Water Papers Water Papers 67321-AFR December 2011 More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa Meike van Ginneken, Ulrik Netterstrom, and Anthony Bennett AFTUW Africa Region The World Bank Water Papers are published by the Water Unit, Transport, Water and ICT Department, Sustainable Development Vice Presidency. Water Papers are available on-line at www.worldbank.org/water. Comments should be e-mailed to the authors. Approving Manager Alexander Bakalian Contact Information The paper is available online at http:www.worldbank.org/water. Authors may also be contacted through the Water Help Desk at whelpdesk@worldbank.org. Disclaimer – World Bank © 2012 International Bank for Reconstruction and Development / International Development Association or The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the govern- ments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to 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. About the Authors Meike van Ginneken works as a sector leader for sustainable development at the World Bank and is based in Yaoundé, Camer- oon (mvanginneken@worldbank.org). Ulrik Netterstrom is an independent consultant working on water and sanitation and is based in Copenhagen, Denmark (un@ rdc.as). Anthony Bennett is an independent consultant working on public financial management issues and is based in London, United Kingdom (tonyun2@aol.com). Acknowledgments We would like to thank Joachim Boko (consultant) for his contributions to the analysis. We are grateful to the authors of the 15 Public Expenditure Reviews upon which this review is based and their efforts in collecting, cross-checking, and analyzing data from the bulk of the work carried out for this review. We would also like to thank colleagues who provided inputs and reviewed earlier drafts of this overview paper including Dominick de Waal, Matar Fall, Vivien Foster, Bill Kingdom, Seema Manghee, Diego Rodriguez, Ahmed Shawky, Caroline van den Berg, and Mike Webster. We would also like to acknowledge the guidance and support received from Shanta Deverajan, Junaid Ahmad, and Alexander Bakalian. This review was financed by the Water Partnership Program. Contents Acronyms and Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi Executive Summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.1 Why Should Governments Spend their Scarce Financial Resources on WSS? . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.2 The Sample of Countries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.3  A Reading Guide for Various Audiences. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.4 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2. Trends in ACCESS to Water and Sanitation Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 2.1 Access to Water in Rural Areas. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 2.2 Access to Water in Urban Areas. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 2.3 Access to Sanitation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 3. Who is Spending What? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 3.1 Expenditure as a Share of GDP. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 3.2 Expenditure Per Capita. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 3.3 Sector Budget as a Share of Government Budget. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 3.4 Sector Funding Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 3.5  Trends over Time. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 4. How is money Being Spent?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 4.1  Capital and Current Expenditure. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 4.2  Salary Component . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 4.3  Targeting. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 5. Budget Execution Rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 5.1  Trends in Budget Execution. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 5.2 Systemic Problems Causing Low Rates of Budget Execution. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 5.3 Sector Specific Reasons for Low Rates of Budget Execution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 6. Exploring the Link between Public Expenditure and Improved Access . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 iii 7. Underlying Public Economy Dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 7.1  What Is the Rationale for Public Spending in the WSS sector? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 7.2  Do Actual Public Spending Patterns Match their Rationale?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 7.3  What are the Reasons behind the Misallocation and Poor Implementation of Public Resources in WSS?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 8. Concluding Remarks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 8.1  PERs are a Useful Tool for Policy Makers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 8.2  More, Better, or Different Spending?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Annexes Annex 1: PERs in African Countries with Chapters on Water Supply and Sanitation. . . . . . . . . . . . . . . . . . . . . . . 42 Annex 2: Remarks on Data Limitations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Annex 3: WSS Budget Execution Rates — 2000–2008. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 Annex 4: References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 Annex 5: Detailed Data Sheets for Countries included in the Review. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 Figures Figure A: Public Expenditure as a Percentage of GDP and on a Per Capita Basis . . . . . . . . . . . . . . . . . . . . . . . . . . viii Figure 4.1: Expenditure as a Percentage of GDP and on a Per Capita Basis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Figure 4.2: Donor Expenditure Per Capita (US$) Ordered by Country GDP/capita (US$). . . . . . . . . . . . . . . . . . . . 18 Figure 5.1: WSS Budget Execution Rates 2000–08 (%). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 Figure 6.1: Per Capita Spending and Access Rates to Drinking Water for All Reviewed Countries . . . . . . . . . . 27 Figure 6.2: Example of Relationship between Spending and Access (Mali 2001–06) . . . . . . . . . . . . . . . . . . . . . . . 28 Tables Table 1.1: Key Indicators for Countries Included in this Review. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Table 2.1: Access to Drinking Water in Rural Areas—2000−08 (% of population). . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Table 2.2: Access to Drinking Water in Urban Areas, 2000–08 (% of population) . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Table 2.3: Key Performance Indicators for Utilities in Reviewed Countries. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Table 2.4: Urban and Rural Access to Improved Sanitation in 2008 (%). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Table 3.1: Expenditure as a Percentage of GDP. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Table 3.2: Expenditure Per Capita (US$). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 Table 3.3: WSS Budget as a Share of Overall Government Budget (%). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 Table 3.4: Externally Funded Expenditure as a Share of Total WSS Expenditure (%). . . . . . . . . . . . . . . . . . . . . . . . 14 Table 4.1: WSS Recurrent Expenditure as a Percentage of Total WSS Expenditure. . . . . . . . . . . . . . . . . . . . . . . . . 15 Table 4.2: Salaries as a Share of Recurrent Expenditure (%). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 Table 5.1: Budget Execution Rates for WSS Budgets (%). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Table 5.2: Domestic WSS Budget Transfers to Subnational Governments, 2002–09. . . . . . . . . . . . . . . . . . . . . . . . 24 Table 6.1: Comparison between Annual Expenditure and Increase of Water Access Rate in Selected Countries during Study Period. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 iv  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa Boxes Box 1.1: How Representative Are the 15 Reviewed Countries of Africa as a Whole?. . . . . . . . . . . . . . . . . . . . . . . . 3 Box 2.1: Post-construction Support in Ghana. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Box 2.2: Private Sector Participation in the Operation and Maintenance of Rural Water Schemes in Niger. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Box 3.1: Comparing the PER Findings with the Africa Infrastructure Country Diagnostic (AICD). . . . . . . . . . . 12 Box 5.1: Addressing Systemic and Sector-Specific Problems in the Budget Execution Chain in Benin. . . . . 20 Box 5.2: Overcoming Procurement Bottlenecks in Ethiopia. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 Box 5.3: Matching Decentralization with Increased Financing to Local Governments in Tanzania . . . . . . . . 24 Box 5.4: Decentralization of Budgets in Mali. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 Box 6.1: Linking Public Spending to Increased Access in Tanzania. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 Contents | v Acronyms and Abbreviations AfDB African Development Bank MIC middle-income country AICD Africa Infrastructure Country Diagnostic MTEF Medium-Term Expenditure Framework AMCOW African Ministers Council on Water NGO nongovernmental organization COFOG Common Functions of Government O&M operation and maintenance (UN system used by IMF-GFS and PEFA) OECD Organisation for Economic Co-operation and CSOs Country Status Overviews Development DAC Development Assistance Committee PEFA Public Expenditure and Financial Accountability GDP gross domestic product PER Public Expenditure Review GFS Government Finance Statistics PETS Public Expenditure Tracking Surveys HDR Human Development Report PRSP Poverty Reduction Strategy Paper HIPC highly indebted poor countries RWSN Rural Water Supply Network IB-NET The International Benchmarking Network for UN United Nations Water and Sanitation Utilities UNDP United Nations Development Programme IBT increased block tariff UNICEF United Nations Children’s Fund IMF-GFS International Monetary Fund, Government WDI World Development Indicator Finance Statistics WDR World Development Report JMP Joint Monitoring Program WHO World Health Organization LICs low-income countries WSP Water and Sanitation Program MDG Millennium Development Goal WSS water supply and sanitation vi  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa Executive Summary I mprovements in water supply and sanitation access in Sub- The Rationale for Public Spending on Saharan Africa have fallen short of national and international WSS targets. Past explanations for this slow progress toward the Millennium Development Goals (MDGs) include lack of capac- This overview paper tests current public spending patterns ity, low levels of financing, inappropriate technologies, and against the economic rationale for such spending, including institutional and governance challenges. This overview paper reducing disparities in service delivery and overcoming market zooms in on the public financing for water supply and sanita- failures. Reducing the disparities in access to basic WSS is a tion (WSS) in the anticipation that knowledge of the quantity, responsibility of government. Individuals have little incen- quality, and drivers of public expenditure is a prerequisite for tive to build and maintain extensive WSS infrastructure, but governments and donors to adjust their policy and practice, communities and societies do. Targeted public spending which is itself a prerequisite to improved access to WSS. benefitting households that otherwise would be unable to afford those services can be a component of a broader This overview paper aims to determine the size and composition social policy agenda to redistribute resources to the poor. of the flow of funds, assess the quality of public spending, and Several market features call for government intervention in identify common issues in public expenditure in the water sec- the WSS sector: tor. How much public money (domestic resources and do- nor financing) was budgeted for the sector? Was it spent in ■■ WSS is a mixed public and private good; while services a timely manner? Who benefits from public spending? What benefit individuals, they also have considerable public are the major bottlenecks in increasing expenditure and the health and environmental benefits. efficiency of these expenditures? ■■ WSS is a good example of a natural monopoly, as WSS network infrastructure cannot practically be duplicated. This overview paper presents an assessment of the findings of ■■ The sector is characterized by a high degree of sunk public expenditure reviews (PERs) conducted by the World Bank costs. in 15 countries in Sub-Saharan Africa in the past few years. Eight ■■ The sector suffers from imperfect information, which of these PERs concentrated on rural WSS only (“rural-only can lead to less than desirable investment and con- countries” in this executive summary) and seven considered sumption. both urban and rural WSS (“rural-and-urban countries”). The scope of the present review includes expenditures by public Public Expenditure Falls Short of institutions (at central and local government levels) on domes- Government Statements tic resources and grants or loans provided by external funding agencies. The review does not include off-budget spending Public expenditure on WSS averaged 0.39 percent of gross do- by water utilities. In other words, while the numbers in this mestic product (GDP) ($1.71 per person) in rural-and-urban review include public subsidies to utilities, they do not in- countries, and 0.26 percent ($1.21 per person) in rural-only clude utility spending, including spending funded by cost- countries (figure A). This is well below the 1 percent bench- recovery from consumers. mark suggested by the 2006 Human Development Report, vii Figure A: Public Expenditure as a Percentage of forward, economic growth and increasing the tax base of GDP and on a Per Capita Basis governments is, therefore, a key ingredient to progress in the sector. 1.00% 5 rural only rural and urban 0.75% 4 Donor financing was not targeted at countries with low levels 3 of access. Donor targeting is guided by factors such as po- 0.50% litical stability, adherence to principles of good governance, 2 and sound project financial-management processes. Donors 0.25% 1 often target relatively rich countries that have already made 0.00% 0 some progress on WSS access. Burkina Faso Cameroon Côte d’Ivoire Ethiopia Ghana Madagascar Mali Niger Central Africal Rep. Congo, Dem. Rep. of Mozambique Congo, Rep. of Sierra Leone Tanzania Togo Public-Expenditure Targeting Exhibited a Bias toward Capital Budget, Water as % of GDP Per capita (in USD) Supply, and Capital Cities Source: PER Reports. According to national PERs, public expenditure favored develop- ment (87.3 percent of total sector expenditures) over recurrent expenditure (12.7 percent), half of which was subsequently al- located to salaries. Underfunding of basic maintenance se- and one-tenth of what is needed to meet the MDGs for Sub- verely limits public sector institutions’ ability to fulfill their Saharan African states. operations, maintenance, monitoring, and enforcement duties and weakens sectoral capacity to bolster access rates Many of the countries in the sample are highly donor depen- and reach MDG targets. dent, with on average 61.9 percent of total WSS expenditures coming from donor financing. On average, 2 percent of total In general, the sanitation subsector, maintenance of existing government expenditures go toward the WSS sector. WSS facilities, and areas outside of capital cities appeared to be underfunded. We observed a general upward trend in public expenditures for the sector, both absolutely and as a share of GDP and per capita. Investments in urban water supply have not kept up with ur- Annual expenditure on WSS is extremely volatile, which ob- banization and population growth. Urban access to drink- structs efficient budget execution. ing water has fallen by 1 to 2 percent each year, on aver- age, over the past decade. In many cases, inefficient water utilities are underperforming and are increasingly a drain Needier Countries Spent Less of Their on state coffers, with revenues unable, or barely able, to Own Resources and Received Less cover operations and maintenance (O&M) costs. As a re- Donor Funds sult, water utilities often used transfers from national gov- ernments to maintain, rather than expand, current cover- We noted large disparities in public WSS expenditures between age. This is effectively a subsidy to households that already countries. For instance, per capita expenditure in the urban- have piped water, the majority of whom are in the top and-rural countries ranged from $0.01 per person (Demo- 40 percent of income distribution. Further, funding new cratic Republic of Congo, 2002) to $8.93 per person (Repub- household connections often receives priority over pub- lic of Congo, 2006). lic standpipes. We noted that few countries have specific public expenditure programs for those who are not con- In cross-country comparisons, richer countries spent more per nected to the network. capita on WSS, but their spending comprised a slightly smaller share of overall GDP. Public expenditure on WSS as a share Investment in rural water supply just kept pace with popula- of GDP depends heavily on the total public expenditure as tion growth. Over the past decade, rural access to drinking a share of GDP, and thus on the tax revenue ratio. Going water increased by 1 to 2 percent on average each year in viii  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa the countries surveyed. More than half of rural dwellers still Bottlenecks along the Budget do not have access to drinking water, though coverage in Execution Chain: Budget Ambition Sahelian countries is substantially better than in other Afri- can countries, as is public expenditure on rural water supply. and Volatility, Capacity Constraints, A key obstacle to improved rural access is the breakdown and Incomplete Reforms rate of rural water supply facilities, which was 25 percent or higher in most of the countries surveyed and more than 50 Obstacles and bottlenecks to improving low budget execution percent in postconflict countries. rates occur all along the budget execution chain and are to some extent outside the control of water sector professionals. We dis- Sanitation receives only a small part of public expenditure, even tinguish between level of ambition and volatility of budgets, though only one in five households in the countries considered capacity gaps, incomplete implementation of sector reforms, had access to improved sanitation in 2008. As the sanitation and incomplete decentralization. In most countries it has sector is dominated by household on-site facilities and is been hard to distinguish which obstacle is the most critical. generally financed from household expenditures, limiting public expenditure in this area might be partly justified. Low execution is caused partly by overambitious plans and bud- The data are admittedly limited and uncertain but suggests gets. Several PERs noted that line ministries are not adequately some progress in access to improved sanitation in rural ar- involved in the budgeting process. As a result, line ministries eas over the past decade. Huge public investments will soon are not interested in budget preparation work, which they be required as countries get richer and people move up the view as an exercise involving the ministry of finance only. sanitation ladder and move toward public infrastructure, such as condominial sewerage. The volatility of WSS budgets forms a major obstacle for efficient budget execution as the unspent budget of one year cannot typi- The review showed large regional expenditure disparities with- cally be carried over into the next year. Delays in budget process- in countries. National utilities tend to privilege capital cities, es exacerbate this problem, as contract bid invitations cannot leaving secondary cities underfunded, understaffed, and be issued until the budget is approved, and spending autho- sometimes without functioning facilities altogether. Im- rization may not happen until months into the fiscal year. portant regional disparities persist in rural access to drink- Donor-funded development expenditure has the lowest and ing water, with sizeable gaps between the best served most erratic rates of budget execution. Donor planning and and less served. Public expenditure often goes to where monitoring is not necessarily linked to the government bud- it is most easily spent instead of where it is most urgently get calendar; information on multiple parallel donor systems needed. can be hard to track, disbursement processes can be lengthy and administratively cumbersome, and delays in counterpart contributions can further slow donor disbursements. Chan- Almost Two-Thirds of Water Supply neling funds directly to regional and district authorities helps donors avoid delays and reduces risks of unauthorized cash and Sanitation Budgets Were diversions along the way, but levels bypassed in the process Executed do not always get the information, compounded by still weak local standards of accounting and reporting. Recurrent budgets (70 percent) were executed more often than development budgets (62 percent). Budget execution in rural- Further downstream, another systemic obstacle to budget ex- only countries (66 percent) was higher than that in rural-and- ecution is the lack of capacity, mainly project management and urban countries (47 percent). In the eight countries where contracting capacity in the government and its partners. Well- data were available, execution of domestically funded sector considered plans and budgets can be undermined by trun- expenditure fared better (66) than externally funded expen- cating planning horizons, scrambling projects, splitting con- ditures (57 percent). But the distinction between “develop- tracts, and undercutting economies of scale. Procurement ment” expenditure and “recurrent” expenditure is becoming procedures can be cumbersome, with multiple donor agency increasingly blurred, as donor funding seems to include con- rules and procedures creating huge demands on host gov- siderable rehabilitation expenditures that could be classified ernments with limited capacity. Throughout, costs rise with as either capital or current. delays, inflation, and stop-and-start resource management, Executive Summary | ix and suppliers soon become reticent to bid on government A Note on Data contracts in the future, or they raise their prices in defense. All public expenditure reviews faced serious limitations with respect to data definitions, classifications, and coverage. We Spending public budgets is slowed by unclear responsibilities found incomplete or contradictory data in many countries, due to unfinished implementation and enforcement of water compounded by the fact that “water and sanitation” is not a sector reforms. Nearly all of the countries studied have elabo- distinct stand-alone sector that would enable international rated a comprehensive set of water sector policies and sub- comparison of government finance. Additionally, a large part sector strategies. Multiple sector institutions—some newly of donor resources are off budget, leaving sizeable holes in created—have overlapping or conflicting mandates and the bigger picture of public expenditure for WSS. On the strategies, with insufficient human and financial resources to sector side, access is often measured using different meth- go around. odologies, even within a given country over time. Full stan- dardization of PERs would serve global officials but would While most countries have officially decentralized all or part of limit the use of the PERs for their main audience—in-country WSS service delivery, the devolution of responsibilities to local, stakeholders (including politicians), government officials, regional, and district authorities has not come with the requi- donor representatives, and civil society representatives—be- site authority, budget, staff, and operational support to imple- cause the WSS sector is defined and organized differently in ment them. We found that WSS budgets were almost never different countries. transferred to local authorities, with the notable exception of Tanzania. In rural areas in particular, incomplete decen- tralization has created a dangerous institutional vacuum, where neither national nor local governments have taken Patterns of Public Spending are responsibility for the safe provision of water. Water supply in Stark Contrast with the Public in urban areas has been less disrupted, if only because na- Spending Rationale tional utilities continue to provide water in two-thirds of the countries reviewed. Central governments sometimes Public spending contributes little to overcoming market fail- use the insufficient management capacity of lower levels of ures. Current public spending patterns are not in line with government to rationalize their reticence to give up power, stated public health objectives; they focus on higher-level budgets, and staff. services at the expense of channeling money into cheaper service levels that would have considerably higher health returns per dollar invested, leaving sanitation underfunded. The Current Link between Spending Public spending patterns do not fully reflect the long as- and Services set life of WSS infrastructure, as the upkeep of existing WSS facilities appears to be underfunded. Low water tariffs un- No countries have managed to increase access without a sub- dermine the rationale that governments prefinance capital stantial increase in public expenditure. In other words, increas- investments that will be recovered from consumers over ing public expenditure on WSS is a necessary but insufficient time. tool for improving services. For instance, successful programs to improve access to water supply in Senegal, Burkina Faso, Current spending patterns contribute to inequality, despite the and Niger all included large public investment programs. fact that the gap between rural and urban access is closing. Similarly, progress in rural water supply access in Benin and We found that the poor do not get their fair share of public Mali has followed increases in public expenditure. spending on services, let alone the larger share that might be justified on equity grounds. Spending is skewed to ser- Overall, however, we found no relationship between levels of vices disproportionately used by richer people in capital cit- spending and levels of access to water supply and sanitation. ies at the expense of people in slums, secondary cities, and This might be due to poor data quality, might be an indica- rural areas. Water utility tariff subsidies for WSS are starkly tion that spending is not the key determinant in increas- regressive. Social connection programs that provide subsi- ing access to water supply, might be because the review dies to reduce the costs of connecting to the network are period was too short, or might be a combination of these a progressive alternative to consumption subsidies in some reasons. countries. x  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa Weak utilities act as a buffer between public spending and pub- ing profession form the background of the strong barriers to lic policy, reducing “bang for the buck.” The PER review findings technological innovations. show that a large part of public spending on urban WSS is absorbed by utilities to cover recurring losses caused by rev- Global debates are influencing local decision making. The enues being well below O&M costs. Only 36 percent of the sometimes heated debate on privatization led to blanket op- utilities in Africa have tariff levels to meet their full O&M costs. position to cost-recovery by some politicians and activists. But the global environmental debate drove a push for higher levels of sanitation services, and the recent global move- There are Several Drivers behind the ment for basic sanitation has further rebalanced the debate. Misallocation and Poor Implementation The observed gaps between policies and practice can partly be traced back to policy prescriptions from donors, which of Public Resources in WSS have been adopted superficially but have not been followed Spending patterns in WSS are in line with cross-country evidence through because of local political resistance. In this respect, that clientelism significantly influences the provision of public ser- it is interesting to note that the call for more direct account- vices. The tendency for political patrons to provide private re- ability of service providers by increasing the client’s power wards to clients can help explain the disproportionate spend- has not really taken hold in the sample of countries yet. ing in capital cities. Public money is often spent where the politically powerful reside; this is where elections are won, or at least where potential social discontent has to be controlled. More, Better, and Different Spending Political patronage might also explain low revenue collection There are compelling arguments to increase public spending for caused by uncollected bills and malfunctioning meters. A po- WSS. Redistributive arguments and market failures call for pub- litical economy perspective on public service delivery suggests lic intervention. The investment needs are huge, but this re- that choices in capital spending may be driven by the corrup- view found that current spending patterns are inefficient and tion, employment, and profit opportunities that construction ineffective and do not match the public spending rationale. provides. Utility management decisions are often driven by the Increasing the volume of public expenditure without chang- interests of their employees or organized labor. Politicians and ing the targeting and execution will not have a large impact. central bureaucrats have been allowed to keep public budgets concentrated in national administrations, stalling decentraliza- The findings of this review point to a strong need for better tion and leading to a dangerous institutional vacuum in the budget execution. A major focus should be on solving insti- rural provision of water supply with neither national nor local tutional bottlenecks in WSS public expenditure both within governments fully taking responsibility. This might be conve- and without the control of sector professionals. Such profes- nient for all—except rural people without water. sionals, including international actors, should broaden their scope from the design to the implementation of sector poli- Politicians’ refusal to raise tariffs makes for good political pro- cies and should work with others to address the bottlenecks poor rhetoric but in practice mainly benefits the middle and along the whole of the budget execution chain. The chal- upper classes that are connected to the public water network. lenge for practitioners is not to identify the perfect “magic Recovering full costs from existing customers and using the bullet,” but the more subtle one of building capacity and in- resulting cash flow to accelerate access expansion for the stilling and maintaining appropriate management cultures. poor would substantially increase equity, although it is a The choice of budget execution tools must be appropriate hard political sell. to the current and evolving state of the country and sector. Looking at political dynamics also helps explain why sanitation The need for better targeting is a major conclusion of this review. is an orphan sector, suffering from slow technology change. This includes channeling funds to the sanitation subsector, Low household demand for sanitation results in politicians to areas outside of the capital, and to the upkeep of existing not seeing sanitation as a vote winner, and therefore allo- WSS facilities that currently appear underfunded. cating scarce resources to sectors with higher perceived po- litical rewards. But sanitation is a cheap lifesaver, and as such A review of PERs reveals huge gaps between policy and practice. might merit higher public spending. Mistrust of cheaper PERs can be a useful tool to hold governments accountable service levels and other cultural norms within the engineer- for the implementation of their own policies and promises. Executive Summary | xi At the sector level, we found that while nearly all countries government budget calendar. Transaction costs for govern- have elaborated comprehensive water sector policies and ments are high. Donors should consider rethinking extend- strategies, implementation and enforcement of sector re- ing their financing to not only cover “development” expendi- form strategies remain incomplete, and efforts are needed tures, but also maintenance—especially in rural areas. in terms of capacity building, general public awareness cam- paigns, and further development of the legal framework that Closing the implementation gap and improving the efficiency would facilitate implementation of policies and strategies. of public spending will require addressing the underlying pow- We have also found that, rather than streamline the process, er patterns. Changes in spending patterns will risk being reforms had, in many cases, led to the creation of new in- marginal or temporary if these factors are not properly ad- stitutions with overlapping mandates. Furthermore, decen- dressed. But addressing underlying power patterns should tralization has stalled, with little or no progress in devolving not be an excuse to pursue an exclusively technocratic ap- financial resources to local government. This has created a proach to improving the targeting and execution of WSS dangerous institutional vacuum in the provision of water public spending; instead, technocratic short-term mea- supply, particularly in rural areas, as neither national nor local sures should be complemented by exploring, exposing, governments are fully taking responsibility. and addressing the longer-term drivers behind the current status quo. PERs can be a powerful tool to help change A second implementation gap is seen in donor financing, which social dynamics as they can help to open up debate by is often badly targeted and unpredictable, resulting in execu- showing the lack of efficiency in current public spending. tion rates that are lower than those of internal resources. A sig- Advocates for better WSS services can use the outcomes of nificant opportunity is available for more pro-poor-targeted PERS, as well as this regional review, to expose the capture donor financing by shifting resources to the areas with the of public resources by interest groups and lobby for dif- largest WSS needs. We found that the WSS sector in most ferent and better spending in the sector. As the quality of countries was characterized by a large number of donors op- spending will improve, the argument for more spending erating on terms and conditions specific to their individual will not only be based on the compelling needs in the sec- projects. Thus, donor funding is unpredictable, and donor tor but also grounded in the economic rationale for public planning and monitoring is not necessarily linked to the spending. xii  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa 1. Introduction T his review mines the rich data of 15 Public Expenditure Re- The public expenditure analyses in all reviews focus on WSS ser- views (PERs) conducted in Sub-Saharan Africa and funded vices, although some reports also discuss water resources man- by the World Bank over the past years. From 2003 the agement. Almost all of the PERs, however, are limited to WSS, World Bank has funded more than 40 PERs that contain an thus excluding water resources management and irrigation analysis of the water supply and sanitation (WSS) sector. In issues from the analyses. Furthermore, most of the studies most of these, the WSS sector is discussed alongside other focus on rural water supply and it is only in recent years that sectors. A set of stand-alone PERs specifically addressing the the urban water supply subsector has been included in the WSS sector have also been carried out in African countries.1 scope of the reviews. The purpose of the present review is to provide, based on the 15 Sanitation is defined as the sanitary removal of liquid waste and PERs selected, an overview of: excreta and the promotion of hygiene. The approach to sanita- tion, however, is not fully consistent across the various PERs. ■■ The size and composition of the flow of funds of to Sanitation is a broad concept potentially covering a large sector. range of intervention areas such as household sanitation, ■■ The quality of public spending in the sector including public latrines, waste water treatment, drainage, and solid distributional issues (who benefits from public spend- waste disposal.2 The review is limited to household-based ing? rich/poor, urban/rural, water/sanitation, coastal/ sanitation—that is, latrines—and excluded expenditure re- inland, and so on), efficiency (cost recovery, and so on), lated to other forms of sanitation except where this is not public financial management (tracking budget alloca- separable (for example, the PER for Togo also deals with solid tions, including tools such as public expenditure track- waste disposal). ing surveys [PETS], assessing investment planning and allocation policies, and so on), and trying to link sector The reviewed PERs did not use standard definitions, which has spending to outcomes (access to services, quality of led to some data limitations described later. This review is a services). data mining exercise of country PERs that were written to ■■ Common issues in public expenditure in the water sec- serve in the political dialogue on the challenges in achiev- tor, such as major bottlenecks in increasing expenditure ing the Millennium Development Goals (MDGs) by 2015 and and the efficiency of these expenditures. on bottlenecks in enhancing public finance management performance. These PERs provide a wealth of data on WSS The scope of the present review includes expenditures by public sectors in the surveyed countries; this overview paper tries institutions (at the central and local government levels) on do- to draw regional conclusions from that data. mestic resources and grants or loans provided by external fund- ing agencies. The review does not include off-budget spend- ing by water utilities. In other words, while the numbers in 1 The list of PERs carried out in the WSS sectors in African countries is this review include public subsidies to utilities, they do not included in annex 1. include expenditure by utilities, thus disregarding expendi- 2 This is especially true in francophone countries where the term as- tures paid for by consumer cost-recovery. sainissement covers a large range of services. 1 Why Should Governments Spend 1.1  Table 1.1: Key Indicators for Countries Included in this Review their Scarce Financial Resources on WSS? Land area Population Urbanization (1,000 sq. GDP/capita This review tests WSS public spending patterns against the eco- Country (millions) rate (%) km) ($) nomic rationale for such spending, including reducing disparities Burkina Faso 13.2 18 274 425 in service delivery and overcoming market failures. Reducing the Cameroon 16.3 54 465 1,035 disparities in access to basic WSS services is a responsibility Central African 4.0 38 623 337 of government. Individuals have little incentive to build and Republic maintain extensive WSS infrastructure, but communities and Congo, Dem. 57.5 32 2,267 126 societies do. Targeted public spending benefitting house- Rep. of holds that otherwise would be unable to afford those ser- Congo, Rep. of 4.0 47 342 1,522 vices can be a component of a broader social policy agenda Côte d’Ivoire 18.2 47 318 899 to redistribute resources toward the poor. Several market Ethiopia 71.3 16 1,000 160 features call for government intervention in the WSS sector: Ghana 22.1 48 228 486 ■■ WSS is a mixed public and private good; while services Madagascar 18.6 29 582 271 benefit individuals, they also have considerable public Mali 13.5 31 1,220 387 health and environmental benefits. Mozambique 19.8 35 784 345 ■■ WSS is a good example of a natural monopoly, as WSS Niger 14.0 16 1,267 238 network infrastructure cannot practically be duplicated. Sierra Leone 5.5 37 72 221 ■■ The sector is characterized by a high degree of sunk costs. Tanzania 38.3 24 884 302 ■■ The sector suffers from imperfect information, which can Togo 6.1 40 54 346 lead to less-than-desirable investment and consumption. Source: World Development Indicators (World Bank 2005). 1.2  The Sample of Countries Fifteen countries have been selected for the review. From the agement specialists, and others working across sectors on pub- 40 World Bank–funded PERs, we made a selection based on: lic management and service delivery. The overview paper is (i) similarity of country circumstances (all countries are Sub- structured as follows: Saharan African countries); (ii) period of review (from 2002 to 2008); and (iii) scope and depth of PERs. Table 1.1 presents ■■ Chapter 2 reviews trends in WSS access. some key data on the sample countries. ■■ Chapter 3 identifies and classifies sector expenditures, looking into how much is being spent and the sources About half of the PERs deal only with rural WSS, while the others of funding. deal with both urban and rural WSS. The PERs focusing on rural ■■ Chapter 4 looks at the targeting of sector expenditures, WSS cover the following countries: Burkina Faso, Cameroon, including recurrent and capital expenditures and how Côte d’Ivoire, Ethiopia, Ghana, Madagascar, Mali, and Niger. The these are split by subsector and geography. PERs that cover both urban and rural WSS are: the Central Afri- ■■ Chapter 5 examines budget execution and analyzes the can Republic, the Democratic Republic Congo, Mozambique, budget execution chain to explore the obstacles to bet- the Republic of Congo, Sierra Leone, Tanzania, and Togo. ter budget allocation and execution. ■■ Chapter 6 examines the link between expenditure and access to WSS. A Reading Guide for Various 1.3  ■■ Chapter 7 tests the observed spending pattern against Audiences the economic rationale for public spending on WSS and explores the political economy underlying the spend- This overview paper is structured to serve multiple audiences, ing patterns. including WSS professionals and economists, financial man- ■■ Chapter 8 provides some concluding remarks. 2  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa Box 1.1: How Representative Are the 15 Reviewed Countries of Africa as a Whole? The Africa Infrastructure Country Diagnostic (AICD) uses a four-way country typology to explore the underlying drivers of progress: • Middle-income countries (MICs) have per capita gross domestic products (GDPs) in excess of $745 but less than $9,206. • Resource-rich countries are countries whose behaviors are strongly affected by their endowment of natural resources. • Fragile low-income countries face particularly severe development challenges, such as weak governance, limited administrative capacity, violence, or a legacy of conflict. • Nonfragile, low-income countries compose a residual category of countries with per capita GDP below $745 that are neither resource rich nor fragile. The 15 countries reviewed in this report fall into three out of four of these groupings; no MICs were reviewed. The distribution across the other three groupings is relatively similar to the overall distributions of African countries, although the resource-rich countries are somewhat underrepresented. The sample is thus reasonably representative for the whole of Sub-Saharan Africa. Sub-Saharan Africa (excluding MICs) Sample of 15 PER countries Number % Number % Countries Resource-rich countries 9 24 2 13 Cameroon, the Republic of Congo Fragile, low-income 13 35 5 33 The Central African Republic, Côte d’Ivoire, the Democratic Republic of Congo, countries Sierra Leone, Togo Nonfragile, low-income 15 41 8 53 Burkina Faso, Ethiopia, Ghana, Madagascar, Mali, Mozambique, Niger, Tanzania countries Source: Foster and Briceño-Garmendia 2009. WSS professionals probably might want to focus on chapter 3 on the topic. A references section is included at the end of and higher. While chapter 2 may be of interest to them, it the paper. All data throughout the overview paper are from contains information they will be familiar with. the core data set derived from the 15 PER studies unless oth- erwise indicated. All financial data presented are based on Generalists interested in the WSS sector might be specifically nominal figures. interested in chapters 2 to 4 and chapter 6 and higher. These chapters contain a wealth of information on the current sta- Each of the individual country PERs followed a similar method- tus of the sector and its public expenditures (chapter 2 and ology, combining a review of literature and government docu- 3) and a political economy perspective on the sector (chap- ments with interviews and consultations with sector actors. ter 7). The discussion on the budget execution chain in chap- Budget data were normally obtained from the ministry of ter 5 contains information they will be familiar with. finance and cross-checked with data from the ministry or ministries responsible for WSS. Budget lines were individu- ally examined and assigned to the correct capital or recur- 1.4 Methodology rent expenditure category and by subsector. Expenditure The current study has been conducted purely as a desk study; we relied mainly on the data presented in the 15 PER studies along 3 Databases used include the World Bank’s World Development In- with some other sources that use a standard set of indicators. dicators (http://data.worldbank.org/indicator), data on access to water We used general geographic, economic, and population supply of the Joint Monitoring Program (JMP) (www.wssinfo.org), and data derived from several international databases to com- data from the International Benchmarking Network for Water and Sani- plement the data in the PERs.3 Several recent studies on WSS tation Utilities (IB-NET: www.ib-net.org). were used to compare and validate our findings.4 Chapter 7, 4 Notably, the Africa Infrastructure Country Diagnostic (AICD, www.in- on the underlying political economy considerations in the frastructureafrica.org/) and the Country Status Overviews (CSOs) of the African Ministers Council on Water (AMCOW), the Water and Sanitation WSS sector, follows a slightly different methodology than the Program (WSP), and the African Development Bank (AfDB, www.wsp.org/ rest of the chapters, as it builds on a broader literature review wsp/content/pathways-progress-status-water-and-sanitation-africa). Introduction | 3 estimates were obtained from sector ministries and directly ■■ WSS sector spending performance from agencies and subnational governments wherever nec- ■■ WSS sector budget execution rate essary and possible. Various donor assistance databases— ■■ Execution of WSS sector domestic resources maintained by governments or the United Nations—have ■■ Execution of WSS sector external resources been used to obtain donor data, and where necessary, do- ■■ Recurrent and investment spending as share of to- nor data were obtained directly from the respective donor- tal sector spending agency country offices. Similarly, in many countries, data ■■ Salary costs as share of total recurrent spending on expenditure by nongovernmental organizations (NGOs) ■■ Rural/urban investment spending as share of total were obtained directly from NGOs. Data from various sourc- WSS investment spending. es were cross-checked to avoid double counting. Draft PERs were discussed during meetings in all countries. Govern- This review faced considerable data limitations. As a result, all ments were requested to provide written comments on final analyses in this review are based on partial and sometimes drafts, which were incorporated in the final PERs. PER teams unreliable data and should be used with care. Data scarcity combined water sector expertise with experience in public poses problems well beyond this review. To set and monitor financial management. In most countries, international con- programs, policy makers require quality and timely disag- sultants spent 40–80 days working on the PER, with a similar gregated data gathered by local jurisdictions. Often, special effort put in by local consultants. information is required to respond to the needs of the poor, who often do not appear as a disaggregated unit in con- We selected the following indicators for the quantitative sumer databases or even in survey and census information. comparison between countries:5 The first conclusion of this review is thus that more atten- tion needs to be paid to WSS information systems. Details on ■■ General trends in WSS public expenditure data limitations are included in annex 2. ■■ Sector budget allocation as a share of total general government budget6 ■■ Sector expenditure as a share of gross domestic product (GDP)7 ■■ Domestic/external sector allocations as shares of total sector public allocations ■■ Sector allocations transferred to subnational levels 5 Data sheets for all countries surveyed are included in annex 4. as share of total sector allocations 6 Government budget includes both internal and external resources. ■■ WSS spending per capita8 7 Expenditure as a share of GDP is calculated, whenever possible, on ■■ Access to WSS services the basis of total executed sector budget (both domestic and internal). ■■ Access to water (national, urban, and rural) 8 Per capita cost is calculated as the total executed budget, both do- ■■ Access to sanitation (national, urban, and rural) mestic and internal, divided by population. 4  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa 2.  Trends in Access to Water and Sanitation Services 2.1  Access to Water in Rural Areas tries (in most countries, more than half of rural dwellers do not have access to safe drinking water), Sahelian countries D espite the priority given by governments to water supply, have substantially better coverage than other African coun- access to drinking water in rural areas has increased only tries (Burkina Faso, 60 percent, 2005; Niger, 62 percent, 2007; marginally over the last decade. As shown in table 2.1, ru- and Mali, 50 percent, 2006). Not surprisingly, coverage in the ral water supply investment has just kept up with population fragile states is extremely low (Central African Republic, 32 growth, with the exception of Tanzania, which experienced a percent, 2008; Sierra Leone, 35 percent, 2008; Democratic decline in access to rural water from 2000 to 2007. Republic of Congo, 17 percent, 2008; Togo, 29 percent, 2007). Rural water supply coverage is higher than average in Sahelian The alarmingly high percentage of nonfunctional facilities is a countries and lower than average in postconflict states. Al- key obstacle to substantially boosting rural access to drinking though access to water in rural areas remains low in all coun- water. In most countries surveyed, the breakdown rate of ru- Table 2.1: Access to Drinking Water in Rural Areas—2000−08 (% of population) 2000 2001 2002 2003 2004 2005 2006 2007 2008 Average increase per year Burkina Faso    46.0    64.2   60.0        2.8 Cameroon     32.0 34.0 36.0 39.0 40.0 45.0   2.2 Central African     17.7 17.6 17.3 17.0 16.8 16.9 32.0 9.4 Republic Congo, Dem. Rep. of           12.0     17.0 1.3 Congo, Rep. of                 15.0   Côte d’Ivoire             65.0       Ghana 41.0 53.0 2.0 Madagascar 22.2 24.0 25.3 27.2 29.5 30.1       1.3 Mali   45.0 45.5 46.2 47.9 49.4 50.3     0.9 Mozambique 24.0           26.0   30.0 0.7 Niger     55.0 57.0 59.0 59.0 58.0 62.0   1.2 Sierra Leone                 35.2   Tanzania 56.0             42.0   −1.8 Togo               29.0     Source: PER Reports. Note: Coverage data for Ethiopia are not available. 5 ral water facilities is at least 25 percent.9 In postconflict coun- tries, such as the Democratic Republic of Congo and the Re- Box 2.2: Private Sector Participation in the public of Congo, breakdown rates exceed 50 percent. The Operation and Maintenance of Rural Water few available data indicate that the rate of nonfunctional fa- Schemes in Niger cilities had not decreased during the study period. In several Since 2006, private operators are providing monitoring services and support to the of the countries surveyed, substantial and systematic efforts O&M of rural piped schemes operated by individuals or private companies in Niger. have been made at the project level to mobilize beneficiary Currently, 79 of 761 facilities are covered by the private operators. Four regions are communities and to promote an understanding of the eco- covered to date. Basic O&M principles include decentralized ownership (communes), delegation to private operators, organization of user groups to defend their interests, nomic value of water and of adequate maintenance of fa- and cost-recovery through tariffs. Outcomes include a rise in the number of cilities. The country PERs observe, however, that the concept functional facilities to approximately 85 percent, an increase in the funds available of community-level management of water facilities has not for major repair works and extensions, enhanced transparency of management and consistently proven successful and faces severe problems operation, enhanced user satisfaction, and conflicts between user groups on one in ensuring adequate operation and maintenance (O&M) of hand and operators and communes on the other. facilities. In Ghana, an interesting approach to postconstruc- Source: www.reseaux-aep.org. tion support in community-managed rural water supply has apparently boosted the functionality of handpumps (box 2.1). But no one has assessed the cost efficiency of the setup. Several attempts have been made to privatize or professionalize 2.2  Access to Water in Urban Areas the operation and maintenance of rural water facilities. It has Urban water supply access has not been able to keep pace with proven difficult to achieve a financial viability for the sector urbanization. Access to water in urban areas has declined in when WSS facilities are geographically dispersed over a large the new millennium in all surveyed countries, confirming area. Lease contracts with private companies have not taken the general trend for the whole of Sub-Saharan Africa, which place on a large scale in the survey countries during the study experienced a 4.2 percent decline in the percentage of the period, although other countries, such as Senegal, have tested urban population with access to drinking water between such models on a large scale. Other approaches have shown 1990 and 2006 (see table 2.2).10 promising signs regarding ensuring functionality of rural wa- ter infrastructure (box 2.2). Likewise, in Mali a combination of The PERs in this sample do not include a comprehensive analysis community-based management and technical and financial of utility performance, but they do show that inefficient utilities postconstruction supervision by private advisers and audi- are a drain on state coffers as many utilities are either unable tors (STEFI concept) of rural piped schemes was introduced or barely able to cover O&M through their revenues. The oper- in 2000. This model has proven viable and contributed to ap- ating cost coverage (total annual operational revenues over proximately 90 percent functionality in 2009 (as compared to total annual operating costs) varies widely between utilities. 66 percent functionality for handpumps in 2006). In neighbor- Table 2.3 below shows selected performance indicators from ing Burkina Faso, where no such system has been introduced, those utilities in 11 of the reviewed countries that have re- the 2005 national survey showed that at least 33 percent of ported performance indicators to the International Bench- the rural piped schemes were nonfunctional. marking Network (IB-NET) for Water and Sanitation Utilities. For 4 out of 10 countries,11 the operating cost coverage is 9 Functionality is not systematically monitored on a regular basis in Box 2.1: Post-construction Support in Ghana any country. The estimates are thus based on isolated studies. In some cases, such as in Burkina Faso, functionality has been assessed more In Ghana, WSS committees have access to a multifaceted system for postconstruction comprehensively through water supply baseline studies in the context support that includes district WSS teams to provide training and support, a network of the preparation of a national water and sanitation program. Lock- of local private mechanics for maintenance and repair work, and a spare-part system, wood and Smits (2011) estimate that 20–40 percent of water points are including a central warehouse and three regional warehouses for the four standard not functional. A recent UNICEF study (cited in RWSN [2009]) showed models of handpump used in the country. A case study in two regions has shown a an average of 36 percent of handpumps in 20 African countries were low handpump breakdown rate (8 percent in Volta and 12 percent in Brong Ahafo). nonfunctional. Other studies and sources reveal similar levels. Source: Bakalian and Wakeman 2009. 10 Source: UNICEF-WHO Joint Monitoring Programme 2008. 11 No data are available for Ethiopia. 6  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa Table 2.2: Access to Drinking Water in Urban Areas, 2000–08 (% of population) 2000 2001 2002 2003 2004 2005 2006 2007 2008 Average annual change Central African Republic 28 Congo, Dem. Rep. of 60 37 38 −2.4 Congo, Rep. of 54 52 52 46 45 46 45 −1.3 Côte d’Ivoire 90 Mozambique 83 71 −1.7 Sierra Leone 81.7 Tanzania 90 80 −1.3 Togo 39 Source: PER Reports. less than one, which means that these utilities do not even past and have been illegally reconnected, with or without cover their operating costs (let alone investment costs). But implication of utility staff. In other countries (for example, Si- even these numbers are probably too positive, as they are erra Leone), the main problem seems to lie not so much with self-reported by utilities, and the worst utilities normally do billing as with the collection of billed amounts. In addressing not report, driving up the average. low financial viability and moving to solutions, it will be criti- cal to prioritize among the various causes of financial losses. The causes of low financial viability vary from one country to another. Some countries (for example, the Central African Re- In most countries, financial viability is compromised by high public and the Democratic Republic of Congo) have a large technical and commercial losses. The volume of produced wa- number of “inactive connections,” which is often a euphe- ter that is actually billed remains low. The IB-NET data show mism for connections that have been disconnected in the that on average 34 percent of produced water is nonrevenue Table 2.3: Key Performance Indicators for Utilities in Reviewed Countries Congo, Burkina Dem. Côte   Faso Rep. of d’Ivoire Ethiopia Ghana Madagascar Mali Mozambique Niger Tanzania Togo 2006 2005 2004 2006 2005 2005 2006 2007 2005 2007 2004 Mean # of utilities in sample 1 1 1 6 1 1 1 5 1 19 1 Nonrevenue water (%) 24 35 21 33 53 34 25 59 19 45 28 34 Staff/1,000 people served 0.3 n.a. 0.21 0.5 n.a. n.a. 0.3 1 n.a. 0.5 0.6 0.49 Continuity of service (hours/day) n.a. 11 24 22.7 11 n.a. 24 19.2 24 8.2 24 19 % sold that is metered (%) 100 n.a. 100 100 n.a. n.a. 100 44 100 100 90 92 Collection period (days) n.a. 1,834 7 87 n.a. n.a. n.a. 334 193 n.a. n.a. 491 Collection ratio (%) 105 n.a. 94 36 n.a. n.a. n.a. 85 n.a. 85 54 77 Operating cost coverage (ratio) 0.89 0.64 1.04 n.a. 1.13 1 1.88 0.82 1.3 1.04 0.69 1.04 Annual bill for households 59 35 32 13 43 14 43 50 39 26 60 38 consuming 6m3 of water/ month (US$/year) Source: The International Benchmarking Network for Water and Sanitation Utilities (IB-NET), www.ib-net.org. n.a. = not applicable. Trends in ACCESS to Water and Sanitation Services  |  7 water—the difference between system input volume and Table 2.4: Urban and Rural Access to Improved billed authorized consumption. This number should be used Sanitation in 2008 (%) with caution, however, because the volume of water sold is only a rough approximation in countries with no or low me- Urban Rural Total tering, and metering rates vary widely. Some countries have Burkina Faso 33 6 11 all connections metered, while in the Republic of Congo, at Central African Republic 43 28 34 the other extreme, only 2.7 percent of consumers are billed Côte d’Ivoire 36 11 23 on actual water consumption. Cameroon 56 35 47 Congo, Dem. Rep. of 23 23 23 Collection rates (the percentage of billed volume that is paid for) Congo, Rep. of 31 29 30 vary widely among countries. In some countries, public institu- Ethiopia 29 8 12 tions’ failure to pay for their water consumption remains a ma- Ghana 18 7 13 jor problem, especially because public institutions consume nearly half of the water billed. In some countries, this problem Madagascar 15 10 11 has been addressed, if perhaps temporarily, by the central- Mali 45 32 36 ized payment of the government’s bills, a conditionality for Mozambique 38 4 17 debt relief or donor budget support (for example, the Central Niger 34 4 9 African Republic and the Democratic Republic of Congo). Sierra Leone 24 6 13 Tanzania 32 21 24 In many countries, water tariffs have not been adjusted for years Togo 24 3 12 and do not cover production and distribution costs. Tariff ad- Total 31 14 19 justment for water is a very sensitive political issue, and gov- Source: JMP (www.wssinfo.org). ernments have proven reluctant to approve increases. For instance, tariffs have remained unchanged in the Republic of Congo since 1994, in the Central African Republic since 1998, and in Togo since 2001. In the surveyed countries, water rates for private consumers range from $0.10 to $0.46 2.3  Access to Sanitation per cubic meter. The PER findings confirm the AICD. Studies In 2008 only 20 percent of households in the reviewed coun- in African countries indicate that the average African water tries had access to improved sanitation according to house- tariff of about $0.67 per cubic meter is well below the full hold survey data collected by the World Health Organization cost of production and distribution of $1.00 per cubic meter (WHO) and United Nations Children’s Fund (UNICEF’s) Joint (Foster and Briceño-Garmendia 2009). Monitoring Program (JMP). Several PERs found access levels well below the figures reported by the JMP. An improved Many utilities are financially inefficient, due in part to overstaff- sanitation facility is one that hygienically separates human ing. High labor costs account for an important part of total excreta from human contact. Possible explanation for the operating expenses in the countries surveyed. According to disparity in numbers includes differences in definitions and one study, labor costs amount to an average of 29.3 percent in methodologies. of water-supply operating expenses in African low-income countries (LICs) (Banerjee and Morella. 2011). Some of the Not more than 10 percent of rural dwellers in the countries sur- countries in our sample, however, have significantly higher veyed have access to improved sanitation facilities, although costs; staff costs in the Democratic Republic of Congo and the few available data on access to improved sanitation in ru- the Republic of Congo, for instance, amount to almost 45 ral areas suggest that some progress has been achieved over percent of operational costs. Though the PERs had little data the last decade. From 2000 to 2008, the percentage of the on staff in professional categories, some figures suggest an total (rural and urban) population without access to an imbalance in staff composition; in the Democratic Republic improved latrine decreased (for example, in Mozambique of Congo, field staff represents only 25 percent of total staff, from 57 to 43 percent), but still remained very high, espe- compared with 42 percent and 62 percent in Mali and Côte cially in rural areas (for example, 96 percent in Mozambique d’Ivoire, respectively. and 89 percent in Burkina Faso). According to household 8  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa surveys in Tanzania, however, the percentage of rural peo- About one-third of urban households have access to an improved ple without access to an improved latrine increased from latrine or a septic tank. Sewerage systems are virtually nonex- 8.2 percent to 9.3 percent from 2000 to 2008. Some coun- istent, and in the few cities where they exist they serve only a tries have introduced a community-led total sanitation ap- small percentage of the population (Freetown, Sierra Leone) proach at a pilot scale (for example, Ghana, Sierra Leone, or are nonoperational (Douala, Cameroon). The limited data Cameroon, and the Central African Republic), but it is too available on access to improved sanitation in urban areas sug- early to measure results. gest that access has been stable over the past decade. Trends in ACCESS to Water and Sanitation Services  |  9 3.  Who is Spending What? 3.1  Expenditure as a Share of GDP Table 3.1: Expenditure as a Percentage of GDP Actual expenditure on the whole water supply and sanitation   2000 2001 2002 2003 2004 2005 2006 2007 2008 Mean (WSS) sector between 2000 to 2008 averaged 0.32 percent of the Rural sector only gross domestic product (GDP), while expenditure on rural WSS Burkina Faso 0.42 0.14 0.19 0.15 0.26 0.49 0.28 in particular was 0.26 percent of GDP. Table 3.1 identifies the eight countries where surveys covered only rural WSS and Cameroon 0.16 0.09 0.07 0.11 0.05 0.11 0.10 the other seven where both rural and urban WSS were cov- Côte d’Ivoire 0.06 0.02 0.02 0.01 0.01 0.03 0.03 ered. The mean (all countries, all years) was 0.26 percent, with Ethiopia 0.26 0.26 a range from nearly zero expenditure as a share of GDP (Côte Ghana 0.15 0.38 0.36 0.17 0.41 0.48 0.33 d’Ivoire, rural sector, 2005–06) to 1.16 percent (Tanzania, rural Madagascar 0.08 0.13 0.06 0.09 0.12 0.26 0.12 and urban, 2005). Actual spending totals are only one-tenth Mali 0.38 0.27 0.18 0.30 0.53 0.60 0.38 of the estimated 2.58 percent of GDP required each year by Niger 0.28 0.36 0.75 0.62 0.40 0.97 0.57 Sub-Saharan African states to meet the water Millennium Rural mean 0.08 0.27 0.19 0.18 0.23 0.32 0.34 0.37 0.26 0.26 Development Goals (MDGs) (Briceño-Garmendia, Smits, and Rural and urban Foster 2008). It is also well below the suggestion by the 2006 Central 0.02 0.06 0.11 0.04 0.14 0.18 0.44 0.14 Human Development Report that all countries should spend African at least 1 percent of their GDP on WSS.12 This assumes, how- Republic ever, that greater funding could be spent effectively. Low Congo, 0.01 0.01 0.20 0.54 0.48 0.69 0.51 0.35 rates of spending out of current budget releases indicate Dem. that more budget releases might not increase spending. The Rep. of focus should then be on capacity building to increase “ab- Congo, 0.12 0.16 0.16 0.09 0.42 0.19 0.16 0.19 sorptive capacity.” Rep. of Mozambique 0.30 0.51 0.88 0.67 1.24 0.82 1.21 1.20 0.85 WSS expenditure has grown over the study period. Annual ex- Sierra Leone 0.18 0.15 0.30 0.50 0.97 0.53 0.38 0.43 penditure is extremely volatile, both absolutely and as a Tanzania 0.35 0.29 0.58 0.54 1.16 0.95 0.64 share of GDP and per capita (see below). But there is a gen- Togo 0.06 0.12 0.17 0.16 0.42 0.05 0.12 0.16 eral upward trend in the sample countries. Over the period Rural and 0.65 0.24 0.39 0.43 0.93 0.84 0.71 0.70 0.39 2002 to 2006, to which most of the data relate, rural WSS urban mean expenditure in seven countries grew from 0.18 percent of Overall 0.08 0.35 0.21 0.27 0.31 0.49 0.57 0.49 0.77 0.32 GDP to 0.24 percent, and total WSS expenditure in four other mean countries grew from 0.15 percent of GDP to 0.64 percent. Source: Country WSS PER papers, World Bank World Development Indicators, and authors calculations. All As a share of GDP, it appears that rural WSS spending grew means are arithmetic unweighted averages. much more slowly than urban sector spending, though the sample is too small to generalize. 12 The 2006 Human Development Report recommends that “countries should spend 1% on CAPEX plus 1% should come from cost-recovery Public expenditure on WSS as a share of GDP depends heavily on and community contributions providing an equivalent amount” (UNDP the total public expenditure as a share of GDP, and thus on the 2006: 65). 11 tax revenue ratio. For instance, in the Central African Repub- 3.2  Expenditure Per Capita lic, tax revenue was only 8.7 percent of GDP in 2009. Even if the government allocated a considerable share of that to the The expenditure per capita shows a positive trend over time. WSS sector, the ratio to GDP would remain low. Among the countries surveyed, average annual per capita expenditure was $1.21 for rural-only WSS, and $1.71 in coun- The different findings of the public expenditure review (PER) tries whose PER covered both rural and urban sectors. Ex- overview paper and the Africa Infrastructure Country Diagnos- penditure in the latter countries ranged from $0.01 per per- tic (AICD) can be explained by their different scopes. While the son (the Democratic Republic of Congo in 2002) to $8.93 per PER data exclude off-budget spending by water utilities, the person (the Republic of Congo in 2006). AICD study included utility spending (see box 3.1). Sector Budget as a Share of 3.3  Government Budget Box 3.1: Comparing the PER Findings with the Africa Infrastructure Country Diagnostic WSS budgets are 2.0 percent of the total government budgets (AICD) of the sample countries. The WSS budgets range widely, from The findings of the PERs are in line with the findings of the AICD taking into account less than 0.1 percent of the total government budget of that the PERs focused on budgetary spending (excluding off-budget spending Cameroon for most of the period covered to 6.5 percent in by water utilities), while the AICD study included utility spending. As a result of Tanzania in 2008. A relatively high share of total budget is this wider scope, one would expect the AICD report to show considerably higher allocated to water in the Sahelian countries of Burkina Faso domestically funded expenditure on recurrent costs for urban water supply. This is and Niger. The average share fell from 2001 to 2004 but has indeed the case. The following differences between the two reports are caused by the differences in scope: risen since 2004. The overview paper found that less than 0.4 percent of GDP was spent on WSS in “rural and urban” countries for only 0.3 percent in “rural only.”This is substantially lower than the 1.2 percent found in the AICD report. 3.4  Sector Funding Sources According to the AICD report, domestically funded spending in the WSS accounted The WSS sector is highly donor dependent, with donors covering for over half of total spending. This contrasts with the findings of the regional overview paper that show a clear predominance of donor funding in the sector more than 60 percent of total sector expenditures in the coun- spending. But the AICD and the overview paper both found a predominant role of tries reviewed. Donor funds are formally targeted at capital donor funding in investment expenditures. expenditure but also include an unknown amount of re- While the regional overview paper concluded on a slight bias of public expenditures current expenditure. Individual countries experienced wide toward urban areas, the AICD report found a more pronounced bias. fluctuations in donor funding, in some cases due to politi- The recurrent expenditures reported in the AICD were three times higher than that in cal and social unrest, such as in Côte d’Ivoire, the Republic the regional overview paper. of Congo, and the Democratic Republic of Congo. In total, donors provided a fairly stable share of funding from year to year. As shown in table 3.4, the sector’s dependency on PER external funding varies from around 20 percent to more than overview AICD 80 percent of total expenditure. paper report Total spending in WSS (as % of GDP) 0.4 1.2 Capital budget execution rate (%) 62.0 75.0 3.5  Trends over Time Share of donor funding (as % of total sector funding) 61.9 53.9 Although WSS budgets have increased relatively quickly in most Recurrent expenditures (as % of total sector 12.7 40.0 countries, public expenditure still falls considerably short of gov- expenditure) ernment commitments. The review showed that annual WSS Investment in sanitation (as % of GDP) — 0.5 expenditure falls short of international and national targets. Share of spending by subnational governments 6 — WSS expenditure is well below what is needed to meet Sub- (% of total) Saharan MDG targets. Source: AICD reports and PER reports. 12  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa Table 3.2: Expenditure Per Capita (US$)   2000 2001 2002 2003 2004 2005 2006 2007 2008 Mean Rural sector only Burkina Faso 1.30 0.49 0.86 0.82 1.48 2.96 1.32 Cameroon 2.15 1.58 1.32 2.31 1.12 2.73 1.87 Côte d’Ivoire 0.68 0.31 0.31 0.09 0.18 0.53 0.35 Ethiopia 0.97 0.97 Ghana 0.57 1.63 1.34 0.85 1.31 1.14 Madagascar 0.26 0.47 0.21 0.38 0.40 0.98 0.45 Mali 1.17 1.03 0.87 1.62 2.97 3.51 1.86 Niger 0.62 0.95 2.12 1.93 1.31 3.57 1.75 Rural mean 0.26 0.98 0.82 0.94 1.13 1.52 1.73 2.28 0.97 1.21 Rural+urban Central African Republic 0.92 0.92 0.92 0.92 0.92 0.92 0.92 0.92 Congo, Dem. Rep. of 0.01 0.02 0.25 0.73 0.76 1.20 1.04 0.57 Congo, Rep. of 1.13 1.66 2.25 1.54 8.93 3.94 4.54 3.43 Mozambique 1.09 2.13 1.96 4.06 2.93 4.43 5.31 3.13 Sierra Leone 0.38 0.32 0.65 1.20 2.65 1.62 1.32 1.16 Tanzania 0.69 0.30 1.02 2.97 2.22 3.73 4.32 2.18 Togo 0.17 0.39 0.61 0.61 1.44 0.18 0.58 0.57 Rural+urban mean 0.63 0.82 1.09 1.72 2.83 2.29 2.58 1.71 Overall mean 0.26 0.98 0.72 0.88 1.11 1.62 2.33 2.29 2.37 1.44 Sources: Country WSS PER papers, World Bank World Development Indicators, and authors calculations. All means are arithmetic unweighted averages. Table 3.3: WSS Budget as a Share of Overall Government Budget (%)   2000 2001 2002 2003 2004 2005 2006 2007 2008 Mean Rural sector only Burkina Faso 2.9 2.5 1.7 1.1 1.8 2.9 3.1 2.3 Cameroun 1.3 0.7 0.4 0.7 0.3 0.5 0.6 Côte d’Ivoire 0.5 0.4 0.3 0.5 0.1 0.1 0.4 0.3 Ethiopia Ghana 0.9 2.5 2.3 1.1 2.1 2.6 1.9 Madagascar 0.6 1.2 0.7 1.6 1.3 1.3 1.4 1.2 Mali 1.9 1.4 1.4 1.6 2.3 2.1 1.8 Niger 4.2 3.2 5.4 4.3 Rural mean 0.6 1.7 1.5 1.3 1.0 1.8 1.8 2.3 0.4 1.8 Rural+urban Central African Republic 0.2 0.4 0.7 0.3 1.2 1.0 4.3 1.2 Congo, Rep. of 0.6 0.1 0.3 0.5 0.9 1.4 2.3 0.9 Congo, Dem. Rep. of Mozambique 2.6 1.6 2.1 3.6 3.5 2.3 3.5 2.7 Sierra Leone Tanzania 2.4 4.0 6.5 4.3 Togo 3.3 2.8 4.9 1.2 1.9 0.8 1.0 2.3 Rural+urban mean 1.7 1.3 2.0 1.4 2.0 1.9 3.5 2.3 Overall mean 0.6 1.7 1.6 1.3 1.4 1.7 1.9 2.1 3.0 2.0 Sources: Country PERs and authors calculations. All means are arithmetic unweighted averages Who is Spending What?  |  13 Table 3.4: Externally Funded Expenditure as a Share of Total WSS Expenditure (%) 2000 2001 2002 2003 2004 2005 2006 2007 2008 Mean Rural sector only Burkina Faso 78.3 78.2 77.6 77.8 72.4 74.6 76.5 Cameroun 22.0 39.8 48.5 50.9 0.0 11.5 28.8 Côte d’Ivoire 52.1 96.8 96.4 83.9 36.9 18.6 64.1 Madagascar 65.4 24.2 39 39 47.5 64.1 46.5 Mali 84.4 86.1 72.9 81.5 89.6 83.7 83.0 Niger 87 79 76.1 81.6 65.2 72.9 77.0 Rural mean 65.4 62.3 60.7 67.5 71.3 73.8 52.1 34.3 60.9 Rural+urban Central African Republic 62.5 86.4 51.2 86.6 85.5 95.4 77.9 Congo, Dem. Rep. of 5.5 15.6 63.1 67.0 93.0 95.8 95.0 62.1 Congo, Rep. of 64.4 75.1 4.3 9.3 1.5 4.6 2.0 23.0 Sierra Leone 54.2 89.2 93.2 93.1 84.8 82.9 Togo 44.6 77.1 79.5 73.6 40.0 66.8 63.6 Rural+urban mean 38.2 57.6 57.5 58.1 62.9 69.8 68.8 59.0 Overall mean (unweighted) 65.4 62.3 53.2 63.54 65.03 66.62 57.47 54.57 68.8 61.9 Sources: Country PERs and authors calculations. Economic growth often caused budget increases over the review Economic growth and an increasing tax base are strong deter- period in countries that use incremental budget procedures. In minants of the level of public WSS expenditure and thus are key some countries increases in budget have happened without to making progress in WSS access. In cross-country compari- a dramatic decline in the budget expenditure ratio, meaning sons, richer countries spend more on WSS on a per capita that the countries were able to absorb a rapid increase in basis, although the higher spending comprises a slightly their budgets. smaller share of overall GDP. Public WSS expenditure as a share of GDP depends heavily on total public expenditure as a share of GDP, and thus on the tax revenue ratio. 14  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa 4.  How is Money Being Spent? 4.1  Capital and Recurrent Expenditure ral-only water supply and sanitation (WSS) sectors, average recurrent expenditures were only 6 percent of total sector In the sample countries, 12.7 percent of total sector expenditure expenditures, as compared to 20 percent in countries with went toward recurrent expenditure, with 87.3 percent going both rural and urban sectors. toward development expenditure (see table 4.1). Shares dedi- cated to recurrent expenditure varied among the countries Low levels of nonsalary recurrent expenditures severely limit considered, from 1.2 percent in Côte d’Ivoire (2002) to 94.5 public sector institutions’ ability to carry out their mandated percent in the Democratic Republic of Congo (2002). In ru- roles. This may be one reason why monitoring of sector prog- Table 4.1: WSS Recurrent Expenditure as a Percentage of Total WSS Expenditure   2000 2001 2002 2003 2004 2005 2006 2007 2008 Mean Rural sector only Burkina Faso 1.9 4.7 4.8 3.6 2.4 1.3 3.1 Cameroon 5.1 8.3 9.8 4.4 12.6 4.4 7.4 Côte d’Ivoire 1.2 2.2 3.5 12.1 5.6 2.5 4.5 Ethiopia Ghana 10.6 13.3 5.0 6.3 13.8 10.9 10.0 Madagascar 9.1 5.2 12.1 6.5 3.8 1.5 6.4 Mali 4.9 6.6 10.0 6.7 3.8 3.5 5.9 Niger 13.0 10.2 5.9 6.4 11.5 4.3 8.6 Rural mean 9.1 5.7 8.0 6.7 5.6 6.3 7.6 3.7 6.6 Rural+urban Central African 48.6 21.4 9.7 28.5 7.7 5.3 2.3 17.6 Republic Congo, Dem. Rep. of 94.5 84.4 5.9 32.9 6.0 1.2 1.3 32.3 Mozambique 56.0 33.6 25.1 17.9 26.8 19.4 13.7 27.5 Congo, Rep. of 4.4 4.1 3.4 5.6 1.7 3.8 7.5 4.4 Sierra Leone 0.0 0.0 0.0 10.8 5.1 6.4 12.3 4.9 Tanzania 13.6 20.8 26.0 89.4 18.4 23.6 22.4 17.5 16.7 27.6 Togo 25.1 9.2 10.9 10.7 6.5 48.8 13.6 17.8 Rural+urban mean 13.6 20.8 42.4 48.4 14.7 26.0 15.2 20.5 9.6 23.5 All countries 11.3 10.8 22.2 20.7 8.1 12.5 9.4 11.4 9.6 12.7 Source: Country PERs and authors calculations. All means are arithmetic unweighted averages. 15 ress and support to water users and local governments have marketing, for instance, is labor intensive), and varying de- been extremely weak in all countries surveyed. These figures grees of political intervention. The data are insufficiently de- should be interpreted with caution for two reasons. In many tailed and robust to make an analysis. cases, high recurrent expenditure levels reflect low external investment funding, particularly in politically and socially There is no norm or international standard for the salary share unstable countries. Also, external expenditures categorized of recurrent WSS expenditures, but the salary share found in as capital expenditures often include recurrent funding. The the public expenditure reviews (PERs) appears high relative volume of this “hidden” contribution to recurrent costs is im- to nonsalary charges, that is, operation and maintenance possible to determine, but is likely to be quite substantial, (O&M) expenditures are too low. Given the insufficiency of relative to recurrent budget allocations inscribed in the na- O&M expenditures evidenced above, a lower share of salaries tional budgets. In any case, this external contribution to the in recurrent expenditure would release more funds for keep- recurrent budget is volatile by nature, as it is dependent on ing WSS facilities functional, thus raising access rates and the lifetime of a specific project. Millennium Development Goal (MDG) achievements. Table 4.2 shows that the average share of recurrent expenditures dedicated to salaries has gone down from 66.7 percent in 4.2  Salary Component 2002 to 54.8 percent in 2007, but the sample is too small and the data too unreliable to conclude that this is a real trend. On average, half of recurrent WSS spending goes toward sala- ries.13 A lack of donor data means that we do not know the comparable share in development spending, but it is prob- 4.3 Targeting ably low, as most development partners no longer allow for salary expenditures. Table 4.2 shows considerable variation 4.3.1  The Rural and Urban Split among countries, ranging from 4.2 percent in the Republic of Congo (2008) to 98.8 percent in the Central African Re- Overall, public expenditures exhibit a slight urban bias. The bias public (2005). This may be due to a combination of varying might be (partially) justified because of the differential unit accounting arrangements (for example, WSS personnel not charged to WSS codes, non-WSS personnel charged to WSS 13 “Salaries” includes wages, allowances, and social security contribu- codes), varying activity and technology mixes (rural social tions. Table 4.2: Salaries as a Share of Recurrent Expenditure (%)   2000 2001 2002 2003 2004 2005 2006 2007 2008 Mean Rural only Burkina Faso 73.1 86.8 64 81.2 86.9 72.3 77.4 Cameroon 66.6 65.2 57.7 69.5 63.6 74.4 66.2 Côte d’Ivoire 38.3 37.8 39.4 43.7 46.4 39.3 40.8 Madagascar 16.8 18.9 18.7 21.4 25.3 25.9 21.2 Mali 79.3 76.4 77.9 79.8 77.2 73 77.3 Niger 82.5 83.1 69.9 72.6 70.6 73 75.3 Rural mean 16.8 57.1 61.6 58.2 58.9 62.6 65.2 62.2 55.3 Rural + Urban Central African Republic 97.8 98.7 98.2 98.8 98.5 98.5 88.2 97.0 Congo, Rep. of 10.1 5.5 7.2 4.2 6.7 Tanzania 31.9 36.6 32.4 33.7 Rural+urban mean 97.8 98.7 98.2 54.45 45.3 47.43 41.6 69.07 Mean (unweighted) 16.8 57.1 66.7 64.01 64.5 60.6 57.7 54.8 41.6 53.8 Source: Country WSS PER papers and authors calculations. 16  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa costs in urban and rural areas. Taking into account the lim- and wastewater treatment (normally financed from public ited number of PERs including both rural and urban water budgets) are virtually nonexistent in many of the countries. supply and the difficulties encountered in disaggregating in- As countries get richer, however, and people move up the vestment data along rural/urban lines, it is difficult to quantify sanitation ladder, huge public investments will be required. the distribution of investment funds to rural and urban areas respectively. Yet, the PER studies suggest that the percent- 4.3.3  Geographical Disparities between Countries age of public expenditure that goes to urban areas is higher than the share of the population that lives there. For instance, Richer countries spend more on WSS per capita but slightly less in Togo, half of total public sector expenditure has been di- as a percentage of gross domestic product (GDP) (figure 4.1). rected toward urban areas in which only 41 percent of the Richer countries can afford to spend more money per capita population resides. In the Republic of Congo, more than 80 to provide their citizens with basic services such as WSS. As percent of actual expenditure between 2002 and 2008 was countries grow richer, however, WSS expenditures do not directed to urban areas that housed 72 percent of the total keep pace with overall GDP growth, resulting in richer coun- population. That said, since 2006, the Republic of Congo has tries spending a relatively lower percentage of their GDP in also seen a substantial increase in funds allocated to rural wa- the WSS sector. ter. In Mozambique the lion’s share of funding goes to urban water, though two-thirds of the population live in rural areas The Sahelian countries have substantially better rural water and get just 12 percent of the total funding. In Sierra Leone, supply coverage than other African countries and a higher however, the relationship between expenditures and the ru- percentage of sector expenditure as a share of GDP. ral/urban population split seemed to be more or less compa- rable: rural water supply accounted for up to 60 percent of Donor financing is not targeted for the greatest impact on WSS; public expenditure in the review period (nearly all of it from per capita donor expenditures are not based on a country’s level donors), and an estimated 62 percent of the population lives of economic development and sectoral need. Instead, external in rural areas. The fact that the average per capita expenditure funding levels depend on other factors, such as political sta- for rural-only PERs ($1.21) is lower than for PERs that cover bility, adherence to good governance principles, and a sound both urban and rural ($1.71) confirms this bias. public financial management reform process. The country with the highest per capita donor contribution to the WSS The bias toward urban areas seems to be partly caused by in- sector is Mozambique, while Madagascar, Togo, Cote d’Ivoire complete decentralization in rural areas, which has led to insti- and the Republic of Congo receive the lowest per capita tutional paralysis. But where this paralysis is resolved, rural expenditure can go up. For instance, in Tanzania, the decen- tralization and devolution process resulted in a significant in- Figure 4.1: Expenditure as a Percentage of GDP crease in the share of the investment budget directed toward and on a Per Capita Basis rural water, whereas allocations for urban water supply have increased far less over the same period (see box 5.4 below). 1.00% $4 0.75% $3 4.3.2  Water Supply versus Sanitation Split 0.50% $2 In most of the countries surveyed, sanitation receives only a small part of public WSS expenditure. This comes as no surprise, as 0.25% $1 the sanitation sector is dominated by household on-site fa- 0.00% $0 cilities (latrines, septic tanks) that are generally financed from 0 250 500 750 1000 1250 1500 1750 household budgets. The amount of expenditure on hygiene GDP per Capita in USD education is particularly hard to assess, as the responsibility is often diffused across various ministries (water, health, ru- as % of GDP - left axis in USD/capita - right axis ral development, and so on). We observed a trend toward Linear (as % of GDP - left axis) Linear (in USD/capita - right axis) more specific sanitation-targeted donor projects in the latter half of the review period, including the introduction of com- munity-led total sanitation in several countries. Sewerage Source: Country WSS PER papers and authors calculations. How is Money Being Spent?  |  17 contributions (figure 4.2). This finding is in line with the Or- In general, capital cities have considerably higher rates of access ganisation for Economic Co-operation and Development/ to water supply than secondary cities. Many of the national utili- Development Assistance Committee (OECD/DAC) reports ties in the reviewed countries do not pay attention to the sec- showing that donor assistance to the water sector does not ondary cities in which they are supposed to operate systems. very well target countries with low levels of WSS access. For in- As a result, schemes outside the capital cities are underfund- stance, an OECD report found that the least developed coun- ed, understaffed, and sometimes without functioning facili- tries only received 25 percent of total aid for water and that ties altogether (for example, the Central African Republic and numerous countries with low levels of WSS access received Democratic Republic of Congo). In Sierra Leone, the utility very little during the period 2001 to 2006 (OEDC, 2008). The for the capital Freetown (GVWC) has a 38 percent household share of global donor aid in the water sector benefiting Sub- connection rate, while the utility for secondary cities (SALWA- Saharan Africa actually decreased over that period. CO) connects only 6 percent of the households in its cities. Sector investment planning and allocation policies have con- 4.3.4  Geographical Disparities within Countries tributed to disparities in access and distribution of WSS services. Within the countries surveyed, there are large regional dispari- Even though these disparities might to some extent reflect ties in rural access to water. Although the PERs do not provide natural constraints to developing the service level (scarcity data to assess whether disparities in rural water coverage of water resources, scattered settlements), there is clearly diminished during the study period, gaps between the best- scope for improving equity and transparency in planning served and less-served regions (the “access gap”) remained and allocating resources for the development of water ser- substantial in 2007−08. In Mozambique the access gap be- vices. In some cases, regions might receive lower levels of tween regions with the highest and lowest levels of access public expenditure for political reasons. But other times, to water supply was 60 percent. The gap was 28 percent in regions with low access to WSS are often difficult to target Ethiopia, 38 percent in Burkina Faso, 29 percent in Madagas- because of distance from the capital city or lack of imple- car, and 26 percent in Ghana. mentation capacity. Access to rural areas in the North Kivu province and, to some extent, the Katanga province in the Democratic Republic of Congo, for instance, has been re- stricted for many years due to insecurity; as a result, WSS Figure 4.2: Donor Expenditure Per Capita (US$) delivery has been negligible. One major challenge for these Ordered by Country GDP/capita (US$) regions’ reconstruction, which started in 2006, is the scarcity 1.00% $4 of both private and public sector capacity to plan and imple- ment sector activities. 0.75% $3 0.50% $2 There is a tension between targeting the needy and rewarding good performance. Targeting public expenditure to needy 0.25% $1 populations is difficult. For instance, Tanzania has developed a formula for allocating sector funds to local government au- 0.00% $0 0 250 500 750 1000 1250 1500 1750 thorities to ensure equity and allocation of funds to the most needy areas. But calculation of the allocation formula is hin- GDP per Capita in USD dered by the quality of underlying data (poverty data and cov- as % of GDP - left axis in USD/capita - right axis erage data), and the formula is not consistently implemented, Linear (as % of GDP - left axis) Linear (in USD/capita - right axis) as some regions get significantly more funds than the water formula would allow. Burkina Faso’s annual sector planning exercise is based on similar principles, but regional allocations Source: Country WSS PER papers and authors calculations. are not strictly respected given the geographic restrictions on virtually all external funding during the study period. 18  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa 5.  Budget Execution Rates 5.1  Trends in Budget Execution ing an average WSS budget execution rate of 75 percent (Ba- W nerjee, Sudeshna, and Morella 2011). The trend over time is ater supply and sanitation (WSS) budgets in the sample flat in the sample countries (see table 5.1; figure 5.1). countries were 63 percent executed over the years cov- ered. Recurrent budgets were 70 percent executed, Averaged across the eight countries for which data are avail- and development budgets were 62 percent executed (see able, domestically funded sector expenditures were 66 percent annex 3). Execution rates were slightly higher in rural-only executed, while externally funded expenditures were 57 percent countries (66 percent) than in rural-plus-urban countries executed. Data deficiencies should be kept in mind, particu- (47 percent). This is probably due merely to the selection of larly on the development budget. Many governments are countries, however, and we cannot infer that execution is struggling to capture all donor disbursements in their ac- higher in the rural sector generally. The WSS budget execu- counts; apparently low execution may be due to a failure to tion rates found in our sample of countries compares to the record some of the donor project expenditures. World Bank’s recent study on Sub-Saharan countries, show- Table 5.1: Budget Execution Rates for WSS Budgets (%)   2000 2001 2002 2003 2004 2005 2006 2007 2008 Mean Rural sector only Burkina Faso 64.1 24.2 49.1 54.5 53.3 62.5 51.3 Mali 67.0 65.3 44.1 63.3 74.8 63.5 63.0 Niger 81.4 30.5 55.7 71.7 63.3 72.9 62.6 Cameroon 98.5 99.3 100.0 100.0 100.0 93.5 98.6 Côte d’Ivoire 42.4 25.2 31.2 5.8 46.0 119.9 45.1 Ghana 145.4 49.0 123.4 185.3 42.7 47.0 98.8 Madagascar 58.9 43.1 31.9 26.6 45.7 83.6 48.3 Mean rural only 58.9 79.9 56.1 56.9 76.5 61.7 63.7 95.4 68.6 Rural+urban Central African Republic 51.6 78.5 99.9 76.0 65.2 113.2 59.5 77.7 Sierra Leone 20.6 54.4 62.2 12.8 39.8 19.9 35.8 35.1 Tanzania 82.9 88.5 18.8 63.4 86.0 73.4 85.4 55.4 69.2 Togo 9.7 22.9 17.6 71.1 101.9 28.5 54.9 43.8 Mean rural+urban 82.9 42.6 43.7 60.8 61.5 70.1 61.7 51.4 59.3 Total (overall mean) 58.9 80.5 51.2 52.1 70.8 61.6 66.3 76.2 51.4 63.2 Source: Country WSS PER papers and authors calculations. 19 Figure 5.1: WSS Budget Execution Rates 2000–08 (%) Box 5.1: Addressing Systemic and Sector- Specific Problems in the Budget Execution 90 Chain in Benin 80 In 2001 Benin began to transition from sector projects to a programmatic approach 70 for the WSS sector. Public expenditure management reforms to overcome 60 systemic problems in the budget execution chain included a transition from a 50 line-item-based budget to a program-based budget. Sector-level analytical and advisory work—including sector PERs—helped sectors to develop programs 40 with supporting program-based budgets, laying the groundwork for shifting 30 spending authority from the ministry of finance to line ministries and progressive 20 decentralization of service delivery. 10 Analytical sector work, including a water PER, facilitated the inclusion of rural water 0 supply in a series of Poverty Reduction Support Credits (PRSC) aligned with budget 2001 2002 2003 2004 2005 2006 2007 2008 cycles. The PRSCs helped the government make considerable progress in addressing structural public sector management issues, increasing public sector implementation Total 80.5 50.0 50.5 66.6 62.0 68.2 80.8 56.6 capacity, and increasing execution rates. Program budgets are now produced Recurrent 80.9 65.3 60.4 71.5 68.0 77.2 79.4 56.0 for most sectors, and details are integrated into the annual budget submitted to Development 67.8 44.3 55.0 64.5 64.5 71.9 74.8 53.3 Parliament. The Ministry of Finance’s budget execution software has been extended to all line ministries and includes comprehensive coverage of donor-financed expenditures. Ministries monitor their programs and produce annual performance reports. This in turn facilitates performance-based contracts between the minister of Source: Country WSS PER papers and authors calculations. finance and line ministers responsible for subprograms. The 2006 and 2007 budgets were prepared in terms of program authorizations and payment appropriations, which enabled multiyear contracts. Ministries regularly Systemic Problems Causing Low 5.2  produce sector performance reports. The increased execution rates as well as better Rates of Budget Execution transparency in public financing convinced several donors to increase their financing to Benin. Low rates of budget execution are due to systemic and sectoral Benin has shown impressive progress and is on track to meet its rural water supply obstacles all along the execution chain. It has been hard to dis- Millennium Development Objective (MDG). Between 2001 and 2008, physical sector output—as measured in the number of water points planned and constructed per tinguish which obstacle is the most critical in most countries. year—has increased more than fourfold. Most governments have made significant improvements to Source: Benin PER and AMCOW 2011. their public financial management systems over the past few years. But many of these improvements have not yet trick- led through to line ministries. The general message is that improvements are needed all along the budget execution chain, and the public expenditure reviews (PERs) can tell us low rate of execution of WSS budgets is partially attributed what needs to be done. For instance, Benin’s rural WSS PER to weaknesses in budget preparation, especially externally was part of a larger suite of analysis that enabled Benin to funded projects. overcome systemic and sectoral problems and, ultimately, improve rural WSS access (see box 5.1). Several PERs noted that line ministries are not adequately in- volved in the budgeting process. When the ministries receive the budget guidelines, the guidelines often already contain Limited Communication during the 5.2.1   an indication of budget envelopes, a situation that prevents Budgeting Process the ministries from communicating their true needs. As a re- Low execution is caused partly by overambitious plans and sult, line ministries are not interested in budget preparation budgets. In Sierra Leone, for example, some projects may work, which they view as an exercise involving the ministry enter the budget without prior government appraisal, while of finance only. others enter from outside of the budget process, and many enter without sufficient scrutiny and debate by the Cabinet Spending limitations may also be due to cash constraints where or by Parliament. There is less scrutiny of the development the ministry of finance is unable to release the full amounts bud- budget than the recurrent budget. Similarly, in Ghana, the geted. There are no data comparing spending releases with 20  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa actual expenditure, but some country evidence indicates and amounts of resources undermines plans and budgets that the operating constraint is capacity to spend rather than and reduces the efficiency of parastatals, local authorities, availability of funds. In Tanzania, for instance, steep increases and central ministerial departments and agencies in the in budget allocations to the water sector, following its identi- WSS sector. The planning horizon is foreshortened, projects fication as a priority, were not matched by corresponding in- are scrambled, contracts are split to keep within cash limits, creases in spending (nor in outcomes). Spending increased and economies of scale are lost. Costs rise with delays, infla- at a much slower pace (Foster and Cecilia Briceño-Garmen- tion, and stop-start resource management. Where controls dia, 2009). don’t work, arrears build up and suppliers have to wait for payment, obtain additional working capital, or reducing the Donor support is still mainly provided on a project basis. The scale of their operations. They raise their prices and become development of national policies and strategic frameworks less willing to bid on government contracts. for WSS delivery has paved the way for a programmatic sec- tor approach based on a single national WSS vision and a Predictability of funds depends on timely budget approval, modus operandi common to all funding sources. In the sur- stable cash flow budgets, and administrative efficiency. Proce- vey period, however, the WSS sector in most countries was dures for the release of cash (or spending authority where still characterized by a large number of donors operating payments are centralized) tend to get longer and more on terms and conditions specific to individual projects re- cumbersome as “gatekeepers” continue to be added in the lated to issues like geographic intervention zones, routing name of control. In Sierra Leone spending authority to water of funds, procurement, monitoring, and project manage- parastatals and ministerial departments and agencies is sup- ment. Transaction costs have thus been quite important in posed to be granted quarterly, but is sometimes piecemeal all countries surveyed. or late. Transfers to local councils have also been unpredict- able. In 2006 hundreds of signatures were required to dis- The Medium-Term Expenditure Framework (MTEF) planning tool burse each quarterly transfer to each of 19 local councils. The has not necessarily led to a greater coherence between resource procedure has since been radically streamlined. allocation and overall sector targets set out by governments. This is likely due to difficulties achieving a macroeconomic Donor funding is even less predictable. Budget execution rates balance as well as to the unpredictability of external funding are lowest and most erratic for donor-funded development (largely off budget) and weak planning and budgeting ca- expenditures. This is partly due to poor information. Donor pacity in technical ministries. Ministries of finance are under- planning and monitoring is not necessarily linked to the standably reluctant to allocate more resources to ministries government budget calendar, while parallel donor systems that have difficulty consuming already allocated funds. make it difficult to obtain full information on aid flows. Dis- bursement applications typically go through a dozen sig- The current budget structure, especially in francophone coun- natures and can take over a month to get approved. If the tries, limits the potential to link budgets to objectives and out- government counterpart contribution is delayed, donor comes set out in longer-term sector plans. In most countries, disbursements are further delayed. This has a negative mul- the national budget is still based on a “classical” budget—that tiplier effect on progress—although the counterpart contri- is, expenditure monitoring on the basis of budgetary clas- butions are generally only a small part of project financing, sification (salaries, operational costs, and investment costs). they can delay the bulk of the financing from donors as well. This weakens the relevance of budget programs based on longer-term sector development plans, as “classical” budgets Some countries, such as Ethiopia and Sierra Leone, have a do- are poorly suited to planning and monitoring objectives and nor-funded basic services delivery program to augments central results for growth and poverty reduction set out in the over- transfers to local authorities for the recurrent costs of water and all national strategy. other basic services. The program uses national allocation cri- teria and administrative arrangements, but with total trans- parency and predictability of amounts and timing of releases. 5.2.2  Volatility of Budgets The impact of the program has not yet been evaluated. Do- The volatility of WSS budgets forms a major obstacle for efficient nors may channel their funds directly to regional and district budget execution because the unspent budget of one year typi- authorities, as in Ethiopia and Sierra Leone to avoid delays cally cannot be carried over into the next year. Uncertain timing while funds are cascaded down from one special account Budget Execution Rates  |  21 to another. This reduces the risk of diversion of funds from Long delays in the approval of contracts add to costs and delay their intended purposes. But it is difficult to ensure that cen- benefits. In some countries, such as Mozambique, approval of tral governments keep informed on progress on the ground, contracts even involves an external audit body. Slow procure- and standards of accounting and reporting at the local level ment has a high impact on the expenditure and performance are still weak. Execution rates are understated. In Ethiopia, of sector agencies. Some costs are explicit in the form of in- sector project management units and regional bureaus of creased penalties and arrears due to the modification and re- finance constantly have to follow up with woredas (districts) negotiation of contracts. Procurement delays also result in in- for statements of expenditure and interim financial reports. efficiencies, lack of transparency and accountability, and poor procurement management, which together increase the cost of procurement and, in turn, of the services delivered. The op- Low Project Management and Contracting 5.2.3   Capacity portunity costs (forgone returns on investment) of delayed completion of projects can also be substantial. Further down the budget chain, the poor state of project man- agement in all the sample countries is a major constraint on Annual spending cycles are an obstacle to efficient spending. spending and results. Lack of technical capacity in line minis- Typically, the unspent budget of one year cannot be carried tries, parastatals, and local authorities limits ex-ante project over into the next year. The first invitations to bid for contracts appraisal and ex-post project evaluation. might not be issued until the budget is approved and warrants or spending authorities are issued, which may be months into Procurement procedures are cumbersome in all the sample the financial year. Contracts are managed individually rather countries and are exacerbated by the multiplicity of donor than as elements of a strategic plan, and expenditure is moni- agencies, each of which has their own procurement rules and tored separately from physical progress. As a consequence, procedures. One water utility in Mozambique, for example, there is a disconnect between expenditure and outputs. had 19 donors in 2008. This places great demands on the limited capacity of the host government. Another problem is that procurement is not planned as part of the planning and budgeting process. Despite a legal requirement for annual Sector Specific Reasons for Low 5.3  procurement plans in Sierra Leone, 51 percent of procure- Rates of Budget Execution ment expenditure in 2007 was for unplanned items. 5.3.1  Incomplete Sector Reforms All countries surveyed, except the Democratic Republic of Con- go, have elaborated a comprehensive set of water sector policies and subsector strategies. Many countries have also passed a Box 5.2: Overcoming Procurement regulatory framework comprised of a water law and comple- Bottlenecks in Ethiopia mentary regulatory instruments. In Oromiya, the largest region of Ethiopia, though the procurement process involves only three agencies—the regional water bureau, Federal Ministry of Water But implementation and enforcement of sector reform strategies Resources, and World Bank— a handpump purchase took 12 months from raising remain incomplete in many cases. Several factors hamper the the bid documents to signing a contract, plus the delivery period. According to the 2009 Ethiopia Public Finance Review, the delay arose from debate on the final full implementation of these reforms. First, human resources, version of the bidding documents. The United Nations Children’s Fund (UNICEF) especially at the ground level, are often inadequate. For ex- attempted to get around this problem by procuring handpumps itself and delivering ample, Burkina Faso has passed a comprehensive set of laws them to the regional water bureaus, but government staff still reported long delays. and regulations related to water supply and water resources One good practice was woreda (district)-level procurement under the Finnish- management over the last decade, but authorities face prob- supported Community Development Fund project. Woredas posted current prices of goods outside the woreda office and facilitated market days, bringing a number of lems in enforcement due to inadequate human and financial communities together on one day to purchase from invited wholesalers. resources. Second, the institutional reform process has led to Ethiopia’s federal government has recruited procurement specialists to regional project the creation of new institutions with mandates that overlap management units to strengthen procurement at the regional and woreda level. With those of existing institutions. For instance, in the Republic of over 700 woredas in Ethiopia, there is a tremendous need for capacity building. Congo, a newly established implementing agency for rural Source: Lockwood and Smits. 2011. water supply was given a mandate very similar to that of the Ministry of Water, leaving the ministry with an unchanged 22  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa mandate but drained of scarce human and financial resourc- of WSS expenditures are still made by the central ministries es. Third, sanctions are often difficult to implement due to a and water parastatals, which have resisted the devolution of lack of bylaws and other implementation-related legal tools. power and the reduction of budgetary resources and staff Fourth, the reform processes have to some extent been donor numbers. This is true in the francophone countries (in par- driven with limited commitment from national stakeholders. ticular Madagascar and West and Central African countries), in which the investment budget was still mostly managed The creation of new sanitation-specific sector agencies may by central government institutions during the review period. have failed to boost sanitation investments. In recognition Tanzania is a notable exception: it has to some extent imple- of the importance of promoting sanitation as a means to mented a decentralized WSS approach, somewhat changing achieving better health conditions, several countries (for the role of central government institutions. Transfers to Tan- example, Mali and Ghana) have taken steps since the mid- zanian local governments reached nearly 40 percent of the 1990s to establish lead agencies for the sanitation sector, sector budget in 2008 from 0 percent in 2005 (box 5.3). Sec- separating the institutional efforts for sanitation from those tor ministries are increasingly becoming facilitators instead of of water supply. Promotion of household sanitation had to a implementers. In Mozambique decentralization has mainly large extent been considered merely an appendix to water been an exercise of deconcentration of expenditure, where- supply infrastructure construction in the past. But it is ques- by deconcentrated government institutions and districts ab- tionable to what extent disconnecting sanitation from water sorb a small portion of the flow of sector funds through the supply has led to a boost in sanitation investments and more central sector ministry to which resources are allocated. effective promotional activities. The share of the domestic WSS budget actually transferred to Some countries developed comprehensive longer-term sector subnational governments is small and unpredictable. Only a development plans during the study period.14 The plans are few countries’ PERs had data on both domestic sector bud- linked to overall growth and poverty reduction frameworks, gets and actual transfers to local governments. What data translating national investments into the projects needed to there were are summarized in table 5.2. achieve the MDGs by 2015. In a few countries (mainly the anglophone countries such as Tanzania, Mozambique, and Tied central government grants are typically the principal source Sierra Leone), sector resource allocation models have been of WSS funds for rural local authorities. The grants may be developed as part of the decentralization process. These supplemented by local revenues, but these are typically not plans have contributed to strengthening sector dialogue, es- more than 5 percent of total resources. Some local authorities pecially between governments and development partners. receive significant external resources, but the main source During the study period most countries had started organiz- of WSS financing remains the central government grants for ing regular joint sector reviews. recurrent and development expenditures. Funding from the central government is not always sufficient, however. In Si- erra Leone, for example, though the law establishes a basis 5.3.2  The Limbo of WSS Delivery Decentralization for equitable annual water grants to rural local councils that Little or no progress has been made in devolving financial resourc- are budgeted as a “protected expenditure,” actual transfers es to local government in most of the countries studied. In the fall short. These statutory transfers should be pegged to a past decade, most countries have adopted a legal framework share of revenue (World Bank 2010). for decentralized delivery of WSS facilities, implying an antici- pated devolution of government resources to local govern- Incomplete decentralization has created an institutional vacu- ment authorities and a reorientation of central government um in some countries, with neither national nor local govern- institutional mandates toward overall sector policy planning, ments fully taking responsibility for the provision of rural water monitoring, and preparation and enforcement of the regula- supply. Examples of this include Mali, the Democratic Repub- tory framework. Frequently, however, responsibilities have been devolved to local authorities without the corresponding human and financial resources needed to implement them. 14 That is, during the study period, countries detailing sector invest- ment plans in a 10 to 20 year perspective (identifying needs in terms of investment, recurrent budgets, and maintenance/renewal) included The slow transfer of personnel and budgets to local councils has Madagascar (2005), Burkina Faso (2005), Cameroon (2007), Tanzania, obstructed progress in WSS decentralization. A high proportion Mozambique, and Ghana (2004). Budget Execution Rates  |  23 Table 5.2: Domestic WSS Budget Transfers to Subnational Governments, 2002–09   Domestic WSS budget ($) Actual transfers to subnational governments ($) % transferred Central African Republic, 2002–08 3,200,154 0 0.0 Côte d’Ivoire, 2002–07 20,690,296 266,830 1.3 Madagascar, 2006 7,001,815 11,203 0.2 Mali, 2000–06 18,442,035 255,502 0.1 Sierra Leone, 2006–09 12,488,031 733,523 5.9 Tanzania, 2005–08 768,366,327 73,877,723  9.6 Source: Country WSS PERs. lic of Congo, and the Republic of Congo, where the absence of leadership is highly detrimental, especially to the opera- Box 5.3: Matching Decentralization with tion and maintenance (O&M) of supply facilities. To address Increased Financing to Local Governments poor performance of water facilities in particular, a funda- in Tanzania mental rethinking of current O&M arrangements needs to Tanzania’s 2002 National Water Policy put in place a new institutional framework for build on the decentralization process. The legal ownership the water sector that devolved service provision to the lowest appropriate level. Fiscal of rural water facilities has in most cases been transferred to decentralization has followed suit. In fiscal year (FY) 2003, none of the water sector the local government, which, depending on the technology budget in Tanzania was allocated to local and regional governments. By FY08, 22 model, has delegated O&M to water boards, private opera- percent of the total budget allocation was going to local governments and another 15 percent to regional governments. But while budget allocations have increased tors, or beneficiary communities through formal or informal rapidly, actual expenditures have lagged behind as execution rates at subnational agreements. But the low levels of control, supervision, and levels were lower than at the central level. support to the rural operators and managers put the sustain- The local governments’ development budget is essentially for the delivery of rural ability of the facilities at risk. water supply services, while the largest part of the ministry’s budget is allocated to urban water supply, with the water resource management subsectors making up National utilities provide urban water supply services in two- a smaller portion. Regional authorities undertake supporting services; the largest thirds of the reviewed countries. Ten of the 15 countries re- part of their budget is linked to feasibility studies. The increasing share of funding available from regional and local governments tends to result in an overall larger part viewed have national utilities; only two countries (Tanzania of the development budget allocated for rural areas. and Ethiopia) have fully decentralized their urban water sup- Regional government spending has increased rapidly due to one foreign-funded ply services to local authorities. The other three countries rural water project. It is not clear whether this regional allocation is structural or have a mixed picture: in Sierra Leone, one utility provides temporary. Regional budget execution in FY07/08 was 63 percent, with large water to the capital, while a second national utility is respon- variations between regions. sible for water supply in secondary cities. Decentralization Local government expenditure doubled between FY04/05 and FY07/08, but budget is ongoing in Mozambique and Madagascar, which have a allocations increased fivefold over the same period. Local governments’ water mixture of one national and multiple municipal utilities. The sector budgets are almost entirely funded through central government transfers. rate of decentralization of urban water supply is somewhat In FY07/08, only half of local government budget allocations were actually spent because of significant delays in the release of budget funds. lower than the overall rate across Africa. Overall, national utilities still serve just over half of the countries in Africa, with A system of WSS budget transfers was introduced in FY05/06 using a formula based on a combination of indicators including coverage rates, technologies used in the municipal service provision in only 15 percent of countries. district, and poverty incidence. The actual calculation of the formulas is hindered by poor data quality and lack of disaggregation. The difference between budget Hygiene education is formally decentralized, but local authori- allocations and actual allocated budgets at the regional and the district level shows ties do not have the means to carry out their responsibilities. that the water formula is not consistently implemented. Some regions and districts While sewerage in urban areas is often the responsibility get significantly more funds than the water formula calculation would allow for, while in others the opposite holds true. of national utilities, responsibility for on-site sanitation and hygiene education has mostly been decentralized. But local Source: Tanzania PER. governments do not have adequate financial and human re- sources to carry out his mandate. 24  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa The pace of decentralization of the WSS sector is slowed by resis- tions of the water ministry during the study period is very tance from central-level government institutions claiming insuf- dependent on external funding (WSS investment projects), ficient capacity at local government level. This is a valid argu- but the provided support has mainly been targeted on ar- ment to the extent that local governments have generally eas likely to facilitate the investment program. Obviously this not been provided with the human or financial resources to adversely affects the possibility of the regional water offices take up their new responsibilities. But in addition to the ob- to provide adequate support to local governments in carry- vious unwillingness to lose control of funds, this resistance ing out their statutory roles and to ensure that monitoring also reflects the inability of central ministries to adapt to their and control functions are carried out. Capacity building at new roles. Central government WSS staff are still mainly en- provincial or district levels was supposed to be supported by gineers and technicians, with limited skills and experience in the deconcentrated staff of central government bodies. The planning, monitoring, and sector coordination. political decision to decentralize WSS has not led to a sub- stantial transfer of central government staff to lower levels Capacity building at provincial or district levels was sup- (province/regions/districts) and deconcentrated staff num- posed to be supported by the deconcentrated staff of cen- bers remain modest. tral government bodies. But in reality the political decision to decentralize WSS has not led to a substantial transfer of central government staff to lower levels (provinces/regions/ districts), and deconcentrated staff numbers remain mod- Box 5.4: Decentralization of Budgets in Mali est. At the extreme, in the Central African Republic, only 4 out of 81 officials of the directorate responsible for water are At the subnational level, WSS in Mali is the responsibility of nine regional offices of posted outside the capital. the Ministry of Water. Each office covers an average area of approximately 137,000 square kilometers. From 2001–06, the allocated recurrent budget (excluding salaries) of the regional offices amounted to $4,800 per office per year. This constituted only 22 Additionally, very limited financial resources have been allocat- percent of the total recurrent budget (excluding salaries) of the WSS sector and only ed for operating purposes at the deconcentrated ministry level. 0.2 percent on average of the total public WSS budget. The offices were only able to Mali reflects a general trend in surveyed countries in this re- function with support from externally funded projects operating in the regions. spect (box 5.4). The functionality of the regional representa- Source: Mali PER. Budget Execution Rates  |  25 6.  Exploring the Link between Public Expenditure and Improved Access W e found no overall relationship between levels of spend- ing public WSS expenditure is a necessary but insufficient ing and levels of water supply and sanitation (WSS) ac- tool for improving services. For instance, the successful pro- cess. In figure 6.1, which plots the per capita expen- grams to improve access to urban water supply in Senegal, diture against water supply access rates, no trend between Burkina Faso, and Niger all included large public investment access and expenditure can be distinguished. Likewise, we programs. Similarly, progress in rural water supply access in found no relationship between expenditures and increases Benin and Mali has followed increases in public expenditure. to access, or between capital expenditure and additional population served. In rural areas, access to drinking water increased alongside per capita spending. The Sahelian countries have substantially This missing link between levels of spending and levels of WSS better rural water supply coverage than other African coun- access is probably due to underlying methodological difficul- tries and a higher percentage of sector expenditure as a ties. Firstly, we are comparing access rates and spending data share of gross domestic product (GDP). But the causality of that are both aggregate at a country level. This means that more spending resulting in better access is hard to establish; the analysis does not pick up on local increased spending it could be that higher levels of access leads to more spend- contributing to localized service improvements. Secondly, ing (see figure 6.2). the analysis does not take into account different unit costs for different types of service provided. The review period The cost of increased rural access to water varied, often due might also be too short to see impacts from increases in ex- to technology choices. As mentioned above, comparisons penditure. For instance, Tanzania’s public expenditure review (PER) found an increase in expenditure starting in 2000 and Figure 6.1: Per Capita Spending and Access Rates an increase in access from 2004 onwards (box 6.1). PERs are to Drinking Water for All Reviewed Countries an instrument that has been mainly used for sectors in which an increase in recurrent expenditure can lead to improved 75% services in a short time period, such as health and educa- tion. The water supply and sanitation is capital intensive and 50% Access level assets have a long asset life. The analysis did not take into account the inherited asset stock prior to the study period. 25% The high level of expenditure on rehabilitation of infrastruc- ture suggests that a considerable part of capital spending is 0% used for deferred maintenance to maintain current access 0 1 2 3 4 and service levels. But the lack of correlation could also be Per capita expanditure (USD) an indication that spending is not the key determinant in in- creasing access to water supply. rural only rural + urban No countries have managed to increase access without a sub- stantial increase in public expenditures. In other words, increas- Source: PER reports. 27 Figure 6.2: Example of Relationship between Box 6.1: Linking Public Spending to Spending and Access (Mali 2001–06) Increased Access in Tanzania 4.0 51 The PER for Tanzania’s water sector found that a sharp increase in funding started to 3.5 50 have an impact on access to improved water sources between 2004 and 2008. The 49 Expenditrure per capita 3.0 Acces to rual water WSS sector’s share of the total priority sector budget increased from 3.7 percent in 48 2.5 47 FY00/01 to 9.8 percent in FY07/08. The budget allocations for the sector are relatively 2.0 46 high compared to other countries in the region. Nevertheless, the PER notes that the 1.5 45 progress has been slower than expected and that increasing resources alone is not 1.0 44 enough. Even though budget allocations increased rapidly, budget releases lagged 0.5 1.17 1.03 0.87 1.62 2.97 3.51 43 significantly behind. Water utilities are still heavily dependent on budgetary support 0.0 42 with a significant part of the budget allocated for operations, maintenance, and 2001 2002 2003 2004 2005 2006 rehabilitation. Expendit. /capita (USD) Access rural areas The PER lists a number of issues that explain why the goal of improving access to WSS services is moving slower than expected. Some of these issues are systemic in nature, that is, they are mostly outside the control of the water sector, but still influence progress in the sector. Other issues are within the direct control of the Source: Country WSS PER papers and authors calculations. water sector and need to be addressed to ensure that the sector can move more quickly to achieve its goals. The following sector measures need to be taken to ensure that spending translates to better access to services: (i) improve sector investment planning, (ii) improve the capacity in the sector to conduct procurement and disbursement, (iii) focus on The relationship between expenditures and access is less clear in including incentives in the allocation of funding, (iv) improve efficiency to lower urban water supply. Public expenditures on urban water sup- operating and capital costs, and (v) promote sustainable tariffs while guarding the ply mostly concern water utilities. The fact that urban water affordability of access. supply coverage has not increased in the past years means At the same time, the PER also lists a number of systemic measures to be taken that much of the investment has gone to maintaining the mainly related to (i) improving budget procedures; (ii) harmonizing procurement, current coverage of water utilities rather than expanding ac- disbursement, and monitoring procedures, and (iii) improving the predictability and cess. Eighty percent of urban households with piped water reliability of donor funding to the sector. connections are in the top two quintiles of income distribu- Source: Tanzania PER. tion, while only 10 percent of poor households have piped water. As this subsidy mainly benefits the relatively rich, sub- sidies to urban utilities are effectively consumer subsidies to the rich. Few countries have specific public expenditure between countries are difficult. But spending efficiency in programs for those that are not connected to the network. In the Sahelian countries of Mali and Niger seems to be lower than in Cameroon and Madagascar (table 6.1). Obviously, natural conditions in the Sahel region adversely impact unit Table 6.1: Comparison between Annual costs (for example, lower success rates on borehole drilling, Expenditure and Increase of Water Access Rate in pumping as the only feasible technology, low density of Selected Countries during Study Period population, and so on). Additionally, service-level definitions differ among countries, and it would thus be hazardous to Average annual expenditure Average yearly increase draw conclusions on spending efficiency in one country per capita in US$ of water access rate compared to another. The 2010 Country Status Overview Cameroon 1.9 2.2 reports that technology choice has been the key determi- Madagascar 0.5 1.3 nant for aid per beneficiary. Aid per beneficiary was higher Mali 1.9 0.9 in countries opting for rural piped water schemes (Maurita- Niger 1.8 1.2 nia, Senegal, Côte d’Ivoire, the Gambia, and South Africa). By contrast, countries that have had a policy of low cost so- Source: Country WSS PER papers and authors calculations. lutions such as Burkina Faso, Ethiopia, Uganda, and Malawi have considerably lower aid per beneficiary costs. A similar 15 An ongoing project called WASHCOST is looking into the true life- argument can be made for public financing in general.15 cycle costs of sustainable services (www.washcost.info). 28  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa some countries, donors finance programs focused on small- Congo, no public standpipes have been constructed in the scale providers, but these are often only at pilot scale. As a area served by the national urban water company. On the result, poorer segments of the urban population pay a much other hand, some of the most successful urban water supply higher price for a service level that is lower than the segment programs have been through social connection programs, of the population with access to a private water connection. in which (often small diameter) household connections are made available for poor households at much reduced costs. It is worth noting that some of the largest increases in urban For instance, in Senegal over 2.5 million people got access to water supply access in Sub-Saharan Africa have been made water supply, nearly exclusively through house connections, through programs focused on subsidized house connections at most of which were subsidized social connections. Burkina a relatively high unit cost. Unsubsidized house connections, at Faso also relied heavily on household connections to expand a cost of $250–$300, are financially beyond the reach of many access to 95 percent of urban households. Both Burkina urban dwellers. In countries with a focus on house connec- Faso and Senegal depended heavily on donor financing to tions but without subsidized social connection programs, increase access, with aid of $189 per beneficiary in Burkina this has meant that access figures have been stagnant. For Faso and $141 per beneficiary in Senegal (AMCOW 2011). instance, in the Democratic Republic of Congo’s Katanga Province, only 0.38 percent of total urban water consump- For sanitation, data availability is too limited to establish a rela- tion is sold through public standpipes. And in the Republic of tionship between expenditure and access levels. Exploring the Link between Public Expenditure and Improved Access  |  29 7.  Underlying Public Economy Dynamics T his chapter explores the political economy of water supply and environmental benefits of WSS services also make WSS and sanitation (WSS) spending patterns. After further de- services a public good. Lack of WSS services accounts for veloping the WSS public spending rationale first intro- 5.5 percent of death and illness in high-mortality develop- duced in chapter 1, we test whether actual patterns match ing countries (WHO 2002). For example, the costs of envi- the rationale. We then explore the reasons behind the misal- ronmental and health degradation due to inadequate WSS location and poor implementation of public WSS resources, services has been estimated at more than 1 percent of gross including a discussion on political economy factors. Unlike domestic product (GDP) in Colombia, 0.6 percent in Tunisia, the bulk of the overview paper which was based almost and 1.4 percent in Bangladesh.16 exclusively on the findings of the 15 public expenditure re- views (PERs), this chapter broadens the scope to draw on WSS service provision is a good example of a natural monop- other sources as well. oly, as the necessary network infrastructure cannot practically be duplicated. WSS can be considered nonrival (nonsub- tractable), as the sector is subject to large economies of den- What Is the Rationale for Public 7.1  sity. For a given distribution network, increasing the number Spending in the WSS Sector? of households connected reduces the network’s average and marginal costs. Similarly, maintenance of nonnetwork rural Two justifications for WSS public spending are correcting mar- water supply facilities, such as handpumps, is most efficiently ket failures and reducing disparities in service delivery. Without provided through a monopolistic service provider. public spending, the market could fail to provide a socially optimal level of WSS services. Reducing disparities in access The WSS sector is characterized by a high degree of sunk to basic WSS services is a government responsibility (WDR costs. Capital costs generally make up 66–80 percent of the 2004). costs of supplying services. Infrastructure investments are substantial and very long term—many water pipes, if well Individuals have little incentive to build and maintain WSS net- maintained, can last for decades, or even centuries. works, but communities and societies do. Individuals might invest in on-site facilities, such as wells or latrines. But the The sector suffers from imperfect information. For instance, expense and complexity of piped networks requires collec- lack of knowledge of the benefits of hand washing or proper tive action. Markets, therefore, fail to provide adequate WSS use of latrines can lead to less than desirable investment and services on their own. Reasons for this include: consumption. WSS are a mixed public and private good. Water can be con- Reducing disparities in access to basic WSS services is a govern- sidered a private good as benefits accrue to individuals. In ment responsibility. The poor face huge barriers to being con- theory, consumers could be excluded by cutting off service nected to WSS networks (a form of “market participation”) in- provisions. In practice, however, it is often culturally or po- litically unacceptable to exclude people from at least a mini- 16 World Bank, Country Environmental Analyses—Colombia (2006), mum level of WSS access. The considerable public health Bangladesh (2006), Tunisia (2004). 31 cluding establishing legal tenure on the necessary land. One to standposts (Banerjee and Morella 2011). But this would mechanism for closing the gap in service delivery is therefore require changing the institutional arrangements associ- to target public spending at households that otherwise would ated with standposts in African cities. At present, because be unable to afford those services. In addition, water supply standposts contribute a negligible portion of revenue for subsidies can be presented as a component of a broader so- most utilities, they have no financial incentive to expand the cial policy agenda to redistribute resources toward the poor. service. Although operating costs are considerably higher Particularly in countries where means tested cash transfers are for standposts than for piped connections, a recent survey hard to implement, consumer utility subsidies can offer an al- found that the official consumer standpost tariffs of half of ternative for delivering transfers to the poor. More equitable Africa’s water utilities were lower than those for consumers service delivery in turn can contribute to social cohesion, par- of small volumes of piped water (Banerjee and others 2010). ticularly important in fractionalized societies. In practice, retail standpost prices are several orders of mag- nitude higher than the utility-imposed price in most African The rationale for public spending on WSS is often not purely eco- countries because of high rent-seeking behavior on the part nomic; many consider water access to be a political, emotional, of operators and reselling through informal channels. or religious matter. Governments—and the societies they represent—often see WSS services as a human right.17 They Only a small part of public expenditure goes to sanitation, are galvanized in this by international endorsement of the a cheap life saver. As the sanitation sector is dominated by Millennium Development Goals (MDGs). The human rights household on-site facilities, limiting public expenditure in argument, however, is often mistranslated as a rallying cry this area might be partly justified. But one rationale for pub- for free water. Water tariffs and privatization of water supply lic spending—health externalities and overcoming informa- systems were at the core of the heated globalization debate tion imperfection—would argue for increasing spending of the 1990s, and WSS services are at a crossroads of capital- on hygiene education and sanitation marketing as well as intensive infrastructure and value-loaded social services. The subsidies for latrines. Going forward, it should be noted that politics of water services are thus a reflection of much larger environmental public good reasons can have a perverse politics of welfare. incentive to prioritize investments in expensive wastewa- ter collection and treatment systems for the rich few above investments in more latrines and septic tanks for the many Do Actual Public Spending 7.2  unserved poor, which are considered a private good despite Patterns Match their Rationale? their higher health impacts. We found that patterns of public spending stand in stark con- Public spending patterns do not fully reflect the long asset life trast to the public spending rationale outlined above. Spend- of WSS infrastructure. While the rationale for public spend- ing is often not properly targeted at extending services to ing outlined above calls for prioritizing public spending on the poor or addressing health and environmental externali- operations and maintenance (O&M), the reality is the oppo- ties. The rationale is also undermined by the low efficiency site, with upkeep of existing WSS facilities appearing to be of spending—in other words, spending does not provide a underfunded. Under maintenance of infrastructure leads to very high “bang for the buck” toward public policy objectives. higher life-cycle costs, because the present value of rehabili- tating infrastructure is substantially higher than that associ- ated with a sound preventive maintenance regime (Briceño- Does Public Spending Help Overcome 7.2.1   Garmendia, Smits, and Foster 2008). About half of recurrent Market Failures? spending in the sector is on salaries, which crowds out other Current public spending is not in line with stated public health essential recurrent expenditure, especially maintenance. As objectives. Public spending is mostly focused on higher-level a result, infrastructure dilapidates rapidly, leading to an ex- services, such as household water supply connections, at the pensive cycle of neglect and rehabilitation. We found break- expense of channeling money into cheaper levels of service, such as standposts, that would have considerably higher 17 See, for instance, UNDP (2006) and the General Comment 15 to the health returns per dollar invested. Utilities could therefore UN’s International Covenant on Economic, Social and Cultural Rights. double the rate of service expansion in Africa by shifting There is a heated international debate on the significance of declaring their investment budgets from piped-water connections water a human right, including its implications for cost recovery. 32  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa down rates of rural water supply facilities: at least 25–30 per- network. A recent review of African water utilities found that cent in most of the countries surveyed, and more than 50 in almost three-fourths of cases, even consumers connected percent in postconflict countries.18 Urban areas are locked in to the piped network with water intake at the survival level a vicious spiral of weak performance, low willingness to pay pay the same or more per unit than average consumers be- by consumers, and insufficient maintenance funding lead- cause of the fixed and minimum consumption charges and ing to deterioration of assets. The result is that a consider- the definition of the IBT blocks (Banerjee and others 2010). able share of public spending goes to rehabilitation (that is, deferred maintenance). Public subsidies to reduce the costs of connecting to the net- work are a progressive alternative to consumption subsidies. Low water tariffs undermine the rationale that governments can The majority of African water utilities levy piped-water con- prefinance capital investments to be recovered from consumers nection charges in excess of $100, an insurmountable bar- over time. A recent World Bank study found that less than rier for low-income households (Banjeree and Morella 2011). 25 percent of utilities in middle- and low-income countries Several countries have started social connection programs to charge tariffs that are sufficient to cover efficient O&M while subsidize connections, often using smaller diameter pipes to making some contribution to cover investment costs (Komi- target low consumption households (for example, Burkina ves and others 2005). Our review found that in most coun- Faso, Côte d’Ivoire, Senegal). These programs often also allow tries tariffs have not been adjusted for many years. utilities for the first time to connect households without legal land tenure. In other countries, governments remain reluc- tant to provide services in unauthorized settlements because Does Public Spending Help Close the Gap 7.2.2   they fear services imply a permanent right to occupy land. between the Served and the Unserved? Though the gap between rural and urban WSS access is clos- Weak Utilities Act as a Buffer between Public 7.2.3   ing, current spending leads to more inequality and could poten- Spending and Public Policy Outcomes tially erode social cohesion. We found that poor people do not get their fair share of public spending on services, let alone A large part of public spending on urban WSS is absorbed by utili- the larger share that might be justified on equity grounds. ties to cover recurring losses caused by revenues that are well be- Spending is skewed to services disproportionately used by low operation and maintenance costs. Just 36 percent of the richer people in capital cities at the expense of people in utilities in Africa have tariff levels that meet their full operation slums, secondary cities, and rural areas. Within the countries and maintenance costs, and only 9 percent have some money we surveyed, important and persistent regional disparities left over to go toward their capital costs.19 Inefficiencies include remain between rural and urban WSS access. The uneven- underutilization of existing capacity in water treatment plants, ness in spending we found was not so much between urban poorly designed plants, excessively high losses in distribution and rural, but rather between capital cities and the hinter- networks, low billing and collections rates, and overstaffing. land. And even within capitals, most spending benefits only But it should be recognized that, given the capital intensity of those who are connected to the network, inevitably the rich. water services, even large gains in operating efficiency have relatively small impacts on the total cost of service. Operating Water utility tariff subsidies are starkly regressive. About 80 per- costs account for about one-third of water supply costs. Thus, cent of utilities use an increased block tariff (IBT) in which the achieving a challenging 25 percent reduction in operating price charged per unit increases at higher volumes of con- costs would reduce the total cost of service provision by only sumption in an attempt to subsidize those consuming small about 10 percent (Komives and others 2005). volumes. Poor households, however, capture only half as much of the value of IBT subsidies as they would if the subsi- 18 Lockwood and Smits (2011) estimate that 20–40 percent of water dies were distributed randomly across the entire population points are not functional. A recent United Children’s Fund (UNICEF) because of low levels of connection among the poor and IBT study (cited in RWSN 2009) showed an average level of nonfunction- subsidy design issues (Komives and others 2005). Quantity- ing handpumps of approximately 36 percent in 20 countries in Africa). Other studies and sources reveal similar levels. based subsidy programs are similarly ineffective if access 19 Banerjee and others 2010. The paper assumes a threshold of $0.4/ to the network is low (this is true of all countries in our PER m3 for O&M cost recovery and is limited to analyzing the possibility of sample); most poor households are excluded from subsidy tariffs to cover costs, and not whether the utilities are actually raising programs altogether because they are not connected to the the revenues to meet the costs (assuming 100 percent collection ratio). Underlying Public Economy Dynamics  |  33 Reductions in capital costs have a much greater potential for often the worst nonpayers, building up extensive arrears improving the efficiency of public spending. It is striking, there- and considering themselves exempt from bills and immune fore, that alternative technologies with the potential for ma- from cutoffs. Elites receiving better services at a low cost jor cost reductions are only pursued in a few countries. Most can provoke others to access services illegally. This could countries in our review still strive to provide a single stan- help explain the large numbers of “illegal” or “informal” con- dard of service, often following engineering standards lifted nections reported in sample countries. Other expressions of directly from industrialized countries. Similarly, demand-side discontent include vandalism. In turn, governments might management to reduce network losses is strikingly absent respond by reducing or avoiding investment in the expan- given their potential to increase the efficiency of public sion of services to “difficult” communities, blaming users for spending. Reducing capital costs will also require improve- the problems that arise (Muller, Simpson, and Van Ginneken ments in the planning, design, and execution of capital proj- 2008). Notably the public spending bias toward capital cit- ects, which could result in significant savings. ies was particularly stark in postconflict states (for example, the Democratic Republic of Congo and Central African Re- public). This might be partly due to practical obstacles to im- What are the Reasons behind 7.3  proving services in the hinterland. But it might also point to the Misallocation and Poor politics based on identities and patronage. Implementation of Public A political economy perspective on public service delivery sug- Resources in WSS? gests that choices in capital spending may be driven by the cor- ruption, employment, and profit opportunities that construc- This review has revealed large variations in the effectiveness tion provides. There are, in contrast, few systemic political in- of spending. Previous chapters have identified the various centives to ensure that systems are kept running over time. technical and managerial challenges of translating political Regularly monitoring services and attributing breakdown or statements into sector expenditures, and in turn into better continued operation is difficult. In turn, it is harder for politi- WSS services. But what are the political economy factors that cians to claim credit for these services. might explain the gap between WSS as an expressed priority and actual changes in budget allocation and disbursement? The impact of corrupt practices within organizations with con- This section explores underlying factors influencing observed trol over investment projects is hard to quantify. The review spending patterns, recognizing that the global water crisis is noted large—and often unexplained—differences between rooted in power, poverty, and inequality (UNDP 2006). unit costs. Procurement costs for capital works may often be unnecessarily high because of a lack of competition and the 7.3.1  Political Factors Drive Spending Patterns prevalence of kickbacks to corrupt officials. A recent study Spending patterns in WSS are in line with international evidence on the political economy of sanitation found some evidence that clientelism significantly influences the provision of pub- of rent seeking and corruption in all case studies, though it lic services (World Bank 2003). The tendency for political pa- was not identified as the predominant feature distorting pro- trons to provide private rewards to clients can help explain poor sanitation investments (WSP Sanitation Global Practice the disproportional spending in capital cities. Public money Team 2010). is often spent where the politically powerful reside; this is where elections are won, or at least where potential social Middle and Upper Class Voters Resist Tariff 7.3.2   discontent has to be controlled. Reforms Political patronage might also explain low revenue collection Politicians’ refusal to raising tariffs makes for good political pro- caused by uncollected bills and malfunctioning meters. Increas- poor rhetoric but in practice mainly benefits the middle and ing collection is normally seen as a technocratic “quick win,” upper class that are connected to the public water network. It as it has an immediate impact on the financial position of the has been estimated that achievement of full-cost recovery utility without requiring much investment. We found anec- would require manifold tariff increases for residential con- dotal evidence suggesting that nonpayment is sometimes sumers. But in most of the countries in our sample, only the indirectly encouraged by government agencies and political rich are connected to the network, which might render this leaders. In the PER countries, government departments were affordability argument politically convenient. Tariffs that fully 34  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa recover capital costs would be affordable for only half of also hired less on merit and more on how best to appor- the population in Africa (Banerjee and Morella 2011). Poor tion patronage, and absenteeism is not penalized in several people currently pay many times the official water tariffs, as countries. This patronage system often leads to a status quo they buy their water from private providers. This leads to the in which services do not improve. In some cases, particular conclusion that recovering full costs from existing custom- interest groups take control of public resources for private ers and using the resulting cash flow to accelerate access benefit, including reported cases of utility staff involved in expansion for the poor would substantially increase equity, the informal resale of water at exorbitant prices. although it is a hard political sell. 7.3.5  Professional Resistance to Change 7.3.3  Resistance to Decentralization Professional resistance to change, in the form of mistrust of cheap- The political dynamics in rural WSS seem to be slightly different er (and sometimes unproven) service levels and other norms of than in the urban water supply sector dominated by utilities. the professional engineering culture, creates a barrier to techno- Here, the most striking finding is how little or no progress logical innovation. Service quality standards and technology has been achieved in devolving financial resources to local choice are often also restricted by outdated and complex laws, governments despite official policies—often embedded in regulations, and construction codes. Revising such technical the constitution—to decentralize WSS services. Instead, poli- standards would be an easy and cost-effective way of improv- ticians and central bureaucrats have been allowed to keep ing access for the poor. But the politically well-connected public budgets concentrated in state administrations. The re- construction industry might object as a change in standards sulting dangerous institutional vacuum in the rural provision would allow new (smaller) enterprises to enter the market. of water supply, with neither national nor local governments fully taking responsibility, might be convenient for all—ex- cept for rural people without water. In interviews for the PERs 7.3.6  Low Political Interest in Sanitation in our sample, local decision makers consistently blamed the Looking at political dynamics also helps explain why sanitation lack of progress on the slow transfer of personnel and bud- is an orphan sector. The low demand from communities and gets to local councils. Central government representatives, households for sanitation investment is well documented. meanwhile, claimed insufficient capacity at the local gov- As a result, politicians do not perceive sanitation as a vote ernment level. A telling example of this negative dynamic is winner and allocate scarce resources to sectors with higher found in South Africa. Here, after municipalities complained perceived political rewards. But sanitation is a cheap lifesaver, about the central government “hanging on” to water-service and as such might merit higher public spending. A recent functions, a date was set for the transfer of responsibility. But study on the political economy of sanitation found that the same municipalities then called for the process to be de- a strong civil-society push focused on health benefits and layed because of inadequate preparation (Muller 2007). dignity increased political incentives (in the shape of career advancement or electoral support) for extending sanita- As rural water supply systems are less technically complicated, tion coverage to the poor in Brazil and India. The need to services are often directly managed by local authorities rather target such campaigns is evident from a case study in Sen- than special-purpose utilities. As a result, the politics of rural egal. Many technocrats there support a shift towards lower- WSS can rarely be isolated from the broader issues of public cost technologies after seeing the successful use of on-site management. Cash from tariffs will often be used to fund sanitation and condominial systems in semi-urban areas of other functions of local government that lack a ready source Dakar. But many politicians still support investments in tra- of revenue, even if this is formally prohibited. At an even ditional sewage treatment plants (WSP Sanitation Global more localized level, the choice of where water pumps are Practice Team 2010). located often reflects local power and social relations. 7.3.7  Global Debates Influence Local Decision 7.3.4  The Self-interest of Utilities Making The interests of employees or organized labor often drive utility Local debates on WSS often reflect global political processes management decisions. Key posts in utilities and ministries rather than local concerns. During the 1990s, the Washington are often filled by politically trusted people. Personnel are Consensus included a strong push for full cost-recovery, trig- Underlying Public Economy Dynamics  |  35 gering a campaign against the commoditization of water to customization or implementation, may similarly have con- that became a spearhead of the antiglobalization movement. tributed to the unfinished sector reforms and multiple insti- This global debate trickled down into many national and lo- tutions with overlapping mandates. cal decision-making processes. It pushed many politicians and local civil society leaders to rally against cost-recovery The call to increase customer power and make service providers in absolute terms—even from rich consumers. At the same more directly accountable has not really taken hold in the sam- time, willingness-to-pay studies often showed that even the ple of countries yet. This idea was presented as a short route poor were willing to pay for a connection or standpipe to the to accountability as opposed to the long route of account- public grid as a cheaper alternative to other more expensive ability in which poor people contact the policy maker, who water sources. Political decision-making processes thus be- in turn ensures that the service provider delivers services to came paralyzed, missing the opportunity for a discussion on its customers (World Bank 2003). For now, neither account- how to make the rich pay for services while freeing up public ability route provides citizens a voice to ensure that the WSS funds to benefit the poor. budget is allocated and implemented effectively and effi- ciently in a way that matches the public spending rationale, For a long time, the global environmental debate drove a push which would lead to increased WSS access. for higher levels of sanitation services, but the recent global movement for basic sanitation has rebalanced the debate. The If the drivers of current WSS spending patterns are not ad- environmental debate often mixes the issues of water re- dressed, any changes that are achieved risk being marginal or sources management and basic service delivery. While high- temporary. But this should not be an excuse to stop pursu- er levels of sanitation services, such as sewerage wastewater ing technical improvements in the targeting and execution treatment, benefit downstream environments, local popu- of WSS public spending. It should merely lead to recogni- lations would mainly benefit—especially in public health tion of the limitation of short-term technocratic measures terms—from receiving basic services such as improved la- without exposing and addressing the longer-term drivers trines. The rise of the basic sanitation lobby was marked by that are the cause of the status quo. One way to begin to the adoption of a new MDG sanitation target to complement address the spending patterns is through the PERs. Clien- the existing water target, still the only target to be added to telism and patronage curve spending to benefit the few the MDGs since their adoption. Again, much of the discourse and obstruct institutional and technical renewal. The PERs is not indigenous to the countries that need the most work. can help to open the debate by showing this lack of effi- ciency in public spending. Advocates for better WSS servic- The gaps between policy and practice can partly be traced back es can use the PERs’ findings, as well as this regional review, to donor policy prescriptions that have been superficially ad- to expose the misallocation or misuse of public resources opted but have not been followed through due to local political and lobby for different and better spending in the sector. As resistance. The international community’s advocacy has con- the quality of spending improves, the emotional argument tributed to service delivery decentralization in some coun- for more spending will be strengthened by proving that tries. But budgets have not followed suit. Donor advocacy public spending contributes to better and more equitable to fix sector frameworks, suffering from insufficient attention WSS services for all. 36  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa 8.  Concluding Remarks PERs are a Useful Tool for Policy 8.1  of analytical tools could help authors of future PERs. While stan- dardization would serve global officials, it might limit the use Makers of the PERs for their main audience—in-country stakehold- R eviewing and tracking public expenditure are not new con- ers, including politicians, government officials, donor repre- cepts, but public expenditure reviews (PERs) for the water sentatives, and representatives of civil society. The WSS sec- supply and sanitation (WSS) sector are relatively recent in- tor is defined differently in different countries (for example, it novations. This review demonstrates that thoughtful applica- might include rainwater drainage, solid waste management, tion of PERs can contribute substantially to an understand- or water resources management), and countries organize ing of what is happening in the sector. Knowledge of the their sector differently. International actors will have to ad- quantity and quality of public spending is a prerequisite for just to each country’s specific system. The findings of this governments and donors adjusting their policies and prac- review endorse the ongoing World Bank effort to develop a tices. This, in turn, will be required to improve WSS access. set of tools and resources to evaluate the allocation of fiscal resources in a more consistent manner. The focus of this review has been to describe and assess the find- ings of PERs of 15 Sub-Saharan counties. We hope it will enable the authors of future PERs to benchmark their findings against More, Better, or Different 8.2  other countries in the region with more comparative data. This will provide decision makers with a better sense of where Spending? they are relative to other countries. In fact, data from this re- Successful advocacy for more spending in the WSS sector depends view were already being used in five recent country PERs. on improving the quality of spending. This in turn will require recognizing and changing the underlying power dynamic. But there is much more work to be done to improve the qual- ity of WSS information systems, including data definitions. All the PERs we reviewed faced serious data limitations. There 8.2.1  More Spending? is a critical need to give more attention and resources to There are compelling arguments to increasing public spending monitoring and evaluation—quality data is a prerequisite for WSS. Redistributive arguments and market failures call for ensuring both equity in distribution of investment re- for public intervention. The investment needs are huge. The sources and sustainability of investments. Another major overall price tag for reaching the Millennium Development obstacle to expenditure tracking is that a large part of do- Objective (MDG) target for access to WSS is estimated at nor resources are still off budget. The parallel donor data- $22.6 billion per year for Sub-Saharan Africa, or 3.5 percent of bases in many countries are only a second-best solution Africa’s gross domestic product (Banjeree and Morella 2011). to moving donor resources on budget and into the bigger But this review also found that current spending patterns are picture. both inefficient and ineffective and do not match that public spending rationale. Increasing the volume of public expen- Standardizing the methodology for PERs in the WSS sector is diture without changing its targeting and execution will not neither feasible nor desirable, but the development of a core set have a large impact. 37 Better Spending—Improving Budget 8.2.2   household on-site facilities and is generally financed from Execution household expenditure, this might be partly justified. But Budgets need to be better executed. A major focus should be huge public investments will soon be required as coun- on solving institutional bottlenecks in WSS public expendi- tries get richer and people move up the sanitation ladder. ture, both within and outside the control of sector profes- About half of recurrent spending in the sector is on salaries, sionals. Poor execution of budgets is caused by a variety of which crowds out other essential recurrent expenditure. The obstacles, such as overambitious plans and budgets, unpre- distinction between “development” expenditure and “recur- dictability and late receipt of resources, cumbersome pro- rent” expenditure is becoming less clear, however, as donor curement procedures, and lack of capacity, mainly project funding seem to include considerable rehabilitation expen- management and contracting capacity of the government ditures that could be classified as either capital or current. and its partners. Typically, the unspent budget of one year cannot be carried over into the next year, so the first invita- Changing the Game to Close the 8.2.4   tions to bid for contracts may not be issued until the budget Implementation Gap is approved and warrants or spending authorities are issued, which may be months into the financial year. Better budget There are huge gaps between policy and practice. The PERs can execution will require capacity building. therefore be a useful tool to hold governments accountable for the implementation of their own policies and promises. Sector professionals, including international actors, should At the sector level, we found that while nearly all countries broaden their scope from designing sector policies to imple- have elaborated comprehensive water sector policies and menting them, while working with others to address the bottle- strategies, implementation and enforcement of sector re- necks along the whole of the budget execution chain. The chal- form strategies remain incomplete, and efforts are needed lenge for practitioners is not to identify the perfect “magic in terms of capacity building, general public awareness cam- bullet,” but more subtly to build capacity, instilling and main- paigns, and further development of a legal framework that taining appropriate management cultures. The choice of can facilitate implementation of policies and strategies. We budget execution tools must be appropriate to the current also found that, rather than streamline the process, reforms and evolving state of the country and sector. Many of those had, in many cases, led to the creation of new institutions tools are not defined by sector ministries but by ministries of with overlapping mandates. Furthermore, decentralization finance and others. Water sector professionals should take an has stalled with little or no progress in devolving financial active interest in addressing all parts of the budget execu- resources to local government. This has created a dangerous tion chain, including participating more actively in upstream institutional vacuum in the provision of water supply, par- planning exercises and in the definition of the regulations ticularly in rural areas, as neither national nor local govern- relevant to their work, such as budgeting and procurement ments are fully taking responsibility. legislation. A second implementation gap is seen in donor financing, which is often badly targeted and unpredictable, resulting in execution 8.2.3  Different Spending—Enhancing Targeting rates that are lower than those of internal resources. A signifi- The need for better targeting is a strong conclusion of this re- cant opportunity is available to increase pro-poor targeted view. In general, areas outside of the capital, the sanitation donor financing by shifting resources to areas with the larg- subsector, and funds for the upkeep of existing WSS facili- est WSS needs. In the survey period, the WSS sector in most ties appear to be underfunded. Within the countries we sur- countries was characterized by a large number of donors op- veyed, important and persistent regional disparities remain erating on terms and conditions specific to their individual in access to water in rural and urban areas. Sector invest- projects. Donor funding is unpredictable and donor plan- ment planning and allocation policies have contributed to ning and monitoring is not necessarily linked to the govern- disparities in access and distribution of WSS services. Overall, ment budget calendar. Transaction costs for governments there seems to be a slight bias in public expenditure toward are high. Solving these problems should be given high pri- urban areas, which is at least partly caused by paralysis in ority to enable governments to take more responsibility in rural WSS following incomplete decentralization of capacity, line with the Paris Declaration and the Accra High Level Fo- funds, and control. Only a small part of public expenditure rum on Aid Effectiveness. Donors should consider rethinking goes to sanitation. As the sanitation sector is dominated by their current policy of financing only “development” expen- 38  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa ditures to include recurrent expenditures. At present, a sub- targeting and execution of WSS public spending. Instead, stantial share of donor funding is spent on the rehabilitation technocratic short-term measures should be complement- of badly maintained infrastructure. Preventive maintenance ed by exploring, exposing, and addressing longer-term driv- is cheaper than periodic rehabilitation. ers that are maintaining the current status quo. The PERs can help to open a debate by showing the lack of efficiency in Closing the implementation gap and improving the efficiency public spending. Advocates for better WSS services can use of public spending will require addressing underlying power the outcomes of the PERs, as well as this regional review, to patterns. This review exposed various drivers that explain lobby for different and better spending patterns. As the qual- the current spending patterns in the WSS sector. Changes in ity of spending improves, the argument for more spending spending patterns risk being marginal or temporary if these will not only be based on the human needs in the sector but issues are not properly addressed. This should not be an ex- will also be grounded in a compelling economic rationale for cuse to pursue a technocratic approach to improving the public spending. Concluding Remarks | 39 Annexes 41 Annex 1 PERs in African Countries with Chapters on Water Supply and Sanitation FY Country Region Type of PER FY03 Mozambique AFR National PER with chapter/volume on water FY03 Tanzania AFR National PER with chapter/volume on water FY03 Uganda AFR National PER with chapter/volume on water FY04 Benin AFR National PER with chapter/volume on water FY04 Ethiopia AFR National PER with chapter/volume on water FY04 Tanzania AFR National PER with chapter/volume on water FY06 Cape Verde AFR National PER with chapter/volume on water FY06 Egypt MNA National PER with chapter/volume on water FY07 Madagascar AFR National PER with chapter/volume on water FY07 Algeria MNA National PER with chapter/volume on water FY08 Cape Verde AFR National PER with chapter/volume on water FY08 Ethiopia AFR National PER with chapter/volume on water FY08 Burkina Faso AFR RWSS PER FY08 Ghana AFR RWSS PER FY08 Mali AFR RWSS PER FY09 Ethiopia AFR RWSS PER FY09 Tanzania AFR Water PER FY09 Mozambique AFR Water PER FY09 Cameron AFR RWSS PER FY09 Côte d’Ivoire AFR RWSS PER FY09 Cross River AFR RWSS PER FY09 Niger AFR RWSS PER FY09 Libya MNA National PER with chapter/volume on water FY10 Central African Republic AFR WSS PER FY10 Congo, Dem. Rep. of AFR WSS PER FY10 Congo, Rep. of AFR WSS PER FY10 Togo AFR WSS PER FY10 Sierra Leone AFR WSS PER 42  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa Annex 2 WSS is more often a sector in country-specific poverty reduction strategy papers (PRSPs), but most of the sample countries had Remarks on Data Limitations problems in preparing budgets and reporting expenditures in their own PRSP classifications. Typical omissions in reporting Growth of WSS access over time has, for most countries sur- are expenditures on WSS facilities in schools, clinics, and hos- veyed,20 proven difficult to determine in the PERs. There are three pitals that are classified to other functions. WSS may include main reasons for this difficulty: wider expenditures on water resource management, envi- ronmental protection, and community development. Sani- 1. No regular monitoring of access to water in rural areas tation may or may not include solid waste removal. has been carried out by government institutions. This limitation is most pronounced in countries in which the Most of the PERs included in this review deal primarily with proj- bulk of WSS expenditure is off-budget (NGO funding, ects and expenditure channeled through central public institu- some bi- and multilateral funding). tions that have WSS as their prime mandate. The sector is served 2. Different data classifications and norms make com- by multiple institutions, both government and non-govern- parison difficult. Regular national household surveys mental, at all levels. It has proven impossible, however, to gain are conducted by different ministries using different, access to data from all institutions and to separate WSS sec- non-conforming sets of classifications and norms for tor expenditure from other expenditure in multi-sector proj- access to drinking water (e.g., potable versus non-pota- ects implemented by other sector ministries (e.g. ministries of ble water sources, improved versus traditional latrines, health). Yet, the case of Burkina Faso shows that WSS expendi- distance and functionality criteria). Classifications used ture incurred by other public institutions can be quite impor- by ministries of health, for instance, are different from tant.22 In the Ghana PER, the government budget estimate is the internationally adopted classifications used by WSS- based on discretionary budget, donor aid, and HIPC resources, sector ministries. As a result, there are huge differences excluding other statutory funds and the spending of internally from one source to another with respect to estimates of generated revenues, though these latter funds and revenues access rates, especially in urban areas. accounted for 17 percent of Ghana’s total budget in 2006. In 3. Comparisons between countries should be treated cau- Mozambique, provincial, municipal, and district allocations tiously, as basic service levels differ from one country that served multiple purposes had to be excluded from the to another. The basic service level for rural water supply analysis, as there was insufficient information to determine in Burkina Faso is, for instance, 500 persons per public how much of their allocations was actually spent on water. standpipe, while the norm in Madagascar is 250 persons per standpipe. The review takes into account donor databases, now govern- ment-maintained in many countries, with the help of the UN, Most PERs apply estimates for access to water supply based in parallel to the budget processes. These databases include on an inventory of existing infrastructure, while for sanitation, off-budget funding, which bypasses sector agencies, and access figures are normally derived from household surveys. includes services funded by many NGOs active in the water Methodologies based on household surveys (e.g. the UNI- sector. Data from donor databases and budgets are often CEF-WHO Joint Monitoring Programme) normally result in inconsistent. Off-budget donor-financing expenditure data higher access rates in urban areas. The main reason for these disparities is that the household surveys include in their cal- culations water fetched from private wells and water resold 20 With the exception of Mali, Cameroon, Madagascar, and Niger. by households with a water connection to neighbors. 21 While WSS is not a distinct sector in COFOG functional expenditure classification it could easily be picked out of the COFOG classification. Data limitations also apply to expenditure data. “Water and The problem rather is that no governments in SSA other than SA re- sanitation” is not a distinct and separate sector according to port on COFOG. The IMF-GFS standard is an economic classification and the UN standard Common Functions of Government (CO- does not allow distinguishing water and sanitation expenditure. Even where there is a start at COFOG classification this is done by retrospec- FOG). Moreover, governments in the sample countries are tively mapping agencies to the classification system. unable to classify their expenditure according to COFOG and 22 The PER for Burkina Faso estimates that 26% of all WSS expenditure only use the IMF-GFS system for the international compari- between 1997 and 2007 was channeled through public institutions son of government finance.21 other than the sector ministry. Annexes | 43 are not consistently available, so actual water sector spend- pendence on foreign aid, they increase misreporting on the ing is likely to be underestimated. Actual WSS expenditures actual use of donor fund. This failure to account for all current are under-stated more than budgeted expenditures, how- costs makes the policy impossible to monitor. ever, so execution rates are also under-stated. The source of funds can be classified as domestic or external, After estimating total WSS expenditure from available sources, with external sources further divided by loan or grant, and by the next hurdle is the distribution of expenditure by sub-sector, donor partner. Tracking donor funds is a problem in all the by geographical area, by capital or current nature, and by source sample countries, because funds often bypass the ministry of funds. of finance and move directly to the beneficiary institutions. This practice has reduced over time but is still a major issue. Two sub-sectors, water and sanitation, combine with a rural Where donors will not disburse to a single treasury account, and urban geographical split to comprise four quadrants for a second-best solution is to ensure the reporting of all such expenditure analysis: urban water, rural water, urban sanita- flows to the center. In some countries, as in Ethiopia, an off- tion, and rural sanitation. The finer the subdivision, the more set policy applies where government transfers to a district expenditure is common to more than one category. Even are reduced by expected aid flows to that district. The offset the primary breakdown between water and sanitation is not policy prompts districts (and donors) to be less transparent always easy to ascertain, as in Sierra Leone, where project ex- in their transactions. This, in turn, makes expenditure data at penditure data are very poor, especially for donor projects. the center less complete. The rural/urban split is sometimes complicated by the use of a third “peri-urban” category and the varying definitions, An attempt to compare the public spending patterns of Sub- even within a single country, of what constitutes a city and Saharan African countries with their peers in other regions but what constitutes a rural area. at a similar level of GDP was abolished because of lack of data. The current data set of 15 PERs in Sub-Saharan Africa, which The capital/current split divides expenditures whose benefits last have been conducted in a relatively short period of time, more than a year from those whose benefits are received during is unique. Twenty WSS PERs were conducted in other parts the year. All the sample countries maintain separate budgets of the world over the same time period, often in countries and accounts for capital (development) expenditure and that were not comparable to Sub-Saharan Africa. Eight of current (recurrent) expenditure. These categories should be the 20 were from in Latin America, five were in the Middle used with some caution, however, as development expen- East and North Africa, five were in Europe and Central Asia diture might include donor-funded operating costs (current region, three were in the East Asia–Pacific region, and one expenditure), while a significant portion of recurrent expen- was in South Asia. However, many of these reviews com- diture may be spent on capacity building, community sensiti- prised only a short chapter or section on water in broader zation, and other expenditures that carry long-term benefits. national PERs, often including irrigation, drainage, and wa- Rehabilitation expenditures are classified as capital or current, ter resources management. We found only three standalone depending on the accounting manual of the funding agency. WSS PERs—Mexico, Egypt, and Lebanon. All three are from Where governments adopt a fiscal policy of funding all cur- middle-income countries that differ considerably from those rent costs from domestic resources, thereby reducing de- included in the current review. 44  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa Annex 3 WSS Budget Execution Rates — 2000–2008 (in percent)   2000 2001 2002 2003 2004 2005 2006 2007 2008 Mean Burkina Faso Total 0.0 64.1 24.2 49.1 54.5 53.3 62.5 0.0 0.0 51.3 Recurrent 0.0 82.5 66.3 61.9 95.3 91.4 97.6 0.0 0.0 82.5 Development 0.0 60.4 20.6 46.4 48.6 46.0 55.7 0.0 0.0 46.3 Cameroon Total 0.0 0.0 98.5 99.3 100.0 100.0 100.0 93.5 0.0 98.6 Recurrent 0.0 0.0 98.1 99.2 100.0 100.0 100.0 95.8 0.0 98.8 Development 0.0 0.0 100.0 99.6 100.0 100.0 0.0 79.2 0.0 95.8 Central African Total 0.0 0.0 51.6 78.5 99.9 76.0 65.2 113.2 59.5 77.7 Republic Recurrent 0.0 0.0 51.6 57.7 99.0 60.8 93.8 91.3 39.8 70.6 Development 0.0 0.0 0.0 100.0 100.0 100.0 62.2 118.0 61.0 77.3 Congo, Dem. Rep. of Total 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Recurrent 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Development 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Congo, Rep. of Total 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Recurrent 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Development 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Côte d’Ivoire Total 0.0 0.0 42.4 25.2 31.2 5.8 46.0 119.9 0.0 45.1 Recurrent 0.0 0.0 60.1 3.1 4.7 4.5 35.0 125.8 0.0 38.9 Development 0.0 0.0 32.4 33.1 39.5 6.1 100.0 99.4 0.0 51.8 Ethiopia Total 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 63.7 63.7 Recurrent 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Development 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Ghana Total 0.0 145.4 49.0 123.4 185.3 42.7 47.0 0.0 0.0 98.8 Recurrent 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Development 0.0 128.3 32.9 116.9 126.2 25.4 44.2 0.0 0.0 79.0 Madagascar Total 58.9 43.1 31.9 26.6 45.7 83.6 0.0 0.0 0.0 48.3 Recurrent 64.7 85.7 55.9 75.0 61.6 69.0 0.0 0.0 0.0 68.7 Development 56.3 16.9 19.1 13.2 35.6 94.7 0.0 0.0 0.0 39.3 Mali Total 0.0 67.0 65.3 44.1 63.3 74.8 63.5 0.0 0.0 63.0 Recurrent 0.0 74.5 81.0 75.0 81.7 97.1 98.1 0.0 0.0 84.6 Development 0.0 65.8 63.3 38.3 60.3 72.9 59.5 0.0 0.0 60.0 Mozambique Total     78.2 244.1 145.3 147.9 93.7 183.4 127.6 145.8 Recurrent 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Development 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Niger Total 0.0 0.0 81.4 30.5 55.7 71.7 63.3 72.9 0.0 62.6 Recurrent 0.0 0.0 81.0 42.5 55.5 68.6 58.7 63.4 0.0 61.6 Development 0.0 0.0 81.5 28.3 55.8 72.5 66.0 77.2 0.0 63.6 Sierra Leone Total     20.6 60.9 80.1 17.6 56.1 28.3 50.8 44.9 Recurrent 0.0 0.0 20.6 54.4 62.2 12.8 39.8 19.9 35.8 35.1 Development 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 (Continued on next page) Annexes | 45 WSS Budget Execution Rates — 2000–2008 (Continued) (in percent)   2000 2001 2002 2003 2004 2005 2006 2007 2008 Mean Tanzania   Total 0.0 82.9 88.5 18.8 63.4 86.0 73.4 85.4 55.4 69.2 Recurrent 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Development 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Togo  Total 0.0 0.0 9.7 22.9 17.6 71.1 101.9 28.5 54.9 43.8 Recurrent 0.0 0.0 73.4 75.0 83.0 107.8 94.3 80.3 92.4 86.6 Development 0.0 0.0 4.7 18.9 14.7 63.3 116.0 0.0 45.7 43.9 Total    58.9 80.5 50.0 50.5 66.6 62.0 68.2 80.8 56.6 62.8 Recurrent 64.7 80.9 65.3 60.4 71.5 68.0 77.2 79.4 56.0 69.7 Development 56.3 67.8 44.3 55.0 64.5 64.5 71.9 74.8 53.3 61.9 Source: Country WS&S PERs and authors calculations 46  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa Annex 4 World Bank. 2009. Cote d’Ivoire: Revue des Dépenses Pub- liques Secteur de l’Eau et de l’Assainissement en Milieu References Rural. Africa Region, World Bank. The 15 Public Expenditure Reviews World Bank. 2009. Ethiopia: Public Expenditure Review. World Bennett, Anthony, Darrell Thompson and Meike van Ginneken. Bank, Washington DC. 2011. Sierra Leone: Public Expenditure Review for Water World Bank. 2009. Niger: Revue des Dépenses Publiques and Sanitation 2002–2009. World Bank, Washington, DC. Secteur de l’Eau et de l’Assainissement en Milieu Rural Jacobsen, Carl Christian, and Meike van Ginneken, 2011. Ré- (2002–2007). Africa Region, World Bank. publique Centrafricaine: Revue des Dépenses Publiques World Bank. 2010. Mozambique: Public Expenditure Review of Approvisionnement en Eau Potable et Assainissement, the Water Sector Africa Region, World Bank. 2002–2008. World Bank, Washington, DC. Jacobsen, Carl Christian. Meike van Ginneken, Michel Ananou Other References Edorh and Kinvi Attiogbé. 2011. Togo: Revue des Dépens- AMCOW. 2011. Pathways to Progress: Transitioning to Coun- es Publiques Approvisionnement en Eau Potable, As- try-Led Service Delivery Pathways to Meet Africa’s Water sainissement et Gestion des Déchets Solides 2002–2008. Supply and Sanitation Targets, African Ministers Council Washington DC, World Bank. on Water and Water and Sanitation Program. Netterstrom, Ulrik, Gilles Gauvreau, Bernard Massamba and Bakalian, Alexander, and Wendy Wakeman (editors), 2009. Rufin Mantsie. 2011. République du Congo: Revue des Post-Construction Support and Sustainability in Com- Dépenses Publiques Approvisionnement en Eau Potable munity-Managed Rural Water Supply (Case Studies in et Assainissement 2002–2008. World Bank, Washington, Peru, Bolivia and Ghana). World Bank, Washington, DC. DC. Banerjee, Sudeshna, and Elvira Morella. 2011. Africa’s Water Netterstrom, Ulrik, Gilles Gauvreau, Gabriel Kataliko and Jona- and Sanitation Infrastructure—Access, Affordability, and than Mwimba. 2011. République Démocratique du Con- Alternatives. Africa Region, World Bank. go: Revue des Dépenses Publiques Approvisionnement en Eau Potable et Assainissement 2002–2008, World Banerjee, Sudeshna, Vivien Foster, Yvonne Ying, Heather Skill- Bank, Washington, DC. ing, and Quentin Wodon. 2010. Cost Recovery, Equity, and Efficiency in Water Tariffs: Evidence from African Util- Van den Berg, Caroline, Eileen Burke, Leonard Chacha and Flo- ities. AICD Working Paper 7, World Bank, Washington, DC. ra Kessy. 2009. Tanzania: Public Expenditure Review of the Water Sector. World Bank, Washington DC. Briceño-Garmendia, Cecilia, Karlis Smits, and Vivien Foster. 2009. Financing Public Infrastructure in Sub-Saharan World Bank, 2008. Burkina Faso: Public Expenditures Review Africa: Patterns and Emerging Issues. AICD Background Rural Water and Sanitation Sector. Africa Region, World Paper 15, Africa Region, World Bank, Washington, DC. Bank. Foster, Vivien, and Cecilia Briceño-Garmendia, eds. 2009. Af- World Bank, 2008. Mali, Revue de Dépenses Publiques Secteur rica’s Infrastructure: A Time for Transformation. Paris and de l’Eau et de l’Assainissement en Milieu Rural. Africa Re- Washington, DC: Agence Française de Développement gion, World Bank. and World Bank. World Bank, 2009. Cameroun: Revue des Dépenses Publiques Komives, Kristin, Vivien Foster, Jonathan Halpern, and Quen- Secteur de l’Eau et de l’Assainissement en Milieu Rural. tin Wodon. 2005. Water, Electricity, and the Poor. Who Africa Region, World Bank. Benefits from Utility Subsidies? Washington, DC: The World Bank. 2007. Madagascar: Revue de Dépenses Publiques World Bank. Volume 4: Eau et Assainissement. Washington DC. World Lockwood, Harold, and Stef Smits. 2011. Lessons for Rural Bank. Water Supply: Moving toward a Service Delivery Ap- World Bank. 2008. Ghana: Public Expenditures Review Rural proach. IRC International Water and Sanitation Centre, Water and Sanitation Sector. Africa Region, World Bank. The Hague, and Aguaconsult, Wivenhoe. Annexes | 47 Muller, Mike, Robin Simpson, and Meike van Ginneken. 2008. United Nations Development Programme. 2006. Human De- Consumers Count: How Water and Sanitation Utilities velopment Report 2006—Beyond Scarcity: Power, Poverty Can Become More Accountable to Their Users. World and the Global Water Crisis. New York. Bank, Washington, DC. World Bank. 2003. World Development Report 2004. Wash- Muller, Mike. 2007. Parish pump politics: the politics of water ington, DC: Oxford University Press and The World Bank. supply in South Africa. Progress in Development Studies 7, World Bank. 2010. Decentralization, Democracy and Devel- 1 pp. 33–45. opment: Recent Experience from Sierra Leone. Ed. by OECD. 2008. Creditor Reporting System, Aid Activities in Sup- Yongmei Zhou. World Bank. Washington DC. port of Water Supply and Sanitation, 2001–2006. Paris. WSP Sanitation Global Practice Team. 2011. The Political RWSN (Rural Water Supply Network). 2009. Myths of the Rural Economy of Sanitation: How can we increase invest- Water Supply Sector, Perspectives no. 4. St Gallen. ment and improve service for the poor? Operational experiences from case studies in Brazil, India, Indonesia, and Senegal. Water and Sanitation Program. 48  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa Annex 5 Annexes | 49 Detailed Data sheets for Countries included in the Review Burkina Faso No. Indicators Unit 2000 2001 WS&S Public Overall Trends 1 Sector Budget Allocation (share of total general gvt. budget) % 0.0 2.9 2 Sector Budget Allocation (share of GDP) % 0.0 0.7 3 Domestic sector allocations (share of total sector public allocations) % 0.0 16.9 4 External allocations (share of total sector public allocations) % 0.0 83.1 5 Sector allocations transferred to sub-national levels (share of total sector allocations) % 0.0 0.0 6 WS&S spending per capita US$ 0.0 1.3 Access to WS&S Services 7 Overall access to water % 8 Access to water – Urban areas % 9 Access to water – Rural areas % 46.0 10 Overall access to sanitation % 11 Access to sanitation – Urban areas % 12 Access to sanitation – Rural areas % 13 Gap between province/region with highest & lowest access rate to water % point WS&S Sector Spending Performance 14 WSS Sector Budget Execution Rate % 0.0 64.1 15 Execution of WS&S sector domestic resources % 0.0 82.5 16 Execution of WS&S sector external resources % 0.0 60.4 17 Recurrent spending (share of total sector spending) % 0.0 1.9 18 Investment spending (share of total sector spending) % 0.0 98.1 19 Salary costs (share of total recurrent spending) % 0.0 73.1 20 Rural investment Spending (share of total WS&S investment spending) % 0.0 100.0 21 Urban investment Spending (share of total WS&S investment spending) % 0.0 0.0 Notes to Indicator Data Ind. No. Year(s) Comment 20+21 Rural Sector only 50  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa 2002 2003 2004 2005 2006 2007 2008 2009 2.5 1.7 1.1 1.8 2.9 3.1 0.0 0.0 0.6 0.4 0.3 0.5 0.8 0.9 0.0 0.0 8.0 17.8 12.7 16.1 16.3 12.6 0.0 0.0 92.0 82.2 87.3 83.9 83.7 87.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.5 0.9 0.8 1.5 3.0 0.0 0.0 0.0 69.2 64.2 60.0 20.3 38.0 24.2 49.1 54.5 53.3 62.5 0.0 0.0 0.0 66.3 61.9 95.3 91.4 97.6 0.0 0.0 0.0 20.6 46.4 48.6 46.0 55.7 0.0 0.0 0.0 4.7 4.8 3.6 2.4 1.3 0.0 0.0 0.0 95.3 95.2 96.4 97.6 98.7 0.0 0.0 0.0 86.8 64.0 81.2 86.9 72.3 0.0 0.0 0.0 100.0 100.0 100.0 100.0 100.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Annexes | 51 Memo Unit 2000 2001 Total population inhbt 9,085,326 GDP Loc. Curr. 2,062,000,000,000 GDP US$ 0 2,812,949,088 General gvt. budget Loc. Curr. 461300000000 General gvt. budget US$ 0 629,298,455 Total sector budget allocations Loc. Curr. 0 13,466,000,000 Total sector budget allocations US$ 0 18,370,113 Domestic sector budget allocations Loc. Curr. 2,270,000,000 Domestic sector budget allocations US$ 0 3,096,700 External sector budget allocations Loc. Curr. 11,196,000,000 External sector budget allocations US$ 0 15,273,413 WS&S transfers to sub-national levels Loc. Curr. 0 WS&S transfers to sub-national levels US$ 0 0 Executed sector budget Loc. Curr. 0 8,631,149,000 Executed sector budget US$ 0 11,774,482 Executed domestic sector budget Loc. Curr. 1,872,789,000 Executed domestic sector budget US$ 0 2,554,830 Executed external sector budget Loc. Curr. 6,758,360,000 Executed external sector budget US$ 0 9,219,652 Executed recurrent budget Loc. Curr. 161,164,000 Executed recurrent budget US$ 0 219,857 Executed investment budget Loc. Curr. 8,469,985,000 Executed investment budget US$ 0 11,554,625 Executed salary costs Loc. Curr. 117,822,000 Executed salary costs US$ 0 160,731 Rural Investments Loc. Curr. 8,469,985,000 Rural Investments US$ 0 11,554,625 Actual WSS expenditure/GDP % 0.0% 0.42 Share of exp that is externally funded % 0.0% 78.3% Recurrent exp per capita $ 0.02 Domestic-funded sector budget execution % 0 0.83 External-funded budget execution % 0.60 52  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa 2002 2003 2004 2005 2006 2007 2008 2009 9,270,741 9,459,940 9,653,000 9,850,000 10,047,000 10,247,940 2,293,000,000,000 2,482,000,000,000 2,698,000,000,000 2,961,000,000,000 3,162,000,000,000 3,451,000,000,000 3,289,869,166 4,270,472,573 5,107,093,659 5,613,609,163 6,047,159,703 7,200,582,484 0 0 528200000000 581100000000 673800000000 806800000000 864100000000 960700000000 757,832,051 999,827,402 1,275,448,372 1,529,571,048 1,652,546,078 2,004,520,311 0 0 13,125,000,000 9,611,000,000 7,672,000,000 14,431,000,000 24,886,000,000 30,256,000,000 0 0 18,831,022 16,536,467 14,522,469 27,358,998 47,593,174 63,129,766 0 0 1,045,000,000 1,711,000,000 974,000,000 2,327,000,000 4,053,000,000 3,799,000,000 1,499,308 2,943,908 1,843,702 4,411,641 7,751,151 7,926,692 0 0 12,080,000,000 7,900,000,000 6,698,000,000 12,104,000,000 20,833,000,000 26,457,000,000 17,331,714 13,592,560 12,678,767 22,947,357 39,842,023 55,203,075 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3,180,781,000 4,721,307,000 4,180,798,000 7,696,987,000 15,554,367,000 0 0 0 4,563,608 8,123,373 7,913,909 14,592,326 29,746,914 0 0 0 693,120,000 1,059,467,000 928,028,000 2,127,572,000 3,956,500,000 994,450 1,822,895 1,756,681 4,033,555 7,566,599 0 0 0 2,487,661,000 3,661,840,000 3,252,770,000 5,569,415,000 11,597,867,000 3,569,158 6,300,478 6,157,228 10,558,770 22,180,314 0 0 0 150,788,000 227,457,000 151,338,000 187,762,000 198,410,000 216,342 391,357 286,470 355,968 379,449 0 0 0 3,029,993,000 4,493,850,000 4,029,460,000 7,509,225,000 15,355,957,000 4,347,266 7,732,016 7,627,439 14,236,357 29,367,465 0 0 0 130,914,000 145,460,000 122,880,000 163,248,000 143,480,000 187,828 250,275 232,602 309,494 274,398 0 0 0 3,029,993,000 4,493,850,000 4,029,460,000 7,509,225,000 15,355,957,000 4,347,266 7,732,016 7,627,439 14,236,357 29,367,465 0 0 0 0.14 0.19 0.15 0.26 0.49 0.00 0.00 0.00 78.2% 77.6% 77.8% 72.4% 74.6% 0.0% 0.0% 0.0% 0.02 0.04 0.03 0.04 0.04 0.66 0.62 0.95 0.91 0.98 0.00 0.21 0.46 0.49 0.46 0.56 0.00 Annexes | 53 Cameroun No. Indicators Unit 2000 2001 WS&S Public Overall Trends 1 Sector Budget Allocation (share of total general gvt. budget) % 0.0 0.0 2 Sector Budget Allocation (share of GDP) % 0.0 0.0 3 Domestic sector allocations (share of total sector public allocations) % 0.0 0.0 4 External allocations (share of total sector public allocations) % 0.0 0.0 5 Sector allocations transferred to sub-national levels (share of total sector allocations) % 0.0 0.0 6 WS&S spending per capita US$ 0.0 0.0 Access to WS&S Services 7 Overall access to water % 8 Access to water – Urban areas % 9 Access to water – Rural areas % 10 Overall access to sanitation % 11 Access to sanitation – Urban areas % 12 Access to sanitation – Rural areas % 13 Gap between province/region with highest & lowest access rate to water % point WS&S Sector Spending Performance 14 WSS Sector Budget Execution Rate % 0.0 0.0 15 Execution of WS&S sector domestic resources % 0.0 0.0 16 Execution of WS&S sector external resources % 0.0 0.0 17 Recurrent spending (share of total sector spending) % 0.0 0.0 18 Investment spending (share of total sector spending) % 0.0 0.0 19 Salary costs (share of total recurrent spending) % 0.0 0.0 20 Rural investment Spending (share of total WS&S investment spending) % 0.0 0.0 21 Urban investment Spending (share of total WS&S investment spending) % 0.0 0.0 Notes to Indicator Data Ind. No. Year(s) Comment 20+21 Rural Sector only 54  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa 2002 2003 2004 2005 2006 2007 2008 2009 1.3 0.7 0.4 0.7 0.3 0.5 0.0 0.0 0.2 0.1 0.1 0.1 0.0 0.1 0.0 0.0 78.4 60.3 51.5 49.1 100.0 86.4 0.0 0.0 21.6 39.7 48.5 50.9 0.0 13.6 0.0 0.0 1.2 1.9 2.5 1.4 2.9 1.2 0.0 0.0 2.2 1.6 1.3 2.3 1.1 2.7 0.0 0.0 32.0 34.0 36.0 39.0 40.0 45.0 13.5 30.0 71.0 98.5 99.3 100.0 100.0 100.0 93.5 0.0 0.0 98.1 99.2 100.0 100.0 100.0 95.8 0.0 0.0 100.0 99.6 100.0 100.0 0.0 79.2 0.0 0.0 5.1 8.3 9.8 4.4 12.6 4.4 0.0 0.0 94.9 91.7 90.2 95.6 87.4 95.6 0.0 0.0 66.6 65.2 57.7 69.5 63.6 74.4 0.0 0.0 100.0 100.0 100.0 100.0 100.0 100.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Annexes | 55 Memo Unit 2000 2001 Total population inhbt GDP Loc. Curr. GDP US$ 0 0 General gvt. budget Loc. Curr. General gvt. budget US$ 0 0 Total sector budget allocations Loc. Curr. 0 0 Total sector budget allocations US$ 0 0 Domestic sector budget allocations Loc. Curr. Domestic sector budget allocations US$ 0 0 External sector budget allocations Loc. Curr. External sector budget allocations US$ 0 0 WS&S transfers to sub-national levels Loc. Curr. WS&S transfers to sub-national levels US$ 0 0 Executed sector budget Loc. Curr. 0 0 Executed sector budget US$ 0 0 Executed domestic sector budget Loc. Curr. Executed domestic sector budget US$ 0 0 Executed external sector budget Loc. Curr. Executed external sector budget US$ 0 0 Executed recurrent budget Loc. Curr. Executed recurrent budget US$ 0 0 Executed investment budget Loc. Curr. Executed investment budget US$ 0 0 Executed salary costs Loc. Curr. Executed salary costs US$ 0 0 Rural Investments Loc. Curr. Rural Investments US$ 0 0 Actual WSS expenditure/GDP % 0.00 0.00 Share of exp that is externally funded % 0.0% 0.0% Recurrent exp per capita $ Domestic-funded sector budget execution % 0 0.00 External-funded budget execution % 0.00 56  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa 2002 2003 2004 2005 2006 2007 2008 2009 7,778,000 7,847,000 7,914,000 8,026,560 8,137,910 8,247,843 7,383,000,000,000 7,917,000,000,000 8,334,000,000,000 8,901,000,000,000 9,581,000,000,000 9,894,000,000,000 10,592,718,731 13,621,809,574 15,775,581,378 16,874,952,774 18,323,161,643 20,644,034,512 0 0 933,000,000,000 1,098,000,000,000 1,277,000,000,000 1,424,000,000,000 1,861,000,000,000 2,251,000,000,000 1,338,616,630 1,889,193,749 2,417,256,710 2,699,689,108 3,559,065,214 4,696,757,801 0 0 11,839,000,000 7,244,000,000 5,520,000,000 9,771,000,000 4,747,000,000 11,522,000,000 0 0 16,985,940 12,463,861 10,448,909 18,524,341 9,078,389 24,040,890 0 0 9,276,000,000 4,371,000,000 2,842,000,000 4,798,000,000 4,747,000,000 9,958,000,000 13,308,690 7,520,643 5,379,674 9,096,284 9,078,389 20,777,572 0 0 2,563,000,000 2,873,000,000 2,678,000,000 4,973,000,000 0 1,564,000,000 3,677,250 4,943,218 5,069,235 9,428,058 0 3,263,318 0 0 137,000,000 137,000,000 137,000,000 137,000,000 137,000,000 137,000,000 196,560 235,719 259,330 259,731 262,005 285,853 0 0 11,665,000,000 7,196,000,000 5,520,000,000 9,771,000,000 4,747,000,000 10,775,000,000 0 0 16,736,295 12,381,273 10,448,909 18,524,341 9,078,389 22,482,259 0 0 9,102,000,000 4,335,000,000 2,842,000,000 4,798,000,000 4,747,000,000 9,536,000,000 13,059,045 7,458,702 5,379,674 9,096,284 9,078,389 19,897,060 0 0 2,563,000,000 2,861,000,000 2,678,000,000 4,973,000,000 0 1,239,000,000 3,677,250 4,922,571 5,069,235 9,428,058 2,585,199 0 0 593,000,000 597,000,000 539,000,000 429,000,000 599,000,000 472,000,000 850,803 1,027,185 1,020,283 813,319 1,145,556 984,838 0 0 11,072,000,000 6,599,000,000 4,981,000,000 9,342,000,000 4,148,000,000 10,303,000,000 15,885,491 11,354,089 9,428,626 17,711,022 7,932,833 21,497,421 0 0 395,000,000 389,000,000 311,000,000 298,000,000 381,000,000 351,000,000 566,724 669,305 588,698 564,963 728,643 732,369 0 0 11,072,000,000 6,599,000,000 4,981,000,000 9,342,000,000 4,148,000,000 10,303,000,000 15,885,491 11,354,089 9,428,626 17,711,022 7,932,833 21,497,421 0 0 0.16 0.09 0.07 0.11 0.05 0.11 0.00 0.00 22.0% 39.8% 48.5% 50.9% 0.0% 11.5% 0.0% 0.0% 0.11 0.13 0.13 0.10 0.14 0.12 0.98 0.99 1.00 1.00 1.00 0.96 1.00 1.00 1.00 1.00 0.79 Annexes | 57 Central African Republic No. Indicators Unit 2000 2001 2002 WS&S Public Overall Trends 1 Sector Budget Allocation (share of total general gvt. budget) % 0.0 0.0 0.2 2 Sector Budget Allocation (share of GDP) % 0.0 0.0 0.0 3 Domestic sector allocations (share of total sector public allocations) % 0.0 0.0 100.0 4 External allocations (share of total sector public allocations) % 0.0 0.0 0.0 5 Sector allocations transferred to sub-national levels (share of total sector allocations) % 0.0 0.0 0.0 6 WS&S spending per capita US$ 0.0 0.0 0.1 Access to WS&S Services 7 Overall access to water % 8 Access to water – Urban areas % 9 Access to water – Rural areas % 17.7 10 Overall access to sanitation % 11 Access to sanitation – Urban areas % 12 Access to sanitation – Rural areas % 13 Gap between province/region with highest & lowest access rate to water % point WS&S Sector Spending Performance 14 WSS Sector Budget Execution Rate % 0.0 0.0 51.6 15 Execution of WS&S sector domestic resources % 0.0 0.0 51.6 16 Execution of WS&S sector external resources % 0.0 0.0 0.0 17 Recurrent spending (share of total sector spending) % 0.0 0.0 48.6 18 Investment spending (share of total sector spending) % 0.0 0.0 51.4 19 Salary costs (share of total recurrent spending) % 0.0 0.0 97.8 20 Rural investment Spending (share of total WS&S investment spending) % 0.0 0.0 0.0 21 Urban investment Spending (share of total WS&S investment spending) % 0.0 0.0 100.0 Notes to Indicator Data Ind. No. Year(s) Comment No details on per capita cost per year 58  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa 2003 2004 2005 2006 2007 2008 2009 0.4 0.7 0.3 1.2 1.0 4.3 0.0 0.1 0.1 0.0 0.2 0.2 0.7 0.0 50.9 13.7 61.1 9.3 17.9 6.9 0.0 49.1 86.3 38.9 90.7 82.1 93.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.2 0.4 0.1 0.5 0.7 2.0 0.0 25.0 28.0 17.6 17.3 17.0 16.8 16.9 32.0 66.0 43.0 78.5 99.9 76.0 65.2 113.2 59.5 0.0 57.7 99.0 60.8 93.8 91.3 39.8 0.0 100.0 100.0 100.0 62.2 118.0 61.0 0.0 21.4 9.7 28.5 7.7 5.3 2.3 0.0 78.6 90.3 86.1 92.3 98.6 112.4 0.0 98.7 98.2 98.8 98.5 98.5 88.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 100.0 100.0 100.0 100.0 100.0 100.0 0.0 Annexes | 59 Memo Unit 2000 2001 2002 Total population inhbt 3,745,000 3,820,000 3,890,000 GDP Loc. Curr. 612,295,517,000 GDP US$ 878,487,633 General gvt. budget Loc. Curr. 146,344,917,000 General gvt. budget US$ 209,967,567 Total sector budget allocations Loc. Curr. 277,528,000 Total sector budget allocations US$ 398,182 Domestic sector budget allocations Loc. Curr. 277,528,000 Domestic sector budget allocations US$ 398,182 External sector budget allocations Loc. Curr. 0 External sector budget allocations US$ 0 WS&S transfers to sub-national levels Loc. Curr. WS&S transfers to sub-national levels US$ 0 Executed sector budget Loc. Curr. 0 0 143,173,000 Executed sector budget US$ 205,417 Executed domestic sector budget Loc. Curr. 143,173,000 Executed domestic sector budget US$ 205,417 Executed external sector budget Loc. Curr. 0 Executed external sector budget US$ 0 Executed recurrent budget Loc. Curr. 69,528,000 Executed recurrent budget US$ 99,755 Executed investment budget Loc. Curr. 73,645,000 Executed investment budget US$ 105,662 Executed salary costs Loc. Curr. 68,028,000 Executed salary costs US$ 97,603 Rural Investments Loc. Curr. Rural Investments US$ 0 Actual WSS expenditure/GDP % 0.00 0.00 0.02 Share of exp that is externally funded % 0.0% 0.0% 0.0% Recurrent exp per capita $ 0.03 Domestic-funded sector budget execution % 0 0.00 0.52 External-funded budget execution % 0.00 0.00 60  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa 2003 2004 2005 2006 2007 2008 2009 3,959,000 4,029,000 4,101,000 4,178,000 4,257,000 4,339,000 651,378,210,000 669,745,370,000 711,727,450,000 772,076,690,000 821,274,440,000 877,687,860,000 1,120,746,487 1,267,773,289 1,349,327,840 1,476,556,309 1,713,606,012 1,959,976,684 113,654,731,000 105,458,540,000 110,571,150,000 129,345,865,000 136,818,300,000 150,949,146,000 195,551,737 199,624,404 209,626,214 247,367,205 285,474,197 337,086,589 471,704,000 753,061,000 352,288,000 1,612,977,000 1,341,092,000 6,501,214,000 811,603 1,425,483 667,885 3,084,734 2,798,216 14,517,949 240,204,000 103,061,000 215,288,000 150,654,000 240,654,000 445,654,000 413,290 195,086 408,154 288,118 502,130 995,196 231,500,000 650,000,000 137,000,000 1,462,323,000 1,100,438,000 6,055,560,000 398,314 1,230,397 259,731 2,796,616 2,296,087 13,522,753 0 0 0 0 0 0 370,104,000 752,061,000 267,801,000 1,051,232,000 1,517,984,000 3,869,520,000 636,792 1,423,590 507,710 2,010,426 3,167,305 8,641,078 138,604,000 102,061,000 130,801,000 141,287,000 219,757,000 177,395,000 238,479 193,193 247,979 270,204 458,528 396,143 231,500,000 650,000,000 137,000,000 909,945,000 1,298,227,000 3,692,125,000 398,314 1,230,397 259,731 1,740,222 2,708,777 8,244,934 79,204,000 73,061,000 76,288,000 80,654,000 80,654,000 90,129,000 136,277 138,299 144,631 154,247 168,286 201,268 290,900,000 679,000,000 230,511,000 970,578,000 1,496,763,000 4,350,523,000 500,516 1,285,292 437,014 1,856,180 3,123,027 9,715,212 78,154,000 71,768,000 75,388,000 79,454,000 79,454,000 79,454,000 134,470 135,851 142,924 151,952 165,782 177,430 0 0 0 0 0 0 0.06 0.11 0.04 0.14 0.18 0.44 0.00 62.5% 86.4% 51.2% 86.6% 85.5% 95.4% 0.0% 0.03 0.03 0.04 0.04 0.04 0.05 0.58 0.99 0.61 0.94 0.91 1.00 1.00 1.00 1.18 Annexes | 61 Cote d’Ivoire No. Indicators Unit 2000 2001 2002 WS&S Public Overall Trends 1 Sector Budget Allocation (share of total general gvt. budget) % 0.0 0.0 0.5 2 Sector Budget Allocation (share of GDP) % 0.0 0.0 0.1 3 Domestic sector allocations (share of total sector public allocations) % 0.0 0.0 33.8 4 External allocations (share of total sector public allocations) % 0.0 0.0 66.2 5 Sector allocations transferred to sub-national levels (share of total sector allocations) % 0.0 0.0 0.2 6 WS&S spending per capita US$ 0.00 0.00 0.68 Access to WS&S Services 7 Overall access to water % 8 Access to water – Urban areas % 9 Access to water – Rural areas % 10 Overall access to sanitation % 11 Access to sanitation – Urban areas % 12 Access to sanitation – Rural areas % 13 Gap between province/region with highest & lowest access rate to water % point WS&S Sector Spending Performance 14 WSS Sector Budget Execution Rate % 0.0 0.0 42.4 15 Execution of WS&S sector domestic resources % 0.0 0.0 60.1 16 Execution of WS&S sector external resources % 0.0 0.0 33.4 17 Recurrent spending (share of total sector spending) % 0.0 0.0 1.2 18 Investment spending (share of total sector spending) % 0.0 0.0 98.8 19 Salary costs (share of total recurrent spending) % 0.0 0.0 38.3 20 Rural investment Spending (share of total WS&S investment spending) % 0.0 0.0 100.0 21 Urban investment Spending (share of total WS&S investment spending) % 0.0 0.0 0.0 Notes to Indicator Data Ind. No. Year(s) Comment 3 Inconsistency in figures given in different tables on sector allocations 20+21 Rural Sector only 62  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa 2003 2004 2005 2006 2007 2008 2009 0.4 0.3 0.5 0.1 0.1 0.4 0.0 0.1 0.1 0.1 0.0 0.0 0.1 0.0 26.2 23.8 20.5 83.0 77.6 23.8 0.0 73.8 76.2 79.5 17.0 22.4 76.2 0.0 0.5 0.6 0.3 1.0 0.9 0.0 0.0 0.31 0.31 0.08 0.18 0.53 0.00 0.00 76.0 90.0 65.0 37.0 26.0 25.2 31.2 5.7 46.0 119.9 0.0 0.0 3.1 4.7 4.2 35.0 125.8 0.0 0.0 33.1 39.5 6.1 100.0 99.4 0.0 0.0 3.2 3.5 12.2 5.6 2.5 0.0 0.0 96.8 96.5 87.8 94.4 97.5 0.0 0.0 37.8 39.4 43.7 46.4 39.3 0.0 0.0 100.0 100.0 100.0 100.0 100.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Annexes | 63 Memo Unit 2000 2001 2002 Total population inhbt 9,491,168 GDP Loc. Curr. 8,006,000,000,000 GDP US$ 0 0 11,486,564,562 General gvt. budget Loc. Curr. 1,942,000,000,000 General gvt. budget US$ 0 0 2,786,273,842 Total sector budget allocations Loc. Curr. 0 0 10,558,122,000 Total sector budget allocations US$ 0 0 15,148,208 Domestic sector budget allocations Loc. Curr. 3,567,677,000 Domestic sector budget allocations US$ 0 0 5,118,705 External sector budget allocations Loc. Curr. 6,990,445,000 External sector budget allocations US$ 0 0 10,029,503 WS&S transfers to sub-national levels Loc. Curr. 19,600,000 WS&S transfers to sub-national levels US$ 0 0 28,121 Executed sector budget Loc. Curr. 0 0 4,478,128,000 Executed sector budget US$ 0 0 6,424,970 Executed domestic sector budget Loc. Curr. 2,145,438,000 Executed domestic sector budget US$ 0 0 3,078,155 Executed external sector budget Loc. Curr. 2,332,690,000 Executed external sector budget US$ 0 0 3,346,814 Executed recurrent budget Loc. Curr. 52,633,000 Executed recurrent budget US$ 0 0 75,515 Executed investment budget Loc. Curr. 4,425,495,000 Executed investment budget US$ 0 0 6,349,455 Executed salary costs Loc. Curr. 20,160,000 Executed salary costs US$ 0 0 28,924 Rural Investments Loc. Curr. 4,425,495,000 Rural Investments US$ 0 0 6,349,455 Actual WSS expenditure/GDP % 0.00 0.00 0.06 Share of exp that is externally funded % 0.0% 0.0% 52.1% Recurrent exp per capita $ 0.01 Domestic-funded sector budget execution % 0 0.00 0.60 External-funded budget execution % 0.00 0.33 64  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa 2003 2004 2005 2006 2007 2008 2009 9,684,866 9,882,516 10,084,200 10,290,000 10,500,000 7,984,000,000,000 8,179,000,000,000 8,631,000,000,000 9,081,000,000,000 9,487,000,000,000 10,238,000,000,000 13,737,088,246 15,482,179,037 16,363,073,519 17,366,937,781 19,794,820,640 22,862,616,892 0 2,002,000,000,000 1,986,000,000,000 1,734,000,000,000 1,948,000,000,000 1,920,000,000,000 2,129,000,000,000 3,444,595,525 3,759,335,807 3,287,402,327 3,725,448,166 4,006,119,493 4,754,298,824 0 7,029,311,000 5,208,315,000 7,860,958,000 2,117,601,000 2,242,077,000 8,013,434,000 0 12,094,472 9,858,915 14,903,190 4,049,801 4,678,140 17,894,908 0 1,844,380,000 1,239,774,000 1,613,263,000 1,758,103,000 1,740,035,000 1,905,736,000 3,173,398 2,346,791 3,058,503 3,362,280 3,630,619 4,255,725 0 5,184,931,000 3,968,541,000 6,247,695,000 359,498,000 502,042,000 6,107,698,000 8,921,074 7,512,124 11,844,687 687,521 1,047,521 13,639,183 0 33,100,000 30,000,000 21,200,000 21,200,000 21,200,000 0 56,951 56,788 40,192 40,544 44,234 0 0 1,772,960,000 1,626,874,000 448,393,000 973,991,000 2,688,454,000 0 0 3,050,515 3,079,540 850,085 1,862,707 5,609,515 0 0 57,051,000 57,922,000 67,864,000 614,493,000 2,189,309,000 98,161 109,642 128,660 1,175,186 4,568,038 0 0 1,715,909,000 1,568,952,000 380,529,000 359,498,000 499,145,000 2,952,354 2,969,898 721,426 687,521 1,041,476 0 0 56,488,000 57,576,000 54,875,000 54,825,000 68,134,000 97,192 108,987 104,035 104,850 142,163 0 0 1,716,472,000 1,569,298,000 393,518,000 919,166,000 2,620,320,000 2,953,323 2,970,553 746,051 1,757,857 5,467,352 0 0 21,360,000 22,680,000 24,000,000 25,440,000 26,760,000 36,752 42,931 45,500 48,653 55,835 0 0 1,716,472,000 1,569,298,000 393,518,000 919,166,000 2,620,320,000 2,953,323 2,970,553 746,051 1,757,857 5,467,352 0 0 0.02 0.02 0.01 0.01 0.03 0.00 0.00 96.8% 96.4% 84.9% 36.9% 18.6% 0.0% 0.0% 0.01 0.01 0.01 0.01 0.01 0.00 0.03 0.05 0.04 0.35 1.26 0.33 0.40 0.06 1.00 0.99 Annexes | 65 Ethiopia No. Indicators Unit 2000 2001 WS&S Public Overall Trends 1 Sector Budget Allocation (share of total general gvt. budget) % 0.0 0.0 2 Sector Budget Allocation (share of GDP) % 0.0 0.0 3 Domestic sector allocations (share of total sector public allocations) % 0.0 0.0 4 External allocations (share of total sector public allocations) % 0.0 0.0 5 Sector allocations transferred to sub-national levels (share of total sector allocations) % 0.0 0.0 6 WS&S spending per capita US$ 0.0 0.0 Access to WS&S Services 7 Overall access to water % 8 Access to water – Urban areas % 9 Access to water – Rural areas % 10 Overall access to sanitation % 11 Access to sanitation – Urban areas % 12 Access to sanitation – Rural areas % 13 Gap between province/region with highest & lowest access to water % point WS&S Sector Spending Performance 14 WSS Sector Budget Execution Rate % 0.0 0.0 15 Execution of WS&S sector domestic resources % 0.0 0.0 16 Execution of WS&S sector external resources % 0.0 0.0 17 Recurrent spending (share of total sector spending) % 0.0 0.0 18 Investment spending (share of total sector spending) % 0.0 0.0 19 Salary costs (share of total recurrent spending) % 0.0 0.0 20 Rural investment Spending (share of total WS&S investment spending) % 0.0 0.0 21 Urban investment Spending (share of total WS&S investment spending) % 0.0 0.0 Notes to Indicator Data Ind. No. Year(s) Comment 7 2008 Per JMP 2008. Sector adm data say 53.9% (PASDEP 2008) 66  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa 2002 2003 2004 2005 2006 2007 2008 2009 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 0.0 0.4 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 0.0 0.0 100.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.0 0.0 31.9 28.0 0.0 0.0 0.0 0.0 0.0 0.0 63.7 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 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 35.0 41.0 46.0 48.0 52.0 51.0 53.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Annexes | 67 Memo Unit 2000 2001 Total population inhbt GDP Loc. Curr. GDP US$ 0 7,887,600,000 General gvt. budget Loc. Curr. General gvt. budget US$ 0 0 Total sector budget allocations Loc. Curr. 0 0 Total sector budget allocations US$ 0 0 Domestic sector budget allocations Loc. Curr. Domestic sector budget allocations US$ 0 0 External sector budget allocations Loc. Curr. External sector budget allocations US$ 0 0 WS&S transfers to sub-national levels Loc. Curr. WS&S transfers to sub-national levels US$ 0 0 Executed sector budget Loc. Curr. 0 0 Executed sector budget US$ 0 0 Executed domestic sector budget Loc. Curr. Executed domestic sector budget US$ 0 0 Executed external sector budget Loc. Curr. Executed external sector budget US$ 0 0 Executed recurrent budget Loc. Curr. Executed recurrent budget US$ 0 0 Executed investment budget Loc. Curr. Executed investment budget US$ 0 0 Executed salary costs Loc. Curr. Executed salary costs US$ 0 0 Rural Investments Loc. Curr. Rural Investments US$ 0 0 Actual WSS expenditure/GDP % 0.00 0.00 Share of exp that is externally funded % 0.0% 0.0% Recurrent exp per capita $ Domestic-funded sector budget execution % 0 0 External-funded budget execution % 0 68  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa 2002 2003 2004 2005 2006 2007 2008 2009 70,557,000 7,428,800,000 8,029,600,000 9,484,700,000 11,373,300,000 13,315,400,000 19,182,000,000 25,899,000,000 0 0 0 0 0 0 0 0 0 0 0 0 0 1,032,000,000 0 0 0 0 0 0 0 107,502,893 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 0 657,000,000 0 0 0 0 0 0 68,439,342 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 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.00 0.00 0.00 0.00 0.00 0.00 0.26 0.00 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0 0 0 0 0 0 0 0 0 0 0 0 Annexes | 69 Ghana No. Indicators Unit 2000 2001 WS&S Public Overall Trends 1 Sector Budget Allocation (share of total general gvt. budget) % 0.0 0.9 2 Sector Budget Allocation (share of GDP) % 0.0 0.1 3 Domestic sector allocations (share of total sector public allocations) % 0.0 9.9 4 External allocations (share of total sector public allocations) % 0.0 90.1 5 Sector allocations transferred to sub-national levels (share of total sector allocations) % 0.0 0.0 6 WS&S spending per capita US$ 0.00 0.00 Access to WS&S Services 7 Overall access to water % 8 Access to water – Urban areas % 9 Access to water – Rural areas % 41.0 10 Overall access to sanitation % 11 Access to sanitation – Urban areas % 12 Access to sanitation – Rural areas % 13 Gap between province/region with highest & lowest access rate to water % point WS&S Sector Spending Performance 14 WSS Sector Budget Execution Rate % 0.0 145.4 15 Execution of WS&S sector domestic resources % 0.0 301.2 16 Execution of WS&S sector external resources % 0.0 128.3 17 Recurrent spending (share of total sector spending) % 0.0 10.6 18 Investment spending (share of total sector spending) % 0.0 89.4 19 Salary costs (share of total recurrent spending) % 0.0 33.9 20 Rural investment Spending (share of total WS&S investment spending) % 0.0 0.0 21 Urban investment Spending (share of total WS&S investment spending) % 0.0 100.0 Notes to Indicator Data Ind. No. Year(s) Comment 10 2006 This is the % using flush toilet, pit latrine, pan/bucket (37.4%) or KVIP. No rural/urban breakdown. Rural population The 2002 figure includes semi-urban population 14 and 16 2001–06 These are based on the main executing agency ((the CWSA) only Executed salary costs These are in fact budgeted salary costs in the main executing agency (CWSA) 6 2001–06 This is average annual investment spending only, $133.8 mn over 12.6 mn rural people over 6 years (2.6 mn people served) Executed sector budget = budget per table 6 * Execution rate for CWSA per Table 8 70  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa 2002 2003 2004 2005 2006 2007 2008 2009 2.5 2.3 1.1 2.1 2.6 0.0 0.0 0.0 0.4 0.4 0.2 0.4 0.5 0.0 0.0 0.0 3.9 4.8 10.3 4.2 3.8 0.0 0.0 0.0 96.1 84.3 82.5 91.6 93.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.57 1.63 1.34 0.85 1.31 0.00 0.00 [‘/////// 52.9 53.0 80.4 25.7 49.0 123.4 185.3 42.7 47.0 0.0 0.0 0.0 447.1 513.4 785.7 468.3 153.9 0.0 0.0 0.0 32.9 116.9 126.2 25.4 44.2 0.0 0.0 0.0 13.3 5.0 6.3 13.8 10.9 0.0 0.0 0.0 86.8 95.1 93.8 86.2 89.1 0.0 0.0 0.0 46.6 45.7 42.7 37.9 37.7 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 100.0 100.0 100.0 100.0 0.0 0.0 0.0 Annexes | 71 Memo Unit 2000 2001 Total population inhbt 18,900,000 GDP Loc. Curr. 38,071,000,000,000 GDP US$ 0 5,314,913,565 General gvt. budget Loc. Curr. 6,329,000,000,000 General gvt. budget US$ 0 883,561,975 Total sector budget allocations Loc. Curr. 0 56,700,000,000 Total sector budget allocations US$ 0 7,915,621 Domestic sector budget allocations Loc. Curr. 5,600,000,000 Domestic sector budget allocations US$ 0 781,790 External sector budget allocations Loc. Curr. 51,100,000,000 External sector budget allocations US$ 0 7,133,831 WS&S transfers to sub-national levels Loc. Curr. WS&S transfers to sub-national levels US$ 0 0 Executed sector budget Loc. Curr. 0 82,467,000,000 Executed sector budget US$ 0 11,512,831 Executed domestic sector budget Loc. Curr. 16,866,000,000 Executed domestic sector budget US$ 0 2,354,583 Executed external sector budget Loc. Curr. 65,601,000,000 Executed external sector budget US$ 0 9,158,248 Executed recurrent budget Loc. Curr. 8,762,000,000 Executed recurrent budget US$ 0 1,223,222 Executed investment budget Loc. Curr. 73,705,000,000 Executed investment budget US$ 0 10,289,609 Executed salary costs Loc. Curr. 3,286,000,000 Executed salary costs US$ 0 414,681 Rural Investments Loc. Curr. Rural Investments US$ 0 0 Rural population Actual WSS expenditure/GDP % 0.00 0.22 Share of exp that is externally funded % 0.0% 79.5% Recurrent exp per capita $ 72  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa 2002 2003 2004 2005 2006 2007 2008 2009 20,475,000 20,955,000 21,435,000 21,915,000 22,393,000 22,871,000 23,351,000 23,837,000 48,862,000,000,000 66,158,000,000,000 79,888,000,000,000 97,261,000,000,000 118,404,000,000,000 6,166,197,192 7,632,755,292 8,881,429,354 10,731,905,210 12,919,828,792 0 0 7,456,000,000,000 10,442,000,000,000 13,005,000,000,000 18,528,000,000,000 22,359,000,000,000 940,918,633 1,204,710,402 1,445,811,495 2,044,403,612 2,439,735,583 0 0 187,600,000,000 239,800,000,000 139,400,000,000 396,600,000,000 571,800,000,000 0 0 0 23,674,401 27,666,113 15,497,587 43,761,360 62,392,809 0 0 7,300,000,000 11,600,000,000 14,400,000,000 16,500,000,000 21,800,000,000 921,232 1,338,311 1,600,899 1,820,631 2,378,739 0 0 180,300,000,000 228,200,000,000 125,000,000,000 380,100,000,000 550,000,000,000 22,753,169 26,327,802 13,896,689 41,940,728 60,014,069 0 0 0 0 0 0 0 0 0 92,009,400,000 295,968,000,000 258,365,000,000 169,370,000,000 268,755,000,000 0 0 0 11,600,457 34,139,984 28,717,029 18,686,101 29,324,620 0 0 32,641,400,000 59,549,000,000 113,139,000,000 77,264,000,000 33,548,000,000 4,119,220 6,870,264 12,578,060 8,525,410 3,660,640 0 0 59,368,000,000 236,419,000,000 145,226,000,000 92,106,000,000 235,207,000,000 7,492,014 27,276,042 16,145,284 10,163,096 25,664,962 0 0 12,226,000,000 14,702,000,000 16,156,000,000 23,415,000,000 29,300,000,000 1,542,874 1,696,193 1,796,119 2,583,642 3,197,113 0 0 79,783,000,000 281,266,000,000 242,208,000,000 145,954,000,000 239,455,000,000 10,068,309 32,450,113 26,927,113 16,104,754 26,128,489 0 0 5,703,000,000 6,712,000,000 6,895,000,000 8,880,000,000 11,053,000,000 719,697 774,374 766,541 979,831 1,206,065 0 0 0 0 0 0 0 0 0 15,400,000 11,600,000 0.19 0.45 0.32 0.17 0.23 0.00 0.00 0.00 64.6% 79.9% 56.2% 54.4% 87.5% 0.0% 0.0% 0.0% 0.08 0.08 0.08 0.12 0.14 0.00 0.00 Annexes | 73 Madagascar No. Indicators Unit 2000 2001 2002 WS&S Public Overall Trends 1 Sector Budget Allocation (share of total general gvt. budget) % 0.6 1.2 0.7 2 Sector Budget Allocation (share of GDP) % 0.1 0.3 0.2 3 Domestic sector allocations (share of total sector public allocations) % 31.5 38.1 34.8 4 External allocations (share of total sector public allocations) % 68.5 61.9 65.2 5 Sector allocations transferred to sub-national levels (share of total sector allocations) % 0.0 0.0 0.0 6 WS&S spending per capita US$ 0.3 0.5 0.2 Access to WS&S Services 7 Overall access to water % 8 Access to water – Urban areas % 9 Access to water – Rural areas % 22.2 24.0 25.3 10 Overall access to sanitation % 11 Access to sanitation – Urban areas % 12 Access to sanitation – Rural areas % 13 Gap between province/region with highest & lowest access rate to water % point WS&S Sector Spending Performance 14 WSS Sector Budget Execution Rate % 58.9 43.1 31.9 15 Execution of WS&S sector domestic resources % 64.7 85.7 55.9 16 Execution of WS&S sector external resources % 56.3 16.9 19.1 17 Recurrent spending (share of total sector spending) % 9.1 5.2 12.1 18 Investment spending (share of total sector spending) % 90.9 94.8 87.9 19 Salary costs (share of total recurrent spending) % 16.8 18.9 18.7 20 Rural investment Spending (share of total WS&S investment spending) % 100.0 100.0 100.0 21 Urban investment Spending (share of total WS&S investment spending) % 0.0 0.0 0.0 Notes to Indicator Data Ind. No. Year(s) Comment 20+21 Rural Sector only 74  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa 2003 2004 2005 2006 2007 2008 2009 1.6 1.3 1.3 1.4 0.0 0.0 0.0 0.3 0.3 0.3 0.3 0.0 0.0 0.0 21.6 39.0 43.4 38.5 0.0 0.0 0.0 78.4 61.0 56.6 61.5 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.4 0.4 1.0 0.0 0.0 0.0 0.0 27.2 29.5 30.1 44.0 29.0 26.6 45.7 83.6 0.0 0.0 0.0 0.0 75.0 61.6 69.0 0.0 0.0 0.0 0.0 13.2 35.6 94.7 0.0 0.0 0.0 0.0 6.5 3.8 1.5 0.0 0.0 0.0 0.0 93.5 96.2 98.5 0.0 0.0 0.0 0.0 21.4 25.3 25.9 0.0 0.0 0.0 0.0 100.0 100.0 100.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Annexes | 75 Memo Unit 2000 2001 2002 Total population inhbt 12,236,820 12,486,551 12,741,379 GDP Loc. Curr. 5,248,400,000,000 5,968,600,000,000 6,008,400,000,000 GDP US$ 3,877,661,518 4,529,563,091 4,397,276,671 General gvt. budget Loc. Curr. 1,299,700,000,000 1,459,800,000,000 1,595,500,000,000 General gvt. budget US$ 960,253,920 1,107,840,398 1,167,674,410 Total sector budget allocations Loc. Curr. 7,300,000,000 18,100,000,000 11,200,000,000 Total sector budget allocations US$ 5,393,440 13,736,067 8,196,774 Domestic sector budget allocations Loc. Curr. 2,300,000,000 6,900,000,000 3,900,000,000 Domestic sector budget allocations US$ 1,699,303 5,236,401 2,854,234 External sector budget allocations Loc. Curr. 5,000,000,000 11,200,000,000 7,300,000,000 External sector budget allocations US$ 3,694,137 8,499,666 5,342,540 WS&S transfers to sub-national levels Loc. Curr. 0 0 0 WS&S transfers to sub-national levels US$ 0 0 0 Executed sector budget Loc. Curr. 4,301,100,000 7,801,400,000 3,572,300,000 Executed sector budget US$ 3,177,770 5,920,473 2,614,405 Executed domestic sector budget Loc. Curr. 1,487,800,000 5,911,000,000 2,180,800,000 Executed domestic sector budget US$ 1,099,227 4,485,851 1,596,029 Executed external sector budget Loc. Curr. 2,813,300,000 1,890,400,000 1,391,500,000 Executed external sector budget US$ 2,078,543 1,434,622 1,018,376 Executed recurrent budget Loc. Curr. 391,500,000 408,800,000 431,100,000 Executed recurrent budget US$ 289,251 310,238 315,503 Executed investment budget Loc. Curr. 3,909,600,000 7,392,600,000 3,141,200,000 Executed investment budget US$ 2,888,519 5,610,235 2,298,902 Executed salary costs Loc. Curr. 65,800,000 77,200,000 80,800,000 Executed salary costs US$ 48,615 58,587 59,134 Rural Investments Loc. Curr. 3,909,600,000 7,392,600,000 3,141,200,000 Rural Investments US$ 2,888,519 5,610,235 2,298,902 Actual WSS expenditure/GDP % 0.08 0.13 0.06 Share of exp that is externally funded % 65.4% 24.2% 39.0% Recurrent exp per capita $ 0.02 Domestic-funded sector budget execution % 0.65 0.86 0.56 External-funded budget execution % 0.56 0.17 0.19 76  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa 2003 2004 2005 2006 2007 2008 2009 13,001,407 13,266,742 13,537,492 13,813,767 6,778,600,000,000 8,155,600,000,000 10,095,100,000,000 11,794,600,000,000 5,473,995,440 4,363,948,854 5,039,925,013 5,505,573,834 0 0 1,444,000,000,000 1,699,100,000,000 2,386,400,000,000 2,792,500,000,000 1,166,088,782 909,164,929 1,191,397,515 1,303,504,564 0 0 23,100,000,000 21,800,000,000 31,800,000,000 39,000,000,000 0 0 0 18,654,190 11,664,879 15,875,981 18,204,719 0 0 5,000,000,000 8,500,000,000 13,800,000,000 15,000,000,000 4,037,704 4,548,233 6,889,577 7,001,815 0 0 18,100,000,000 13,300,000,000 18,000,000,000 24,000,000,000 14,616,487 7,116,646 8,986,404 11,202,904 0 0 0 0 0 24,000,000 0 0 0 11,203 0 0 6,149,100,000 9,966,900,000 26,574,200,000 0 0 0 0 4,965,649 5,333,150 13,267,028 0 0 0 3,752,000,000 5,236,300,000 9,527,000,000 3,029,893 2,801,872 4,756,304 0 0 0 2,397,100,000 4,730,600,000 17,047,200,000 1,935,756 2,531,279 8,510,724 0 0 0 400,200,000 383,400,000 398,800,000 323,178 205,152 199,099 0 0 0 5,748,900,000 9,583,500,000 26,175,400,000 4,642,471 5,127,998 13,067,929 0 0 0 85,700,000 97,100,000 103,300,000 69,206 51,957 51,572 0 0 0 5,748,900,000 9,583,500,000 26,175,400,000 4,642,471 5,127,998 13,067,929 0 0 0 0.09 0.12 0.26 0.00 0.00 0.00 0.00 39.0% 47.5% 64.1% 0.0% 0.0% 0.0% 0.0% 0.02 0.02 0.01 0.00 0.00 0.00 0.75 0.62 0.69 0.13 0.36 0.95 Annexes | 77 Mali No. Indicators Unit 2000 2001 2002 WS&S Public Overall Trends 1 Sector Budget Allocation (share of total general gvt. budget) % 0.0 1.9 1.4 2 Sector Budget Allocation (share of GDP) % 0.0 0.6 0.4 3 Domestic sector allocations (share of total sector public allocations) % 0.0 14.0 11.2 4 External allocations (share of total sector public allocations) % 0.0 86.0 88.8 5 Sector allocations transferred to sub-national levels (share of total sector allocations) % 0.0 0.2 0.2 6 WS&S spending per capita US$ 0.0 1.2 1.0 Access to WS&S Services 7 Overall access to water % 8 Access to water – Urban areas % 9 Access to water – Rural areas % 45.0 45.5 10 Overall access to sanitation % 0.5 0.5 11 Access to sanitation – Urban areas % 12 Access to sanitation – Rural areas % 13 Gap between province/region with highest & lowest access rate to water % point WS&S Sector Spending Performance 14 WSS Sector Budget Execution Rate % 0.0 67.0 65.3 15 Execution of WS&S sector domestic resources % 0.0 74.5 81.0 16 Execution of WS&S sector external resources % 0.0 65.8 63.3 17 Recurrent spending (share of total sector spending) % 0.0 4.9 6.6 18 Investment spending (share of total sector spending) % 0.0 95.1 93.4 19 Salary costs (share of total recurrent spending) % 0.0 79.3 76.4 20 Rural investment Spending (share of total WS&S investment spending) % 0.0 100.0 100.0 21 Urban investment Spending (share of total WS&S investment spending) % 0.0 0.0 0.0 Notes to Indicator Data Ind. No. Year(s) Comment 20+21 Rural Sector only 78  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa 2003 2004 2005 2006 2007 2008 2009 1.4 1.6 2.3 2.1 0.0 0.0 0.0 0.4 0.5 0.7 1.0 0.0 0.0 0.0 16.0 14.3 8.0 10.6 0.0 0.0 0.0 84.0 85.7 92.0 89.4 0.0 0.0 0.0 0.2 0.2 0.1 0.1 0.0 0.0 0.0 0.9 1.6 3.0 3.5 0.0 0.0 0.0 46.2 47.9 49.4 50.3 0.5 0.5 0.5 0.5 9.0 30.0 44.1 63.3 74.8 63.5 0.0 0.0 0.0 75.0 81.7 97.1 98.1 0.0 0.0 0.0 38.3 60.3 72.9 59.5 0.0 0.0 0.0 10.0 6.7 3.8 3.5 0.0 0.0 0.0 90.0 93.3 96.2 96.5 0.0 0.0 0.0 77.9 79.8 77.2 73.0 0.0 0.0 0.0 100.0 100.0 100.0 100.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Annexes | 79 Memo Unit 2000 2001 2002 Total population inhbt 8,622,251 8,798,215 GDP Loc. Curr. 1,928,000,000,000 2,330,000,000,000 GDP US$ 0 2,630,148,323 3,342,954,713 General gvt. budget Loc. Curr. 587,899,111,000 679,330,540,000 General gvt. budget US$ 0 802,003,040 974,665,764 Total sector budget allocations Loc. Curr. 0 11,082,000,000 9,690,000,000 Total sector budget allocations US$ 0 15,117,896 13,902,674 Domestic sector budget allocations Loc. Curr. 1,554,000,000 1,087,000,000 Domestic sector budget allocations US$ 0 2,119,943 1,559,567 External sector budget allocations Loc. Curr. 9,528,000,000 8,603,000,000 External sector budget allocations US$ 0 12,997,953 12,343,107 WS&S transfers to sub-national levels Loc. Curr. 21,800,000 22,000,000 WS&S transfers to sub-national levels US$ 0 29,739 31,564 Executed sector budget Loc. Curr. 0 7,422,000,000 6,329,000,000 Executed sector budget US$ 0 10,124,980 9,080,498 Executed domestic sector budget Loc. Curr. 1,157,000,000 881,000,000 Executed domestic sector budget US$ 0 1,578,362 1,264,010 Executed external sector budget Loc. Curr. 6,265,000,000 5,448,000,000 Executed external sector budget US$ 0 8,546,618 7,816,488 Executed recurrent budget Loc. Curr. 362,000,000 415,000,000 Executed recurrent budget US$ 0 493,835 595,419 Executed investment budget Loc. Curr. 7,060,000,000 5,914,000,000 Executed investment budget US$ 0 9,631,145 8,485,079 Executed salary costs Loc. Curr. 287,000,000 317,000,000 Executed salary costs US$ 0 391,521 454,814 Rural Investments Loc. Curr. 7,060,000,000 5,914,000,000 Rural Investments US$ 0 9,631,145 8,485,079 Actual WSS expenditure/GDP % 0.00 0.38 0.27 Share of exp that is externally funded % 0.0% 84.4% 86.1% Recurrent exp per capita $ 0.07 Domestic-funded sector budget execution % 0.00 0.74 0.81 External-funded budget execution % 0.00 0.66 0.63 80  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa 2003 2004 2005 2006 2007 2008 2009 8,982,579 9,171,745 9,365,836 9,564,978 2,568,000,000,000 2,610,000,000,000 2,753,000,000,000 2,905,000,000,000 4,418,442,211 4,940,516,846 5,219,272,552 5,555,660,638 0 0 0 737,552,452,000 787,198,778,000 850,631,194,000 1,307,152,654,000 1,269,015,922 1,490,102,998 1,612,668,377 2,499,861,118 0 0 0 10,306,000,000 12,360,000,000 19,637,000,000 27,599,000,000 0 0 0 17,732,268 23,396,471 37,228,789 52,781,645 0 0 0 1,645,000,000 1,773,000,000 1,576,000,000 2,922,000,000 2,830,349 3,356,144 2,987,858 5,588,172 0 0 0 8,661,000,000 10,587,000,000 18,061,000,000 24,677,000,000 14,901,919 20,040,326 34,240,930 47,193,472 0 0 0 22,600,000 23,300,000 25,800,000 32,600,000 38,885 44,105 48,913 62,346 0 0 0 4,548,000,000 7,830,000,000 14,689,000,000 17,537,000,000 0 0 0 7,825,185 14,821,551 27,848,127 33,538,596 0 0 0 1,233,000,000 1,449,000,000 1,531,000,000 2,866,000,000 2,121,472 2,742,839 2,902,545 5,481,075 0 0 0 3,315,000,000 6,381,000,000 13,158,000,000 14,671,000,000 5,703,713 12,078,712 24,945,582 28,057,521 0 0 0 457,000,000 521,000,000 552,000,000 618,000,000 786,304 986,210 1,046,509 1,181,893 0 0 0 4,092,000,000 7,309,000,000 14,137,000,000 16,918,000,000 7,040,602 13,835,340 26,801,619 32,354,791 0 0 0 356,000,000 416,000,000 426,000,000 451,000,000 612,525 787,454 807,632 862,514 0 0 0 4,092,000,000 7,309,000,000 14,137,000,000 16,918,000,000 7,040,602 13,835,340 26,801,619 32,354,791 0 0 0 0.18 0.30 0.53 0.60 0.00 0.00 0.00 72.9% 81.5% 89.6% 83.7% 0.0% 0.0% 0.0% 0.09 0.11 0.11 0.12 0.00 0.00 0.75 0.82 0.97 0.38 0.60 0.73 Annexes | 81 Mozambique No. Indicators Unit 2000 2001 2002 WS&S Public Overall Trends 1 Sector Budget Allocation (share of total general gvt. budget) % 0.0 0.0 2.6 2 Sector Budget Allocation (share of GDP) % 0.0 0.0 0.7 3 Domestic sector allocations (share of total sector public allocations) % 0.0 0.0 0.0 4 External allocations (share of total sector public allocations) % 0.0 0.0 100.0 5 Sector allocations transferred to sub-national levels (share of total sector allocations) % 0.0 0.0 0.0 6 WS&S spending per capita US$ 0.0 0.0 1.1 Access to WS&S Services 7 Overall access to water % 38.0 8 Access to water – Urban areas % 83.0 9 Access to water – Rural areas % 24.0 10 Overall access to sanitation % 10.0 11 Access to sanitation – Urban areas % 32.0 12 Access to sanitation – Rural areas % 2.0 13 Gap between province/region with highest & lowest access rate to water % point >80 WS&S Sector Spending Performance 14 WSS Sector Budget Execution Rate % 0.0 0.0 78.2 15 Execution of WS&S sector domestic resources % 0.0 0.0 0.0 16 Execution of WS&S sector external resources % 0.0 0.0 0.0 17 Recurrent spending (share of total sector spending) % 67.2 61.2 56.0 18 Investment spending (share of total sector spending) % 32.8 38.8 44.0 19 Salary costs (share of total recurrent spending) % 0.0 0.0 0.0 20 Rural investment Spending (share of total WS&S investment spending) % 0.0 0.0 0.0 21 Urban investment Spending (share of total WS&S investment spending) % 100.0 100.0 100.0 Notes to Indicator Data Ind. No. Year(s) Comment Ex’d sector budget 2002–2008 Per Ministry of Finance Ex’d recurrent 2000–2008 Expenditure as reported by sector (varies from exp reported by Min of Finance , mostly above, up to 2x) 82  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa 2003 2004 2005 2006 2007 2008 2009 1.6 2.1 3.6 3.5 2.3 3.5 0.0 0.4 0.5 0.8 0.9 0.7 0.9 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 100.0 100.0 100.0 100.0 100.0 100.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2.1 2.0 4.1 2.9 4.4 5.3 0.0 43.0 71.0 70.0 26.0 30.0 19.0 48.0 7.0 60.0 244.1 145.3 147.9 93.7 183.4 127.6 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 0.0 0.0 33.6 25.1 17.9 26.8 19.4 13.7 0.0 66.0 74.9 82.1 73.2 80.6 86.3 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 0.0 0.0 100.0 100.0 100.0 100.0 100.0 100.0 0.0 Annexes | 83 Memo Unit 2000 2001 2002 Total population inhbt 19,259,465 GDP Loc. Curr. GDP US$ 0 3,697,200,000 4,091,700,000 General gvt. budget Loc. Curr. 24,871,000,000 General gvt. budget US$ 0 0 1,050,386,245 Total sector budget allocations Loc. Curr. 0 0 636,000,000 Total sector budget allocations US$ 0 0 26,860,426 Domestic sector budget allocations Loc. Curr. Domestic sector budget allocations US$ 0 0 0 External sector budget allocations Loc. Curr. External sector budget allocations US$ 0 0 0 WS&S transfers to sub-national levels Loc. Curr. WS&S transfers to sub-national levels US$ 0 0 0 Executed sector budget Loc. Curr. 72,917,000 229,051,000 497,356,000 Executed sector budget US$ 4,788,586 11,063,320 21,005,022 Executed domestic sector budget Loc. Curr. 58,376,000 155,532,000 296,763,000 Executed domestic sector budget US$ 3,833,653 7,512,302 12,533,303 Executed external sector budget Loc. Curr. 14,541,000 73,519,000 200,593,000 Executed external sector budget US$ 954,933 3,551,018 8,471,719 Executed recurrent budget Loc. Curr. 49,011,000 140,116,000 278,300,000 Executed recurrent budget US$ 3,218,638 6,767,699 11,753,548 Executed investment budget Loc. Curr. 23,906,000 88,935,000 219,056,000 Executed investment budget US$ 1,569,949 4,295,621 9,251,474 Executed salary costs Loc. Curr. Executed salary costs US$ 0 0 0 Rural Investments Loc. Curr. Rural Investments US$ 0 0 0 Actual WSS expenditure/GDP % 0.00 0.30 0.51 Share of exp that is externally funded % 19.9% 32.1% 40.3% Urban population 6205400 Domestic-funded sector budget execution % 0 0 0 External-funded budget execution % 0 0 0 GDP deflator 95.04 External funding % 19.9 32.1 40.3 Recurrent exp per capita $ 0.61 84  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa 2003 2004 2005 2006 2007 2008 2009 19,783,627 20,310,610 20,834,379 21,352,466 21,869,362 22,382,533 4,789,400,000 5,903,900,000 6,823,000,000 7,608,200,000 8,030,000,000 9,867,000,000 25,044,000,000 29,503,000,000 36,912,000,000 48,569,000,000 58,256,000,000 64,067,000,000 1,053,053,499 1,306,521,080 1,600,626,860 1,912,106,699 2,254,459,374 2,636,432,352 410,000,000 618,000,000 1,320,000,000 1,694,000,000 1,366,000,000 2,261,000,000 0 17,239,735 27,367,726 57,239,582 66,690,868 52,863,079 93,042,808 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 996,564,000 897,773,000 1,952,701,000 1,586,463,000 2,505,818,000 2,885,946,000 0 42,088,260 39,757,291 84,675,598 62,457,257 96,973,099 118,760,070 377,798,000 322,561,000 455,299,000 554,236,000 750,024,000 677,055,000 15,885,701 14,284,403 19,743,276 21,819,646 29,025,313 27,861,609 618,766,000 575,212,000 1,497,402,000 1,032,227,000 1,755,794,000 2,208,891,000 26,017,956 25,472,888 64,932,322 40,637,612 67,947,786 90,898,461 336,119,000 225,169,000 350,332,000 424,642,000 486,474,000 395,896,000 14,133,177 9,971,462 15,191,559 16,717,676 18,826,144 16,291,586 660,445,000 672,604,000 1,602,369,000 1,161,821,000 2,019,344,000 2,490,050,000 27,770,481 29,785,829 69,484,039 45,739,581 78,146,955 102,468,484 0 0 0 0 0 0 0 0 0 0 0 0 0.88 0.67 1.24 0.82 1.21 1.20 0.00 61.8% 64.1% 76.7% 65.1% 70.1% 76.5% 0.0% 6524640 6852800 7187861 7533503 7886092 8245725 0 0 0 0 0 0 100 107.47 116.91 127.78 137.38 147.98 62.1 64.1 76.7 65.1 70.1 76.5 0.71 0.49 0.73 0.78 0.86 0.73 Annexes | 85 Niger No. Indicators Unit 2000 2001 2002 WS&S Public Overall Trends 1 Sector Budget Allocation (share of total general gvt. budget) % 0.0 0.0 0.0 2 Sector Budget Allocation (share of GDP) % 0.0 0.0 0.3 3 Domestic sector allocations (share of total sector public allocations) % 0.0 0.0 13.1 4 External allocations (share of total sector public allocations) % 0.0 0.0 86.9 5 Sector allocations transferred to sub-national levels (share of total sector allocations) % 0.0 0.0 0.0 6 WS&S spending per capita US$ 0.0 0.0 0.6 Access to WS&S Services 7 Overall access to water % 8 Access to water – Urban areas % 9 Access to water – Rural areas % 55.0 10 Overall access to sanitation % 11 Access to sanitation – Urban areas % 12 Access to sanitation – Rural areas % 13 Gap between province/region with highest & lowest access rate to water % point WS&S Sector Spending Performance 14 WSS Sector Budget Execution Rate % 0.0 0.0 81.4 15 Execution of WS&S sector domestic resources % 0.0 0.0 81.0 16 Execution of WS&S sector external resources % 0.0 0.0 81.5 17 Recurrent spending (share of total sector spending) % 0.0 0.0 13.0 18 Investment spending (share of total sector spending) % 0.0 0.0 87.0 19 Salary costs (share of total recurrent spending) % 0.0 0.0 82.5 20 Rural investment Spending (share of total WS&S investment spending) % 0.0 0.0 100.0 21 Urban investment Spending (share of total WS&S investment spending) % 0.0 0.0 0.0 Notes to Indicator Data Ind. No. Year(s) Comment 20+21 Rural Sector only Heavy inconsistency in financial figures 86  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa 2003 2004 2005 2006 2007 2008 2009 0.0 0.0 4.2 3.2 5.4 0.0 0.0 1.2 1.4 0.9 0.6 1.3 0.0 0.0 15.1 24.0 19.2 37.5 31.2 0.0 0.0 84.9 76.0 80.8 62.5 68.8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.0 2.1 1.9 1.3 3.6 0.0 0.0 57.0 59.0 59.0 58.0 62.0 6.0 30.5 55.7 71.7 63.3 72.9 0.0 0.0 42.5 55.5 68.6 58.7 63.4 0.0 0.0 28.3 55.8 72.5 66.0 77.2 0.0 0.0 10.2 5.9 6.4 11.5 4.3 0.0 0.0 89.8 94.1 93.6 88.5 95.7 0.0 0.0 83.1 69.9 72.6 70.6 73.0 0.0 0.0 100.0 100.0 100.0 100.0 100.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Annexes | 87 Memo Unit 2000 2001 2002 Total population inhbt 9,715,571 GDP Loc. Curr. 1,499,100,000,000 GDP US$ 0 0 2,150,825,498 General gvt. budget Loc. Curr. General gvt. budget US$ 0 0 0 Total sector budget allocations Loc. Curr. 0 0 5,190,000,000 Total sector budget allocations US$ 0 0 7,446,324 Domestic sector budget allocations Loc. Curr. 680,000,000 Domestic sector budget allocations US$ 0 0 975,626 External sector budget allocations Loc. Curr. 4,510,000,000 External sector budget allocations US$ 0 0 6,470,698 WS&S transfers to sub-national levels Loc. Curr. WS&S transfers to sub-national levels US$ 0 0 0 Executed sector budget Loc. Curr. 0 0 4,226,000,000 Executed sector budget US$ 0 0 6,063,230 Executed domestic sector budget Loc. Curr. 551,000,000 Executed domestic sector budget US$ 0 0 790,544 Executed external sector budget Loc. Curr. 3,675,000,000 Executed external sector budget US$ 0 0 5,272,686 Executed recurrent budget Loc. Curr. 550,000,000 Executed recurrent budget US$ 0 0 789,109 Executed investment budget Loc. Curr. 3,676,000,000 Executed investment budget US$ 0 0 5,274,121 Executed salary costs Loc. Curr. 454,000,000 Executed salary costs US$ 0 0 651,374 Rural Investments Loc. Curr. 3,676,000,000 Rural Investments US$ 0 0 5,274,121 Actual WSS expenditure/GDP % 0.00 0.00 0.28 Share of exp that is externally funded % 0.0% 0.0% 87.0% Recurrent exp per capita $ 0.08 Domestic-funded sector budget execution % 0.00 0.00 0.81 External-funded budget execution % 0.00 0.00 0.81 88  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa 2003 2004 2005 2006 2007 2008 2009 10,029,158 10,353,053 10,687,602 11,033,163 11,388,099 1,534,300,000,000 1,539,400,000,000 1,755,200,000,000 1,872,200,000,000 2,004,200,000,000 2,639,881,575 2,913,958,480 3,327,594,327 3,580,484,629 4,181,804,525 0 0 358,400,000,000 376,600,000,000 491,600,000,000 0 0 679,472,315 720,227,813 1,025,733,512 0 0 18,231,000,000 20,796,000,000 15,189,000,000 11,951,000,000 26,739,000,000 0 0 31,367,843 39,365,130 28,796,052 22,855,663 55,791,473 0 0 2,751,000,000 4,987,000,000 2,915,000,000 4,481,000,000 8,339,000,000 4,733,308 9,439,984 5,526,400 8,569,678 17,399,495 0 0 15,480,000,000 15,809,000,000 12,274,000,000 7,470,000,000 18,400,000,000 26,634,535 29,925,146 23,269,652 14,285,984 38,391,978 0 0 0 0 0 0 0 0 0 5,554,000,000 11,589,000,000 10,895,000,000 7,560,000,000 19,500,000,000 0 0 9,556,086 21,937,031 20,655,276 14,458,105 40,687,151 0 0 1,169,000,000 2,768,000,000 2,001,000,000 2,629,000,000 5,287,000,000 2,011,355 5,239,598 3,793,594 5,027,825 11,031,434 0 0 4,385,000,000 8,821,000,000 8,894,000,000 4,931,000,000 14,213,000,000 7,544,731 16,697,433 16,861,682 9,430,280 29,655,717 0 0 567,000,000 684,000,000 700,000,000 867,000,000 848,000,000 975,567 1,294,756 1,327,094 1,658,092 1,769,369 0 0 4,987,000,000 10,905,000,000 10,195,000,000 6,693,000,000 18,652,000,000 8,580,518 20,642,274 19,328,181 12,800,013 38,917,782 0 0 471,000,000 478,000,000 508,000,000 612,000,000 619,000,000 810,392 904,815 963,091 1,170,418 1,291,556 0 0 4,987,000,000 10,905,000,000 10,195,000,000 6,693,000,000 18,652,000,000 8,580,518 20,642,274 19,328,181 12,800,013 38,917,782 0 0 0.36 0.75 0.62 0.40 0.97 0.00 0.00 79.0% 76.1% 81.6% 65.2% 72.9% 0.0% 0.0% 0.10 0.13 0.12 0.15 0.16 0.00 0.42 0.56 0.69 0.59 0.63 0.28 0.56 0.72 0.66 0.77 Annexes | 89 Republic of Congo No. Indicators Unit 2000 2001 2002 WS&S Public Overall Trends 1 Sector Budget Allocation (share of total general gvt. budget) % 0.0 0.0 0.6 2 Sector Budget Allocation (share of GDP) % 0.0 0.0 0.17 3 Domestic sector allocations (share of total sector public allocations) % 0.0 0.0 0.0 4 External allocations (share of total sector public allocations) % 0.0 0.0 100.0 5 Sector allocations transferred to sub-national levels (share of total sector allocations) % 0.0 0.0 0.0 6 WS&S spending per capita US$ 0.0 0.0 1.1 Access to WS&S Services 7 Overall access to water % 8 Access to water – Urban areas % 54.0 9 Access to water – Rural areas % 10 Overall access to sanitation % 11 Access to sanitation – Urban areas % 12 Access to sanitation – Rural areas % 13 Gap between province/region with highest & lowest access rate to water % point WS&S Sector Spending Performance 14 WSS Sector Budget Execution Rate % 0.0 0.0 69.9 15 Execution of WS&S sector domestic resources % 0.0 0.0 0.0 16 Execution of WS&S sector external resources % 0.0 0.0 0.0 17 Recurrent spending (share of total sector spending) % 0.0 0.0 4.4 18 Investment spending (share of total sector spending) % 0.0 0.0 95.6 19 Salary costs (share of total recurrent spending) % 0.0 0.0 0.0 20 Rural investment Spending (share of total WS&S investment spending) % 0.0 0.0 10.0 21 Urban investment Spending (share of total WS&S investment spending) % 0.0 0.0 90.0 Notes to Indicator Data Ind. No. Year(s) Comment 20+21 Share of rural investments estimated at 10% of total investments 90  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa 2003 2004 2005 2006 2007 2008 2009 0.1 0.3 0.5 0.9 1.4 2.3 0.0 0.0 0.1 0.1 0.2 0.4 0.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 100.0 100.0 100.0 100.0 100.0 100.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.7 2.2 1.5 8.9 3.9 4.5 0.0 52.0 52.0 46.0 45.0 46.0 45.0 15.0 27.1 5.0 402.0 231.3 68.8 170.7 42.9 27.3 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 0.0 0.0 4.1 3.4 5.6 1.7 3.8 7.5 0.0 95.9 96.6 94.4 98.3 96.2 92.5 0.0 0.0 0.0 10.1 5.5 7.2 4.2 0.0 10.0 10.0 10.0 10.0 10.0 10.0 0.0 90.0 90.0 90.0 90.0 90.0 90.0 0.0 Annexes | 91 Memo Unit 2000 2001 2002 Total population inhbt 3,228,600 GDP Loc. Curr. 2,104,900,000,000 GDP US$ 0 0 3,019,993,723 General gvt. budget Loc. Curr. 584,100,000,000 General gvt. budget US$ 0 0 838,034,269 Total sector budget allocations Loc. Curr. 0 0 3,635,000,000 Total sector budget allocations US$ 0 0 5,215,296 Domestic sector budget allocations Loc. Curr. Domestic sector budget allocations US$ 0 0 0 External sector budget allocations Loc. Curr. External sector budget allocations US$ 0 0 0 WS&S transfers to sub-national levels Loc. Curr. WS&S transfers to sub-national levels US$ 0 0 0 Executed sector budget Loc. Curr. 0 0 2,540,800,000 Executed sector budget US$ 0 0 3,645,399 Executed domestic sector budget Loc. Curr. 905,700,000 Executed domestic sector budget US$ 0 0 1,299,448 Executed external sector budget Loc. Curr. 1,635,100,000 Executed external sector budget US$ 0 0 2,345,951 Executed recurrent budget Loc. Curr. 110,700,000 Executed recurrent budget US$ 0 0 158,826 Executed investment budget Loc. Curr. 2,430,100,000 Executed investment budget US$ 0 0 3,486,573 Executed salary costs Loc. Curr. Executed salary costs US$ 0 0 0 Rural Investments Loc. Curr. 243,010,000 Rural Investments US$ 0 0 348,657 Actual WSS expenditure/GDP % 0.0% 0.00 0.12 Share of exp that is externally funded % 0.0% 0.0% 64.4% Recurrent exp per capita $ 0.05 Domestic-funded sector budget execution % 0 0 0 External-funded budget execution % 0 0 0 92  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa 2003 2004 2005 2006 2007 2008 2009 3,317,300 3,408,600 3,502,300 3,598,600 3,697,600 3,799,300 2,031,800,000,000 2,455,800,000,000 3,210,700,000,000 4,042,600,000,000 3,664,400,000,000 4,801,900,000,000 3,495,868,725 4,648,628,839 6,087,002,682 7,731,261,169 7,645,845,974 10,723,188,128 0 606,100,000,000 656,300,000,000 775,800,000,000 1,127,100,000,000 1,201,400,000,000 1,237,900,000,000 1,042,841,832 1,242,322,301 1,470,799,726 2,155,519,830 2,506,745,812 2,764,371,308 0 794,000,000 1,750,000,000 4,137,900,000 9,836,900,000 16,293,400,000 28,325,900,000 0 1,366,138 3,312,607 7,844,834 18,812,557 33,996,514 63,254,952 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3,191,700,000 4,048,000,000 2,848,800,000 16,794,000,000 6,989,600,000 7,727,700,000 0 5,491,566 7,662,533 5,400,895 32,117,647 14,583,944 17,256,832 0 794,400,000 3,872,000,000 2,585,200,000 16,539,900,000 6,668,100,000 7,571,000,000 1,366,827 7,329,380 4,901,149 31,631,694 13,913,128 16,906,903 0 2,397,300,000 176,000,000 263,600,000 254,100,000 321,500,000 156,700,000 4,124,740 333,154 499,746 485,953 670,816 349,929 0 129,900,000 136,900,000 158,800,000 291,400,000 267,000,000 583,200,000 223,503 259,141 301,061 557,287 557,101 1,302,352 0 3,061,800,000 3,911,100,000 2,690,000,000 16,502,600,000 6,722,600,000 7,144,500,000 5,268,063 7,403,393 5,099,834 31,560,360 14,026,843 15,954,480 0 16,000,000 16,000,000 19,200,000 24,600,000 0 0 30,334 30,599 40,061 54,935 0 306,180,000 391,110,000 269,000,000 1,650,260,000 672,260,000 714,450,000 526,806 740,339 509,983 3,156,036 1,402,684 1,595,448 0 0.16 0.16 0.09 0.42 0.19 0.16 0.00 75.1% 4.3% 9.3% 1.5% 4.6% 2.0% 0.0% 0.07 0.08 0.09 0.15 0.15 0.34 0 0 0 0 0 0 Annexes | 93 Sierra Leone No. Indicators Unit 2000 2001 2002 WS&S Public Overall Trends 1 Sector Budget Allocation (share of total general gvt. budget) % 0.0 0.0 0.0 2 Sector Budget Allocation (share of GDP) % 0.0 0.0 0.9 3 Domestic sector allocations (share of total sector public allocations) % 0.0 0.0 100.0 4 External allocations (share of total sector public allocations) % 0.0 0.0 0.0 5 Sector allocations transferred to sub-national levels (share of total sector allocations) % 0.0 0.0 0.0 6 WS&S spending per capita US$ 0.00 0.00 0.38 Access to WS&S Services 7 Overall access to water % 54.0 57.0 8 Access to water – Urban areas % 9 Access to water – Rural areas % 10 Overall access to sanitation % 63.0 39.0 11 Access to sanitation – Urban areas % 12 Access to sanitation – Rural areas % 13 Gap between province/region with highest & lowest access rate to water % point WS&S Sector Spending Performance 14 WSS Sector Budget Execution Rate % 0.0 0.0 20.6 15 Execution of WS&S sector domestic resources % 0.0 0.0 20.6 16 Execution of WS&S sector external resources % 0.0 0.0 0.0 17 Recurrent spending (share of total sector spending) % 0.0 0.0 0.0 18 Investment spending (share of total sector spending) % 0.0 0.0 100.0 19 Salary costs (share of total recurrent spending) % 0.0 0.0 0.0 20 Rural investment Spending (share of total WS&S investment spending) % 0.0 0.0 0.0 21 Urban investment Spending (share of total WS&S investment spending) % 0.0 0.0 100.0 Notes to Indicator Data Ind. No. Year(s) Comment 10 2000–08 The definition of improved sanitation changes from one M&E program to another, and from year to year. 14–16 2002–09 No budget data for WSS externally funded expenditure. Execution rates are solely for government expenditure, which is almost entirely recurrent.. 94  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa 2003 2004 2005 2006 2007 2008 2009 0.0 0.0 0.0 0.0 0.0 0.0 2.4 0.3 0.2 0.4 0.2 0.2 0.2 0.2 100.0 100.0 100.0 100.0 100.0 100.0 100.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 9.7 1.3 6.4 6.7 0.32 0.65 1.20 2.65 1.62 1.32 1.37 53.0 47.0 50.5 81.7 35.2 30.0 13.0 26.1 6.5 54.4 62.2 12.8 39.8 19.9 35.8 30.3 54.4 62.2 12.8 39.8 19.9 35.8 30.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 10.8 5.1 6.4 12.3 0.0 100.0 100.0 89.3 95.0 93.5 88.5 0.0 0.0 0.0 0.0 23.0 42.0 27.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 100.0 100.0 100.0 100.0 100.0 100.0 0.0 Annexes | 95 Memo Unit 2000 2001 2002 Total population inhbt 4,368,000 4,540,000 GDP Loc. Curr. 1,964,695,698,342 GDP US$ 0 805,700,000 936,000,000 General gvt. budget Loc. Curr. General gvt. budget US$ 0 0 0 Total sector budget allocations Loc. Curr. 0 0 17,483,300,000 Total sector budget allocations US$ 8,329,213 Domestic sector budget allocations Loc. Curr. 17,483,300,000 Domestic sector budget allocations US$ 0 0 8,329,213 External sector budget allocations Loc. Curr. External sector budget allocations US$ 0 0 0 WS&S transfers to sub-national levels Loc. Curr. WS&S transfers to sub-national levels US$ 0 0 0 Executed sector budget Loc. Curr. 0 0 3,596,500,000 Executed sector budget US$ 0 0 1,713,407 Executed domestic sector budget Loc. Curr. 3,596,500,000 Executed domestic sector budget US$ 0 0 1,713,407 Executed external sector budget Loc. Curr. Executed external sector budget US$ 0 0 0 Executed recurrent budget Loc. Curr. Executed recurrent budget US$ 0 0 0 Executed investment budget Loc. Curr. 3,596,500,000 Executed investment budget US$ 0 0 1,713,407 Executed salary costs Loc. Curr. Executed salary costs US$ 0 0 0 Rural Investments Loc. Curr. Rural Investments US$ 0 0 0 Actual WSS expenditure/GDP % 0.00 0.00 0.18 Share of exp that is externally funded % 0.0% 0.0% 0.0% 96  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa 2003 2004 2005 2006 2007 2008 2009 4,733,000 4,926,000 5,107,000 5,271,000 5,420,000 5,560,000 5,696,000 2,323,757,867,500 2,894,439,378,333 3,518,361,740,000 4,274,034,927,500 4,967,349,226,667 5,828,861,157,042 6,574,932,300,000 989,700,000 1,071,500,000 1,217,600,000 1,443,000,000 1,664,000,000 1,955,000,000 1,942,000,000 0 0 0 0 0 0 0 5,837,600,000 4,937,400,000 10,882,000,000 5,010,800,000 6,433,500,000 6,528,200,000 8,240,300,000 2,781,089 2,352,225 5,184,290 2,387,193 3,064,982 3,110,098 3,925,758 5,837,600,000 4,937,400,000 10,882,000,000 5,010,800,000 6,433,500,000 6,528,200,000 8,240,300,000 2,781,089 2,352,225 5,184,290 2,387,193 3,064,982 3,110,098 3,925,758 0 0 0 0 0 0 687,600,000 114,700,000 592,900,000 787,400,000 0 0 0 232,148 38,423 198,859 264,094 3,554,400,000 8,624,100,000 17,763,200,000 41,398,900,000 26,246,200,000 21,870,300,000 27,323,800,000 1,513,837 3,192,679 6,146,609 13,971,654 8,792,544 7,336,322 7,826,450 3,554,400,000 3,954,100,000 1,911,200,000 2,811,900,000 1,820,200,000 3,317,300,000 4,759,800,000 1,513,837 1,463,779 661,409 949,354 609,744 1,112,622 1,189,950 4,670,000,000 15,852,000,000 38,587,000,000 24,426,000,000 18,553,000,000 22,564,000,000 0 1,728,900 5,485,200 13,027,300 8,182,800 6,223,700 6,636,500 1,911,200,000 2,118,200,000 1,681,500,000 2,696,100,000 3,958,400,000 0 0 661,409 715,147 563,282 904,272 3,554,400,000 8,624,000,000 15,852,000,000 39,293,000,000 24,531,500,000 19,366,549,700 23,365,500,000 1,513,837 3,192,541 5,485,904 13,266,106 8,217,746 6,495,541 487,100,000 706,100,000 729,000,000 864,800,000 0 0 0 164,455 236,535 244,507 290,054 0 0 0 0 0 0 0.15 0.30 0.50 0.97 0.53 0.38 0.40 0.0% 54.2% 89.2% 93.2% 93.1% 84.8% 84.8% Annexes | 97 Tanzania No. Indicators Unit 2000 2001 2002 WS&S Public Overall Trends 1 Sector Budget Allocation (share of total general gvt. budget) % 0.0 0.0 0.0 2 Sector Budget Allocation (share of GDP) % 0.00 0.35 0.29 3 Domestic sector allocations (share of total sector public allocations) % 0.0 0.0 0.0 4 External allocations (share of total sector public allocations) % 0.0 100.0 100.0 5 Sector allocations transferred to sub-national levels (share of total sector allocations) % 0.0 0.0 0.0 6 WS&S spending per capita US$ 0.0 0.78 0.69 Access to WS&S Services 7 Overall access to water % 66.0 8 Access to water – Urban areas % 90.0 9 Access to water – Rural areas % 56.0 10 Overall access to sanitation % 11 Access to sanitation – Urban areas % 96.8 12 Access to sanitation – Rural areas % 91.7 13 Gap between province/region with highest & lowest access rate to water % point WS&S Sector Spending Performance 14 WSS Sector Budget Execution Rate % 0.0 82.9 88.5 15 Execution of WS&S sector domestic resources % 0.0 0.0 0.0 16 Execution of WS&S sector external resources % 0.0 0.0 0.0 17 Recurrent spending (share of total sector spending) % 13.6 20.8 26.0 18 Investment spending (share of total sector spending) % 86.4 79.2 74.0 19 Salary costs (share of total recurrent spending) % 0.0 0.0 0.0 20 Rural investment Spending (share of total WS&S investment spending) % 0.0 0.0 0.0 21 Urban investment Spending (share of total WS&S investment spending) % 100.0 100.0 100.0 Notes to Indicator Data Ind. No. Year(s) Comment 10 all The PER provides sanitation access in only one year, using JMP data. National HBS data are much higher (rural 91.7, urban 96.8 in FY2002, rural 90.5 and urban 97.7 in FY2007). 2 all Actual for 2005/06 and 2006/07. Budget figures from 2007/08 98  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa 2003 2004 2005 2006 2007 2008 2009 0.0 0.0 0.0 2.4 4.0 6.5 4.4 0.58 0.54 1.16 0.95 0.00 0.00 0.00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 100.0 100.0 100.0 100.0 100.0 100.0 0.0 0.0 0.0 11.2 12.1 10.3 7.7 0.0 0.30 1.02 2.97 2.22 3.73 4.32 0.00 53.0 80.0 42.0 33.0 93.0 97.4 90.5 18.8 63.4 86.0 73.4 85.4 55.4 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 0.0 0.0 89.4 18.4 23.6 22.4 17.5 16.7 0.0 10.6 81.6 76.4 77.6 82.5 83.3 0.0 0.0 0.0 0.0 31.9 36.6 32.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 100.0 100.0 100.0 100.0 100.0 100.0 0.0 Annexes | 99 Memo Unit 2000 2001 2002 Total population inhbt 35,026,000 35,958,000 GDP Loc. Curr. GDP US$ 0 9,440,900,000 9,758,100,000 General gvt. budget Loc. Curr. General gvt. budget US$ 0 0 0 Total sector budget allocations Loc. Curr. 0 28,715,925,500 26,914,572,800 Total sector budget allocations US$ 0 32,765,339 27,845,078 Domestic sector budget allocations Loc. Curr. Domestic sector budget allocations US$ 0 0 0 External sector budget allocations Loc. Curr. External sector budget allocations US$ 0 0 0 WS&S transfers to sub-national levels Loc. Curr. 0 0 0 WS&S transfers to sub-national levels US$ 0 0 0 Executed sector budget Loc. Curr. 24,886,471,838 23,793,604,668 23,821,237,780 Executed sector budget US$ 31,092,213 27,148,891 24,644,798 Executed domestic sector budget Loc. Curr. Executed domestic sector budget US$ 0 0 0 Executed external sector budget Loc. Curr. Executed external sector budget US$ 0 0 0 Executed recurrent budget Loc. Curr. 3,377,959,838 4,939,597,668 6,194,858,780 Executed recurrent budget US$ 4,220,295 5,636,161 6,409,031 Executed investment budget Loc. Curr. 21,508,512,000 18,854,007,000 17,626,379,000 Executed investment budget US$ 26,871,918 21,512,729 18,235,768 Executed salary costs Loc. Curr. Executed salary costs US$ 0 0 0 Rural Investments Loc. Curr. Rural Investments US$ 0 0 0 Actual WSS expenditure/GDP % 0.00 0.29 0.25 Share of exp that is externally funded % 0.0% 0.0% 0.0% Recurrent exp per capita $ 0.18 Domestic sector budget execution rate % 100  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa 2003 2004 2005 2006 2007 2008 2009 36,930,000 37,945,000 39,007,000 40,117,000 41,276,000 42,484,000 43,739,000 10,282,800,000 11,351,400,000 11,578,700,000 12,783,800,000 0 0 0 0 0 0 0 0 0 0 62,141,902,700 66,182,327,400 152,106,949,300 152,035,837,972 224,683,273,335 396,844,848,026 0 59,842,802 60,754,810 134,735,002 121,444,078 180,463,352 331,723,895 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 17,066,100,000 18,429,200,000 23,126,314,533 30,463,947,759 0 0 15,117,002 14,720,984 18,574,824 25,464,913 0 11,674,499,185 41,964,946,102 130,832,912,171 111,663,849,232 191,932,002,032 219,694,921,171 0 11,242,571 38,523,461 115,890,647 89,195,504 154,157,859 183,643,697 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 10,442,472,207 7,700,793,116 30,854,986,324 25,010,163,051 33,612,905,139 36,583,285,458 10,056,126 7,069,262 27,331,076 19,977,765 26,997,548 30,580,087 0 1,232,026,978 34,264,152,986 99,977,925,847 86,653,686,181 158,319,096,893 183,111,635,713 1,186,445 31,454,199 88,559,570 69,217,739 127,160,311 153,063,610 0 7,971,032,111 12,315,954,664 11,869,534,134 0 0 0 6,367,148 9,892,051 9,921,782 0 0 0 0 0 0 0 0 0.11 0.34 1.00 0.70 0.00 0.00 0.00 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.27 0.19 0.70 0.50 0.65 0.72 Annexes | 101 Togo No. Indicators Unit 2000 2001 2002 WS&S Public Overall Trends 1 Sector Budget Allocation (share of total general gvt. budget) % 0.0 0.0 3.3 2 Sector Budget Allocation (share of GDP) % 0.0 0.0 0.6 3 Domestic sector allocations (share of total sector public allocations) % 0.0 0.0 7.3 4 External allocations (share of total sector public allocations) % 0.0 0.0 92.7 5 Sector allocations transferred to sub-national levels (share of total sector allocations) % 0.0 0.0 0.0 6 WS&S spending per capita US$ 0.0 0.0 0.17 Access to WS&S Services 7 Overall access to water % 8 Access to water – Urban areas % 9 Access to water – Rural areas % 10 Overall access to sanitation % 11 Access to sanitation – Urban areas % 12 Access to sanitation – Rural areas % 13 Gap between province/region with highest & lowest access rate to water % point WS&S Sector Spending Performance 14 WSS Sector Budget Execution Rate % 0.0 0.0 9.7 15 Execution of WS&S sector domestic resources % 0.0 0.0 73.4 16 Execution of WS&S sector external resources % 0.0 0.0 4.7 17 Recurrent spending (share of total sector spending) % 0.0 0.0 25.1 18 Investment spending (share of total sector spending) % 0.0 0.0 66.2 19 Salary costs (share of total recurrent spending) % 0.0 0.0 140.2 20 Rural investment Spending (share of total WS&S investment spending) % 0.0 0.0 52.0 21 Urban investment Spending (share of total WS&S investment spending) % 0.0 0.0 48.0 Notes to Indicator Data Ind. No. Year(s) Comment Report also includes solid waste disposal. Disaggregation on sub-sectors not feasible in all cases. 19 Salary costs are based on allocated budget 20 Based on rural/urban investment distribution in 2008 (??) 102  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa 2003 2004 2005 2006 2007 2008 2009 2.8 4.9 1.2 1.9 0.8 1.0 0.0 0.5 0.9 0.2 0.4 0.2 0.2 0.0 7.0 4.4 17.4 64.9 35.5 19.7 0.0 93.0 95.6 82.6 35.1 64.5 80.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.39 0.61 0.61 1.44 0.18 0.58 0.00 34.0 39.0 29.0 22.9 17.6 71.1 101.9 28.5 54.9 0.0 75.0 83.0 107.8 94.3 80.3 92.4 0.0 18.9 14.7 63.3 116.0 0.0 45.7 0.0 9.2 10.9 10.7 6.5 48.8 13.6 0.0 82.2 84.7 80.7 89.1 0.0 84.9 0.0 175.1 96.4 94.8 91.2 67.8 74.3 0.0 52.0 52.0 52.0 52.0 0.0 52.0 0.0 48.0 48.0 48.0 48.0 0.0 48.0 0.0 Annexes | 103 Memo Unit 2000 2001 2002 Total population inhbt 4,989,988 GDP Loc. Curr. 1,026,248,000,000 GDP US$ 0 0 1,472,403,686 General gvt. budget Loc. Curr. 183,201,877,000 General gvt. budget US$ 0 0 262,847,888 Total sector budget allocations Loc. Curr. 0 0 5,961,943,000 Total sector budget allocations US$ 0 0 8,553,865 Domestic sector budget allocations Loc. Curr. 436,943,000 Domestic sector budget allocations US$ 0 0 626,902 External sector budget allocations Loc. Curr. 5,525,000,000 External sector budget allocations US$ 0 0 7,926,963 WS&S transfers to sub-national levels Loc. Curr. WS&S transfers to sub-national levels US$ 0 0 0 Executed sector budget Loc. Curr. 0 0 579,208,000 Executed sector budget US$ 0 0 831,015 Executed domestic sector budget Loc. Curr. 320,638,000 Executed domestic sector budget US$ 0 0 460,034 Executed external sector budget Loc. Curr. 258,570,000 Executed external sector budget US$ 0 0 370,982 Executed recurrent budget Loc. Curr. 145,638,000 Executed recurrent budget US$ 0 0 208,953 Executed investment budget Loc. Curr. 383,570,000 Executed investment budget US$ 0 0 550,325 Executed salary costs Loc. Curr. 204,163,000 Executed salary costs US$ 0 0 292,922 Rural Investments Loc. Curr. 199,456,400 Rural Investments US$ 0 0 286,169 Actual WSS expenditure/GDP % 0.00 0.00 0.06 Share of exp that is externally funded % 0.0% 0.0% 44.6% Recurrent exp per capita $ 0.04 Domestic-funded sector budget execution % 0.00 0.00 0.73 External-funded budget execution % 0.00 0.00 0.05 104  |  More, Better, or Different Spending? Trends in Public Expenditure on Water and Sanitation in Sub-Saharan Africa 2003 2004 2005 2006 2007 2008 2009 5,147,720 5,313,512 5,487,479 6,469,738 6,660,410 6,859,622 972,643,000,000 1,023,211,000,000 1,113,072,000,000 1,160,112,000,000 1,212,824,000,000 1,427,249,000,000 1,673,507,355 1,936,854,859 2,110,216,541 2,218,653,554 2,530,582,223 3,187,209,132 0 179,181,413,000 199,742,937,000 202,773,652,000 254,101,424,000 259,627,485,000 326,915,062,000 308,295,451 378,097,067 384,428,244 485,955,690 541,718,088 730,038,466 0 5,080,066,000 9,709,206,000 2,469,585,000 4,770,918,000 2,009,680,000 3,233,015,000 0 8,740,646 18,378,734 4,681,960 9,124,131 4,193,239 7,219,690 0 355,066,000 423,206,000 429,585,000 3,094,918,000 713,680,000 637,015,000 610,918 801,094 814,428 5,918,869 1,489,108 1,422,527 0 4,725,000,000 9,286,000,000 2,040,000,000 1,676,000,000 1,296,000,000 2,596,000,000 8,129,727 17,577,640 3,867,532 3,205,262 2,704,131 5,797,163 0 0 0 0 0 0 0 0 1,161,110,000 1,712,201,000 1,754,709,000 4,863,046,000 573,035,000 1,773,743,000 0 1,997,779 3,241,057 3,326,663 9,300,321 1,195,649 3,960,970 0 266,270,000 351,421,000 462,929,000 2,918,906,000 573,035,000 588,343,000 458,138 665,211 877,644 5,582,255 1,195,649 1,313,837 0 894,840,000 1,360,780,000 1,291,780,000 1,944,140,000 0 1,185,400,000 1,539,641 2,575,845 2,449,020 3,718,066 0 2,647,133 0 106,673,000 186,706,000 188,023,000 314,266,000 279,511,000 241,493,000 183,539 353,419 356,463 601,017 583,205 539,281 0 954,840,000 1,450,780,000 1,416,778,000 4,333,780,000 0 1,506,000,000 1,642,876 2,746,208 2,685,997 8,288,128 0 3,363,069 0 186,776,000 179,916,000 178,159,000 286,568,000 189,540,000 179,320,000 321,363 340,566 337,763 548,046 395,479 400,442 0 496,516,800 754,405,600 736,724,560 2,253,565,600 0 783,120,000 854,295 1,428,028 1,396,719 4,309,826 0 1,748,796 0 0.12 0.17 0.16 0.42 0.05 0.12 0.00 77.1% 79.5% 73.6% 40.0% 0.0% 66.8% 0.0% 0.04 0.07 0.06 0.09 0.09 0.08 0.75 0.83 1.08 0.19 0.15 0.63 Annexes | 105 WORLD BANK 1818 H Street, NW Washington, DC 20433