37529 WATE R SU PPLY AND SANI TATION SE CTOR BOAR D DI SCU S SI ON PAPE R SE RIE S P A P E R N O . 6 J U N E 2 0 0 6 Explanatory Notes on Key Topics in the Regulation of Water and Sanitation Services Eric Groom, Jonathan Halpern, and David Ehrhardt THE WORLD BANK GROUP Bank Netherlands Water Partnership ACKNOWLEDGEMENTS DISCLAIMERS This report, which is a product of the Energy and Water The findings, interpretations, and conclusions expressed in Department (EWD) and the Infrastructure Economics and this paper are entirely those of the authors, and should not Finance Department (IEF) of the World Bank, was funded be attributed in any manner to the World Bank, to its affiliated by the Bank-Netherlands Water Partnership (BNWP), a organizations, or to members of its Board of Executive facility that enhances World Bank operations to increase Directors or the countries they represent. The material in this delivery of water supply and sanitation services to the poor work is copyrighted. No part of this work may be reproduced (for more information, see http://www.worldbank.org/wat- or transmitted in any form or by any means, electronic or san/bnwp), and by the Public-Private Infrastructure Advisory mechanical, including photocopying, recording, or inclusion Facility (PPIAF), a multi-donor technical assistance facility in any information storage and retrieval system, without the aimed at helping developing countries improve the quality prior written permission of the World Bank. The World Bank of their infrastructure through private sector involvement encourages dissemination of its work and will normally grant (for more information, see http://www.ppiaf.org). permission promptly. For permission to photocopy or reprint, please send a request with complete information to the The authors would like to acknowledge the valuable inputs Copyright Clearance Center, Inc, 222 Rosewood Drive, of their colleagues Mr. Doug Andrew, Mr. Aldo Baietti, Danvers, MA 01923, USA, fax 978-750-4470. All other Ms. Georgina Dellacha, Mr. Jeff Delmon, Mr. Antonio queries on rights and licenses, including subsidiary rights, Estache, Mr. Jan Janssens, Mr. Bill Kingdom, Mr. Alain should be addressed to the Office of the Publisher, World Locussol, Mr. Gustavo Saltiel, Mr. Chris Shugart, Mr. Bernie Bank, 1818 H Street N.W., Washington DC, 20433, fax Tenenbaum, Ms. Meike van Ginneken, and Mr. Richard 202-522-2422, e-mail: pubrights@worldbank.org. Verspyck. CONTACT INFORMATION To order additional copies, please contact the Water Help Desk at whelpdesk@worldbank.org. This paper is available online at http://www.worldbank.org/watsan. Eric Groom, Senior Regulatory Specialist, World Bank: egroom@worldbank.org, tel. 202-458-8558 Jonathan Halpern, Adviser, World Bank: jhalpern@worldbank.org, tel. 202-458-4442 David Ehrhardt, Chief Executive, Castalia Limited: David.Ehrhardt@Castalia.fr, tel. +64 4 913 2800 WATER SUPPLY AND SANITATION SECTOR BOARD DISCUSSION PAPER SERIES P A P E R N O . 6 J U N E 2 0 0 6 Explanatory Notes on Key Topics in the Regulation of Water and Sanitation Services Eric Groom, Jonathan Halpern, and David Ehrhardt THE WORLD BANK GROUP Bank Netherlands Water Partnership Copyright © 2006 The International Bank for Reconstruction and Development/The World Bank. All rights reserved INTRODUCTION .......................................................................................................1 NOTE 1 -- DEFINING ECONOMIC REGULATION FOR WATER SUPPLY SERVICES........5 Overview.................................................................................................................5 Defining Economic Regulation in the Water Sector.....................................................5 Economic Regulation Addresses Monopoly Power ......................................................6 Economic Regulation versus Regulation Generally .....................................................8 Economic Regulation versus Other Interventions ........................................................9 Summary...............................................................................................................12 Further Reading .....................................................................................................12 NOTE 2 -- DESIGNING ECONOMIC REGULATION FOR WATER SUPPLY SERVICES: A FRAMEWORK .................................................................................15 Overview...............................................................................................................15 Steps in Designing Economic Regulation .................................................................15 Is Regulation Part of the Solution?...........................................................................18 Defining Regulatory Functions.................................................................................18 Allocating Regulatory Functions to Organizations and Instruments............................20 Further Reading .....................................................................................................21 NOTE 3 -- CHOOSING ORGANIZATIONS AND INSTRUMENTS FOR ECONOMIC REGULATION OF WATER SUPPLY SERVICES ....................................25 Overview...............................................................................................................25 Organizational Architecture....................................................................................26 Regulatory Instruments ...........................................................................................28 Legal and Organizational Design for Regulation .....................................................30 Further Reading .....................................................................................................32 NOTE 4 -- REGULATION AND PRIVATE PARTICIPATION CONTRACTS .......................33 Overview...............................................................................................................33 Two Different Traditions ..........................................................................................34 Regulation with Various Private Participation Contracts.............................................35 Concession Contracts.............................................................................................36 Management Contracts..........................................................................................38 Lease-Affermage Contracts ....................................................................................41 Approaches to Combining the Traditions.................................................................43 NOTE 5 -- COST OF SERVICE AND TARIFFS FOR WATER UTILITIES...........................47 Overview...............................................................................................................47 Estimating the Cost of Service.................................................................................48 Operating Expenses ...............................................................................................48 Capital Costs.........................................................................................................50 Setting the Maximum Allowed Revenue ...................................................................54 Cost of Service and Social Impact...........................................................................54 What to Do When Efficient Costs Are Below Actual Costs.........................................55 Conclusions on the Maximum Allowed Revenue ......................................................56 Setting the Tariff Structure.......................................................................................57 iii The Tariff Control Regime.......................................................................................59 Conclusion ............................................................................................................61 Further Reading .....................................................................................................62 NOTE 6 -- REGULATING GOVERNMENT-OWNED WATER UTILITIES .........................63 Overview...............................................................................................................63 Why Regulate What You Own? ...............................................................................63 Governments Adding Regulation to Ownership .......................................................65 Regulation for Government-Owned Utilities.............................................................67 Conclusion ............................................................................................................70 Further Reading .....................................................................................................71 NOTE 7 -- REGULATING WASTEWATER SERVICES IN DEVELOPING COUNTRIES.......73 Overview...............................................................................................................73 Sanitation in a Developing City...............................................................................73 Policy and Environmental Regulation of Decentralized Systems.................................74 Regulation of Centralized Systems...........................................................................76 Summary...............................................................................................................80 Further Reading .....................................................................................................80 TABLES Table 1.1: Economic Regulation and Other Policies and Instruments ..............................10 Table 2.1: Is Regulation Part of the Solution? ................................................................19 Table 2.2: Allocating Regulatory Functions ....................................................................22 Table 3.1: Organizational Architectures in Three Regulatory Jurisdictions .......................27 Table 3.2: Legal Instruments for Regulation in Three Jurisdictions...................................29 Table 5.1 Approaches to Calculating Capital Costs .......................................................52 FIGURES Figure 1.1: Defining Water Sector Economic Regulation ..................................................9 Figure 2.1: Framework for Thinking about Regulation ...................................................17 Figure 4.1: Concession Contract ..................................................................................36 Figure 4.2: Management Contract................................................................................39 Figure 4.3: Lease Contract...........................................................................................41 Figure 5.1: The Building-Block Approach to Tariff Setting...............................................49 BOXES Box 5.1: Benchmarking Labor Costs -- Guyana Water..................................................50 Box 5.2: Tariffs Compared with Operations and Maintenance Costs ..............................51 iv INTRODUCTION NOTE 1 -- DEFINING ECONOMIC REGULATION FOR WATER SUPPLY SERVICES NOTE 2 -- DESIGNING ECONOMIC REGULATION FOR WATER SUPPLY SERVICES: A FRAMEWORK NOTE 3 -- CHOOSING ORGANIZATIONS AND INSTRUMENTS FOR ECONOMIC REGULATION OF WATER SUPPLY SERVICES NOTE 4 -- REGULATION AND PRIVATE PARTICIPATION CONTRACTS NOTE 5 -- COST OF SERVICE AND TARIFFS FOR WATER UTILITIES NOTE 6 -- REGULATING GOVERNMENT-OWNED WATER UTILITIES NOTE 7 -- REGULATING WASTEWATER SERVICES IN DEVELOPING COUNTRIES INTRODUCTION INTRODUCTION Considerable confusion has arisen about what regulation means in the context of water supply and sanitation (WSS) services. In particular, there are questions about the application of the "independent regulator" model to WSS in the developing world. What types of problems can it can address effectively? What is its relevance, especially as provision and oversight of these services are often the responsibility of subnational governments with limited resources? The Explanatory Notes on Key Topics in the Regulation of Water and Sanitation Services provide a consistent set of principles and practices that respond to these questions. Such information will be of interest to service providers, policy makers, and development practitioners interested in improving the perform- ance of WSS services in urban areas. The notes draw upon current regulatory thinking and research, but are intended to be accessible to those who are not regulatory experts. These are the first outputs of a program of work on regulation in the water supply and sanitation sector funded by the Public-Private Infrastructure Advisory Facility (PPIAF), the Bank-Netherlands Water Partnership (BNWP), and the World Bank. We will add additional notes as that work progresses. Each of the notes can be read separately, and together the notes provide an integrated framework for the development of practical approaches to the regulation of WSS. The seven notes address the following topics: 1. Defining Economic Regulation for Water Supply Services 2. Designing Economic Regulation for Water Supply Services: A Framework 3. Choosing Organizations and Instruments for Economic Regulation of Water Supply Services 4. Regulation and Private Participation Contracts 5. Cost of Service and Tariffs for Water Utilities 6. Regulating Government-Owned Water Utilities 7. Regulating Wastewater Services in Developing Countries Explanatory Notes 1­3 provide an integrated view of regulatory functions and the principles and practice underlying the design of regulatory systems in the WSS sector. These notes stress the following: · Economic regulation is the set of rules and organizations that set, monitor, enforce, and change allowed tariffs and service standards for water providers. This looks beyond whether there is a regulatory body or what that body may do. Economic regulation can function well for extended peri- ods without a "regulator." In such cases, the regulatory mechanism may be a contract with a privately owned service provider, a process for decision making by a department or minister, or a performance contract/license with a publicly owned service provider. Furthermore, economic regulators have often been asked to do far more than economic regulation (for example, 1 INTRODUCTION resolving customer disputes or fulfilling policy roles, such as formulation of financing and subsidy policies). · Economic regulation should be clearly defined. While there is overlap with other functions (for example, consumer dispute resolution and social policy), the domain of economic regulation should be kept narrow, clearly specified, and distinguished from the policy and governance functions. · Often the challenge in a developing country context is to increase average prices that are "too low" and distorted because of political factors, rather than to constrain a monopolist from charging prices that exceed the cost of services. This is the reverse of many of the textbook models of regulation. The role and efficacy of regulation in these circumstances need careful con- sideration. · Designing effective regulation starts with an identification of the WSS objec- tives and a careful consideration of both the extent to which regulation can facilitate achievement of these goals and its attendant costs. This evaluation should consider the full range of regulatory and policy instruments and is a task for policy makers (rather than for regulatory bodies). · WSS services typically require economic and technical regulation, but it is not necessary that all regulatory functions be undertaken by a stand-alone regulatory body. Legal rules and instruments can be used to set key regula- tory parameters (such as the initial price path and key elements in its subse- quent resetting). This applies particularly to privately owned WSS utilities. Assignment of the functions should consider the country's social, political, and legal traditions; the capability of existing agencies; and potential impacts on sector reform programs. Different jurisdictions can use quite dif- ferent organizational structures to perform similar functions. There is no sin- gle "best practice" model for the allocation of functions to agencies or instruments: ­ Where in-country capacity is scarce, there may be opportunities to use existing organizations, international panels of experts, or regional bodies. ­ Where legal and governance traditions are supportive, contracts can be an effective regulatory mechanism. However, care should be taken to avoid inconsistencies if contracts are combined with the creation of regu- latory agencies. Explanatory Note 4 explores regulation of services provided by privately operated utilities. There are two distinct traditions: one that relies on courts or arbitrators to fulfill the regulatory functions when the parties cannot agree; and another that relies upon government-established regulatory agencies. In the former, services are typically provided under contract to the government, which retains ownership of the assets, whereas the latter approach arose in the con- text of investor-owned utilities. This note outlines the range of contractual options available for delivery of WSS by privately owned utilities and highlights the problems that have occurred where there are both contracts and an inde- pendent regulatory agency. 2 INTRODUCTION Explanatory Note 5 provides a brief analysis of consistent approaches to resetting tariffs for WSS services. Most successful approaches have used a cost building-block approach that sets average prices or revenues on the basis of forecasts of reasonable costs by broad categories (operational expenditures, depreciation or renewals expenditures, and return on assets). Because it is for- ward looking, it still provides incentives for the utility to improve its efficiency, and because it is reset on the basis of utility specific costs, it provides some assurance that the utility will be able to recover reasonable costs incurred (including the cost of capital). The note explores the key issue of what to do where actual costs exceed the assessed efficient costs. The note emphasizes the need for caution. Explanatory Note 6 looks at the conundrum of regulating government-owned water utilities and poses the question: given that monopoly power can be con- trolled by the government (as owner), is there a need for separate regulation of government-owned utilities? However, if the governments can control their utilities for the public good, why are the outcomes often so poor (for example, selective meeting of customers' needs; use of service provision for short-term political aims or personal ends; capture of ministers by service providers)? In some countries, governments have established independent regulators for their water utilities. This can increase transparency, reinforce the positive incentives for utilities operating within a framework of good governance, and create more political space for tariff increases. But the history of these agencies shows the difficulty of the task. Further work has been commissioned on this issue, and the note highlights that a separate regulator is not a panacea. Whether a separate regulatory agency should be established and the extent of separation between the governance, policy, and regulatory functions will depend on the sector objectives, governance and incentive structures, and institutional and capacity constraints within the country. Explanatory Note 7 provides guidance on the role of regulation in improving wastewater services. Access to wastewater services often lags well behind access to water services. There are strong public health benefits from provid- ing wastewater services, but the provision of a centralized network can be pro- hibitively expensive. Improving wastewater services may be a matter of improving or extending existing small-scale systems: for example, septic tanks, latrines, and small-scale local systems. In this case, economic regulation may not be critical, but centralized environmental regulation could be necessary to ensure that health objectives are achieved. Centralized sanitation systems may be able to exert monopoly power; hence, some economic regulation may be necessary. Two key points to consider are the level and structure of the charges. Recovering full costs from charges on consumers may not be possible (because of the impact on bills) or desirable (because of the community health benefits). Hence, there may be a role for government subsidies, but these should ideally be based on measurable outputs. Volumetric sewerage charges (based on a percentage of metered water sales) have become more popular, but the note concludes that they are not necessarily efficient or equitable. 3 NOTE 1 -- DEFINING ECONOMIC REGULATION FOR WATER SUPPLY SERVICES NO Overview TE There is some confusion In the past decades, water sector reforms worldwide have focused attention on 1 over what regulation is regulation of the sector.1 But is it not always clear what is meant by "regula- and what it can do. tion" or which problems regulation is able to solve. Sterile debates have raged on topics such as whether regulation by contract is or is not "regulation." Some assert that regulation is not possible without a regulator and define reg- ulation as whatever the regulator does. Others use "regulation" to mean almost any form of government control of the water sector and assume it to be the answer to any water sector problem. This note aims to provide This is the first in a series of notes designed to bring greater clarity to economic clarity ... regulation of the water sector. This note's role is simply to define what economic regulation in the water sector is and what it is not. Clarity on this point means that later notes can address how to design economic regulatory regimes effec- tively, using well-understood concepts. ... in definitions. Economic regulation is best thought of as the legal controls on water providers intended to overcome the problem that water is an essential, monopoly serv- ice. A core definition of economic regulation is the rules and organizations that set, monitor, enforce, and change the allowed tariffs and service standards for water providers. Although other closely related functions (such as controlling asset condition) can usefully be considered part of economic regulation in some cases, some things (such as policy, ownership, governance,2 and coordination) are not reg- ulation. Such concepts are best kept distinct. Defining Economic Regulation in the Water Sector Regulation is not just One way to define economic regulation would be to survey what regulators "what regulators do." around the world do. However, this would be unhelpful because of the following: 1 The phrase "water sector" refers to the provision of water supply services and also the collection, treatment, and disposal of wastewater. 2 "Governance" refers to the relationship between the owners, directors, and managers and the rules, laws, policies, and customs that define this relationship and ensure that the managers and directors are account- able to the owners for the pursuit of objectives consistent with those of the owners and that the entity com- plies with all laws and regulations. 5 · It is precisely the absence of a ready consensus on what constitutes appro- priate regulation that motivated this note; hence, a descriptive approach would provide little guide to good practice. 1 · A descriptive survey would confuse regulatory rules with the organizations TE charged with making and enforcing those rules. Regulation can be imple- NO mented through a variety of organizations and is more than just "what reg- ulators do." · Regulation can exist where there is no regulator. For example, if we observe that ETOSS, the water regulator in Buenos Aires, claimed the right to direct particular investments by the utility, while in Azerbai- jan the Tariff Council does not direct investments, but does set tariffs, this tells us little about what regulation is or should be. We need a definition that We need a definition that makes it easy to develop regulation that plays an guides good policy. appropriate role in water sector reform. Such a definition starts with an under- standing of the problems that economic regulation should be used to solve and of the differences between regulation and other interventions that could be used to solve those problems. In developing such a definition, we must consider both "economic" and "regulation." Economic Regulation Addresses Monopoly Power Economic regulation Economic regulation is needed to address the problem of natural monopoly. is about stopping In a competitive market, customers can choose between suppliers, so suppliers monopoly abuse. try to offer the products and services that customers want. Competition between suppliers keeps the prices charged in line with costs. For example, in many countries, bread is an essential, but any baker who provided poor qual- ity or overcharged would soon lose business to his or her competitors. Equally, a baker who undercharged would also lose money and have to raise prices or go out of business. In most markets, competition ensures that providers offer what customers want and charge a price that reflects efficient costs. Water utilities are Water utilities are natural monopolies. This means that customers cannot monopolies, and can choose between competing suppliers, so there is no competitive pressure to provide bad service ... ensure that they provide the services that customers want. ... and charge prices well Water is generally worth a lot more than it costs to supply. In other words, the above costs ... value of water piped to the premises is so great and the cost of alternatives so high that customers are often willing to pay several times the reasonable cost of the service, rather than go without the service completely. 6 ... either to make profits Left to themselves, providers could take advantage of this to make high profits or cover inefficiencies. at the expense of consumers. Government-owned providers might also take advantage of consumers by charging too much and would typically dissipate NO the excess charges in inefficiencies (such as low labor productivity or corrup- TE tion). Or they might charge low prices for a poor service, when customers would prefer a good-quality service, even if it meant paying more. 1 Traditional regulation tries For a long time economic regulation focused on private providers in devel- to keep prices down. oped countries, where the concern was that the provider would charge too much. The tools of traditional regulation are therefore largely concerned with stopping prices from rising too high. But regulation can also However, we often observe that publicly owned providers, particularly in devel- be used to help oping countries, charge too little. Charging below cost is meant to help con- government-owned sumers, but is generally counterproductive. When tariffs are below cost, the providers charge more. provider must either rely on government subsidies or cut back on service, maintenance, and investment. Subsidies are seldom sufficiently large and reliable to allow a provider to func- tion at the level that customers want. Even if subsidies are provided, they tend to undermine the customer focus of the provider without necessarily promoting equity (because the taxpayers who fund the subsidy are not necessarily less deserving than the customers who receive it). Commonly, tariff below costs result in poor service, asset deterioration, and an inability to invest to meet growing demand. The damage this does exceeds any benefits from the low tariff, so governments in both developed and devel- oping countries have adapted regulation to bring tariffs up to a level that covers reasonable costs. Adapting regulation like this can be difficult. Clearly, a regulator that keeps tariffs down will be more popular than one that raises them. The need for public acceptance means that regulators are most likely to be effective in creating space for tariff increases where they have already earned public confidence and where they have political support. Economic regulation aims In other words, economic regulation can usefully be thought of as mimicking the to ensure that providers pressures that competition provides in other markets. That is, regulation should offer good service at make providers offer services their customers want at reasonable tariffs. Reason- reasonable prices. able tariffs, in this sense, are tariffs that cover the reasonable cost of providing the service, including a reasonable return on capital used, but no more. 7 Economic Regulation versus Regulation Generally There are other problems, We often use regulation as shorthand for economic regulation. More gener- 1 besides monopoly ally, regulation means legal restrictions on the normal freedom of operation of TE abuse, that regulation people and enterprises. Governments use regulation in pursuit of many objec- NO can tackle, ... tives, not just control of monopoly power. In developing regulatory regimes, it is helpful to distinguish between economic and other types of regulation, including the following: · Environmental. Water providers and other businesses have little incentive to care about the environment. They may overabstract water resources or discharge untreated pollutants. Environmental regulation can stop this. In some countries (such as the United Kingdom), all abstraction from, and discharge to, the environment is controlled by the Environment Agency, while in other countries, there are specific controls that apply only to the water utility. · Safety. Even in competitive markets, information problems may mean that consumers do not know which services are safe and which are not. Gov- ernments often impose product safety standards to combat this problem. For example, food safety standards impose purity requirements on bread and other foods, just as drinking water standards can be used to ensure that water is safe to drink · Consumer Protection. Similarly, governments may regulate for other forms of consumer protection (such as arrangements for handling complaints) both in monopoly and in competitive markets. In Barbados, the Fair Trading Commission deals with customer complaints against all businesses and also regulates utilities. In other countries (for example, Jamaica), the utility regu- lator responds to complaints against utilities. · Social Objectives. Finally, governments may regulate for social objectives, to ensure that service is available to certain groups, redistributing benefits from one group of people to another. ... and the boundary As figure 1.1 shows, economic regulation overlaps with other areas of regula- between economic and tion, making the boundaries somewhat unclear. The core--the area without other forms of regulation the overlaps--is a narrow definition of economic regulation as simply setting, can be blurred. monitoring, and enforcing rules on tariffs and service quality (in particular, pressure and reliability). In the blurred area around the core, a choice is needed as to whether a par- ticular regulatory function should be considered part of economic regulation or dealt with in another way. Table 1.1 lists many of the common overlap areas and provides the arguments for and against treating them as economic regulation. The right approach will differ from country to country and depend on the general regulatory regime, 8 the levels of organizational capacity, and the types of problems that must be addressed. NO Economic Regulation versus Other Interventions TE Economic regulation Governments have a range of tools they can use to limit monopoly power and 1 must be distinguished to achieve social, environmental, safety, and consumer protection objectives. from other government These include the following: interventions. · Ownership. Governments can own water service providers and achieve their desired objectives by telling the providers what to do. How govern- ments tell the providers they own what to do is called "governance." · Fiscal Incentives. Governments can influence providers through subsidies and tax incentives. For example, governments can offer subsidies for extending service to poor households. · Regulation. Governments can use the power of the law to instruct providers to do certain things and can enforce these instructions through penalties and other forms of compulsion. Economic regulation must be distinguished from other types of control (in par- ticular, ownership). Figure 1.1: Defining Water Sector Economic Regulation Source: Castalia. 9 Regulation can support Controlling a water service provider by owning it is not regulation. In fact, reg- public sector ownership ... ulation is, in a sense, a substitute for control through ownership. 1 Because ownership seems to give government complete control and flexibility, TE one might ask, "Why bother with regulation? Why not just own the water com- NO pany?" One answer is that in practice, governments have difficulty in getting their water providers to serve the public interest. Governments try to make state-owned water companies serve the public purpose though "governance Table 1.1: Economic Regulation and Other Policies and Instruments Is this economic regulation? Regulatory function Yes No Controlling drinking Essential part of the serv- Health issue, best dealt water standards ice specification with by health authorities and experts Effluent discharge Essential service specifi- Environmental issue, best standards cation for wastewater dealt with by environmen- services tal authorities Monitoring the utility's Monopolies have little Helping consumers deal response to consumer incentive to treat cus- with merchants is an econ- complaints tomers well. omywide function and requires an economywide Complaints on billing and response (such as a con- service standards can pro- sumer affairs bureau for vide information for moni- all sectors). toring utility performance. Service coverage targets Monopolies may limit Extending service to service by charging high unserved areas is a policy prices, so regulation is decision involving social required to make them objectives and subsidies. offer widespread service. Controlling tariff Monopolies may price- Tariff structure may be structure (in addition discriminate in unjustified used to ensure cross- to the average tariff) ways or set inefficient tariff subsidies and achieve structures. social objectives. Input-based controls such To keep costs at efficient The provider should be as the following: levels and to ensure that given the incentives to · Specifying asset service is sustainable, provide good service at conditions operating efficiency and reasonable cost, and then · Specifying efficiency or asset serviceability may investment and operating performance targets need to be controlled decisions should be left to (such as NRW or staff- directly. provider management. per-connection ratios) Source: Castalia. 10 mechanisms," including appointing the board of directors or management and giving the company instructions or directions. But governance arrange- ments often fail because the government cannot adequately monitor or moti- NO vate management to act in the public interest. For this reason, governments TE may choose to establish a regulator for a public company. The government of Victoria in Australia recently brought all water providers in the state under the 1 jurisdiction of the Essential Services Commission, even though the water providers are publicly owned (see Explanatory Note 7 for a discussion of the pros and cons of regulating publicly owned providers). ... or substitute for it. Alternatively, governments may bring in a private firm to run the water service and subject it to regulation. The U.K. government did this in 1989, and since then many other countries, including Chile, Malaysia, and Romania, have done the same thing. Similarly, governments may subsidize private or publicly owned providers to provide service that would otherwise be uneconomic, in addition to, or instead of, imposing service requirements through regulation. To design good regulation, Overall, clear thinking about regulation demands clarity about what regulation we must recognize is and what it is not. Vital government roles in the water sector that comple- when something is not ment regulation, but are distinct from it, include the following: regulation. · Policy Making. Water policy defines the "ends and means" for the sector (that is, it defines sector objectives and principles and sets out who should do what to achieve those objectives). The extent to which consumers or tax- payers should pay for water services and infrastructure is a policy decision, as is the ownership of the providers and the general strategy for controlling tariffs and service standards. · Ownership, Service Provision, and Governance. Water provider performance is driven largely by four factors: who owns the water assets (asset ownership), who owns the service provider (utility ownership), who is responsible for delivering service (service provision), and how the owner exercises control over the utility's management (governance). In most developing countries, water utilities and assets are owned by the government. The government may retain responsibility for service provision or transfer it to a private provider. A government may establish good governance procedures by exer- cising effective control over the utility through a well-functioning board. Get- ting these four things right is critical to sector performance. They must align with the regulatory design, but they are not themselves regulation. · Coordination. Governments must coordinate the water sector. This involves ensuring that policy decisions and implementation plans are consistent, managing input from the various bodies involved in water sector activities 11 and coordinating water development with other public expenditure priori- ties. The regulatory regime must be coordinated with other interventions, but coordination is not regulation. 1 TE Summary NO To sum up, economic regulation in water involves setting and enforcing rules to address the problem of monopoly in the water sector. This produces the fol- lowing core definition of economic regulation: the rules and organizations that set, monitor, enforce, and change the allowed tariffs and service standards for water providers It may be useful to include other functions in our definition of economic regu- lation. Controlling drinking water quality, effluent discharge, customer service, coverage, and asset condition may be a reaction to a problem of monopoly and therefore come under the heading of economic regulation. However, con- trols in these areas may address wider concerns (such as social and environ- mental objectives). Whether or how these issues should be integrated with the system of economic regulation must be decided case by case. Regulation is definitely distinct from policy, governance, ownership, and sub- sidy arrangements. Successful water sector reform may require action in all these areas, but planning and implementing subsidy regimes or changes in ownership are not regulation. Reform will be more successful if the definitions of the various reform instru- ments are kept separate. Then the interrelationships between regulation and the other reform instruments can be clearly seen and the right mix selected to achieve sector objectives. Further Reading Braetigam, R. 1989. "Optimal Policies for Natural Monopolies." In R. Schmalensee and R. Willig (eds): Handbook of Industrial Organization. The Netherlands: Elsevier Science. Harris, Clive and I. Alexander. 2005. The Regulation of Investment in Utilities: Concepts and Applications. World Bank. Laffont, Jean-Jacques. 2005. Regulation and Development. Cambridge Uni- versity Press. Midttun, Atle and Erik Svindland. 2001. Approaches and Dilemmas in Eco- nomic Regulation. Macmillan. Palgrave. 12 Public Utility Research Center, University of Florida, Université de Toulouse and Pontificia Universidad Católica de Peru. 2004. Annotated Reading List for a Body of Knowledge on the Regulation of Utility Infrastructure and Services. Pre- NO pared for the World Bank, under PPIAF funding. TE 1 13 NOTE 2 -- DESIGNING ECONOMIC REGULATION FOR WATER SUPPLY SERVICES: A FRAMEWORK Overview This note provides a We know that regulation in the water sector is important. However, we some- framework for design of times struggle with what regulation is, what problems it can solve, and how to economic regulation ... design effective regulatory systems that will really work. NO TE ... consisting of simple This note outlines a simple, high-level set of steps that can help with designing 2 logic steps. economic regulation in many countries: 1. Define the problems and objectives in the sector. 2. Determine whether regulation is well suited to the objectives. 3. Define the specific regulatory functions needed to achieve those objectives. 4. Decide which legal instruments are best suited to embody the regulatory rules and which organizations are best suited to perform the regulatory functions. This will generally be Although these steps are simple, they are often not followed. Rather, policy a better approach than makers short-circuit the process, saying, "We know we need regulation, so we importing regulatory had better create a regulator," and importing regulatory designs from else- models designed for where. The resulting regime may be doubly ill adapted, in the senses that it is other countries ... not designed to solve the problems the country really has and also that it does not take into account the political, legal, and organizational cultures and capacities in the country. ... and allow for regula- This note shows that well-designed regulatory regimes can use widely varying tory designs more suited legal and organizational arrangements to achieve similar objectives. This sug- to each country's needs gests that regulatory design must pay more attention to local circumstances and traditions. and traditions than has been done in the past. Steps in Designing Economic Regulation Regulatory design is not Regulation is often a key component of water sector reform, but it is some- a matter of "checking times treated as an end in itself. In an effort to "check the regulation box," the box." governments may pass laws and create regulatory bodies without defining the purpose of economic regulation or how it fits with other issues and organiza- tions in the country and without considering the best way to deliver regulation within each country's legal and institutional culture. 15 It is worth giving an example of just how poorly thought out regulatory reform processes can be. In the mid-1990s, Trinidad and Tobago abolished its public utilities commission (PUC). It did this in part because the PUC had been inef- fective in achieving rational tariffs for the state-owned water utility. The PUC had had no discernable positive impact on service standards or efficiencies and had itself become bloated and expensive. Before the decade was out, however, development agencies interested in help- ing Trinidad and Tobago in reforming its water sector had insisted that it "cre- 2 ate an independent regulator." The Regulated Industries Commission (RIC) TE was established, an entity with legal powers and structure very similar to the former PUC. The RIC is even housed in the same building that the PUC was. NO While the RIC has bright staff working hard to improve the water sector, five years after the establishment of the RIC, water tariffs remain well below cost, water service is intermittent, efficiency low, and investment inadequate. We have not picked this example because it is particularly bad, but because it is illustrative of an approach to regulatory reform that has become all too prevalent. Figure 2.1 outlines a better framework for developing workable economic regulation in water. The steps involve As illustrated in figure 2.1, governments should first identify the water sector the following: define objectives and issues. Without a clear idea of what is to be achieved in the objectives, ... water sector, it will be impossible to develop an effective solution. This seems obvious, but surprisingly often this step is missed or lacks the rigor to allow proposed regulatory and policy reforms to be evaluated against clear objec- tives. Such objectives may be to provide service to people who do not have service now, to make sure that water is drinkable and supplied 24 hours a day, to reduce government expenditure on water, and to make sure that as many people as possible can afford water services. ... analyze what regulation Having identified the sector objectives, governments must decide whether eco- can contribute, ... nomic regulation will help to achieve them. Economic regulation may be the solution, only part of a solution, or not a solution at all. For example, eco- nomic regulation is well suited to keeping tariffs in line with reasonable costs, but cannot by itself achieve social objectives (such as extending service to large numbers of customers who cannot afford to pay the full cost of service). To decide whether regulation is part of the solution, governments must know what economic regulation is. Explanatory Note 1 in this series defines regula- tion as "the rules and organizations that set, monitor, enforce, and change the allowed tariffs and service standards for water providers." ... specify the key Economic regulation has many facets. Effective regulatory design specifies regulatory functions, ... exactly what regulatory functions must be performed to achieve sector objec- 16 Figure 2.1: Framework for Thinking about Regulation Set WSS Objectives Formulate Policies & Strategies NO TE 2 Will Regulation N Choose Other Solutions Help to Achieve (such as subsidies or Sector governance changes) Objectives? Y Define Regulatory Model Identify Regulatory Functions · Price Control · Service Standards Implement Supporting Identify Existing Organizations Policies or & Instruments Governance Arrangements Assign Establish New Functions to Organizations Existing & Instruments Organizations Source: Castalia. tives. These may include controlling prices, setting service standards, defining asset serviceability indicators, and so on. ... and choose the Once regulatory functions have been defined, it is necessary to allocate them legal instruments and to appropriate organizations, and to select legal instruments to embody the organizations in which to regulatory rules. People sometimes assume that an "independent regulator" embed the functions. should perform all regulatory functions. In reality, different functions may be allocated to different organizations. Explanatory Note 3 provides more detailed guidance on this step.) Finally, regulation alone cannot solve all water sector issues. Governments must identify the complementary policies or governance arrangements to com- plete the reforms. For example, in Armenia, regulatory developments and pri- 17 vate participation arrangements have proceeded in parallel, supporting each other. In contrast, in Trinidad and Tobago, regulation might be more effective if it were accompanied by reforms to the governance arrangements for the state-owned water utility. Is Regulation Part of the Solution? Policy makers should not As shown in figure 2.1, before developing regulation, government should assume that regulatory assess whether regulation can help achieve sector objectives. This should not 2 reform is needed. be assumed. It requires empirical testing. For example, in Guyana, Trinidad TE and Tobago, and many other countries, regulation has not been able to over- come political unwillingness to allow water utilities to charge cost-recovery tar- NO iffs. In Azerbaijan, like many other countries, regulation has not been effective in increasing the efficiency of service provision. In the Comoros, regulation crumbled as political order broke down. Table 2.1 shows some common water problems and things to consider in deciding whether economic regulation has a role in solving them. In some cases (such as keeping tariffs at no more than cost-reflective levels), regulation is generally effective, but may need to be supported by complementary poli- cies such as reforming governance. In other cases (such as achieving social objectives), regulation can do little, and government policy and subsidy provi- sion must take the lead. Defining Regulatory Functions "Regulatory functions" Once governments are clear what regulation can do to help solve water sector generally include control- problems, they will be able to define the required regulatory functions. By ling tariffs and service "regulatory functions" we mean what regulation will actually do. For example, standards ... when water companies are privatized (as in Santiago, Chile), the following are clear: · Tariffs must be limited to no more than reasonable levels. · Minimum service standards must be set and enforced. Controlling tariffs and service standards are common, core, regulatory functions. ... and may extend to Economic regulatory functions can be drawn wider than this core, depending controls on asset on the circumstances, as the following examples show: condition, efficiency parameters, coverage · In Manila, the Metropolitan Waterworks and Sewerage System (MWSS) targets, and the like. Regulatory Office has recently created a regime to directly encourage reductions in nonrevenue water levels. · In Vanuatu, the regulatory regime created under the concession contract for Port Vila includes a mechanism for deciding on network extensions. 18 Table 2.1: Is Regulation Part of the Solution? Limits on the effectiveness Problem How regulation could help of regulation The utility is effi- Limiting tariffs to no more than Where the provider is not able cient, but average costs. to reduce costs to efficient lev- tariffs are above els, regulation will causes cost. losses for the company. This may lead to service-standard reductions and increased subsi- NO dies, especially if the provider is publicly owned. TE 2 The utility is effi- Providing a neutral and author- Governments nevertheless hold cient, but average itative view on reasonable cost- tariffs below costs, especially tariffs are below recovery tariffs provides for publicly owned companies. actual cost. legitimacy for tariff increases. This worked for the state of New South Wales (Australia) and in Colombia. The utility is inef- Regulation can support effective In this situation, regulation can- ficient. Average governance and incentive struc- not simultaneously keep tariffs tariffs are below tures to provide pressures for in line with reasonable costs actual costs, but efficiency. It can also allow tar- and allow the provider to be above efficient iffs to rise to cover actual costs. financially viable. If tariffs are costs. to be kept to reasonable costs, the owner will have to be will- ing to cover the utility's losses while efficiency improves. Tariffs are at Regulation can assist in this Cross-subsidies should involve cost, but some case by allowing cross-subsidies a policy decision. There may customers cannot between customer categories. not be enough consumers able afford service. to pay above cost to subsidize all those who need subsidies. In such cases, a taxpayer-funded subsidy may be the only option. Water service Setting a minimum level of If the provider lacks the funds, provision is service and applying penalties motivation, or ability to unreliable. for not meeting it can improve increase service, regulatory service provision. penalties will simply increase the provider's losses. Utility operations Regulation can give incentives If the utility does not respond are inefficient. to reduce costs while maintain- to incentives, this will not be ing service quality. effective. System coverage Regulation can mandate If people cannot afford to pay is poor. increased coverage targets. for service extension, govern- Regulation can allow tariffs to ment policy decisions (such as recover full costs of service, subsidizing service extension) thus making service extension will be needed. viable. Source: Castalia. 19 · In many countries, the regulatory regime also serves to control tariff struc- ture, fulfilling social as well as economic objectives. Controls on asset condition, operating efficiency parameters, coverage, and tariff structure may be appropriate regulatory functions, depending on a coun- try's circumstances and objectives (see Explanatory Note 1 for a discussion of what should and should not be thought of as economic regulation). Allocating Regulatory Functions to Organizations and Instruments 2 TE Regulatory design involves The final step is to allocate regulatory functions to organizations and legal examining the countries' instruments. Before doing this, governments should consider their "regulatory NO existing "regulatory endowment": Which existing organizations have the capacity to do the sort of endowment" of organiza- work required in utility regulation? How has the country regulated utility service tions and traditions, ... providers in the past, and how well has this worked? What attributes of cultural, legal, and administrative traditions in the country could be important in the design of regulation? For example, a country with a civil law tradition may look to civil law models of utility regulation, rather than Anglo-American approaches. A society in which public discussion is traditional in reaching deci- sions may want to enshrine a role for such discussions in its regulatory system. This is where the difference between the approach outlined in figure 2.1 and the conventional "check the regulatory box" approach is most apparent. ... assigning regulatory Government need not create a "regulator" to carry out all regulatory func- functions to particular tions. Governments should consider which organizations are best suited to agencies, ... perform the regulatory functions. A well-functioning ministry, for example, may be a better choice for monitoring provider performance than a new and untested agency. An expert panel, like that used in the Sofia concession, may be a better choice than a public utilities commission for resetting tariffs. ... and embedding rules Similarly, governments should not assume that regulation must be embodied in legal instruments. in any particular legal instrument, such as a statute or license. The better approach is to choose which instruments would be most effective in making the regulatory rules predictable and enforceable in each case. For example, in Azerbaijan, an attempt to give a utility regulator direct legal powers risked undermining the government's plans for the sector and was shelved. In con- trast, contracts between government ministries and private providers do an effective job in controlling tariffs in many Western African and Latin American countries. 20 Various allocations can To emphasize the point that radically different allocations of functions to achieve functionally organizations and instruments can achieve the same functional result, we equivalent results, ... compare the regulatory regimes in England and Wales; Vanuatu; and Welling- ton, New Zealand. As is well known, in the United Kingdom, the Water Services Regulation Authority (Ofwat) was established by statute and given independent responsi- bility for setting, monitoring, and enforcing tariffs and service standards. Table 2.2 shows that on the surface, the regime for economic regulation of NO water in Port Vila, the capital of Vanuatu, could hardly be more different from TE that in England and Wales. In Vanuatu, tariffs and service-standard changes 2 are negotiated between the government and the private operator. A govern- ment ministry is responsible for monitoring and enforcing these standards. If disputes arise between the government and the utility, they are settled by an arbitrator or the normal courts. In Wellington, New Zealand, the City Council owns the water utility. The City Council also sets the water charges and decides on service standards (above certain minimum standards set by the national government). Despite the dissimilarities between the three approaches, each system performs the same basic functions of keeping tariffs and the remuneration of the opera- tor broadly in line with reasonable costs, providing incentives to the utility to be efficient, maintaining and improving service standards, and supporting provider sustainability. ... and the choice should The choice of organizations and instruments to perform regulatory functions be informed largely by the should depend in large part on a country's social, political, and legal tradi- specific institutional and tions, as well as on sector organization and ownership. (Explanatory Note 3 legal environment of each discusses choosing regulatory organizations and instruments in more detail.) country. Table 2.2 on the next page, followed by "Further Reading." 21 Table 2.2: Allocating Regulatory Functions Regulatory England Wellington, functions & Wales Vanuatu New Zealand Set tariffs Ofwat Agreed on between City Council utility and Cabinet; embodied in a conces- sion contract Monitor and enforce Ofwat A government ministry City Council 2 tariffs TE Change tariffs Ofwat Agreement between City Council NO utility and Cabinet Control the remu- Ofwat Controlled by the con- Not applicable neration received by cession contract negoti- (municipal-owned the private operator ated between the utility utility) and the government Set service standards Ofwat Controlled by the con- City Council (for example, water cession contract negoti- pressure and relia- ated between the utility bility) and the government Monitor and enforce Ofwat A government ministry City Council service standards (for example, water pres- sure and reliability) Change service stan- Department for Agreement between City Council dards (for example, Environment, the utility and the gov- water pressure and Food, and Rural ernment, subject to reliability) Affairs (DEFRA), arbitration if they can- Welsh Assembly not agree government; Ofwat responsi- ble for changing level-of-service indicators Resolve disputes Ofwat responsi- Arbitration or the City Council is between provider ble for disputes courts both owner and and regulator or between con- "regulator" of the government sumers and the utility, so disputes utility; price con- would not arise; trol disputes or if they did, the referred to City Council itself Competition would resolve Commission them. Source: Castalia. 22 Further Reading Brown, Ashley C., Jon Stern, Bernard Tenenbaum, and Defne Gencer. 2006. Handbook for Evaluating Infrastructure Regulatory Systems. The World Bank. Washington, D.C. Baron, D. 1989. "Design of Regulatory Mechanisms and Institutions." In R. Schmalensee and R. Willig (eds): Handbook of Industrial Organization. The Netherlands: Elsevier Science. NO Eberhard, A. A. 2006. Infrastructure regulation in developing countries: an TE exploration of hybrid and transitional models. 3rd Annual Conference. Africa 2 Forum of Utility Regulators. Windhoek, Namibia. Estache, Antonio. 1997. Designing Regulatory Institutions for Infrastructure-- Lessons from Argentina. Public Policy for the Private Sector 114. The World Bank. Washington, D.C. Joskow, Paul L. 1998. Regulatory Priorities for Reforming Infrastructure Sectors in Developing Countries. Paper prepared for the Annual World Bank Confer- ence on Development Economics. Washington, D.C. Kessides, I. N. 2004. Reforming Infrastructure: Privatization, Regulation, and Competition. A World Bank Policy Research Report. Lavey, Warren G. 2002. Making and Keeping Regulatory Promises. Federal Communications Law Journal 55. Stone and Webster. 2003. Introduction to Economic Regulation of Water Sup- ply and Wastewater Utilities. Asian Development Bank: Capacity Building for the MWSS Regulatory Office TA 3703-PHI. Young, A. 1997. Consumer Choice? Social Obligations, Cross-Subsidies and Competition in the Privatized Utilities. Center for Management under Regula- tion. World Bank. 2006. Approaches to Private Participation in Water Services: A Toolkit. Funded by PPIAF and The World Bank. Washington, D.C. 23 NOTE 3 -- CHOOSING ORGANIZATIONS AND INSTRUMENTS FOR ECONOMIC REGULATION OF WATER SUPPLY SERVICES Overview Policy makers must decide Economic regulation in the water sector3 puts legal limits on water service which regulatory functions providers to control monopoly power. Core regulatory functions include set- (such as controlling tariffs ting, monitoring, enforcing, and changing the maximum tariffs that water and service standards) providers are allowed to charge and the service standards that they are are needed ... required to provide. Other economic regulatory functions can include control- ling tariff structures, setting coverage targets, or ensuring that asset serviceabil- ity remains above specified levels (see Explanatory Note 1). ... and then allocate these Policy makers must decide which economic regulatory functions are needed in NO functions to organizations their country's water sector. After that, regulatory design involves deciding the and legal instruments. TE following: 3 · Which organizations should have responsibility for which regulatory functions? · Which legal instruments should be used to embody the regulatory rules (such as limits on tariffs, or procedures and powers to change tariffs)? A wide variety It is sometimes erroneously assumed that all regulatory functions must be per- of organizational ... formed by a "regulator". This note shows the diversity of organizational arrangements that can achieve functionally similar regulatory results. ... and legal architectures In choosing instruments for regulation, some familiar with (Anglo-American) are possible. common law tradition may consider it an anathema for regulation to be con- tained in a contract. Those familiar with French civil law traditions may be equally uncomfortable with statutes that give a government agency unilateral power to set tariffs for a private company. In fact, a wide range of legal archi- tectures can give functionally similar results. The right design will This note shows that many organizational and legal architectures can be used often depend on local to achieve similar results. How then to choose the right option in any given sit- institutional capabilities uation? Often, the best architecture will be the one that makes the best use of and legal traditions. existing organizational capacities and achieves consonance with local legal and administrative traditions. 3 The term "water sector" refers to the provision of clean water supply, as well as the collection, treatment, and disposal of wastewater. 25 Organizational Architecture Different jurisdictions Regulatory systems that seem structurally different may carry out the same regu- perform similar latory functions to solve similar problems. This note considers three examples: functions through quite different organizational · The Public Services Commission (PSC) of Florida, a typical U.S. regulator architectures. · The regulatory regime for the Manila water concessions, comprising the Metropolitan Waterworks and Sewerage System (MWSS) board (a self-regu- lating government corporation); its Regulatory Office, which regulates pri- vate concessionaires in accordance with concession contracts and the MWSS statute; and an international appeals tribunal · Water regulation in Colombia, where several organizations are responsible for regulatory functions, including the Comisión de Regulación de Agua Potable y Saneamiento Básico (CRA), the Superintendencia de Servicios Públicos Domiciliarios (SSP), the Ministry of Economic Development, and the Ministry of the Environment (these organizations regulate both public 3 and private service providers at the municipal level) TE NO Table 3.1 summarizes how regulatory functions are allocated to organizations in each of these jurisdictions. Florida has a classic The Florida PSC seems like a classic utility regulator. Established by statute, it "independent regulator." has broad discretion to set, change, monitor, and enforce limits on tariffs. However, the regulatory function of setting standards for water pressure is the job of the Department for Environmental Protection, which also controls stan- dards for drinking water and effluent discharge. In Manila, the Regulatory The Regulatory Office in Manila looks like an independent regulator on the Office is constrained by U.S. or U.K. model, however, the reality is more complex. The Regulatory the concession contracts Office's discretion is limited by the contract, which sets out the rules for tariff and requires approval adjustment. This contract was agreed on between the board of MWSS (the from the asset-owning body that owns the assets and has statutory responsibility for water supply) and company board. the concessionaires that provide water services. Tariff changes recommended by the Regulatory Office must be agreed on by the board and may be appealed to the appeals panel (an arbitration panel). The Department of Health controls drinking water standards, while effluent discharge standards are set by the Department of Environment and Natural Resources. 26 Table 3.1: Organizational Architectures in Three Regulatory Jurisdictions Regulatory Florida, Manila, functions United States Philippines Colombia Set tariffs Public Service Com- Base tariff set during CRA (Comisión de Regulación de Agua y mission (PSC) Divi- bidding Saneamiento Básico -- Regulatory Com- sion of Economic mission for Water and Basic Sanitation Regulation Services) Monitor and PSC Regulatory Office SSP (Superintendencia de Servicios enforce tariff Division of Economic Públicos Domiciliarios -- Public Services limits Regulation Superintendent) Change tariffs PSC Regulatory Office Final CRA (Comisión de Regulación de Agua y Division of Economic approval by MWSS Saneamiento Básico -- Regulatory Regulation board, subject to Commission for Water and Basic Sanita- private law arbitration tion Services) in event of dispute NO Set service stan- Water pressure -- MWSS, set in contract Ministry of Economic Development TE dards (pressure Florida Department 3 and reliability) of Environmental Protection Monitor and Water pressure -- Regulatory Office SSP (Superintendencia de Servicios enforce service Florida Department Públicos Domiciliarios -- Public Services standards (pressure of Environmental Superintendent) and reliability Protection standards) Change service Water pressure -- Regulatory Office Ministry of Economic Development standards (pressure Florida Department MWSS board has final and reliability stan- of Environmental approval dards) Protection Resolve disputes Office of the General Appeals panel SSP (Superintendencia de Servicios Counsel, Courts Públicos Domiciliarios -- Public Services Superintendent) Handle consumer PSC Regulatory Office SSP (Superintendencia de Servicios complaints Division of Regula- Públicos Domiciliarios -- Public Services tory Compliance and Superintendent) Consumer Assistance Set drinking water Florida Department Department of Health Ministry of Economic Development standards of Environmental Protection Set effluent Florida Department Department of the Ministry of the Environment discharge of Environmental Environment and standards Protection Natural Resources Source: Castalia. PSC = Public Service Commission; MWSS = Metropolitan Waterworks and Sewerage Services. Note: All RO decisions are subject to final approval or veto by the MWSS board of trustees. 27 In Colombia, setting tariffs In Colombia, the distribution of regulatory functions across various organiza- is the responsibility of tions is even more apparent. The Comisión de Regulación de Agua y one body, while another Saneamiento Básico (CRA -- the Regulatory Commission for Water and Basic body monitors and Sanitation Services) established the tariff-setting methodology. Providers set enforces compliance. their own tariffs, in accordance with this methodology (or apply to the CRA to set the tariffs a different way). Service standards are set by the Ministry of Eco- nomic Development. The Superintendencia de Servicios Públicos Domiciliarios (SSP ­ Public Services Superintendent) monitors the providers to check that they follow the tariff-setting rules and comply with the service standards. This separation of regulatory powers is deliberate; Colombian administrative tradi- tion requires that a single body should not be responsible for both making and enforcing rules. However, where private providers operate under conces- sion contracts with a municipality, the general practice is for the contract to set service standards and tariffs and to be enforced by the municipality. 3 TE These examples illustrate These examples are only a few among the variety of possible organizational NO the possible range of architectures for regulation. While it would clearly be wrong to conclude that organizational architec- all systems for allocating regulatory responsibilities between organizations tures for regulation. work equally well, it is also wrong to imagine that unified, independent regula- tory agencies patterned after U.S. or U.K. models are the only effective regula- tory organizations. Rather than relying on imported models, the key may be to allocate organizational responsibilities in a manner consonant with organiza- tional capabilities and administrative and legal traditions. Regulatory Instruments Regulation must Economic regulation in the water sector consists of legal controls on water be embodied in legal service providers. These controls are applied by legal instruments. For exam- instruments. ple, in Manila, a decision of the Regulatory Office setting a maximum tariff derives its legal force from the concession contract. From a technical and legal perspective, if a concessionaire charged more than the allowed tariff, it would be a breach of contract. In contrast, a U.S. public utilities commission is typically created by a statute, and the statute typically makes it illegal for a regulated water provider to charge tariffs that have not been approved by the commission. Different jurisdictions There are many ways to make regulatory rules legally enforceable. These use widely different include the following: instruments ... · Statutes. These are legally binding documents passed by a legislature. Statutes may contain detailed regulatory rules themselves, or they may con- fer the power on another body (typically, a minister or a regulatory commis- sion) to make such rules. 28 · Contracts. These are legally binding agreements between two or more par- ties, usually between the government and a private water provider. Terms can be changed only with consent from all parties. Contracts often contain formulas controlling tariffs, as well as minimum service standards, and may also stipulate the mechanisms by which these limits can be changed. · Licenses. Typically, licenses are issued by a minister or executive agency under statute. Like a contract, a license may contain detailed regulatory rules, but it has a more unilateral character than a contract, in that it may provide power for the issuing authority or another government agency to change aspects of the license unilaterally (as is the case with the U.K. water licenses). · Executive Orders. In some countries, executive agencies can issues orders with legal force. Presidential decrees in some of the Former Soviet Union countries or in the Philippines of the Marcos era are examples. Different countries have chosen different legal instruments to implement similar NO regulatory rules. Table 3.2 illustrates the choices that some countries have made. TE 3 ... to achieve similar ends. There is no "right" choice of legal instruments. As these examples show, a sin- gle regulatory system can use several instruments to good effect. More impor- tant, different systems achieve similar results with different legal architectures. Again, the right choice may be a matter of fitting with existing legal and administrative traditions. Table 3.2: Legal Instruments for Regulation in Three Jurisdictions Instrument Manila, Cartagena, purpose Typical U.S. PUC Philippines Colombia Creates major Statute Statute created MWSS, a self- Statue regulatory regulating utility. Concession organization contract mandated the cre- ation of the Regulatory Office. Controls tariffs Decisions (orders) Concession contract and the Lease-affermage of the PUC, given MWSS statute contract legal force by statute Controls Varies Concession contract, statutes, Lease-affermage service and regulations contract standards Source: Castalia. 29 Legal and Organizational Design for Regulation Policy makers want to There are a variety of ways to allocate regulatory functions to organizations know how to choose from and to choose legal instruments to embody regulatory rules. Many arrange- the variety of possible ments work reasonably well and achieve functionally similar results. How arrangements. should policy makers choose one design over another? While each case is unique, the choice should typically be based on whether any proposed design can be expected to do the following: · Perform the necessary regulatory functions competently and predictably · Perform them in a way that does not strain the country's organizational capabilities · Be consonant with legal and administrative traditions 3 TE Where in-country capacity When selecting regulatory organizations, governments should consider the NO is scarce, countries could country's human resources capacity and capabilities. If the country has a small consider using existing population or limited secondary or tertiary education, it may not be sensible to organizations or outsourc- create a separate, independent regulatory body. In these countries, it may be ing regulation to regional better to use staff in existing organizations with appropriate skills or to out- or international bodies. source the functions (for example, to a regional body or a specially created panel of international experts). Many Western African countries conserve water sector expertise by placing regulatory functions within sector ministries or combining asset ownership and regulation in one statutory body. In keeping with the Francophone tradition, the regulatory rules are embodied in contracts with private operators, allowing a reasonable degree of predictability in their application, although the ministry staff are answerable to the government of the day. Several small countries in the Eastern Caribbean have addressed capacity issues through implementing a regional regulatory body for the telecommuni- cations sector, and this model also has promise for water regulation. Contract-based regulation In deciding on the appropriate legal architecture, existing legal traditions and may be more compatible jurisprudence are important. Countries can generally be divided into two cate- with existing jurisprudence gories: those with a tradition of civil law and those with a common law tradi- in civil law countries, while tion. Countries in continental Europe (such as France and Spain) and their common law countries former colonies (for example, many North and West African or Latin American may be more comfortable countries) generally use civil law. The United States, the United Kingdom, and with statute-based inde- many of the latter's former colonies (for example, many Caribbean and East pendent regulators. and Southern African countries), use common law. 30 These different traditions have given rise to two distinct forms of regulation: civil law or French regulation, which evolved from a model of private partici- pation contracts operating under specialized administrative law, and common law or Anglo-American regulation, a tradition of independent regulators that exercise discretion in the public interest. These traditions evolved over a num- ber of years and are based on the specific legal and political arrangements in their countries of origin. Hybrid systems are It is possible to mix and match regulatory concepts from these two traditions; possible, but can lead to however, sometimes the resulting hybrids have not worked as well as hoped. In unexpected problems. Manila, for example, the Regulatory Office was intended as an independent regulator of the concession contracts. The Regulatory Office was bound to fol- low the rules in the concession contracts, but it was also intended to be inde- pendent and to exercise discretion. When a sudden devaluation of the NO Philippines peso put the concession contracts under great strain, the chief reg- ulator thought that the Regulatory Office should use judgment and discretion TE in trying to find a workable solution. Some of his deputies thought that their 3 job was simply to enforce the terms of the contract. These tensions crippled the Regulatory Office. They followed from a failure to define how an inde- pendent regulator can coexist with rules defined in contracts and subject to final decisions by binding arbitration. The constitution of a coun- Regulatory systems must be predictable, especially if private investment is try, as well as its judicial sought. While it may seem useful to government to have the flexibility to and administrative tradi- change the rules easily, this flexibility can in fact be counterproductive because tions, influences which providers may not act in a manner consistent with the existing rules if they instruments will best pro- think that the rules can easily be changed. Stability and commitment are mote regulatory stability. important. In some countries with clear and easily enforceable contract law, contracts offer a good choice for legal commitment because they cannot be changed unless both parties agree. In other countries, governments may not be con- strained by contracts that they have signed, so other instruments are more appropriate. In systems with a separation of powers between the legislature and the execu- tive, and especially in those with bicameral legislatures, statute law is hard to change, once passed, and so can provide a stable basis for regulation. Again, it is a question of choosing the instrument that will work best in the particular legal and administrative traditions of the country concerned. 31 Further Reading ADB. 2001. Regulatory Systems and Networking: Water Utilities and Regula- tory Bodies. Asian Development Bank: Proceedings of the Regional Forum. Manila. Bertolini, Lorenzo. 2006. How to improve regulatory transparency. PPIAF (Pub- lic-Private Infrastructure Advisory Facility) Gridlines Note No.11. Johannsen, Katja Sander. 2003. Regulatory Independence in Theory and Prac- tice: A Survey of Independent Energy Regulators in Eight European Countries. Energy Research Programme and the Danish Research Training Council. Copenhagen. Levy, B. and P. Spiller. 1996. Regulation, Institutions and Commitment. Cam- bridge University Press. 3 TE Samarajiva, Rohan, A. Mahan, and A. Barendse. 2002. Multisector Utility Reg- NO ulation. Discussion Paper 0203. World Dialogue on Regulation for Network Economies (WDR). Lyngby. Smith, W. 1997. "Utility Regulators -- The Independence Debate"; "Utility Regulators -- Roles and Responsibilities"; "Utility Regulators -- Decision-mak- ing Structures, Resources, and Start-up Strategy." The Private Sector in Infra- structure: Strategy, Regulation and Risk. The World Bank. Spiller, P. and W. Savedoff. (eds). 1999. Spilled Water: Institutional Commit- ment in the Provision of Water Services. Inter-American Development Bank, Washington, D.C. Srivastava, Leena. 2000. Issues in Institutional Design of Regulatory Agencies. Paper presented in the South Asia Forum for Infrastructure Regulation (SAFIR) Core Training Course on Infrastructure Regulation and Reform. New Delhi. Tremolet, Sophie; P. Shukla, and C. Venton. 2004. Contracting Out Utility Reg- ulation. The World Bank, Washington D.C. 32 NOTE 4 -- REGULATION AND PRIVATE PARTICIPATION CONTRACTS Overview Economic regulation and Private participation in water is often based on contract. This note looks at private participation in good regulatory design for contract-based private participation. water often go together. Private participation can help water service providers to increase efficiency, invest in infrastructure, and improve service. At the same time, private providers may seek to charge tariffs above cost, skimp on investment, and provide inadequate service. Economic regulation is intended to ensure that the drive for profits leads to lower costs and better service, not higher tariffs and worse service (see Explanatory Note 1). There are two distinct There are two distinct traditions in private participation in water. In the Anglo- traditions in regulation American tradition, the water utility is privately owned, but regulated by an and private participation: independent government agency. This regulator controls the provider's prices the French and the and services. The regulator uses its judgment to set tariff and service standards Anglo-American. at levels that it believes will serve the public interest. NO In the French, contract-based tradition, water infrastructure is publicly owned, TE and the supply of services remains a public responsibility. The public authority 4 contracts with a private firm, allowing it to use the infrastructure and requiring it to provide services at a price stipulated in the contract. Mixing the two Both traditions harness private management and capital to serve the public traditions' designs can interest, but do so in different ways. Problems can arise when the traditions are cause problems, ... combined. Around the world, private participation in water has generally followed the French, contract-based model. However, regulation of the resulting arrange- ment has generally been based on Anglo-American designs. This risks regula- tory confusion. In the French model, many of the regulatory rules are embodied in the private participation contract. In the Anglo-American model, an independent regulator has discretionary power to direct the utility, which may conflict with the operator's contractual rights. Too often governments receive poor advice, resulting in regulatory arrangements that undermine, rather than support, the private participation plans. 33 ... although in principle, In principle, elements of the two traditions can be combined to improve on cross-fertilization either one. The French tradition could be supplemented with a dedicated con- between the two tract-monitoring body and public proceedings, for example. For instance, the traditions is possible. Societe Nationale des Eaux du Senegal (SONES) in Senegal is the contract- monitoring body, as well as the asset holder. The Anglo-American approach might increase certainty with greater use of low-discretion rules that the regu- lator is required to follow, similar to contracts. However, successful hybridiza- tion requires a deep understanding of the two systems and how they might interact with the legal and administrative traditions of the country concerned. Two Different Traditions Private participation Private participation in water is usually done through contract: the government through contract ... retains ownership of the water assets and contracts with a private firm to man- age the systems to deliver water services to customers. There are many types of contract, but in all cases, the responsibilities, rights, and remuneration of the private operator are defined by the contract, and the operator is obliged to the government to deliver the services specified in the contract. Examples include the concession contracts in Manila (Philippines) and Côte d'Ivoire, the lease-affermage contracts in Brno (Czech Republic) and Senegal, and the management contracts in Gaza and Trinidad and Tobago. 4 TE ... derives from France The contractual models commonly used are derived from a French approach to NO private provision of infrastructure that has evolved over more than 100 years. ... and differs markedly In contrast, in the United States and England, private participation in water from the Anglo-American commonly involves a private firm that invests in and owns the assets. Like any tradition of privately firm, it would (were it not for regulation) be free to use its assets as it wished, owned utilities supply whatever service it wanted, and charge whatever prices its customers would pay. ... on which most conven- Economic regulation arrangements in the United Kingdom and United States tional regulatory models share a set of common features. They are based on an autonomous govern- are premised. ment entity known as a regulator (Ofwat in the United Kingdom, a state public utilities commission [PUC] in the United States), and a statute that gives the regulator legal authority to determine maximum allowed tariffs and minimum service standards. The regulator is expected to act in the public interest, but has considerable discretion in its decisions. While the similarities are more important than the differences between the two models, it may be worth mentioning that the key differences include the 34 type of decision maker (an individual in the United Kingdom, a committee in the United States), the decision-making process (executive in the United King- dom, quasi-judicial in the United States), and the type of tariff regulation (price cap in the United Kingdom, cost of service in the United States). The U.S. and U.K. models have been widely copied in both water and electric- ity. The U.K. strain (which typically involves a single decision maker setting price caps by executive decision) has taken root in Australia, Jamaica, and Malawi, to name just a few examples. Barbados, Canada, and the Philippines have long had regulators modeled closely on U.S. PUCs (with decision making by a multimember commission, through a quasi-judicial process). New ones are being created all the time. Recent examples in the water sector include Armenia (operational) and Azerbaijan (proposed). The two traditions differ in The assumptions and machinery embodied in the Anglo-American regulatory fundamental ways, ... tradition differ in fundamental ways from those underlying the French tradition of private participation through contract. For a start, the question of control- ling the profit-seeking behavior of a private firm does not arise in the same way with private participation through contract, because there is no question of the private provider acting solely in its own private interest: the private firm provides the service only because of its contract with the government, which NO confers public service obligations. TE 4 Another fundamental difference is that a contractual approach assumes an agreement between equals. Generally, neither party has the power to unilater- ally alter the relationship. Often the tariffs and service standards are funda- mental contractual terms, and the agreement of both parties is required to change them. This is quite different from a model in which an autonomous government agency has discretion to set tariffs and service standards in what it judges to be the public interest. ... so attempts to merge Too often, Anglo-American style regulatory models have been layered on top of them cause problems. contract-based private participation, without sufficient thought as to how to make them compatible. Regulation with Various Private Participation Contracts The right regulatory There are several types of private participation contract. The regulatory objec- approach depends on the tives--good service, reasonable tariffs, investment, and efficiency--are much type of PSP contract. the same in all situations, but the approach to achieving those objectives depends on the type of contract. 35 The following sections look at a typical concession contract, a typical manage- ment contract, and a typical lease-affermage contract. For each type of con- tract, they discuss how regulatory objectives are achieved purely within the contract-based approach to private participation. They then give examples of what can go wrong when an Anglo-American regulatory approach is applied unthinkingly, before discussing some ways in which ideas from the Anglo- American approach could be adapted to perhaps improve the contract-based approach. Concession Contracts The diagram at left illustrates a concession contract. In some senses, a conces- sion contract is similar to a fully private utility that owns the assets. The conces- sionaire is responsible for all aspects of service provision, and its shareholder(s) or parent company is rewarded with the profit from the utility, after all operating and debt service costs are paid. Figure 4.1: Concession Contract 4 TE NO 36 In a classic concession, the contract sets the service standards and tariff rules. Economic regulation--in the sense of protecting customers by controlling tar- iffs and service standards--is subsumed into the design and monitoring of the concession contract. The regulatory roles in this case are the following: · Monitor performance under the concession contract. · Resolve disputes under the contract. · Provide a mechanism to fill in contractual incompleteness by exercising dis- cretion in a principled and predictable way in those cases (such as tariff resets and response to new information) where discretion is unavoidable. Traditional concession Traditional concession contracts, such as those in Côte d'Ivoire and Vanuatu, contracts contain their own have no special regulatory organizations. Contract monitoring is done by the regulatory framework. sector ministry, a municipality, or an asset holder set up as a statutory body. Tariff resets are agreed on by negotiation between the concessionaire and the government. Arbitration governs disputes, including failure to agree on tariff resets. While these arrangements are not perfect, they are often successful in delivering water services that are of higher quality and more efficient that in similar countries that do not have concession contracts. NO TE 4 Adding a conventional It should be clear that imposing an Anglo-American regulatory framework on Anglo-American regulator a concession contract will not work. The essence of a concession is that the will not work. rules determining the tariff are embodied in a contract between the investor and the government. Because the rules are in a contract, they cannot be changed without the investor's consent. This gives the investor the confidence to invest in water infrastructure, knowing that its tariff expectations are legally protected. On the other hand, a classic Anglo-American style regulator has the legal right to set the tariff at the level it considers reasonable. If a regulator is given such a legal right, it implies that it can override the tariff-setting rules in the contract. This removes the contractual certainty that the investor sought. Alter- natively, if the regulator is bound to follow the rules in the contract, then it becomes a contract-administration unit. This model can provide greater cer- tainty for the future investor, but it is not an independent regulator. In some countries, one group of consultants has been engaged to establish an Anglo-American style regulator, while at the same time the government has tried to develop a traditional concession contract for water. These countries have typically not been able to attract investors to the concession contract or have had to make last-minute changes to the regulatory regime. 37 There are several cases where concession contracts and the Anglo-American traditions have been combined in ways that worked, but in hindsight may have done more harm than good. In Manila, combining a In Manila, the water system was transferred to private firms under two con- Regulatory Office with a cession contracts. Service standards and tariff-setting rules are embodied in concession contract the contract. Influenced by the Ofwat example, the government created a caused confusion. quasi-autonomous Regulatory Office. There has been confusion over the proper role of the Office. For example, when the Asian currency crisis struck, the regulatory rules needed to be changed if the concessionaires were to remain viable. Yet it was not clear whether the Regulatory Office should take the lead in adapting the rules (an approach consistent with an independent agency using its discretion in the public interest) or whether its role was to strictly enforce the terms of the contract and leave any negotiation to the board of MWSS. The board was not intended to have any role in tariff set- ting or regulation, but was the legal signatory to the contract. In the end, a confusing mess of regulatory players were involved, including the Regulatory Office, the board of MWSS, the president of the Philippines, and the arbitra- tor under the contract. 4 TE But there may be ways to Some concession contracts have had more success in creating both special NO combine elements of the government organizations charged with administering the contract and dedi- two traditions. cated mechanisms for making binding decisions at periodic tariff resets. Exam- ples include the Bucharest and Sofia concessions, where each established dedicated units with clearly defined functions. These units point the way toward possible successful cross-fertilization between the two traditions. Management Contracts Management contracts are completely different from concessions in their "regulatory" approach. Under a management contract, the private operator is typically paid a fixed fee for managing the utility, plus a performance fee for meeting financial and service improvement targets. In this scheme, the management contract provides the incentives to improve performance. The targets and payments in the management contract will determine how the operator manages the utility. What happens in this case to regulation as it is conventionally understood (that is, controlling the relationship between the utility and its customers by setting service standards and tariff levels)? 38 Figure 4.2: Management Contract NO TE 4 The management It will still be necessary to set tariffs and service standards for the utility. But in contract itself provides most cases, the regulatory regime that sets the tariffs will have very little impact incentives for efficiency. on the management contractor--and so very little effect on the way the utility is managed. For this reason, in the pure contract-based approach, the tariffs and service standards continue to be set by the government, at its discretion. The theory is that the contract gives the management team incentives to improve the utility. The contract governs the management contractor's remuneration. The government remains as representative of the consumers and sets tariffs at a level that strikes the right balance between financial viability and social acceptability, quite independent of its arrangements with the management contractor. 39 Anglo-American style Again, overlaying a management contract with conventional Anglo-American regulation of the utility is style regulation is likely to be counterproductive. Consider a utility that is both likely to impose risks on subject to a management contract and regulated by a price cap. government, without increased incentives for Price caps are intended to create incentives to increase efficiency. Under a efficiency or performance. price cap, reducing costs can increase utility profits. If the utility is privately owned, this gives the private firm an incentive to reduce costs and so increase the private firm's profits. However, under a typical management contract, the operator's fees do not depend on the utility's profits. In this situation, a price cap does not give the private firm an incentive to increase efficiency. It would increase the risk that the utility would suffer financial distress, at least if com- pared with a cost-plus approach to tariff setting. In Guyana, regulation Guyana implemented a comprehensive water sector reform, including a man- had little effect on a agement contract and Anglo-American style regulation. The Guyanese govern- government utility under ment mandated Guyana Water to sign a management contract with Severn management contact. Trent Water International, a United Kingdom­based firm. The government also gave the public utilities commission authority over the water utility. The inten- tion was that the private managers would make the utility more commercial and efficient, while the regulator would ensure that tariffs reflected reasonable 4 costs and that service quality improved. TE Things have not worked out as planned: Tariffs are well below cost. For social NO and political reasons, the government, which still owns the company, does not want a tariff increase; therefore, the government instructed the board to delay filing for an urgently needed tariff increase. Guyana Water is not meeting the service standards in its license. The public utilities commission would like to enforce compliance with service standards; however, imposing penalties on Guyana Water would only increase its operat- ing deficit. Because the company is publicly owned, the deficit must ultimately be funded by taxpayers. Regulatory models suited A utility operated under a management contract is publicly owned. Taxpayers, to publicly owned utilities not the management contractor, typically bear most of the risks of the business. might be adapted to In looking for a regulatory model, models suited for public utilities may turn out work with management to be a better starting point than those intended for private utilities. (Explanatory contracts. Note 6 discusses options for regulating publicly owned companies.) 40 Lease-affermage contracts Lease-Affermage Contracts also expose the private firm to only some of the Lease-affermage contracts (such as that in Brno, Czech Republic) are intermedi- risk and responsibilities of ate between a concession contract and a management contract. The operator providing the service. takes risk on the operation of the business, as with a concession contract. But like a management contract, the public authority retains responsibility for invest- ment, and this creates a gap between the profits and risks of the service as a whole and the profits and risks of the private operator. In a lease-affermage, the tariff revenue is typically divided into two parts: The first part covers operating and maintenance costs, which are to be retained by Figure 4.3: Lease Contract NO TE 4 41 the private operator. The second part of the tariff goes to the public sector to help finance investment. This complicates tariff regulation. The operator por- tion of the tariff must be governed by the lease-affermage contract; however, the government typically chooses to retain discretion over its portion of the tar- iff, and so over the final tariff faced by customers. Service-standard regulation is also complicated. Many service levels will be jointly determined by operating and capital decisions, but the private operator does not usually control the capital expenditure decisions. For example, to increase reliability, leaking pipes must be fixed. This can be done by patching leaks as they occur (maintenance) or replacing entire sections of the network (capital expenditure). The private operator may argue that reliability standards are being missed because the public sector is falling behind on its pipe replacement program, while the government may argue that the operator is to blame for not doing adequate maintenance work. The lease contract The logic of the classic contract-based system is that the lease-affermage provides incentives for provides incentives for operating efficiency and also gives the operator pre- operating efficiency, while dictability as to what it will earn. The government is accountable to consumers the public authority and therefore determines the appropriate level for the final tariff and for the 4 determines tariffs and investment program. investment. TE Again, simply applying a classic Anglo-American regulatory regime to a lease- NO affermage contract would not work. Consider tariff regulation: A classic inde- pendent regulator would examine operating efficiency and capital needs. The regulator would then set the tariff in line with its estimate of reasonable oper- ating and capital costs. Assume that a regulator reviews a utility under a lease contract and concludes that its capital costs are reasonable, but its operating costs are excessive. The regulator would order a tariff reduction. Under a lease contract, how this reduction in income was allocated between the private operator and the public authority would depend on the details of the contract. There is no guarantee that the inefficient operator, rather than the public authority, would suffer the consequences of the regulatory decision. Alternatively, if the regula- tor has the authority to determine the remuneration of the private operator directly, it would mean that it had the power to override the contract, creating risk and uncertainty. 42 In Azerbaijan, the plan It might seem obvious that basing private participation on a contract and then was to create a regulator creating a regulator with unilateral power to override the contract is a recipe with powers to override for failure. But it happens surprisingly often. In Azerbaijan, for example, the private participation con- government planned to introduce contract-based private participation to the tracts, but reforms have WSS sector. Its advisers drafted statutes to create a United States­style regula- stalled. tory authority with the powers to change the contracts without the consent of the contracting parties. No transactions have yet taken place. If a transaction were attempted, the statutory regulatory system would make it difficult to attract bidders and could result in extensive disputes between the government, the regulator, and the contracted parties. Key regulatory design issues in a lease contract include the following: · Holding the private operator to account for performance when respon- sibility for the system is divided · "Regulating" the public sector component (Conventional regulatory tools harness an operator's profit motive to provide incentives for good perform- ance. Different mechanisms are needed to promote efficiency and well- targeted investment in a public sector agency.) Possible improvements on the pure contract-based system could include the following: NO · A dedicated contract-monitoring unit to manage the lease contract for the TE public authority 4 · A unit to "regulate" the performance of the public authority by assessing the adequacy of its capital investment planning and by benchmarking its efficiency against other similar agencies · A body with the responsibility for publishing the performance of both the private operator and the public authority compared with their contractual obligations and other relevant standards or benchmarks to promote increased transparency and accountability Approaches to Combining the Traditions Where private Regulation is generally intended to complement and support private participa- participation is based on tion. Where private participation is based on a contractual arrangement, contract, the contract is the contract is fundamental--it must be respected. In contract-based private the foundation of the participation, private firms are not buying an asset. Firms are entering an arrangements. arrangement in which their risks, rewards, cash flows, and obligations are determined by the contract. The contract is the deal. The contract is also the primary legal instrument that gives the private firm the predictability and enforceability it needs and so leads it to commit management and investment resources and to take risk. All this means that any additional regulatory mech- anisms must be consistent with, and supportive of, the contract. 43 But contract-based Each approach--reliance on contracts and reliance on independent regula- approaches have some tors--has it own flaws, and each may benefit from drawing on the other. common weaknesses ... Wherever private participation arrangements are based on contract, the approach should be to start with the logic of the contract-based approach, identify its weaknesses, and then look for mechanisms to offset those weak- nesses. Common weaknesses in the contract-based tradition include the following: · Lack of a dedicated unit with the expertise to monitor, enforce, and (where necessary) renegotiate the contract · Lack of transparency in regulatory processes, where they are treated as commercial contractual matters to be settled between the government and the utility behind closed doors · No structured process for public consultation or for the public to contribute and be heard in the regulatory process · In countries where water is a municipal responsibility, concession contracts for each town are the responsibility of the local government. Often, there is no mechanism to reduce costs and increase quality through adopting similar regulatory approaches and sharing information between municipalities. 4 TE ... that may be overcome In thinking about how to overcome these problems, regulatory designers may NO by techniques from the benefit from ideas from the Anglo-American regulatory tradition, including the Anglo-American tradition. following: · Creating an autonomous body to monitor the contract, enforce it, and pro- vide the analytic input when tariffs or other aspects of the contract are reset or renegotiated. Such a unit may be delegated the job of agreeing on changes with the private operator (subject to arbitration in disputes), or it may simply advise the gov- ernment. · Giving a government entity the responsibility to publish information about the performance of the private operator against its contract and about the performance of government agencies involved in the sector to increase transparency and accountability · Establishing a national unit to benchmark the performance of a number of municipal systems and perhaps also to monitor and enforce contracts or reset tariffs on behalf of the municipalities · Creating a customer complaints unit with a mandate to assist customers with complaints they cannot resolve directly with the utility · Finding ways to involve the public in regulatory decision making (for example, by allowing the public to make submissions or ask questions of the utility and to attend public sessions in which the utility presents its case for tariff changes to the government) 44 Further Reading Bakovic, T., B. Tenebaum, and F. Woolf 2003. Regulation by Contract. The World Bank: Working Paper No. 14. Brocklehurst, Clarissa and Jan Janssens. 2004. Innovative Contracts, Sound Relationships: Urban Water Sector Reform in Senegal. Water Supply and Sani- tation Sector Board Discussion Paper 1. The World Bank. Washington, D.C. Burns, Phil and Antonio Estache. 1999. Infrastructure Concessions, Information Flows, and Regulatory Risk. Public Policy for the Private Sector 203. The World Bank. Washington, D.C. Castalia. 2004. Final Report on Key Contract Provisions for Long Term PPP in the Water and Sanitation Sector. Volume I, main report, Report to World Bank and Operator Roundtable. Castalia Strategic Advisors. Crampes, Claude and Antonio Estache. 1996. Regulating Water Concessions: Lessons from the Buenos Aires Concession. Public Policy for the Private Sector 91. World Bank. Washington, D.C. Guasch, J. L. 2004. Granting and Renegotiating Infrastructure Concessions. World Bank Institute. NO TE Kerf, M, R. D. Gray, T. Irwin, C. Lvelesque, R.Tayplor, and M. Klein 1998. 4 Concessions for Infrastructure: A Guide to Their Design and Award. World Bank Technical paper no. 399. Finance, Private Sector, and Infrastructure Network. The International Bank for Reconstruction and Development. The World Bank. Washington, D.C. 45 NOTE 5 -- COST OF SERVICE AND TARIFFS FOR WATER UTILITIES Overview Regulation tries to set Regulation usually aims to set tariffs at a level that allows the utility to cover its tariffs equal to reasonable reasonable costs, but no more, over the medium term. This is enormously chal- costs. lenging. Regulators often do not know what the reasonable cost of service is. Often the reasonable cost of service is above current tariff levels. Increasing tar- iffs to a level that covers reasonable costs is socially and politically challenging. This is difficult because At the same time, many utilities in developing countries are inefficient. Their utilities are inefficient and actual costs are more than what would be deemed reasonable. So tariffs set tariffs often start to cover reasonable costs may condemn these utilities to a vicious cycle of well below costs. losses, underinvestment, and deterioration in both efficiency and service. This note offers an approach to these problems. It describes how to calculate the reasonable cost of service for a water utility and how to use that calcula- tion in controlling tariffs. It suggests ways to deal with inefficient utilities and with social issues in tariff setting. In general, the analysis is applicable to both publicly and privately owned utilities. The note also argues that the debate over the merits of rate-of-return regulation compared with price-cap regulation is overdone. Almost all successful regula- tory approaches base tariffs on estimates of the reasonable cost of service. NO At the heart of most Figure 5.1 illustrates the underlying approach common to most successful reg- TE successful regulatory ulatory regimes. In essence, these are the key steps: 5 regimes is a process for estimating the reasonable · Estimate the total cost that the utility would incur in providing the required cost of service, based service efficiently. We refer to this as the "reasonable cost of service." The on a number of cost cost of service comprises a number of building blocks (that is, the various "building blocks." operating and capital cost components). · Set the actual tariff to be charged (that is, the fixed charges, metered charges, and so on) for each class of customers. The various types of charge and the relativities between them are called the "tariff structure." The tariff structure should be set so that when applied to actual customer numbers and demand, the revenue generated is equal to the cost of service. · Put in place the tariff control regime (that is, the rules that specify how the utility may change its tariffs). This regime may be, for example, a price cap or a United States­style mandated tariff. 47 This note focuses on the first of these three steps--estimating the cost of serv- ice--before discussing the tariff structure and control regime. Estimating the Cost of Service To estimate the cost Estimating the cost of service is conceptually simple. The regulator just adds up of service, add up all all the costs that the utility must incur to provide the required services. This starts the costs ... with summing the actual costs recorded by the service provider, because these are all that can be observed. ... and adjust for costs The regulator must know whether these costs are reasonable; therefore, it that are not reasonable. adjusts actual costs by · ensuring that all costs are properly recorded; · checking whether the utility is inefficient in some areas (in which case, its actual costs are more than is reasonable and must be reduced); and · checking whether actual costs are too low because service levels are too low (in which case, costs must rise if the utility is to provide the required level of service). This process is illustrated in figure 5.1. Regulators in Australia have developed a clear and effective way of setting tariffs using this method, which they refer to as the "building-block" approach. This is similar to the approach used in the United Kingdom, with perhaps the difference that Australian regulators tend to be more open about the cost-plus nature of their methodology. The Australian approach is also similar to the U.S. cost-plus approach, although regulators in Australia focus on forecasts of costs while regulators in the 5 United States may pay more attention to historic costs. The key point is that TE the process shown in figure 5.1 is common to most successful regulatory sys- tems. The steps in the process are discussed below, first for operating NO expenses, then for capital costs. Operating Expenses Operating costs must be This section discusses three of the key operating cost components: labor, correctly recorded and electricity, and provisions for bad debts. These typically account for most of reviewed for possible a utility's operations and maintenance costs. efficiency gains. Labor and electricity costs are usually recorded accurately in the utility's accounts, so the regulator's4 focus is simply on ensuring that the levels of these costs are reasonable. This involves checking how efficient the utility is 4 In this note (Explanatory Note 5), the term "regulator" means the body deciding on the tariffs that the utility is allowed to charge. As discussed in Explanatory Note 2, this could be a conventional independent regulator, another kind of organization (such as a ministry overseeing a publicly owned utility), or an arbitration panel setting tariffs under a contract. 48 Figure 5.1: The Building-Block Approach to Tariff Setting Source: Castalia. Note: CAPEX = capital expenditure; OPEX = operating expenditure. NO (for example, whether costs are too high, or whether the costs are being held TE too low, thereby causing service problems). Box 5.1 gives an example. 5 Many utilities are in arrears on their electricity bills; therefore, in reviewing elec- tricity expenditure, the first job is to ensure that actual bills are being recorded in the accounts. The next step is to review whether efficiency can be improved (for example, by replacing old pumps or reducing leakage). On the other hand, if the company is trying to increase hours of service, this may require additional pumping, which could push up electricity costs. Similarly, many utilities do not collect all the revenue owed to them. The regu- lator first must check that uncollected bills are reflected in the utility's accounts through an accurate provision for bad and doubtful debts. Then the regulator must examine whether the collections rate should be improved so that the rea- sonable provision for bad debts could be reduced. 49 Box 5.1: Benchmarking Labor Costs -- Guyana Water A recent review of the tariffs charged by Guyana Water Inc. (GWI) for the Guyana Public Utilities Commission benchmarked staff numbers per thousand connections against other utilities in the Caribbean and elsewhere. The review found that on this simple indicator, GWI performed well (as this graph shows), but went on to caution: "The number of staff per 1,000 connections is only one measure of labor efficiency. In particular, the measure does not evaluate whether the number of staff employed is sufficient to allow the provider to operate effec- tively and provide acceptable levels of service. Indeed, GWI ... emphasizes that more staff are required to identify and counter system leaks." Source: Castalia. Estimating the reasonable Estimating the reasonable cost of service is an art as much as a science. The cost of service involves regulator must balance the increases in service standards it imposes against judgments on the level of their impact on costs. 5 service required and the TE efficiency gains achievable. Even more difficult are judgments on the rate at which the utility can increase efficiency. If the regulator sets targets for cost reduction that the NO utility cannot meet, tariff revenue will not cover costs. The utility will become financially distressed, making it impossible to improve service levels and effi- ciency. Rather than risk setting unachievable efficiency targets, one option is to set tariffs in line with actual costs initially. A well-run company should be able to reduce costs and increase its cash flows as a result, making it easier to attract investment. At a later tariff review, the regulator might then be able to set tariffs at a lower level, to pass on to customers the benefits of the lower costs. Capital Costs Capital costs are usually Providing water and wastewater services requires substantial capital investment the biggest element of the in pipes and other fixed assets. This capital investment results in two types of cost of service ... capital cost: 50 Box 5.2: Tariffs Compared with Operations and Maintenance Costs This figure plots Asian water utilities according to their average tariff (on the vertical axis) and their average operating and maintenance cost (on the horizontal axis). The 45-degree line indicates the point at which tariffs equal average operating and maintenance costs. Util- ities below the line are making an operating loss. Those above the line are earning more in revenue than they are spending on operations and maintenance (that is, they are covering at least some of their capital costs). Of the 14 utilities shown, 9 are located on or below the line. Only the utilities in Colombo and Phnom Penh are far enough above the 45-degree line to come close to covering all their capital costs. While it is impossible to generalize, many utilities need tariffs that are twice their opera- tions and mainte- nance costs to achieve full cost recovery. Source: C. T. Andrews and C. E. Yniguez, eds., Water in Asian Cities: Utilities' Performance and Civil Society Views, ADB Water for All Series, Vol. 10 (Manila: Asian Development Bank, 2004). NO TE 5 · The cost of replacing old assets when they reach the end of their useful life (called "return of capital") · The cost of providing a return on the capital tied up in the assets (called "return on capital") ... but they are often not Many utilities' tariffs do not allow them to cover their capital costs at all (as accurately recorded. box 5.2 illustrates). That may be fine if the government is reliably providing subsidies to fund the capital. Often though, capital costs are neither recovered through the tariff nor adequately funded by government. The result is decaying infrastructure and declining service. To avoid such decay, it is advisable to start by accurately measuring capital costs. Only when true costs are known can an informed decision be made on the extent to which they should be covered by tariffs or by subsidies. 51 Table 5.1 Approaches to Calculating Capital Costs Depreciation Infrastructure renewals plus a return accounting plus a return on assets on assets Cash needs Return of Depreciation Infrastructure renewals Loan principal payments capital charge plus depreciation on plus cash-financed capital operating assets expenditure Return on Cost of capital Cost of capital times asset Interest payments on loans capital times asset valuation valuation Source: Castalia. There are three ways to calculate capital costs: · Depreciation plus a return on assets · Infrastructure renewals accounting plus a return on assets · Cash needs The table below shows how each of these methods addresses return of, and return on, capital. The first priority is Perhaps the most commonly recommended approach is depreciation plus a to record capital costs. return on assets. Depreciation records the reduction in value of an asset over 5 Various approaches are time. For example, a pump that costs $5 million and has a life of five years TE possible. might be depreciated at $1 million a year. This is a way of recognizing the decline in value of the pump over time. NO The return on assets recognizes that capital tied up in water infrastructure could have an alternative use (for example, it could be invested in income- earning assets instead) and therefore has a cost associated with it. If the return earned on investments with a similar degree of risk is, say, 10 percent, then the cost of capital for the water utility is 10 percent. Using this approach, the total annual cost of the $5 million pump would be $1.5 million: $1 million in depreciation (return of capital) and $0.5 million as the required return on cap- ital invested. Depreciation is the common accounting approach to recognizing the loss in value as assets wear out; however, depreciation is not necessarily well suited as a measure for water infrastructure. A pipe network does not wear out over a predictable life and then get replaced; it is typically repaired and renewed in sections. Infrastructure renewals accounting, which is used by water companies 52 in the United Kingdom, addresses this problem by recording as a cost the medium-term average expenditure required to maintain the network at existing levels of serviceability. This may provide the best estimate of the actual costs involved in keeping the system working and would be used instead of depreci- ation for the infrastructure assets. The return on capital would then be added to this level of expenditure to make up the capital cost figure. The third option is the cash needs approach. Under this approach, the capital cost measure is simply the amount of debt service the utility incurs each year through interest and principal payments on loans. While at first sight this seems to be completely different from the depreciation-plus-return-on-assets method, the two can in fact be equivalent. For example, if the $5 million pump was financed by a $5 million loan at a 10 percent interest rate, then the annual debt service could be $1.5 million per year--the same capital cost as determined using the depreciation-plus-return-on-assets approach. Whichever approach is used, the regulator must determine the regulated asset base, which comprises the value of existing assets and the value of new capi- tal expenditure. Existing assets may Setting the value of existing assets is often difficult. It seems natural to set it need to be valued at equal to the book value of the utility's assets when it enters the regulatory sys- well below book value to tem; however, in many cases, the utility has not previously been earning a avoid tariff shocks. return on those assets. If the cost of service is calculated to include a reason- able return on the book value of assets, the result might be a doubling or tripling of tariffs. A more practical approach is to set the value of existing assets, based on the profits they generate under current tariffs. When the water NO companies in England and Wales were privatized, the regulatory value of existing assets was set at around 10 percent of the current cost book value of TE the assets. If a utility had zero operating profits, this approach would imply a 5 zero regulatory value for existing assets. If the utility will have to service the debt incurred to construct the assets, a good approach may be to set the reg- ulatory value of the existing assets equal to the amount of debt that will have to be serviced. New assets should The regulator will also want to be sure that new capital expenditure is prudent be recorded at cost, but and efficient. This usually involves some combination of (a) requiring regula- with some check on the tory approval of investments before they are made and (b) allowing for ineffi- need for the investment cient or unnecessary capital expenditure to be "disallowed" after it has been and its cost. made. Whatever capital investments are allowed or approved by the regulator, the utility is then allowed to earn a return on them, and this is added to the reasonable cost of service (in a cash-needs approach, the regulator may approve the loans as well as the capital expenditure). 53 The utility should have the Finally, in assessing the reasonable cost of service, it is not enough to just look right balance of operating at operating and capital costs in isolation. Many capital investments may be and capital expenditure to justified in part because they reduce operating costs. For example, replacing reduce total costs. pipes may reduce leakage and thereby cut pumping and chemical costs. The regulator will want to be satisfied that the utility has chosen the right mix of operating and capital expenditures to reduce total costs over the medium term. Setting the Maximum Allowed Revenue The maximum allowed Once the regulator has estimated the reasonable cost of service, it must revenue is how much the decide how much the utility will be allowed to earn (that is, its maximum utility will be allowed to allowed revenue). At first sight, it would seem that the utility should simply be earn. allowed to recover the reasonable cost of service. But things are not always so simple. When the cost of service is higher than current tariffs, increasing tariffs to cover the cost of service could have a severe social impact and political backlash. Equally, if a utility's actual costs are higher than reasonable costs, then limiting the utility's revenue to the reasonable cost levels could cripple the utility's financial viability. These competing tensions can make the regulator's job very difficult. In this section, we outline a systematic way of addressing these tensions. Cost of Service and Social Impact Where reasonable costs The real cost of service may be much higher than the current average tariffs. are above current tariff As box 5.2 illustrated, it is not uncommon for tariffs to be half the current cost levels, increasing tariffs to of service. And while efficiency gains could reduce the cost of service, the cover reasonable costs will need for service improvements can more than offset this. This raises the ques- be difficult. tion, how should the regulator use the cost of service in setting tariffs? Simply increasing tariffs to full costs is often unpalatable. Water tariff increases are 5 always politically sensitive, and rapid and substantial increases in bills can TE cause genuine hardships for some customers. NO The first and essential point to recognize is that if the service is to be provided, the cost of service must be paid by someone: either customers or taxpayers. Many regulators keep tariffs below the cost of service out of concern for cus- tomers' ability to pay. If the gap is not filled by reliable taxpayer-funded subsi- dies, then service will suffer, and the utility will not be able to finance expansion to new areas. Because (for most customers) water service is worth a lot more than it costs, a regulatory approach that prevents the utility from pro- viding the desired service does more harm than good. 54 The regulator may The implication of this is that if the regulator wants some customers to pay less recommend that the than the cost of service, then the regulator must do one of the following: government provide a subsidy to keep tariffs · Be sure that the government will cover the shortfall (for example, by funding lower or build a cross- the infrastructure or providing other direct subsidies). subsidy into the tariff. · Create a tariff structure in which some customers are charged more than the cost of service so that others can pay less. This is commonly referred to as "cross-subsidization," which is discussed in the section Setting the Tariff Structure. In other words, the cost of service is what it is. Social impact issues cannot be addressed by changing cost-of-service estimates. They can be addressed through direct subsidies or possibly by cross-subsidies in the tariff structure. What to Do When Efficient Costs Are Below Actual Costs Actual costs are often Often the efficient cost of service will be lower than a utility's actual costs. higher than reasonable In Baku (Azerbaijan), for example, the utility had bad debts of around 25 costs. percent of revenue, when international experience suggested that an efficient utility would have bad debt provisions of only around 5 percent of revenue. In this situation, regulators are often inclined to set the maximum allowed revenue at efficient cost levels, on the grounds that the utility should not be entitled to recover the costs of its inefficiencies from its customers. In theory, a utility subject to rigorously enforced service and coverage requirements would continue to provide the required service and absorb its losses until it can increase efficiency and return to profitability. In practice, utilities with revenues held below actual costs generally allow NO service and maintenance to decline while trying to cut costs. New areas go unserved, while existing customers suffer increasing inconvenience as service TE deteriorates. 5 Worse still, when the utility is government-owned, the losses will be funded by government, as owner of the company. Losses absorbed by government are ultimately a cost to taxpayers. Often the taxpayers are much the same people as the customers, so what a household saves in lower water bills it pays later in higher taxes. This "money-go-round" would be justified if the losses motivated the govern- ment (as owner) to change the way the utility is managed, increasing efficiency and reducing future losses. But very commonly, governments are not able or willing to do this, and losses simply continue to be absorbed, often in an unplanned way (for example, when the utility is bailed out when it is in danger of defaulting on its obligations to creditors). 55 Limiting tariffs to reason- The reality is that increasing efficiency in a water utility is a difficult and time- able costs may cripple the consuming process that often requires initial increases in expenditure. Typical utility if it cannot increase "quick wins" in improving efficiency include increasing collections and reduc- efficiency fast enough. ing electricity costs by installing more-efficient pumps. But to increase collec- tions may require a new billing system, often costing millions of dollars and taking months or years to procure and install. Staff may need to be retrained or laid off, or new staff hired, incurring additional costs and delays. New pumps often have a payback period of only a few years, but still require up- front capital expenditure. When the utility is bleeding cash and credit lines are exhausted, the quick wins may be unobtainable. This suggests that a hard-line regulatory approach of limiting allowable rev- enue to the efficient cost of service may be counterproductive. A better approach may be A more pragmatic approach may be for the regulator to determine, with the to allow the utility to pass utility and its owners, a realistic and phased program for improving efficiency. on some of its inefficien- This may involve agreeing with the owner (often the government) on the losses cies to customers during a the owner will absorb and on the investment it will make in efficiency improve- transition period. ments. The regulator may then allow the utility to recover the remaining ineffi- ciencies and investment costs from customers. This may seem unpalatable for those schooled in the belief that a regulator's job is to stop a utility from passing on its inefficient costs to customers. How- ever, the alternative of a utility spiraling into deteriorating service and diminish- ing efficiency is often worse for customers. Conclusions on the Maximum Allowed Revenue 5 TE In summary, having estimated the reasonable cost of service, the regulator must translate this into the maximum allowed revenue that the utility will be NO allowed to recover. Where the utility is inefficient, the regulator must decide whether to set the maximum allowed revenue at efficient costs immediately or to allow a transi- tion in which maximum allowed revenue starts higher than the efficient cost of service and converges over time. Unless the utility has owners able to absorb losses and make the investments needed to increase efficiency, the latter is often the better strategy. Where tariffs are below the cost of service, the regulator may be tempted to set maximum allowed revenue below the cost of service to limit tariff increases. This is counterproductive because it will lead to deterioration of service. 56 A better approach may be to encourage government to provide a subsidy suf- ficient to address the social impact concerns and then set maximum allowed revenue equal to the cost of service minus the subsidy provided. Failing this, the regulator may choose to adjust the tariff structure so that those who are better able to pay will absorb a greater share of the total costs than do those who are poorer. The result should be a maximum allowed revenue that is equal to the reason- able cost of service, plus any inefficiencies allowed for a transition period, minus any subsidies provided by government. Setting the Tariff Structure The tariff structure Up to now, this note has focused on the cost of service and the maximum determines which allowed revenue (that is, the total amount of money the utility should be customers pay what. allowed to recover from customers). The next question must be, how much should each customer pay? This is determined by the tariff structure. The simplest tariff structure, workable when customers have meters, would be a single price per cubic meter of water consumed. Another simple tariff struc- ture, frequently used where customers do not have meters, is for each cus- tomer to pay in proportion to the value of his or her property. Most tariff structures are more complex than these simple examples. Cus- tomers' bills may include both a fixed monthly component that does not vary with consumption and a volumetric charge. The charges may differ between customer classes (such as residential, commercial, and industrial). The charge per cubic meter consumed may change as the volume consumed changes, either rising or falling. NO TE 5 A good structure must Amidst this plethora of possibilities, which approach is best? This depends on provide financial viability, the specific circumstances and the regulatory and policy objectives. Generally, reflect costs, and be these objectives include the following: socially acceptable. · Financial viability: in our terminology, ensuring that the maximum allowed revenue can be recovered · Cost-reflectiveness: charging customers in a way that reflects the costs they impose on the system, thus giving them an incentive to limit consumption to efficient levels · Social acceptability: ensuring that charges seem reasonable and that all customers--even those with low incomes--are able to receive at least basic service 57 Setting volumetric tariffs Cost-reflectiveness can generally be approximated by calculating the long-run at the long-run marginal marginal cost of supplying an additional cubic meter of water and setting the cost will signal to charge per cubic meter equal to this. If the revenue generated by such a volu- consumers the true cost metric charge would be less than the maximum allowed revenue, the shortfall of their consumption. can be made up from a fixed monthly charge. If marginal cost pricing would see the utility recover more than the maximum allowed revenue, the regulator may reduce the volumetric charge until expected revenues are equal to the maximum allowed revenue. Social objectives are Considerations of social acceptability often lead to a tariff structure intended often pursued through to make water services cheaper for low-income residential customers. These cross-subsidies in the tariff structures seek to allocate a more-than-proportional share of the maxi- tariff structure (such as mum allowed revenue to better-off or nonresidential customers. The most lifeline blocks or residen- common approaches are the following: tial tariffs that are lower than industrial tariffs). · Charge industrial customers a higher rate to allow residential customers to pay less. · Charge a lower amount for a consumption level assumed to reflect a household's basic needs (referred to as a "lifeline block"), with a higher charge for consumption in excess of this amount. In practice, these These approaches may be well intentioned, but in practice, they seldom approaches tend not to achieve the objective of helping the less well-off. One unintended conse- be effective in helping quence is that they discourage utilities from extending service into poorer resi- poor people. dential areas because the utility will lose money by doing so. 5 Quantity-based consumption subsidies tend to benefit the better-off customers. TE Put more precisely, the subsidy benefits of underpricing tend to be concen- trated in the top four income deciles.5 NO If possible, direct subsidies Overall, a regulator will want to ensure that the tariff structure is cost-reflective are a better approach. and promotes cost recovery. This will minimize costs and promote service for all customers over the medium term. Social equity issues can be addressed through the tariff structure in principle, but in practice, they should be viewed with some skepticism, given their poor performance in practice to date. 5 "Water, Electricity and the Poor: Who Benefits from Utility Subsidies?" Komives, K., Foster, V., Halpern, J., Wodon, Q, 2005­10 58 The Tariff Control Regime The maximum allowed revenue and the tariff structure together determine the tariffs that a utility is allowed to charge immediately after the regulator has made its determination. But tariff controls apply for more than a single moment. How should the controls be extended into the future, and for how long? How do the ideas reviewed so far relate to conventional regulatory con- cepts (such as price caps and rate-of-return regulation)? Once tariffs are set to The simplest approach is for the regulator to set the actual tariffs the utility is recover the maximum allowed to charge and to require the utility to charge those tariffs for the indef- allowed revenue, they may inite future. In this approach, if the utility wished to change its tariffs, it must be left unchanged until the apply to the regulator, asking for tariffs to be reviewed and reset. utility or a customer applies for a review. This simple approach is used by many U.S. regulators. It is associated with rate-of-return regulation and has been adopted in other regimes influenced by the U.S. regulatory tradition, including Barbados and Guyana. In the U.S. tra- dition, customers can also request a tariff review if they think that tariffs exceed the cost of service. This creates a This simple United States­derived approach has a number of drawbacks: United States­style "cost-plus" regime. · It can be inflexible (for example, the utility would need the regulator's permis- sion to reduce tariffs or to introduce a tariff for a new service, which might be unduly restrictive). NO · When input prices are rising rapidly, the U.S. approach can harm the utility's financial health because needed tariff increases will be delayed TE while tariff applications are considered. Similarly, a utility embarking on 5 a major investment program to improve quality of service might have to keep filing for tariff reviews to ensure that it can earn a return on the new capital invested. · It may not promote cost efficiencies on the part of the utility. If the regulator is not easily able to tell whether cost increases or new investments are justi- fied, the utility may simply apply for a tariff increase whenever costs rise, rather than working hard to keep costs down to efficient levels. Conversely, if a utility does reduce costs, customers could then file for a tariff reduction, thus capturing the financial benefits of the cost reductions. 59 Alternatively, the tariff The price-cap approach to tariff control overcomes these problems. The most may be indexed to input famous example of a price cap in water is the RPI+K formula introduced in prices and other factors England and Wales when the regional water authorities were privatized in and reviewed after a 1989. Similar formulas had been used for some time in lease and concession set period, creating a contracts in the French tradition. Price caps are now widely used, including in price cap. Australia and Jamaica (where they are applied to publicly owned utilities). Similar approaches are used in concession contracts in Manila (Philippines), Monteria (Colombia), and Port Vila (Vanuatu). A typical price-cap approach differs from the traditional United States­style tariff control in a number of ways: · The tariff control applies to the average tariff charged. The utility is gener- ally free to change the tariff structure (at least within certain limits) if the average tariff remains below the cap. The maximum allowed average tariff may be calculated by dividing the maximum allowed revenue by forecast demand. · The maximum allowed average tariff is indexed to increases in the price of inputs. In England, the utility may increase its tariffs in line with inflation. More complex formulas, such as the one used in Vanuatu, track a weighted average of changes in wage rates, electricity costs, and the price of other major inputs and allow the utility to increase its average tariff accordingly. · The tariff control may be set on a forward-looking basis. For example, when the regional water authorities were privatized in England and Wales, it was clear that substantial new investments were required. A financial model of the companies was constructed, and the real tariff increase that would be required to allow the companies to finance the new investments was calculated. This real tariff increase entered the formula as the "K" fac- tor. This is simply the amount in excess of inflation by which each company 5 was allowed to increase its average tariff each year. TE · The price cap applies for a set number of years (typically, five), and reviews occur only at the end of the set period. This contrasts with the U.S. approach, NO in which reviews can be requested at any time. Having a fixed review period or "regulatory lag" may increase incentives for cost efficiencies because the utility knows that cost reductions will increase its profits (and cost increases reduce them), at least until the time set for the next review. Whether a price cap In some cases, the traditional U.S. approach of approving a detailed tariff or cost-plus regulation schedule and then requiring that the utility stick with it may be the better is better depends on option. It has the advantage of being clear and easily understood. In other the circumstances. cases, the advantages of indexation, or a forward-looking approach to tariffs, may point toward a price-cap approach. 60 Conclusion Much has been written on the differences between price caps and rate-of- return regulation. But in practice, both approaches depend on calculating a reasonable cost of service, deriving from that a maximum allowed revenue, and then limiting tariffs to a level that is expected to generate that amount of revenue. The Australian regulatory approach of constructing price caps based on cost building blocks (as illustrated in figure 5.1) makes this clear. The Australian regulators are doing explicitly what U.K. regulators have generally done implicitly. Tariff controls in concession contracts are similarly set based on estimates of efficient costs of service. Given this, someone involved in setting up a regulatory framework would be well advised not to start this task with a discussion on whether to use a rate- of-return or a price-cap approach, but to focus on these key steps: · Estimate the reasonable cost of service, which involves reviewing each of the building blocks of operating costs and capital costs to ­ ensure that actual costs are correctly recorded, ­ check whether actual costs are reasonably efficient and estimate reason- able cost levels if they are not, and ­ analyze whether costs must rise to allow the utility to deliver the required standards of service. · Set the maximum allowed revenue, which could differ from the reasonable cost of service if ­ government provides a subsidy to reduce the amount of revenue that must be collected from customers or NO ­ the utility's costs are above efficient levels and the regulator decides to allow the utility to recover those costs from customers TE for a transition period while efficiency is improved. 5 · Determine the tariff structure, which will involve the following: ­ Considering whether to charge some customers less, and others more, for social reasons (lifeline blocks, or industrial charges that are higher than commercial charges, are common examples of such socially moti- vated tariff structures; however, empirical analysis shows that these structures are seldom effective in benefiting poor people, so they should be used with care) ­ Deciding on the relative mix of fixed and volumetric charges ­ Ensuring that the tariff structure chosen is likely to allow the utility to generate revenue equal to the maximum allowed revenue · Determine the tariff control mechanism, which involves deciding whether the regulator will require the utility to use the exact approved tariff levels and structure or whether the regulator will just set the maximum average tariff and allow the utility freedom to change its tariff structure to stay at or 61 below this average. Choosing the tariff control mechanism will involve deciding the following: ­ Whether the utility will be allowed to charge tariffs that are lower than the maximum allowable amounts ­ Whether the maximum allowed tariffs should be indexed to input costs ­ Whether the maximum allowed tariffs should include a factor (such as a "K" factor) specifying the amount by which tariffs should increase or decrease in real terms each year ­ Whether reviews should be allowed only after a certain period of time or whether a tariff review may be triggered whenever the utility or a cus- tomer applies for one Further Reading Andrews, C. T., and C. E. Yniguez (eds.). 2004. Water in Asian Cities: Utilities' Performance and Civil Society Views. ADB Water for All Series, Vol. 10. Manila: ADB (Asian Development Bank). Chisari, Omar O., Antonio Estache, and Catherine Waddams Price. 2001. Access by the Poor in Latin America's Utility Reform: Subsidies and Service Obligations. Discussion Paper 2001/75. World Institute for Development Eco- nomics Research. United Nations University. Helsinki. ERA. 2004. Inquiry on Urban Water and Wastewater Pricing. Economic Regu- latory Authority. Perth. Green, Richard, and Martin Rodriguez Pardina. 1999. Resetting Price Controls for Privatized Utilities: A Manual for Regulators. The World Bank. Washington, D.C. 5 TE ICRC (Independent Competition and Regulatory Commission). 2003. Investi- gation into Prices for Water and Wastewater Services in the ACT. Issues Paper, NO ICRC, ACT (Australian Capital Territory). Komives, Kristin, Vivien Foster, Jonathan Halpern, and Quentin Wodon. 2005. Water, Electricity, and the Poor: Who Benefits from Utility Subsidies? The World Bank. Washington, D.C. McIntosh, A. C. 2003. Asian Water Supplies -- Reaching the Urban Poor. IWA Publishing. PPIAF (Public-Private Infrastructure Advisory Facility) and World Bank. 2006. Approaches to Private Participation in Water Services: A Toolkit, chapters 5­6. Washington, DC: World Bank 62 NOTE 6 -- REGULATING GOVERNMENT-OWNED WATER UTILITIES Overview Most water utilities In most countries, water utilities are owned by governments, which hope that are government-owned. by owning the utility, they can make it operate in the public interest. Should they also Some governments are considering independent regulation of government- be regulated? owned utilities. But if government ownership and regulation each aim to make a utility provide good service at reasonable prices, do we need both mecha- nisms? Can they complement and reinforce each other? Or will they duplicate or conflict with each other? There are times when regulation can complement government ownership. This note describes ways in which government ownership can fail in its mission to make a water utility truly serve the public and the way in which regulation could fill the gap. On the other hand, regulation as commonly applied to private companies may fail with government-owned companies. The rewards and punishments of conventional regulation do not stop with the managers or shareholders of a government-owned utility, but are passed through to customers and taxpayers. Regulation can only complement governance, not replace it. Caution is warranted before applying private models to government companies; however, there are techniques that might work for government companies. These include creating competing streams of advice, providing trusted comparative information, and increasing transparency and public participation. Why Regulate What You Own? Private utilities are Privately owned water utilities are motivated to increase profits, so govern- regulated to control their ments regulate them. The profit motive makes private utilities provide service NO monopoly power. and control costs, but can also cause private companies to put up tariffs and TE (possibly) provide inadequate service. Regulation is intended to make private firms operate in the public interest while allowing the private company to 6 make money if it delivers the required service efficiently. Government ownership Instead of regulating a private utility, governments may own and operate the is another way of doing utility. Public utility ownership is another way of addressing the monopoly prob- the same thing ... lem. Governments direct the utilities they own to achieve social, environmen- tal, safety, and consumer protection objectives. 63 ... through the gover- Governments do this though the governance mechanism. In utilities that are nance mechanism, ... departments of ministries or municipalities, this is done through the normal line of command in civil service. For statutory bodies and government-owned companies, the main governance mechanism is a board, usually appointed by government. The board monitors utility management and sets the strategic direction for the utility. ... which is distinct from This raises the question, what is the difference between governance and regu- regulation. lation? These notes define economic regulation as the organizations and rules that set allowed tariffs and required service standards. Generally, in a govern- ment-owned utility, the board or government decides tariffs and service stan- dards. Governments often do not distinguish between regulation, ownership, and policy: all three functions are conflated in the relationship between the government and the management of the utility. One might expect owner- Governments are (or should be) accountable to their citizens--the totality of ship and governance to water consumers and the unserved; therefore, the interests of the owners and be enough ... the customers should be aligned. Seen in this light, a government-owned water utility is similar to a cooperative, in which the utility is owned by its customers. A cooperative may deal with market power issues without an external regu- lator. For example, La Cooperativa de Servicios Públicos Santa Cruz Ltda (SAGUAPAC) in Santa Cruz, Bolivia, is a consumer cooperative that is gov- erned by its customers, who are also its owners. Until 1998, SAGUAPAC operated as a de facto self-regulated utility. The utility moved to cost recov- ery at the initiative of its owners--in contrast to other public utilities in Bolivia.6 Similarly, one can argue that the customers elect the gov- ernment, which owns and operates the business of a public water utility in the consumers' interests. 6 ... but often it isn't. Yet in many countries, government-owned utilities are inefficient and provide TE poor service. Governments are often unable to make their utilities perform the NO way the government and the people would like. Why is this? There are some systemic reasons: 1. Selective representation of customer needs. Water is a basic need, so utility customers are diverse, spanning widely different social and financial cir- cumstances. However, governments may represent the interests of some 6 Since 1999, SAGUAPAC has had to have its tariffs approved by the Superintendencia de Saneamiento Básico (SISAB), the same as any other regulated water utility in the country. 64 constituencies more than those of others. Often poor or other marginal groups are not represented. 2. Short-term political aims. Higher water tariffs are immediately unpopular, while long-term deterioration in the viability of a water utility and the service it provides is less noticeable. Short-term political motives often drive gov- ernment owners to hold water tariffs below cost or provide subsidies to politically powerful groups. This erodes the financial viability and efficiency of the utility in the long term, and ultimately the quality of service that cus- tomers receive. 3. Capture of the utility for personal ends. Governments may interfere in man- agement of the utility in an ad hoc way intended to benefit themselves or their friends--a minister or mayor may tell the utility to hire his or her friends, to extend tertiary mains to an influential person's new house far from existing mains, or to buy meters from a company owned by a politi- cian's friends. Managers of the utility may be similarly tempted. 4. Provider capture. Government-appointed boards or managers are at risk of being captured by the companies they administer. In other words, instead of acting in the interests of the customers, boards and managers begin to sys- tematically favor the interests of the utility. Often, a board and its manage- ment act in the interests of themselves or unionized staff, using the utility's monopoly power to benefit management and employees at the expense of customers through high pay or low productivity. Tough decisions (such as changing working styles and demanding good performance) are not taken because the benefit to the consumer is outweighed by the difficulties involved in disrupting established behavior in the utility. Governments Adding Regulation to Ownership Therefore, some countries Some governments have recognized these problems and have tried to separate regulation from encourage greater accountability and better utility performance by separating governance ... governance, policy, and regulatory functions. Regulation becomes the responsibility of an autonomous organization that operates at arm's length from the utility and often from existing government structures. NO ... by creating The regulatory organization most frequently recommended is an Ofwat- or TE independent regulators. PUC-style "independent regulator." For example, Jamaica's Office of Utilities Regulation (OUR) regulates the government-owned National Water Commis- 6 sion. In Colombia, the Regulatory Commission for Water and Basic Sanitation Services (CRA) sets tariffs for municipal-owned water utilities. Municipally owned water companies are regulated by public utilities commissions in some states of the United States. In these models, elected officials (or the bureaucrats who report to them) con- tinue to oversee the operations of the utility. However, the utilities must now 65 also comply with the decisions of a separate regulatory agency on tariffs and service standards. This can make sense for There are a number of circumstances in which such separation of roles may commercialized utilities ... make sense: 1. Commercialized utility. Many governments have tried to increase efficiency by making their utilities independent from day-to-day political considera- tions and more profit-oriented. In such settings, public utilities are asked, in effect, to pursue similar objectives to those of private utilities. Hence, public utilities may need to be regulated in the same way and for the same reason as private utilities. ... or where an 2. Political space for tariff increases. An independent regulator may protect gov- independent body can ernments from political pressure, making necessary tariff increases easier. For insulate the sector from example, in the state of New South Wales (Australia), the state government political pressures ... knew that water tariffs should be restructured to reduce the subsidies to households; however, past attempts had failed because they were politically too difficult. The government brought the water sector under the jurisdiction of the Independent Pricing and Regulatory Tribunal (IPART), the state's independ- ent regulatory tribunal, to "take the politics out of price determination." This allowed a transparent, arms-length approach to pricing that engaged with stakeholders and helped insulate the water sector from short-term political pressures. The outcome was much-needed tariff reform.7 ... or ensure that the 3. Information and transparency. Often the utility is the only source of informa- utility is benchmarked tion on the water sector. Consumers and politicians both end up distrusting and scrutinized. the utility, but lacking independent information to assess whether costs, tar- iffs, and services are reasonable. A competent independent body can be an alternative source of information, benchmarking and scrutinizing the utility. A regulatory agency can also manage hearings that allow customers' views to be heard. A regulator can force the utility to disclose information and 6 answer criticisms. TE NO But there is a However, independent regulation of public utilities has often failed to deliver serious problem: lack the expected outcomes. The principal problem is the inability to apply sanc- of sanctions. tions. Effective regulation requires the ability to reward good performance and 7 The reforms involved removal of a free water allowance and the property tax component of charges, increases in the usage charge, and a rebalancing of prices between residential and business users to remove the cross-subsidies paid by business. 66 punish poor performance. When a privately owned utility is inefficient, regula- tors may refuse to grant a tariff increase. This hurts the private owners by reducing their profits. In response, they may change the management team or make other reforms to increase the utilities' efficiency. The key point is that the cost of the penalty is borne by the shareholders, and so the penalty motivates them to action. This logic does not apply in a publicly owned utility. If the regulator punishes a publicly owned utility for inefficient performance by refusing it a tariff increase, the government-owner will have to cover this deficit through its funds, which are generated through taxes. Alternatively, the utility will cut back on expendi- ture, worsening service. In either case, the public suffers. Although in principle, the government could change the board and management of the company or take other steps to improve performance, in practice, this seldom happens. Also, government may Another problem in many cases is that government as owner retains control of still make short-term, the tariffs actually charged and the services actually delivered. For example, in populist decisions. Trinidad and Tobago, the Water and Sewerage Authority is under the jurisdic- tion of the Regulated Industries Commission. Tariffs are well below costs, and the Commission would in all likelihood grant a tariff increase to cover reason- able costs; however, the government, as owner of the utility, has decided that it should not file for a tariff increase. In other words, the short-term political pres- sures to keep tariffs down still dominate, despite the independent regulator. Regulation for Government-Owned Utilities An independent regulator Clearly, creating an independent regulator of a public utility does not auto- is no panacea. matically increase service, efficiency, or cost recovery. So, when is it sensible to have a separate institution regulate publicly owned water utilities? Rather than assuming that "regulation" is the answer and that this requires an "independ- ent regulator," a more subtle approach is warranted. There are various degrees of separation of regulatory and ownership forms of control that the government exercises over a public utility. NO TE There is a spectrum of There is a spectrum of options. At one end is the classic department with no 6 options to consider. regulatory oversight. We could call this the "unitary" end of the spectrum because a single mechanism combines the role of regulator and provider, owner, policy maker, and consumer representative. At the opposite, "dualist" end of the spectrum, a government-owned utility is regulated by an independent agency that tries to treat the government-owned entity on an arm's-length basis, similar to the way in which regulators typically treat a private utility. Both regulation and policy making are separated from 67 ownership. This is the situation in New South Wales and Victoria (Australia), Colombia, Jamaica, and Scotland. In between, a number of options create some of the tensions inherent in the dualist models, without going to the extreme of treating the public company as though it were a private, profit-maximizing entity. For example, shifting up slightly from the unitary end of the spectrum, the gov- ernment could ask a unit in a government ministry to develop a real compe- tence in water utility monitoring. This unit could benchmark the utility and engage external expertise to scrutinize its operations and services. This would provide an independent source of information and advice, making the utility more accountable. In Scotland, for instance, the Water Industry Commissioner (WIC) is responsible for ensuring that the public utility Scottish Water meets its service standards. However, its role is only as an adviser. WIC also advises on Scottish Water's four-yearly revenue cap. Closer to the dualist end of the spectrum would be an independent body that held public hearings and issued public reports on the efficiency and service performance of the utility, but that did not itself set tariffs and service stan- dards, leaving this to government. Choosing a model should The right position on the spectrum between unitary and dualist--or to put it be based on the specific another way, the right degree of separation between governance, policy, and problems you're trying to regulatory mechanisms--depends on a country's circumstances. It is necessary solve. to first determine what is wrong with the existing system and then define a solution that best fixes the problem, as in these examples: · Poor accountability stemming from lack of information may be fixed by pro- viding more information. In some instances, lack of accountability to cus- tomers stems from lack of public information about utilities' performance. In this case, an agency that provides independent information and assess- ment may be a good approach. This would help the government make the right decisions about improving the utility. If government fails to make the 6 right decisions, consumers equipped with the new information will be better TE able to hold the government to account. NO · Lack of expertise may be addressed by an autonomous expert body or panel. Some regulatory activities are complex. While the ministry responsi- ble for the government's ownership of a water utility may be able to per- form routine regulatory functions, such as monitoring service standards, it may not have the necessary expertise for something as complex as a peri- odic tariff review. Establishing an independent regulator does not necessar- ily solve the problem of competence. An alternative solution could be to appoint a panel of experts who would be called upon from time to time. The ministry would still make the final decision, based on advice from the 68 panel. To ensure transparency and accountability, the expert panel's advice could be made public before the ministry's decision. · Short-termist tariff setting may be addressed by giving an independent body powers to set tariffs. If the problem is that political time horizons always result in tariffs being set below costs, the solution may be to give an inde- pendent body power to set tariffs. For this to work, the independent body must have the power to override political decisions on tariffs. It should not have to wait for the utility to file an application before it orders a tariff increase. The political credibility and durability of the body also must be considered. A body with established public credibility and political backing (such as the New South Wales Regulatory Tribunal or Jamaican OUR) should be able to make its tariff decisions stick. But a newly created entity whose first job was to raise tariffs in the face of popular and political oppo- sition might not be sustainable. Regulation should build on Whatever the position on the spectrum chosen, it makes sense to use existing existing competencies, ... organizational competencies in carrying out the new role. Assume that the model chosen is to publish independent benchmarking information on the util- ity, to help the government and the public hold the utility accountable. If a respected regulatory commission were already monitoring electricity and telecommunications companies, it might make sense to give that commission the job of benchmarking water utilities. But it would not necessarily make sense to create an independent regulatory commission solely to benchmark the water agency. If no commission existed, another competent government agency, which might be the Ministry of Water, the Ministry of Finance, or the Auditor General, could be given the job, possibly overseen by an external panel. ... and if the decision If the government decides that government-owned utilities should be regulated is made to regulate like private utilities, it is essential to build in ways to punish poor performance. public utilities like private This is difficult when the government is the owner, because often only the cus- ones, the regulator needs tomers suffer. NO some teeth. TE Countries that have used regulation as a positive force for public utilities have introduced it gradually and as part of a wider set of reforms. As a first step, 6 responsibilities were introduced without specific sanctions. This can both enhance performance by prompting the parties to the contract to focus on results and strengthen the relationship between parties by giving them periodic opportunities to discuss progress and problems. Legally binding rules that include sanctions can be introduced only when performance evaluation sys- tems are functioning properly. Overall, it is easier to introduce positive sanc- tions ("carrots") for good performers before negative ones ("sticks") for bad performers. Establishing a new culture of doing business through incentives is 69 easier than changing an existing culture of noncompliance. Introducing overly ambitious unenforced contracts might create a tradition that is hard to trans- form later on. It may be possible for a regulator to take a more active approach to underper- formance. For example, where there are a number of water utilities, such as in Colombia, the agency could identify the bottom 10 percent of performers and intervene directly to reorganize the management of the worst performers. One option would be to link managers' pay to the performance of the utility. Another option is to allocate scarce investment funds to those utilities that per- form best, using decentralization to introduce competition. In developing countries, there is hardly ever enough investment to fund all required infra- structure investments. This gives governments an option to reward better per- formance without starving others more than they would anyway. For example, Ecuador is pioneering a system in which the national government offers some 220 municipalities free technical assistance and financial incentives if they agree to delegate the provision of water supply and sanitation services to autonomous (public and private) operators. These suggestions may seem radical. But without rewards and sanctions, the regulatory mechanisms used to control private utilities are unlikely to be effective in changing the behavior of underperforming, publicly owned water utilities. Conclusion In summary, the government can address the problem of natural monopoly in the water sector through ownership or through economic regulation. This note considered under what circumstances the government should use both instruments. In general, applying independent regulation to government-owned water utili- ties is not a panacea for underperformance. There are difficulties in coordinat- ing regulation with public sector governance. In certain circumstances, however, it is useful to separate the regulatory responsibilities from the govern- ment's responsibilities as owner and service provider. 6 TE Governments must choose the degree to which they would like economic NO regulatory activities to be performed by an independent organization. Choos- ing the right degree of separation involves determining what problem must be solved and considering the country's existing institutional constraints and capacity. 70 Further Reading Eberhard, A. A., and M. Mtepa. 2003. Reform and regulation of a low-price utility: The case of Eskom in South Africa. International Journal of Regulation and Governance, Vol 3, No 2, 77-102, 2003. Irwin, T. and C. Yamamoto. 2004. Some Options for Improving the Gover- nance of State-Owned Electricity Utilities. World Bank: Energy and Mining Sector Board Discussion Paper, No. 11. Kingdom, Bill and Meike van Ginneken. 2004. From Best Practice to Best Fit: Reforms to Turn Around and Institutionalize Good Performance in Public Utili- ties. Briefing Note for World Bank-WaterAid workshop. NO TE 6 71 NOTE 7 -- REGULATING WASTEWATER SERVICES IN DEVELOPING COUNTRIES Overview What is the role of economic regulation in improving urban wastewater serv- ices in developing countries? In many developing cities, wastewater services are provided mostly by decentralized systems (such as septic tanks). There is seldom a need for economic regulation of decentralized solutions. Improved health and urban environments will come mostly from better environmental regulation, which must be coupled with good policy and planning to manage the transition from decentralized to centralized systems in many densely popu- lated areas. Centralized wastewater systems are often monopolies, justifying economic reg- ulation; however, the cost structures, beneficiaries, and willingness to pay for wastewater services differ from those for water services. This means that gov- ernment subsidies and property-tax­based systems must be considered along- side user-pays charging in deciding how wastewater services should be paid for and regulated. Sanitation in a Developing City Developing cities may Imagine a developing city: More than 70 percent of the households have run- have good water services, ning water, supplied by a centralized water system; yet, fewer than 20 percent but very minimal are connected to a centralized sewer system. The sewers are in the old part of sewerage services. town, in poor condition and often blocked. The sewage discharges untreated into the bay, where poor people go to wash themselves and fishermen catch fish for sale in the market. People may need to rely In the newer areas of town, well-laid-out residential developments have all the on private septic tanks or modern utility services, including cable TV, but no sewer connection. The may have no safe houses and apartments have flush toilets, but these discharge into under- disposal sites at all. ground septic tanks. Even in the squatter settlements, residents have (legally or illegally) connected to the water and electricity networks. But when they want to defecate, they must go down the street to a communal toilet or use the "bag and throw" method.8 A new, centralized NO Urban sanitation is essential to public health and environmental quality. In network would ensure densely populated areas, a centralized sewer network to collect and treat TE that sewerage is safely wastewater is ideal; however, for many developing cities, such a system is 7 disposed of ... 8 Defecating in a cheap plastic bag, tying the bag, and throwing it into waste ground. 73 ... but such a system can decades away. The cost of digging through existing streets and neighborhoods be prohibitively expensive. to retrofit them with sewer networks can seem prohibitive. Generally, the cost of a centralized wastewater system will be higher than the cost of a water sys- tem serving the same area, sometimes by a factor of two or three times. Plan- ning issues and disruption to traffic increase the challenges. Policy and Environmental Regulation of Decentralized Systems In such an environment, what is the role for economic regulation? We define economic regulation as the rules and organizations that set allowed tariffs and service standards. What role is there for a law or government agency imposing tariffs and service standards on decentralized wastewater service providers? Wastewater services Start by considering how wastewater services are actually provided: Many can take several apartment buildings and middle class homes have septic tanks that may need different forms ... to be periodically emptied. Septic-tank emptiers typically transport the resulting "septage" by tanker and dispose of it in a treatment facility or--untreated-- into a body of water. Poorer households may dig and use pit latrines or use slightly more sophisticated facilities built by local masons. ... and be operated Property developers may put in small treatment works to serve new housing with varying degrees of developments. The wastewater from the houses in the development is carried responsibility. in sewers to a single point, where it is treated before being discharged into a local watercourse. Some of these systems are operated and maintained responsibly. In other cases, neither the developers nor the homeowners con- sider themselves responsible for keeping the system in good repair, and the In reality, many treatment plant breaks down, after which the wastewater is discharged without wastewater systems are proper treatment. In the late 1990s, the Water and Sewerage Authority of poorly maintained ... Trinidad and Tobago estimated that there were as many as 600 such small systems (in a country of 1.1 million people) and that most of these were in a poor state of repair. ... or provide poor levels Pit latrines, septic tanks, and other decentralized solutions are often inade- of sanitation. quate from a health and environmental point of view. Pit latrines may not ade- quately isolate waste from the people and properties nearby. Flies may travel from the latrine to the kitchen, contaminating food. Septic tanks and latrines 7 often allow waste to leak into the surrounding groundwater. In densely popu- TE lated areas, this can drain into rivers (as in Manila, Philippines) or contami- nate aquifers with nitrates (as has happened in Kingston, Jamaica). NO 74 Economic regulation Economic regulation cannot solve these problems, which are not those of will not solve these a monopoly provider setting tariffs too high or failing to provide services. problems, ... It makes no sense to think of controlling the price of self-dug pit latrines or regulating the quality of service provided to householders by their own septic ... and price controls tanks. A similar logic applies to decentralized third-party providers. There are seem unnecessary for no significant economies of scale in septic-tank emptying, for example, so small providers. competition between the providers should ensure that the service provided to the paying consumer is reasonable and that prices are competitive.9 But there IS a role for The real problems of decentralized systems are not that those who pay for environmental and health the service are being exploited, but that third parties suffer. A household that regulation. pays for its septic tank to be emptied considers the service complete when the tanker drives the septage away. If the truck then discharges the waste untreated into the harbor, it is those who use the harbor who suffer. Similarly, when nitrates seep into an aquifer, rendering it unusable, the whole commu- nity suffers. Stopping such environmental and public health problems is a mat- ter for environmental and public health regulation, not economic regulation. Developing environmental Without intending to generalize about the right environmental and regulatory regulation requires ... approach to wastewater, this note lists the following steps to be taken in many cases: ... identifying emissions, ... · Identify emissions into the environment that are socially harmful. · Set targets to reduce those emissions to levels that are socially optimal. ... their optimal levels, ... · Develop ways to bring actual emissions down to those targets. ... and strategies for Emissions include the leaching of nitrates into groundwater, the discharge of emission reduction. untreated wastewater into bodies of water, and even the transfer of fecal mat- ter by flies and rodents out of unsanitary latrines into a neighbor's property. It is through these emissions that patterns of service provision that seem privately beneficial become socially harmful. Care must be taken in Setting the optimal level of discharge involves technical and economic analy- setting targets: ... sis. In Kingston, Jamaica, the value of the aquifer that has become polluted by nitrates would perhaps have justified rules to limit discharges from latrines and septic tanks, but this analysis was not done. Flies and rodents moving in and NO out of unimproved latrines harm public health, so minimizing such emissions will generally be justified; however, not all emissions should be prevented. The TE 7 9 In some cases, tanker operators may collude to form a cartel; however, in this case, the better response may be to stop the cartelization rather than attempt to control the prices charged and services offered by the cartel. 75 ... sometimes the cost of World Bank investigated discharge of septage into the lagoon around the city reducing emissions can be of Lagos, Nigeria. The study found that provided the septage was discharged greater than the benefits. only on the ebb tide, it would not have any serious impact on water quality in the lagoon.10 In that case, limiting septage discharge would not be justified. Technological Once a target level for emissions has been set, ways must be found to bring innovations can also emissions down to that level. Sometimes this will involve environmental regula- help reduce emissions tion (for example, householders might be required to install septic tanks and without regulations, ... latrines that comply with specified technical standards). Other times, nonregu- latory approaches will be better. In Dar es Salaam, Tanzania, many house- holds rely on self-dug latrines that are periodically cleaned out by latrine emptiers. Recently, the city assisted the latrine emptiers to develop a mecha- nized hand pump and vacuum system for emptying the latrines, which reduced the cost of emptying the latrines and also reduced the risk of contamination.11 ... but a centralized In most cities, the best long-term solution will be construction of a centralized system may be the sewerage system to serve the densely populated areas. Reaching that point will best way to address both require a number of policy and planning decisions: first, whether a centralized environmental problems system is indeed warranted, and then, decisions on how it is to be planned, and service needs. financed, built, and operated. For many cities, decentralized and centralized systems will coexist, and the policy challenge is to move both systems in the direction of providing a cleaner, healthier environment. Regulation of Centralized Systems Monopoly operators Where there are centralized wastewater systems, economic regulation may be of centralized wastewater needed. This is especially the case where households are required either to con- systems may need to nect to the wastewater system or to pay the wastewater charge even if they do be regulated. not connect. In these cases, the wastewater service provider has a legal monop- oly, and it is reasonable to have regulatory mechanisms to ensure that service is adequate and that charges are no higher than the reasonable cost of service. Wastewater services It often seems natural to regulate wastewater services in the same way that could be regulated in water services are regulated. The reasonable cost of service can be calculated, the same way that adding together required operating and capital costs. Users can then be water services are ... required to pay a tariff that allows the utility to recover its cost of service. 7 TE NO 10 BNWPP, Practical Wastewater Treatment Requirements -- Lessons from activities supported by the BNWPP Wastewater Window. World Bank (mimeo, April 2003). 11 Water Resources and Environment Technical Note D.2 Water Quality Management: Wastewater Treatment World Bank p.13. 76 Because the wastewater discharged by a typical household or small business is generally proportionate to the water consumed on that property, the waste- water tariff can be set on a volumetric basis. The customer's bill is then calcu- lated by multiplying the wastewater tariff by the reading on the water meter. ... such as by controlling User-pays (that is, volumetric charging for wastewater services) is gaining in volumetric prices. popularity. In 1995, Vancouver, Canada, moved from charging for wastewater as part of the property tax to a largely volumetric charge. In China, the new Draft Guidelines for Wastewater Tariffs state, This is a popular "Wastewater services should be financed from user charges, and there should approach, ... be a progressive move to full cost recovery. º All domestic and institutional customers and most commercial customers should pay for wastewater services based on a uniform price per cubic meter of water supply." ... but it is not always the While volumetric (user-pays) charging is suitable for some countries, it is not best approach. necessarily the best choice for all, for three main reasons: · Volumetric charging is not cost-reflective and does not generally send effi- cient price signals. Volumetric charges are · The beneficiaries of wastewater systems are often not those who connect to not efficient and seldom the system--they might have been quite happy with their existing on-site recover the full cost of methods--but the wider community that benefits from reduced contamina- connections. tion of the environment. · The expense of installing new centralized systems is such that it is often socially and politically impossible to recover the full cost through an incre- ment to the water charge. As a practical matter, if the system is to be financed, other ways of paying for it must be found. The cost of wastewater Volumetric charging for households and small businesses is not, in fact, disposal is driven more cost-reflective. The main cost driver in the wastewater business is not the by pollution load and volume of wastewater; rather, the pollution load (essentially, the amount of network size than by organic material in the wastewater) is the biggest determinant of treatment water volume, ... costs. The costs of the collection network itself are driven largely by the NO length of the network and the levels of groundwater and rainwater inflow and infiltration into the sewer network. Inflows and infiltration determine the TE required pipe diameters, holding tank capacities, and so on. For these 7 reasons, the state of New South Wales, Australia, issued the following wastewater tariffs guidelines: 77 ... so prices based on "Pay-for use sewerage pricing is not warranted for residential customers due to wastewater volume are a lack of net benefits from such pricing. The costs of sewage collection and not cost-reflective. transfer are largely driven by hydraulic capacity which is dependent on wet weather flow, and the cost of treatment works is driven by biological and sus- pended solids loads which relate to the number of people serviced." Volumetric user charging If volumetric charging for wastewater is in place and working well, it makes may still be retained, ... sense to retain that system. Volumetric charging may also be justified as a sec- ond-best strategy: where water charges are well below the marginal cost of water, adding a volumetric wastewater charge can help to improve customers' incentives to use water wisely. But the notion that volumetric user charging automatically reflects costs or increases efficiency is simply wrong. ... but some other source For developing cities that must expand their centralized wastewater system, the of funding is likely to be costs are such that often the real issue is finding a realistic, socially acceptable required for cost recovery. way of paying for the system. Because a wastewater system generally costs as much or more than a water system, recovering the costs of a new wastewater system will generally at least double tariffs. In fact, because most water tariffs Fully cost-reflective do not recover the capital costs of the system, the tariff impact will often be charges may be higher. It is common for the total costs of a centralized wastewater system to unacceptably high, ... exceed US$1 per cubic meter. Where adding a charge of this amount to the water bill would be socially unacceptable, funding from other sources must be found if the system is to go ahead. ... and because waste- Moreover, it is not clear that recovering the full costs from those connected to water systems have social the system is the right choice, from either an efficiency or equity perspective. In benefits, their costs many cases, customers who are offered the choice to connect to the waste- could be spread over the water system and pay its costs prefer not to. In this case, it is hard to say that wider community. the customers are the beneficiaries because, from their perspective, the bene- fits of the system are less than its costs. And yet, the community may still decide that it is worth having a wastewater system and requiring people to connect. In this situation, it might be reasonable to spread the costs across the broader community, and not simply those connected to the system. 7 For example, sanitation One option is to require all those whose properties are passed by the sewer TE charges could be applied network to pay, whether they do or do not connect to it. This has the advan- to ALL properties, ... tage of encouraging people to connect and reflecting the fact that everyone in NO an area benefits from a system that removes wastewater from that area. The 78 disadvantage of this approach is that it can generate resistance from those who do not connect to the system. ...or the cost could Other approaches include recovering the cost of wastewater systems through be recovered through local property taxes or national government grants. Payment through property property taxes or taxes reflects the fact that everyone in the area benefits from wastewater general taxation. removal and treatment, and this may in fact increase property values. Grants from the national government reflect the broader benefits of wastewater treat- ment. Treating wastewater in an upstream municipality benefits downstream municipalities, which may justify funding wastewater treatment on a national, or at least a river-basin-wide, basis. Economic regulation sets Economic regulation involves setting service standards, as well as controlling standards for good cus- tariffs; however, the most important standards for a wastewater provider are tomer service, not those that govern the quality of the effluent discharged. These are generally environmental safety. best regarded as a matter for environmental regulation, rather than economic regulation. Economic regulation, however, should be concerned with the serv- ice as experienced by the customer. This may involve setting standards to ensure that sewers do not block and back up into people's properties, plus standards about responses to complaints and enquiries. Service standards In setting service standards, economic regulation must consider the various should be set to reflect technologies available, the cost of different standards, and people's willing- community demand. ness to pay. For example, condominial sewerage systems have been devel- oped in Brazil to reduce the cost of sewage collection.12 These systems have smaller-diameter pipes than a conventional system does and may run across users' property, rather than being buried in the street. While cheaper, condominial systems are also somewhat less con- venient and more prone to blockage. Also, rapid, widespread expansion of service seems desirable, but is costly. The regulatory system must weigh the costs and benefits of alternative expansion plans and technologies in setting both tariffs and service standards for centralized providers. NO 12 Jose Carlos Melo, "The Experience of Condominial Water and Sewerage Systems in Brazil: Case Studies TE from Brasilia, Salvador, and Parauapebas," report for Water and Sanitation Program, Latin America and the Caribbean (WSP-LAC), World Bank, and for Bank-Netherlands Water Partnership Program (BNWPP) (Wash- 7 ington, DC: World Bank, 2005). 79 Summary In most developing cities, wastewater services are largely decentralized. Improving wastewater services is therefore a matter of improving the sanitary characteristics of septic tanks, pit latrines, and small systems serving discrete housing developments. This is not really an issue for economic regulation, but rather a question for environmental regulation and policy. In many cases, the ideal long-term solution will be a centralized system, and the big policy challenge will be deciding how such a system can be paid for, installed, and managed. Most cities have at least some centralized wastewater collection and treatment network. These systems may be natural or legal monopolies, so there is a role for regulation in setting allowed tariffs and service standards. It may not be possible or sensible to recover the full costs of the system from the customers connected to it. In many cases, there is a role for local or national govern- ment contributions to the cost of the system to reflect its community benefits. The economic regulatory task then is to calculate the costs of service, exclude those costs covered by government, and allow the provider to recover the remaining costs from customers on some reasonable basis. Further Reading BNWPP (Bank-Netherlands Water Partnership Program). 2003. Practical Wastewater Treatment Requirements: Lessons from Activities Supported by the BNWPP Wastewater Window. The World Bank. Washington, D.C. Burian, Steven J., Stephan J. Nix, Robert E. Pitt, and S. Rocky Durrans. 2000. "Urban Wastewater Management in the United States: Past, Present, and Future." Journal of Urban Technology 7 (3): 33­62. Department of Land and Water Conservation. 2002. Water Supply, Sewerage and Trade Waste Pricing Guidelines. New South Wales, Australia Melo, Jose Carlos. 2005. The Experience of Condominial Water and Sewerage Systems in Brazil: Case Studies from Brasilia, Salvador, and Parauapebas. Report for Water and Sanitation Program, Latin America and the Caribbean (WSP-LAC), World Bank, and for Bank-Netherlands Water Partnership Program (BNWPP), The World Bank. Washington, D.C. UNEP (United Nations Environment Programme). 2001 (draft). Guidance on Municipal Wastewater: Practical Guidance for Implementing the Global Pro- 7 gramme of Action for the Protection of the Marine Environment from Land- TE based Activities (GPA) on Sewage. UNEP/GPA Coordination Office. The Hague, Netherlands. NO 80 World Bank. 2003. Water Resources and the Environment: Technical Note D2: Water Quality Management, Wastewater Treatment. The World Bank. Wash- ington, D.C. NO TE 7 81 LIST OF WATER SUPPLY AND SANITATION SECTOR BOARD DISCUSSION PAPER SERIES 1. Innovative Contracts, Sound Relationships: Urban Water Sector Reform in Senegal. Clarissa Brocklehurst and Jan G. Janssens. January, 2004 2. Can the Principles of Franchising be used to Improve Water Supply and Sanitation Services? -- A Preliminary Analysis. Meike van Ginneken, Ross Tyler and David Tagg. January, 2004 3. Ten Years of Water Service Reform in Latin America: Towards an Anglo-French Model. Vivien Foster. January, 2005 4. Financing Water Supply and Sanitation Investments: Utilizing Risk Mitigation Instruments to Bridge the Financing Gap. Aldo Baietti and Peter Raymond. January, 2005 5. Water for the Urban Poor: Water Markets, Household Demand, and Service Preferences in Kenya. Sumila Gulyani, Debabrata Talukdar and R. Mukami Kariuki. January, 2005 THE WORLD BANK GROUP Bank Netherlands Water Partnership The World Bank 1818 H Street N.W. Washington, D.C. 20433 USA