December 2006 Working Paper 46760 Tackling Corruption in the Water and Sanitation Sector in Africa Starting the Dialogue Janelle Plummer Piers Cross 2 Tackling Corruption in the Water and Sanitation Sector in Africa: Starting the Dialogue Contents Abstract 5 1. Introduction 6 The Challenge 6 Corruption, reform and access in the WSS sector in Africa 7 2. Diagnosing Corruption In The Water And Sanitation Sector In Africa 9 2.1 Understanding corruption in water and sanitation services 9 2.2 Promoting a comprehensive view of sector corruption 11 2. Developing a Corruption Interaction Framework 12 2.4 Identifying concentrated areas of corruption 19 3. Tackling Corruption In The WSS Sector In Africa 19 .1 Understanding and addressing incentives ­ a key starting action 19 .2 Key dimensions of good governance in WSS sector anti-corruption activity 20 . Aligning potential anti-corruption mechanisms to corruption in the WSS sector 22 .4 A cautionary tale for the water sector 0 4. Moving Forward 31 5. Concluding Remarks 35 References 37 Tables, Figures and Boxes LIST OF TABLES Table 1. Value Chain Framework ­ WSS Sector Corrupt Interactions 1 Table 2. Key Public Actors involved in the WSS sector 15 Table . Tackling hotspots in public-public corruption 24 Table 4. Tackling hotspots in public-private corruption 26 Table 5. Tackling hotspots in public-consumer/civil society corruption 28 Table 6. Next Steps LIST OF FIGURES Figure 1. Corruption, Water Reform, GNI per capita and Access to Water 8 Figure 2. The Spectrum of Utility and non-utility water providers 11 LIST OF BOXES Box 1. Utility officials extort bribes from small water providers in Nairobi 18 Box 2. Institutional reforms improve utility performance in Nairobi 27 Box . Intensive efforts to curb corruption in village infrastructure development: Lessons from Indonesia 29 4 Tackling Corruption in the Water and Sanitation Sector in Africa: Starting the Dialogue Abstract For the past three decades a substantial number of governments, donors and NGOs have focused efforts on a range of institutional, financial, technical and social interventions aimed at bringing about much-needed improvements in the delivery of water and sanitation services in rural and urban areas of Africa. Yet the attainment of the water and sanitation MDGs is unlikely in the majority of African countries ­ the stability, investment and capacity needed to meet significant and growing demand is lacking. But even if additional finance was to become available, the unacceptable level of leakage of existing resources brings into question current processes and, perhaps, the wisdom of increasing resource flows to the sector. Much of the funding available in ministries, local governments, utilities and village administrations is being used by public office for private gain. Despite the complexity, leakage, and the potential impacts on the poor, there is currently only a limited understanding of the extent and nature of corruption in the water and sanitation sector in Africa, and limited knowledge of the policies and mechanisms that are required to tackle it. To address this concern, and to help the sector `catch up', the purpose of this paper is to promote more comprehensive understanding of sector corruption and potential anti-corruption mechanisms among a broad audience of WSS stakeholders. The paper describes the plural nature of corruption in the WSS sector corruption by setting out, in a structured framework, the network of corrupt practices prevalent in the sector. Emphasizing the need to build a braid-based coalition for change it collects together the many types of WSS corruption into typologies of public to public, public to private, and public to consumer interactions, and considers these interactions at each stage of a WSS value chain. It then describes the range of anti-corruption policies and mechanisms that have been developed to prevent or counter anti-corruption activity in the sector ­ mapping these over the corrupt interactions ­ and thus linking the framework of corrupt practices to the menu of existing solutions. Notwithstanding this attempt to promote a more comprehensive understanding of corruption, the paper emphasizes the need to undertake rigorous diagnostics to identify areas of concentrated corruption, and to focus efforts in the future on improving sector understanding of what anti- corruption strategies are most appropriate. While it cautiously suggests that efforts should be made in the development of greater transparency and accountability mechanisms, supported by continuing more targeted efforts in WSS sector policy and institutional reform, it argues for more data and learning, for context specificity, and for efforts to develop appropriate methodologies and models for sector interventions in the different economic, governance, and WSS contexts of the African region. Janelle Plummer and Piers Cross Water and Sanitation Program ­ Africa 5 1. INTRODUCTION 1.1 The Challenge For the past three decades a substantial number of governments, donors and NGOs have focused efforts on a range of institutional, financial, technical and social interventions aimed at bringing about much-needed improvements in the delivery of water and sanitation services in rural and urban areas in developing countries. Recognizing the impacts of low levels of access for the poor, approaches have become increasingly targeted and service-oriented: responding to demand from users, identifying entry points with clients, reacting to signals in a developing water and sanitation market, and of course steering this course with the ebb and flow of donor funding. In more recent years, governments have embarked upon a process of establishing road maps (action plans) that plot the long paths of sector reform and service improvement needed to meet the MDGs. In a number of well-performing states, it looks like steady progress is being made.1 Yet the attainment of the water and sanitation MDGs is unlikely in the majority of African countries2 ­ the stability, investment and capacity needed to meet significant and growing demand is lacking. But even if additional finance was to become available, an unacceptable level of leakage of existing resources brings into question current processes and, perhaps, the wisdom of increasing resource flows to the sector. Much of the funding available in ministries, local governments, utilities and village administrations is being used by public office for private gain. Be it in decision-making over the allocation of water resources, or bribery and fraud in procurement or construction, corrupt practices are endemic to most water supply and sanitation (WSS) institutions and transactions in Africa. This corruption varies substantially in size and incidence, but it is clear that significant WSS sector finances is being lost to those tasked with the decision-making and delivery of water and sanitation services.4 If the estimated 6.7 billion USD needed to reach the MDGs in sub-Sahararan Africa was mobilized annually, a 0 per cent leakage would represent a loss of over 20 billion USD from the sector over the next decade.5 The struggle with corruption in the water sector is of course fundamentally part of a broader governance problem and is characterized by the dynamics of both sector and national level reform processes. Policy reform promoting decentralization and private sector participation, and new funding paradigms such as SWAps and direct budget support may have provided a more fertile environment for new (sometimes higher) levels of corruption of national or donor funds, but simultaneously, WSS sector reform in many African countries has removed conflicts of interests in sector management, improved transparency and accountability, and created the potential for models and the promise of change. Placing sector-specific anti-corruption initiatives in the broader context of governance and anti-corruption reform is therefore key to understanding the opportunities and limitations of sector anti-corruption reform. Notwithstanding the importance of the governance framework, the sector needs also to focus on diagnostics and solutions specific to WSS service delivery if it is going to improve progress towards the MDGs. The dysfunctionality of the water 1 This paper is a revision of an earlier paper Tackling Corruption in the Water and Sanitation Sector in Africa. We are particularly grateful to Ed Campos for his support, to the Roundtable reviewers of this version, Randi Ryterman, Francesca Recanatini. and Vinay Bhargava, and to Scott Guggenheim, Charles Kenny, Jean Doyen, Patrik Stalgren, Fook Chuen Eng, Alain Locussol, Antonio Estache, Donal O'Leary, Chris Heymans, Steve Burgess, Juliet Pumpuni , Vivek Srivastava and Jonathon Halpern for helpful comments on earlier drafts. The views expressed are entirely those of the authors and should not be attributed to WSP-Africa, EWD or the World Bank, their Executive Board, or management. 2 Joint Monitoring Program, 2002. World Bank definition, 1997. This definition is now under much discussion. TI have adopted a definition `entrusted power for private gain', a recent paper by Kaufmann challenges the conventional definition that distinguishes between legal and illegal forms of corruption (Kaufmann and Vicente, 2005), and the emerging debate in the World Bank calls for more precise definitions (PREM week, May 2006). 4 This estimate is drawn from a variety of unofficial sources and measurements (such as Davis, 200; World Bank, 200) in other regions. A key task in the forthcoming stage of work is to ascertain a clearer estimate for different WSS contexts and countries of Africa. 5 This figure is based on an estimate of 6.7 billion USD for annual expenditure requirements to meet the MDGs in sub-Saharan Africa. Of this 2.6 bn USD is intended for capital investments. Mehta, et al, 2005. Of course not all these leaked funds would necessarily make their way into sector investment, and there is no way to measure what would. 6 Tackling Corruption in the Water and Sanitation Sector in Africa: Starting the Dialogue sector in Africa is distinctive. The major failure of public service provision is exacerbated by widespread financial disorder, few service providers are accountable to their customers, and sector resources are frequently not ring-fenced. Sector providers are also characterized by diversity: formal and informal, large and small, delivering different types of services. In addition, the sector shares high-risk characteristics with a number of other sectors. Being about infrastructure, basic services, and being predominately public, corruption in water supply and sanitation is multi-faceted. Evidence suggests that WSS sector corruption is beset by the opportunity for massive distortion in resource allocation and significant procurement-related corruption (as a construction sector); the daily opportunities for petty corruption (as a service delivery sector); and opaque budgeting and financial management practices of weak institutions (typical of the civil service). Despite this complexity, leakage, and the potential impacts on the poor ­ and despite significant efforts reforming policies and institutions and improving efficiencies by a range of supporting agencies ­ there is currently only a limited understanding of the extent and nature of corruption in the water and sanitation sector in Africa, and limited knowledge of the policies and mechanisms that are required to tackle it. There is virtually no data, and the data that exists is piecemeal, produced for other purposes. There is an urgent need to launch the dialogue and begin the diagnostics and reforms that are needed. With some caution, the purpose of this paper is to synthesize, for a broad audience of WSS stakeholders, the known dimensions of sector corruption and anti-corruption activity. The paper first provides an overview of the African water context in relation to corruption perception, water reform, and access to water supply. Section 2 describes the plural nature of corruption in the WSS sector corruption by setting out, in a structured framework, the network of corrupt practices prevalent in the WSS sector. In so doing, this framework collects together the many types of WSS corruption into a typology of public to public, public to private, and public to consumer interactions, and considers these interactions at each stage of a WSS value chain. Section then describes the range of anti-corruption policies and mechanisms that have been emerged over recent years to prevent or counter anti-corruption activity in the sector ­ mapping these over the corrupt interactions ­ and thus linking the framework of corrupt practices to a menu of existing solutions. The paper concludes with a discussion of key concerns the WSS sector will need to address as it moves forward with the anti-corruption agenda. Notwithstanding this effort to promote a comprehensive understanding of corruption in the sector in Africa, the paper emphasizes the need for sound diagnostics to identify areas of concentrated corruption, and a focus on improving sector understanding of appropriate and viable anti-corruption strategies. While it cautiously suggests that efforts should be made in the development of greater transparency and accountability mechanisms supported by more targeted efforts in WSS sector reform, it argues for more data and learning, for context specificity, and for efforts to develop appropriate methodologies and models for sector interventions in the different economic, governance, and WSS contexts of the African region. 1.2 Corruption, reform and access in the WSS sector in Africa Access to water and sanitation in Africa, especially rural areas drags well behind other regions. Progress is slow. More than 42 per cent or 00 million Africans lack access to an improved water supply, and 64 per cent or 477 million lack access to adequate sanitation (JMP, 2002). The African problem is characterized by low coverage of services, a slow rate of improvement, and a growing diversity. Averages hide a wealth of problems and gaps. If we consider instead some of the lowest figures in Ethiopia only 22 per cent of the population has access to improved drinking water and 6 per cent to adequate sanitation. Many of the poorest states recovering from natural disasters and humanitarian crises are still politically and economically unstable, have few functioning assets, little capacity, and high levels of corruption. Access to WSS services are propped up by relief measures that are highly susceptible to leakage or service delivery is non-existent. Yet, the story of corruption, reform and limited access in the 7 WSS sector in Africa is not limited to these post-crisis countries. At any one time, 0 to 50 per cent of rural water supplies are out of order, and 80 per cent of urban water utilities in Africa are considered financially unviable.6 Corruption in Africa is significant, unabated and country specific, driven by conditions ripe for unaccountable and less than transparent behavior. Of the 4 African countries ranked in the Corruption Perception Index (CPI) produced by Transparency International7 in 2004, only six African countries8 were ranked in the top 50 per cent of the 146 country index, 15 countries were ranked in the next 25 per cent, and 1 countries, many post-conflict states, were ranked with the former Eastern European (and other exceptional countries) in the bottom 25 per cent of the ranking. In the African region, the correlation between corruption (as measured by the CPI) and growth (as measured by GNI per capita) is variable.9 While South Africa and Ethiopia show correlations at either end of the scale, in an (unrepresentative) basket of countries, the country-by-country correlation between corruption and growth is less steady, perhaps supporting the thesis that a range of country-specific factors determine the growth-corruption relationship. Water sector reform seems stronger in countries with lower levels of corruption. In this set of African countries a general correlation is found between perceived corruption levels and the status of water reform. Notwithstanding the range of other influencing factors, countries with less corruption seem to have made better progress in WSS sector reform.10 Figure 1 shows the Transparency International Corruption Perception Index (CPI) together with a Water Reform Ranking (WRR)11 for eleven African countries. While South Africa leads regionally in WSS reform, Benin, Senegal and Uganda also show significant progress towards sector reform, whereas DRC has only recently initiated the reform process and struggles with post conflict levels of corruption. Stronger WSS sector reform and lower corruption correlates with higher levels of access to water supply. From the limited information available (and notwithstanding the limitations of measurement of the indices or the existence of a range of factors influencing access) there is an expected correlation between higher levels of access to water supply and countries that have made progress in WSS sector reform. Analysis of country level characteristics of the WSS sectors in these countries also reveals the time lag for reforms to be translated into better outcomes. Uganda particularly stands out as having moved forward with water reforms, but has not yet seen this reflected in increased levels of access. The correlation between sector reform, lower corruption and higher rates of access, is supported by utility specific studies and cross-cutting global studies. Evidence provided in investment climate surveys (Kenny, 2006) that measure the perception of petty corruption in WSS delivery supports the finding that corruption seems to be strongly correlated with lower levels of WSS coverage.12 In their assessment of the efficiency of African utilities, Estache and Kouassi, (controlling for other variables) found that corruption is significant, estimating that if water utilities were operating in corrupt-free environments, efficiency would be 64 per cent higher (or costs could be reduced by 64 per cent).1 6 Depending on the country and regions within each country, hand pump failure rates can be anywhere from 15-50%, averaging around 0% continent-wide (Sutton, 2004). 7 The TI Corruption Perceptions Index (CPI) ranks countries in terms of the degree to which corruption is perceived to exist among public officials and politicians. It is a composite index, drawing on corruption-re- lated data in expert surveys carried out by a variety of reputable institutions. It reflects the views of business people and analysts from around the world, including experts who are locals in the countries evaluated. The Corruption Perceptions Index provides a snapshot, with less capacity to offer year-to-year trends. Transparency International, 2005. 8 The limitations of these indices are recognized in this broad brush overview, efforts are needed to develop more robust sector indicators. For limitations on the CPI see www.transparency.org. 9 There are many exceptions to the general trend that growth is stronger in less corrupt countries. Indonesia's economic miracle occurred in the context of historically unprecedented corruption and an authoritar- ian state (Timmer, 2006). In Governance and Growth: Causality which way? Kaufmann and Kraay, examine the linkages between growth and governance, and find that there is a strong causal effect better gov- ernance to higher per capita income, confirming the importance of good governance and economic development, but that there is a lack of a virtuous circle whereby higher incomes result in better governance. Kaufmann and Kraay, 200. 10 These findings are not statistically significant, such analysis has not been systematically carried out in the sector ­ a recommendation of this paper is that this analysis be undertaken. 11 The Water Reform Ranking has been drawn from data obtained in 20 Country Status Overviews in Africa conducted to ascertain sector progress towards the MDGs. The criteria for this ranking includes: (i) sector restructuring to create transparency, separation of policy, regulation and implementation; (ii) financial policies and ring-fencing for viability; (iii) openness to leveraging private sector and other local stakeholder involvement; (iv) policy towards servicing the poor; (v) sector coordination and sector financing instruments, water included in PRSP. WSP (forthcoming) Country Status Overviews of Africa's Progress Towards the WSS MDGs. 12 In 16 recent Investment Climate Surveys carried out by the World Bank, companies were asked if they made informal payments to get a water connection. Although statistically insignificant, results in 16 countries (controlling for GDP per capita) suggest a strong correlation. Interestingly, the micro-level survey (with firms, not households) challenges the earlier correlation that corruption in the sector (as measured by bribe payments for a water connection) is correlated with general measures of corruption, but the fragility of the results further highlights the recommendation for improved measurement at the sector and sub-sector level. 1 This exceeds the total gain achieved from privatization. Estache and Kouassi, 2002, quoted in Collier and Hoeffler 2005. 8 Tackling Corruption in the Water and Sanitation Sector in Africa: Starting the Dialogue Figure 1. Corruption, Water Reform, GNI per capita and Access to Water CPI - Corruption WRR - Water Reform GNI per capita (2004) access to improved Perception Index (score) Ranking (score) water supply 90 900 80 70 access 60 600 %/ 50 capita 40 per WRR/ 0 00 $US 20 CPI 10 0 0 Africa Ghana Benin DRC Senegal anzaniaT Uganda Zambia Kenya Ethiopia South Mozambique Source: TI 2005, WSP 2006, JMP 2004, WDI Database 2004 The vast differences in the African continent suggest the need for country specificity and better understanding of regional patterns and trends. Regional typological differences are apparent in terms of economies (be they coastal, landlocked or resource-rich), governance (be they fragile, emerging, or capable) and political systems (authoritarian, established and emerging democracies). This variation creates a multitude of contexts, and suggests a mixed basket of solutions. The variation is also evident at the WSS sector level. Institutional capacity and institutional frameworks vary (national and decentralized, regulatory and provider agencies with and without autonomy), as do service delivery models (public or privately managed utilities, municipal and district water departments, small local providers, and community management). Understanding what can be done in the best (economic, governance) and worst cases, and how this impacts on potential action at the sector level will be useful for lesson learning across the region. 2. DIAGNOSING CORRUPTION IN THE WATER AND SANITATION SECTOR IN AFRICA 2.1 Understanding corruption in water and sanitation services The corruption that occurs in the WSS sector can generally be understood in terms of: bureaucratic or petty corruption in which a vast number of officials abusing public office extract small bribes and favors; grand corruption meaning the misuse of vast amounts of public sector funds by a relative small number of officials; or state capture, seen in the collusion between public 9 and private actors for private benefit (Schacter and Shah, 2000). In WSS, these corrupt practices, big and small, take the form of (i) abuse of resources ­ theft and embezzlement from budgets and revenues, (ii) corruption in procurement which results in overpayment and failure to enforce quality standards, (iii) administrative corruption in payment systems, and (iv) corruption at the point of service delivery. Corruption in the water and sanitation sector varies by country (and regions within a country), by governance, by WSS systems, and a multitude of other factors. In urban areas, the type, size and incidence of corruption in water and sanitation service delivery may be a function of the path of legislative reform or of the leadership that the sector has seen. Or it might be an outcome of decentralization, the role of social structures and civil society, or the nature of the water market in difficult locations. In rural and peri-urban areas, similar factors, plus the structural shift towards community driven development approaches, the highly opaque construction and management processes, the unmonitored, uncompetitive and low capacity contexts in remote areas, and the nature of traditional social structures, all potentially contribute to local level corruption, capture and collusion. Who is involved? WSS sector corruption involves, to some degree, a vast range of stakeholders. The list of actors includes international actors (both donor representatives and private companies and multinationals), national and local construction companies, consultancy firms and suppliers, large and small-scale operators, a range of middlemen, consumers and CSOs as well as national and sub- national politicians, and all grades of civil servants and utility staff. Corrupt activities between these partners occur at a range of institutional levels, with different stakeholders often involved in one or more types of corruption. What are the causes? Like all corruption in developing and transitional economies, corruption in WSS in Africa is founded in historical, political and social realities, the causes of corruption are not sectoral. Corrupt practices take hold and are manifest in different contexts in very different ways and legal frameworks, institutional structures and bureaucratic systems strongly influence how elected, managerial and technical officials behave. Klitgaard's suggested approach to fighting corruption: Corruption = Monopoly + Discretion ­ Accountability, provided in the context of municipal service delivery in Bolivia (Klitgaard, 1998) is very relevant to an understanding of the WSS sector in Africa in that it highlights the aggregate effect of monopoly and discretionary power.14 The WSS sector has long grappled with its monopolistic past and the traits (such as high capital costs and economies of scale) that tend to keep it that way. A strong characteristic of agencies and officials involved in the sector (and outlined in table 2 below) is their enormous discretion in the planning, design, contracting, implementation and monitoring of water and sanitation service delivery (compounded by a lack of clarity of rules and regulations). To this it must be added that demand for accountability for services, although improving in many contexts, is typically a missing element in service-provider (Gray and Kaufmann, 1998) and water-user relationships in Africa. Is the water sector prone to corruption? It is difficult not to follow the lead of other sectors and emphasize the enormity of the problem of WSS sector corruption.15 While we know it is significant however, we do not currently know if the WSS sector is more or less prone to corruption, or whether 14 A number of anti-corruption advocates including Klitgaard and Ackerman identify four key factors that engender opportunities for corruption : monopoly power, wide discretion, weak accountability and lack of transparency. 15 On health, see TI 2006; on power, see World Bank, 2006; on forestry, see FIN 2001; and on the construction industry, see TI 2005. 10 Tackling Corruption in the Water and Sanitation Sector in Africa: Starting the Dialogue such a generalization could be made when we know country contexts, institutions and policies vary so greatly.16 Nevertheless it is possible to posit a number of characteristics which make WSS services17 susceptible and a cause for grave concern for all stakeholders. Many of the fundamental issues, such as low capacity, low wages, dysfunctional institutions, and large-scale procurement are common to public service delivery, but the WSS sector is also part of the construction sector, globally thought to be the most corrupt of all sectors (TI, 2005), and it does aggregate a number of other dimensions which suggest relatively high potential for corruption. These include, inter alia, (i) the large flow of public money, often uncoordinated donor, national and local funds; (ii) the opacity, political interference and discretion in investment decisions; (iii) the monopolistic nature of service delivery, coupled with the failure of sector financing and cost recovery, problematic tariffs and subsidies, and the increasing role of the informal market; (iv) the cost of sector assets;18 (v) the asymmetry of information between user and provider; and (vi) the complexity of sector stakeholders, systems, levels of service, institutional roles and functions. How much is corruption costing the WSS sector? Hypotheses on the scope and incidence of corruption in the WSS sector in Africa are largely untested. Leakage can be roughly estimated through comparative and limited sector studies,19 but to date it has not been measured in the WSS sector in Africa in any systematic way. As a proportion of sector expenditure, the high levels of petty corruption, in the aggregate, constitute a substantial figure across the continent, but we do not yet have regional or country estimates based on empirical studies. The figure of 20-5 per cent provided by Davis in the context of the service delivery in South Asia (Davis, 200), provides a sector not a regional indication, but it should be noted that this estimate is limited to petty corruption and does not account for high level abuse or diversion of resources. This is an urgent area of future work. 2.2 Promoting a comprehensive view of sector corruption A comprehensive approach to sector diagnosis and action first requires recognition that water service delivery is heterogeneous at the sector, city/district and household levels, and involves a number of formal and informal stakeholders from the public and private sectors and civil society. Figure 2 provides a simplified picture of the water market place. While a utility may be producing half of the water consumed, only a proportion of the population are connected formally from source to consumption (illustrated by the top line of arrows to point A). This might represent the means by which a majority of non-poor are served in urban areas. In practice water produced by a utility might be distributed by the utility and bought or stolen during the secondary or tertiary distribution (illustrated by the diversion to point B). Alternatively, water might be diverted immediately after production, by being bought or stolen by other providers willing to distribute it from the point or near the point of production and following a series of lines along a route of non-utility distribution to point C. The fourth line of production-distribution is that shown at the bottom of the diagram reaching point D where neither the production, or distribution is carried out by the utility or local government ­ this excludes the role of state agencies in delivery (and is largely outside the public office definition of corruption). The typical forms of delivery for each of these supply chains are shown in the figure below at the right hand side. They include: the utility itself, community-managed provision, formal intermediate providers, informal providers, and householders. One of the problems being addressed by this paper, and the key issue motivating its structure and content, is that perceptions 16 There are some comparative findings in the WBI country governance and corruption surveys. In Mozambique for instance, water is perceived to be the most corrupt service (above health and education, significantly higher than sanitation but not as corrupt as other municipal services). Austral Consultoria a Projectos, (no date). 17 The need to disaggregate water and sanitation is well recognized. Some contend that sanitation in rural projects will be subject to less corruption than water as it is a public good (Guggenheim, 2006) and there is some indications in the WBI Mozambique survey that public perceptions see water as being significantly more corrupt than sewerage. More research however is needed before any conclusions can be drawn. Future analysis should disaggregate, not only water and sanitation, but the service delivery system (community-based, utility, household etc). 18 The specificity of assets in the water industry is -4 times that in telcom and power, and a higher cost when compared with electricity) Kirkpatrick et al. 2004 19 The limitation of the figures commonly quoted in the sector is a key concern addressed in this paper. These types of (20-40%) figures are often derived only from leakages in utilities, but they are actually similar to the levels of corruption quoted generally and in similar programs (such as rural development programs, roads projects) elsewhere. 11 Figure 2. The Spectrum of Utility and non-utility water providers supply. UTILITY · Household connections, of water form DISTRIBUTION utility standpipes, utility tanker supply is only one utility water UTILITY provision Utility DISTRIBUTION utility water · Domestic reselling NON-UTILITY · Private water carriers UTILITY DISTRIBUTION · Private/CSO managed PRODUCTION utility water standpipes/kiosks networks NON-UTILITY · Water stolen at DISTRIBUTION neighborhood level utility water · NGO / Community NON-UTILITY NON-UTILITY managed water supply PRODUCTION DISTRIBUTION · Private water providers non-utility water utility water · Water bought or stolen at NON-UTILITY CONSUMERS city/district level DISTRIBUTION non-utility water NON-UTILITY · Private rural water supply DISTRIBUTION · Urban residents with In the water supply/delivery process, non-utility water private wells corrupt interactions can occur at any one of these points. Diagnosis is needed at the country level to determine the hotspots. Source: Adapted from Plummer, 200. of corruption among WSS stakeholders tend to be very narrow. Individual stakeholders seem to each hold subjective views on what corruption is, often focused on one type of corruption or another;20 and the formal corruption debate has mostly focused on procurement, construction, and operational inefficiencies. This narrow understanding limits the development of effective reform. To move forward with a pro-poor strategy for action in WSS in Africa, it is necessary to consider the whole network of corrupt practices and expand narrow views of corruption to encompass the diversity of its forms. The need for sector level (rather than project level) analysis and action is key to a coordinated and meaningful sector response.21 To help meet this objective, this section sets out a comprehensive framework to unbundle and differentiate types of sector corruption. This framework can be used to tell us what corrupt practices exist, who is involved, and at what stage of WSS service delivery they occur. It can be also used in each country context to locate the areas of corruption concentration, to identify unknowns, to plot shifts in corruption activity, and identify linkages within the corruption matrix. Ultimately the goal of this sort of information organizing exercise is to attempt to provide a robust framework that is relevant and applicable to the sector, integrates project level and cross-cutting governance diagnostics and is useable as a tool for understanding and promoting change.22 20This observation comes from discussions with a wide range of stakeholders and sector professionals that few in the setor have acknowledged the many and varied types of corruption that exist. Perceptions of corruptions are not homogeneous but they are narrow, limiting the development of effective reform. 21Governance, sector, and project levels of anti-corruption mechanisms were used to describe post-tsunami reconstruction efforts in the WSS sector in Aceh in Plummer, 2005. 22Other diagnostic work has informed the approach. Focusing on areas of activity, the first version of this paper provided a framework of these interactions but structured by the levels of the delivery agency. One 12 Tackling Corruption in the Water and Sanitation Sector in Africa: Starting the Dialogue 2.3 Developing a Corruption Interaction Framework A value chain approach The corruption framework (illustrated in table 1) is structured around interactions and a value chain. This approach is driven by a need to engage with and build broad stakeholder commitment and a coalition for change, and a strong belief that the corruption problem in the WSS sector should be articulated in terms of the actions of all public, private and civil society actors and institutions. It is also pursued knowing that policy actions are more likely to influence the public sector and that more policy options are available to this end.2 Beginning with the definition of corruption as the use of public office for private gain, the approach places the public or entrusted office at the core of the interaction framework and notes that the public officer/ agency will interact with one of three different types of actors: (i) other public actors/agencies, (ii) private actors/companies, or (iii) consumers/civil society.24 The framework highlights these interactions in terms of the functions to be performed in water and sanitation services ­ a cycle of policy-making and regulation, budgeting and planning, financing, program design and management, tendering and procurement, construction, operation and maintenance, and monitoring functions. Public to Public interactions The `public' actor or government is made up of a range of actors from international, national and sub-national departments and agencies in both water and non-water (but influential) functions. Corrupt practices within government, typically involve interactions between public actors (although it may, in rare cases, be achieved by an individual alone). These interactions are `vertical', within the hierarchy of water institutions, and/or `horizontal' involving various line departments and agencies at a similar level of government (Table 2). At the higher levels of government, corruption is opaque and complex, but distortions in the allocation of resources are only achieved by collaboration within water departments and between line departments such as Finance, Planning and Water Affairs or Public Works. Officials are expected to `play the game' and their status and power base is dependent on their willingness to work within the established system. Typical interactions where corrupt practices are concentrated within the public sector are found in policy-making functions. Politicians and officials responsible for water sector policies seek to influence the focus of policy (which is likely to determine investment priorities) to set up future opportunities for rent seeking. In those countries where the regulatory function has been defined, regulators can be bought (defined by the term regulatory capture), by politicians and other stakeholders to determine standards and regulations, or allowing projects to bypass established standards or procedures. Grand corruption occurs among politicians and senior officials in the selection of WSS projects: during planning and budgeting processes, capital-investment projects are favored over lower investment alternatives, sector investments that guarantee higher very useful initiative developed for the World Bank-funded Kecamatan Development Program (KDP) in Indonesia, (which supports village-level community-based infrastructure), is the `corruption map'. This map- ping tool walks through the project cycle, considering the design, implementation and monitoring stages, and setting out the incentives of actors, the forms of corrupt behaviour and the risks associated with each potential leakage. This map is then used for the project to develop specific responses to each medium to high-level risk, and later for facilitators to structure their oversight of the various processes. The mapping process has been critical in establishing transparency and empowering communities to monitor implementation (and ongoing operation and maintenance) at the project level. A similar approach is also taken forward in a number of roads projects that have also used the project cycle as the foundation for assessing potential areas of corruption, and is similar to that originally presented by Klitgaard in the participatory diagnosis approach developed for the municipality in La Paz, Bolivia that focused on corrupt systems and aimed to identify the size, winners, losers, causes and cures of the different types of corruption identified through the participatory process (Klitgaard, 1998). These are however not carried out at the sector level. Another useful conceptual approach is the adaptation of the accountability framework presented in the WDR 2004 which examines the relationships between key stakeholders in the service delivery process, highlighting the importance of the `voice' accountability relationship between citizens and politicians, the `compact' relationship between politicians/managers and service delivery agents, and the `service/client power' relationship between the service provider and the citizen/ consumer. Each side of this triangular framework explores a set of complex interactions enabling corruption and the key mechanism for improving accountability. 2 This issue is taken up in the evaluation method used in Huther and Shah, 2004. 24 Civil society and their representative organizations. While it is possible that private-private interactions are also prevalent in the sector (e.g. bribery or fraud between contractors and sub-contractors as de- scribed in (TI-Colombia, 2005), it is assumed that in Africa where most WSS and WRM transactions involve the public sector, that this type of practice ultimately is embodied in public-private interactions. In those situations where the private sector has been granted concessions, they have `entrusted office'. 1 levels of return, are favored over those that do not. Public resources are diverted to WSS projects where there are greater levels of potential kickbacks, with the greatest incidence, at the lowest possible risk. Experience suggests that these are, more often than not, regressive in their impact. Sourcing water from surface rather than ground water alternatives where they are available, is a typical illustration of decision-making that, while legal, creates opportunities for both grand and petty corruption. The need for the construction of costly water treatment plants and ongoing procurement of chemicals, (and thus opportunity for recurrent bribery, extortion and fraud) such as that seen in Kinshasa and elsewhere, is characteristic of the types of decisions made within the water sector that have structural effects on corruption (WSP-Africa, 2006). The use and abuse of resources is found in the manipulation of budgets and in dubious decision-making in planning processes. This practice is exacerbated by a disconnect between policy objectives, planning and implementation. At the delivery agency and village level, this type of corruption might involve the diversion of the inputs themselves (such as chemicals for water treatment) for resale or other use ­ all resulting in lower quality or quantity of water supply. In some country contexts, fiscal transfers, irrespective of sector, present a series of Table 1. Value Chain Framework ­ WSS Sector Corrupt Interactions Public-Public Public-Private Public-Consumers/Civil Policy-making · Policy capture (competition and · Policy capture monopolies) · Regulatory capture (e.g. waivers to and Regulation regulations and licensing) · Distortions in decision-making · Bribery to influence allocation of by politicians (affecting location + resources types of project investments) · Bribery in sector budgeting · Corruption in national and sector management (influencing, planning and budget management distortions in funding allocation) (misuse of funds, inter-ministerial (national and local) Planning and bribery for fund allocation, Budgeting collusion/bribery in selection and project approval) · Corruption in local budget management (fraud, falsification of accounts /documents, village level collusion) · Donor-Government collusion in · Donor and national private operator negotiations to meet spending collusion (outside legal trade Donor /funding targets agreements) financing · Donor ­ Government collusion/ fraud wrt to progress and quality. · Bribery,rentseeking and kickbacks Fiscal to ensure fund transfers between transfers MoF and sector ministries · Corruption in personnel · Influence project decision-making management · Bribery for preferential treatment, ­ payments for preferred elite capture) candidates (e.g. utility · Distortionary decision-making Management directorships) (project level­ site selection, and ­ payments for promotions, and equipment, construction) Program transfers, salary perks) · Distortionary decision-making Design (collusion with leaders in selection + approval of plans/schemes) · Corruption in LG and departmental planning and budget management table continued on next page... 14 Tackling Corruption in the Water and Sanitation Sector in Africa: Starting the Dialogue Table 1. Value Chain Framework ­ WSS Sector Corrupt Interactions Public-Public Public-Private Public-Consumers/Civil · Administrative corruption (fraud, · Bribery to influence contract /bid falsification of documents, silence organization payments) · Corruption in delegating · Inter-department / agency collusion management: Fraud to over/ over procurement and construction underestimate assets; selection Tendering and and type, award of concessions; Procurement decisions over duration, exclusivity, tariffs, subsidies. · Corruption in procurement: Inflated estimates for capital works, supply of chemicals, vehicles, equipment · Falsification of documentation · Corruption in construction ­ bribery · Corruption in community based and fraud wrt construction (with similar types ­ Not building to specification, of practices as for public-private concealing sub-standard work, interactions. Construction materials ­ Failure to complete works­ Underpayment of workers · Fraud - in invoicing ­ Marked up pricing, over billing by suppliers · Over billing by suppliers, theft/ · Administrative corruption for water diversion of inputs (chemicals) (access to water ­ installing/ Operation · Avoiding compliance with concealing illegal connections, and regulations, specifications, health avoiding disconnection, illicit supply maintenance and safety rules. using utility vehicles). · Falsification of accounts · Administrative corruption for speed (or preferential treatment) ­ repairs/ new connections. Payment · Administrative corruption (for services) repayment / billing ­ fraudulent meter reading, avoidance or partial payment, over-charging. opportunities for fraud and extraction as well as extortion by departments and units mandated with the approval for funding WSS programs. In some cases approval will pass through various units of various ministries ­ Planning, Finance, Public Works Ministries, and Water departments and a number of sub-national agencies and their own sub-departments. Corruption between the tiers of government, irrespective of the sector, frequently concern personnel management: bribes for promotions, appointments, transfers, and a multitude of perks. Buying senior appointments is standard practice throughout the region and the prices paid for utility directorships or municipal engineers are often common knowledge and calculable, based on sector norms.25 Many argue that these types of practices, common throughout the civil service, lie at the core of the incentive and patronage system and propagate other forms of corruption. Corrupt politicians and managers might also appoint willing personnel to lucrative positions on the condition they pass on a portion of the income the office earns. 25Informal feedback in more than one African country suggests that this could be as much as a few years salary for large utilities . Bribes for appointments are thought to be made to both politicians and to management staff although it is not known in what proportion. Davis found in the context of South Asia this was mostly bribery of politicians and local leaders to exert influence on decision-makers. 15 Table 2. Key Public Actors involved in the WSS sector Public actors non-water sector Public actors water sector · Government leadership National political actors · Policy-makers · Judiciary National agencies · Enforcement agencies · Regulators · AC forums National line departments · Directors in Finance, Planning, Health, Interior / · Directors in Public Works, Water ministries management Home departments · Deputy directors (finance, planning) · Deputy directors (finance, operations, customer National planning and support) · Departmental heads at national provincial levels implementing staff (planning and budgeting) · Procurement staff Provincial / local · District/ local level leaders · Public works officials Political actors · Mayors / Town Clerks · Municipal engineers Management · Municipal procurement staff · Utility directors Utility staff · Engineering staff · Utility procurement and project management Middle level staff Local implementing staff · Senior operations staff · Technical managers, Engineers, technicians, supervisors, Facilitators Community leaders and · Village leaders · Management user committees quasi government officials · Lower level public officials · Lower level public officials Public to public interactions might also include the collusion between government officials and international donor representatives in the targeting of donor financing.26 International donors are under pressure to disperse loans or grants, and to maintain relationships. In the water sector, this can lead to an emphasis on quantity over quality, and speed over specification. Furthermore, donor representatives, like all employees, have incentives to deliver and to be seen to be managing successful projects delivered on time and cost. On the government side, donor-funding channels large flows through inadequate financial management systems, often dwarfing annual budgets and the capacity of recipient departments, utilities or district offices. Funding also includes for allowances that have a multitude of perverse personnel management effects, giving senior public officials discretion to top up staff salaries by an order of magnitude and developing unhealthy incentive structures and relationships within recipient agencies. Public to Private interactions Procurement ­ requiring the interaction between the public and private sectors ­ is the most publicized face of corruption. Every level of government, and every type of government agency has to purchase goods and services, normally from the private sector. In WSS, a number of public actors may be involved depending on the size and type of project: national and local government politicians and managers, municipal engineers, operations staff, project managers, procurement officers, and a set of private actors that might include suppliers, contractors operators and/or local and national consultants. On large loans or grants it may also involve the collusion of donors seeking preferable terms for donor-country firms or operators.27 26 For a decade or more, governance has led the agenda of bi- and multi-lateral development agencies and support for anti-corruption has been recognized as a key concern in meeting both development and risk management objectives. At the same time, few would argue that development assistance, by its nature, exacerbates inherent corruption. 27 This practice was perhaps more associated with the letting of concessions to international firms during the nineties, but still applies to large construction contracts. 16 Tackling Corruption in the Water and Sanitation Sector in Africa: Starting the Dialogue It is a norm of economic life for private actors to seek to reduce competition, and private stakeholders do so in the WSS sector at all levels ­ it is clear that the supply chain creates a number of concentrated opportunities for private gain.28 The Lesotho Highlands project provides useful insights into procurement related corruption in the water sector (Darroch, 2004; Earle and Turton 2005). Corruption in relation to public procurement seeks to influence the selection of contracts for WSS services and supplies, payment schedules, profit margins and the outcomes of the regulatory process. In urban water supply, much of the attention on public-private interactions has inevitably focused on transactions for WSS operations. The practice of creating public-private partnerships behind closed doors created unworkable agreements in Africa, muddied incentive structures and undermined the possible benefits of reform attained through private sector efficiency and effectiveness. Corruption in design, tendering and procurement, described in detail in Chapter xx, typically occurs in relation to the construction of WSS infrastructure and facilities:29 · Before contracts are awarded, high level officials can influence the way a contract is let, determining the nature of the project (e.g. higher investment projects), and then the type of contract (e.g. DBL, concession or franchise service area). Purchasing officials of a utility, municipality or district office can tailor specifications to suit favorite suppliers. · During tendering corrupt practices restrict /provide information about contracting opportunities, create excuses for sole sourcing or uncompetitive selection, breach confidentiality or disqualify suppliers and accept/ solicit bribes to influence tender lists or selection procedures. Contractors and operators falsify records and documentation to ensure bids look competitive and officials either encourage them to do so, or turn a blind eye. It is also common for private suppliers of pipes, chemicals and other inputs, consultants and contractors to collude among themselves, take turns in bid-winning or to mark up pricing. Collusion and the falsification of records are often known to the procurement official who receives a kick-back for his/her silence. Corruption in construction occurs: · After contracts are awarded, bribery and fraud is similar to other parts of the construction sector. Oversight officials are bribed or extort payments to ignore instances when specifications are not adhered to (e.g. depth of pipe work, foundation materials) or works are not completed, or when lower quality materials are used (e.g. type of pipes), and provide fraudulent documentation. Typically these practices help contractors minimize costs and result in sub-standard works, affecting sustainability and quality of WSS service delivery.0 · Claims for payment. Supervising officials are bribed or extort payments to agree to falsified claims and/or accounts, to facilitate the speedy approval of payments; and regulators are bribed or extort payments to keep silent or permit waivers of works not completed to standard. Corruption in operations and maintenance can occur in all service arrangements and varies according to the actors involved: Publicly operated WSS services: Be it in cities or small towns, the public sector (utilities, municipalities, district governments and line departments) is mainly responsible for operating and maintaining the water treatment and distribution systems and sewerage or sanitation services1 and have interacted with the private sector mainly through the supply of goods (chemicals, pipes, meters and other hardware) but increasingly do so in the provision of services which can be unbundled from the main water supply function. This includes for instance, earthworks, billing and collection, or security, cleaning etc. The interactions 28 The lack of detailed knowledge of public-private interactions in the water sector is exacerbated because diagnosis has not always included the right actors ­ those paying the bribes. There has been inad- equate critical engagement with how mechanisms work on the ground and why it is likely that they will continue to do so (Sohail, 200). The sector is acutely aware that addressing the actions of private firms, the bribing and solicitation process, is as important as addressing government actors but more diagnosis of public-private interactions is needed. 29 Bribery in procurement and construction is described in detail in Transparency International, 2005. The water sector does not differ to other construction sectors. 0 The construction of rural and urban water supply and sewerage projects are thought to reflect the same types of practices exposed in the roads sector (DFID, 200). 1 Only 5% of water delivered in Africa is operated by the private sector, and this is predominately found in West Africa in concession type arrangements (WUP, 2001). 17 Box 1. Utility officials extort bribes from small water providers in Nairobi The sprawling squatter settlement of Kibera lies in an area of Nairobi not served by the city water utility and the lack of water provision has resulted in a market of water kiosks - tanks with taps, providing a neighborhood source of water to the half a million residents. Householders queue for water during the times the utility allows the water to get through the bulk supply points established by the providers, and pay -20 Shillings per 20l can. The price is fixed and competitive within the squatter settlement, although it is 5? times the price of utility water, and varies according to the season and availability of water. The small scale private providers are increasingly organized, have formed an association and developed a code of ethics to ensure they all follow a set of agreed rules, and to create a platform with the capacity and weight to interact effectively with the utility. They see this as being a critical vehicle to counter the regular petty corruption of Nairobi Water Utility officials in meter reading, billing and collection. Bills are erratic and hugely inconsistent. There appears little the providers can do to bring the bills back in line. And so they tip the officials to revise the bills. The irregularity of the bulk water supply to the provider kiosks provides the utility with leverage over the providers and incentive for them to grin and bear the extortive demands. The losers are the poor who pay a higher price for their water each time this `surcharge' is levied. Source: WSP-Africa 2005b; Author interviews, March 2005. between the contractors and suppliers providing these goods and services falls prey to the same types of interactions seen in the procurement and construction, be they part of a recurrent budget, that may or may not be tendered. Concessions: In the context of large cities where the water supply function itself has been delegated to private operators, opportunities for bribery and fraud are created by the way the deal is structured at the outset and played out throughout the duration of the contract. Experience suggests that the types of contract can be influenced by private-public negotiations (concessions, DBL, leases, management contracts), the concession area and bundle of services, their duration and conditions (such as exclusivity), capacity can be over or underestimated, assets over or under valued, the level and process of tariff setting, manipulated, and the targets and subsidies for serving the poor determined through public-private deals. Small private providers: In small and large towns and peri-urban areas, an alternative set of corrupt interactions occur in the water market between public (local government and utility) officials and small private providers of water. Investigation into the actions of small private water providers in squatter settlements reveal that their ability to function is often dependent on the deals they do with local officials. In squatter settlements in Nairobi for instance, legal water kiosk operators reported they struggle with inconsistent, irregular billing (Plummer, Mehrotra and Collignon, 2005). Legal operators bribe officials to obtain more reliable and longer daily bulk supply, while those functioning illegally pay officials to connect into the network or deliver bulk water that they then distribute in a competitive market. In the sanitation sector, small-scale private operators pay local government officials to allow them to dump waste on inappropriate sites, with health and environmental consequences. Public to Consumer/Civil Society interactions Corrupt interactions between consumers and public officials, mostly in the form of bribery, are a wholly different matter. For the briber (the consumer), water is the desired outcome, and the incentive is to obtain a much-needed basic service. The corrupt interactions that take place between public water sector officials and poor water consumers are petty, frequent and systemic (they may be either extortive or collaborative).2 2 Extortive behaviour is a reflection of an imbalance of power, and typically can mean officials are exploiting the poor. For the poor, the outcome may be access to water service, but the transaction price is high, probably not at `market value'. This often occurs in a monopolistic situation where consumers are dependent on decision-making officials, and is found in network and non-network situations, community or utility-managed, rural and urban. Collaborative transactions between the poor and public service delivery officials are voluntary transactions that often provide the poor with a service they would not have had otherwise, and at a price that is competitive with other local alternatives. In squatter settlements for instance, officials provide illegal connections or tanker water at a price and arrangement the consumer is willing to pay. These are frequent and widespread. 18 Tackling Corruption in the Water and Sanitation Sector in Africa: Starting the Dialogue Common corrupt practices at the point of service delivery in the operations and maintenance of water services, include for instance, officials providing illegal connections, using utility water for resale in utility vehicles, or offering preferential treatment for repairs or new services (speed money). · In rural areas, corruption affects the delivery of community-based and NGO supported water supply and sanitation projects in their design, implementation and ongoing maintenance. Although there is a lack of systematic assessment in Africa, collusion between village leaders and government overseers in ways that detrimentally affect the poor are frequently visible. In the initial stage of projects, for instance, it is common to see design decisions (location of pumps or tanks) benefiting the elite through efforts to influence allocation. In implementation, efforts by community leaders to increase profit reflect typical public-private procurement and construction fraud and bribery (theft of materials, not building to specification), and in project management involves malfeasance: fraudulent documentation, accounting and reporting by those tasked with managing finances. The cost of rural bore holes in Africa ­ up to fourfold the cost of some parts of Asia ­ is considered by sector professionals to be a prime hotspot for further investigation (Narkevic, 2005). · In urban areas, community-based WSS projects suffer from similar patterns of behavior, distorting the type of installation selected, and ongoing management. Where the poor are served by utilities, they frequently pay bribes to officials to obtain access to services household connections and/or repairs off the utility-books, (and sewage disposal services). In squatter areas, the level of the bribe may be pitched at a level the poor can afford and thus becomes part of the cost of accessing services (and some poor households will be able to afford it better than others). In other situations where the poor live in mixed-income settlements, and the water market is differentiated, higher-income households are prepared to pay more and the bribe is likely to be higher, marginalizing the poor and placing them at the end of the queue (Collignon, 2005). Other common public-consumer interactions concern administrative corruption (bribes, fraud) in payment systems: falsifying meter reading, irregular billing, or avoiding officials over-charging (Davis, 200). Typically where poor consumers are involved, the bribe is demanded rather than offered. Most of these services result in commercial leakage adding to the inefficiencies of dysfunctional agencies. In surveys conducted in Mozambique4 12 percent of households reported that it was always necessary to bribe officials for services, over 20 percent of user-enterprises reported that they paid bribes in over 25 percent of transactions, and almost half reported that it made no difference which official was involved ­ suggesting an institutional norm. These figures however are low. In the TI survey conducted in Nairobi, over 60 per cent of households and institutional customers reported they had experienced corruption in relation to utility officials (TI, 2006) This corruption however is part of a series of failures (weak policies and institutions) that result in a lack of services and inevitably create a market for corruption at the point of service delivery. It is vital that this corruption be viewed as a part of a system, it is far more complex than a simple picture of officials forcing consumers to pay bribes to obtain a service they are entitled to obtain free of charge. Where there is no alternative supply (and especially where small private providers have not developed an informal supply market), it is common that poor and non-poor households create the demand for `corrupt water'. They need the officials to provide the water and are not so concerned with the terms. Similar to costly `informal' water, corrupt water fills a gap created by ineffective agencies (Plummer and Cross, 2005).5 In these cases corrupt officials, acting as informal providers, provide the Despite the undeveloped NGO sector in WSS in Africa, NGO service delivery is a part of a range of vehicles to mobilize services to the poor. But nor should one assume that all Government/ NGO transactions are free of corruption, the assumption that NGOs are corrupt-free and held accountable needs investigation (ref PREM week, 2006). 4 World Bank Institute Governance and Anti-corruption Survey conducted by Austral Consultoria a Projectos, undated. 5 It is recognized that this practice also contributes, in some cases, to the inefficiencies of the utilities. But in many case it probably does not. To some extent too, the problem is caused outside the water sector. A number of characteristics typify the situations in which the poor rely on corrupt water. Typically tenants and illegal squatters have less WSS options and are more likely to tap into the service a corrupt official (or a landlord) might offer. The illegality of their dwelling makes this group vulnerable and susceptible to extortion and high priced water. Other households with changing needs and capacities are also susceptible as they may wish to opt in or out of a higher level of service creating the opportunity for extortive or collaborative bribery. 19 poor with services they may not have otherwise obtained. This presents the sector with a dilemma, at least in the short-term, that needs to be carefully managed to ensure poor household's ongoing access to water. 2.4 Identifying concentrated areas of corruption In practice each WSS context will see different areas of concentration or vulnerability to corruption in policy-making, planning and budgeting, fiscal transfers, procurement, personnel management, construction and service delivery, all of which respond to the factors determining the local and sector corruption scene. This section has tried to provide a picture of the range of corrupt practices in WSS, to promote comprehensive diagnosis, it does so to `locate' the vulnerable spots on which action can be concentrated, and to stress the importance of inter-linkages and connections ­ not to advocate for a total system of reform. In practice too, corruption varies significantly between countries, within countries, between agencies/institutions and within sectors (Recanatini et al, 2005; Guggenheim, 2006). Understanding the weighting (intensity/incidence) of corruption at any one point in this matrix in the sector and institutional context is vital to ensure anti-corruption efforts concentrate on major points of leakage ­ good policy and effective mitigation efforts can therefore only come from good diagnosis. Country level diagnosis is vital. 3. TACKLING CORRUPTION IN THE WSS SECTOR IN AFRICA Over the last decade, a number of theories and mechanisms have been developed and tested to tackle corruption in its various forms. Policy changes that support anti-corruption activity generally range from removing trade barriers to more targeted actions minimizing regulation and licensing, and strengthening accounting and auditing standards. Institutional reform strategies have included the strengthening of judicial and legal systems, improving budgeting procedures and financial management as well as mainstreaming civil service reform, and focusing on incentives, competition, and internal checks. More specific anti-corruption strategies have included enforcement, watchdogs and ombudsman, awareness campaigns and anti-corruption commissions. Despite this menu of anti-corruption options and the increasing resources from development agencies devoted to the fight against corruption, many commentators highlight the lack of rigor and specificity in the development of anti-corruption programs (Shah and Schacter, 2004). In many cases corruption has not been identified accurately or matched with appropriate interventions. This section on anti-corruption interventions aims to briefly consider the array of instruments for tackling corruption and position these in relation to the WSS corruption interaction framework provided above. To focus the discussion, it concentrates on the most prevalent area of corruption for each of the public-public, public-private and public-consumer/civil society interactions. This is done with some caution, knowing that generic lessons are continually emerging, that sector lessons are few and far between, and in the full knowledge that any anti-corruption strategy would be developed at the country level to respond to the country- specific nature and causes of corruption. 3.1 Understanding and addressing incentives ­ a key starting action Irrespective of the actors involved, corruption flourishes when the incentives exist for it to do so. Corruption is driven by need, greed or opportunity for money or power (Klitgaard, 2002) or the need for water services (Plummer and Cross, 2005). Public incentives might suggest that lower-level poorly paid officials have the need to supplement their income, middle level managers have ample opportunity, and politicians, senior managers and directors are driven by greed. Yet public officials at all levels 20 Tackling Corruption in the Water and Sanitation Sector in Africa: Starting the Dialogue shoulder a range of responsibilities and must meet the obligations that come with their position (Burgess, 2006). Individual and organizational incentives are complex, closely aligned with socio-political and institutional structures and often driven by factors outside the sector. An understanding of incentive structures developed in governance and civil service reform initiatives (Huther and Shah, 2000) can be applied to water sector institutions and relationships. The key concern for any corrupt official is that the potential gains of their actions outweigh the potential losses. To change behavior of officials it is necessary to lower expected gains and increase expected penalties: · reducing the incidence of transactions ­ through policies and organizational reforms that reduce discretion, monopolies, bureaucratic procedures and clarify functional responsibilities for regulation, policy-making and delivery; by increasing service standards and decentralizing government services. · reducing the gain from each transaction ­ by scaling down large projects or improving the contract management of those projects (e.g. creating transparent decision-making processes), by de-monopolizing public services and promoting sector competition. · increasing the probability of detection or penalty ­ by clarifying procedures and streamlining operational roles, by increasing accountability and transparency through citizen /consumer participation and monitoring of WSS services, establishing citizens charters, specifying standards, media independence, promoting transparency in interactions between public and private sectors, and providing support for whistleblowers.6 · increasing the magnitude of the penalty ­ the key to enforcement is meaningful penalties (but these are currently rare at the sector level in developing countries). 3.2 Key dimensions of good governance in WSS sector anti-corruption activity Experience in anti-corruption activity to date has provided a set of anti-corruption mechanisms fundamental to good governance and applicable to water sector ant-corruption reform. Best practice suggests that more demand-side efforts are needed (see detailed discussion in chapter xx) to support the technical approaches to improved sector performance carried out in many countries over the last decade. Further consideration is needed of the applicability and impact of all approaches, their blending and sequencing within the water sector, and in particular contexts. Key areas of action include: 1. Diagnosis and measurement Effective anti-corruption policy depends on sound diagnosis and understanding of the sector context (as outlined in the previous section). In addition, measurement systems that enable sector benchmarking, and monitor relative progress can be used to raise awareness and focus efforts. There is much debate over the appropriateness and accuracy of corruption indexes: the Global Perception Index (TI, 2006), the corruption measurement approach (Olkren, 2005), and the measurement of anti-corruption instruments and policies such as the Global Integrity Index (Heller, 2006) similar in objective to the Water Reform Ranking developed by WSP-Africa (WSP-Africa, 2006), provide three possible approaches for sector consideration. 6Although outside the sector, strengthening the rule of law so both individuals and the media do not fear reprisals (Huther and Shah, 2000) 21 Demand-side 2. Transparency and access to information Promoting greater transparency around the actions of politicians and water sector officials creates disincentives for their engagement in corrupt transactions. Transparency can be developed in various forms at the project, community or sector level ­ publicizing utility accounts, budgets, contracting arrangements and annual reports, and public hearings by regulators ­ are all tangible mechanisms for improving transparency in WSS services. Access to information is essential to improve demand for accountability. Typically consumers have no knowledge of recurrent and capital costs making it possible for public officials to make decisions that misallocate resources, or tap into limited budgets, and they need to be able to access information about complaints mechanisms and their rights as consumers and citizens. A significant contribution to political accountability and a transparent operating environment is also made by increasing the role of the media (Stapenhurst 2000), and by utilizing e- government for transparent record management. . Accountability The development of accountability, be it transparency, professionalism, honesty, or competence, is central to tackling corruption in WSS, and applies to all parts of the service delivery framework.7 Efforts focused at improving accountability in the WSS sector occur on both supply and demand sides. On the demand side it includes the institutionalization of surveys, mechanisms to strengthen civil society's role in monitoring (for instance through consumer voice and client power through report cards such as the Bangalore citizen report card) and consumer associations (Thampi, 2005). Introducing processes that rely on a consumer/ citizen role requires efforts to support consumers to play this role. This is particularly important in Africa where civil society tends to be relatively immature and low in capacity. 4. Education and advocacy Lack of awareness is currently a key factor preventing action in Africa. Better awareness of the causes and consequences of corruption is urgently needed by politicians, high-level officials, the media, and the general public but the most effective means to generate that awareness however in countries with weak governance is yet to be found ­ most advocacy campaigns have failed.8 Evidence also suggests, in rural India (reference)9, that higher levels of education and literacy correlates with lower corruption, more accountability, better targeting, and less political capture. Supply-side 5. Institutional and policy reform Efforts should not only focus on the demand-side. Reforms addressing the complexities and ambiguities of country policy, regulatory and institutional frameworks and decentralization are vital, and are underpinned by civil service reform in the areas of personnel, organizational and financial management. Efforts to create an environment of accountable service provision are central to these reforms, all create the structural change needed in WSS service delivery systems and reduce opportunity for corruption. 7 The World Development Report, Making Services Work for the Poor, provides a triangular framework of accountability relationships between policy makers / politicians, service providers, and citizens. It high- lights the need to shorten and strengthen accountability relationships of voice, compact and client power. World Bank, 2004. 8 Shah and Schacter argue for instance that corruption awareness campaigns have universally failed in developing countries, that advocacy is only useful in countries where governance is strong. Shah and Schacter, 2004 op cit. 9 Reference from PREM week presentation? 22 Tackling Corruption in the Water and Sanitation Sector in Africa: Starting the Dialogue These technical approaches include for instance: · better sector planning and policy-making · alignment of the functions of different tiers and departments · separating provider, regulatory and financing responsibilities · corporate governance (e.g. reform of utility management) · internal procedures (e.g. contract management) · addressing monopolistic and uncompetitive systems; and · partnerships with private operators to improve efficiency. · results based performance approaches.40 6. Leadership The need to embed the anti-corruption struggle in broader reform and strong leadership and good management is critical.41 Central to the anti-corruption institutional reform agenda is leadership: the struggle will be carried by leaders needing to cast the drive against corruption as a part of their effort to expand and improve services. Successful attacks against corruption have been frequently led by outstanding leaders as seen in the municipal reforms by the Mayor of Le Paz from 1985 to the late 1990's (Maclean-Abaroa, 2006). 7. Integrity Efforts to promote integrity and ethical behavior among institutions and actors involved in transactions have been developed and tested with some success in developed countries and are gaining momentum in developing countries. The `integrity pact' promoted by Transparency International binds actors in a transaction (such as bidders and government agencies in a contractual relationship) to not offer or accept bribes in public contracting. Codes of conduct or codes of ethics and training around ethics issues have been developed by government and professional associations with the aim of binding members to corrupt-free behavior. Other mechanisms include business principles, conflict of interest laws and rules, and whistleblower protection. Most of these approaches are potentially useful to clean up the international private sector that operates in developing country water sectors, but may have limited applicability in developing countries until transparency and accountability frameworks are established. 8. Enforcement and regulation A large number of countries have introduced anti-corruption legislations, regulations, rules and procedures aimed at controlling corruption. However, in most developing countries, these sanctions-based approaches are rarely accompanied by effective enforcement and many have proven ineffective.42 Ombudsman and complaints offices, oversight committees, `watchdog agencies' and independent auditing introduced to formalize processes and enforce punishment; and special anti-corruption and fraud agencies to detect corruption and promote preventative measures have all been tested, but with little success in the context of weak governance (Shah and Schacter, 2004). At the international level, the UN Convention against Corruption provides a common framework for 40On the supply side efforts to strengthen the accountability between WSS service providers and policy makers, better functional allocation and establishing results-oriented approaches, (e.g. OBA mechanisms) are gaining momentum (Halpern and Mumssen; Tremolet and Halpern, 2006). 41Corruption is closely linked to legal but misguided policies such as low tariffs or water sources and many basic remedies are synonymous with the efforts that are part of what effective utilities are doing to improve performance. 42The police and judiciary are themselves riddled with corruption ­ making sanctions impossible without extra-judicial action (such as high-level political drives). 2 countries to tackle corruption and provides for cross-border co-operation. International Finance Institutions have also established rules and sanctions, but to date a lack of harmonization between financiers/donors has made efforts less than effective.4 3.3 Aligning potential anti-corruption mechanisms to corruption in the WSS sector This set of anti-corruption instruments work at general, sector-specific and project levels. Many mechanisms cut across sectors, institutions and agencies and address many forms of corruption ­ these often improve governance generally, and may be effective in creating an enabling environment that discourages corruption in all sectors. To provide an overview of the tools available and their relevance, the following discussion describes the types of mechanisms that might be applicable to each of the interactions, described in Table 1, at each of the relevant points in the delivery process. Tackling corruption occurring within government (public-public interactions) To date there has been little effort to tackle corruption involving water sector officials directly. In the past, corrupt interactions internal to the public sector ­ within or between government tiers, departments, WSS agencies and individuals ­ have generally been addressed through indirect initiatives aimed at civil service reform and improved accountability through decentralization and other political reform processes. While it is not clear the extent to which these have been successful in combating corruption,44 action has focused on: · WSS sector restructuring, policy reform and organizational change (openness to leveraging private sector and other local stakeholder involvement, competition, as well as civil service size, stronger leadership, sector coordination, separation of policy, regulation and implementation, more clarity in the allocation of functions); · Decentralization and accountability reforms (improved political accountability, mechanisms to improve accountability for basic services at local level ­ citizen report cards, service surveys, increased public awareness, improved accountability between tiers of government); · WSS planning and financial management (financial policies, ring-fencing, increased transparency, cost recovery and improvements in metering, billing and collection, as well as oversight, reporting, auditing processes); · Personnel management reform (pay structures, promotions/appointments/recruitment, transfers, performance-based management, terms and conditions, enforcement and sanctions). These instruments may have mixed impact in the reform of sector agencies, especially where mandates are limited. In practice, a hierarchy of officials from a ladder of agencies interact with each other ­ within and outside water agencies. In many countries it will be futile to build capacity and accountability in service delivery agencies (utilities, district administrations and village water committees) without also tackling higher levels of government, or local government owners and decision-makers. Similarly it may be unproductive to work on developing accountability in a Ministry of Water, without the implicit agreement or participation of political leaders and the Ministry of Finance. 4 In the Lesotho Highlands corruption case not all donors blacklisted the companies found guilty of corruption. Earle and Turton, 2005. 44 Recanatini's work highlights a number of features as being characteristic of less corrupt institutions/agencies ­ all strongly reinforcing of WSS sector reform efforts: decisions being regularly audited by external or internal auditors, maintaining open and transparent procedures and basing personnel decisions on criteria of merit and professional competence, and appointing leadership and managers through a political rather than a popular process (Recanatini et al, 2005). 24 Tackling Corruption in the Water and Sanitation Sector in Africa: Starting the Dialogue Table 3. Tackling hotspots in public-public corruption Public-Public Interactions Early Warning Indicators Potential Responses · Bribery of decision-makers to · Low tariff structures · Policy and Tariff Reform influence policy priorities. · Monopolies and a resistance to Policy-making introduce competition · Influencing regulations and · No division of regulator and · Separation of regulator and licenses provider roles provider roles, incl.PSP. · Bribery to bypass constraining · Renegotiation of contracts · Development and publication of regulations. (frequency and timing) minimum standards. Regulating · Appointment of regulator · Public oversight of negotiations with operators · Distortions in decision-making · Lack of coordination between · Independent auditing of by politicians (affecting location Ministries of Planning and decisions made + types of project investments) Finance ­ no. inconsistencies · Organizational and procedural · Corruption in national and · Speed and complexity of change in budgeting and finance sector planning and budget budget allocation ­ no of hands functions management (inter-ministerial approving plans and budgets, · Clarify, legislate and implement Planning and bribery for fund allocation, time taken. decentralized functions collusion/bribery in selection and · Overlapping roles and · Review options for alternative Budgeting project approval) responsibilities in delivery stage delivery systems (e.g. PSP, · Corruption in local budget · % spending on capital intensive community) management (fraud, falsification / large projects · Citizen involvement and demand of accounts /documents, village · Inappropriate discretion of MOF for accountability in planning and level collusion. and planning officers budgeting. · Media involvement Donor · Donor ­ Government collusion · Unexpected change in donor · Transparency in negotiations, financing in negotiations to meet spending support / choice budgets and proposed plans /funding targets. · Donor ­ Government collusion/ fraud wrt to progress and quality. Fiscal transfers · Bribery and kickbacks to ensure · Long process /time for fund · Performance standards fund transfers between MoF and transfers · Auditing sector ministries. · Unexpected release of funds · Corruption in personnel · No. unqualified senior staff · Staffing reforms promote Management management · Poorly paid staff with significant competition, performance /merit and ­ bribery for preferred extras, living beyond means based career structures. Program candidates (e.g. utility · No. of unplanned transfers · Disclosure of assets Design directorship, municipal · Lack of autonomy of LG and · Transparent appointment of engineering positions) utility management qualified administrative leaders ­ bribery for promotions, · Conflict of interest on (and election of political leaders) appointments and transfers, management board · Ringfencing finances of utilities, salary perks). · Increase in price of informal separation of roles between LG · Distortions in decision-making water and utilities. (in selection + approval of plans/schemes) at and between central, local and village levels 25 Table 4. Tackling hotspots in public-private corruption Public-Private Interactions Early Warning Indicators Potential Responses · Bribery to influence contract /bid · No. same tenderers short listed, · Procurement reform organization. selected (possible cartels) · Integrity pacts with tenderers · Corruption in delegating · Dropping out of bidders and their representatives, management: Fraud to over/ · % Single source supply (accompanied by sanctions) underestimate assets; selection · % Standard design unchanged · Simplification of tender and type, award of concessions; · % Higher unit costs documentation decisions over duration, · lowest tender repeatedly not · Review role of middlemen and Tendering and exclusivity, tariffs, subsidies. selected local consultancy `commissions' Procurement · Corruption in procurement for: · % Immediate renegotiation of · Transparency in public-private ­ Capital works contracts interactions - publication of ­ Supply of chemicals, vehicles, tender documentation and equipment etc tenders received. ­ Marked up pricing · Independent tender evaluation · Falsification of documentation · Audits and reporting · Reduced size of contracts · Citizen oversight / monitoring · Support for improving integrity of business (professional associations, codes of conducts) · Corruption in construction · Resistance to meeting standard · Citizen role in oversight and ­ bribery and fraud wrt specifications monitoring ­ Not building to specification, · No. changes in specification · Technical auditing, spot checks Construction concealing sub-standard work, · No. variation orders in site works of infrastructure constructed materials · Citizen auditing, public hearings ­ Failure to complete works · Performance based, · Underpayment of workers management contracts · Fraud and bribery in invoicing ­ Marked up pricing ­ Over billing by suppliers · Corruption in operations · % Single source supply · Integrity pacts (in countries with · Corruption in supply chain · % Change in quality or coverage capacity for sanction) · supply of chemicals, vehicles · Increase in informal price of · De-monopolizing operations and and other inputs water maintenance · Avoiding compliance with · Financial and technical audits regulations, specifications, health and reporting and safety rules. · Performance based contracts · Falsification of accounts / with defined minimum standards evidence · Transparency in public-private interactions -- publication of Operation Small / informal providers tender documentation and and · Inconsistent, irregular billing to small scale providers tenders received. maintenance · Extortion to enable businesses · Citizen role in oversight and to continue monitoring, auditing, public hearings, complaint mechanisms · Benchmarking utility performance Small / informal providers · Improve interface between formal and informal systems ­ SSPP legitimization, · Formal bulk supply, pricing, competitive tenders for area franchises. 26 Tackling Corruption in the Water and Sanitation Sector in Africa: Starting the Dialogue Box 2. Institutional reforms improve utility performance in Nairobi For decades the Nairobi City Council (NCC) Water and Sewerage Services Department was responsible for the delivery of WSS services to the residents of Nairobi. In 2002, because it was widely considered ill-equipped to delivery quality service due to corruption, the water service functions and assets were relocated to a newly founded Nairobi Water and Sewerage Corporation (NWSC) incorporated as a private company. The NWSC is now in existence ­ run by a board of 12 Directors drawn from private sector organizations and professional bodies and a number of ex-officers from the NCC. The provision of a supportive legislation (the Water Act 2002 as well as the provisions of the LGA) provided for the reform of the institutional framework and enabled the establishment of an independent entity for water supply in the city. The essential reform elements were to separate out policy, service provision and regulatory functions improving the climate of transparency and accountability on the sector and creating 7 autonomous water boards who appoint water service providers. By 2004, the NWSC was formed in Nairobi ­ inheriting 2200 staff and the operational structures of the former local government Water Service Department. The policies and reforms put in place by the NWSC saw the emergence of a reform-minded organizational culture made some immediate difference to the perceptions of consumers and suppliers. The creation of the Nairobi Water Company brought new leadership, clearer accountabilities, introduction of a commercial culture with improved service conditions, greater transparency of operations and better communication policies. Nevertheless those mandated with reform still saw that corruption as a major obstacle to efficiency and profitability and saw the need for better understanding of this corruption. The assessment conducted under this new management exposed potential corrupt practices of staff in relation to both public-private and public-consumer interactions. Although the sample was small, the assessment conducted by Transparency International, pointed towards a number of corruption hotspots and areas of action. Corrupt practices between officials and consumers The assessment revealed that the major forms of corruption with regard to consumers were: (i) at the point of service delivery ­ bribery to avoid being disconnected or to get reconnected, and solicitation for bribes / rent-seeking for illegal connections; and (ii) in relation to billing and collection ­ solicitation for bribes to get hold of documents, records or to get solutions for inflated bills. About 65% of those domestic and institutional customers surveyed had experience of corrupt practices with NWSC, mostly with the lower levels of officials. Inflated billing was also found to be a major cause for illegal connections, particularly from domestic consumers. Over 50% reported a lack of fairness regarding consumption versus billing, 6% in some for of dispute, and no disputes being satisfactorily resolved. There was a recurrent accusation that NCC and now NWSC took advantage of its monopoly status, insisting on payment of inflated bills without record or justification. The assessment has pointed towards the need for immediate improvement in record keeping and accuracy and frequency of water meter reading, but also to practical steps toward better dispute resolution over inflated bills. To this end the design and implementation of a consumer liaison process and strategy has been recommended. Corrupt practices between officials and private companies The survey comparing the `before' and `after' corporatization also included the perceptions of NWSC suppliers. The findings included: (i) the incidence of corrupt practices has decreased since NWSC took over operations, ostensibly because institutional reforms were working; and (ii) these was still a lack of transparency and possibly corruption in the way tenders were awarded ­ some of the suppliers admitted to having inside information. In response to the charges that its procurement procedures were opaque, it was recommended that NWSC should reform the procedures determining the nature of the public-private interface through the revision of procedures to all its suppliers and business associates, ensuring the code of ethics and conduct apply to all staff and requiring up-front action concerning conflict of interest. Source: Transparency International-Kenya, 2006. Tackling corruption between government and private individuals/firms The private sector, in its tendering, supply, construction and operating roles, is a key actor in the determination of corruption outcomes in the WSS sector, and cleaning up the interface between the public and private sectors is paramount to corruption reform. Corrupt interactions between government and private sector companies have been addressed implicitly in the sector through sector reform efforts which strengthen the policy and enabling environment, and through specific contracting and procedural mechanisms within the construction sector in public procurement, construction and operations. To date general efforts to curb corruption in tendering and procurement have focused on prescriptive improvements to procurement environments ­ introducing anti-corruption laws, charters, performance standards, and establishing rules, principles 27 Table 5. Tackling hotspots in public-consumer/civil society corruption Public-Consumer - Civil Early Warning Indicators Potential Responses Society Interactions · Community-based construction · % Loss of materials · Citizen role in oversight and and management: · Resistance to meeting standard monitoring ­ Fraud and bribery in specifications · Technical auditing, spot checks ­ Theft of materials by village of infrastructure constructed Construction leaders · Performance based contracts ­ Fraudulent documentation, accounting and reporting · Administrative corruption · Changes in unaccounted for · Legitimization of illegal for water (access to water water connections ­ installing/ concealing · Unofficial usage of tankers · Legitimization of public officials illegal connections, avoiding · Lack of interest in connection extending services disconnection, non-network campaigns · Review of connection costs (tankers) illicit supply using · Low rate of faults · Performance contracts for Operation public assets). · No. faults reported / complaints speed of repairs · Administrative corruption for · No. connections versus increase · Citizen participation in and speed (speed or preferential in water consumed monitoring and oversight maintenance treatment ­ repairs, new · Unexplained zonal variations · Transparency and reporting of connections) performance requirements · Report cards and other survey and feedback mechanisms · Complaints redressal · Administrative corruption · Unexplained variations in · Information and awareness for billing ­ fraudulent meter payment campaigns reading, avoidance or partial · Complaints from consumers and · Citizen participation in payment, over-charging ... small scale providers monitoring and oversight · Participatory corruption assessments, corruption Payment mapping Systems · Commercialization / Reform to customer interface: metering billing, collection. · Performance contracts for % billed and collected · Complaint redressal · Women cashiers and practices dictating procurement procedures and auditing (see chapter xx; TI, 200). Ideally this would mean that at the point of contact between the private sector and public officials, project and procurement staff were working at a new level of professionalism with less discretion, in a commercial environment where actions were overseen and sanctions enforced if necessary. In practice however the reform of the procurement environment is difficult and takes a great deal of time to be resolved (see Box 2). A second set of efforts focusing at either the project and departmental level aim to improve the localized tendering and procurement `space'. This includes for instance the introduction of integrity pacts dissuading contractors and officials to offer and accept bribes, and establishing sanctions for non-compliance. Transparency in public-private interactions, such as the 28 Tackling Corruption in the Water and Sanitation Sector in Africa: Starting the Dialogue publication of tender documentation and tenders received, independent evaluation of tenders, auditing, monitoring of unit rates, and public participation in negotiations and so on, create a different environment from the opacity that characterized private sector participation in the nineties. In Pakistan, a clean and open bidding process instigated by the Karachi Water and Sewerage Board and monitored by Transparency International showed how the application of a no-bribes integrity pact could be applied to contracts for consultancy services and all physical works and supplies and resulted in a net saving estimated at about 75 per cent of the cost of the contract (TI, 200). A third set of efforts, initiatives with multi-national companies and national companies operating in developing countries have mostly focused on achieving a greater level of integrity and professionalism among members through professional associations, codes of conduct, monitoring and benchmarking, and integrity pacts. (In Latin America the integrity pact has been used successfully in the water sector in Colombia and Argentina and is being transferred to other countries in the region). Transparency International has spearheaded efforts to establish minimum standards for public contracting. The World Economic Forum PACI initiative, the Extractive Industries Transparency Initiative, and construction-industry initiatives in the UK and Europe, seek to improve integrity in private companies, national governments and construction companies respectively (EITI, 2006; WEF, 2006). During construction, private companies appointed to carry out a range of WSS works (construction of treatment plants, networks, boreholes for instance) again interface with officials, be it different actors in different types of corrupt interactions. Anti-corruption measures to clean up the corrupt practices described above have been limited in WSS in developing countries, and the common practice of `looking the other way' is still common. Contractors are mostly national (large or small) working within sector norms. There is little possibility that either party would be brought to justice, and those exposing the fraud or bribery often find themselves in a worse position by doing so. Some efforts to measure and control corruption in construction provide lessons for African village water supply. Evidence from Indonesia suggests technical audits (spot checks on built pipe work), community oversight and monitoring, as well as the simple announcement or threat of auditing (when combined with community participation) seem to reduce corruption in rural development projects (Olkren, 2005). In both urban and rural, small and large projects the introduction of performance-based projects which define minimum standards and requirements and reimburse after the water is flowing show promise at improving accountability and outcomes (GPOBA, 2005). In operations, there has been little explicit focus on reducing corruption but significant effort at reform and efficiency improvement. Reforms at the sector and utility level have focused on improving leadership and management, financial management, ring fencing and delegating roles of operations and maintenance, and increasing citizen participation in planning, budgeting, oversight and monitoring to improve accountability. The separation of provider roles from policy-makers and regulators has been pursued in conjunction with decentralization to some degree, although it is still cloudy at the local level where local government and utility boundaries are less than defined. Monitoring and measurement has improved and benchmarking approaches such as that seen in Uganda [aim to promote efficiency through competition. All such mechanisms aim at establishing an environment less susceptible to corruption. Although it forms a relatively small segment of the WSS sector in Africa, privately operated utilities provide some lessons for consideration. Lessons from Cote d'Ivoire for instance suggest that the delegation of operations to SODECI in the late 1980s led to a shift in the focus of investment (from capital intensive production units) to the rapid extension of distribution networks; the shift to 100% cost recovery from 100% debt financing; from public-private deal-making to commercial procedures (WUP, 200). Efforts `at the interface' between local private providers and those operators responsible for city wide service delivery have started emerging in different forms in many cities in Africa and are central to developing more effective and efficient linkages between public 29 and local private water sector actors. Efforts to form associations of suppliers, develop constitutions and mechanisms for dialogue, formal bulk supply and other technological solutions that result in better bulk supply for the providers (and revenue for the utility), and competitive tender processes for area franchises, all formalize the interface with utilities and municipal water departments, and create more predictable environments for water businesses to flourish (Plummer, Collignon and Mehrotra, 2005). The local private sector is also the home of the middle-men that facilitate bribes between multi-nationals/national companies and governments ­ much greater focus is needed on understanding and developing the integrity of the local private sector. Tackling corrupt interactions between public officials and consumers/civil society To date, anti-corruption mechanisms tackling corrupt interactions at the point of service delivery have largely focused on improving the efficiency of the utility or delivery agency. The sector has long been aware of the various types of corruption that occur `at the tap'. The problem of illegal connections has been addressed through efficiency drives or, in more innovative situations, legitimization programs, while a focus on improved meter-reading, billing and collection, has implicitly addressed some of the leakage that occurs through payment systems. In operation and maintenance, interactions between public officials and consumers, like others are tempered by citizen oversight and monitoring as a part of water governance efforts aimed to stimulate accountability of service providers. On the consumer side, the important work on report cards developed by the Public Affairs Centre in Bangalore (Thampi, 2005), has been applied to the WSS sector in the WSP-supported initiative developing consumer report cards on water and sanitation services in Nairobi, Mombassa and Kisumu in Kenya (WSP-Africa, 2005). Still in the development stage, this initiative is aimed at empowering households, be they poor and non-poor, to monitor the utility services delivered to them. A critical aspect however is the existence or development of an effective complaints redressal system. At a broader level, corruption surveys such as those conducted by WBI, have (to a limited degree) included utility or municipal service delivery, and have provided some insights into community perceptions of bribery between officials and consumers (Austral Consultoria, Center for Democracy and Development, 2000). These are invaluable inputs for policy-makers. Improving utility-consumer interactions will however be limited in direct impact. Only an estimated 2 per cent (WSP-Africa, 2006) of the African population is served by utilities, and few of these are poor. Community-based WSS delivery systems, the preferred donor approach to WSS service provision, has escaped the attention of sector efficiency drives and has beebypassed in the debate over sector corruption. This model of WSS service delivery has suffered from a somewhat naive assumption that community involvement will, by definition, produce accountable and efficient outcomes. Investigation into community-managed rural development programs with sizeable WSS components have, however, uncovered that community management often results in high levels of corruption. Poorly paid defacto public officials frequently act in a non-transparent and unaccountable manner, collude with the project overseers, contractors and suppliers, and engage in a range of practices regarding procurement, construction, payment, as well as decisions which distort project benefits (Woodhouse, 200). Surveys and reports cards and other forms of citizen monitoring can also be applied to community-managed water supply. The consultative citizen's report card survey have proven effective in Maputo in Mozambique, where citizens have pressured for improved urban services (including water) from the municipal authority (Winter, 2004). The innovative introduction of corruption mapping, community monitoring and complaints redressal in the Kecamatan Development Program in Indonesia provides a model for strengthening citizen voice, improving accountability and reducing corruption (see Box ). This initiative raises important issues for WSS public-consumer interactions and community based service delivery in all continents. Assessments have shown that corruption varies in relation to public versus private goods, and moreover that information and participation have proven successful in reducing corruption in relation to private goods (e.g. 0 Tackling Corruption in the Water and Sanitation Sector in Africa: Starting the Dialogue Box 3. Intensive efforts to curb corruption in village infrastructure development: Lessons from Indonesia The Kecamatan Development Program (KDP) is a $1.2 billion World Bank community-driven development project in Indonesia, commenced in 1998 during the East Asia financial crisis that now funds infrastructure development and small loans and has been rapidly scaled up to over 20,000 villages nationwide. Its approach to combating corruption is based on an analysis of the political economy of corruption in Indonesian villages and is two-pronged. First, it aims to change the conditions that breed corruption in villages by breaking existing monopolies over information, resources, and access to justice. Second, it aims to prevent corruption in the project itself by skewing the incentives of the project structure against corrupt behavior. The case is useful to understand what works to limit corruption in a large, rural development project in a country with endemic corruption, a weak legal system and a history of top-down political control by a powerful state bureaucracy. · At the heart of KDP's anti-corruption approach is the principle that villagers themselves have decision-making power over planning, procurement and management of funds. Some of the concrete measures of its approach include: · simplifying financial formats so that they can be understood easily by villagers · transferring funds directly into collective village bank accounts · insisting that all financial transactions have at least three signatures and that at least three quotations are found for the procurement of goods, to be shared publicly at village meetings · insisting that details of all financial transactions are posted on village notice-boards · requiring that regular village meetings are held to account for project funds at which villagers have the right to suspend further disbursements of funds if irregularities are found · providing village-level sources of information and channels for complaints independent of local government · intensive field-level supervision by elected village facilitators and sub-district level project facilitators · independent monitoring of the project by NGOs and local journalists Although these measures have had some success, corruption in KDP persists. Assessments (including in-depth, ethnographic interviews, field experience reviews, and analysis of the incentive structures throughout the project cycle) determined actors' interests, motivations and constraints. The study found that corruption is primarily a problem of incentives, and can be fought effectively only by changing the costs and benefits attached to corrupt behavior ­ that local context and social norms are key to understanding how these incentives can be changed in order to reduce corruption. Corruption in KDP takes several forms, including budget mark-ups, collusion, bribes and kick-backs to local officials. The elements of the project most effective in limiting corruption are transparency, community participation, and the provision of independent channels for resolving complaints. Information and local control are key elements in both preventing and fighting corruption: the most successful strategies for fighting corruption in KDP have hinged on publicizing anti-corruption activities, garnering wide local support, and using sanctions credibly. Project facilitators are also key to fighting corruption: they provide a channel of information to villagers that is independent of local government and, because they are backed by the central KDP structure, they have more protection from threats and intimidation than ordinary villagers. There is evidence also of some governance spillovers from KDP, illustrated by examples of villagers using their experience of KDP as a precedent for protesting against corruption in other projects. The incentives analysis of the project cycle identified the stages of the project cycle most vulnerable to corruption. These lie at the stages of proposal preparation (formation of false borrower groups for small loans); release of funds (collusion among bank account signatories to embezzle funds); and implementation (collusion and corruption in procurement). The analysis highlights several ways in which corruption in KDP could be better prevented. These include improving information dissemination; working with social sanctions to make the incentive structure less conducive to corruption; increasing incentives for KDP staff to fight corruption; instituting measures at specific stages of the project cycle intended to limit monopoly, clarify discretion, and improve accountability; and supporting the capacity of project facilitators to come up with flexible local solutions to their problems. Source: Adapted from Woodhouse, 200 rural water supply) but for public goods (e.g. sanitation, roads), the use of audits, sanctions and enforcement has proven more effective. The program has also raised understanding of `corruption horizons' in a community-managed development program: as corruption is tackled in observable spheres, it moves elsewhere. In the case of KDP, this a shift occurred to less detectable but perhaps less profitable forms of nepotism (Olkren, 2005). The development of more effective payment systems in water and sanitation service delivery has been an integral part of utility efficiency improvements, programs of meter replacement, and professionalization of billing and collection systems designed to reduce fraud and bribery, which change the nature of the customer interface. In some contexts there has been efforts made to 1 ensure these systems are appropriate to poor households (battery meters, collective billing, localizing payment offices). Taking the corruption out of payment systems however requires more informed consumers. More effort is needed to provide households and communities with the information and capacity they need to know if billing is correct, and the systems to appeal if it is not (the supporting role of consumer associations has been explored in East Africa).45 Some innovative approaches in Benin and Cote d'Ivoire have seen the delegation of billing and collection to private contractors ­ who work on a performance contract ­ to improve performance (although their use of entrusted office is equally susceptible) and the introduction of women cashiers who have shown they are less inclined to accept bribes or defraud the utility (WUP, 200). 3.4 A cautionary tale for the water sector Despite this multitude of anti-corruption instruments relevant to the sector, there is still much debate over what works and what does not in developing countries. Cross-cutting activity over the last decade has provided lessons, but there is a need for much greater understanding how they will work at the sector level.46 Models and pilots are needed in the WSS sector to test approaches that promise change in the context of weak African institutions and civil society. Over the last decades a number of anti-corruption mechanisms have been seen to fail or back fire in developing countries. The work of WBI and others has provided analysis of the progress of specific anti-corruption instruments. This analysis emphasizes that the proliferation of anti-corruption commissions, corruption watchdog agencies, and ethics agencies across Africa, the drafting of new anti-corruption laws, decrees and codes of conduct have had little impact (Kaufmann, 2005; ADF, 2004 consensus statement) being more appropriate in countries where public accountability and transparency has been established. A second key lesson developed over recent years is that a focus on the public sector alone is not enough ­ increased salaries do not necessarily lower the incentive of officials to engage in corrupt practices (World Bank, 200), that anti-corruption needs to include the private sector. The problem with `getting it wrong'47 is that anti-corruption activity, if wrongly focused, can result in significant `fallback' in the reformed institutions because corruption becomes more robust, adapts and reestablishes elsewhere when old opportunities are closed down (for instance stepped up enforcement might lead to less incidence, but a greater size of bribes).48 More constructively, experience over recent years has illustrated that more recognition is needed of the level of governance when embarking on anti-corruption initiatives and that it is governance that is key to making decisions about what anti-corruption activities should be embarked upon (Kaufmann, 2005; Shah and Thompson, 2004) ­ that different anti-corruption mechanisms will be applicable in different governance contexts (Shah and Schacter, 2004). It is clear that prescriptive anti-corruption efforts need to be treated with caution in countries where the rule of law is weak, the state has little legitimacy, institutions responsible for service delivery are not accountable, and commitment of national leaders is questionable. What is the message for future anti-corruption activity in the water sector? The cautionary tale for the water sector is to look before you leap (Shah and Schacter, 2004). There is little experience of direct anti-corruption reform in the sector, and while it is possible to learn lessons from other sectors, the specificity of country, sector and even local contexts cannot be overemphasized. The African WSS sector needs to approach the anti-corruption agenda cautiously, draw on lessons from the broader governance agenda, and expand on its knowledge about WSS sector institutions, and the means to demand high and low level demand for reform. 45 A first set of lessons on capacity building of consumer bodies to engage in WSS sector reform has been developed by WSP in collaboration with Consumers International. 46 Over the last 5 years, there has been a shift in focus from supply to demand side efforts. 47 See Shah and Schacter's article `Look before you leap' which warns against ad hocism in anti-corruption work and suggests the lack of progress in eradicating corruption could be due to misguided strategies. 48 In relation to public-private interactions, if sanctions can not be enforced, more robust corruption can result through tighter more sophisticated processes, or firms adhering to integrity mandates can be margin- alized within the market, providing strong disincentive for others to follow suit. 2 Tackling Corruption in the Water and Sanitation Sector in Africa: Starting the Dialogue Table 6. Next Steps Country And Regional Level Actions Diagnosing country The development, implementation and prioritization of anti-corruption policies and mechanisms needs to be based level problems and on sound analysis and understanding of the problem if anti-corruption efforts are to be justified and effectively operating frameworks targeted. Thorough diagnosis is needed at the country (or city/district) level to describe how corruption works at different levels (from high level grand corruption down to the corruption at the tap); in different interactions (public sector, public-private, public-consumer); in the different systems of water provision in African countries (public and private utilities, local operators, community-management); to measure the scope and extent of corruption, and identify the areas where corruption in concentrated. It is important we work out what matters most in any one context with particular emphasis on developing an understanding of the impacts on the poor. Developing anti-corruption measures in the water sector will require an understanding of the basic governance and anti-corruption systems in place at the country level (the pre-conditions for effective reform) and the willingness and capacity to put these basics in place (political commitment, reform/stabilization of the legal and regulatory framework, market regulation). Developing an understanding of the sector context (institutional, regulatory, policy environment), causes and incentive structures should form a central part of the diagnostic work. We need to understand how the general state of play affects sector level activity. What are the pre-conditions, sequences, combinations for reform? What is viable at the sector level? What are the best entry points? Measuring corruption It is vital we start to measure and monitor sector corruption and establish robust corruption measurement tools and benchmarking and indicators at the sector and sub-sector levels. A number of indexes are available and provide lessons and experience. These need to be carefully considered and tested. Do we measure the perception of corruption in the sector (as developed by TI and others)? Do we seek to measure where possible the actual corruption (Olkren, 2005)? Or do we measure anti-corruption policies and mechanisms (Global Integrity, 2006)? At the sector level the water reform ranking being developed by WSP-Africa and global benchmarking initiatives (IBNet) are key starting points for regional initiatives. We need to know the size of the problem in the African region, we need baselines against which to measure the impact of interventions and we need to track the effectiveness and cost of specific anti-corruption strategies. Building a platform Efforts are needed to sensitize stakeholders (public, private, civil society) to WSS corruption issues, to create of informed anti- a platform of anti-corruption advocates and to build the capacity of that platform. The current group of anti- corruption advocates corruption advocates is small given the scale of the problem, and the limited NGO capacity will hinder efforts to establish strong civil society engagement. Empowering and engaging local, national and regional organizations is necessary to stimulate, coordinate and optimize activity. Country-level anti-corruption working groups should be developed during the diagnostic stage. The focus should be on building alliances ­ at the country level in strategic working partnerships for diagnostics and anti-corruption activity on the ground; and at the regional level in the formation of a regional task group that promotes the agenda at the regional level and engages donors in harmonize efforts in developed and developing countries. Developing country With the benefit of diagnostics and country level working groups, efforts are needed to develop country level level strategies with strategies. Coordinated with MDG roadmaps and sector reforms, country anti-corruption strategies should government and civil harmonize sector and cross-cutting action, identify high level government and civil society leadership, and society mobilize the stakeholder groups needed to make sustainable change. Where possible they should be an explicit, additional and mainstreamed dimension of sector reform at all stages (diagnostic, action plan and monitoring). Recognizing the difficulties of implementation, strategies need to focus on selected parts of the corruption map. It is likely that these corruption hotspots (areas of concentrated corruption activity) will vary in different countries, sub-regions, institutional contexts ­ strategies need to build corruption reform scenarios from the causes and incentives, constraints and capabilities at the country level. table continued on next page... Launching advocacy, The corruption struggle needs to become an African initiative, spearheaded at the country level. Advocacy awareness building and awareness building is needed at the regional, national and local levels with high level officials to develop: (i) and capacity building committed leaders and (ii) informed communities. But efforts should learn from the mixed success of awareness drives building campaigns. The identification of appropriate CSO partners will be critical, and specific initiatives are needed to build CSOs with the tools, vision and knowledge to represent consumer voice effectively, and building on existing work, efforts are needed to strengthen the effectiveness of consumer organizations. Disseminating tools, The range of existing and emerging lessons and experience in addressing corruption need to be developed for methodologies, best sector use and disseminated to provide information on diagnostic tools and anti-corruption methodologies that practice and lessons can be adapted to country and sector context. In particular, lessons on transparency (e.g. the disclosure of learnt education budgets in Uganda), empowering citizens and developing voice (e.g. the Bangalore report cards) and developing accountability processes at the program level (e.g. Kecamaten Development Program) all provide successes the water sector can use. Linkages with the newly formed Water Integrity Network will provide dissemination opportunities at the regional level. Thematic Areas Of Investigation Understanding The effects of decentralization on corruption are the subject of much debate (Huther and Shah, 2000; Fismana and decentralization and Gatti, cf. Garing, 2006). Efforts are needed to understand better the processes at work in the water sector in the corruption in WSS various decentralized contexts of Africa. Decentralization is a potential driver for change, but incentive structures as well as the specific modalities of fiscal and functional responsibilities affect efforts to control the shift of corrupt practices to the local level. There is a short window of opportunity to establish new accountability and ensure perverse incentives are not embedded in new institutions. We know the basic requirements for establishing better governance in decentralized settings, but we need to understand how decentralization affects corruption within the WSS sector in African countries. We need to know how decentralization has affected the regulatory environment and how decentralization frameworks can mainstream anti-corruption objectives. We also need to understand better the balance between incapacity and corruption among less skilled local institutions and officials. Unbundling and The various economic, political and governance systems present a basket of variation in Africa that limits diagnosing corruption generalization. At the sector level however we need to unpack the problem of corruption to understand it in WSS delivery better ­ country level efforts will be informed by more sector knowledge. Typologies can be explored to better systems understand the generic structures and incentive systems for corruption in different types of delivery institutions and systems. This might include understanding better the areas of concentration and modalities of corruption in, inter alia, public utility supply systems, private concessions, donor/NGO-led projects and/or community- managed WSS projects, and identifying the commonalities and differences between rural and urban, national and local, water and sanitation. This analysis can then be used to develop guidelines for diagnostics and reform methodologies that can be adapted as needed to country contexts. Understanding the In order to plan pro-poor anti-corruption strategies for the water and sanitation sector, we need to understand impacts of corruption the impacts on the poor and how the poor perceive corrupt practices of WSS officials. Diagnosis needs to in WSS on the poor include participatory corruption assessments that reveal the specifics of corruption in the lives of different types of poor households, and this information and perception used to develop interventions that are pro-poor e.g. What types of corruption have the most negative impact on the poor? How much do they pay? What are the net effects on the poor? Are there any benefits in unofficial rather than official supply? How can the detrimental impacts of both corruption and anti-corruption actions on the poor be mitigated?5 Diagnosis that focuses on the water market serving the poor ­ the mechanisms of demand and supply and the movement of money within this system ­ is key to establishing sensible anti-corruption approaches. A worst-case scenario of increased WSS anti-corruption activity would be that the initial target of anti-corruption activity becomes the poor. While it is true that the easiest set of actions will involve the least powerful stakeholders, it is vital that the quick win is not at the expense of the poor obtaining water. Further consideration is needed in this area. table continued on next page... 4 Tackling Corruption in the Water and Sanitation Sector in Africa: Starting the Dialogue Enhancing Best practice suggests that transparency helps reduce corruption, it increases the likelihood of exposure and transparency in the reduces the discretion of public officials. Transparency related reforms however, remain checkered around the water sector globe and notably low in Africa (Kaufmann, 2005). The WSS sector in Africa has some experience in developing transparency mechanisms in areas such as tendering and procurement, and the publication of utility accounts. But in practice, the sector is relatively immature in this area and the scope of activity has been limited to isolated cases that have not been focused on corruption. Better understanding is needed as to how the basic ingredients of transparency and access to information (for ant-corruption) can be adopted and mainstreamed into WSS sector reform processes: disclosure of assets, fiscal and public financial transparency of national and local budgets, sector expenditure tracking, access to information on WSS on local/agency/village revenues and expenditures, public meetings and budgetary disclosure the potential role of the media, transparency indexing, e-governance, social monitoring, all contribute to more open and accountable service delivery. How can current reform approaches be strengthened with a significant transparency focus? What are the dynamics and constraints of increasing sector transparency in differing contexts? In what situations would transparency in the sector harden corruption? The sector needs to learn how to identify and prioritize interventions: At the country level it is important to first understand capacity and governance, and to understand incentives, the impact of corruption and anti-corruption mechanisms (prescriptive, preventative), the effectiveness of differing processes (shock or gradual change), the combinations of anti-corruption mechanisms needed (e.g. matching transparency reforms with increased citizen roles) and the sequencing of interventions that will work best. The sector needs to test, build and disseminate experience: Working within this broader sphere of governance, accountability and transparency efforts to get sector governance right will be critical in any context. Strategy development should, in the context of better diagnosis should look at prioritization of: (i) improving demand, improving voice and participation, (ii) improving transparency and access to knowledge and information, and (iii) building on efforts in relation to reform water institutions and institutional frameworks. Understanding how these areas of activity can be more effectively focused on anti-corruption ends seems a critical step for the sector. 4. MOVING FORWARD A number of issues should be kept in mind as the water and sanitation sector embarks on anti-corruption reform: There is a notable lack of information on the scope, nature, impact and costs of corruption in the WSS sector. For decades, donors and other supporting agencies have carried on regardless of endemic corruption distorting decision-making and leaking sector investment. This tolerance has meant that there have been few attempts to define, unpack or delve into the key dimensions of sector corruption. While there are islands of information, most stakeholders acknowledge that this lack of comprehensive information on the scope and nature of corruption hinders future action. Furthermore we know little about impacts and costs. There is only limited understanding of the relative levels of sub-sector corruption (water supply, sanitation, cf. WRM), and no accurate measure of the relative levels of WSS sector corruption between countries in the region. Data is urgently needed to support proposed actions. 5 Corruption in the water sector is a part of the broader governance problem. Corruption in the water sector is interlinked with overall and sector governance: few would argue that the main challenges for improving water sector governance and tackling sector corruption do not lie in the water sector at all (Estache and Kouassi, 2002), or that corruption is embedded in the problems of a dysfunctioning sector. Yet while many dimensions of governance ­ the rule of law and political stability ­ lie outside the remit of the sector, many solutions ­ strong leadership and the emergence of social groups demanding change ­ are common to all. A key to effective sector anti-corruption activity will also lie in an understanding of the primary `interaction space' in which corruption takes place, in particular, how far corrupt water interactions extend beyond water institutions and stakeholders.49 Developing better understanding of the scope and content of viable sector action, in particular what is possible with ring-fenced service providers will be important to sustainable interventions. Decentralization has created a new set of risks and opportunities, and more effort is needed to develop accountability at the outset. Decentralization provides a window of opportunity for the development of transparency and accountability in sub-regional governments. To date the process of decentralization have produced mixed results both in terms of corruption and service delivery,50 there is the lack of predictability and understanding of the impacts of decentralization on WSS or on the corruption levels in WSS. This can be attributed at least in part to the heterogeneity of the reforms in the region. Decentralization in Africa is a mix of political, fiscal, administrative delegation carried out by massively different processes (big bang, gradual, top- down, bottom-up, in different time frames) ­ the picture is one of diversity (Shah and Thompson, 2004). In practice too, local, often poorly skilled sector officials, previously without access to decision-making or budgets, have seized the opportunity for rent-seeking created by delegated fiscal and functional management. The lack of mandates clarifying functional allocations and relationships creates. A lack of transparency and failures in accountability. Better understanding of the linkages between corruption and decentralization processes is needed at the sector level. In particular, how decentralization has affected sector corruption, which areas are most prone, and how the momentum presented by decentralization can be harnessed to hinder corruption reemerging. It takes two or more to bribe ­ the focus of anti-corruption efforts needs to include the briber in WSS transactions. With a significant profit motive, many private WSS sector stakeholders (be they international, national or local) have strong incentive to ensure their companies are included in tenders, win contracts, avoid unnecessary delays in construction and find ways to cut corners to higher profit margins. Typically, staff incentive systems to bring back profit to shareholders or owners encourage corrupt individual behavior. But this is not always the case ­ not all companies and not all individuals join in. Understanding incorruptible behavior might provide pointers for actions to reduce private sector bribing (Sohail, 2005). More detailed knowledge is also needed about the role of middlemen that facilitate payments. Anti-corruption efforts have generally focused on `cleaning-up' offending governments, but these efforts to promote accountability in government are undermined if pressures on staff are exogenous. It is vital to develop anti-corruption mechanisms that reduce bribers soliciting and paying bribes in water sector transactions. 49 There are very mixed views on the level to which anti-corruption activity can be undertaken in a ring-fenced manner within the WSS sector, and there is an urgent need for testing models and strategies. There are those that argue that efforts can proceed at the local (delivery agency) level, while others argue that it is likely that anti-corruption efforts may be futile, counterproductive or at the margins, if sector efforts are not coordinated with the broader governance agenda. The Kenyan experience too is sobering. Reforming one branch of government and not another created serious problems in Kenya, deepening levels of corruption. (Nussbaum, 2006). 50 The different views on the impact of decentralization can be seen for instance in World Bank 200, and Fisman and Gatti, no date. Garing for instance argues that no country has ever solved its corruption problems through decentralization, highlighting the fact that in the UK in the 1800's corruption was solved by recentralizing (Garing, 2006). This is sobering evidence for a sector where decentralization is the hope for improved accountability of service delivery. 6 Tackling Corruption in the Water and Sanitation Sector in Africa: Starting the Dialogue The political realities of fighting corruption in the water sector are sobering: many constraints and opposing stakeholders block the way forward. Despite increasing political rhetoric and high profile commissions and investigations, in many countries there is still a reticence to discuss corruption and anti-corruption activity at the sector level.51 There are only a few stakeholders interested in analyzing corruption, sharing their knowledge or proposing anti-corruption tools and techniques difficult, and advocacy is hindered by the absence of a strong CSO sector in the African region.52 Outside government there is a strong set of incentives and disincentives affecting stakeholder willingness to engage in anti-corruption activity: large construction and engineering companies are concerned with shareholder profit, consultants with their client base, and individuals fear reprisals. Bilateral and multilateral agencies need to meet spending and lending targets and maintain the status quo, but they also fear anti-corruption activity will sour relationships with government partners. At the sector level, in WSS organizations, the advocacy base is also narrow ­ not all those who we work with on a day-to-day basis are convinced that corruption is something we should confront. Acknowledging that corruption is a problem, developing a broader platform of advocates, building awareness and creating a safe space for dialogue will all be critical to effective sector action. The net effects of corruption and anti-corruption activity in WSS for the poor are not really known. The structural impacts of corruption on the poor summarized by Kaufmann and others,5 are now well disseminated ­ lower investment and growth, less pro-poor growth, less progress in service delivery and the development of human assets. To this can be added the many non-economic consequences of corruption ­ weakening of new and emerging democracies, social injustices, environmental degradation, heightening of insecurity, and undermining trust in public institutions. But at the sector level more detailed analysis is needed on the short, medium and long-term impacts on the poor. Medium to long-term impacts on the poor result from distortions and delays created by corrupt officials steering investment in WSS away from the poor toward opportunities that are likely to bring them the most personal gain. Typically this means inappropriate investment in large capital- intensive infrastructure projects such as water treatment plants, bulk supply or networks. In the short term however corrupt activities by low-level officials selling or providing illicit utility water might fill gaps in the services available to the poor households, providing them with a service they would not otherwise obtain. In some cases, moonlighting providing private service with public assets (Collignon, 2005) may not necessarily be inefficient when considered at the sector level and in the short term.54 Developing pro-poor anti-corruption activity within the WSS sector should be informed by more widespread and detailed demand-side assessment. Much of the anti-corruption activity launched to date has proceeded on the assumption that any anti-corruption intervention will automatically improve efficiency and effectiveness and thus create benefits for the poor. But this is not always the case, and while `corrupt' water is clearly not optimal in the long run, focusing reform on low level officials misusing assets may have negative immediate impact on those who need water services most. This level of anti-corruption activity will require complimentary demand-side assessments and actions that are vital to reveal and mitigate against any likely impacts of anti-corruption 51 In other countries (such as Indonesia) the situation is more nuanced, there may be dialogue about corrupt practices, a group of anti-corruption champions, and public willingness to speak openly about corrup- tion, but there might be reticence to speak about installing anti-corruption mechanisms (that may affect friends/colleagues/family income). 52 Best practice in anti-corruption activity reveals the importance of civil society organisations taking a lead role in demanding change. This is more difficult in the African region where the CSO movement is weak. The position of CSOs is especially weak in the government-dominated water sector in Africa. Even where the state can not deliver there is the public perception that WSS provision is a state, not a civil society responsibility. CSO WSS support in Africa generally consist of scattered, specific projects. This may develop momentum in stimulating public demonstration, as in Ghana in the movement against water privatiza- tion, but they do not play a central role in large scale service provision (as, for example in Bangladesh). Efforts will be needed to work with, and develop the capacity of non-water CSOs such as the country chapters of Transparency International. 5 See the Synthesis Matrix on Poverty and Governance in Kaufmann, 2000. 54 See overview of the types of water available to the poor including `corrupt' water (cf. informal, private, network or self supply) water in Plummer and Cross, 2005. 7 mechanisms. To ensure reforms are pro-poor, it is essential that anti-corruption advocates understand the interactions between long-term structural changes and short term contingent ones, and plan for both positive and negative impacts on water supply to the poor. With this recognition, legitimization rather than eradication or supplementary instruments that provide the poor with water in the short term might, for instance, become an important pro-poor aspect of tackling corruption at the point of service delivery. The following table provides a set of country, regional and thematic actions for anti-corruption stakeholders able and willing to take forward the agenda. Donor engagement is essential for funding, harmonization and creating a stronger motive for reform. 5. CONCLUDING REMARKS Although the challenges in tackling corruption in the WSS sector in Africa are significant, a number of opportunities have emerged in recent years that have heightened the issue of corruption on the WSS agenda. First, a growing number of governments are indicating their willingness to discuss corruption openly and engage in programs and anti-corruption discussion. Second, there is strengthened commitment in the World Bank, the African Development Bank, sector donors and other sector agencies to address corruption in donor-funded initiatives and in their own systems. Third, there is growing commitment to government and donor coherence on the anti-corruption agenda.55 A key factor to recognize in taking forward an anti-corruption agenda in the water and sanitation sector is that most of what is being done now in the water governance agenda ­ policy, institutional, financial management reform, reducing inefficiencies, improving leadership and demand-side capacity building ­ is central to anti-corruption activity, but unfortunately has not shifted (to any degree) corrupt practices in the sector. It is vital the sector come to understand what adjustment is needed to recharge these efforts and focus them more effectively on tackling corruption. It is critical in this regard to improve understanding of corruption in the sector, and to develop linkages between the sector and broader governance constraints and initiatives. This chapter has described, through a framework of corrupt interactions between public, private and consumer/civil society actors, the types of corruption that occur in the various stages of sector policy, planning, budgeting , delivery and implementation. It has argued that corruption is neither singular nor homogenous in any one setting, that understanding the network of corrupt activity, and identifying the areas of concentration of corruption within this larger framework is critical to effective policy making and strategy development. Overlaid on this matrix of corrupt interactions, the paper has set out the many and varied anti-corruption mechanisms, including those that are generic and create an environment that deters or mitigates against the risk of corruption, and those that target specific types of anti-corruption activity. This menu of actions should be taken in with caution. A key concern is that we do not yet know conclusively what works in which situations, what combinations of interventions are needed, and what sequencing of reforms and anti-corruption activity will optimize efforts in tackling the various forms of corruption. Most of all, the water and sanitation sector in Africa urgently needs to enter into a phase of diagnosis and testing that provides the empirical basis for action. 55Paris Declaration on Aid Effectiveness, Paris, March 2005 8 Tackling Corruption in the Water and Sanitation Sector in Africa: Starting the Dialogue REFERENCES Gray, C and D. Kaufmann, (1998) Corruption and Development in Finance and Development, March 1998 Austral Consultoria a Projectos, Governance and Anti- Fisman, R and R. Gatti, [date] Decentralization and Corruption: corruption Diagnostic Survey: Mozambique, Draft for Evidence Across Countries? Columbia Business School World Bank Institute, Washington D.C. [no date] and Development Research Group, The World Bank Bellver, A. and D. Kaufmann (2005) Transparenting Garing, J. (2006) Discussion in Governance and Growth in Transparency: Initial Empirics and Policy Applications, Africa, Presentation and Discussion, PREM Week, 2006 Preliminary Draft, World Bank, Washington, D.C. Guggenheim, S. (2006) Discussion regarding the Center for Democracy and Development (2000) The Kecamaten Development Progam in Indonesia, Addis Ghana Governance and Corruption Survey Evidence Ababa, January 2006 from Households, Enterprises and Public Officials Commissioned by the World Bank, August 2000 Halpern, J. and Y. Mumssen (2006) Lessons Learned in Infrastructure Services Provision: Reaching the Poor, OBA Collier, P. and A. Hoeffler (2005) `The Economic Costs of Working Paper Series, Paper No. 6, April 2006. Corruption in Infrastructure' in Transparency International, 2005, Global Corruption Report. Heller, N. (2006) Addressing Governance and Corruption in Latin America: Lessons from Field Research, Presentation Collignon, B. (2005) Draft Discussion on Corruption in Water World Bank, PREM Learning Week, May 2006. Services in Africa, Nairobi, June 2005 Hellman, J, G. Jones and D. Kaufmann (2001). Seize the Darroch, F (2004) Lesotho Highlands Water Project- State, Seize the Day: State Capture, Corruption and Corruption and Debarment Prepared for TIRI Influence in Transition. World Bank Policy Research Working Paper 2444, Washington. D.C. Davis, J. (200) Corruption in public service delivery: Experience from South Asia's water and sanitation sector. Huther, J. and A Shah (2000) Anticorruption Policies and MIT, Cambridge Programs: A Framework for Evaluation Dec 2000, OED, World Bank Doyen, J.(2006) Informal comments on feedback on tackling corruption in WSS in Africa, April 2006 JMP, 2002, Meeting the MDG Drinking Water and Sanitation Targets: A Mid Term Assessment of Progress. WHO/UNICEF Earle, A. and A. Turton, (2005) `No Duck No Dinner: How Sole Sourcing triggered Lesotho's Struggle against Corruption, Kaufmann, D. (1998) Challenges in the Next Stage of Anti- Paper presented at World Water Week 2005 in Stockholm, corruption. In D. Kaufmann, (Ed.), New Perspectives seminar on `Corruption in the Water Sector: How to fight on Combating Corruption. Washington, DC: Economic it?' African Water Issues Research Unit (AWIRU), CiPS, Development Institute. University of Pretoria Kaufmann, D. (2000) Governance and Poverty: A Brief Note Estache, A, and E. Kouassi, (2002) Sector Organization, on the Power of Empirics, Quality of Growth, Chapter 6 Governance, and the Inefficiency of African Water Governance and Corruption Draft August 2000, World Bank Utilities, World Bank, Policy Research Working Paper September 2002 Kaufmann, D. and A. Kraay (2002) Growth Without Governance. Economia. Fall 2002. EITI (2006) The Extractive Industries Transparency Initiative website. see www.eitransparency.org/ Kaufmann, D. (2005) Back to Basics ­ 10 Myths About Governance and Corruption , Finance and Development, Sept, 2005, Volume 42, Number . 9 Kaufmann, D., A. Kraay and P. Zoido-Lobatón (2005) Olkren, B. (2004) Monitoring Corruption in CDD: Main Findings Governance Matters IV ­Updated Indicators for 2000/01. and Implications, KDP Note, World Bank, Indonesia. World Bank Institute, Washington, D.C. Olkren, B. (2005) Corruption Perception Versus Corruption Kaufmann, D. and A. Kraay (200) Governance and Growth, Reality, Harvard and NBER Causality which ay? Evidence for the World in Brief. World Bank Institute Plummer, J. (200) Small scale providers in water and sanitation Background Paper for WDR 2004, World Bank, Kaufmann, D. G. Mehrez and T. Gurgur (2002) Voice or Washington D.C. Processed. Public Sector Management? An Empirical Investigation of Determinants of Public Sector Performance based on a Plummer, J. (2005) Anti-corruption efforts in the Post-tsunami Survey of Public Officials World Bank, IMF, and World Bank Reconstruction of Water and Sanitation Infrastructure and Services in Aceh, Indonesia, Water and Sanitation Program, Kaufmann, D. and P.C. Vicente (2005) Legal Corruption, World Jakarta, Indonesia. Bank Institute. Plummer, J and P. Cross (2005) Combating Corruption: Kaufmann, D. , Montoriol-Garriga and F. Recanatini, How Does Developing an Anti-corruption Strategy for the Water and Bribery affect Public Service Delivery? Micro-evidence form Sanitation Sector in Africa, Presentation for the Preparatory service users and public officials in Peru, World Bank Institute. meeting of the Water Integrity Network (WIN), Delft November 2005, Water and Sanitation Program­Africa Kenny, C. (2006) Corruption in Infrastructure, mimieo, World Bank Plummer, J., B. Collignon and S. Mehrotra (2005) Supporting Kirkpatrick, C, Parker, D. and Y.F Zhang, State versus private the market that serves the urban poor: Emerging sector provision of water services in Africa: A statistical, responses to enhance the role of local private sector dea and stochastic cost frontier analysis, University of providers. Background paper for the Domestic Private Manchester, Paper No. 70, June 2004 Sector Participation Initiative Workshop held in Nairobi, 20- 21 June 2005 Klitgaard, R. (1988). Controlling Corruption. Berkeley, CA: University of California Press. Recanatini, F. , A. Prati, and G. Tabellini (2005) Why are some public agencies less corrupt than others? Lessons from Klitgaard, R., Maclean-Abaroa, R., & Parris, H.L. (2000). institutional reform from survey data. IMF, November 2005 Corrupt Cities: A Practical Guide to Cure and Prevention. Oakland, CA: Institute for Contemporary Studies Press. Rose-Ackerman, S. (1999) Corruption and government: Causes, consequences, and reform. New York: Cambridge Locussol, A. and Fook Chuan Eng (2006) Informal comments on University Press. feedback on tackling corruption in WSS in Africa, April 2006 Shah, A. and M. Schacter (2004) Combating Corruption: Look Mehta, M., T. Fugelsnes and K. Virjee (2005) Financing the Before You Leap Finance & Development; Dec 2004 Millennium Development Goals for Water and Sanitation: What will it take? International Journal of Water Resources Shah, A. and Thompson (2004) Implementing Decentralized Development. 21:2:29-252. Local Governance Policy Research Working Paper Series 5, World Bank, Washington D.C. Nussbaum, D. (2006) Building Accountability for Shared Growth, Presentation World Bank, PREM Week, April 2006. Sohail Khan (2005) Comments in Preparatory Meeting for Water Integrity Network, Delft November 24, 2005 OECD (1999) Public sector corruption: An international survey of prevention measures. OECD, Paris. Stapenhurst, R. (2000) The Media's Role in Curbing Corruption, World Bank Institute, Washington D.C. 40 Tackling Corruption in the Water and Sanitation Sector in Africa: Starting the Dialogue Sutton, S. Preliminary Desk Study of Potential for Self Supply Woodhouse, A. (2002) Village corruption in Indonesia, Fighting in Sub-Saharan Africa, WaterAid and the Rural Water corruption in the World Bank's Kecamatan Development Supply Network, October 2004, Program, World Bank, Indonesia. Tendler, J. (1997) Good Government in the Tropics. Baltimore: World Bank (1997) Helping Countries Combat Corruption: The Johns Hopkins University Press. Role of the World Bank. Washington, DC: The World Bank. Thampi, GK (2005) Community Voice as an Aid to World Bank (2004) Making Services Work for the Poor, World Accountability. Experiences with Citizen Report Cards in Development Report 2004, World Bank, Washington D.C. Bangalore. Bangalore: Public Affairs Foundation. World Bank (2005) "Public Sector Governance: The Grand Timmer, p. (2006) Paying Suharto as if he were an American Challenge?" in Economic Growth in the 1990s: Learning CEO. Center for Global Development, processed. From a Decade of Reform, PREM, Washington, D.C. Transparency International (2005) Global Corruption Report, World Bank (2006) An anticorruption Agenda for the Power Pluto Press, London Sector: A suggested Bank Approach Discussion Draft, January 26, 2006, World Bank, Washington D.C. Transparency International (2006) Global Corruption Report, Pluto Press, London World Economic Forum (2006) www.weforum.org/site/ homepublic.nsf/Content/Partnering+Against+Corruption Transparency International (200) Business Principles for Countering Bribery, TI, Berlin WSP-Africa (forthcoming) Country Status Overviews of Africa's Progress Towards the WSS MDGs, WSP-Africa, Nairobi. Transparency International (2005) Integrity Pact: A Pakistan Success Story Greater Karachi Water Supply Scheme, TI WSP-Africa (2004a) Moving from Protest to Proposal: Building Pakistan, Karachi the Capacity of Consumer Organizations to Engage in Urban Water Supply and Sanitation Sector Reform in Transparency International (2006) Nairobi Water and Sewerage Africa, WSP-Africa, Nairobi. Company Ltd. A Survey. April-May 2005, TI Kenya Nairobi WSP-Africa (2004b) Self Supply: A Fresh Approach to Water Trémolet, S. and J. Halpern (2006) Regulation of water and for Rural Populations, Field note, WSP-Africa, Nairobi. sanitation services: Getting better service to poor people, OBA Working Paper Series, Paper No. 8, June 2006 WSP-Africa (2004c) Solutions for Reducing Borehole Costs in Rural Africa, Field note, WSP-Africa, Nairobi. UN-Habitat and TI 2004, Tools to Support Transparency in Local Governance, Urban Governance Toolkit Series, UN Nairobi. WSP-Africa (2005a) Proposal for the Development of Report Cards, WSP-Africa, Nairobi. UNESC (200) Perspective on Governance in Central Africa. Accra Ghana, December 200 WSP-Africa (2005b) Rogues No More? Water Kiosk Operators Achieve Credibility in Kibera, Field note, WSP-Africa, Nairobi. Wei, Shang-Jin (1999) Corruption in Economic Development: Beneficial Grease, Minor Annoyance, or Major Obstacle? WUP (2001) Status of Reforms in the Water and Sanitation World Bank Policy Research Working Paper 2048. World Sector in Africa, Cote d'Ivoire. Bank, Washington D.C. WUP (200) Better Water and Sanitation for the Urban Poor in Winter, C, and L. Rodriguez, (2006) Municipal Service Delivery Africa www.wupafrica.org in Maputo City, Mozambique, mimieo, World Bank, Washington D.C. 41