestic Tariffs at Various Levels of Consumption Minimum Price Price consumption Fixed first last Country Utility Connection fee (m3) charge block block 4 m3 5 m3 6 m3 8 m3 10 m3 20 m3 30 m3 50 m3 100 m3 Benin SONEB 202.00 0 0.41 0.85 0.41 0.41 0.48 0.57 0.63 0.74 0.78 0.81 0.83 Burkina Faso ONEA 204.90 0 2.05 0.39 2.13 0.90 0.80 0.73 0.75 0.76 0.78 0.79 1.33 1.73 Cape Verde ELECTRA 24.24 0 2.67 4.67 2.67 2.67 2.67 2.93 3.09 3.88 4.14 4.35 4.51 Chad STEE 0.00 0 0.22 0.47 0.22 0.22 0.22 0.22 0.22 0.28 0.34 0.39 0.43 Congo, Dem. Rep. REGIDESO 0.00 0 0.05 0.12 0.05 0.05 0.05 0.05 0.05 0.06 0.07 0.08 0.04 Côte d'Ivoire SODECI 256.35 9 0.16 0.00 1.20 0.04 0.03 0.03 0.02 0.06 0.30 0.45 0.57 0.71 Ethiopia AWSA 14.44 0 0.19 0.42 0.19 0.19 0.19 0.21 0.24 0.30 0.34 0.37 0.40 ADAMA 8.89 0 0.26 0.44 0.26 0.26 0.27 0.28 0.29 0.35 0.38 0.40 0.42 Dire Dawa 43.33 0 0.14 0.34 0.14 0.14 0.15 0.17 0.17 0.19 0.22 0.24 0.29 Ghana GWC 0.00 0 0.52 0.73 0.52 0.52 0.52 0.52 0.52 0.52 0.59 0.65 0.69 Kenya NWASCO 34.41 0 0.18 0.52 0.18 0.18 0.18 0.18 0.18 0.23 0.24 0.31 0.40 KIWASCO 104.72 0 0.60 0.60 0.60 0.60 0.60 0.60 0.60 0.48 0.46 0.47 0.52 Lesotho WASA 208.20 0 0.29 1.18 0.40 0.37 0.39 0.41 0.43 0.64 0.79 0.94 1.06 Madagascar JIRAMA 0.00 0 0.30 0.03 0.08 0.11 0.09 0.08 0.07 0.06 0.07 0.07 0.08 0.08 Malawi LWB 0.00 0 2.42 0.30 0.61 0.91 0.79 0.71 0.61 0.54 0.49 0.48 0.53 0.57 BWB 0.00 5 0.48 0.00 0.52 0.12 0.10 0.16 0.24 0.29 0.40 0.43 0.47 0.49 CRWB 76.04 0 2.33 0.00 0.00 0.58 0.47 0.39 0.29 0.23 0.12 0.08 Mozambique AdeM Beira 239.25 10 3.83 0.00 0.66 0.96 0.77 0.64 0.48 0.38 0.44 0.48 0.55 0.61 Adem Maputo 239.25 10 3.83 0.00 0.71 0.96 0.77 0.64 0.48 0.38 0.53 0.58 0.64 0.67 AdeM Nampula 239.25 10 3.83 0.00 0.58 0.96 0.77 0.64 0.48 0.38 0.42 0.45 0.50 0.54 AdeM Pemba 239.25 10 3.83 0.00 0.57 0.96 0.77 0.64 0.48 0.38 0.40 0.43 0.49 0.53 AdeM Quelimane 239.25 10 3.83 0.00 0.57 0.96 0.77 0.64 0.48 0.38 0.40 0.42 0.48 0.53 Namibia Walvis Bay 0.00 0 0.71 3.48 0.71 0.71 0.71 0.71 0.71 0.83 1.06 1.38 1.87 Windhoek 238.36 0 0.80 2.46 1.45 1.32 1.23 1.26 1.27 1.30 1.31 1.43 1.94 Oshakati 23.66 0 3.85 1.01 1.94 1.97 1.78 1.65 1.53 1.46 1.41 1.46 1.57 1.76 Niger SEEN 245.88 0 1.02 0.26 0.92 0.52 0.47 0.43 0.39 0.36 0.47 0.50 0.60 0.76 Nigeria FCT WB 235.29 0 0.39 0.39 0.39 0.39 0.39 0.39 0.39 0.39 0.39 0.39 0.39 Kaduna 15.69 0 0.16 0.19 0.16 0.16 0.16 0.16 0.16 0.16 0.16 0.17 0.18 Katsina WB 47.06 0 0.19 0.28 0.19 0.19 0.19 0.19 0.19 0.19 0.19 0.21 0.22 Rwanda ELECTROGAZ 146.72 0 0.44 1.09 0.44 0.44 0.46 0.48 0.50 0.52 0.59 0.65 0.92 Senegal SDE 153.68 0 0.37 1.46 0.37 0.37 0.37 0.37 0.37 0.37 0.65 0.92 1.19 South Africa Drakenstein 325.01 0 1.52 0.00 1.86 0.00 0.00 0.00 0.25 0.25 0.38 0.42 0.57 0.85 Tygerberg 203.60 0 2.02 0.00 1.86 0.00 0.00 0.00 0.34 0.35 0.52 0.72 0.94 1.40 eThekwini 337.13 0 6.89 0.00 1.77 0.00 0.00 0.00 1.08 1.04 0.96 0.94 1.27 1.52 Johannesburg 339.95 0 0.00 1.40 0.00 0.00 0.00 0.15 0.24 0.56 0.77 0.98 1.61 Sudan NWC Khartoum 137.06 0 0.37 0.73 0.37 0.37 0.37 0.37 0.37 0.37 0.37 0.37 0.52 NWC South Darfur 198.74 0 0.64 0.64 0.64 0.64 0.64 0.64 0.64 0.64 0.64 0.64 0.64 NWC Upper Nile 6.36 0 0.59 0.59 0.59 0.59 0.59 0.59 0.59 0.59 0.59 0.59 0.59 Tanzania DAWASCO 20.55 0 0.39 0.52 0.39 0.39 0.41 0.43 0.45 0.48 0.50 0.50 0.51 DUWS 16.60 0 3.95 0.00 0.51 0.99 0.79 0.66 0.49 0.40 0.40 0.42 0.45 0.48 MWSA 15.81 0 1.11 0.24 0.28 0.51 0.46 0.42 0.38 0.35 0.29 0.28 0.27 0.26 Uganda NWSC 30.58 0 0.92 0.65 0.65 0.88 0.83 0.80 0.77 0.74 0.70 0.68 0.67 0.66 Zambia SWSC 33.49 0 0.30 0.47 0.30 0.30 0.30 0.30 0.30 0.31 0.33 0.35 0.41 LWSC 50.00 0 1.24 0.25 0.55 0.56 0.50 0.45 0.42 0.39 0.34 0.33 0.35 0.36 NWSC 0.00 0 0.25 0.37 0.25 0.25 0.25 0.25 0.26 0.27 0.28 0.30 0.34 Source: Banerjee, Foster, and others 2008. 353 Table A4.3 Cost Recovery at Various Levels of Consumption 354 4 m3 10 m3 40 m3 Price O&M Capital Meets Meets Price O&M Capital Meets Meets Price O&M Capital Meets Meets of cost cost O&M cost capital cost of cost cost O&M cost capital cost of cost cost O&M cost capital cost Country Utility 4 m3 threshold threshold threshold threshold 10 m3 threshold threshold threshold threshold 40 m3 threshold threshold threshold threshold Benin SONEB 0.41 0.70 0.80 No No 0.63 0.70 0.80 No No 0.79 0.70 0.80 Yes No Burkina Faso ONEA 0.90 0.75 0.80 Yes Yes 0.76 0.75 0.80 Yes No 1.12 0.75 0.80 Yes Yes Cape Verde ELECTRA 2.67 0.80 Yes 3.09 0.80 Yes 4.27 0.80 Yes Chad STEE 0.22 0.80 No 0.22 0.80 No 0.38 0.80 No Congo, Dem. Rep. REGIDESO 0.05 0.70 0.80 No No 0.05 0.70 0.80 No No 0.07 0.70 0.80 No No Côte d'Ivoire SODECI 0.04 0.63 0.80 No No 0.06 0.63 0.80 No No 0.53 0.63 0.80 No No Ethiopia AWSA 0.19 0.32 0.80 No No 0.24 0.32 0.80 No No 0.36 0.32 0.80 Yes No ADAMA 0.26 0.32 0.80 No No 0.29 0.32 0.80 No No 0.39 0.32 0.80 Yes No Dire Dawa 0.14 0.18 0.80 No No 0.17 0.18 0.80 No No 0.24 0.18 0.80 Yes No Ghana GWC 0.52 0.80 No 0.52 0.80 No 0.63 0.80 No Kenya NWASCO 0.18 1.13 0.80 No No 0.18 1.13 0.80 No No 0.29 1.13 0.80 No No KIWASCO 0.60 0.49 0.80 Yes No 0.60 0.49 0.80 Yes No 0.45 0.49 0.80 No No Lesotho WASA 0.40 0.16 0.80 Yes No 0.43 0.16 0.80 Yes No 0.88 0.16 0.80 Yes Yes Madagascar JIRAMA 0.11 0.70 0.80 No No 0.06 0.70 0.80 No No 0.08 0.70 0.80 No No Malawi LWB 0.91 0.57 0.80 Yes Yes 0.54 0.57 0.80 No No 0.51 0.57 0.80 No No BWB 0.12 0.41 0.80 No No 0.29 0.41 0.80 No No 0.45 0.41 0.80 Yes No CRWB 0.58 0.23 0.80 Yes No 0.23 0.23 0.80 Yes No 0.23 0.80 No No Mozambique AdeM Beira 0.96 0.51 0.80 Yes No 0.38 0.51 0.80 No No 0.53 0.51 0.80 Yes No Adem Maputo 0.96 0.73 0.80 Yes No 0.38 0.73 0.80 No No 0.62 0.73 0.80 No No AdeM Nampula 0.96 0.35 0.80 Yes No 0.38 0.35 0.80 Yes No 0.48 0.35 0.80 Yes No AdeM Pemba 0.96 0.53 0.80 Yes No 0.38 0.53 0.80 No No 0.46 0.53 0.80 No No AdeM Quelimane 0.96 0.42 0.80 Yes No 0.38 0.42 0.80 No No 0.46 0.42 0.80 Yes No Namibia Walvis Bay 0.71 0.80 No 0.71 0.80 No 1.26 0.80 Yes Windhoek 1.45 2.08 0.80 No No 1.27 2.08 0.80 No Yes 1.32 2.08 0.80 No Yes Oshakati 1.97 1.44 0.80 Yes No 1.46 1.44 0.80 Yes Yes 1.48 1.44 0.80 Yes Yes Niger SEEN 0.52 0.46 0.80 Yes No 0.36 0.46 0.80 No No 0.52 0.46 0.80 Yes No Nigeria FCT WB 0.39 0.80 No 0.39 0.80 No 0.39 0.80 No Kaduna 0.16 0.80 No 0.16 0.80 No 0.17 0.80 No Katsina WB 0.19 0.06 0.80 Yes No 0.19 0.06 0.80 Yes No 0.20 0.06 0.80 Yes No Rwanda ELECTROGAZ 0.44 0.51 0.80 No No 0.50 0.51 0.80 No No 0.63 0.51 0.80 Yes No Senegal SDE 0.37 0.85 0.80 No No 0.37 0.85 0.80 No No 0.78 0.85 0.80 No No South Africa Drakenstein -- 0.70 0.80 Yes Yes 0.25 0.70 0.80 No No 0.51 0.70 0.80 No No Tygerberg -- 1.21 0.80 Yes Yes 0.35 1.21 0.80 No No 0.83 1.21 0.80 No Yes eThekwini -- 1.56 0.80 Yes Yes 1.04 1.56 0.80 No Yes 1.14 1.56 0.80 No Yes Johannesburg -- 1.50 0.80 Yes Yes 0.24 1.50 0.80 No No 0.87 1.50 0.80 No Yes Sudan NWC Khartoum 0.37 0.28 0.80 Yes No 0.37 0.28 0.80 Yes No 0.37 0.28 0.80 Yes No NWC South Darfur 0.64 0.49 0.80 Yes No 0.64 0.49 0.80 Yes No 0.64 0.49 0.80 Yes No NWC Upper Nile 0.59 0.73 0.80 No No 0.59 0.73 0.80 No No 0.59 0.73 0.80 No No Tanzania DAWASCO 0.39 0.80 No 0.45 0.80 No 0.50 0.80 No DUWS 0.99 0.42 0.80 Yes Yes 0.40 0.42 0.80 No No 0.44 0.42 0.80 Yes No MWSA 0.51 0.19 0.80 Yes No 0.35 0.19 0.80 Yes No 0.27 0.19 0.80 Yes No Uganda NWSC 0.88 0.60 0.80 Yes Yes 0.74 0.60 0.80 Yes No 0.67 0.60 0.80 Yes No Zambia SWSC 0.30 0.30 0.80 No No 0.30 0.30 0.80 No No 0.34 0.30 0.80 Yes No LWSC 0.56 0.27 0.80 Yes No 0.39 0.27 0.80 Yes No 0.34 0.27 0.80 Yes No NWSC 0.25 0.20 0.80 Yes No 0.26 0.20 0.80 Yes No 0.29 0.20 0.80 Yes No Source: Banerjee, Foster, and others 2008. 355 Table A4.4 Structure of Nondomestic Tariffs 356 Government/ Industrial public institutions Commercial Comparison of commercial to residential price Residential Commercial Ratio of commercial Connection Fixed Number Price first Fixed Number Price first Fixed Number Price first price at price at to residential price Country Utility charge charge of blocks block charge of blocks block charge of blocks block 100 m3 100 m3 at 100 m3 Benin SONEB No 1.00 0.85 No 1.00 0.85 No 1.00 0.85 0.828 0.850 1.027 Burkina Faso ONEA Yes 1.00 2.13 Yes 1.00 2.13 Yes 1.00 2.13 1.729 2.151 1.245 Cape Verde ELECTRA No 1.00 0.78 No 1.00 No 1.00 0.78 4.509 4.533 1.005 Chad STEE No 2.00 0.22 No 2.00 0.22 No 2.00 0.22 0.433 0.433 1.000 Congo, Dem. Rep.REGIDESO No 1.00 No 1.00 0.00 No 3.00 0.01 0.040 0.006 0.144 SODECI Côte d'Ivoire Yes 4.00 0.48 No 1.00 1.07 4.00 0.48 0.707 Ethiopia AWSA No 1.00 0.42 No 1.00 0.42 No 1.00 0.42 0.397 0.422 1.064 ADAMA 0.424 Dire Dawa 0.294 Ghana GWC No Yes 2.20 0.687 2.198 3.199 Kenya NWASCO 4.00 0.18 0.403 0.435 1.078 KIWASCO Yes 5.00 0.60 0.521 0.479 0.920 Lesotho WASA Yes Yes 1.00 0.69 Yes 1.00 0.69 Yes 1.00 0.69 1.060 0.690 0.651 Madagascar JIRAMA No 2.00 0.23 0.078 Malawi LWB Yes 2.00 0.49 Yes 2.00 0.45 Yes 2.00 0.49 0.572 0.540 0.944 BWB 0.494 CRWB Mozambique AdeM Beira No 2.00 15.69 No 2.00 15.69 No 2.00 15.69 0.606 4.395 7.247 Adem Maputo No 2.00 16.75 No 2.00 16.75 No 2.00 16.75 0.674 4.689 6.960 AdeM Nampula No 2.00 13.88 No 2.00 13.88 No 2.00 13.88 0.542 3.885 7.173 AdeM Pemba No 2.00 15.02 No 2.00 15.02 No 2.00 15.02 0.530 4.207 7.935 AdeM Quelimane No 2.00 15.22 No 2.00 15.22 No 2.00 15.22 0.528 4.261 8.065 Namibia Walvis Bay No 4.00 1.99 No 4.00 1.99 No 4.00 1.99 1.869 1.993 Windhoek Yes No 1.00 1.63 No 1.00 1.63 No 1.00 1.63 1.945 1.628 0.837 Oshakati Yes Yes 3.00 17.70 Yes 3.00 17.70 Yes 3.00 17.70 1.758 36.810 20.934 Niger SEEN No 3.00 0.85 No 1.00 0.87 No 1.00 0.87 0.759 0.871 1.148 Nigeria FCT WB No 1.00 7.84 No 2.00 0.47 No 1.00 0.78 0.392 0.784 2.000 Kaduna No 3.00 0.55 No 2.00 0.19 No 2.00 0.55 0.181 0.549 3.030 Katsina No 1.00 1.57 No 2.00 0.20 No 1.00 1.57 0.221 1.569 7.092 Rwanda ELECTROGAZ No 3.00 0.44 No 3.00 0.44 No 3.00 0.44 0.921 0.691 0.751 Senegal SDE No 1.62 1.62 1.193 1.616 1.355 South Africa Drakenstein 0.848 Tygerberg Yes 1.00 0.82 Yes 1.00 0.82 Yes 1.00 0.82 1.401 0.841 0.600 eThekwini Yes 1.00 0.88 Yes 1.00 0.88 Yes 1.00 0.88 1.518 0.953 0.628 Johannesburg 1.606 0.375 0.233 Sudan NWC Khartoum No 1.00 0.73 No 1.00 0.73 No 1.00 0.73 0.519 0.734 1.415 NWC South Darfur No 1.00 1.41 No 1.00 1.41 No 1.00 1.41 0.636 1.407 2.212 NWC Upper Nile No 1.00 1.35 No 1.00 1.35 No 1.00 1.35 0.587 1.346 2.292 Tanzania DAWASCO No 3.00 0.57 No 3.00 0.57 No 3.00 0.57 0.510 0.573 1.123 DUWS No 1.00 13.04 No 1.00 13.04 No 1.00 13.04 0.484 1.767 3.649 MWSA No 1.00 0.47 No 1.00 0.28 No 1.00 0.40 0.264 Uganda NWSC No 1.00 1.05 No 3.00 0.80 No 1.00 1.05 0.660 1.050 1.590 Zambia SWSC LWSC Yes 3.00 0.37 0.359 0.587 1.633 Source: Banerjee, Foster, and others 2008. 357 358 Africa's Water and Sanitation Infrastructure Table A4.5 Structure of Sanitation Tariffs (Only Utilities with Wastewater Responsibility) Connection Tariff part of % of water Fixed Block Country Utility cost water bill bill fee tariff Burkina Faso ONEA Yes Cape Verde ELECTRA Côte d'Ivoire SODECI Yes Ethiopia AWSA Yes Kenya NWASCO Yes KIWASCO Yes Yes Lesotho WASA Yes Yes 0.85 Madagascar JIRAMA Namibia Walvis Bay Yes Yes Windhoek Oshakati Yes Yes Nigeria FCT WB meter Senegal SDE with sanitation Yes South Africa Drakenstein Tygerberg eThekwini Johannesburg Sudan NWC Khartoum NWC South Darfur NWC Upper Nile Tanzania DAWASCO Yes 0.80 DUWS Yes Yes 0.40 MWSA Yes Yes 0.50 Uganda NWSC Yes Yes 0.75 Zambia KWSC Yes 0.30 LWSC Yes Yes 0.30 NWSC Yes 0.30 Source: Banerjee, Foster, and others 2008. Tariffs 359 Table A4.6 Structure of Standpost Tariffs Official piped Ratio of Ratio of official Official Unofficial water unofficial piped water standpost standpost price to official price at 4 m3 price price at 4 m3 standpost to official Country Utility 3 3 (US$/m ) (US$/m ) (US$/m3) price standpost price (1) (2) (3) (2)/(1) (3)/(1) Benin SONEB 0.41 1.91 0.41 4.66 0.99 Burkina Faso ONEA 0.51 0.48 0.90 0.94 1.76 Cape Verde ELECTRA -- 9.44 2.67 -- -- Chad STEE -- -- 0.22 -- -- Congo, Dem. Rep. REGIDESO 0.05 1.02 0.05 20.40 0.93 Côte d'IvoireSODECI 0.45 0.93 0.04 2.06 0.09 Ethiopia AWSA 0.19 0.87 0.19 4.55 1.02 ADAMA 0.26 Dire Dawa 0.14 Ghana GWC 3.64 5.51 0.52 1.52 0.14 Kenya NWASCO -- 1.73 0.18 -- -- KIWASCO 0.60 Lesotho WASA n.a. 2.58 0.40 n.a. n.a. Madagascar JIRAMA 0.14 1.24 0.11 8.60 0.75 Malawi LWB 0.91 BWB 0.29 1.16 0.12 4.00 0.41 CRWB 0.58 Mozambique AdeM Beira 0.96 Adem Maputo 0.31 0.98 0.96 3.17 3.09 AdeM Nampula 0.96 AdeM Pemba 0.96 AdeM Quelimane 0.96 Namibia Walvis Bay 0.71 Windhoek 1.41 n.a. 1.45 n.a. 1.02 Oshakati 1.97 Niger SEEN 0.24 0.48 0.52 1.97 2.13 Nigeria FCT WB 0.39 Kaduna -- -- 0.16 -- -- Katsina WB 0.19 Rwanda ELECTROGAZ 0.44 1.79 0.44 4.07 1.00 Senegal SDE 0.54 1.53 0.37 2.83 0.69 South Africa Drakenstein 0.00 Tygerberg 0.00 eThekwini 0.00 Johannesburg n.a. n.a. 0.00 n.a. n.a. Sudan NWC Khartoum 0.92 1.15 0.37 1.25 0.40 (continued next page) 360 Africa's Water and Sanitation Infrastructure Table A4.6 (continued) Official piped Ratio of Ratio of official Official Unofficial water unofficial piped water standpost standpost price to official price at 4 m3 price price at 4 m3 standpost to official Country Utility 3 3 (US$/m ) (US$/m ) (US$/m3) price standpost price (1) (2) (3) (2)/(1) (3)/(1) NWC South Darfur 0.64 NWC Upper Nile 0.59 Tanzania DAWASCO 0.58 0.87 0.39 1.51 0.67 DUWS 0.99 MWSA 0.51 Uganda NWSC 0.39 1.40 0.88 3.63 2.28 Zambia SWSC 0.30 LWSC 0.19 1.67 0.56 9.03 3.02 NWSC 0.25 Source: Banerjee, Foster, and others 2008. Note: n.a. = not applicable. Table A4.7 Scorecard on Efficiency, Equity, and Cost Recovery Cost recovery Efficiency Equity Equity Efficiency Cost-recovery Total Country Utility 1 2 3 4 5 6 7 8 9 score score score score Cape Verde ELECTRA 1 1 1 1 1 1 1 1 3 3 2 8 Chad STEE 0 0 1 1 1 1 1 1 1 4 3 0 7 Benin SONEB 1 0 1 1 1 1 0 0 1 2 3 1 6 Namibia Oshakati 1 1 1 1 1 0 1 1 3 2 6 Namibia Windhoek 1 1 1 1 1 0 1 0 1 3 2 6 Nigeria Katsina WB 0 0 1 1 1 1 1 1 3 3 0 6 Rwanda ELECTROGAZ 1 0 1 1 0 1 1 1 0 3 2 1 6 Burkina Faso ONEA 1 0 0 1 1 0 1 0 1 2 2 1 5 Ethiopia AWSA 0 0 1 0 0 1 1 1 1 4 1 0 5 Ghana GWC 1 0 1 1 0 1 0 1 2 2 1 5 Kenya KIWASCO 1 0 1 0 1 1 1 0 2 2 1 5 Lethoso WASA 1 0 1 1 0 1 1 0 0 2 2 1 5 Namibia Walvis Bay 1 0 0 1 1 1 1 2 2 1 5 Nigeria FCT WB 0 0 1 1 0 1 1 0 1 3 2 0 5 Sudan NWC Upper Nile 1 0 1 0 1 1 0 0 1 2 2 1 5 Sudan NWC South Darfur 1 0 1 0 0 1 1 1 3 1 1 5 Tanzania DAWASCO 1 0 1 0 0 1 0 1 1 3 1 1 5 Uganda NWSC 1 0 0 1 0 0 1 1 1 3 1 1 5 South Africa eThekwini 1 1 0 0 1 1 1 0 2 1 2 5 South Africa Johannesburg 0 0 1 1 1 1 1 0 2 3 0 5 Kenya NWASCO 0 0 1 0 0 1 1 1 3 1 0 4 Nigeria Kaduna 0 0 1 0 0 1 1 1 3 1 0 4 Senegal SDE 0 0 1 0 0 1 0 1 1 3 1 0 4 South Africa Tygerberg 0 0 0 1 1 1 1 0 2 2 0 4 South Africa Drakenstein 0 0 0 0 1 1 1 1 3 1 0 4 361 Zambia NWSC 0 0 1 0 0 1 1 1 3 1 0 4 (continued next page) Table A4.7 (continued) 362 Cost recovery Efficiency Equity Equity Efficiency Cost-recovery Total Country Utility 1 2 3 4 5 6 7 8 9 score score score score Ethiopia Dire Dawa 0 0 1 1 0 1 0 1 2 0 3 Ethiopia ADAMA 0 0 1 0 0 1 1 2 1 0 3 Mozambique AdeM Quelimane 0 0 0 1 0 0 1 0 1 2 1 0 3 Niger SEEN 0 0 0 0 0 0 1 1 1 3 0 0 3 Sudan NWC Khartoum 0 0 1 0 0 1 0 1 2 1 0 3 Zambia SWSC 0 0 1 0 0 1 1 2 1 0 3 Côte d'Ivoire SODECI 0 0 0 0 0 1 0 1 2 0 0 2 Congo, Dem. Rep. REGIDESO 0 0 1 0 0 1 0 0 1 1 0 2 Mozambique AdeM Pemba 0 0 0 1 0 0 0 1 1 1 0 2 Malawi BWB 0 0 0 1 0 1 1 1 0 2 Malawi CRWB 0 0 0 1 1 0 0 0 2 0 2 Malawi LWB 1 0 0 1 0 0 0 0 1 1 2 Tanzania DUWS 0 0 0 0 0 0 1 1 2 0 0 2 Zambia LWSC 0 0 0 0 0 0 1 1 2 0 0 2 Mozambique AdeM Beira 0 0 0 0 0 0 0 1 1 0 0 1 Mozambique Adem Maputo 0 0 0 0 0 0 0 1 1 0 0 1 Mozambique AdeM Nampula 0 0 0 0 0 0 0 1 1 0 0 1 Tanzania MWSA 0 0 0 0 0 0 1 1 0 0 1 Madagascar JIRAMA 0 0 0 0 0 0 0 0 0 0 0 Source: Banerjee, Foster, and others 2008. Note: IBT = increasing block tariff, LRMC = long-run marginal cost, O&M = operations and maintenance. The scorecard is compiled on the basis of cost recovery, efficiency, and equity criteria. The scorecard adds the score against each criterion. The utility scores 1 against a specific criterion according to (1) Cost recovery: O & M cost recovery; (2) Cost recovery: Capital cost recovery; (3) Efficiency: No fixed charge or minimum consumption charge; (4) Efficiency: Metering ratio is higher than sample average (77%); (5) Efficiency: The price of the last block meets the capital cost; (6) Equity: Small piped consumers (at 4 m3) pay less than average piped consumers (at 10 m3); (7) Equity: Stand-post consumers pay less than small piped consumers (at 4 m3); (8) Equity: Connection cost as a share of GNI per capita is lower than sample average (27%); (9) Equity: Residential consumers pay less than nonresidential consumers at 100 m3 of consumption. Tariffs 363 Reference Banerjee, S., V. Foster, Y. Ying, H. Skilling, and Q. Wodon. 2008. "Achieving Cost Recovery, Equity and Efficiency in Water Tariffs: Evidence From African Utilities." AICD Working Paper 7, World Bank, Washington, DC. APPENDIX 5 Affordability of Water and Sanitation 365 Table A5.1 Contribution of Food to Total Spending 366 Expenditure budget (2002 US$) Share of household budget Country Year National Rural Urban Q1 Q2 Q3 Q4 Q5 National Rural Urban Q1 Q2 Q3 Q4 Q5 Angola 2000 102 112 37 22 56 85 121 194 46.32 45.80 58.92 58.22 59.48 56.66 53.80 38.78 Benin 2002 48 45 54 26 38 44 51 66 55.22 60.70 49.59 62.07 64.20 62.23 60.15 47.84 Burkina Faso 2003 58 55 70 33 44 53 62 80 47.92 54.20 35.57 67.88 66.00 61.90 55.60 36.30 Burundi 1998 47 45 91 13 29 39 54 81 71.83 76.80 43.47 71.84 77.13 77.67 78.33 66.29 Cameroon 2004 69 65 85 31 46 58 76 106 61.71 65.80 52.06 63.98 66.12 66.30 66.12 56.86 Cape Verde 2001 62 59 68 43 52 61 61 75 50.68 69.49 38.23 68.23 64.55 61.56 56.92 40.51 Congo, Dem. Rep. 2005 79 64 117 33 50 65 83 126 71.43 66.50 79.85 72.90 74.30 73.92 74.57 68.45 Congo, Rep. 2002 60 40 85 20 39 49 63 96 27.73 29.06 27.18 28.99 33.79 33.11 33.26 23.99 Gabon 2005 175 150 181 34 164 202 215 205 39.19 57.69 36.77 35.99 51.43 47.28 43.06 32.51 Ghana 1999 94 83 113 41 67 85 105 131 55.71 60.29 50.82 63.56 61.95 59.98 56.88 51.47 Guinea-Bissau 2005 81 72 103 35 55 65 81 138 54.41 52.25 58.59 49.44 53.71 54.55 55.50 54.84 Kenya 1997 87 81 109 42 62 77 97 119 62.35 70.03 47.30 76.96 76.42 74.72 70.91 51.77 Madagascar 2001 173 157 220 69 106 135 184 294 61.15 66.64 51.65 74.13 76.02 73.60 70.28 51.13 Malawi 2003 39 37 59 20 27 33 40 61 56.53 59.76 45.31 61.82 62.26 61.80 60.53 50.94 Mauritania 2000 114 88 150 55 79 102 125 169 50.76 50.16 51.26 59.95 58.19 55.45 53.97 44.66 Morocco 2003 191 168 209 84 138 183 237 375 43.09 54.16 38.37 52.60 51.81 48.50 44.19 35.65 Niger 2005 84 78 112 31 47 61 79 155 67.83 73.39 53.93 62.56 67.46 69.38 68.76 67.72 Nigeria 2003 43 42 45 17 32 42 50 59 50.08 57.32 43.88 55.82 61.25 60.54 56.73 40.95 Rwanda 1998 57 51 116 22 37 47 61 108 56.70 67.05 35.08 71.73 73.47 71.97 67.04 44.72 São Tomé and Príncipe 2000 127 110 141 58 81 95 120 217 60.57 70.66 55.48 77.64 75.57 71.99 68.68 51.73 Sierra Leone 2003 55 52 61 27 42 52 61 97 50.73 62.61 38.18 61.62 63.73 61.49 55.40 38.62 Tanzania 2000 39 36 51 20 29 36 42 56 65.92 68.83 59.38 71.78 70.58 71.33 68.60 60.60 Zambia 2002 62 60 67 26 42 54 67 99 63.00 74.57 49.58 70.99 72.50 0.00 70.20 54.38 Source: Banerjee, Wodon, and others 2008. Note: Q = quintile. Year refers to year of the survey. Table A5.2 Spending on Water Services Expenditure budget (2002 US$) Share in household budget Country Year National Rural Urban Q1 Q2 Q3 Q4 Q5 National Rural Urban Q1 Q2 Q3 Q4 Q5 Angola 2000 1 1 0 0 0 0 1 2 0.3 0.3 0.0 0.2 0.3 0.3 0.2 0.3 Burkina Faso 2003 0 2 0 2 2 2 0 0 0.2 2.2 0.0 5.0 3.0 2.8 0.2 0.0 Cameroon 2004 7 10 5 1 2 2 3 10 5.9 10.3 3.1 1.6 2.2 1.8 2.2 5.3 Cape Verde 2001 2 1 2 2 2 2 2 2 1.6 1.2 1.1 2.5 2.2 1.8 1.7 1.1 Chad 2001 9 4 11 3 4 6 8 14 2.5 1.7 2.6 3.4 2.7 2.6 2.8 2.1 Congo, Dem. Rep. 2005 2 1 3 1 1 1 2 3 1.9 0.9 1.7 1.8 1.7 1.4 1.7 1.6 Congo, Rep. 2002 2 1 4 1 1 2 2 4 1.1 0.6 1.3 0.9 1.2 1.3 1.3 0.9 Côte d'Ivoire 2005 4 2 4 3 4 4 4 5 1.9 1.3 1.6 3.2 2.7 2.3 1.7 1.2 Ethiopia 2000 1 1 1 1 1 1 1 1 1.5 1.5 1.3 2.0 1.6 1.6 1.5 1.2 Gabon 2005 11 6 11 1 8 9 13 12 2.4 2.4 2.2 1.4 2.4 2.1 2.7 1.9 Ghana 1999 1 0 2 0 1 1 1 2 0.7 0.2 0.7 0.1 0.5 0.4 0.5 0.7 Kenya 1997 2 1 3 1 1 1 2 3 1.7 0.8 1.5 1.6 1.7 1.2 1.3 1.3 Madagascar 2001 1 1 1 3 0 1 1 1 0.4 0.4 0.3 3.7 0.4 0.4 0.2 0.2 Malawi 2003 0 0 3 0 0 0 0 1 0.7 0.2 2.5 0.2 0.2 0.3 0.3 1.2 Mauritania 2000 11 5 14 1 5 11 10 14 4.9 2.9 4.8 1.6 3.7 5.8 4.2 3.6 Morocco 2003 9 4 10 5 6 8 9 14 2.1 1.3 1.8 2.9 2.2 2.0 1.8 1.3 Mozambique 2003 3 2 3 1 1 1 2 5 4.5 3.8 2.7 4.0 3.0 2.6 3.2 3.2 Niger 2005 5 4 7 2 3 4 5 7 4.1 3.7 3.2 4.7 4.5 4.3 4.6 2.9 Nigeria 2003 1 1 1 1 1 1 1 2 1.5 2.0 1.3 3.6 2.4 1.7 1.3 1.1 (continued next page) 367 368 Table A5.2 (continued) Expenditure budget (2002 US$) Share in household budget Country Year National Rural Urban Q1 Q2 Q3 Q4 Q5 National Rural Urban Q1 Q2 Q3 Q4 Q5 Rwanda 1998 8 4 8 1 1 1 9 8.1 5.6 2.6 2.9 1.6 1.5 3.5 São Tomé and Príncipe 2000 5 0 10 0 0 1 1 18 2.6 0.2 3.8 0.1 0.2 0.9 0.5 4.2 Senegal 2001 4 2 5 2 2 3 4 6 1.9 1.4 1.7 2.0 1.7 1.8 1.8 1.5 Sierra Leone 2003 2 1 2 0 0 2 2 2 1.5 0.7 1.1 0.3 0.3 2.9 1.8 0.6 South Africa 2000 6 1 8 1 1 2 4 13 1.0 0.4 1.2 0.8 1.0 1.2 1.2 1.0 Tanzania 2000 2 2 1 1 2 2 2 2 2.9 4.6 1.3 3.0 4.9 3.7 2.9 1.9 Uganda 2002 3 2 3 1 1 2 2 5 3.1 3.1 2.1 4.3 3.0 2.7 2.4 2.1 Zambia 2002 2 0 3 1 1 1 1 4 2.5 0.5 2.0 1.6 1.3 1.3 1.5 2.4 Source: Banerjee, Wodon, and others 2008. Note: Q = quintile. Year refers to year of the survey. Table A5.3 Affordability of Piped Water at 5 Percent Budget Threshold for Urban Households (% of households for which 5% of household budget is less than the cost of minimum consumption) Cost of minimum consumption (US$) Country 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 Benin 0.0 0.0 0.0 2.0 3.0 4.0 7.0 12.0 21.0 33.0 41.0 45.0 53.0 60.0 65.0 71.0 75.0 82.0 85.0 Burkina Faso 0.0 0.0 1.0 4.0 8.0 20.0 24.0 34.0 42.0 47.0 56.0 62.0 69.0 72.0 75.0 78.0 82.0 85.0 88.0 Burundi 1.0 7.0 17.0 29.0 45.0 53.0 67.0 72.0 76.0 82.0 86.0 90.0 94.0 97.0 99.0 100.0 100.0 100.0 100.0 Cameroon 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.0 1.0 2.0 4.0 7.0 11.0 17.0 21.0 27.0 34.0 Congo, Dem. Rep. 0.0 9.0 31.0 49.0 67.0 79.0 87.0 91.0 98.0 98.0 99.0 99.0 99.0 100.0 100.0 100.0 100.0 100.0 100.0 Congo, Rep. 0.0 0.0 0.0 0.0 1.0 3.0 3.0 5.0 9.0 12.0 17.0 21.0 23.0 28.0 33.0 35.0 36.0 43.0 49.0 Côte d'Ivoire 0.0 0.0 0.0 0.0 0.0 1.0 1.0 2.0 3.0 3.0 5.0 5.0 7.0 7.0 8.0 10.0 15.0 19.0 23.0 Ethiopia 1.0 40.0 73.0 87.0 93.0 95.0 98.0 99.0 99.0 99.0 99.0 99.0 99.0 99.0 100.0 100.0 100.0 100.0 100.0 Ghana 0.0 0.0 1.0 2.0 3.0 7.0 10.0 11.0 23.0 30.0 36.0 46.0 50.0 55.0 61.0 67.0 76.0 80.0 85.0 Guinea-Bissau 0.0 0.0 1.0 6.0 22.0 38.0 56.0 65.0 73.0 81.0 85.0 89.0 89.0 91.0 92.0 93.0 96.0 96.0 98.0 Kenya 0.0 0.0 0.0 0.0 4.0 5.0 13.0 20.0 28.0 36.0 49.0 62.0 67.0 72.0 77.0 78.0 80.0 83.0 86.0 Madagascar 0.0 0.0 5.0 16.0 23.0 28.0 38.0 47.0 53.0 61.0 64.0 68.0 74.0 78.0 82.0 85.0 86.0 89.0 90.0 Malawi 0.0 2.0 13.0 32.0 49.0 66.0 71.0 78.0 81.0 87.0 90.0 92.0 93.0 93.0 94.0 94.0 95.0 95.0 95.0 Mozambique 0.0 0.0 5.0 16.0 20.0 32.0 41.0 47.0 52.0 59.0 64.0 68.0 72.0 75.0 76.0 78.0 82.0 84.0 85.0 Niger 0.0 1.0 4.0 11.0 20.0 28.0 41.0 55.0 61.0 70.0 74.0 79.0 86.0 89.0 92.0 93.0 93.0 95.0 96.0 Nigeria 0.0 3.0 7.0 10.0 18.0 23.0 27.0 35.0 46.0 57.0 69.0 78.0 85.0 89.0 93.0 95.0 96.0 97.0 97.0 São Tomé and Príncipe 0.0 0.0 0.0 2.0 5.0 13.0 20.0 29.0 36.0 46.0 57.0 64.0 72.0 77.0 78.0 81.0 84.0 87.0 87.0 Senegal 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.0 1.0 1.0 1.0 2.0 2.0 Sierra Leone 0.0 0.0 2.0 4.0 7.0 16.0 23.0 30.0 40.0 44.0 49.0 54.0 57.0 62.0 65.0 67.0 69.0 71.0 73.0 South Africa 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 Tanzania 0.0 1.0 1.0 8.0 15.0 25.0 38.0 55.0 69.0 75.0 84.0 89.0 94.0 96.0 97.0 98.0 98.0 99.0 99.0 Uganda 0.0 2.0 5.0 17.0 32.0 45.0 55.0 65.0 77.0 82.0 88.0 90.0 94.0 96.0 96.0 97.0 98.0 98.0 98.0 369 Zambia 0.0 0.0 1.0 4.0 11.0 18.0 28.0 35.0 41.0 50.0 55.0 58.0 61.0 67.0 72.0 76.0 78.0 82.0 84.0 Source: Banerjee, Wodon, and others 2008. Note: Year refers to year of the survey. 370 Africa's Water and Sanitation Infrastructure Reference Banerjee, S., Q. Wodon, A. Diallo, T. Pushak, H. Uddin, C. Tsimpo, and V. Foster. 2008. "Access, Affordability and Alternatives: Modern Infrastructure Services in Sub-Saharan Africa." AICD Background Paper 2, Africa Infrastructure Country Diagnostic, World Bank, Washington, D.C. APPENDIX 6 Funding Gap for Water Supply and Sanitation 371 372 Table A6.1 Water and Sanitation Expansion and Rehabilitation Water Sanitation Total CAPEX CAPEX US $ million (Expansion + (Expansion + (Annual) Expansion Rehabilitation Rehabilitation) OPEX Total Expansion Rehabilitation Rehabilitation) OPEX Total CAPEX OPEX Total Angola 158 70 228 88 316 75 58 133 125 258 361 213 574 Benin 131 40 171 86 257 16 5 22 3 25 193 90 283 Botswana 22 42 64 70 134 3 3 6 1 7 70 71 141 Burkina Faso 65 45 110 44 154 44 3 47 7 54 157 50 208 Burundi 30 30 60 15 75 28 1 29 1 30 89 16 105 Cameroon 101 126 227 110 337 44 40 84 13 97 311 123 434 Cape Verde 4 2 5 3 9 5 1 6 9 15 12 12 24 CAR 54 40 94 43 137 10 7 17 2 19 111 45 156 Chad 36 31 67 21 88 44 7 51 6 57 118 27 145 Congo, Dem. Rep. 785 259 1,044 395 1,440 97 82 179 20 199 1,223 416 1,639 Congo, Rep. 90 38 129 54 182 22 8 30 4 34 159 57 216 Côte d'Ivoire 119 80 199 134 332 250 62 312 130 442 511 264 774 Equatorial Guinea 7 1 8 4 12 2 4 6 1 7 14 4 19 Eritrea 56 32 88 37 125 18 2 20 2 22 108 39 147 Ethiopia 542 153 694 337 1,031 123 33 156 19 175 850 356 1,206 Gabon 20 16 36 31 68 4 1 5 1 6 42 32 74 Gambia 17 13 30 14 44 10 5 15 17 31 45 31 76 Ghana 135 131 266 127 393 64 6 70 9 80 337 136 473 Guinea 20 42 62 25 88 18 9 27 8 35 89 33 123 Kenya 767 437 1,204 534 1,738 89 87 176 22 198 1,380 556 1,936 Lesotho 11 15 26 14 40 8 4 12 3 15 38 16 54 Liberia 34 18 52 23 74 24 7 30 17 47 82 40 122 Madagascar 290 97 387 151 538 55 2 57 17 75 444 168 612 Malawi 63 70 133 45 178 8 20 28 3 31 162 47 209 Mali 89 77 166 74 240 31 31 63 8 71 229 82 311 Mauritania 27 22 49 22 72 6 4 10 1 11 60 24 83 Mauritius 14 27 41 47 88 3 3 6 19 25 47 66 113 Mozambique 107 53 160 55 215 73 66 140 15 155 300 70 370 Namibia 58 71 129 126 255 12 4 16 19 36 145 145 290 Niger 112 54 166 46 212 45 2 47 6 53 213 52 266 Nigeria 875 259 1,134 426 1,560 448 244 692 88 780 1,827 514 2,340 Rwanda 35 16 51 21 71 27 7 34 4 38 85 25 110 Senegal 101 53 154 95 249 40 9 50 43 93 204 138 342 Sierra Leone 103 31 134 50 184 15 1 16 2 18 149 52 202 South Africa 563 565 1,128 1,057 2,184 319 204 523 613 1,136 1,651 1,670 3,320 Sudan 634 583 1,217 601 1,818 127 113 239 33 273 1,457 634 2,091 Swaziland 9 6 15 9 24 3 5 8 1 9 23 10 33 Tanzania 410 211 621 248 869 84 71 155 29 184 776 277 1,053 Togo 235 109 344 123 467 171 6 177 23 199 521 146 666 Uganda 58 98 156 36 191 72 65 137 20 157 293 56 348 Zambia 139 95 234 102 337 36 46 82 52 134 317 154 471 Zimbabwe 81 159 240 131 371 34 12 46 10 56 286 142 427 Total 7,225 4,327 11,553 5,686 17,239 2,617 1,352 3,969 1,432 5,401 15,522 7,118 22,640 Source: Briceño-Garmendia, Smits, and Foster 2008. 373 374 Africa's Water and Sanitation Infrastructure Table A6.2 Indicative Water and Sanitation Spending Needs US$ (million per year) GDP (percentage per year) Total Total spending spending Country Capital O&M needs Capital O&M needs Angola 361 213 574 1.18 0.70 1.87 Benin 193 90 283 4.50 2.09 6.60 Botswana 70 71 141 0.66 0.68 1.34 Burkina Faso 157 50 208 2.90 0.93 3.83 Burundi 89 16 105 11.21 1.97 13.18 Cameroon 311 123 434 1.88 0.74 2.62 Cape Verde 12 12 24 1.18 1.21 2.38 Central African Republic 111 45 156 8.22 3.35 11.57 Chad 118 27 145 2.01 0.46 2.47 Comoros -- -- -- -- -- -- Congo, Dem. Rep. 1,223 416 1,639 17.22 5.85 23.08 Congo, Rep. 159 57 216 2.61 0.94 3.55 Côte d'Ivoire 511 264 774 3.12 1.61 4.74 Equatorial Guinea 14 4 19 0.19 0.06 0.25 Eritrea 108 39 147 11.11 4.04 15.15 Ethiopia 850 356 1,206 6.91 2.89 9.80 Gabon 42 32 74 0.48 0.37 0.85 Gambia, The 45 31 76 9.73 6.68 16.41 Ghana 337 136 473 3.14 1.27 4.41 Guinea 89 33 123 2.73 1.03 3.76 Guinea-Bissau -- -- -- -- -- -- Kenya 1,380 556 1,936 7.37 2.97 10.34 Lesotho 38 16 54 2.65 1.15 3.80 Liberia 82 40 122 15.44 7.51 22.94 Madagascar 444 168 612 8.81 3.33 12.15 Malawi 162 47 209 5.67 1.66 7.33 Mali 229 82 311 4.31 1.54 5.86 Mauritania 60 24 83 3.25 1.28 4.53 Mauritius 47 66 113 0.75 1.05 1.80 Mozambique 300 70 370 4.56 1.07 5.63 Namibia 145 145 290 2.33 2.33 4.66 Niger 213 52 266 6.40 1.58 7.98 Nigeria 1,827 514 2,340 1.63 0.46 2.09 Rwanda 85 25 110 3.57 1.05 4.62 São Tomé and Príncipe -- -- -- -- -- -- Senegal 204 138 342 2.35 1.59 3.93 Seychelles -- -- -- -- -- -- Sierra Leone 149 52 202 12.29 4.29 16.59 (continued next page) Funding Gap for Water Supply and Sanitation 375 Table A6.2 (continued) US$ (million per year) GDP (percentage per year) Total Total spending spending Country Capital O&M needs Capital O&M needs South Africa 1,651 1,670 3,320 0.68 0.69 1.37 Sudan 1,457 634 2,091 5.32 2.32 7.63 Swaziland 23 10 33 0.88 0.37 1.25 Tanzania 776 277 1,053 5.49 1.96 7.45 Togo 521 146 666 24.18 6.77 30.95 Uganda 293 56 348 3.35 0.64 3.99 Zambia 317 154 471 4.31 2.10 6.41 Zimbabwe 286 142 427 8.36 4.15 12.51 Sub-Saharan Africa 15,522 7,118 22,640 2.42 1.11 3.53 Low-income, fragile 3,282 1,249 4,531 8.55 3.25 11.80 Low-income, nonfragile 5,682 2,128 7,810 5.15 1.93 7.08 Middle-income 1,990 1,996 3,987 0.73 0.74 1.47 Resource-rich 4,605 1,759 6,364 2.07 0.79 2.86 Source: Briceño-Garmendia, Smits, and Foster 2008. Note: GDP = gross domestic product, O&M = operations and maintenance, -- = not available. Table A6.3 Existing Financing Flows to Water and Sanitation Sectors 376 US$ (million per year) GDP (percentage per year) Capital Capital O&M PPI O&M PPI Public Public Non-OECD Household Total Total Public Public Non-OECD Household Total Total Country sector sector ODA financiers PPI self-finance CAPEX spending sector sector ODA financiers PPI self-finance CAPEX spending Angola -- -- 15.80 70.37 0 -- -- -- -- -- 0.05 0.23 0.00 -- -- -- Benin 15 11.43 59 0.00 0 13 83 98 0.35 0.27 1.37 0.00 0.00 0.30 1.93 2.28 Botswana 95 172 1 0 0 -- 172 268 0.91 1.64 0.01 0.00 0.00 -- 1.64 2.55 Burkina Faso 17 2 37 0 0 33 72 90 0.32 0.03 0.69 0.00 0.00 0.61 1.33 1.65 Burundi -- -- 11 0 0 -- -- -- -- -- 1.40 0.00 0.00 -- -- -- Cameroon 48 36 23 1 0 19 79 127 0.29 0.22 0.14 0.01 0.00 0.12 0.48 0.77 Cape Verde 1 11 9 0 0 -- 21 21 0.06 1.11 0.91 0.02 0.00 -- 2.05 2.11 Central African Republic -- -- 1 0 0 -- -- -- -- -- 0.10 0.00 0.00 -- -- -- Chad 0 1 27 1 0 10 39 39 0.00 0.01 0.47 0.01 0.00 0.17 0.66 0.67 Comoros -- -- 1 1 0 -- -- -- -- -- 0.27 0.24 0.00 -- -- -- Congo, Dem. Rep. -- -- 48 2 0 62 -- -- -- -- 0.68 0.02 0.00 0.87 -- -- Congo, Rep. 28 19 0 1 0 -- 20 48 0.45 0.31 0.00 0.02 0.00 -- 0.33 0.79 Côte d'Ivoire 54 13 1 1 0 92 107 162 0.33 0.08 0.00 0.01 0.00 0.56 0.66 0.99 Equatorial Guinea -- -- 1 0 0 -- -- -- -- -- 0.01 0.00 0.00 -- -- -- Eritrea -- -- 6 12 0 -- -- -- -- -- 0.64 1.26 0.00 -- -- -- Ethiopia 123 74 92 0 0 94 260 383 1.00 0.60 0.75 0.00 0.00 0.76 2.11 3.11 Gabon -- -- 12 0 0 -- -- -- -- -- 0.13 0.00 0.00 -- -- -- Gambia, The -- -- 7 3 0 -- -- -- -- -- 1.45 0.56 0.00 -- -- -- Ghana 53 23 98 0 0 31 151 204 0.49 0.21 0.91 0.00 0.00 0.29 1.41 1.90 Guinea -- -- 14 1 0 -- -- -- -- -- 0.42 0.04 0.00 -- -- -- Guinea-Bissau -- -- 1 0 0 -- -- -- -- -- 0.45 0.00 0.00 -- -- -- Kenya 12 34 97 2 0 23 155 167 0.07 0.18 0.52 0.01 0.00 0.12 0.83 0.89 Lesotho 7 3 14 2 0 2 21 28 0.48 0.20 1.01 0.12 0.00 0.17 1.49 1.98 Liberia -- -- 4 0 0 -- -- -- -- -- 0.73 0.00 0.00 -- -- -- Madagascar 0 5 19 0 0 46 70 70 0.01 0.11 0.37 0.00 0.00 0.90 1.38 1.39 Malawi 15 3 16 0 0 6 25 40 0.53 0.12 0.54 0.00 0.00 0.22 0.88 1.41 Mali 26 7 34 1 0 -- 42 68 0.48 0.13 0.64 0.02 0.00 -- 0.79 1.27 Mauritania -- -- 10 29 0 -- -- -- -- -- 0.55 1.57 0.00 -- -- -- Mauritius -- -- 9 6 0 -- -- -- -- -- 0.14 0.10 0.00 -- -- -- Mozambique 4 9 55 0 0 35 99 103 0.07 0.13 0.84 0.00 0.00 0.53 1.50 1.56 Namibia 131 12 6 0 0 3 21 152 2.11 0.19 0.09 0.00 0.00 0.05 0.33 2.44 Niger 11 28 23 1 0 3 56 66 0.32 0.85 0.68 0.03 0.01 0.10 1.67 1.99 Nigeria 14 355 100 0 0 295 751 766 0.01 0.32 0.09 0.00 0.00 0.26 0.67 0.68 Rwanda 6 0 17 0 0 20 37 43 0.25 0.00 0.71 0.00 0.00 0.85 1.56 1.81 São Tomé and Príncipe -- -- 0 0 0 -- -- -- -- -- 0.14 0.00 0.00 -- -- -- Senegal 5 9 37 14 0 49 110 114 0.05 0.11 0.43 0.16 0.00 0.57 1.26 1.32 Seychelles -- -- 1 0 0 -- -- -- -- -- 0.09 0.07 0.00 -- -- -- Sierra Leone -- -- 8 0 0 -- -- -- -- -- 0.65 0.00 0.00 -- -- -- South Africa 1,874 115 60 0 2 -- 177 2,051 0.77 0.05 0.02 0.00 0.00 -- 0.07 0.85 Sudan -- -- 12 5 7 -- -- -- -- -- 0.05 0.02 0.02 -- -- -- Swaziland -- -- 1 0 0 -- -- -- -- -- 0.05 0.00 0.00 -- -- -- Tanzania 15 33 143 4 1 28 209 224 0.10 0.23 1.01 0.03 0.01 0.20 1.48 1.59 Togo -- -- 2 0 0 -- -- -- -- -- 0.08 0.00 0.00 -- -- -- Uganda 1 1 47 3 0 41 93 -- 0.01 0.01 0.54 0.04 0.00 0.47 1.06 -- Zambia 35 67 47 1 0 9 123 158 0.48 0.91 0.63 0.02 0.00 0.12 1.68 2.16 Zimbabwe -- -- 1 0 0 -- -- -- -- -- 0.03 0.00 0.00 -- -- -- Sub-Saharan Africa 3,112 1,252 1,227 163 10 2,125 4,778 7,890 0.48 0.20 0.19 0.03 0.00 0.33 0.74 1.23 Low-income, fragile 128 30 105 20 0 165 313 441 0.33 0.08 0.27 0.05 0.00 0.43 0.81 1.15 Low-income, nonfragile 307 243 783 55 2 451 1,533 1,840 0.28 0.22 0.71 0.05 0.00 0.41 1.39 1.67 Middle-income 2,186 324 101 8 2 206 641 2,827 0.81 0.12 0.04 0.00 0.00 0.08 0.24 1.04 Resource-rich 188 717 238 80 7 522 1,564 1,753 0.08 0.32 0.11 0.04 0.00 0.23 0.70 0.79 Source: Briceño-Garmendia, Smits, and Foster 2008. 377 Note: CAPEX = capital expenditure, GDP = gross domestic product, O&M = operations and maintenance, ODA = official development assistance, OECD = Organisation for Economic Co-operation and development, PPI = private participation in infrastructure, -- = not available. 378 Africa's Water and Sanitation Infrastructure Table A6.4 Annual Budgetary Flows (not Traced) Country US$ (million per year) GDP (percentage per year) Angola -- -- Benin 26 0.61 Botswana 267 2.54 Burkina Faso 19 0.35 Burundi -- -- Cameroon 84 0.51 Cape Verde 12 1.17 Central African Republic -- -- Chad 1 0.02 Comoros -- -- Congo, Dem. Rep. -- -- Congo, Rep. 46 0.76 Côte d'Ivoire 67 0.41 Equatorial Guinea -- -- Eritrea -- -- Ethiopia 197 1.60 Gabon -- -- Gambia, The -- -- Ghana 75 0.70 Guinea -- -- Guinea-Bissau -- -- Kenya 46 0.24 Lesotho 10 0.68 Liberia -- -- Madagascar 6 0.11 Malawi 18 0.64 Mali 32 0.61 Mauritania -- -- Mauritius -- -- Mozambique 13 0.20 Namibia 143 2.30 Niger 39 1.18 Nigeria 370 0.33 Rwanda 6 0.25 São Tomé and Príncipe -- -- Senegal 14 0.16 Seychelles -- -- Sierra Leone -- -- South Africa 1,988 0.82 Sudan -- -- Swaziland -- -- Tanzania 47 0.34 Togo -- -- (continued next page) Funding Gap for Water Supply and Sanitation 379 Table A6.4 (continued) Country US$ (million per year) GDP (percentage per year) Uganda 2 0.02 Zambia 102 1.39 Zimbabwe -- -- Sub-Saharan Africa 4,364 0.68 Low-income, fragile 158 0.41 Low-income, nonfragile 550 0.50 Middle-income 2,509 0.93 Resource-rich 906 0.41 Source: Briceño-Garmendia, Smits, and Foster 2008. Note: GDP = gross domestic product, -- = not available. Nontraced spending refers to all available spending in the sector. Table A6.5 Public Infrastructure Spending by Sector and Institution US$ (million per year) GDP (percentage per year) 380 OPEX CAPEX OPEX CAPEX Country On-budget Off-budget On-budget Off-budget On-budget Off-budget On-budget Off-budget Angola -- -- -- -- -- -- -- -- Benin 0.00 14.85 0.00 11.43 0.00 0.35 0.00 0.27 Botswana 72.50 22.85 169.87 2.02 0.69 0.22 1.62 0.02 Burkina Faso 0.00 17.50 0.00 1.64 0.00 0.32 0.00 0.03 Burundi -- -- -- -- -- -- -- -- Cameroon 0.08 48.25 4.55 31.41 0.00 0.29 0.03 0.19 Cape Verde 0.64 0.00 8.71 2.47 0.06 0.00 0.87 0.25 Central African Republic -- -- -- -- -- -- -- -- Chad 0.19 0.00 0.78 0.00 0.00 0.00 0.01 0.00 Comoros -- -- -- -- -- -- -- -- Congo, Dem. Rep. -- -- -- -- -- -- -- -- Congo, Rep. 1.90 25.60 14.55 4.31 0.03 0.42 0.24 0.07 Côte d'Ivoire 6.97 47.37 12.96 0.00 0.04 0.29 0.08 0.00 Equatorial Guinea -- -- -- -- -- -- -- -- Eritrea -- -- -- -- -- -- -- -- Ethiopia 122.92 0.00 74.34 0.00 1.00 0.00 0.60 0.00 Gabon -- -- -- -- -- -- -- -- Gambia, The -- -- -- -- -- -- -- -- Ghana 3.55 48.99 3.68 18.83 0.03 0.46 0.03 0.18 Guinea -- -- -- -- -- -- -- -- Guinea-Bissau -- -- -- -- -- -- -- -- Kenya 12.28 0.00 33.56 0.00 0.07 0.00 0.18 0.00 Lesotho 1.47 5.41 1.67 1.18 0.10 0.38 0.12 0.08 Liberia -- -- -- -- -- -- -- -- Madagascar 0.29 0.00 5.32 0.00 0.01 0.00 0.11 0.00 Malawi 0.75 14.32 0.96 2.37 0.03 0.50 0.03 0.08 Mali 3.51 22.03 2.07 4.73 0.07 0.42 0.04 0.09 Mauritania -- -- -- -- -- -- -- -- Mauritius -- -- -- -- -- -- -- -- Mozambique 4.29 0.00 8.57 0.00 0.07 0.00 0.13 0.00 Namibia 13.16 118.25 2.37 9.27 0.21 1.90 0.04 0.15 Niger 0.02 10.72 1.83 26.61 0.00 0.32 0.06 0.80 Nigeria 14.37 0.00 355.49 0.00 0.01 0.00 0.32 0.00 Rwanda 0.64 5.25 0.05 0.00 0.03 0.22 0.00 0.00 São Tomé and Príncipe -- -- -- -- -- -- -- -- Senegal 1.30 3.33 9.35 0.00 0.01 0.04 0.11 0.00 Seychelles -- -- -- -- -- -- -- -- Sierra Leone -- -- -- -- -- -- -- -- South Africa 1,543.27 330.31 82.21 32.65 0.64 0.14 0.03 0.01 Sudan -- -- -- -- -- -- -- -- Swaziland -- -- -- -- -- -- -- -- Tanzania 10.81 3.94 32.66 0.00 0.08 0.03 0.23 0.00 Togo -- -- -- -- -- -- -- -- Uganda 0.77 0.00 0.76 0.00 0.01 0.00 0.01 0.00 Zambia 28.59 6.50 66.28 0.47 0.39 0.09 0.90 0.01 Zimbabwe -- -- -- -- -- -- -- -- Sub-Saharan Africa 2,216.17 895.78 1,072.60 179.53 0.35 0.14 0.17 0.03 Low-income, fragile 16.37 111.31 30.46 0.00 0.04 0.29 0.08 0.00 Low-income, nonfragile 163.85 143.30 176.09 66.73 0.15 0.13 0.16 0.06 Middle-income 1,691.14 494.38 274.59 49.34 0.62 0.18 0.10 0.02 Resource-rich 67.75 120.61 662.89 54.32 0.03 0.05 0.30 0.02 381 Source: Briceño-Garmendia, Smits, and Foster 2008. Note: CAPEX = capital expenditure, GDP = gross domestic product, OPEX = operating expenditure, -- = not available. 382 Table A6.6 Size and Composition of Funding Gap US$ (million per year) GDP (percentage per year) Capital O&M Capital O&M Country expenditure gap expenditure gap Funding gap expenditure gap expenditure gap Funding gap Angola -- -- -- -- -- -- Benin 106 72 177 2.46 1.67 4.14 Botswana 0 0 0 0.00 0.00 0.00 Burkina Faso 72 28 100 1.33 0.52 1.85 Burundi -- -- -- -- -- -- Cameroon 230 74 305 1.39 0.45 1.84 Cape Verde 0 7 7 0.00 0.68 0.68 Central African Republic -- -- -- -- -- -- Chad 79 27 105 1.34 0.45 1.79 Comoros -- -- -- -- -- -- Congo, Dem. Rep. -- -- -- -- -- -- Congo, Rep. 117 25 143 1.93 0.42 2.35 Côte d'Ivoire 320 166 486 1.96 1.02 2.97 Equatorial Guinea -- -- -- -- -- -- Eritrea -- -- -- -- -- -- Ethiopia 567 224 792 4.61 1.82 6.43 Gabon -- -- -- -- -- -- Gambia, The -- -- -- -- -- -- Ghana 185 83 269 1.73 0.78 2.50 Guinea -- -- -- -- -- -- Guinea-Bissau -- -- -- -- -- -- Kenya 1,162 516 1,678 6.20 2.76 8.96 Lesotho 12 7 19 0.82 0.48 1.30 Liberia -- -- -- -- -- -- Madagascar 322 144 466 6.39 2.86 9.24 Malawi 117 28 144 4.09 0.97 5.06 Mali 126 38 165 2.38 0.72 3.10 Mauritania -- -- -- -- -- -- Mauritius -- -- -- -- -- -- Mozambique 171 56 226 2.59 0.85 3.44 Namibia 107 12 119 1.72 0.19 1.92 Niger 147 39 186 4.41 1.17 5.58 Nigeria 966 448 1,414 0.86 0.40 1.26 Rwanda 41 17 58 1.73 0.69 2.43 São Tomé and Príncipe -- -- -- -- -- -- Senegal 69 98 168 0.80 1.13 1.93 Seychelles -- -- -- -- -- -- Sierra Leone -- -- -- -- -- -- South Africa 604 0 604 0.25 0.00 0.25 Sudan -- -- -- -- -- -- Swaziland -- -- -- -- -- -- Tanzania 512 237 749 3.62 1.67 5.30 Togo -- -- -- -- -- -- Uganda 179 49 228 2.05 0.56 2.61 Zambia 158 97 255 2.14 1.32 3.46 Zimbabwe -- -- -- -- -- -- Sub-Saharan Africa 8,648 3,225 11,873 1.3 0.5 1.8 Low-income, fragile 2,627 993 3,620 6.8 2.6 9.4 Low-income, nonfragile 3,673 1,612 5,285 3.3 1.5 4.8 Middle-income 312 0 312 0.1 0.0 0.1 Resource-rich 2,696 1,393 4,089 1.2 0.6 1.8 383 Source: Briceño-Garmendia, Smits, and Foster 2008. Note: GDP = gross domestic product, O&M = operations and maintenance, -- = not available. 384 Table A6.7 Reducing the Funding Gap US$ (million per year) GDP (percentage per year) Gain from inefficiencies Gain from inefficiencies Total Spending Gain (Funding Potential Total Spending Gain (Funding Potential spending traced to from inef- Capital Operational Cost gap) or for reallo- spending traced to from inef- Capital Operational Cost gap) or for reallo- Country needs needs ficiencies execution inefficiencies recovery surplus cation needs needs ficiencies execution inefficiencies recovery surplus cation Angola (574) -- -- -- -- -- -- -- 1.87 -- 0.00 -- -- -- -- -- Benin (283) 98 8 0 2 6 (177) 0 6.60 2.28 4.27 2.46 1.67 0.13 4.14 0.00 Botswana (141) 141 26 6 3 18 26 126 1.34 2.55 0.17 0.00 0.00 0.17 0.00 1.20 Burkina Faso (208) 90 18 0 14 4 (100) 0 3.83 1.65 1.92 1.33 0.52 0.07 1.85 0.00 Burundi (105) -- -- -- -- -- -- -- 13.18 -- 0.00 -- -- -- -- -- Cameroon (434) 127 2 2 -- -- (305) 0 2.62 0.77 1.84 1.39 0.45 -- 1.84 0.00 Cape Verde (24) 12 5 2 3 0 (7) 9 2.38 2.11 0.68 0.00 0.68 0.00 0.68 0.87 Central African Republic (156) -- -- -- -- -- -- -- 11.57 -- 0.00 -- -- -- -- -- Chad (145) 39 1 1 0 -- (105) 0 2.47 0.67 1.79 1.34 0.45 -- 1.79 0.00 Comoros -- -- -- -- -- -- -- -- -- -- 0.00 -- -- -- -- -- Congo, Dem. Rep. (1,639) -- 150 -- 53 97 -- -- 23.08 -- 1.37 -- -- 1.37 -- -- Congo, Rep. (216) 48 25 0 4 22 (143) 0 3.55 0.79 2.71 1.93 0.42 0.36 2.35 0.00 Côte d'Ivoire (774) 162 127 3 3 121 (486) 0 4.74 0.99 3.71 1.96 1.02 0.74 2.97 0.00 Equatorial Guinea (19) -- -- -- -- -- -- -- 0.25 -- 0.00 -- -- -- -- -- Eritrea (147) -- -- -- -- -- -- -- 15.15 -- 0.00 -- -- -- -- -- Ethiopia (1,206) 383 32 0 9 23 (792) 0 9.80 3.11 6.62 4.61 1.82 0.18 6.43 0.00 Gabon (74) -- 21 -- 1 21 -- -- 0.85 -- 0.24 -- -- 0.24 -- -- Gambia, The (76) -- -- -- -- -- -- -- 16.41 -- 0.00 -- -- -- -- -- Ghana (473) 204 1 1 -- -- (269) 0 4.41 1.90 2.50 1.73 0.78 -- 2.50 0.00 Guinea (123) -- -- -- -- -- -- -- 3.76 -- 0.00 -- -- -- -- -- Guinea-Bissau -- -- -- -- -- -- -- -- -- -- 0.00 -- -- -- -- -- Kenya (1,936) 167 90 18 40 32 (1,678) 0 10.34 0.89 9.13 6.20 2.76 0.17 8.96 0.00 Lesotho (54) 28 8 0 1 7 (19) 0 3.80 1.98 1.76 0.82 0.48 0.46 1.30 0.00 Liberia (122) -- 1 -- 1 -- -- -- 22.94 -- 0.00 -- -- -- -- -- Madagascar (612) 70 76 1 13 62 (466) 0 12.15 1.39 10.47 6.39 2.86 1.23 9.24 0.00 Malawi (209) 40 25 0 14 10 (144) 0 7.33 1.41 5.43 4.09 0.97 0.37 5.06 0.00 Mali (311) 68 79 0 8 70 (165) 0 5.86 1.27 4.43 2.38 0.72 1.33 3.10 0.00 Mauritania (83) -- 5 -- 5 -- -- -- 4.53 -- 0.00 -- -- -- -- -- Mauritius (113) -- -- -- -- -- -- -- 1.80 -- 0.00 -- -- -- -- -- Mozambique (370) 103 41 2 18 21 (226) 0 5.63 1.56 3.75 2.59 0.85 0.32 3.44 0.00 Namibia (290) 152 19 0 5 14 (119) 0 4.66 2.44 2.13 1.72 0.19 0.22 1.92 0.00 Niger (266) 66 13 0 1 12 (186) 0 7.98 1.99 5.95 4.41 1.17 0.38 5.58 0.00 Nigeria (2,340) 766 161 75 52 34 (1,414) 0 2.09 0.68 1.29 0.86 0.40 0.03 1.26 0.00 Rwanda (110) 43 9 0 9 0 (58) 0 4.62 1.81 2.43 1.73 0.69 0.00 2.43 0.00 São Tomé and Príncipe -- -- -- -- -- -- -- -- -- -- 0.00 -- -- -- -- -- Senegal (342) 114 59 2 0 57 (168) 0 3.93 1.32 2.59 0.80 1.13 0.66 1.93 0.00 Seychelles -- -- 12 -- 0 12 -- -- -- -- 1.61 -- -- 1.61 -- -- Sierra Leone (202) -- -- -- -- -- -- -- 16.59 -- 0.00 -- -- -- -- -- South Africa (3,320) 1,847 870 0 400 470 (604) 204 1.37 0.85 0.44 0.25 0.00 0.19 0.25 0.08 Sudan (2,091) -- 103 -- 47 56 -- -- 7.63 -- 0.20 -- -- 0.20 -- -- Swaziland (33) -- -- -- -- -- -- -- 1.25 -- 0.00 -- -- -- -- -- Tanzania (1,053) 224 80 13 47 20 (749) 0 7.45 1.59 5.44 3.62 1.67 0.14 5.30 0.00 Togo (666) -- 3 -- 3 -- -- -- 30.95 -- 0.00 -- -- -- -- -- Uganda (348) -- -- -- -- -- -- 0 3.99 -- 2.61 2.05 0.56 -- 2.61 -- Zambia (471) 158 58 14 21 23 (255) 0 6.41 2.16 3.77 2.14 1.32 0.31 3.46 0.00 Zimbabwe (427) -- -- -- -- -- -- -- 12.51 -- 0.00 -- -- -- -- -- Sub-Saharan Africa (22,640) 7,890 2,877 168 1,259 1,450 (11,873) 0 3.53 1.23 2.08 1.35 0.50 0.23 1.85 0.00 Low-income, fragile (4,531) 441 471 6 106 358 (3,620) 0 11.80 1.15 10.36 6.84 2.58 0.93 9.43 0.00 Low-income, nonfragile (7,810) 1,840 685 39 265 381 (5,285) 0 7.08 1.67 5.13 3.33 1.46 0.35 4.79 0.00 Middle-income (3,987) 2,637 1,037 8 492 537 (312) 189 1.47 1.04 0.31 0.12 0.00 0.20 0.12 0.07 Resource-rich (6,364) 1,753 522 137 172 214 (4,089) 0 2.86 0.79 1.93 1.21 0.63 0.10 1.84 0.00 Source: Briceño-Garmendia, Smits, and Foster 2008. Note: GDP = gross domestic product, -- = not available. 385 386 Africa's Water and Sanitation Infrastructure Reference Briceño-Garmendia, C., K. Smits, and V. Foster, V. 2008. "Financing Public Infrastructure in Sub-Saharan Africa: Patterns and Emerging Issues." Background Paper No. 15, Africa Infrastructure Country Diagnostic (AICD), World Bank, Washington, DC. Index Boxes, figures, and tables are indicated with b, f, and t following the page number. A supply-side vs. demand-side analysis, 39­42, 40b, 41t, 42f Abidjan (Nigeria), water resale in, surface water, 49­52, 276­77t 108, 109b water collection, 50­51b access to sanitation, 9, 27, 63­81 wells and boreholes, 33­37, 49­59, challenges across countries, 66­69 274­75t coverage growth, 69, 70f, 72­73f, accountability in sanitation market, 75f, 79­80t 115­20, 116f improved latrines. See latrines Accra (Ghana), standposts in, 102 by income groups, 14­16, 16f ADAMA (Nazareth Water Company), 176 on-site sanitation, 63­66 Addis Ababa Water Services Authority open defecation, 284­85t (AWSA), 176 patterns of, 13­16, 15t, 17t, 65t, ADeM (Águas de Moçambique), 100 66­68, 69t, 281­91t affermage policy, 88b progress in countries, 71­80, 74t affordability of services. See spending traditional pit latrines. See latrines Africa Infrastructure Country Diagnostic access to water, 9, 27, 33­61. See also (AICD) urban water markets data sources, 6­9 coverage of sources, change in, 53, DHS/MICS Survey Database, 28 56­57f, 286­88t expenditure survey database, 29­30 by income groups, 13, 15f, 18 fiscal database, 9 patterns of, 10­13, 11­12t, 58f methodologies, 4­6, 6f piped water, 37­42, 42f, 270­71t sample, 34, 38 rural water market, 49­52, 51t, sanitation sector analysis, 260 52f, 288t unit costs and, 211 septic tanks, 278­79t water sector analysis, 251 standposts, 272­73t 387 388 Index Africa Infrastructure Country Diagnostic open defecation, 68 (AICD) (continued) piped water, 37, 56, 58, 59 Water Supply and Sanitation rural water supply, 113, 115 (WSS) Survey sanitation, 117, 193 on institutional quality, 84 spending needs, 217 on latrine construction and tariffs, 130 operation, 117­18 toilet facilities, 65 methodology, 7­9, 8t urban reform index, 89 snapshot, 3, 27­28 utilities, 85 on standposts, 44, 103 water coverage, 193 on utility operations and Blackden, C. M., 50b management, 123, 158­59 BOCC (Basket of Construction on water reselling, 47 Components approach), 201b African Conference on Sanitation and boreholes. See wells and boreholes Hygiene, 265 Botswana African Development Bank, 236 funding gap, 239 Águas de Moçambique (ADeM), 100 sanitation, 193 AICD. See Africa Infrastructure spending needs, 217, 244 Country Diagnostic water coverage, 193 Angola Briceño-Garmendia, Cecilia, 9 sanitation, 193 budgeting. See also spending spending needs, 244 funding gap and, 220­24, 221­23t, water coverage, 193 223f, 378­79t water resale, 102­3 household, 161­62, 163t, 231, 234 "Angola mode," 237, 248n1 public expenditure and, 254­55 Arab funds, interest rates of, 239 bulk-water production, 87 Area Performance Contracts, 156b Burkina Faso AREQUAPCI (water resellers bulk-water production, 87 association), 48b coverage rates, 42, 191 Asia. See also individual countries institutional reforms, 252 financial crisis of 1997, 236 latrines, 67, 68, 71, 75, 76 infrastructure investment, 233 open defecation, 68 nonrevenue water, 137 piped water, 58, 59 tariffs, 168 rural water supply, 113 AWSA (Addis Ababa Water Services sanitation, 117, 132, 134b, 193, 265 Authority), 176 spending needs, 244 standposts, 44, 102, 164, 167 toilet facilities, 65 B utilities, 100 Banerjee, S., 6 water resale, 48 banking systems, 237­38. See also water sector regulation, 93­94 individual banks wells and boreholes, 259 base scenario assumptions, 196, 197f, Burundi 244, 246t, 248t ODA, 234 Basket of Construction Components sanitation, 193 (BOCC) approach, 201b water coverage, 193 Benin household size, 38b C improved latrines, 68 institutional reforms, 252­53 Cameroon as low-income country, 31 latrines, 68 ODA, 234 private sector, 87 Index 389 as resource-rich country, 30 concessions for utilities, 87 sanitation, 193 Congo, Democratic Republic of (DRC) spending needs, 162, 244 as fragile state, 30 water coverage, 36, 193 institutional reforms, 252­53 Cape Verde rural water access, 51 as middle-income country, 30, 34 sanitation, 117, 193 non-OECD financiers, 237 spending needs, 206, 244 private sector water supply, 87 standposts, 44 rural water supply, 110, 113­14 subsidies, 179­80 spending needs, 244 surface water, 53, 56 tariffs, 168 toilet facilities, 65 utilities, 84, 98, 167 traditional latrines, 68, 76 water sector regulation, 93, 94 utilities, 124, 145 capital costs. See also unit costs water collection, 50b from different sources, 237­38, 238f water coverage, 56, 59, 191 per capita, 198 Congo, Republic of recovery of, 143, 144f, 159n3. See also coverage rates, 42 cost recovery subsidies, 179­80 subsidies for, 18, 179­80, 180f traditional latrines, 68 capital investments, 220f water collection, 50b Caribbean, tariffs in, 168 connection charges, 176­79, 177­78f, CBOs. See community-based 184­87, 258 organizations connection rates for utilities, 39, 40b, Central African Economic and Monetary 41, 43­44, 126­27 Community (CEMAC), 138, 141 construction index, 200, 201b, 202 Central African Republic consumption of water. See water latrines, 68 consumption sanitation, 193 cost recovery water coverage, 42, 193 of capital costs, 143, 144f, 159n3 Central Region Water Board (CRWB), 130 consumption and, 173­74, 354­55t Chad efficiency and, 361­62t open defecation, 68 in middle-income countries, 17 rural water supply, 110 operating costs, 173­76, 174t, 175f, sanitation, 117, 193 176t, 180f spending needs, 217, 244 policy options for, 257­58 traditional latrines, 68 sanitation sector and, 265 water coverage, 42, 193 tariffs and, 17­18, 19f, 147b, 161, water resale, 12, 48 174, 179 children, water collection by, 49, 50­51b Côte d'Ivoire China, People's Republic of connection charges, 177 "Angola mode," 237, 248n1 coverage rates, 42, 53, 193 economic assistance from, 237 as fragile state, 30 interest rates, 239 institutional reforms, 252 China Export-Import Bank, 248n1 open defecation, 68 chlorine tests, 138, 336­39t piped water, 59 collection-efficiency ratios for utilities, private sector, 24, 25, 87 145­46, 146f rural water supply, 113, 115 Common Market for Eastern and sanitation, 117, 193, 261 Southern Africa (COMESA), 145 septic tanks, 68 community-based organizations (CBOs), sewerage systems, 63 103, 106­8, 118 standposts, 256 Comoros, traditional latrines in, 68 tariffs, 130 390 Index Côte d'Ivoire (continued) in utilities management, 139­42, traditional latrines, 76 140­41t urban reform index, 89 ELECTRA (utility company), 84, 168 utilities, 98, 154 ELECTROGAZ (utility company), 85, water law, 85 100, 130, 165b water resale, 48, 48b, 108, 109b ENDA Tiers Monde, 107­8 water sector regulation, 91, 92, 93 Equatorial Guinea country grouping methodology, 9­10, 10t, coverage rates, 191 30­31. See also fragile states; low- spending needs, 206, 244 income countries; middle-income Eritrea countries; resource-rich countries sanitation, 193 Cross River State (Nigeria), rural water water coverage, 193 supply in, 110, 111­12b Ethiopia CRWB (Central Region Water Board), 130 boreholes, 51 connection charges, 177 D as low-income country, 31 open defecation, 68, 71, 76, 77­78b Dakar (Senegal), sewerage systems in, piped water, 59 265 sanitation, 71, 117, 193, 262 Dar es Salaam Water and Sewerage spending needs, 206, 217 Company (DAWASCO), 130 subsidies, 179­80 debt-service ratios, 145 traditional latrines, 68, 71, 76 defecation. See open defecation water collection, 50­51b demographic and health surveys (DHSs), water coverage, 42, 53, 193 4­5, 28­29 European Commission, 236 Diallo, A., 41 diarrheal illnesses, 1 Direction de l'Hydrolique, 91 F direct water connections, 243, 244 Federal Capital Territory (FCT), 100 Dire Dawa (utility company), 176 financial crisis of 1997 in Asia, 236. distribution systems for water utilities, See also global financial crisis 136­38, 137f, 328­35t financing of services. See spending domestic finance, 231­34, 232t flush toilets. See toilet facilities DRC. See Congo, Democratic Republic of food spending, 162, 366t Foster, V., 6, 9 E fragile states East African Community (EAC), 125, definition of, 30 134, 138 efficiency savings, 224 East Asia, infrastructure investment in, 233 funding gap, 226 Economic Community of West African household self-finance, 234 States (ECOWAS), 125, 135, lower-cost technologies, 244 142, 145 MDG attainment, 241, 243 efficiency nonrevenue water losses, 136 cost recovery, 361­62t piped water, 125 financial, 142­46, 143t, 344­47t spending needs, 22, 216­17, funding gaps and, 22, 23t, 24, 224­31, 221, 223 225t, 227f, 239 utilities, 124­25, 128, 138, 142, 144 operations and pricing, inefficiencies in, water consumption, 134, 135 146­52, 147b, 149f, 150t, 152f funding gaps, 9, 27, 215­49. See also savings in, 224­31, 225t, 227f, 239 spending state-owned enterprises and, 139, budget execution, 220­24, 221­23t, 156­57b 223f, 378­79t Index 391 efficiency savings and, 22, 23t, 24, standposts, 102 224­31, 225t, 227f, 239 toilet facilities, 65 funds, increasing of, 239­40 utilities, 124 MDGs, other ways of reaching, water collection, 50b 240­48 water coverage, 36, 193 delaying attainment, 240­43, water sector regulation, 93­94 241­42f, 243t global financial crisis, 231, 233, technologies, lower-cost, 243­48, 236, 239 245f, 246t, 247f, 248t Global Rural-Urban Mapping Project by O&M and capital expenditure, (GRUMP), 209 226, 231f Global Water Intelligence, 159n3 raising additional finance, 231­39, governance index for SOEs, 85, 86t, 94, 376­77t 100, 101f, 306­17t costs of capital, 238­39, 238f gross domestic product (GDP) domestic finance, 231­34, 232t domestic finance and, 232­33 household self-finance, 231, 234 efficiency savings and, 224, 226 local sources, 231, 237­38, 237t hidden costs and, 148 ODA, 234­36, 235f high-end scenario and, 247 private sector and non-OECD lower-cost technologies and, 244 financiers, 236­37 MDG coverage gap and, 202, reducing, 215, 247­48, 384­85t 206, 209 size and composition of, 226, 230­31t, MDG financing and, 18­19, 22 382­83t ODA and, 234 water and sanitation spending, 215­20, subsidies and, 179­80 216t, 218f, 219t, 220f, 226, WSS spending and, 215, 216­17, 220 228­29f, 374­75t groundwater, 258­60 GRUMP (Global Rural-Urban Mapping Project), 209 G Guinea Gabon lease contracts, 25 concessions for utilities, 87 sanitation, 193 coverage rates, 42 toilet facilities, 65 as middle-income country, 34 traditional latrines, 68 piped water, 34 utilities, 154 private sector water supply, 87 water coverage, 193 septic tanks, 68 Gulf States, interest rates in, 239 spending needs, 244 subsidies, 180 H traditional latrines, 68 urbanization, 34 hand pumps, 51­52 water coverage, 193 Henderson, J. Vernon, 209, 211 The Gambia hidden cost estimates, 25, 123, 146­48, ODA, 234 147b, 150t, 152, 153, 153f spending needs, 244 high-density settlements, sewerage in, water coverage, 193 264­65. See also urban areas GDP. See gross domestic product high-end scenario assumptions, 244, 247 Ghana hookup rates. See connection rates for hand pumps, 52 utilities latrines, 68 household budgeting, 161­62, 163t, private sector water supply, 87 231, 234 rural water supply, 113, 115 household size, 38, 38b sanitation, 193 hygiene education, 117, 261­62 392 Index I spending needs, 206 tariffs, 130, 132 improved latrines. See latrines traditional latrines, 68 income groups urban reform index, 89 affordability threshold, 173­74 utilities, 85, 97, 98, 100 household budgets, 161­62 water coverage, 193 sanitation and, 14­16, 16f, 66, 66f, water sector regulation, 94 70­71, 177 Khartoum (Sudan), standposts in, 102 water access, 13, 15f, 18, 172­73 Kigali (Rwanda), piped water in, 165b increasing block tariffs (IBTs), 129, Kingdom, W., 142 168­70, 184, 187 Kinshasa (Democratic Republic of Congo), index factors. See individual indexes standposts in, 44 (e.g., governance index for SOEs) Kisumu Water and Sewerage Company India (KIWASCO), 130, 132 hygiene education, 261 Komives, K., 169, 181 interest rates, 239 Indonesia, public investment in, 233 inefficiencies in operations. See efficiency L in situ sanitation. See on-site sanitation land tenure, 39 services Latin America institutions and institutional reform infrastructure investment, 233 hidden costs and, 123 sewerage systems, 264­65 MDGs and, 24­27 tariffs, 168 performance improvement for utilities, latrines. See also on-site sanitation services 27, 153­57, 153f, 155t base scenario assumptions and, 196, 197 policy models, 255­57 improved private sector and, 24, 26t, 153­54, access patterns, 13, 15, 68, 69, 155t, 253­54 75, 76, 280­81t in water sector, 24­27, 251, 252­54 connection costs, 177, 179 Internally Delegated Area Management defined, 4­5 Contracts (IDAMCs), 156­57b management of, 117­18, 119t International Benchmarking Network for sanitation policy options, 261­62 Water and Sanitation Utilities traditional pit (IBNET), 7, 9 access patterns, 13, 15­16, 63­67, International Comparison Program (ICP), 69­71, 196, 282­83t 201b defined, 4­5 International Development Association goals for, 261, 262­64 (IDA), 239 progress measurement and, 266­67 ventilated improved pit (VIP), 4­5 J women and, 78b lease contracts, 253­54 Jiro sy Rano Malagasy (JIRAMA), 84­85 Lesotho Joint Monitoring Programme (JMP), 4­6, latrines, 68 5t, 6f, 266­67 as middle-income country, 30, 34 rural water supply, 113, 115 K sanitation, 117, 264 Kenya septic tanks, 71 hand pumps, 52 sewerage systems, 63 institutional reforms, 252 spending needs, 244 sanitation, 117, 193 standpipes, 103, 105 septic tanks, 68 tariffs, 132 sewerage systems, 63 utilities, 94, 100 Index 393 water coverage, 44, 56, 59, 193 water sector regulation, 92 water resale, 47 maintenance. See operations and water sector regulation, 92, 93 maintenance wells and boreholes, 58 Malawi Liberia coverage rates, 191 sanitation, 193 efficiency savings, 226 spending needs, 206, 244 institutional reforms, 252­53 water coverage, 193 latrines, 67, 68 Lilongwe Water Board, 145 piped water, 56, 59 line ministries, 89, 92 rural water supply, 51, 112 local finance sources, funding gaps and, sanitation, 193 231, 237­38, 237t sewerage systems, 63 lower-cost technologies for water and subsidies, 179­80 sanitation services, 243­48, 245f, tariffs, 130 246t, 247f utilities, 107 low-income countries. See also fragile states water sector regulation, 93 base scenario, 247­48 wells and boreholes, 259 coverage rates for, 194, 196 Mali definition of, 31 concessions for utilities, 87 domestic finance, 232 efficiency savings, 226 efficiency savings, 239 piped water, 56, 58 funding gap, 24, 226 sanitation, 117, 193 household self-finance, 234 standposts, 59 lower-cost technologies, 244 traditional latrines, 68 MDG coverage gap, 202, 205t, 206, 209 water coverage, 193 non-OECD financiers, 236 wells and boreholes, 259 ODA, 217 Maputo (Mozambique) piped water, 13, 34, 36 piped water, 49 pragmatic scenario, 247­48 water resale, 47 spending needs, 19, 22, 216, 220­21 marginal costs of public funds, 238­39, standposts, 256 248n3 tariffs, 257­58 Maseru (Lesotho) utilities, 11, 128, 140, 142, 144, piped water, 44 151, 152 water resale, 47 water consumption, 134 Mauritania Lusaka Water and Sewerage Company open defecation, 68 (LWSC), 100 sanitation, 193 traditional latrines, 68 water collection, 50b M water coverage, 193 Madagascar water resale, 12, 48 efficiency savings, 226 water spending, 162 latrines, 68, 71, 75 Mauritius, spending needs of, 244 rural water, 51, 110, 115 MDGs. See Millennium Development sanitation, 117, 193 Goals septic tanks, 71 metering of water. See water sewerage systems, 63 consumption spending needs, 217 MICSs (multiple-indicator cluster surveys), standpipes, 103 4, 28­29 toilet facilities, 65 middle-income countries water collection, 50b connection charges, 176 water coverage, 193 cost recovery, 17 394 Index middle-income countries (continued) non-OECD financiers, 237 coverage rates for, 194 open defecation, 68, 76 definition of, 30 piped water, 49 domestic finance, 232 standpipes, 108 efficiency savings, 224, 226 tariffs, 130 funding gap, 24, 226 traditional latrines, 68 high-end scenario, 247 Unit Costs of Infrastructure Projects lower-cost technologies and, 244, 248 Study, 212b MDG attainment, 241, 243 urban reform index, 89 MDG coverage gap, 202, 205t, water collection, 51b 206, 209 water resale, 47 non-OECD financiers, 237 water sector regulation, 92, 93 nonrevenue water losses, 136 water supply, 87, 113 piped water, 13, 34, 36, 125 water utilities, 99, 100, 169 spending needs, 19, 22, 128, 216, 217, wells and boreholes, 259 220­21, 223 multiple-indicator cluster surveys utilities, 11, 124, 138, 140, 142, 144­45, (MICSs), 4, 28­29 152, 153 water consumption, 134, 135 N Millennium Development Goals (MDGs), 1­32 Namibia AICD surveys, 28, 29­30 bulk-water production, 87 attainment delay of, 240­43, connection charges, 177 241­42f, 243t coverage rates, 191 costs, tariffs, and subsidies, 16­18 institutional reforms, 252 country typology, 30­31 as middle-income country, 34 coverage expansion, 191­94 open defecation, 68 coverage gaps in, 202­9, 203t, piped water, 49, 59 207­8f rural water supply, 113 data sources and methodologies, 4­10 sanitation, 193 country categories, 9­10, 10t septic tanks, 67, 68 data sources, 6­9, 8t sewerage systems, 14, 63 JMP and AICD, 4­6, 6f standpipes, 103, 105 financing of, 18­24, 20­21t, 191 tariffs, 132 funding gaps and, 240­48 Unit Costs of Infrastructure Projects institutional reform and, 24­27, 252 Study, 212b monitoring and evaluation, 4­6, 5t, utilities, 16, 97, 124, 140, 142 83, 266­67 water sector regulation, 94 in rural areas, 193 National Sanitation Program (Lesotho), sanitation, 2­3, 3t, 13­16, 15t, 17t, 264 63, 69 National Water and Sewerage Company snapshot of, 3, 27­28 (NWSC), 98, 99, 100, 129, synthesis for, 3­4 156­57b in urban areas, 193 natural resource savings accounts, water access, 10­13, 11­12t, 33, 49 239­40 water gap, 2, 2t, 191, 192f Nazareth Water Company (ADAMA), Morella, E., 6 176 Morocco, water collection in, 50b NGOs. See nongovernmental organizations Mozambique Niger connection charges, 177 bulk-water production, 87 coverage rates, 42, 191 connection charges, 177 institutional reforms, 252 coverage rates, 191 Index 395 improved latrines, 68 "off-grid" consumers, 124 institutional reforms, 252 Office National de l'Assainissement du open defecation, 68 Sénégal (ONAS), 116­17 sanitation, 193 Office Nationale des Eaux et spending needs, 217 d'Assainissement (ONEA), subsidies, 180 98­99, 117, 145 utilities, 98, 100 official development assistance (ODA), water collection, 51b 215, 217, 234­36, 235f water resale, 12, 48 on-site sanitation services. See also water sector regulation, 93, 94 latrines water supply, 87, 115 accountability, 115, 116f Nigeria groundwater and, 259­60 coverage rates, 53, 191 limitations of, 264 domestic finance, 232 predominance of, 63­66 efficiency savings, 226 sanitation index and, 117, 118f, household investment, 234 319­21t institutional reforms, 252 spending on, 217 ODA, 234 unit cost of, 177, 200, 201t open defecation, 71, 76 open defecation piped water, 36­37, 56, 59 access to sanitation, 284­85t as resource-rich country, 30 numbers practicing, 13, 64, 66, 71, rural water supply, 110, 111­12b, 72f, 76, 77­78b 113, 115 policy options, 260, 261­62 sanitation, 71, 117, 193 in rural areas, 68, 70, 193 spending needs, 206, 244 sanitation index and, 117 standpipes, 103 in urban areas, 68, 70 tariffs, 130 operating expenses traditional latrines, 68, 71 cost recovery of, 173­76, 174t, 175f, Unit Costs of Infrastructure Projects 176t, 180f Study, 212b subsidies for, 179­80 utilities, 100, 124, 148 operations and maintenance (O&M) water collection, 50b, 51b funding gap, 226, 231f water resale, 12, 48, 108, 109b inefficiencies in, 23, 146­52 wells and boreholes, 58, 259 MDG attainment and, 243 Nikana Water and Sewerage Company MDG coverage gap and, 202 (NWSC), 145 quantifying of, 213­14 nongovernmental organizations (NGOs), spending on, 16, 18, 197, 216, 221, 103, 106­8, 110, 115, 118 223­24, 223t, 380­81t non-OECD financiers, 236­37 tariffs for, 128, 130 nonpiped water services, 102­10. See also Organisation for Economic water resale Co-operation and Development nonrevenue water (NRW), 88b, 132, (OECD) 136­38, 137f, 147b donor contributions, 217 NWSC. See National Water and Sewerage non-OECD financiers, 236­37 Company ODA and, 234, 236 official development assistance, 215 Ouagadougou (Burkina Faso), standposts O in, 44, 102, 164, 167 O&M. See operations and maintenance outsourcing, 99­100 ODA. See official development assistance overstaffing in utilities management, OECD. See Organisation for Economic 147, 147b, 148, 151, 152, Co-operation and Development 340­43t 396 Index P Q performance contracts and monitoring, quality of services. See service provision 27, 99, 99f, 100, 102, 102f and quality performance indicators for utilities, 181­83, 182­83f, 185f, 186 R piped water. See also access to water rainwater harvesting, 43, 165b access to, 37­42, 42f, 270­71t reform, institutional. See institutions and coverage for, 33­34, 34f, 36t, 37f, institutional reform 53, 55­56, 57t, 58­59 reform index price of, 162, 164, 164f, 165b, 167, rural, 113, 114f, 115, 218t 171­72 urban, 85, 86t, 89, 90f, 294­99t in rural areas, 10, 34, 59 Régie de Production et de Distribution sewerage systems and, 63­64 d'Eau (REGIDESO), 145 tariffs for, 128­31, 131f regulation index, 85, 86t, 89, 100, in urban areas, 34, 43, 53, 369t 300­305t pit latrines. See latrines regulation of water sector, 25, 89­94, Plateau (utility company), 148 91t, 92­93f, 96f policy options, 27, 251­67 rehabilitation needs for sanitation sector, 260­67 backlog in, 19, 206, 243 champion for issue, 266 cost of, 17, 197 measuring progress, 266­67 MDG coverage gap and, 202 open defecation, 261­62 quantifying of, 213­14 sewerage in high-density resellers of water. See water resale settlements, 264­65 resource-rich countries spending, 265 coverage rates, 196 traditional latrines, 262­64 definition of, 30 for water sector, 251­60 domestic finance, 232 cost recovery and social policies, efficiency savings, 239 257­58 funding gap, 226 groundwater in urban water supply, high-end scenario, 247 258­60 household self-finance, 234 institutional models, 255­57 local finance sources, 237 institutional reforms, 252­54 MDG coverage gap, 206, 209 public expenditure, 254­55 natural resource savings accounts, population densities, 209, 210t, 211 239­40 population growth, 38, 38b, 123, 264 non-OECD financiers, 236­37 poverty lines, 186 spending needs, 19, 22, 216­17, pragmatic scenario assumptions, 244, 220­21, 224 245f, 246t, 248t utilities, 138, 141­42, 144, 145 private sector rural areas financiers, 236­37 coverage gaps, 202 institutional reform and, 24, 26t, household budgets, 161 153­54, 155t, 253­54 institutional reforms, 252­53 water supply contracts, 87, 88b MDG targets, 193 production capacities of water open defecation, 68, 70, 193 utilities, 128 piped water, 10, 34, 59 public expenditures, 254­55. See also policy options for, 251­52 spending reform index, 113, 114f, 115, 218t public-private partnerships, 88b sanitation, 67­68, 291t public sector utility providers, 94, 97 standposts, 49 purchasing power parity estimates, 201b surface water, 10 Index 397 unit costs, 198, 200 septic tanks, 67, 68, 71, 76 urban-rural divide, 64, 202, 204f sewerage systems, 14, 63, 265 rural water markets, 49­53, 51t, 52f, spending needs, 217, 244 110­15, 111­12b, 112t, standposts, 256 113­14f, 288t utilities, 98, 100, 107 RUWATSSA (State Rural Water Supply water coverage, 193 and Sanitation Agency), 111b water sector regulation, 94 Rwanda Sénégalaise des Eaux (SDE), 100 coverage rates, 42 septic tanks latrines, 68, 71, 75 access to water, 278­79t piped water, 56, 59, 165b cost of, 15, 66 sanitation, 193 flush toilets and, 63 standposts, 256 health benefits of, 243 tariffs, 130 in urban areas, 13, 67­68, 261 utilities, 85, 100 use of, 13, 69­71, 76 water coverage, 193 service provision and quality water sector regulation, 92 unit cost of, 197­202, 199t water spending, 162 of water utilities, 138­39, 139t wells and boreholes, 259 sewerage systems condominial, 264­65 in high-density settlements, 264­65 S prevalence of, 14, 63­64, 64f SADC. See Southern African Development tariffs and, 131­32, 133t Community unit cost of, 200, 202 sale of water. See water resale Sierra Leone Sanitation Advocacy Campaign coverage rates, 191 (Ethiopia), 262 sanitation, 193 sanitation gap, 193, 194f water collection, 50b sanitation index, 117, 118f, 319­21t Skilling, H., 6 sanitation ladder, 6, 7f, 63, 71, 75f, 76 SNE (Société Nationale des Eaux), 87 sanitation sector. See also access to social accountability index, 93­94, sanitation; on-site sanitation 95f, 120n1 services; water supply and social value premiums, 238­39 sanitation; individual types of Société de Distribution d'Eau de Côte facilities (e.g., latrines) d'Ivoire (SODECI), 48b, 98, accountability, 115­20, 116f 100, 109b, 145 spending for, 265 Société de Exploitation des Eaux du Niger Sanitation Strategic Plan (Burkina Faso), (SEEN), 87, 98, 100 134b Société de Patrimoine des Eaux du Niger SDE (Sénégalaise des Eaux), 100 (SPEN), 87 SEEN. See Société de Exploitation des Société Nationale des Eaux (SNE), 87 Eaux du Niger Société Nationale des Eaux du Benin Senegal (SONEB), 130 coverage rates, 42, 53 Société Tchadienne d'Eau et d'Électricité improved latrines, 68 (STEE), 84 institutional reforms, 252 socioeconomic groups. See income as low-income country, 31 groups piped water, 34, 36, 56, 58, 59 SODECI. See Société de Distribution private sector, 24, 87, 88b d'Eau de Côte d'Ivoire rural water supply, 49, 113, 115 SOEs. See state-owned enterprises sanitation, 116­17, 177, 179t, SONEB (Société Nationale des Eaux du 193, 266 Benin), 130 398 Index South Africa staffing issues. See overstaffing in utilities bulk-water production, 87 management connection charges, 177 standpipes. See also standposts coverage rates, 191 management of, 103­8, 104f, 104t, 106f financial markets, 248n2 subsidies and, 187­88 local capital markets, 231 standposts. See also standpipes as middle-income country, 30, 34 access to water, 272­73t piped water, 34, 49 estimates on, 211 private sector water supply, 87 institutional models and, 255­56 rural water supply, 110 piped water through, 33 sanitation, 117, 193 price variations, 162, 164, 164f, 165b, septic tanks, 67, 68 167­68 sewerage systems, 14, 63 in rural areas, 49 spending needs, 206, 217, 244 tariffs, 164, 359­60t standpipes, 103, 105 in urban areas, 11, 43­44, 47, 47t tariffs, 130 state-owned enterprises (SOEs) traditional latrines, 68 governance index, 85, 86t, 94, 100, 101f, utilities, 16, 85, 97, 99, 100, 124, 306­17t 140, 142 inefficiencies in management, 151 water collection, 50­51b institutional reform and, 254 Southern African Development spending by, 215, 221, 223 Community (SADC), 134, 138, utilities and, 97, 139, 156­57b 140, 142 State Rural Water Supply and Sanitation Southern Regional Health Bureau, 77b Agency (RUWATSSA), 111b SPEN (Société de Patrimoine des Eaux STEE (Société Tchadienne d'Eau et du Niger), 87 d'Électricité), 84 spending, 27, 191­214. See also budgeting; stimulus packages, 240 funding gaps subsidies. See also tariffs affordability threshold for WSS, capital expenses, 18, 179­80, 180f 173­75, 176 for connections, 176­79, 184­86, connection charges, 176­79, 177­78f, 185f, 258 184­87, 258 design factors, 184­85 coverage expansion, 191­97 IBT structure and, 170 on food, 162, 366t for latrines, 263 funding gaps and, 215­20, 216t, 218f, MDGs and, 16­18 219t, 220f, 226, 228­29f, 374­75t monthly spending on water, 161­62 GDP and, 215, 216­17, 220 operating costs recovery, 173­76 MDG coverage and, 18­24, 20­21t, price variations in urban water market, 202­9, 203t, 207­8f 162­68 operating cost recovery, 173­76, 174t, standpipes, 187­88 175f, 176t utility, 180­84 on operations and maintenance, 18, Sudan 216, 221, 223­24, 223t, 380­81t connection charges, 176 raising additional, 231­39, 376­77t as fragile state, 30 rehabilitation and O&M needs, open defecation, 68 quantifying of, 213­14 private investment, 236 for sanitation sector, 265 rural water supply, 113 subsidies and, 179­80, 180f sanitation, 193 unit costs of, 197­202, 199t, 209­13, spending needs, 206 211­13b, 214n1 standposts, 102 on water, 142­52, 161­62, 162f, subsidies, 179­80 171­72f, 171­73 traditional latrines, 68 Index 399 urban reform index, 89 water consumption levels, 143, 144f utilities, 99, 148 water sector regulation and, 92 water coverage, 193 taxation water sector regulation, 93 revenue collections, 232­33 supply chains, 51­52, 252, 253, 261 sanitation tax, 132, 134b, 265 surface water technical assistance, 255 access to water, 49­52, 276­77t technologies for water and sanitation reliance on, 12­13, 14f, 33, 56, 59, 252 services, lower-cost, 243­48, in rural areas, 10 245f, 246t, 247f in urban areas, 53 Togo surveys. See demographic and health sanitation, 193 surveys; multiple-indicator cluster spending needs, 206, 244 surveys; Water Supply and utilities, 148 Sanitation Survey water coverage, 193 Swaziland Togolaise des Eaux, 148 spending needs, 244 toilet facilities, 14, 65, 65f, 67t water coverage, 193 traditional pit latrines. See latrines Tynan, N., 142 typologies, country, 30­31 T Tanzania U budget allocations, 255­56 coverage rates, 42, 53, 193 Uganda institutional reforms, 252 bulk-water production, 87 latrines, 67, 68 coverage rates, 42, 193 ODA, 234 hand pumps, 52 piped water, 56, 59 institutional reforms, 252­53 sanitation, 193 latrines, 67, 68, 76 spending needs, 206 as low-income country, 31 tariffs, 130 performance contracts, 27 Unit Costs of Infrastructure Projects piped water, 34, 56 Study, 212b rural water supply, 110, 113, 115 urban reform index, 89 sanitation, 193 utilities, 85, 98, 99, 100 sewerage systems, 63 water collection, 50­51b spending needs, 217 water resale, 48 tariffs, 129, 154, 156b water sector regulation, 93­94 urbanization, 34 tariffs. See also subsidies urban reform index, 89 cost recovery, 17­18, 19f, 147b, 161, utilities, 98­99, 100, 156­57b 174, 179 water collection, 51b efficiency savings and, 226 water sector regulation, 94 increasing block tariffs (IBTs), 129, wells and boreholes, 58, 59, 259 168­70, 184, 187 unbundling of utilities, 24, 85, MDGs and, 16­18 153, 253 for piped water, 128­31, 131f underpricing of water, 257 policy options for, 257­58 unit costs. See also capital costs poverty line and, 186 matrix model, 209­13, 211­13b, sanitation and, 131­32, 133t, 358t 214n1 standposts, 164, 359­60t of sanitation, 177, 200, 201t two-part, 168­70f, 168­71 of service provision, 197­202, 199t utilities, 128­32, 131f, 133t, 146, Unit Costs of Infrastructure Projects 258, 350­53t, 356­57t Study, 211­13b 400 Index United Nations Children's Fund, 4 water supply and sanitation (WSS) Upper Nile Water Corporation, 148 access to sanitation, 63­81. See also urban areas. See also urban water markets access to sanitation access to water, 39, 41, 290t access to water, 33­61. See also access to groundwater, 258­60 water household budgets, 161 funding gaps, 215­49. See also funding institutional models for, 255­56 gaps institutional reform, 253 MDGs, 1­32. See also Millennium MDG targets, 193 Development Goals open defecation, 68, 70, 193 policy options, 251­67. See also policy piped water, 34, 43, 53, 369t options reform index, 85, 86t, 89, spending, 191­214. See also spending 90f, 294­99t Water Supply and Sanitation (WSS) septic tanks, 13, 67­68, 261 Survey sewerage, 264­65 on institutional quality, 84 standposts, 11, 43­44, 47, 47t on latrine construction and operation, surface water, 53 117­18 underpricing of water, 257 methodology, 7­9, 8t unit costs, 198, 200 snapshot, 3, 27­28 urbanization, 34, 36­37, 38b, 240 on standposts, 44, 103 urban-rural divide, 64, 202, 204f on utility operations and management, urban water markets, 84­110, 84f 123, 158­59 affordability threshold for WSS, on water reselling, 47 173­75, 176 water truckers, 48, 102­3, 110, 110t heterogeneity of, 84­102 water utilities, 9, 27, 123­60. See also institutional arrangements in, 84, 84f urban water markets nonpiped services, 102­10 access issues, 124­26, 125­26t, 324­27t. patterns in, 42, 43t, 45t, 287t See also access to water players in, 42­49 in AICD WSS database, 158­59 price variations in, 162­68, 165b, collection-efficiency ratios, 145­46, 166t, 167f 146f reforms across countries, 85­89 concessions for, 87 regulation, 25, 89­94, 91t, 92­93f, 96f connection rates, 39, 40b, 41, 43­44, resellers, 108­10 126­27 water tankers, 110, 110t cost of, 16­18, 162­68, 165b, 166t, water utilities, 94­102, 97­99f 167f coverage expansion, 126­28, 127f V debt-service ratios, 145 distribution system, 136­38, 137f, vendor water. See water resale 328­35t end users' water consumption, W 132­36, 136t Walvis Bay (Namibia), 132 financial efficiency, 142­46, 143t, Water and Sanitation Program, 107 344­47t water bills, 131­32, 169, 265 institutions and institutional reform, water consumption, 136t, 138, 159n1, 153­57, 153f, 155t, 156­57b, 253 168­70, 168t, 169f, 173­74, operations and pricing, inefficiencies in, 354­55t 146­52, 147b, 149f, 150t, 152f water gaps, 191, 192f overstaffing, 147, 147b, 148, 151, water kiosks, 43 152, 340­43t water resale, 11­12, 47­48, 48b, 102­3, performance indicators, 181­83, 108­10, 109t, 256­57 182­83f, 185f, 186 Index 401 piped water and, 128­31, 131f, 369t World Health Organization (WHO), production capacity, 128, 129t 4, 173 service quality, 138­39, 139t WSS. See water supply and sanitation sewerage charges linked to water bills, 131­32, 133t Y spending on, 221, 367­68t yard taps, 167, 256 standpipes and standposts, 103­5, 256 subsidies for, 180­84 Z tariffs and, 128­32, 131f, 133t, 146, 258, 350­53t, 356­57t Zambia technical efficiency in, 139­42, coverage rates, 42, 193 140­41t hand pumps, 52 treatment, 138, 336­39t institutional reforms, 252 unbundling of, 24, 85, 153, 253 piped water, 37, 56, 59 undermaintenance of, 151 as resource-rich country, 30 wells and boreholes sanitation, 117, 193 access to water, 33­37, 49­59, 274­75t septic tanks, 68, 71 groundwater from, 258­59 spending needs, 217 as improved water source, 10, 12­13 standpipes, 105 increase in use, 52­59 tariffs, 132 policy options for, 251­52 traditional latrines, 68 in rural water market, 49­52 Unit Costs of Infrastructure Projects unit costs of, 200, 200t Study, 212b WHO (World Health Organization), utilities, 94, 97­98, 99­100 4, 173 water collection, 51b Wodon, Q., 6, 41, 50b water sector regulation, 93­94 women Zimbabwe latrine construction by, 78b improved latrines, 68 water collection by, 49, 50­51b open defecation, 76 World Bank sanitation, 193 concessional funding, 236 septic tanks, 67, 68, 76 rural water assessment in Cross River spending needs, 244 State (Nigeria), 111b water coverage, 193 unit cost estimates, 200, 211 wells and boreholes, 259 ECO-AUDIT Environmental Benefits Statement The World Bank is committed to preserving Saved: endangered forests and natural resources. The · 8 trees Office of the Publisher has chosen to print · 3 million British Africa's Water and Sanitation Infrastructure: thermal units of Access, Affordability, and Alternatives on total energy recycled paper with 50 percent post-consumer · 781 pounds of net waste, in accordance with the recommended greenhouse gases standards for paper usage set by the Green (CO2 equivalent) Press Initiative, a nonprofit program support- · 3,761 gallons of ing publishers in using fiber that is not sourced waste water from endangered forests. For more informa- · 228 pounds of solid tion, visit www.greenpressinitiative.org. waste The Millennium Development Goals (MDG) have called attention to deficiencies in the quantity and quality of water supply and sanitation (WSS) globally. Although most of the world is on track to meet the MDG drinking water target, Africa lags behind. Only 58 percent of the population enjoys access to safe drinking water. According to projections, 300 million more people--almost 38 percent of the region's population, or half the number of people who currently have access to improved water--will need to be covered to meet the MDG target. Similarly, more than 2.5 billion people remain without improved sanitation worldwide; of that total, 22 percent, corresponding to more than half a billion people, lives in Africa. With the MDG deadline fast approaching, it is essential to take stock of the WSS sectors in Africa, analyze their achievements and shortcomings, and identify the sector characteristics that either advance or inhibit the population's ability to access service. Africa's Water and Sanitation Infrastructure--Access, Affordability, and Alternatives integrates a wealth of primary and secondary information to present a quantitative snapshot of the state of the WSS sectors in Africa. It explains the sectoral institutional structures and utility performance and articulates the volume and quality of financing available over time. The authors also evaluate the challenges to the WSS sectors and explore the factors that govern the expansion of coverage over time. Finally, the authors estimate spending needs for WSS, arriving at a funding gap for meeting the MDGs. The proposed directions for the future draw on lessons learned from best practices and present the menu of choices available to African countries, bearing in mind that the challenges differ to a significant extent among countries and solutions must be tailored to national or regional conditions. ISBN 978-0-8213-8457-2 SKU 18457