74051       Water Papers Water Papers November 2012 Investing in Water Infrastructure: Capital, Operations and Maintenance Diego J. Rodriguez, Caroline van den Berg and Amanda McMahon 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 Julia Bucknall Contact Information This paper is available online at http://www.worldbank.org/water. Authors may also be contacted through the Water Help Desk at the email address above. Disclaimer – World Bank © 2012 The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org This document is a product of the staff of the International Bank for Reconstruction and Development/the World Bank. The findings, interpretations, and conclusions expressed in this paper do not necessarily re- flect the views of the executive directors of the World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, col- ors, 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 at- tribution 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. A summary of this document was prepared as a contribution to the United Nations World Water Assessment Program. Acknowledgments This document was prepared by Diego J. Rodriguez, Caroline van den Berg, and Amanda McMahon. Research assistance was provided by Olusola Ikuforiji. The authors would like to thank the following for their contributions: Alessandra Campanaro, Alexander Bakalian, and Dominick Revell de Waal for their construc- tive feedback in peer reviewing the document; Julia Bucknall for her support and approval; Graciela Testa for editing; and The World Express, Inc. for document layout and design. This work was made possible by the financial contribution of the Water Partnership Program (WPP) – http://water.worldbank.org/water/wpp. CONTENTS Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii Section 1: Water in an Uncertain World . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Financial Crisis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Climate Crisis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Food Crisis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Energy Shocks and the Green Response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Section 2: Investment Needs and Funding Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Global Investment Needs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Funding Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Public Contributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Official Development Assistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Public Expenditure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Private Contributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 Private Sector Participation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 Private Equity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Households, Geography and the Poor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 Section 3: Financing the Gap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 Barriers to Sector Finance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 The Reform Cycle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Service Providers: Deliver Services More Efficiently . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Non-Revenue Water . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Billing and Collections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 Technology Choice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Managing Uncertainty . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 iv Investing in Water Infrastructure: Capital, Operations and Maintenance Price Water Properly . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 What is Cost Recovery? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 Incentives for Conservation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 Tariff Reform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 Governments: Improve Public Expenditure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 Public Budgets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 Subsidies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 All Stakeholders: Develop Sound Sector Governance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 Governments and Donors: Leverage Resources to Crowd in Private Investment . . . . . . . . 28 Municipal Bonds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 Concessional Loans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Guarantees . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 Pooled and Revolving Funds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 Private Sector Participation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 Section 4: Tools for the Way Forward . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 Public Expenditure Reviews (PERs) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 Lessons Learned from PERs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 Results-Based Financing (RBF) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Section 5: Conclusions and Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Service Providers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Public Expenditure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 Private Contributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Summary A ll countries face a growing funding gap as they try to keep up with the reha- bilitation, operation, and maintenance of aging water infrastructures. New water systems must also be built to cope with growing populations, shift- ing consumption patterns, and a changing climate. In developing countries,1 pub- lic and private investment has not kept up with demand for water infrastructure, which increases costs in the long-run. The financial crisis threatens to bring even more uncertainty to an already underfunded sector with inherently low capacity to attract investment. Private investors prefer to work in middle-income countries where the risk is lower, leaving the poorest countries dependent on volatile public budgets and donor commitments. An estimated 75 percent of water investment in developing countries comes from public sources. Slower growth and lower tax revenues imply that donor commitments, public budgets, and household contributions are at risk of diminishing. As a result, the resources available must be maximized through reforms. The overarching goal of water sector reform is to promote sustainable service delivery by incre- mentally moving the burden of infrastructure finance from the public sector to shared investment by the public and private sector. Under a national reform strategy, all stakeholders should participate in the reform cycle by adjusting their performance, and each will start at a different point in the cycle commensurate with existing challenges and capacities. Successful reform depends on strong leader- ship to coordinate a series of sometimes parallel and sometimes sequenced transformations. Most countries can improve their financial control and ability to weather the current financial cli- mate by helping utilities to move toward cost recovery and by improving public spending. From target- ed subsidies to risk pooling and guarantees, public contributions lay the groundwork for sustainable water service delivery. Improving the way public funds are allocated and transferred will help gov- ernments leverage private finance in the long run. When supported by sound governance frame- works, these contributions can foster mutually beneficial partnerships between the public and private spheres that can help fill the financing gap. Public Expenditure Reviews and Results Based Financing Developing countries are defined here as low-income, low middle-income and upper middle-income countries with 1  2008 per capita GNI less than $12,275, as per the World Bank definition. vi Investing in Water Infrastructure: Capital, Operations and Maintenance are two tools that countries can use to begin generation of water infrastructure is lower main- identifying their most egregious challenges and tenance, less expensive, and more efficient. piloting more sector development programs that Assessment of tradeoffs at the national level and are driven by results. across water using sectors can yield demand The international donor community has an side interventions that are more cost-effective important role to play in brokering public-pri- than new large-scale infrastructure. Such inter- vate collaboration and in promoting the con- ventions can decrease the fiscal burden on poor sideration of greener planning so that the next countries. Introduction T his paper provides background information for development practitioners in the water and other infrastructure sectors. It outlines the major challeng- es related to financing the gap in global water infrastructure, including those systems that provide urban and rural water supply, and sanitation and irrigation services. Water infrastructure finance includes costs for capital works as well as the operations and maintenance costs that motivate sustainable service delivery. The paper seeks to synthesize the extensive body of literature on this subject into a broad over- view, providing some examples of the historical trends in financing, and taking lessons learned from developed to developing countries. Most of the published literature on this topic emphasizes the need for additional financial resources to respond to increasing demand for services. Furthermore, the stud- ies are limited primarily to the water supply and sanitation sector and rely heavily on illustrating means of increasing private sector participation and private financing. In contrast, this paper defines new challenges in the wake of the recent global financial crisis and provides insight into improving the effi- cacy of water supply, sanitation, and irrigation infrastructure finance from public and private sources. Given that in developing countries around 75 percent of investments in water are from public re- sources (loans, grants, technical assistance), this paper emphasizes the importance of efficiencies in the investment and monitoring of public spending. In many instances, additional financial resources will not necessarily result in increased access and better services, but efficiency improvements will re- duce overall financing needs; a crucial factor in this era of financial insecurity. Section 1 introduces the linkages between water infrastructure and growing global challenges, in- cluding food and energy security as well as climate change. Section 2 describes investment needs in the sector and details various traditional funding sources. Section 3 proposes a 5 step reform cycle for making better use of limited funding in the sector. Tools for making these improvements are out- lined in Section 4. The paper concludes with Section 5, a summary of the challenges and recommen- dations for the way forward. A summary version of this paper was published in the 4th edition of the UN World Water Development Report (WWDR4) in 2012. Section 1 Water in an Uncertain World A dequate and well-maintained water infrastructure is a necessary condition for economic growth and poverty reduction. From water supply and sani- tation to irrigation, flood control and hydropower, investments in water in- frastructure need to keep up with global demand. New systems must be built for growing and urbanizing populations, changing consumption and income patterns, and food and energy security demands. At the same time, deteriorating structures require rehabilitation just to maintain current levels of service. Background Research shows that limited access to infrastructure, including those that provide water and energy services, has substantial implications for the poor, ranging from ill health to child mortality (Fay et al. 2005). According to the World Health Organization (WHO), between 85 and 90 percent of diarrheal diseases in the developing world are caused by unsafe water, poor sanitation, and a lack of hygiene ed- ucation (Pruss-Ustun et al. 2002). Women and girls benefit immensely from well-sustained water sup- ply and sanitation services. For example, the provision of latrines in schools increases the enrollment of girls, and improvements in safe water sources frees women from spending hours every day drawing and carrying water home (WaterAid 2005). Thus, the Millennium Developments Goals (MDGs), with their emphasis on safe water, remain cornerstones of human and economic development. Between 1990 and 2010 more than 2 billion people gained access to improved sources of drink- ing water, making drinking water one of the first MDG targets to be met. However, progress towards reaching the sanitation target has been discouraging. Though the number of people with access to im- proved sanitation rose from 43 percent in 1990 to 63 percent in 2010, in many developing countries, especially those in South Asia and sub-Saharan Africa, it has only reached 30 or 40 percent (UNICEF and WHO 2012). Halving by 2015 the number of people without access to improved drinking water and sanitation is a necessary condition for reaching other targets, including poverty reduction. Furthermore, providing the right quantity and quality of water for food and energy production will promote economic growth and stability. As a critical input in production and in services, water is directly affected by the current global financial, energy, and food crises (Winpenny et al. 2009). 2 Investing in Water Infrastructure: Capital, Operations and Maintenance Financial Crisis However, the ability of IFIs to follow through on their commitments is uncertain because it de- According to the Global Monitoring Report 2010: pends on future impacts of the financial crisis on MDGs after the Crisis, the global financial crisis the economies of their donors. Across all sec- increased the number of people living in extreme tors, development assistance fell by 3 percent in poverty by an estimated 50 million in 2009. An 2011, which was the first major drop since 1997. additional 64 million people were more likely to fall into extreme poverty by the end of year the report was published (2010). The report’s worst- Climate Crisis case scenario projects that an additional 100 million people may lose access to drinking water Just as public and private financial flows to the by 2015 (see Box 1). sector are dwindling, climate change is provid- Winpenny et al. (2009) highlights the spe- ing another set of economic challenges. More cific impacts of the crisis on various sources of people are experiencing droughts and floods funding. Public resources are more limited and that impact water quantity and quality. The tariff revenues fall as poverty deepens, weaken- Intergovernmental Panel on Climate Change ing the financial position of utilities, and decreas- (IPCC) expects serious shortages of water in ing their ability to access private finance (i.e. semi-arid regions, which will result in an increase loans, bonds and equity). As predicted, private in the frequency of droughts (Bates et al. 2008). investment (stocks, bonds and project finance) These changes in the spatial and temporal pat- in the water infrastructure and services of devel- tern of water availability make planning for fu- oping countries has taken a hit since 2008. ture water supply more challenging. Projecting In response, international financial institu- changes in runoff and streamflow is complex tions (IFIs) have made a renewed commitment and laden with uncertainty. Incorporating those to increase assistance to the water sector to off- changes into the hydrologic design of water re- set these changes. These institutions play a key sources projects and making real investment de- role in mitigating the risks associated with the fi- cisions is even more difficult. nancial and other crises. IFI commitment to pre- Yet, because the macroeconomic impacts venting more people from falling into poverty has are considerable, countries cannot afford to ig- resulted in more technical assistance, grants, nore hydrological variability. Extreme water and loans in water. For example, the World Bank events affect almost everyone, but the poor suf- Group committed more than US$100 billion in 2 fer the most because of their locations, low in- 2009 to help countries who had cut spending in comes, insufficient infrastructure, and greater services during previous crises to maintain and reliance on climate-sensitive sectors like agri- expand infrastructure. The new funds include a culture. For example, during a three year peri- $10 billion Infrastructure Crisis Facility to man- od, Kenya was hit but an extreme flood that cost age short-term liquidity problems in private its economy 16 percent of gross domestic prod- systems, and the Infrastructure Recovery and uct (GDP), and by an extreme drought that cost Assets Platform, which will provide an addition- 11 percent of GDP (World Bank 2004a). These al $15 billion to many sectors, including water. dramatic losses will only be exacerbated by poor water management. Adapting to these changes may require sub- 2  Unless otherwise noted, all values are in current US dollars. stantial investments in infrastructure and will Water in an Uncertain World   3 The Impact of the Financial Crisis on Access to Improved Water Supply Box 1:   Sources: Three Alternative Scenarios The Global Monitoring Report 2010 created three possible scenarios to analyze the effects that the financial crisis would have on gross domestic product (GDP) growth in developing countries: the post-crisis trend, the pre-crisis (high growth) trend and the low growth scenario. These were used in projecting the percentage of the population in develop- ing countries who would not have access to improved water sources. The post-crisis scenario shows the effects on GDP assuming a relatively rapid economic recovery starting in 2010. This is the report’s base case forecast. The pre-crisis (high growth) scenario shows what the effects on GDP would have been had developing countries continued the impressive growth pattern that occurred between 2000 and 2007. The impact that the crisis had on the MDGs can thus be measured by comparing the post-crisis trend with the pre-crisis trend. The low growth scenario assumes that the things that got worse because of the financial crisis will continue to ad- versely affect GDP in the medium term, resulting in little or no growth for about five years, followed by a slow recovery. Percent of population without access to improved water source 2015 Target 1990 2006 2015 Estimate Pre-Crisis (High  Region       Post-Crisis Growth) Low Growth East Asia and Pacific 16 32 13 3.3 0.6 4.1 Europe and Central Asia 5 10 5 0 0 1.8 Latin America and the 8 16 9 5.4 4.5 7.1 Caribbean Middle East and North Africa 6 11 12 8.3 7.4 10 South Asia 13 27 13 9.3 5.1 10.2 Sub-Saharan Africa 26 51 42 39.1 38.8 39.8 All developing countries 12 24 14 10.1 9.6 11 Source: World Bank (2010a). depend on the capacity and resources available more innovative and more expensive infrastruc- in each country. In order to adapt the water re- ture, such as wastewater reuse, and the desali- sources sector to climate change, the developing nation of brackish aquifers for drinking water. nations would need between $13 and $17 billion annually. This amount would only cover hard in- frastructure for water supply and riverine flood Food Crisis protection (World Bank 2010b). Extreme vari- ability could stretch the infrastructure and in- The food crisis is a result of population growth, stitutional limits of systems that manage water economic growth, water variability, and the surge across sectors and even national boundaries in energy prices. These combine to put upward (Alavian et al. 2009). To meet demand in wa- pressure on food prices in the short term, push- Section 1 ter scarce regions, countries will need to look to ing more people below the poverty line. With the 4 Investing in Water Infrastructure: Capital, Operations and Maintenance global population estimated to increase to 9 billion supplies, which will subsequently affect water by 2050, Hanjra and Qureshi (2010) estimate a use and water costs. First, mitigation policies 3,300 km per year water gap for food production. 3 may require a shift in energy production modal- Producing more food will require sustain- ities, from conventional power plants to renew- able water management systems that use water able energy facilities, which can require even more productively. Rosegrant et al. (2002) pre- more water to generate electricity. Integrating dicts a severe food crisis by 2025 unless fun- water considerations into low-carbon plan- damental policy changes are made that alter ning will be essential. Such a shift could also re- future water use. If current water policies dete- duce local energy supplies in the short run and riorate further, such as declining public invest- increase the price of energy for all, including ment in water infrastructure, 2025 prices for pumping for irrigation and the cost to treat and staple crops, like rice and wheat, could be dou- deliver water services to end users. Second, the ble 1995 levels. Higher prices would lower de- generation potential of hydropower (one such re- mand for food in the long run, with a damaging newable resource), will become unpredictable as affect on nutrition. However, new investments rainfall patterns shift. in irrigation infrastructure and improvements A green economy agenda is being promoted in water productivity can minimize the impact across developed and developing countries alike of water scarcity and partially meet water de- to address these challenges. The aim is to trans- mands for food production (Falkenmark and form the way countries do business, including Molden 2008). the planning and design of new infrastructure. A green economy would rely on fit-for-purpose infrastructure that makes more efficient use Energy Shocks and the Green of natural resources. For example, combining Response physical and natural capital can save costs and help cities guard against natural disasters. Multi- The spike in energy prices between 2003 and purpose projects that deliver benefits to water 2008 had an impact on the cost of water service users, farmers, and energy producers can like- delivery in most countries. Most directly, higher wise help countries cope with low water levels energy prices increase the cost of pumping wa- during droughts and share risks and benefits ter for treatment and conveyance, but they also across sectors. have an indirect impact on building infrastruc- The green economy presents both a chal- ture by making inputs more expensive. Energy lenge for improving the design of our water in- demand has fallen as a result of the financial cri- frastructure and an opportunity to reverse the sis but is expected to rebound as the economy trend of over-consumption. Estache (forthcom- recovers, with developing countries making up ing) argues that low access to water services 90 percent of the demand by 2020 (McKinsey has further degraded natural resources (for ex- and Company 2009). Thus, service providers ample, people without adequate sanitation pol- will remain subject to variable, and probably lute aquifers). At the same time, those without higher, costs for energy inputs. Beyond the cost, access have a unique advantage: they do not the interplay between water and energy is about face the opportunity cost of rehabilitating exist- to get much more complicated. ing infrastructure in lieu of using a new, greener At the global level, changes in temperature technology. In other words, developing coun- and precipitation will have an impact on energy tries have an economic advantage in pioneering Water in an Uncertain World   5 green growth strategies. More developed coun- Fostering green investment in developing tries, on the other hand, will face a larger cost in countries will require strong incentives including switching from old infrastructure to new, more expanding funding, enhancing political will, es- efficient technologies. tablishing a well-defined institutional and regula- But for all countries, green growth will come tory framework and more importantly, initiating at a cost. For example, Korea’s National Strategy policy reforms that would reduce harmful subsi- for Green Growth and its Five Year Plan (2009– dies across and within sectors. It is also critical 2013) is expected to cost 2 percent of GDP to increase global environmental research and (OECD 2011a). Estache estimates the cost of development (R&D), facilitate technology trans- greening infrastructure at an additional 20 per- fer for clean technologies, and design proper in- cent of current sector investment needs for the centive structures that solicit optimal behavioral poorest countries. responses. Section 1 Section 2 . Investment Needs and Funding Sources W ater services provide important economic, health, and environmental benefits but are severely underfunded on a global scale. Funding sourc- es, from private investment to utility revenues and public expenditures, have been inadequate to meet past needs and are becoming scarcer as a result of the financial crisis. Nevertheless, investment needs in the sector are large and in- creasing alongside population growth and urbanization. The financing gap for water, sanitation, and irrigation infrastructure is difficult to estimate, but ap- proximating future costs will help countries and donors activate more funding and help financiers un- derstand the potential market for private investment. Fay et al. (2010) asserts that a thorough analysis of investment needs requires four distinct steps: 1. Understand how much is being spent and how that relates to current quantity and quality of infrastructure. 2. Set a target and have it priced. The infrastructure gap is the difference between current spend- ing and the target. 3. Determine how much of the gap can be bridged through improved efficiency. 4. Look at the balance to see the needed additional spending (financing gap). There are difficulties in each of the four steps. Countries and financial institutions do not account for infrastructure investment in a clear way in national accounts, and inefficiencies in the system are difficult to estimate. Much more information is available on recent capital expenditures than on exist- ing infrastructure stocks. This paper uses estimates from a variety of data sources, and focuses on in- formation related to sub-Saharan Africa, which is a priority region for water infrastructure finance and for which recent data and analysis is available. Global Investment Needs The OECD (2012b) estimates that by 2025 water will make up the lion’s share of global infrastruc- ture investment. For just the OECD countries and Russia, China, India, and Brazil, water spending will 8 Investing in Water Infrastructure: Capital, Operations and Maintenance top $1 trillion that year, nearly triple the amounts While private sector contributions follow needed for investments in electricity or trans- general market principles, public sector contri- port. For developing countries alone, an esti- 3 butions generally seek to promote specific poli- mated $103 billion per year is needed to finance cy objectives, and are therefore provided free of water, sanitation, and wastewater treatment charge or offered at below market rates. Donors through 2015 (Yepes 2008). 4 and governments often try to promote equal ac- In general, low-income countries need to cess to water infrastructure and its services. As and do invest more (about 70 percent of sector a result, their intent is to fund infrastructure for spending) on capital works to reach a larger seg- populations that otherwise cannot access pri- ment of the population. In contrast, investments vate finance. In this regard, the two types of fund- in middle-income countries focus on the oper- ing target different geographic or social spheres. ation and maintenance of existing infrastruc- ture (about 80 percent of sector spending) to achieve sustainable service delivery (Banerjee Public Contributions and Morella 2011). A lack of centralized and reliable information The only comprehensive estimate for wa- makes it challenging to estimate current public ter investment needs at the regional level is the funding to the water sector. Winpenny (2003) es- Africa Infrastructure Country Diagnostic (AICD). timates that the public sector contributes ap- The AICD estimates that to close the infrastruc- proximately 75 percent of total water supply and ture gap in water supply and sanitation (WSS) sanitation infrastructure costs.5 Public contribu- and meet the corresponding MDG targets in tions include external donor funding and central Africa within 10 years, annual investment of ap- and local government budgets. proximately $22 billion, equal to 2.58 percent of GDP, is required. Nearly $15 billion of this is need- Official Development Assistance (ODA) ed for capital expenditure, and the remaining $7 Official development assistance (ODA) includes billion for operational expenditures. In irrigation, grants, low interest loans, and technical assis- the study estimates that it will take $3.4 billion tance from donors and international financial in- per year to attain the region’s goal of doubling stitutions to developing countries. According to the amount of land under irrigation, with 85 per- the Development Assistance Committee (DAC) cent of the total going to capital works (Foster of the OECD, ODA for water and sanitation has and Briceño-Garmendia 2010). been rising sharply, from average annual com- mitments in 2002–2003 of $3.3 billion to $8.3 billion in 2009–2010 (last period reported by Funding Sources 3  The electricity and transport figures are global. The wa- Funding for water infrastructure is mostly paid ter figure only includes OECD countries and the BRIC 5. for by current or future water users or current or 4  The estimate includes low-income and low- to mid- dle-income countries. Analysis is based on a “top-down future taxpayers (including taxpayers in donor approach of using data on infrastructure services and countries). Public sector contributions comprise parameters for construction and maintenance costs to model investment needs.� (Yepes 2008). government tax revenues and official develop- 5  In the mid-1990s, the sector’s financial sources were ment assistance (ODA), while the private sector estimated to be as follows: domestic public sector 65–70 percent, domestic private sector 5 percent, international contributes in the form of private debt or credit donors 10–15 percent and international private compa- financing and individual household investments. nies 10–15 percent. Investment Needs and Funding Sources    9 DAC). ODA is a key funding mechanism, espe- and the financial crisis impacts the ability of IFIs cially for countries with low public tax revenues and other donors to finance their loans through to spend on sector investments and/or that are capital markets. As borrowing costs increase unable to attract private finance. A recent study for these lenders, it may become more difficult of 15 countries in sub-Saharan Africa showed for them to lend money at affordable terms or to that donors contribute around 60 percent of to- provide grants to developing countries (OECD tal sector spending (Van Ginneken el al. 2011). 2011b). In fact, in 2011, in the aftermath of the Over the period 2009–2010, aid to water global recession, total ODA to all sectors fell by and sanitation primarily targeted regions most 3 percent. in need of improved access. Thus, sub-Saharan Africa received 26 percent of total aid to the sec- Public Expenditure tor, and South and Central Asia received 21 per- Sovereign and sub-sovereign governments cent. In contrast, at the country level, only 40 budget funds to be transferred to local govern- percent of total funds were given to the poorest ments to pay for the provision of water servic- countries (OECD 2012a). Nonetheless, for poor es on an annual basis. These fiscal transfers are countries with sufficient institutional capacity made possible by tax revenues and can come in (low income, non-fragile), ODA still comprises the form of cash payments, grants, subsidies or the majority of all sector finance, at 0.7 percent guarantees. Local governments themselves can of GDP (Banerjee and Morella 2011). also provide tax revenues to water service pro- While donors play a large role in service de- viders. These local contributions are considered livery, they also pose obstacles to sustainability. public expenditures, and do not include reve- Water infrastructure requires long-term invest- nues paid directly to service providers by their ment, yet donor funds fluctuate between years customers. and are often committed on a project basis. 6    While funding needs for the water supply This inconsistency is due to the nature of devel- and sanitation sub-sector average an estimated opment assistance, which is often granted in ac- 2.58 percent of GDP per country in sub-Saharan cordance with specific policy objectives, political Africa, countries often spend much less. A study environments, and recipient country capacities, of 15 countries in the region showed that, on av- all of which are subject to extreme variation from erage, they committed 2 percent of their nation- year to year. al budget to WSS. However, a 2008 sample of Donors, however, also provide high value countries showed that only 66 percent of the do- technical assistance to regulators and service mestic WSS budget was executed, and on aver- providers to improve technical, management, age, expenditures between 2000 and 2008 were and administrative capacities. Technical assis- equivalent to about 0.32 percent of GDP (Van tance is generally paid through grants or the Ginneken et al. 2011).7 grant components of low interest loans, and The majority of public expenditure goes to can be extremely valuable for improving local in- urban areas to the detriment of rural populations. stitutional capacity to carry out policy reform, Many countries have passed decentralization execute budgets, manage finances, and de- sign and implement service and infrastructure 6  In 2010, for example, 76 percent of ODA in WSS was improvements. provided on a project basis. About half of ODA to water supply and san- 7  Includes central and local government expenditures on domestic resources and grants or loans from exter- Section 2 itation is in the form of loans rather than grants nal funding agencies. 10 Investing in Water Infrastructure: Capital, Operations and Maintenance laws in an attempt to give rural communities The total value of private activity fell from authority over planning and implementation of $58 billion in the decade of the 1990s to $29 bil- their own service delivery systems. In reality, de- lion between 2001 and 2010. The decline in pri- centralization policies have left many local gov- vate investment was possibly caused by two ernments without sufficient funding from central factors: a paradigm shift in investment and the governments to carry out their mandate. financial crisis. The paradigm shift resulted in The lack of funding has important implica- more investments with smaller values. For ex- tions for countries with large rural populations. ample, in 1997, the median capacity of a new In Mozambique, for example, where two-thirds of treatment plant with private investment was 300 the population is rural, only 12 percent of the WSS cubic meters per day and by 2010 it had dropped budget is spent outside of cities. Similarly, while to 40 cubic meters per day (Perard 2012). At $18 million was budgeted for WSS in Mali between the same time, and perhaps for similar rea- 2000 and 2006, only $250,000 or one-tenth of sons, funding dried up during the financial cri- one percent of the budget was transferred to sub- sis: the number of new projects opening peaked national governments. Each of the nine regional in 2007, and had fallen by two thirds by 2010 (see offices mandated to provide rural water services Figure 2). Furthermore, the financial crisis limit- received only 22 percent of their recurrent cost ed the amount of financing available and lenders budget (excluding salaries) from the central gov- became less tolerant of risky investments, water ernment (Van Ginneken et al. 2011). being one of them. Along with the decrease in overall invest- ment, lending has become more concentrated, Private Contributions both in terms of where money goes and which Like public funding, private investment is also countries it comes from. Since 2001, the top 20 difficult to track as there are various types of sponsors (countries investing) have provided 46 finance (debt, equity, project finance, micro- percent of private investment. The majority of finance) and no source that provides compre- funds go to China, Latin America, and the Middle hensive data on all private funds. This paper East and North Africa. There has been no private uses information from the Private Participation activity in Europe and Central Asia or Africa since in Infrastructure (PPI) database managed by the 2008 and in South Asia since 2010. World Bank. 8 In 2011, the most recent year reported by PPI, there was an 8.3 percent increase in lend- Private Sector Participation ing and 24 percent more projects than the pre- Over the last 10 years, private activity 9 in the wa- vious year. However, the increase again showed ter sector in developing countries has averaged high concentrations in both the regions where $2.5 billion annually, or about 3 percent of the investments needed for water supply and sani- 8  The Private Participation in Infrastructure (PPI) da- tation. This is a drastic decline from the decade tabase provides information on 6,000 infrastructure projects dating from 1984 to 2010, owned or managed leading up to 2000 which saw several large pri- by energy, telecommunications, transport, or water vate contracts in treatment infrastructure and companies. 9  Private activity is any large scale private investment in utility management, averaging $3.7 billion per water and wastewater infrastructure or its management. year, and sometimes totaling up to $14 billion per This includes a variety of contract types carrying varying degrees of risk for the private investor or private opera- year when new and existing private activity is in- tor, including concessions, divestitures, greenfield proj- cluded (See Figure 1). ects, and management and lease contracts. Investment Needs and Funding Sources    11 projects are implemented, and the sponsors Figure 1:  Investment Commitments in Water Projects providing the investments. China was the largest in Developing Countries by Region, 1990–2011 sponsor of private activity in the world in 2011, Projects mostly investing in domestic treatment plants, 90 and Mexico and Spain were heavily involved in 80 several projects throughout Latin America, pro- viding about a third of all investment. 70 Overall, private participation in water is 60 concentrated in a handful of projects in mid- 50 dle-income countries. Compared to other in- frastructure sectors (energy, telecom, and 40 transport), the water sector represents just 1 30 percent of total private investment in develop- 20 ing countries. Moreover, private participation is 10 coming more often from domestic, rather than international sources. Jimenez and Perez-Foguet 0 1990 1995 2000 2005 2010 2011 (2009) estimate that the portion of private inter- national flows to water supply and sanitation fell EAP ECA LAC MENA SA AFR by 6 percent between 1995 and 2005, while lo- cal private flows expanded by 10 percent. This Source: World Bank and PPIAF    Reproduced from Militaru 2012. trend is expected to continue following the glob- al financial crisis. Private Equity Figure 2:  Water Projects with Private Participation, The recent financial crisis has also kept inves- by Region, 1990–2011 tors from purchasing equity in water compa- 2011 US$ billions nies. Over the course of 2008, an index of Asian 16 water stocks dropped by 47.5 percent (OECD 14 2011b). That same year, the OECD reported that developing country water companies, such as 12 Maynilad in the Philippines and Nova Cerae in 10 Brazil, had to postpone their initial public offer- 8 ings due to poor market conditions. Uncertainty about the future supply of water 6 *Adjusted by resources and changes in demand may make in- US CPI 4 vestors even more apprehensive about investing in water infrastructure. A 2010 study by Water 2 Asset Management (a global equity investor in 0 public and private water-related companies and 1990 1995 2000 2005 2010 2011 assets) and the non-profit organization Ceres es- timated the risk to six US water utilities resulting EAP ECA LAC MENA SA AFR from changes in water availability through 2030 Source: World Bank and PPIAF    Section 2 (Leurig 2010). The report informs municipal Reproduced from Militaru 2012. 12 Investing in Water Infrastructure: Capital, Operations and Maintenance bond investors of the potential risks to utility per- levels, as more money goes to capital cities than formance associated with changes in hydrologic rural areas or peri-urban slums. variability. These risks are not currently reflect- In places where rural water and sanitation ed in the bond ratings issued by the three larg- services are not provided through public fund- est ratings agencies. Ceres suggests that ratings ing, households purchase what they need in the agencies endorse over-use of water by reward- private market, often paying much higher prices ing utilities that sell more water (and attain high- than the relatively wealthier customers of pub- er revenues) despite very real supply constraints lic water and wastewater utilities. There are large in the medium term. consumer and producer surpluses and large For example, large populations in Arizona government revenues to be gained by the for- and Nevada rely on Lake Mead as a primary wa- malization of these private water providers, or ter source, but a decade-long drought is reduc- the conversion of their services into piped net- ing available supplies. On the other side of the works and household connections, where econ- country, the city of Atlanta may have to reduce omies of scale can be achieved. supplies by 40 percent as a result of a new judi- Household contributions to water and sani- cial order to make more water available for en- tation infrastructure (i.e. toilets, septic tanks) are vironmental services. While each utility has a also not well documented, but are a large portion different capacity to manage such risks, their of overall investment. Four countries respond- ability to attract financing remains more or less ing to the 2012 Global Analysis and Assessment unscathed because these issues go unreport- of Sanitation and Drinking Water (GLAAS) survey ed. Ceres’ analysis sheds light on the real need reported that household contributions were be- to factor climate risks and uncertainty into long- tween one-third and two-thirds of all sector fund- term planning, financing and tariff adjustments, ing. Another source puts household contributions as part of comprehensive adaptation plans. For in sub-Saharan Africa closer to one-third of all developing countries, failure to address the un- sector funding, at 0.3 percent of GDP (Banerjee certainty of water supplies today will only exac- and Morella 2011). erbate risks and curtail effective service delivery Urban slums receive similarly low levels of for the generations to come. public investment. Unlike rural areas, however, slum dwellers do not own the land on which they Households, Geography and the Poor live and thus do not qualify for access to servic- There are large disparities in funding for water es. Moreover, if these residents want to purchase sector investments, resulting in lower access WSS services from private providers the dense rates for water supply and sanitation in rural ar- layout of slums can be an obstacle for building eas than in urban areas. Public spending exac- below ground infrastructure such as septic tanks erbates the rift between urban and rural service or water pipes. Section 3 . Financing the Gap D eveloping countries need to invest around $103 billion a year to meet sector demands in WSS, on top of the estimated $15 billion needed for climate change adaptation measures. The financial crisis has reduced the amount of money available from both private and public sources and may continue to do so. Countries need to make use of all available resources through more efficient and effective com- binations of funding. Each stakeholder, from service providers to regulators and cen- tral governments, has a role to play in securing long-term sector finance. Each should bring its contributions to the table to leverage the resources of the others. Barriers to Sector Finance In an ideal world, service providers would recover their costs and remain in a healthy financial posi- tion to provide sustainable services in the long run. This would be done by attracting capital from the private sector to make needed investments, and generating sufficient revenue from users to service loans and pay recurring costs. In reality, this rarely happens because the water sector faces several inherent challenges that re- strain private investment in its infrastructure and services. Potential investors perceive many risks in the delivery of water supply, sanitation and irrigation services. First, water is a politicized commodity that is considered a public good on one hand and a high-value economic input on the other. To reach health, economic, poverty or other goals, governments could change water policies and plans at any given moment. Future water policies (whether wastewater has to be treated or drinking water has to be paid for) are uncertain, and tariff levels are subject to political interference. Second, many utilities in developing countries have sub-par operational management. For example, assets are not inventoried, the location of pipe networks is not fully known, and the current and future customer base is undocumented. This makes it very difficult to estimate costs and revenues or design long-term business plans. Given the lack of information and the potential lack of control over price set- ting, local governments and providers do not have credit ratings that investors can use to estimate risk. Finally, most water infrastructure is built for a service period of around 20 years. This means that the upfront costs are large and the repayment periods are long, making them unfavorable to com- mercial lenders. Thus, even when trying to attract finance on a project basis, providers have difficulty 14 Investing in Water Infrastructure: Capital, Operations and Maintenance making the case to financiers that theirs is a The reform cycle has five parts: “bankable� project. The rate at which they could borrow on the private market is too high given 1. Service providers deliver services more the tariff rates they can charge and the inherent efficiently: Service providers must reduce risks of the sector. a range of inefficiencies to increase reve- nues and lower costs. In the case of new infrastructure, this includes selecting the The Reform Cycle appropriate technology (given existing operational capacity and regulatory en- This section proposes a five part reform cycle to vironment) that is affordable to maintain. address these unique water challenges in an ef- 2. Regulators price water properly: Gove­ fort to narrow the financing gap (see Figure 3). rnments must recognize that water ser- The reform cycle should be used as part of an vices will be most sustainable when overarching sector plan that incorporates the treated as an economic good. Regulatory needs of all sector stakeholders. The agenda is a agencies should consider financial, re- circular process, meaning that a stakeholder can source, and environmental costs when enter at any of the steps depending on the exist- pricing water so that the resource ing conditions in a given service area. The parts is conserved for long-term growth. are like moving pieces that require the coopera- Environmental costs should be progres- tion of various stakeholders, and as such, some sively incorporated based on the im- reform steps can be implemented in parallel with provement in the provision of services. others. The majority of successful reform cases 3. Governments improve public expendi- in the water sector have been the result of long- ture: Governments have an important term restructuring of sector institutions, policies role to play in correcting market failures and incentives, supported by strong leadership. to improve access and equity across Figure 3:  The Reform Cycle 1) Deliver services more efficiently 3) Improve public expenditure • Reduce operating • Strategic planning and capital costs • Budget execution • Increase revenue • Subsidy targeting 5) Public resources leveraged Providers move toward cost (bonds, loans, guarantees, recovery and are less reliant pooled/revolving funds) to on external finance • Internalize • Transparency attract private finance externalities • Independence • Keep up with • Accountability inflation • Procurement • Conservation 2) Price water properly 4) Bolster sector governance Financing the Gap   15 income groups. Subsidies can be better overstaffing, high distribution losses, and un- targeted by making them more explic- der-collection of revenues (Foster and Briceño- it and tied to policy objectives. Budget Garmendia 2010). The AICD concludes that transfers can be more predictable, more Africa’s resources would go considerably further strategic, and better executed. if various inefficiencies (amounting to $2.7 billion 4. All stakeholders work to improve sector a year) could be addressed. and corporate governance. This requires Reducing non-revenue water, improving a sound framework for interaction among billing and collections, and selecting the right stakeholders including roles and respon- technology are three of the most important effi- sibilities as well as separation between ciency improvements that any service provider policy formulation and service delivery. can make. All of these are mechanisms for “green- 5. Governments and donors leverage re- ing� existing infrastructure, which at the same sources to crowd in private investment. time free up more revenue to be used for rehabili- In line with their political and institution- tation or to expand capital outlays. These reforms, al capacity, governments should pool risk however, should be conducted as part of a long- and funding needs across local govern- term improvement plan that considers undertak- ments to help them access private mar- ing the other four steps of the reform cycle in order kets at lower rates. The implicit subsidies to ensure the sustainable delivery of service. garnered from municipal bonds and con- While non-revenue water (NRW), billing and cessional loans can be better targeted. collections of tariffs, and technology choice can Donors can continue to provide guaran- have catalytic impacts on cost recovery, there tees and other risk mitigation instruments. are many other efficiency factors in the man- agement and operations of a service provider. When planning for costs savings, it is important Service Providers: Deliver Services to unbundle the source of inefficiencies to deter- More Efficiently mine which ones are the most costly and, hence, Inefficiencies in water provision come in many make the most sense to tackle. different guises and can cause low water pro- ductivity, high rates of drinking water consump- Non-Revenue Water (NRW) tion, and increased flows of wastewater that Non-revenue water is one of the largest sources need to be collected and treated. Beato and Vives of inefficiency in water utilities (see Box 2). Almost (2008) argue that removing technical inefficien- $14 billion is wasted each year in lost revenue. cies (water losses) and managerial inefficiencies This includes, in developing countries, an esti- (in metering, billing and collections, asset man- mated 45 million m3 of treated water that physi- agement, corruption) can generate resources, cally leaks daily from urban water supply systems enhance profits, and reduce the need for new in- and another 30 million m3 that is delivered to con- vestments. These inefficiencies act as an implicit sumers but is not billed due to pilferage, corrup- subsidy that must be gradually removed as long tion, and poor metering (Kingdom et al. 2006). as the marginal benefits to efficiency gains ex- An estimated 40–50 percent of the water ceed the marginal costs of their implementation. produced for delivery in developing countries In Africa alone, water utilities loose approx- is lost as NRW (Kingdom et al. 2006). The ben- imately $1 billion a year due to operating inef- efits from reducing non-revenue water are ob- Section 3 ficiencies associated with poor maintenance, vious; utilities in these countries urgently need 16 Investing in Water Infrastructure: Capital, Operations and Maintenance Reducing Non-Revenue Water in Brazil Box 2:   Inaccurate metering, unauthorized consumption, and leakages all result in non-revenue water. NRW reductions should be considered within the broader context of utility reform in order to ensure that appropriate funds and resources are allocated. The full scope of the problem should be identified at the onset by characterizing the sources of NRW through a baseline assessment. The private sector has much to offer in devising and implementing solutions for different NRW challenges. Private players can develop new technology and provide investments and incentives for project performance. Options range from delegated management under public private partnership (PPP) contracts, to technical assistance contracts and outsourcing of key utility functions to improve performance. Under a performance-based services contract, a private firm’s remuneration is based on enforced operational performance measures. This strategy was adopted by the Companhia de Saneamento Básico do São Paulo (SABEP), the water utility that serves the São Paulo Metropolitan Region in Brazil. The private contractor assisted the utility in improving the production and delivery of water through activities such as better micro metering. This increased reve- nues and reduced the debt load. The outcome of the 3-year contract led to an increase in total volume of metered con- sumption by 45 million m3 and a revenue increase of $72 million. Source: Kingdom et al. (2006) additional revenues to finance expansion and to with system maintenance, achieved a collection ameliorate intermittent supplies and poor wa- rate less than 30 percent (Easter and Liu 2005). ter quality. But such programs require strong Moreover, when local or regional government is institutional capacity and substantial financial the payee some users, such as public entities or resources. Reducing NRW is not only a technical other politically connected enterprises, could get issue. If water tariffs are too low, the costs relat- preferential treatment for their water bill. ed to reducing NRW may exceed the benefits of Autonomous organizations that have strong saving water, as any water “saved� can only be user participation and are transparent in the way sold at very low costs. they set water charges are more likely to achieve high collection rates. The case of the Kyrgyz Billing and Collections Republic demonstrates how when water user as- Improving billing and collections goes beyond sociations collectively pay for shared infrastruc- delivering invoices. Transferring management of ture investments in irrigation, service delivery billing and collections to a service provider that and subsequent farm output improve, creating has financial autonomy will give the provider a di- a virtuous cycle of paying for and receiving good rect financial link to the users, who can leverage quality water services (see Box 3). that relationship to demand better services. In Another key to collections is the ability of the an irrigation scheme in Awati, China, for exam- service provider to disconnect non-paying cus- ple, tying staff salaries to the fee collection rate tomers as an incentive to pay. Tolerance of ar- resulted in a 98 percent collection rate. In con- rears in customer payments or low collection trast, government managed irrigation schemes rates act as an implicit tax on utilities. Likewise, in Nepal, where water fees were provided to the tolerance of pilferage is an implicit subsidy to national treasury rather than the entity charged consumers. These policies reduce the revenue Financing the Gap   17 needed to carry out maintenance or invest in new lower-cost, and often greener, options for man- infrastructure. Forty years of experience in Cote aging demand. For example, large quantities d’Ivoire has shown that a policy of disconnecting of water can be saved in India through the use when in arrears, and quickly re-connecting after of more drip irrigation technologies in order to bills were settled led to a 98 percent collection avoid the exploitation of new raw water sources. rate from residential users (McPhail et al. 2012). But this can only be achieved as long as the ag- ricultural frontier is not expanded, which could Technology Choice lead to an increase in water use. This demon- Technology choice has a significant impact strates the complexity and linkages of policies on the cost of capital expenditures as well as that span one sector. In China, industrial water long-run operations and maintenance costs. reuse could save water and reduce the need to Moreover, as stated in the World Bank’s Inclusive build expensive conveyance systems. Green Growth report, “Infrastructure choices Many of these technologies already exist and have long-lived and difficult-to-reverse impacts are in use in developing countries. Service provid- on the carbon, land and water intensity of future ers, donors and governments should analyze the patterns of development� (World Bank 2012). costs and benefits of new and different technolo- Where water infrastructure is built, industry and gies, and promote better alternatives to conven- populations tend to follow. That is why the de- tional water infrastructure where sensible. This cision to use one technology over another can can be difficult in countries where procurement have implications for several generations of wa- rules prevent entry of new engineering and con- ter users, and requires consideration of social, fi- struction firms, where sourcing new materials is nancial, and environmental costs. cost prohibitive, or where technical capacities ex- The cost of supplying water and sanitation ist for only a limited number of technologies. Many services varies widely with the level of service countries also face corruption and nepotism chal- provided, especially in rural areas where popula- lenges given historical ties between politicians tion densities are lower and transport costs are and suppliers and developers of infrastructure higher. These large cost differentials, coupled and services. Improved governance and trans- with the fact that an expensive, high-quality ser- parency in procurement, and in some cases re- vice is likely to be used only by richer consumers, form of procurement rules, will be required to offer a rationale for providing a minimum level enable the use of new solutions (technologies, de- of service to consumers. This means, for exam- sign, and management) for water infrastructure. ple, constructing standposts in lieu of household water connections, or improved latrines in lieu Managing Uncertainty of septic tanks. If such a minimum service level As shown in Box 4, demand management mea- strategy is used, overall spending needs for low- sures can help service providers achieve effi- income, non-fragile countries in Africa could drop ciencies at low cost in the short term. These by nearly 3 percent of GDP, and could reduce the improvements can have the added benefit of funding gap for the whole of sub-Saharan Africa making utilities more resilient to the impacts of by 64 percent (Banerjee and Morella 2011). climate change. However, without such planned Technology can also be transformational in efficiency improvements, some utilities have to filling the water gap for a green economy (see resort to ad hoc measures, like rationing water Box 4). While new treatment systems and dams or interrupting service when faced with a disas- Section 3 can supply more water, there are a variety of ter. According to a 2010 study, this can alienate 18 Investing in Water Infrastructure: Capital, Operations and Maintenance customers, reduce revenue, and increase costs. and recommend they take an integrated approach The study, Climate Change and Urban Utilities: to planning that relies on flexible designs and the Challenges and Opportunities, provides sur- use of climate action plans to mitigate risk. vey data from 20 large water utilities around the world. It shows that while utility managers are concerned about decreased surface water avail- Price Water Properly ability and water quality, and would prefer not Water is a scarce resource, which when deliv- to take ad hoc measures, they also lack the re- ered as a service, should be sold as an econom- sources needed to integrate climate change into ic good. Pricing water to reflect the marginal cost their planning (Danilenko et al. 2010). of service delivery creates a market where people Utilities need to take a long-term approach and industries are willing to pay, and service pro- to climate change, which often requires supply viders can afford to meet demand in the long run. side measures as well as demand management. The optimal way to price water is by using a Adapting to these changes requires planning in- cost recovery model, whereby service providers frastructure to meet future demand in addition can turn a profit through selling their services, and to protecting against potential scarcity or abun- re-invest revenues in long-term system mainte- dance of water. This could require investing in nance and rehabilitation. Tariff regimes that do not new raw water sources to diversify the resource allow for cost recovery provide an implicit subsidy base, expanding treatment facilities to accommo- to consumers that can distort market incentives. date larger flows, or using desalination, recycling However, including economic incentives in or multi-purpose storage facilities. The authors water pricing has proven difficult. Many people of the study warn service providers not to use cli- argue that water should be provided for free or mate change as an excuse to over-design systems, at a price that is below its real financial, resource, Box 3:  Water User Associations: An Essential Component to Improving Cost Recovery in Irrigation Systems in the Kyrgyz Republic An on-farm irrigation project was implemented from 2000 to 2008 in seven administrative regions of the Kyrgyz Republic. The project was aimed at increasing crop production through reliable and sustainable water distribution. On-farm infrastructure was also rehabilitated under the management of water user associations (WUAs). Members from each participating WUA signed an agreement to repay 25 percent of the on-farm rehabilitation costs, raise irrigation fees to support the operation and maintenance activities of their associations, and pay the wa- ter supplier an irrigation service fee for water delivered to the WUA’s head gate. Considerable success was achieved: • Performance of targeted WUAs (166,000 members), managing about 710,000 hectares (70 percent) of the country’s irrigated land, was improved. • Infrastructure that fed 120,000 hectares was rehabilitated, and water delivery to farmers in 80 percent of the rehabilitated systems now closely matches irrigation water demands. • Three agricultural seasons later, irrigation service fees had doubled on average and collection rates by WUAs amounted to close to 100 percent of total assessed fees. • Overall cost recovery for operation and maintenance increased from about 20 percent to 60 percent and at least 80 percent of water users were found to be satisfied with the performance of the WUA. Source : World Bank (2008) Financing the Gap   19 Box 4:  The Cost of Agricultural Water Demand The 2009 report, Charting our Water Future: Economic frameworks to inform decision-making, is a study carried out by the 2030 Water Resources Group and led by the International Finance Corporation and McKinsey & Company. It pro- vides an evaluation of the scale of the water challenge, estimating that by 2030, global water requirements may grow by over 40 percent from 4,500 billion m3 to 6,900 billion m3. Demand for agricultural water, which currently accounts for 71 percent of water used, is expected to rise from 3,100 billion m3 to 4,500 billion m3 by 2030. In percentage terms, this will be a slight drop (to 65 percent of 2020 global water withdrawals), but water for agriculture remains a major challenge. Agricultural demand is projected to be most significant in the poorest regions, such as India (1,195 billion m3), sub-Saharan Africa (820 billion m3), and China (420 billion m3). Closing the water supply–demand gap has financial implications. According to the report, the cost of closing the 2030 gap will range from $0.10/m3 to over $0.50/m3. Without a new, balanced approach, these figures call for an ad- ditional annual investment in upstream water infrastructure of up to $200 billion, which is more than four times cur- rent expenditure. India – Water availability cost curve Cost of additional water availability in 2030 $/m3 National river linking Specified deficit in project (NRLP) between supply and Pre-harvest treatment water requirements 2030 Municipal dams Gap in 2030 = 755,800 million m3 Deep groundwater 0.80 Ag. rainwater harvesting Cost to close gap = USD 5.9 billion Aquifer recharge small 0.10 Infrastructure rehabilitation Large infrastructure 0.08 Shallow groundwater Rain-fed germplasm Wastewater reuse 0.06 Irrigated integrated plant stress mgt. 0.04 Irrigated germplasm Incremental Drip irrigation 0.02 availability 0 Billion m3 –0.02 250 500 750 1,000 1,250 –0.04 –0.06 Industrial levers Increase fertilizer use Rain-fed drainage Reduce transport losses Desalination (thermal) Irrigated drainage Sprinkler irrigation Rain-fed fertilizer balance Artificial recharge Desalination (reverse System of rice intensification Small infrastructure osmosis) (SRI) Genetic crop development – rain-fed On-farm canal Ining Irrigated fertilizer balance Rainfed integrated plant stress mgt. Post-harvest treatment Reduced over-irrigation Last mile infrastructure Rainwater harvesting No-till farming Genetic crop development – irrigated Municipal leakage Agriculture Industry Municipal & Domestic Supply Various interventions for meeting India’s 2030 water demand are presented in the cost curve above. Each inter- vention is represented by a block. The width of the block shows the amount of water that will become available as a re- sult of the intervention, and the height shows the cost per cubic meter. The least cost options (on the left side of the graph) comprise demand-side efficiencies in irrigation. Existing technologies, such as drip and sprinkler irrigation, can be used to save water, so that the more expensive supply-side interventions (building treatment systems and dams) are not needed. The cost curve for China, on the other hand, demonstrates how industrial water use efficiencies can save significant resources at little cost. Each country has to find the appropriate mix of interventions to fund the gap. Source: 2030 Water Resources Group (2009)    Reproduced from Water Resources Group 2009. Section 3 20 Investing in Water Infrastructure: Capital, Operations and Maintenance and environmental cost. Historically, water ser- maintenance costs, down from 111 percent in vices have been systematically underpriced. 2000. This is shown in Table 1, where the global As a result, achieving full cost recovery solely median operating cost coverage ratio (total an- through user charges will now require large tar- nual operational revenues/total annual operat- iff increases that are politically difficult to man- ing costs) is provided. age. However, if service providers become more A large number of utilities with an operating profitable they will be able to offer better quali- cost coverage ratio equal to or less than 1 (1 be- ty services, expand their market, and potentially ing breaking even) increased from 35 percent in cut marginal costs across the board. 2000 to 43 percent in 2008. Over the same time period, average operation and maintenance costs What is Cost Recovery? more than doubled, rising from $0.31 to $0.66 per Cost recovery is the ability of a service provid- cubic meter. Most of the increase happened after er to take in sufficient revenues from customers the fuel crisis when energy costs grew to 4 per- to cover their current and some of their future cent as a portion of total operation and mainte- costs. 10 These include operations and mainte- nance costs (Van den Berg and Danilenko 2011). nance costs (to deliver the service) as well as While most countries do recover their operat- capital costs (including recuperation of asset de- ing costs, many do not, and there is wide variation preciation over time and savings to pay for future among countries. Figure 4 provides a snapshot capital investment needs). of operating cost coverage ratios across regions, Water fees are collected from users for two using the most recently reported data in IBNET. main objectives (see Box 5). The first objective Latin America and the Caribbean and East Asia is to cover the direct financial cost of the service and the Pacific have the highest recovery rates. to guarantee sustainable services. These direct Considering the average revenues they gener- costs cover basic operation and maintenance of ate (see Figure 5), Africa and Europe and Central the service, the renewal of existing infrastruc- Asia have relatively low cost recovery. South Asia ture, and the possible capital expansion of water has the lowest revenue per cubic meter sold and, services. In many countries, most utilities and ir- rigation agencies charge only a fraction of these 10  The long-term financial sustainability of a service pro- vider, or its ability to meet operations and maintenance, direct costs to users. The median utility in the and capital costs, depends on recovering costs from us- developing world barely covers its basic opera- ers and/or receiving predictable and sufficient public/ donor funds. tion and maintenance costs.11 In 2008, operating 11  Most of the utilities participating in IBNET are from de- revenues covered 105 percent of operation and veloping countries. Table 1:  Median Operating Cost Coverage Ratio in Utilities Year 2000 2001 2002 2003 2004 2005 2006 2007 2008 Operating cost coverage 1.11 1.13 1.10 1.11 1.08 1.07 1.07 1.08 1.05 ratio Standard deviation 0.55 0.56 0.58 0.61 0.57 0.56 0.55 0.54 0.50 Number of utilities reporting 579 615 723 999 1,151 1,173 1,379 1,229 930 Note: The data collection cycle of 2008 was not yet complete at the time of publishing the source material.    Reproduced from Van den Berg and Danilenko 2011 Financing the Gap   21 as expected, barely covers its operating costs, Figure 4:  Average Operating Cost Coverage Ratio and the Middle East and North Africa is the worst by Region (2004–2008) performer, bringing in tariffs that are equivalent with the global average, but covering only about 70 percent of operating costs. Developing countries are not the only ones challenged to translate higher costs into higher prices. Even in the United States, where afford- ability concerns are relatively low, some water supply and sanitation utilities do not charge rates high enough to recover costs. Twenty-five percent 0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 of drinking water and 40 percent of wastewater utilities in the country do not charge their con- Africa East Asia & Pacific sumers the full cost of service (Anderson 2010). Europe & Central Asia Latin America and the Caribbean Middle East & North Africa South Asia Cost recovery is perhaps even more elusive Source: IBNET (2012).    in the irrigation sector, where measuring and * 92 countries were included in this data set; Africa (32); EAP (8); monitoring water use proves more difficult at ECA (24); LCR (15); MNA (8); and SA (5). All numbers are as of latest reporting year for each country. Latest reporting year varies from the farm level. Easter and Liu (2005) show that 2004–2009. The ratio is the average ratio for all utilities reporting, which recovering full costs in developing countries is also varies per country. See source for details. rare, with examples ranging from recovery of 12 percent of operation and maintenance costs in Argentina to 70 percent in Tunisia. save water and increase productivity. In addition to tariffs, other demand management measures Incentives for Conservation include quotas and water rights that can be trad- The second objective of collecting water fees is ed between farmers. However, these require more to provide incentives to use water more efficient- sophisticated monitoring and administration. ly. Charging low tariffs can result in unsustain- Water use efficiency will become more im- able levels of water consumption that can cause portant as water scarcity increases. To guarantee the depletion of water resources. In turn, this can economic sustainability, users should be charged raise future production costs as water can only the full supply costs, plus the costs created by be found by drilling at greater depths or convey- any economic and environmental externalities.12 ing from longer distances. Economic externalities include the costs to pro- McPhail et al. summarize the relative advan- ducers and consumers, while environmental ex- tages of different pricing policies in the irrigation ternalities include those imposed on public health sector. Pricing irrigation water based on the vol- and ecosystems. However, quantifying externali- ume of water used rather than the area under irri- ties is difficult and users must be willing to pay the gation can prevent overuse but can also be more additional cost, which is why few systems in the difficult and costly to administer, and can make world include them in their pricing structures. revenues less predictable as users vary water use over time (McPhail et al. 2012). Likewise, charging 12  Externalities can be positive (i.e., benefits) or nega- lower tariffs for water-saving technologies (like tive (i.e., costs). If externalities are positive, the econom- ic costs of the water service are lower than the financial drip irrigation instead of flood irrigation) or for costs; the opposite is true when the externalities are Section 3 growing crops that are less water intensive can negative. 22 Investing in Water Infrastructure: Capital, Operations and Maintenance concluded that for the WSS projects, the factor Figure 5:  Average Revenue per Cubic Meter Water/ Wastewater Sold, by Region (2004–2008) that contributed most successfully to meeting cost recovery targets was improving collection rates (IEG 2009a). Most often, this involved in- creasing the capacity and willingness of water institutions to collect fees from beneficiaries. Increasing water tariffs also had a discernible im- pact on overall project results. Tariff Reform 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 If they cannot sufficiently reduce costs through efficiency improvements, regulators can work Africa East Asia & Pacific with utilities to identify reasonable tariff increas- Europe & Central Asia Latin America and the Caribbean Middle East & North Africa South Asia es to improve cost recovery rates. The timing and sequencing of such tariff reform is critical to suc- Source: IBNET (2012).    * 94 countries were included in this data set; Africa (33); EAP (9); cessful implementation. In some cases increas- ECA (24); LCR (15); MNA (8); and SA (5). All numbers are as of latest reporting year for each country. Latest reporting year varies from es in tariffs are required before service quality 2004–2009. The ratio is the average ratio for all utilities reporting, which improvements can be made (for example, in the also varies per country. See source for details. case of cash strapped service providers). In oth- er cases, tariff increases can only be justified af- Inadequate cost recovery is not only a func- ter improvements have been made (for example, tion of low tariff levels but also of low collection in cases where customers are already unsatis- rates, unaccounted-for water and other opera- fied with service and are unlikely to pay more). tional efficiencies covered earlier. A 2009 evalu- Raising tariffs when service quality is low ation of water sector projects conducted by the can be an especially difficult task, but it has been Independent Evaluation Group of the World Bank done. Colombia in 1990 faced severe problems Box 5:  European Union Water Framework Directive on Water Pricing Article 9 of the European Union Water Framework Directive (WFD) required member states to adopt water pricing pol- icies by 2010 that provided adequate incentives for efficient use of water resources. Water service costs include envi- ronmental costs, and are based on the polluter pays principle whether that user is industry, agriculture or households. The principle aims to mitigate environmental problems through reliance on economic efficiency. The WFD’s concept of cost recovery includes two levels: 1) financial; and 2) environmental and resource costs. • Financial costs (the full cost of supply): This includes the costs of providing and administering water servic- es. It includes all operation and maintenance costs as well as capital costs. • Resource costs: These represent opportunities lost to other uses as a result of the depletion of the resource beyond its natural rate of recovery (for example, losses linked to the over-abstraction of groundwater). • Environmental costs: These refer to the costs of damage that water use imposes on the environment and ecosystems (for example, aquatic ecosystems can be damaged). Source: Garrido and Calatrava (2010); Francois et al. (2010); Commission of the European Communities (2000). Financing the Gap   23 with cost recovery in urban utilities and at the should still be increased to account for inflation. same time needed large-scale capital invest- Not keeping up with inflation can create a per- ments to keep up with demand. A bold decision verse incentive to consume more water as pric- to raise tariffs incrementally from $0.33 per cu- es per cubic meter drop in real terms. bic meter in 1990 to $0.78 in 2001 created sus- In Mali and Madagascar, proper pricing of tainable service delivery in the long run. The WSS tariffs could improve revenue by an esti- price increase for customers caused a large drop mated 1.2 percent of GDP (Banerjee and Morella in consumption, from 34 to 19 cubic meters per 2011), and comprises nearly 80 percent of all po- household per month, which meant that new ma- tential efficiency improvements. Governments, jor capital outlays could be postponed or elimi- however, are under political pressure not to raise nated. The government also used revenues from tariffs to cost recovery levels, and often believe wealthier households and industry to cross-sub- that customers are either unwilling or unable to sidize poorer households (World Bank 2012). pay the full cost of service. Recent evidence from Like Colombia, many developing coun- one of the world’s poorest regions, sub-Saharan tries are raising tariffs to cover increasing costs, Africa, suggests that this is not the case. In low- as data from the International Benchmarking income countries where 10 percent of the pop- Network for Water and Sanitation Utilities ulation is connected to a network, an additional (IBNET) database show. Between 2000 and 30 percent of the population could afford to pay a 2008, the average revenue per cubic meter of connection fee as well as a fee of $10 per month water sold (a proxy for tariffs) nearly doubled (with water priced at $1/cubic meter). The re- from $0.37 to $0.71 in the utilities participating maining 60 percent of the population could pay in IBNET (See Table 2). As discussed earlier, op- $6 per month (Banerjee and Morella 2011). Many eration and maintenance costs have doubled potential network customers are already paying over the same period, and these costs are be- much more than this for water supplied by pri- ing passed on to customers in the form of higher vate tanker trucks. This data suggest that more tariffs. These reforms are also helping to reduce analysis is needed to understand market condi- consumption. Between 2000 and 2008, con- tions in these countries and advocate for extend- sumption in low-income countries fell sharply ing networked services where financially feasible. from 138 liters per capita per day (lcd) to 75 lcd Another way to achieve cost recovery (Van den Berg and Danilenko 2011). through tariffs is to cross-subsidize among dif- Even when nominal operating costs re- ferent classes of water users. Cross subsidies, main the same from one year to the next, tariffs whereby industrial users pay more to support Table 2:  Average Revenues per m3 Water Sold (in US$) – Median Values 2000 2001 2002 2003 2004 2005 2006 2007 2008 Average revenues 0.37 0.34 0.28 0.32 0.37 0.43 0.50 0.63 0.71 Standard deviation 0.34 0.34 0.37 0.42 0.47 0.50 0.53 0.59 0.51 Number of utilities 567 632 725 982 1,137 1,154 1,188 1,203 878 reporting Note: The data collection cycle of 2008 was not yet complete at the time of publishing the source material.    Section 3 Reproduced from Van den Berg and Danilenko 2011. 24 Investing in Water Infrastructure: Capital, Operations and Maintenance residential use, are quite common in WSS, and the reach an explicit agreement on who pays for the direct fiscal repercussions are small. The IBNET uncovered part of the costs of the water services. shows that high levels of cross-subsidies tend to Without such agreements governments risk an be more common in low-income countries than implicit deferral of the real costs of water services in middle-income countries. It also notes that tar- into the future, seriously hampering the short- and iff levels are around 1.35 times more per cubic medium-term sustainability of the services. meter for non-residential users than for residen- In developing countries, most of the financ- tial users (Van den Berg and Danilenko 2011). ing gap is filled by public sources, including utility Whether cross-subsidies can work depends and consumer subsidies. However, these sources largely on the existing tariff structure and re- of revenue are less predictable than tariffs from quires willingness to pay, or lack of a cheaper users because they are affected by a host of ex- substitute, on the part of non-residential water ternal factors. Moreover, government transfers users. Depending on the market structure, some are a mechanism for realizing national planning service providers must take great care in retain- and policy objectives, and have varying degrees ing their high paying users. In 2000, Uganda’s of efficiency in spending. Nonetheless, if well-im- largest utility risked losing its industrial custom- plemented through sound targeting and predict- ers because tariff rates were 230 percent higher able budgeting and execution, public expenditure than residential rates. In order to maintain an op- can provide the cash flow needed to jump-start timal customer base, the utility reduced the in- reforms that can ultimately reduce reliance on dustrial tariff rate gradually as it increased the public funding by attracting private investment. residential rate (USAID 2005). Tariffs help ensure quality service delivery Public Budgets by creating a fair market value for water and by As discussed in Section 2, many central govern- keeping service providers accountable to their ments do not follow through on planned com- customers. In addition, tariffs are often the most mitments to the sector. One reason that budgets concrete revenue source for a service provider fail to be implemented is that they are rarely and thus offer predictable cash flows, which are linked to policy objectives, such as poverty re- not provided by ODA, fiscal transfers or subsidies. duction or economic growth. Public Expenditure Reviews conducted in several African countries show that the budget process is top-down, with Governments: Improve Public the Ministry of Finance rarely consulting line Expenditure ministries while formulating the allocations. This When consumers do not pay the full cost of the means that budgets are not the result of fore- service (either because of tariff levels or an inabil- casts of the investments needed to meet policy ity to collect the fees), the finance gap must be goals, but from competition between water and bridged by contributions from future consumers, other infrastructure, military, and social sectors. current or future taxpayers, or a combination of A second challenge is timing. Budgets are these. Deciding how to allocate the costs of servic- approved on an annual basis and unspent funds es to different groups of consumers and taxpay- often cannot be used once the budget cycle is ers depends on political preferences, but also on concluded. Lengthy procurement processes, the structure of the water market in the area ser- bureaucratic budget approvals, and low capacity viced by irrigation water, drinking water or waste- to implement budgets are also obstacles to time- water collection and treatment. It is important to ly budget execution (Van Ginneken et al. 2011). Financing the Gap   25 Regardless of planned commitments, budgets Service Provider Subsidies themselves can fluctuate because they are gener- Capital subsidies are relatively common in devel- ally a percentage of GDP, a value that, of course, oped countries. In the European Union, utilities varies with economic growth. Therefore, even if receive generous capital grants to assist them in budgets were fully executed, the value of the funds complying with stringent wastewater standards. transferred would change from year to year, mak- In the United States, up to 75 percent of the value ing it difficult for local governments to plan for the of much of the water and wastewater infrastruc- future, especially in the case of capital investment. ture built in the 1960s and 1970s was granted The impact of the financial crisis on GDP is expect- by the federal government under the Federal ed to reduce the rate of access to water supply and Construction Grants Program. Today, to the con- sanitation infrastructure (see Box 1). trary, local governments generally pay 98 per- Larger government transfers, however, will cent of total costs for building and running water not necessarily result in improved access to sus- and wastewater infrastructure, and most of the tainable water services. The efficiency and effec- money comes from user fees (Anderson 2010). tiveness with which they are managed is a critical The remainder is covered by a small volume of factor in whether or not services are sustainable. capital grants through agencies like Housing Countries must look at the incentives and po- and Urban Development, the Department of tential bottlenecks in fiscal and public finance Commerce, and the Department of Agriculture, models. The discussion of subsidies that follows with most of the funds going to rural communi- demonstrates how much of public expenditures ties (Anderson 2010). on water supply and sanitation fall short of ad- The evolution of the water supply and san- dressing market failures and should be re-target- itation sector in the United States shows how ed for different, more effective purposes. capital subsidies can be transformational for meeting policy objectives. If phased out over Subsidies time, they can free up government resources for Public contributions can also be called subsidies other purposes, while making service provision and can be provided either to the service provid- more sustainable.14 er (service provider subsidies) or to the consum- Operating subsidies create severe distortions er (consumer subsidies). Utility subsidies include in both the production and consumption of water cash transfers, grants, and various types of credit services. Operating subsidies should be used as and credit enhancement mechanisms. Consumer a short-term stopgap measure to help with emer- subsidies are essentially discounted tariff rates or gency cash flow concerns and should be tied to connection fees provided to consumers. This sec- performance improvements. Otherwise, local tion focuses on explicit capital and operating sub- authorities come to expect annual transfers irre- sidies to utilities and consumers. spective of yearly performance, having little to no Global water subsidies are estimated at incentive for service improvements, and by ex- $200 to $300 billion per year (McKinsey and tension, sustainable service delivery. Company 2011) as stated in World Bank 2012).13 While generally intended to promote affordabili- 13  This includes only direct cash payments to producers. ty and equity across income groups, subsidies in 14  There are, however, still grave concerns that current developing countries have proved to be regres- levels of service cannot be maintained without more subsidies to urban areas given that much of the infra- sive, benefiting a relatively small, well-off group structure is now near the end of its useful life and new in- Section 3 of consumers. vestments are needed (Anderson 2010). 26 Investing in Water Infrastructure: Capital, Operations and Maintenance Energy is a large part of operations costs for subsidy programs showed that quantity-based both irrigation and WSS services. Energy subsi- subsidies, the most common type of consumer dies lower the cost of energy inputs for the pro- subsidy, do not reach the poor customers they ducer to generate and distribute drinking water, target (Komives et al. 2007). Subsidies that are pump and treat wastewater, or deliver water for geographically targeted have a better record of irrigation. Governments offering electricity at reaching the poor. The AICD confirms this trend below market prices can have an enormous im- by concluding that in Africa, around 90 percent pact on the operations and maintenance costs of those with access to piped water are among of treatment infrastructure and can motivate the wealthiest 60 percent of the population. In over-pumping of groundwater for irrigation. In such an environment, any subsidy to piped wa- India, for example, where agriculture is a main- ter services is largely captured by better-off stay of the economy and electricity makes up a households. large share of overall irrigation costs, the gov- Connection subsidies, on the other hand, are ernment is in favor of providing cheap electric- often targeted to poor households who live close ity to farmers. Lee et al. (2001) estimate that in to an existing system. Evidence from a study in India subsidized electricity for irrigation pumps West Africa suggests that in terms of reaching comprises $4 million of the total $800 million in the poor, subsidizing connections is better than subsidies that go to agricultural producers each subsidizing consumption. However, connection year (as referenced in Ashley and Cashman subsidies can have negative impacts on the vol- 2006). The result can be over-abstraction of ume of wastewater generated and the ability of groundwater and inefficiency in pumping. some poor users to pay subsequent water bills. Operating subsidies will be of greater con- They also cannot necessarily be extended to the cern once developing countries start spending poorest households who have no land holdings a larger portion of total sector funds on opera- (Debomy et al. 2005). tions and maintenance. A sample of countries in In summary, capital and operating subsidies sub-Saharan Africa have shown that recurrent each have their advantages and disadvantages, expenditures make up just 12.7 percent of total and can address different types of market fail- sector spending, with the remainder going to the ures. However, subsidy proliferation should be ap- development of capital works (Van Ginneken et proached with caution, taking into account past al. 2011). Moreover, half of the recurrent costs performance and central and local government are salaries, meaning that just over 6 percent of capacity to affect outcomes. When provided, all sector funds are spent on actual system op- subsidies should be transfered on a regular ba- erations and maintenance. This can lead to de- sis so that service providers and consumers can ferred maintenance, requiring more expensive plan their future cash flow. The World Panel for system rehabilitation in the long run. Financing Water Infrastructure calls for greater assurance that resources for subsidies are bud- Consumer Subsidies geted in advance, as part of a trend towards more Consumer subsidies lower costs for the end “sustainable cost recovery� (Winpenny, 2003). user. In water supply and sewerage systems When large government subsidies are not provid- where coverage is far from universal, consum- ed directly and on a regular basis, water agencies er subsidies benefit only those already connect- tend to postpone maintenance, which shortens ed to the piped system, who tend to be non-poor the expected life of assets, and means that infra- households. A 2007 evaluation of 32 water structure needs to be replaced more often. Financing the Gap   27 All Stakeholders: Develop Sound Official policies are often not fully carried Sector Governance out in countries that have weak governance    Sound governance is part and parcel of the suc- regimes. It is critical to look at the de facto func- cessful implementation of the reform cycle, and tioning of institutions, rather than just at the has a particularly important role in attracting pri- paper policy framework (Locussol and Van vate finance while protecting government inter- Ginneken 2010). ests. Governance has several dimensions, from Separate roles should be created for regula- political stability, rule of law, government effec- tors and service providers. In many countries the tiveness, and regulatory quality, to voice and ac- lines of responsibility are blurred and agencies countability and control of corruption. To promote have similar or identical functions. Policy setting the reform of governance frameworks, donors and oversight should be conducted by the regu- can provide technical assistance on international lator, with water or sanitation delivery performed best practices and governments can provide the by the service provider. This helps to prevent ma- political leadership needed to create a transpar- jor conflicts of interest that can lead to corruption ent and well-functioning governance structure. and other inefficiencies. Because water is a pub- Improving governance structures requires lic good, other rules, such as whether a service identifying the main actors and clarifying their provider can raise capital in private markets or in exact mandate with regard to the key functions foreign currency, are important for safeguarding of: (i) policy formulation; (ii) asset manage- the fiscal positions of local governments. ment and infrastructure development; (iii) ser- Good governance also limits corruption. vice provision; (iv) financing the sector and the Procurement policies for design, construction, development of water infrastructure; and (v) and management contracts should be designed regulation of the service. Clarifying the con- to enable competition among suppliers and en- tractual arrangement under which the parties sure transparent and open bidding procedures. interact with each other, and assessing the ad- All too often entrenched political interests and lo- equacy of the instruments used by the actors cal engineering and construction know-how limit to fulfill their mandates are critical steps in de- the choices available for building and moderniz- veloping successful governance frameworks. ing water infrastructure. Procurement policies Box 6:  Changes in Management Structure Lead Uganda’s Sector Reform The National Water and Sanitation Corporation (NWSC) of Uganda, which supplies water and provides sewerage ser- vices in urban areas, enhanced its financial performance through a series of efficiency improvements. Top-down sector reforms enabled private sector participation in operations and fostered a more commercially-oriented environ- ment at the public utility. In response, NWSC made internal changes to its management structure by providing super- visors with the autonomy they needed to staff positions and allocate other resources. Managers could earn bonus pay for increasing operating income and meeting other performance targets. The number of employees was subsequent- ly reduced from 24 people per 1,000 connections to 10 people per 1,000 connections. NWSC expanded service by 33 percent and improved billing and collections ratios. Moreover, a new culture focused on customer relations improved customer complaint response times and connection speeds. As a result, revenues jumped from 23.7 billion Uganda shillings in 2000 to 40.9 billion just four years later. Source: USAID (2005). Section 3 28 Investing in Water Infrastructure: Capital, Operations and Maintenance have the power to foster the use of new technol- and offering guarantees, and managing pooled ogies and management approaches; the inno- and revolving funds. vation can lead to cost-effective solutions that can put countries on a path to greener growth. Municipal Bonds Procurement and other policies should be de- Service providers, whether in developed or de- signed to suit the specific needs and long-term veloping countries, often lack the creditwor- goals of the country, and should promote efficien- thiness to borrow from commercial banks. To cy and equity in the provision of water services. address this issue, countries have created in- stitutions to intermediate between local gov- ernments and private sector investors. These Governments and Donors: institutions, often called Municipal Development Leverage Resources to Crowd Funds, facilitate borrowing on the private mar- in Private Investment ket by pooling the borrowing capacity and needs In the aftermath of the financial crisis, when all of several local governments and sharing risk. sources of infrastructure finance could poten- Using the combined collateral of several local tially decline, governments and donors should governments the intermediary can then float a focus their resources to build financially solvent bond to raise private capital. In doing so, the in- utilities and sound governance structures. They termediary takes on some of the risk. should also work in concert to build local capaci- Municipal bonds are a mechanism for local ty to plan and execute donor and public budgets, governments to access capital from private sourc- which comprise the majority of utility cash flow es. Furthermore, they can be tax-exempt, provid- in many developing countries. ing an implicit subsidy to the local government. In These fundamental contributions build the the United States, municipal revenue bonds are foundation for the efficient and effective use of the most common financing mechanism for wa- scarce resources and strong institutions will pro- ter and wastewater infrastructure in towns with vide the oversight needed to protect investors a population over 30,000. The interest paid on and service providers. The reform steps taken these municipal bonds is tax-free, providing an will, over time, crowd in more private investment implicit subsidy from the Treasury Department, to help fill the financing gap. which for 2006 was estimated at $2.1 billion, or There are several major contributions that 0.02 percent of GDP (Anderson 2010). donors and governments can make to help Outside of developed countries, municipal jump start private finance in water. These two bonds used to finance water sector infrastructure stakeholders have the convening power to help and operations are rare because they generally bring utilities and private investors to the table. require a sovereign guarantee, which is uncom- Through offering incentives for the public and mon in the water sector. Two exceptions include private sectors to do business, donors and gov- the city of Johannesburg in South Africa, and the ernments can have an enormous impact on im- municipality of Tlalnepantla de Baz in Mexico. proving water infrastructure finance. There are In 2004 the city of Johannesburg issued a many tools that provide incentives by lowering $150 million 12-year bond for infrastructure proj- risk, eliminating perceived risk, and reducing ects, with a guarantee provided by the World transaction costs. The following tools are cov- Bank and the International Finance Corporation’s ered in this section: subsidies provided via mu- (IFC) Municipal Fund and the South African nicipal bonds and concessional loans, brokering Development Bank (World Bank 2004b). The 40 Financing the Gap   29 percent partial credit guarantee helped increase is difficult to measure. Central governments can the municipal bond rating from A– to AA–. The also take on the foreign exchange risk by borrow- bond matures in 2016 and will pay 11.9 percent ing in foreign currency and lending in local cur- interest to creditors. rency, which provides another implicit subsidy to In 2000, a new law in Mexico enabled sub- the local government. national governments to borrow in capital mar- Kingdom et al. (2012) quantifies one such kets but only in peso denominated debt. The subsidy in Vietnam. Compared to a commer- utility in Tlalnepantla de Baz, a municipality near cial loan of similar size, a concessional loan at 4 Mexico City, needed financing for a new waste- percent interest with a 20-year repayment peri- water treatment plant. The utility had historically od and a 5-year grace period provides an implic- covered around 54 percent of its costs with reve- it subsidy that is 72 percent of the total value of nues, but was reliant on the federal government the loan. When the subsidy value of a loan is not for the remainder. The new law allowed the mu- quantified and discussed, stakeholders do not nicipality to issue a $9.6 million municipal bond, get a clear picture of their contributions to the which it did using a partial credit guarantee from sector and, therefore, what they should expect in the IFC and a letter of credit from Dexia Bank. The return (whether it is in units of improved service National Water Commission also provided incen- delivery, greater crop yields, poverty reduction, tives for improvement, which helped strengthen pollution control, etc.). the utility’s balance sheet (USAID 2005). Kingdom et al. argue that concessional loans These examples demonstrate the value of are not perfectly efficient in meeting central gov- municipal bonds in reducing costs and pooling ernment policy targets. In the case of Vietnam, the risk to introduce private investment in local wa- policy was to expand water infrastructure in small- ter service delivery. The instrument, however, re- er cities, but with the central government tak- quires long-term planning through coordination ing the reins in on-lending to local governments, of all sector stakeholders, and guarantees on re- smaller cities received about 12 percent larger payment, which can be difficult to acquire espe- subsidies, and fewer of the subsidies went to those cially in developing countries where few utilities cities that were in the government’s initial plan. are creditworthy. Thus, during the transfer from central to local government, the instrument can lose part of its in- Concessional Loans tended impact. This inefficiency, coupled with the Concessional loans comprise a large portion of high cost of providing the implicit subsidy, make overseas development assistance to the sector. concessional loans expensive and difficult to mea- Donors offer these loans to developing countries sure against results. Kingdom et al. argue that to meet explicit policy objectives and provide where government capacity is adequate, an initial an implicit subsidy. Rather than saving mon- capital subsidy can be used with a parallel com- ey through a tax exemption on interest as with mercial loan instead of a concessional loan. These a municipal bond, concessional loans are pro- subsidies can also be output-based to provide in- vided to central governments at a discounted centives for achieving more long-term outcomes, interest rate, higher grace periods, and longer and the financing arrangement can work to gradu- repayment terms than those available on the pri- ally attract more and more private investment. vate market. The funds are then on-lent to local Service providers with low capacity to man- governments. This preferential treatment trans- age project implementation and loan repay- Section 3 lates into an implicit subsidy, the value of which ment should continue to use concessional loans. 30 Investing in Water Infrastructure: Capital, Operations and Maintenance However, the implicit subsidy on these loans Pooled and Revolving Funds should be quantified and made explicit so that Developed and developing countries alike create the real value is known and the resulting benefits institutions that manage sector investment in the can be measured against the real costs. long run. These pooled or revolving funds help share risk and borrowing needs across groups of Guarantees service providers, such as at the state or even at As in the South Africa case presented above, IFIs the country level. Unlike government transfers, provided a guarantee to the City of Johannesburg these institutions provide resources without any for repayment of a municipal bond, making a annual limitations. When service providers re- contribution to the credit rating of the bond and, pay their debt to the fund, the money becomes therefore, the city’s ability to access private fi- available for investment in other providers. The nance. The World Bank Group’s Multilateral stability and tenure of these funds help attract Investment Guarantee Agency (MIGA) provides investment and make efficient use of a variety of similar guarantees on a project basis in the form funding sources. In the Philippines and India, the of political risk insurance, with water and waste- funds have also been instrumental in building lo- water projects limited to a few middle-income cal capacity to design and implement projects. countries, in addition to a handful of irrigation The US Clean Water State Revolving Fund and hydropower projects. and Drinking Water State Revolving Fund provide USAID’s Development Credit Authority capital grants to states, which are then on-lent to (DCA) also provides partial credit guarantees for local governments for water and wastewater in- water finance. The Philippines Water Revolving frastructure. These state revolving funds mostly Fund (PWRF) (see Box 7) provides a good exam- fund rural community infrastructure rather than ple of coupling a guarantee with a concessional large systems in cities. These funds can also loan to offer below market rates to creditworthy serve to back state-issued bonds to raise more utilities, and giving private creditors access to money for similar works (Anderson 2010). new markets. The PWRF received support from In India, USAID provided DCA munici- USAID through a $37.5 million development pal bond guarantees to the Tamil Nadu Urban credit that re-guarantees loans from the Local Development Fund in 2002. The guarantee cov- Government Unit Guarantee Corporation. ers up to 50 percent ($6.4 million) of bonds is- Governments can also provide loan and sued by the Water and Sanitation Pooled Fund credit guarantees to local water authorities. In for upgrading service in poor areas. The devel- this way, central governments boost the capac- opment fund is now an autonomous financial ity of local providers to borrow in private capi- intermediary managing 500 projects in 90 mu- tal markets by reducing part of the risk (political, nicipalities, and its partial government owner- regulatory, foreign exchange) associated with ship is expected to be phased out to yield full sector investments. However, as the impact private management (USAID 2005). of the financial crisis on central governments spreads, sovereign guarantees will be more dif- Private Sector Participation ficult to obtain. In the case of a currency deval- Over the last 20 years, many developing coun- uation, where the debt repayment in foreign tries have turned to the private sector to bridge currency equivalent becomes much higher than gaps in financing, expertise, and management as what was initially borrowed, it can become more a way to improve the performance of public util- difficult to repay debt. ities. These objectives can be achieved under Financing the Gap   31 Box 7:  The Philippines Water Revolving Fund: The Reform Cycle in Action Much of the water supply in the Philippines is delivered by more than 6,000 small utilities, each with their own finan- cial and operational challenges. The country is currently undergoing major water sector reform to include operational, financial, and regulatory changes toward meeting the MDGs for water supply and sanitation. The cornerstone of the process is the Philippine Water Revolving Fund (PWRF), created in response to a 2004 executive order whereby water utilities were mandated to move from government to market-based financing. The PWRF works from both sides of the challenge, helping potential investors understand water utility business models and assess risk, and helping local governments gain access to private capital. The PWRF acts as a financial in- termediary through which private creditors lend 25–50 percent of a loan, and public and ODA contributions make up the remainder. The creditor applies for a guarantee from a private domestic corporation. The funds are on-lent to local utilities through the Development Bank of the Philippines, usually with a 20-year repayment period. In addition to financial brokering, the country’s water reform encompasses a host of institutional, regulatory, and technical strengthening measures. For example, utility managers receive training in investment planning and market forecasting to produce viable business plans that can be proposed to private lenders. The reform has also included assistance to develop bankable projects for water supply and septage management projects, starting with feasibili- ty studies. Source: Paul (2011). various contractual schemes. Governments and In Burkina Faso, private sector management donors can work together to ensure that the contracts have helped the autonomous water comparative advantages of the private and pub- board, ONEA, gain efficiencies in water cover- lic sector are leveraged, but this requires a strong age, water loss reduction, collections, metering governance structure and proper incentives for and cost recovery. Moreover, experience with both sides. Private entities can be brought in at private contractors can help build the capaci- any stage of the reform cycle depending on the ty of water institutions to develop and manage capacity of local institutions and the investment similar performance-based contracts with pub- climate. Most private contributions are classi- lic entities. Since 1993, ONEA has itself entered fied under management, lease or concession into three-year performance contracts (contract contracts. plans) with the government. Technical, financial, Management contracts and lease contracts and commercial targets are set by means of 34 can incorporate incentives for performance indicators that are monitored on a regular basis improvement and take advantage of interna- by a committee whose members include staff, tional best practices while leaving asset man- government officials, and consumers (Agrawal agement to the public service provider. Lease 2009). These types of contracts can bolster the contracts offer even greater incentives for per- financial solvency of utilities, helping to prepare formance improvements because the private them to become creditworthy entities that can company has the authority to set tariffs and col- attract private finance on their own. lect bills (USAID 2010). Management contracts Concession contracts, on the other hand, can be used to outsource any portion of a utili- require more commercial risk because the pri- ty’s operations, from meter reading to billing and vate party is responsible for making capital in- Section 3 collections. vestments. This can result in non-fulfillment of 32 Investing in Water Infrastructure: Capital, Operations and Maintenance contract terms. Evidence has shown that con- this improves collection ratios and pro- cessions in the water sector have been less vides less resistance to raising tar- successful than other forms of private sector iffs. Experiences from Cote d’Ivoire and participation. Between 1991 and 2010, half of Gabon show that operating efficiently en- all cancelled contracts in the PPI database were abled investments to be funded for more concession contracts (Perard 2012). Requiring than a decade through cash flows, with- private contractors to post performance bonds out needing to incur new debt. as a guarantee for specific capital investments �� Successful water PPPs have to be im- can help mitigate government risk. plemented within a well-conceived, Marin (2009) analyzed more than 65 urban broader sector reform. Successful expe- water projects with public-private partnerships riences in countries such as Colombia, (PPPs) in developing countries over a 15-year Cote d’Ivoire, and Morocco show that in- period. Results suggest that though some proj- troducing PPPs was part of a wider re- ects performed better than others, the overall form to establish a sector framework that performance of water PPPs has been satisfacto- supported financial viability and account- ry. The urban population serviced by private op- ability for performance. These countries erators in the developing countries rose steadily, also had clear policies in place to move to from 94 million in 2000 to more than 160 million cost recovery tariffs in a sustainable and by the end of 2007 (Marin 2009). PPP projects socially acceptable manner. have provided access to piped water to more �� Establishing a good partnership that than 24 million people in developing countries achieves tangible results takes time. It since 1990. Some of the major findings of the took a decade to achieve good results in study include: Senegal. The outcome of a PPP depends heavily on solid collaboration. Government �� The largest contribution of private opera- officials need to move away from old hab- tors was through improved service quality its of interfering in the operations of water and operational efficiency. Improvements utilities, toward an arm’s length relation- achieved through operational efficiency ship based on contractual rules. and quality service depends on the allo- �� Traditional classification of PPP projects cation of responsibilities and risks, which as management contracts, lease-affer- is based on multiple factors such as the mages, and concessions have become incentive structure and the nature of the obsolete. The most sustainable projects arrangement. observed in the study did not fit into any �� Efficient private operators have a posi- of these traditional categories. tive, although mostly indirect, financial contribution. They do this by improving The study is the most comprehensive analy- the creditworthiness of the utility and sis to date in the sector and its recommendations allowing it to secure investment fund- are instrumental in ensuring the proper design of ing for investment more easily and at the next generation of PPP arrangements, partic- better terms. A better service increas- ularly given the fact that local private operators es customers’ willingness to pay and are entering the market more and more. Section 4 . Tools for the Way Forward T his section suggests the use of two tools that can help sector stakeholders enter into the reform cycle. The first is Public Expenditure Reviews (PERs), a sector-level analysis that has proven instrumental in showing governments where their money ends up and the bottlenecks that need to be removed to im- prove efficiency and effectiveness in spending. Evidence from recent PERs in the water sector has shown that annual budgets are rarely part of long-term sector strategies and, thus, do not reflect policy objectives. This creates a “disconnect between expenditure and outputs� (Van Ginneken et al. 2011). The second tool, the incorporation of results-based financing in water investments, seeks to address this challenge. This instrument can be used by donors to provide incentives for achieving project outcomes, or by governments to ensure long-run sustainability of outputs in contracting with public or private operators. Public Expenditure Reviews (PERs) Public Expenditure Reviews (PERs) help a country compare how the flow and use of public funds to the water sector stacks up against budgets. A PER is concerned with public-based (not always gov- ernment) revenues and expenditures as expressions of public policy and public involvement in the economy (World Bank 2009). It entails a careful examination and analysis at the country level of the fundamental drivers of public finance. The recommendations provide guidance to governments on critical reform processes that can be taken to ensure efficiency, efficacy, and transparency in the use of public monies flowing to the water sector. Pradhan (1996) provides a summary of the main elements of a PER, which include discussion and analysis of aggregate public spending in the sector and its allocation; whether public expenditure com- plements or substitutes for private activities; the impacts that public programs have had on the poor; allocations for capital and recurrent expenditure; and budgetary institutions and processes, and their role in promoting fiscal discipline, allocative and technical efficiency, and equity. 34 Investing in Water Infrastructure: Capital, Operations and Maintenance Since 2003, the World Bank has funded 42 First, sector and investment planning in PERs in which the water sector featured in some water require collaboration across line minis- capacity. A quick assessment of some of the wa- tries and between central and local agencies. ter supply and sanitation PERs suggests that Climate change, along with food and energy the efficiency and effectiveness of how govern- security concerns, call for even more integrat- ments allocate, disburse, and use resources in ed planning in the future than has been prac- the sector can be improved. A number of coun- ticed in the past. Cost-benefit analysis can be a tries that have undertaken these exercises have useful tool in prioritizing needed infrastructure, adopted comprehensive budget legislation, re- and when done at the river basin level can pro- duced waste in public expenditure, given great- vide fundamental analysis for water allocation er budget autonomy to local governments, and across sectors. Water variability should be a key attempted to open budgets to public scrutiny concern for long-term planning in both devel- (Deolalikar 2008). oped and developing countries. Sensitivity and Benin is a good case in point. In 2001, Benin risk analysis can help to determine how robust moved to decentralize service delivery. Analytical investments are to changes in circumstances. work, including a water PER, helped uncov- Sector planning should be combined with multi- er bottlenecks in the budget execution chain, year budgeting to ensure that short-, medium-, forming the basis for improvements in public fi- and long-term investments can be implement- nancial management, especially geared to ru- ed properly. ral water supply. Unlike other African countries, Second, governments must improve dis- Benin now produces annual budgets at the sec- bursement functions, which are often a major tor level, which are then incorporated into the source of the inefficiencies that cause higher national budget. Budget execution software en- procurement costs. For example, late release ables tracking of donor financed and public ex- of funds results in low budget execution, which penditures. Line ministries have the authority to may have implications for future access to spend public money and are accountable to the funds. Many developing countries have ineffi- Ministry of Finance through performance-based cient mechanisms to transfer resources from contracts at the subprogram level. The increased central, to regional, and then to local authori- transparency and better budget execution rate ties. Similarly, it is along this chain that a portion have crowded in more donor commitments to of the face value of subsidies is lost. Yearly bud- Benin, which is on track to meet its MDG for rural get cycles often mean that capital works must be water supply (Van Ginneken et al. 2011). contracted and completed within the cycle. Lack of capacity in procurement curtails and delays investments in the sector. Lessons Learned from PERs Third, when allocating funds, governments As some PERs have shown, sector-specific is- should ensure that their policies align with incen- sues at the country level play a major role in ex- tives to improve performance, encourage effi- plaining sector performance. Efficiency can be ciency, and reduce costs. Inflexible procurement enhanced in three ways: (i) improve sector and regulations that narrow the playing field for new investment planning; (ii) improve the capacity to technologies and designs can encourage higher procure, disburse, audit, and monitor sector re- than necessary capital and operating costs that sources; and (iii) maintain a sharper focus on in- impede sustainable service delivery in the long centives in the allocation of funds. run. Public funds should foster the use of lower Financing the Gap   35 cost technologies where possible. Likewise, sub- Output-Based Aid (GPOBA), a donor trust-fund sidies currently channeled to non-poor consum- managed by the World Bank, provides output- ers could be used in much more efficient ways based aid (OBA) to service providers in exchange to meet poverty, health, and economic growth for connecting poor customers to water supply objectives. Doing so would eliminate perverse or sanitation networks. OBA subsidies can either incentives to over-consume. Finally, through buy down the capital cost or cover the difference contracts with public and private services pro- between an affordable user fee and a cost recov- viders and donors, governments can focus on ery user fee. results rather than inputs. GPOBA has approved close to $4 billion in grants. Of these, $137 million are for WSS. There are currently 22 projects with World Bank par- Results-Based Financing (RBF) ticipation that have approximately $140 million allocated to subsidies: fifteen are water sup- Results-based financing (RBF) encompasses a ply schemes, three are sanitation schemes, and range of mechanisms designed to enhance the four provide both water and sanitation (Kumar delivery of infrastructure and social services and Mugabi 2010). Many of these projects are al- through the use of performance-based incen- ready showing promising results. In less than a tives, rewards, and subsidies. A funding enti- year, 6,700 connections were made in Cameroon ty (typically a government or sub-governmental (project target is 40,000); and in India, 77,000 agency) provides a financial incentive, on the connections were made in rural communities condition that the recipient undertakes a set of in Andhra Pradesh. There are, however, crit- pre-determined actions or achieves particular icisms of OBA, including high costs and low le- outputs. Resources are disbursed not against in- verage of commercial funds. Kumar and Mugabi dividual expenditures or contracts on the input (2010) argue that countries with sound regulato- side (as is traditionally done), but against dem- ry frameworks, good capacity for implementing onstrated and independently verified results that programs, and experience with private sector are largely within the control of the recipient. provision have more success than others. RBF mechanisms can be structured in sev- The use of RBF mechanisms in develop- eral ways depending on the objectives and goals ment lending is expected to increase. The World set by the government. There are several types Bank’s new Program for Results (PforR), for ex- of RBF mechanisms, including carbon finance, ample, is an instrument introduced in 2012 that conditional cash transfers, output-based dis- ties financing to achievement of results. PforR in- bursements, and advance market commit- vestments will support government and govern- ments. The application of such standard forms ment-led programs and have a large institutional of RBFs in the water sector is quite limited, but capacity building component to strengthen in- there are a few examples of performance-based country governance and transparency. This contracts between public entities that are based lending will use disbursement-linked indicators on the same characteristics. The cases of NWSC whereby payment is only made after indica- in Uganda (see Box 6) and ONEA in Burkina Faso, tor targets have been met and verified. Funding as discussed earlier, provide two examples. through PforR will be limited for the first few The World Bank has piloted the use of one years but could become an important source of RBF instrument in water using subsidies to cov- finance for the next generation of development Section 4 er access by the poor. The Global Partnership on finance. Section 5 . Conclusions and Recommendations A dditional and improved water supply, sanitation, and irrigation infrastruc- ture will be needed for countries to achieve their development goals, from human health to food security, to energy security and climate resilience. While data on current water infrastructure stocks and sector financing is sparse, estimates of needed investment for developing countries are around $100 million per year. To pay for this infrastructure, developing countries will face serious ob- stacles, including low levels of government support and an inability to attract pri- vate finance. The recent financial crisis has made financing the gap in global water infrastructure even more difficult. All water sector stakeholders can contribute to reforms that help close this gap. Service provid- ers can improve performance and maintain accountability to customers. Governments and donors can stretch their dollars and collaborate to reduce risks for potential private sector investors. The wa- ter reform strategy in the Philippines provides an example of how each stakeholder can make its con- tribution to the reform process, helping to bridge the space between what the public needs and what private investors can provide. Service Providers All stakeholders should focus on helping service providers achieve financial sustainability by improv- ing cost recovery and public expenditure. Cost recovery that enables private investment is the optimal strategy for service providers to pay for current and future water infrastructure. However, it is political- ly difficult to implement even in developed countries, like the United States. Recovering the full cost of delivering water services is complex. Reconciling the economic and financial objectives of cost recov- ery is hard enough, and this is exacerbated by considerations of environmental sustainability and social affordability. The mechanisms and instruments designed to manage these different objectives deter- mine the role that tariffs and user charges can play, which will differ from water service to water service. Service providers can partially close the gap by lowering costs and making efficiency gains. Such efficiencies, even in the absence of full cost recovery, will improve the ability of utilities to adapt to 38 Investing in Water Infrastructure: Capital, Operations and Maintenance future risk, and will make them less dependent Public Expenditure on external funding. Efficiencies will also trans- late into improved service, which can start the When service providers are unable to recov- positive cycle of increasing revenues (for farms) er their costs, a mix of public and private sourc- or productivity (for people), prompting a higher es is needed to fill the financing gap. In developed ability and willingness to pay for services. countries, like the United States, large-scale wa- Tariffs that reflect the cost of inflation can ter systems were built with capital grants and assist in maintaining the trend toward lower per tax-exempt loans from the federal and state gov- capita consumption, while those that account ernments. Today, after decades of public support, for environmental externalities can go one step many urban systems have been able to recov- further by addressing water scarcity and sup- er costs from users to sustain service delivery. porting a green economy. Each and every case Developing countries are experiencing a similar faces different challenges requiring a different process, relying on the government and donors solution. for about 75 percent of total investments in water. Colombia and Uganda approached cost re- For these countries, however, public funding can covery goals from different sides of the equa- be very volatile and local capacity is inadequate tion. While tariffs were increased in Colombia to when it comes to using money efficiently. cut water demand and reduce the need for new Because the majority of funds in the sec- investments, NWSC in Uganda made efficien- tor are public monies, more attention should cy improvements to lower its costs so that tar- be paid to the efficiency and efficacy of pub- iffs could remain affordable for consumers. With lic transfers and subsides. Service providers revenues the most predictable source of finance need better support from government institu- for the sector, providers must continue to focus tions through improved subsidy targeting, more on the portion of funding that they can control, strategic planning, better budget execution, and and make all necessary efficiency improve- guarantees and risk sharing that can help them ments possible. Private management contracts access private funds. Any allocation of costs can play a key role in demonstrating efficiency through stakeholders must take into consider- improvements in the short run. In all cases, un- ation social equity and affordability. Subsidies derstanding the water market and making de- play a critical social function in the distribu- cisions and plans transparent and gradual can tion of equity and should be designed for pre- make successful reform possible. dictability, transparency, and to be phased out Decisions about how to allocate the uncov- over time. Implicit subsidies, including those ered portion of costs depend on political pref- that result from concessional loans and interest- erences, but they also depend on the structure free bonds, should also be made explicit to im- of the local water market. It is imperative that prove targeting. Public Expenditure Reviews are an explicit agreement is reached on who pays a promising tool for countries to identify weak- for what. Without such an arrangement, the real nesses in planning, budgeting, and implementa- costs of water services may be deferred into the tion of public funds. future, seriously hampering short-term and me- Donor contributions also make up a sig- dium-term sustainability. nificant share of total public expenditures, Financing the Gap   39 especially in low-income countries. Technical Trade-offs will need to be made because fi- assistance, guarantees, and concessional loans nancial sustainability is likely to be only one of and grants are vital for maintaining current ac- several objectives (such as service coverage cess levels in some countries. While water sector levels and environmental objectives) that form commitments have increased in recent years, part of a government’s agenda to improve the overall donor funding has dropped. In the future, performance of the water sector. Making these tying development aid more directly to coun- trade-offs more explicit will improve account- try program results (through output-based aid ability and transparency. They may also pro- and other results based financing mechanisms) vide incentives for much-needed reform (such could help countries achieve their long-term pol- as strategies for a green economy and im- icy goals while using fewer resources. proved water and energy conservation). Green growth provides an opportunity to change infra- structure modalities, which can have significant Private Contributions impacts on the long-run cost of operations, help countries mitigate the impacts of climate Private finance has steered away from many low- change, and adapt to its consequences. In income countries toward large urban centers in some cases, climate change could considerably middle-income countries. It is also coming from add to the investment needs of some service more domestic, and fewer international, sourc- providers, and sound forecasting and planning es, and the size of the average project is small- is needed to ensure that funds are used in the er. To attract more private finance, stakeholders most cost-effective manner. should work to eliminate information asym- All sector stakeholders have a role to play metries and the perceived risks that they gen- in helping to bridge the gap between supply and erate. However, it is in the interest of both the demand for capital investments and system op- private sector and national governments to en- eration and maintenance. There is room for im- sure that the appropriate governance structures provement at every turn in the financing chain. and incentives are put in place before bringing The reform cycle presented in this paper reiter- in the private sector. Results-based financing is ates that water agencies can start at their own a mechanism for improving efficiencies in a re- pace and capacity. Reform should be undertak- source-constrained, inefficient sector, but re- en as pieces of a puzzle, with all stakeholders quires the institutional capacity to define and making improvements toward a common goal: monitor outcomes. sustainable water service delivery for all. Section 5 References 2030 Water Resources Group. 2009. Charting Beato, P. and A. Vives. 2008. A Primer for Water Our Water Future: Economic Frameworks to Economics and Financing for Developing Inform Decision-making. Washington DC: Countries. Paper prepared for EXPO 2008 in World Bank. Zaragoza. Washington DC: World Bank. Agrawal, P. C. 2009. Enhancing Water Services Commission of the European Communities. through Performance Agreements. World 2000. Pricing policies for enhancing Bank Water and Sanitation Program (WSP). the sustainability of water resources. A New Delhi: WSP. Communication from the Commission to Alavian, V. et al. 2009. Water and Climate Change: the Council, the European Parliament and Understanding the Risks and Making Climate- the Economic and Social Committee. COM Smart Investment Decisions. Washington (2000) 477, Brussels: EC. DC: World Bank. Danilenko, A., E. Dickson, and M. Jacobsen. 2010. Anderson, R. F. 2010. “Trends in Local Government Climate Change and Urban Water Utilities: Expenditures on Public Water and Wastewater Challenges and Opportunities. Water Note. Services in Infrastructure: Past, Present Washington DC: World Bank. and Future.� U.S. Conference of Mayors. Debomy, S., D. Lauria, and O. Hopkins. 2005. Pro- Washington, DC: Mayors Water Council. Poor Subsidies for Water Connections in West Ashley, R. and A. Cashman. 2006. “The im- Africa. Water Supply & Sanitation Working pacts of change on the long-term future Notes, no 4. Washington DC: World Bank. demand for water sector infrastructure.� Deolalikar, A. B. 2008. Lessons from the World In Infrastructure to 2030: Telecom, Land Bank’s Public Expenditure Reviews, 2000– Transport, Water and Electricity. Paris: OECD 2007, for Improving the Effectiveness Publishing. of Public Spending. Transparency and Banerjee, S. and E. Morella. 2011. Africa’s Water Accountability Project. Washington, DC: The and Sanitation Infrastructure—Access, Brookings Institute. Affordability, and Alternatives. Directions Easter, K. W and Y. Liu. 2005. Cost Recovery in Development Series, Infrastructure. and Water Pricing for Irrigation and Drainage Washington DC: World Bank. Projects. Agriculture and Rural Development Bates, B. C., Z.W. Kundzewicz, S. Wu, and J. Discussion Paper 26. Washington DC: World Palutikof (eds.). 2008. Climate Change and Bank. Water. IPCC Technical Paper VI. Geneva: Estache, A. Forthcoming. Towards a GG- IPCC. compatible Infrastructure Service Delivery. 42 Investing in Water Infrastructure: Capital, Operations and Maintenance Falkenmark, M. and D. Molden. 2008. Wake Kingdom, W. et al. 2006. The Challenge of up to realities of river basin closure. Water Reducing Non-Revenue Water (NRW) in Resources Development, Vol. 24, No. 2, pp. Developing Countries. How the private sector 201–215. can help: A look at performance based con- Fay, M. et al. 2005. “Achieving child-health-relat- tracting. Washington DC: World Bank. ed Millennium Development Goals: The role Kingdom, W., A. Baeumler, and A. Guzman. 2012. of infrastructure.� World Development, Vol. Capital Subsidies Implicit in Concessional 33, pp. 1267–1284. Finance: How to make them More Transparent Fay, M. et al. 2010. Infrastructure and Sustainable and Better Targeted. Washington DC: World Development in Post Crisis Growth and Bank. Development. Washington DC: World Bank. Komives, K. et al. 2007. “Utility Subsidies as Foster, V. and C. Briceño-Garmendia, (eds). Social Transfers: An Empirical Evaluation 2010. Africa’s Infrastructure: A Time for of Targeting Performances.� Development Transformation. Washington DC: World Policy Review, Vol. 25, pp. 659–679. Bank. Kumar, G. and J. Mugabi. 2010. “Output-Based François, D., A. Correljé, and J. Groenewegen. Aid in Water and Sanitation: The Experience 2010. “Cost Recovery in the Water Supply So Far.� OBA Approaches, Note Number 36. and Sanitation sector: A case of competing Washington DC: World Bank. policy objectives?� Utilities Policy, Vol. 18, Lee, T., J. L. Oliver, P. F. Teniere-Buchot, L. Issue 3, pp. 135–141. Travers, and F. Valiron. 2001. “Economic Garrido, A. and J. Calatrava. 2010. “Agricultural and Financial Aspects.� In Frontiers in Urban Water Pricing: EU and Mexico.� Background Water Management – Deadlock or Hope, ed. report for OECD study, Sustainable C. Maksimovic and J.A. Tejada-Guibert. IWA Management of Water Resources in Publishing. Agriculture. Paris: OECD Publishing. Leurig, S. 2010. The Ripple Effect: Water Risk in Hanjra, M. A and M. E. Qureshi. 2010. “Global the Municipal Bond Market. Boston, Mass. Water Crisis and Future Food Security in an and New York: Ceres and Water Asset era of Climate Change.� Food Policy, Vol.35, Management. pp. 365–377 Locussol, A. and M. Van Ginneken. 2010. IBNET (International Benchmarking Network for “Template for Assessing the Governance of Water and Sanitation Utilities). 2012. http:// Public Water Supply and Sanitation Service www.ib-net.org Providers.� Water Working Notes, No. 23. IEG (Independent Evaluation Group). 2009a. Washington DC: World Bank. Water and Development: An Evaluation Marin, P. 2009. Public-Private Partnerships for of World Bank Support 1997–2007, Vol.1. Urban Water Utilities: A review of experienc- Washington DC: World Bank. es in developing countries. Washington, DC: IEG. 2009b. The World Bank Group’s Response World Bank. to the Global Crisis: Update on an Ongoing McPhail, A., A. Locussol, and C. Perry. 2012. IEG Evaluation. Washington, DC: World Bank. “Achieving Financial Sustainability and Jimenez A., and A. Perez-Foguet. 2009. Recovering Costs in Bank Financed Water International Investments in the Water Supply and Sanitation and Irrigation Sector. International Journal of Water Projects.� Water Paper. Washington DC: Resources Development, Vol. 25, pp. 1–14. World Bank. Financing the Gap   43 McKinsey and Company. 2009. “Averting the Private Participation in Infrastructure Database next energy crisis: The demand challenge.� (PPI). 2012. http://ppi.worldbank.org/. McKinsey Global Institute. Public-Private Infrastructure Advisory McKinsey and Company. 2011. “Resource Facility (PPIAF) and the World Bank. Revolution: Meeting the World’s Energy, Rosegrant, M. W., X. Cai, X. and S. A. Cline. 2002. Materials, Food, and Water Needs.� World Water and Food to 2025: Dealing with McKinsey Global Institute. Scarcity. Washington DC: International Food Militaru, A. 2012. “Private investment in water Policy Research Institute (IFPRI). and sewerage recovers in 2011, mostly due UNICEF and World Health Organization. 2012. to activity in China.� PPI Data Update Note Progress on Drinking Water and Sanitation: 76, September 2012. Washington DC: World 2012 Update. New York: UN. Bank. U.S. Agency for International Development OECD (Organisation for Economic Co-operation (USAID). 2005. Case Studies of Bankable and Development). 2011a. Towards Green Water and Sewerage Utilities. USAID. Growth: A Summary for Policy Makers. Paris: Van den Berg, C. and Danilenko, A. 2011. The IBNET OECD Publishing. Water Supply and Sanitation Performance 2011b. Meeting the Challenge of Blue Book. Washington DC: World Bank. Financing Water and Sanitation – Tools and Van Ginneken, M., Netterstrom, U., and Bennett, Approaches. Paris: OECD Publishing. A. 2011. More, Better or Different Spending? 2012a. Financing Water and Sanitation in Washington DC: World Bank. Developing Countries: The Contribution of WaterAid. 2005. “Problems for women.� http:// External Aid. Paris: OECD Publishing. www.wateraid.org/uk/what_we_do/the_ 2012b. Infrastructure to 2030: Telecom, need/206.asp Land, Transport, Water and Electricity. 2006. Winpenny, J. 2003. Financing water for all. Paris: OECD Publishing. Report of the World Panel on Financing Paul, J. N. 2011. Making Water Reform Happen: Water Infrastructure. World Water Council The Experience of the Philippine Water and Global Water Partnership. http://tinyurl. Revolving Fund. A background paper for com/FinancingWaterForAll the OECD Global Forum on Environment: Winpenny, J et al. 2009. The Global Financial Making Reform Happen. OECD Conference and Economic Crisis and the Water Sector. Centre, Paris. 25–26 October, 2011. Report for the Stockholm International Perard, E. 2012. Private sector participation in Water Institute. Stockholm: SIWI. water infrastructure: Review of the last 20 World Bank. 2004a. Towards a Water- years and the way forward. A deliverable Secure Kenya: Water Resources Sector from the PPI Database to IFC. Memorandum. Washington DC: World Bank. Pradhan, S. 1996. Evaluating Public Spending: A World Bank. 2004b. Municipal Fund: Global Framework for Public Expenditure Reviews. Finance for Local Needs. Powerpoint. World Bank Discussion Paper No. 323. October, 2004. Washington DC: World Bank. 2008. “Kyrgyz Republic: On-Farm Pruss-Ustun, A. et al. 2002. “Unsafe Water, Irrigation Project.� Implementation Sanitation and Hygiene.� Comparative Completion and Results Report. Europe and Quantification of Health Risks, Vol 6, Chapter Central Asia Region, World Bank. Section 5 16. World Health Organization. 44 Investing in Water Infrastructure: Capital, Operations and Maintenance 2009. Preparing PERs for Human Report of the Economics of Adaptation to Development: Core Guidance. Washington Climate Change Study. Consultation Draft. DC: World Bank. Washington DC: World Bank. 2010a. Global Monitoring Report 2010: 2012. Inclusive Green Growth: The The MDGs after the Crisis. A Joint Report Pathway to Sustainable Development. of the Staffs of the World Bank and the Washington DC: World Bank. International Monetary Fund. Washington Yepes, Tito. 2008. “Investment Needs for DC: World Bank. Infrastructure in Developing Countries 2010b. “The Cost to Developing Countries 2008–2015.� Latin America and Caribbean of Adapting to Climate Change.� The Global Region, World Bank. Unpublished.