RESTRUCTURING OF ELECTRICITY SECTOR K. K. A. Chandana Samarasinghe B.Sc (Eng), PG Dip. EE, MBA, C.Eng, MIE(SL),MIEE(UK) This thesis was submitted to the department of Electrical Engineering of the University of Moratuwa in partial fulfillment of the requirements for the degree of Master of Engineering Department of Electrical Engineering University of Moratuwa Sri Lanka February 2003 7 7 7 0 7 ABSTRCT Traditionally Electricity Generation, Transmission and Distribution activities were considered as one vertically integrated business in the past. But. by restructuring the system these three areas can be unbundled to three separate units. This will help to introduce increased competition of the supply and choice for the customers. In general, by introducing restructuring, it is expected to bring about lower energy price and lower capital expenditure for governments. Some countries have already achieved successful results by restructuring the electricity sector but some have come across various problems and difficulties. The aim this project is to study the restructuring process of the electricity sector in detail and to suggest a suitable model for Sri Lanka. In order to achieve this different industry structures have been analysed to study the different alternatives of restructuring. Also, legal and economic framework of restructuring, experiences in different countries and various other factors to be considered in restructuring have been discussed. By critically analysing the preset position in the electricity industry in Sri Lanka, important factors to be considered in the process of restructuring the electricity sector have been identified. Finally a suitable model for Sri Lanka has been suggested. This project has been conducted mainly using the published material and information accessible in the environment. ACKNOWLEDGEMENT There are so many who inspired and guided me in completing this thesis. I may not be able to thank every one of them individually in this acknowledgement. Therefore, while appreciating everybody's contribution, I would thank personnel who contributed most for this thesis. Special appreciation goes to my supervisor, Professor Priyan~a D.C. Wijethunge, for his valuable guidance and support, which was a major factor in completing this thesis. Also, my sincere thanks arc due to Professor Rohan Lucas, Head of the Department of Electrical Engineering, University of Moratuwa and the course coordinators of the Master of Engineering course for the year 2000/2001 , Dr. Aruna Ranaweera, Dr. Jehan Peiris, Mrs. Anula Abegunawardane and Dr. Nalin Wickramarachchi for their valuable support in completing this thesis. ... Thanks for all friends and colleagues in the CEB, LECO and Master of Engineering batch 2000/2001 at University of Moratuwa who have helped me in various ways for this thesis. Finally my special thanks go to my wife Yamuna for her inspiring support given to me in following this Master of Engineering course. II TABLE OF CONTENTS ABSTRACT ACKNOWLEDGEMENT TABLE OF CONTENTS LIST OF FIGURES CHAPTER 1 : INTRODUCTION 1.1 Background 1.2 Transaction costs 1.3 Movement to competition 1.4 Restructuring and Privatisation 1.5 Basic models of industry structure 1 .6 The forms of ownership and management 1 . 7 Scope of study CHAPTER 2 : REGULA TORY AND LEGISLATIVE FRAMEWORK 2.1 Introduction 2.2 Economic regulation 2.3 Objectives of economic regulation .J 2.4 Functions and responsibilities of Regulatory Commission 2.5 Key characteristics of a Regulatory Commission 2.6 Legislative Framework 2.7 Conclusion CHAPTER 3 : INDUSTRY STRUCTURES . 3.1 The four models 3.2 Implications for the structure of companies 3.3 Economic efficiency 3.4 Social policy obligations 3.5 Treatment of "Stranded Costs" 3.6 Pressures fo r change CHAPTER 4 : MODEL 1 - MONOPOLY 4.1 Description of the model 4.2 Trading arrangements 4.3 Transmission access 4.4 Should the Distcos be separate? 4.5 Achieving efficiency ' 4.6 Social policy obligations 4.7 Implications for asset values and stranded costs 4.8 Pressure for change 4.9 Examples of model 1 4.10 Conclusions Ill Page Numbers II iii vi 1 1 3 4 6 8 9 11 12 12 12 12 14 16 17 18 13 13 21 22 23 24 24 26 26 26 27 27 28 30 30 31 32 33 CHAPTER 5 : MODEL 2 - PURCHASING AGENCY 35 5.1 Description of the model 35 5.2 Transition mechanisms 36 5.3 Trading arrangements 36 5.4 Transmission access 37 5.5 Should the purchasing agency be separate? 38 5.6 Achieving efficiency 39 5. 7 Social policy obligations 41 5.8 Implications for asset values and stranded costs 41 5.9 Pressure for change 42 5.10 Examples ofmodel2 43 5.11 Conclusions 44 CHAPTER 6 : MODEL 3- WHOLESALE COMPETITIO~ 47 6.1 Description of the model 4 7 6.2 Trading arrangements 48 6.3 Transmission access 49 6.4 Implications for the structure of companies 50 6.5 Achieving efficiency 54 6.6 Social policy obligations 55 6. 7 Implications for asset values and stranded costs 55 6.8 Pressure for change 56 6. 9 Examples of model 3 57 6.10 Conclusions 57 .... CHAPTER 7 : MODEL 4 - RETAIL COMPETITION 60 7.1 Description of the model 60 7.2 Trading arrangements 61 7.3 Implications for the structure of companies 62 7.4 Achieving efficiency 64 7.5 Social policy obligations • 67 7.6 Implications for asset values and stranded costs 68 7.7 Netback Pricing 70 7.8 Examples of model 4 71 7. 9 Conclusions 71 CHAPTER 8 : WHICH MODEL TO BE SELECTED?. 73 8.1 Final thoughts on the structure question 73 CHAPTER 9 : EXPERIENCE IN DIFFERENT COUNTRIES 78 9.1 Introduction 78 9.2 England and Wales 78 9.3 India '.- 79 9.4 Chile 81 9.5 Argentina 82 9.6 Philippines 83 9. 7 Malaysia 84 9.8 General discussion 84 IV CHAPTER 10 : DIFFERENT SITUATINS TO BE CONSIDERED 10.1 Introduction 10.2 Relevance to small systems I 0.3 Costs and benefits of privatisation 10.4 Costs and benefits of vertical separation 10.5 Costs and benefits of horizontal separation in generation 10.6 Current situation in the world CHAPTER 11 : RESTRUCTURING OF ELECTRICITY SECTOR IN SRI LANKA 11.1 11.2 11.3 11.4 11.5 11.6 11.7 11.8 11.9 11.10 11.11 11.12 11.13 Introduction History and Background Government's vision for the power sector The structure of the power sector in Sri Lanka The present status of the Sri Lankan power sector Regulatory frame work Problems faced by Sri Lankan power sector Reasons for the problems in the power sector Why do we need restructuring in the power sector ~ .J Suggested option for restructuring in the Sri Lankan power sector Structure and number of SBUs Recommendations for a Regulatory Commission Conclusion BIBLIOGRAPHY .... v 87 87 87 88 90 91 93 100 100 101 101 102 104 106 106 112 119 119 121 125 127 128 LIST OF FIGURES FIGURE 1 : MODEL 1 -MONOPOLY FIGURE 2 : MODEL 2 - PURCHASING AGENCY FIGURE 3 : MODEL 3 - WHOLESALE COMPETITIO_f' FIGURE 4 : MODEL 4 - RETAIL COMPETITION ... VI Page Numbers 25 34 46 59 CHAPTER 1 INTRODUCTION l.l Background Traditionally electricity industry was considered as a naturaJfinonopoly where all the functions of generation, transmission and distribution of electricity interconnected together and vertically integrated into one single system. In most of the cases it was belonged to the government or to one single firm. First it was realised that generation was no longer a natural monopoly. This in turn was due to the changes in generating costs in the 1980s. The generating portion of the industry had been thought of as a natural monopoly because of the economies of scale ... that could be obtained by purchasing large and more efficient plants. These plants were large with respect to the size of the market. Even as markets got larger, and the use of electricity increased, so did the optimal size of plant. The optimal size of generating unit rose through the period 1970 to 1980, as it had for the previous 50 years. Then things . began to turn around. Technology imported from materials science and the space programme made turbines much more efficient than they had ever been. At the same time, the price of natural gas declined, and the prohibitions on gas burning which the western countries had imposed were repealed. The way was then clear for smaller and cheaper generating units to be built economically; they were the cheapest form of new construction but, more important, in many cases the ali-in cost of a new plant was lower than the customers were paying for the sunk costs of old plant. Customers began to think about building their own plant, and wanted to know why they could not change suppliers to get cheaper products. Also people started to think why they can not have competition and choice in purchasing electricity as any other product. The idea which underlies the new world of competition and choice in electricity is that it is possible and desirable to separate the transportation from the thing transported. That is, electric energy as a product can be separated commercially from transmission as a service. In more simple term~, we have been used to think of electricity as a product that we only use at the point of delivery, and pay for in a single delivered tariff. The question now is, could the bill be "unbundled" into an electricity and a delivery charge? Even if• the delivery service J remained a monopoly, could the customer choose who would supply the electricity over the wires? Could the wires be "common carriers" even though the physics of the system dictate that the product is fully intermingled and indistinguishable? This seemingly simple question is central to understanding what is going on in the electric industry today. But if it is possible to define and separate the transport service, so that it can be provided separately from the electricity itself, electricity becomes a product that can be bought and sold and transported from place to place, much as any other product. ... Electricity markets are opened to alternative producers and alternative purchasers. The economic analysis required for this type of world is the analysis of transactions. What is the product being bought and sold, at what time, in what place, who is the buyer, who is the seller, what is the price, how is it determined, what are the conditions of sale have to be clearly identified. It is certainly true that transporting electricity is physically more complicated than transporting most other goods. Transmission requires split-second timing of electricity flows from producers, or the system will go out of control with disastrous consequences. In physical terms, transport and production are inevitably closely related. Transmission company must have access to electric energy (beyond that transported) to be able to run a transmission system. In the past, the need for central control of production and transportation resulted automatically in "vertical integration": generation and transmission and local distribution were integrated within the same firm. Distribution might be provided by a separate company, but with each distribution company tied to only one generating company by contract. Some have argued that this 2 is necessary because of the physical relationships, as the eggs and cake analogy suggests. However, the more subtle argument is an analysis of the costs and benefits of separating them. 1.2 Transaction costs It is theoretically possible to replace command-and-control relationships (within a firm) with "contractual" relationships (between firms). "Contractual" •relationships in this J context may mean any agreement about the terms on which ·transactions take place between the separate firms. However, the difficulty of fully specifying all the necessary terms of the contract so that all possible situations are covered may be so great, and so expensive to negotiate, execute and litigate, that it is not worth attempting; it is more efficient to keep the activities with a single firm where one manager manages both activities. The technical term for the costs of negotiating, executing and litigating the required contracting mechanisms are transaction costs. Transactions costs are the costs associated with making contracts to replace command and control. The following important aspects in the institutional and technical context of the electric industry have to be noted carefully in contractual relationships: • a regional transmission-coordination systcm'with interconnected generating plants; • a mechanism for dispatching generating plant that recognises the need for physical control second by second, but permits and encourages economic (least cost) dispatch; • some method for coordinating unit commitment and maintenance; • some method for ensuring that adequate generating capacity is built; • some method for ensuring minimum cost investment, system wide; • some method for dealing with emergencies. Then it was widely agreed that these requirements were correct, but that the difficulties were insuperable because the transactions costs of separating transmission from generation and distribution were simply too great. Vertical integration from production 3 to consumption was the natural condition of the industry, because of the transactio ns costs of separating them. 1.3 Movement to competition However, this did not necessarily rule out competition entirely. Although the utili ty needed to maintain control over plant construction decisions and the operation of t he .. transmission system, there could be some competition to build and operate plant. In t he US, the Public Utilities Regulatory Policy Act (1978), known as !uRPA, introduced the idea of competition in generation. Established utilities were required to purcha se power from independent generators at prices that equalled their "avoided costs". Aft er initial skirmishes in the courts, the independent generators (called Independent Power Producers or IPPs) flourished. However, they were not allowed to sell to end consumers but had to sell all their output to the local monopoly utility. Some states overestimate d avoided costs so badly that they induced excessive amounts of new independe nt capacity; as a corrective measure, during the 1980s, -competitive bidding to build an d operate capacity, and contracts for the output of the plants, became standard procedu re for new plants in many states. The growth of the IPPs demonstrated forcefully th at economies of scale in generation were no longer a sufficient consideration to dicta te that generation was a natural monopoly. By 1993: some 50% of new capacity in the US was being constructed by IPPs. Competition in generation was now possible. Even if the utility had become a sort of purchasing agency for generation, there was sti ll no question of giving the US electricity customers a choice of who would supply them. By the mid 1980s there was near-universal agreement that the industry was naturall y vertically integrated (although some competition at the generation level was possible if a purchasing agency coordinated everything). Into this conventional wisdom stumbled the British government in early 1988. 'they published a White Paper proposing that th e electricity privatisation should include breaking up the Central Electricity Generati ng Board (CEGB), the nationalised industry that owned all the generating plants and the 4 transmission system. (The distribution system was also government-owned as twelve separate companies, each with a local monopoly over customers, and each able to purchase electricity from only the CEGB.) Existing plant would be divided between two generating companies; new entry of competing independent generators would be encouraged; a separate transmission company would be established; the distribution companies would provide local transport, and customers would choose their suppliers, to encourage competition. The previous privatisations of the telephone and gas • industries as private sector monopolies had apparently not COI/mced the government that they were reaping all the benefits that private sector disciplines were supposed to provide. So electricity would be made competitive. At that time there was a clear understanding of the conceptual problems of transactions costs and the problems of replacing a command structure with a system of contracts. However, most people thought it impossible at the time. After two years of negotiations, false starts, massive computer programs commissioned and abandoned, the current market structure took i<>rm, and was implemented in March 1990. The new structure, which separates the product from the transportation at all levels, consists of competing generators, regulated transport companies at two levels (transmission and distribution) and competing retailers. The UK system is a highly organised market with more rules than a nom1al market, to ensure system stability. This central commercial structure has worked remarkably well. The complaints about the UK system, of which there have been many, relate to the winners and losers during the changes, and to the small number of competitors. None of the complaints relate to the feasibility of setting up a disaggregated commercial system. It was shown to be feasible to arrange contracts that allowed open access to transmission and distribution wires, at least in a single island with a single transmission company. However, the fact that something is technically feasible does not make it necessarily desirable, the transaction costs may still be too high. At the beginning many countries strongly opposed the move to open access. The European Community Council turned 5 down such a proposal although it was considering opening generation to competitive bidding, believing competition in generation to capture most of the benefits and few of the costs of open access. The European Union also offered some access to large customers. The US has been moving gradually towards choice for at least some customers: particularly for independent distribution companies, known as wholesale customers, who have previously been tied to a single supplier. The US Energy Policy Act of 1992 (EPAct) permitted wholesale customers a choice of supplier, and obligated > utilities to transmit ('•wheel") power across their territory to accomplish this (this is .I known as ·'wholesale wheeling"). However, the same Act prohibited the federal authorities from mandating choice for retail customers ("retail wheeling"), although individual states may permit it, and some are now considering doing so. California announced that it intended to go to "direct access" or competitive markets for all consumers, in April 1994. Restructuring the electric industry is conceptually different from privatising it, which is a change from public to private ownership. In the UK. the restructuring of the commercial relationships was done with the privatisation. Elsewhere (and in other industries in the UK) privatisation has taken place without restructuring. In other countries, notably the US, where the sector has been largely private for many years, restructuring is taking place without changes of ownership. China is considering restructuring without relinquishing government ownership. It is not therefore surprising that confusion reigns in this area. The number of options seems limitless. The tools for addressing them are being developed in a fairly unsystematic way, in response to particular concerns in particular countries. 1.4 Restructuring and Privatisation All over the world, governments and ·regulators are considering whether to restructure and/or privatise their electric industries. Mostly their aim is to increase efficiency through better investment decisions, better use of existing plant. better management and 6 better choices for customers. Sometimes they are driven to it by customers who feel that they can purchase more cheaply elsewhere, sometimes shortages force a search for new sources of capital, sometimes the incumbent utility has become inefficient and the problem is to introduce incentives, sometimes the utilities themselves want to be freed from inhibiting intervention. Whatever the reason, they need to know their alternatives, and the implications of a change. Restructuring and privatisation are different dimensions of change.,· ;I • Restructuring is about commercial arrangements for 'selling energy: separat- ing or "unbundling" integrated industry structures and introducing competi- tion and choice. • Privatisation is a change from government to private ownership, and is the end-point of a continuum of changes in ownership I management. In the UK. when the electric industry was privatised, it was also restructured. The two need not go together. They arc two almost separate dimensions of change. However, there is a practical logic linking the two decisions. [fa government decides it wants to privatise its electric industry (or any industry) it needs to place a value on the assets. The value of the assets will depend upon the revenues the assets can earn. To provide investors with sufficient information to decide what the assets are worth, the government must itself decide what system will be adopted to determine the flow of revenues. In the case of electric industry the sources and certainty of revenues will be crucial. Regulatory systems are put in place to control costs and prices and to make investment decisions in the absence of competition. Regulation is a surrogate for competition, to be used when competition is unworkable. However, once it has been shown that competition is feasible the question must arise as to whether it would make sense to introduce it and how much to introduce. Hence, the question of restructuring inevitably arises in conjunction with considerations of privatization. 7 1.5 Basic models of industry structure The structure of the electric industry of any country can vary according to two dimensions. Those two dimensions are degree of competition and the ownership. Under the degree of competition we can identify four basic models to structure an • electric industry, but there can be a lot of other models which are-combinations and I extensions of these four basic models. These four basic modefs can be defined as follows: • Model I has no competition at all. • Model 2 allows or requires a single buyer or purchasing agency to choose from a number of different producers. to encourage competition in generation. • Model 3 allows Distribution companies to choose their supplier, which brings competition into generation and wh-olesale supply. • Model 4 allows all customers to choose their supplier. which implies full retail competition. < Also the ownership dimension can conveniently be divided into three basic levels: • In some countries, the electric industry is a government department, with no separate accounts, and often with responsibilities that are only remotely connected to electricity production. • The next level is a distinct government-owned company or nationalised industry. • The third level is a privately owned industry. A useful way to look at these two dimensions is as a matrix which a country might be anywhere on the matrix. 8 1.6 The forms of ownership and management Many of the global changes in the electric industry are changes in ownership and management. These changes are concerned with putting pressures on enterprises to behave more commercially, but without necessarily changing the structure of the industry. ~ ;l Owners are defined as those "who are entitled to the profits of the· industry"; owners appoint managers to ensure that the enterprise is run efficiently, give them authority to do so, and hold them accountable for the results. The three most common forms of ownership I management are: 1. Direct government ownership: The government both owns and has direct managerial control over the industry, as in China at present (and as was formerly the case in many countries). The same people are owners, regulators and managers, although sometimes they have different "nameplates'' in their different roles. Investment is done with government appropriations, prices are set by the government and revenues are remitted to the government accounts. The government focuses on central planning, perhaps in conjunction with other industries; it should be concerned with investment appraisal and efficiency, but that is not its primary focus. The industry is viewed as "infrastructure". The government may impose other tasks on the electric industry, such as responsibil ity for schools and hospitals in a region. 2. A government-owned corporation: The government owns a corporation which manages the industry so that government is one step removed from day to day control. The board of the corporation sets the ·goals and appoints different people to the management, to achieve those goals. The corporation may still be required to carry out other government policies such as support of supplying industries, but it is under some 9 obligation to show a profit in its activities. There may be an independent regulatory agency, or the government department may approve prices and investment policy. This is the case with Electricite de France (EDF) in France, and used to be the case in the UK under the Central Electricity Generating Board (CEGB). 3. A privately owned corporation: A third form of ownership is private ownership of the corporation and its assets, as in the US and now the UK. These , companies Goint stock companies) may be listed on the stock exchanges, and are J expected to make profits for their shareholders (who may be the employees of the company). The managers are accountable to the Board, which represents the shareholders. These companies are generally regulated by an independent regulator. These distinctions are never rigid in practice, the government may in effect have total control even over private companies. The level of government control may depend more on the intentions and behaviour of the government than on the organisation of the sector. ... The three most common forms of different levels of government control are: 1. Commercialisation happens when the government relinquishes detailed control, in favour of autonomy for the enterprise and a focus on profitability. This is a change in behaviour rather than organization. It normally involves adoption of commercial accounting practices, economic tariffs, and an effort to separate the core business from other activities. 2. Corporatisation is the formal and legal move from direct government control to a legal corporation with separate management. This may be a government-owned corporation. The ownership of asSets and the capital structure need to be determined before this step is taken. The government also needs to set out the objectives for the corporation, and the process by which public policy objectives are taken into account. 10 Economic regulation may be introduced at this stage to oversee pricing and investment policies. 3. Privatisation is the move from a government corporation to a privately held corporation. Incentives for efficiency are considered even greater if management is subject to the disciplines of stock market valuation of the company, which happens when the enterprise is privatised. Privatisation may also be undertaken to increase the - ~ company's access to capital markets. Privatisation is accomplished· by a flotation on the ;I stock market or a trade sale. This requires a valuation, a prospectus and registration on a stock exchange. It is accompanied by an increase in external regulation of the monopoly elements of the industry. 1. 7 The scope of the study The following areas involved in the restructuring of electricity industry have been concentrated in this study: ... • Regulatory and Legislative framework. • Different industry structures and alternatives of restructuring . . • Different situations to be considered in restructuring. • Experience in different countries. • Comments and suggestions on the Sri Lankan system. 11 ' ' CHAPTER2 REGULA TORY AND LEGISLATIVE FRAMEWORKS 2.1 Introduction During last two decades many countries around the world restructured their electricity industries in order to achieve economic efficiency and: social objectives. ' This transformation from a natural monopoly to a competitivlindustry is associated with transformation in the industry regulatory and legislative frameworks. This chapter examines the objectives, institutional framework, and processes of regulation in the restructured electricity industry. Functions, responsibilities and key characteristics of a regulatory commission are also discussed. Further, the existing legislative framework is identified as an important factor in electricity industry restructuring. ... 2.2 Economic Regulation The economic regulation of a market can be defined as "explicit public or governmental intervention into a market. to achieve a public policy or social objective that the market fails to accomplishes on its own." The fundamental justifications for governments to intervene in electricity industry oversight are universal even though the mode of implementation differs from country to country. The main justification is that electricity is a strategic sector and essential to the well being of society. Secondly, considering the physical interconnectivity of the system, a single firm may be able to provide electricity at a cheaper rate than a combination of firms. On the other hand a single firm can keep prices higher than economically justifiable. Thus there may be a "natural monopoly" 2.3 Objectives of Economic Regulation The main objectives of economic regulation can be classified as follows: 12 • Efficiency Both allocative and productive efficiencies should be considered. Since traditional electric utilities generally do not operate in competitive markets that would impose cost disciplines upon them, regulation must fulfill that function. This objective is promoted by setting rates that reflect, to the greatest extent possible, the marginal costs of production. • Fair prices , Prices should be fair for both consumers and investors. This nleans price regulation is intended to restrict economically unjustifiable profits to the investors but still provide them with a reasonable return on their investments. • Non-discriminatory access to services This refers to providing access to services such as transmission for all market participants. ... • Adequate quality and reliability Quality and reliability in electricity is critically important in modern day society since electricity has become an "essential service". • Other stated public policy objectives Eg. Environmental protection, low income support etc. As an essential and important component in national infrastructure, the electricity sector should support public policy programs. However, a natural monopoly may not be in favor of such programs. Therefore, regulation is an essential mechanism to achieve public goals such as: • Policies on low income customers and rural areas, • Investments in energy efficiency programs to minimize social cost, 13 • Investments in environmentally friendly technologies and minimization of environmental impact from existing system, • Consumer protection and education programs, • Research and development on electricity generation, delivery, use and impacts. Therefore, regulation is important when the strategic nature, monopolistic characteristics and public policy roles of the electricity industry are considered. 2.4 Functions & Responsibilities of Regulatory Commission , In poor countries, the traditional monopoly electricity industt;lmay experience low levels of reliability, an inability to meet customer demand and a lack of capital for expansion. In rich countries, electricity prices may be higher than necessary. Therefore, many countries are presently focussing attention on reforming the sector in order to achieve short and long term economic benefits. Electricity industry reforms generally involve two main steps: • Utility operations are transformed into an enterprise format from the traditional vertically integrated government utility. • A regulating mechanism/commission is established to regulate the transformed industry. A regulatory commission can impose a variety of economic regulations on the sector in order to improve economic efficiency and address associated issues. The Federal Electricity Regulatory Commission (FERC) in the USA and the Office of gas and Electricity Markets (OFGEM) in the UK are examples of regulatory commissions. The main functions and responsibilities of a regulatory commission include: • Rate setting (tariff setting), • General regulatory rule making, • Utility system resource planning, • Environmental impacts of resource utilization, 14 • Conservation and efficient use of utility and societal resources, • Consumer protection, • Maintenance of the utility's financial integrity, • Assuring high system reliability, • Utilization of appropriate tools to assure that utility management is given the proper set of incentives. For example, Office of Gas and Electricity Markets (OFGEM) states its main tasks as; • • ~ ;I Promote competition in all parts of the gas and electricity industries by creating the conditions which allow companies to compete fairly and which enable customers to make an informed choice between suppliers; Regulate areas of the gas and electricity industries where competition is not effective by setting price controls and standards to ensure customers get value for money and a reliable service. Federal Energy Regulatory Commission (FERC) of the United States describes its functions on electric power regulation as "The Commission approves rates for wholesale electric sales of electricity and transmission in interstate commerce for private utilities, power marketers, power pools, power exchanges and independent system operators. The Commission oversees the issuance of certain stock and debt securities, assumption of obligations and liabilities, and mergers. The Commission reviews the holding of officer and director positions between top officials in utilities and certain other firms they do business with. Finally, the Commission reviews rates set by the federal power marketing administrations, such as the Bonneville Power Administration, confers exempt wholesale generator status under the EP Act, and certifies qualifying small power production and cogeneration facilities". These functions and responsibilities· often compete with each other. Therefore. a regulatory commission has to set up a balanced and a workable regulatory framework. 15 2.5 Key Characteristics of a Regulatory Commission The structure, scope and powers of a regulatory commission is key to the success of restructured electricity industry. The key characteristics of a good regulatory commission include; • Independence from the political process, • Independence from the regulated enterprise, • A broad mandate to protect the public interest, ' • Technical expertise in the functions and business of the regulatld enterprise, • Continuing monitoring and enforcement of rules and orders. Independence is the most important factor for a successful regulatory commission. The strategic and essential nature of the electricity industry attracts a lot of political attention. Political influence of the regulatory commission may increase the risk of investment in the sector, resulting in higher financing costs. Being a capital- intensive industry, higher financing costs greatl~ influence electricity prices. Further, decisions of the commission such as pricing may be viewed by the public as political decisions. Therefore, keeping public confidence is also an important factor. The FERC is composed of five members who are appointed by the President of the United States, with the advice and consent of the Senate. Commissioners serve five- year terms, and have an equal vote on regulatory matters. No more than three members may belong to the same political party. One member is designated by the President to serve as Chair and administrative head. Therefore, FERC is not entirely independent from political influence. But restriction of political party affiliation reduces political influence upon FERC. The National Electricity Code Administrator Limited (NECA) and National Electricity Market Management Company (NEMMCO) are bodies set up under National Electricity Law (NEL) in Australia that have some regulatory functions. The governments of states participating in the National Electricity Market (NEM) 16 appoint directors for these two bodies. Therefore, no state government has, control over the affairs of these organizations. The regulatory framework may vary from country to country, depending on the existing electricity industry structure, existing legal structure and the national culture of that country. Therefore, decision on the most suitable regulatory fram~work for a country should address relevant country specific issues. 2.6 Legislative Framework A restructured electricity industry contains both competitiv/processes and regulated monopolies where competition is impossible. The following fundamental properties of electricity should be considered in designing a competitive electricity industry; • Ephemerality: electricity cannot be stored in a cost effective way and therefore only exists for a brief instant between its production and conversion into an end use energy form. • Fungibility: electricity flows through a network accpording to physical laws. Hence electricity cannot be directed from a particular power station to a particular end user. • Quality of supply attributes: maintaining quality of supply is a shared responsibility of generators, network operators and consumers. These fundamental properties distinguish electricity from other market commodities. Therefore, there are many issues, ranging from generator market power to end user equipment, to be addressed in a competitive electricity industry. These issues may not necessarily arise in conventional commodity markets. Hence, either the existing legal framework should be able to address these specific issues or new laws should be introduced to fill the gap. All countries with restructured .electricity markets have legal frameworks covering general business practice as well as specific legal structures in the industry. In Australia National Electricity Market Management Company and The Trade Practices Act of Australia covers general business practice and the National 17 Electricity Law governs the specific issues in electricity industry restructuring in participating states. In the United Kingdom, the Fair Trading Act of 1973 and the Competition Act of 1998 cover general business practices. The Electricity Act of 1989 governs the electricity industry. The Utilities Act of 2000 also is a part of the legislation that governs the electricity industry. This analysis confirms that the legal framework is an essential part of successful implementation of electricity industry restructuring. A combination of general business practice laws as well as electricity sector specific laws i~ · desirable for a restructured electricity industry. .J 2. 7 Conclusion Electricity industry restructuring is an important economic decision for all countries. One of the most critical decisions of electricity industry restructuring is the regulatory framework of the industry. The success or failure of the restructuring process is largely rests on the regulatory structure. ... Countries have adopted different regulatory frameworks. The most common mode of regulation is through an independent regulatory commission such as the Federal Energy Regulatory Commission (FERC) in the United States or the Office of Gas and Electricity Markets (OFGEM) in the · United Kingdom. The regulatory mechanism is dependent on country's existing electricity industry, legal structure, culture etc. Therefore, the most suitable regulatory structures are country specific and should be carefully designed and implemented. Blind adaptation from other countries may cause more problems than answers. A suitable legal framework is an essential part of electricity industry restructuring. A combination of general business practice Jaws and electricity industry specific laws are required for a successful restructurjng process. 18 " IV CHAPTER3 INDUSTRY STRUCTURES 3.1 The four models ~ The four models were chosen because they correspond to varyin,t degrees monopoly, competition and choice in the industry. The models are abstractions and do not describe particular systems. They correspond broadly real electric systems and nature of the structures can be described as follows: • Model t- Monopoly at all levels. Generation is not subject to competition and no one has any choice of supplier, a single monopoly company handles the production of electricity and its delivery over the transmission network to distribution ... companies and/or to final consumers. • Model2 - Purchasing agency. This allows a single buyer, the purchasing agency, to choose from a number of different generators to encourage competition in generation. Access to the transmission wires is not permitted for sales to final consumers. The purchasing agency has a monopoly on transmission networks and over sales to final consumer. • Model 3 - Wholesale competition. This allows Distcos to buy direct from a producer and deliver over a transmission network. Distcos still have a monopoly over final consumers. There is open access to transmission wires. • Model 4 - Retail competition. This allows all customers to choose their supplier. There is open access to transmission and distribution wires. The distribution 19 7770 -1 (delivery) is separate from the retail activity, and the latter is competitive. The models have quite different types of trading arrangements. They require different sorts of contracting arrangements and have different regulatory requirements. They may require different ownership arrangements for the companies operating in the sector. They also have different implications for stranded assets. These dimensions do not define the models. The defining characteristic which distinguishes the models from each other is competition and choice. I In a Model 1 system, no one may buy from an independent generator, so none exists. All final consumers are supplied by the incumbent utility. The first step away from Model I is to introduce competing generators or IPPs (Independent Power Producers). In Model 2, only the purchasing agency is allowed to buy from IPPs, which is why it is sometimes called the "single buyer" model. The design of the power purchase agreements (PPAs) is a major feature ofModel2. In Model 3, Distos (companies which both own the low voltage wires and retail, i.e. traditional distribution companies) are given the right to buy direct from IPPs, but they retain a local franchise over retail customers. The IPP will therefore need access to the transmission network, and there will need to be trading arrangements for the network. The design of trading systems is the main feature in Model 3. In Model 4, retail customers are given the right to buy from an IPP. They can select their suppliers. In this case the trading arrangements and the access provisions may become more complex in practice. Access to distribution networks is required as well as access to transmission. Alternatively, if we look at it from the customer's point of view: In Modell, there is no choice at any level. In Model 2, the purchasing agency chooses which generator it will buy from. The purchasing agency is the wholesaler for any area. The choice may only be exercised when new plant is built if the purchasing agency works by signing long- term contracts with generators. However, it could also purchase spot energy from other generators or from other jurisdictions. In Model 3, Distcos choose whom they will buy 20 from. They can choose to buy from generators or aggregators or utilities or purchasing agencies outside their own area. To do this the Distco needs some form of contract with the transmission company and the generator needs connection to the system. These contacts are commonly called "access to transmission wires". In Model 4, the choice filters down to the final consumer, who may choose to purchase from generators acting as retailers, or from independent retailers, or from other utilities. To do this, "access to distribution wires" is also required. 3.2 Implications for the Structure of Companies I Many of the models will have implications for the structure of existing companies. Some functions will need to be separated to avoid conflict of interest. Large companies do not often willingly break themselves up into smaller ones, but reorganising the industry along the lines of the models described in this study often requires changes to the structure of companies. The restructuring decision involves consideration of the economies of scale and scope that originally led to the creation of integrated companies and which may still be important. However, conflicts of interest, self-dealing, cross- subsidies and market power create problems that offset or overcome the benefits of integration. Increased regulation or breaking up existing companies are common solutions to these problems. Economies of scale mean that larger scale is cheaper than smaller. This used to be the case for generating plant and, as we noted earlier, this was a major reason for the historic monopolisation of generation. Economies of scope mean that different functions can be most efficiently performed by the same organisation. They often occur because of the transactions costs of setting up contracts for the tasks to be done separate I y. Conflicts of interest and self-dealing problems arise when competitors find themselves in competition with the incumbent utility in situations where the incumbent can benefit 21 itself at the competitor's expense, even if the competitor's product is better or cheaper. Cross-subsidisation is possible if a company has a subsidiary in the competitive sector and one in the regulated monopoly sector, particularly if there is cost-plus regulation. The company will have an incentive to load costs on to the regulated accounts. The solutions which have been tried in these situations are: separate accounts, policed by regulators, prohibitions on the incumbent engaging in the problematic activity or divestiture. Market power is the ability of a producer with a large share of th/relevant market to raise the price and keep it there or alternatively to keep competitors out of the market by barriers to entry, including predatory pricing. The remedies for market power include structural remedies, such as breaking up the company. Behavioural remedies such as requiring advance contracting or outcome remedies such as price regulation or profit regulation. 3.3 Economic Efficiency ... Economic efficiency, the traditional concern of economists, is about giving the right incentives to use resources in the way that gives the "biggest bang for the buck" and that avoids waste. The concerns are usually divideo into three types of efficiency: • production and investment (efficient investment maintenance and closure decisions, the best choice of fuel , the right choice of investment type, location and timing, etc.), • usage (consumers get the right signals to use electricity whe.n their value exceeds the cost of production), • allocation (prices should reflect the marginal cost of the resources at different times and locations to ensure that the correct amount is produced, that the most economic producers generate and that production is allocated to the consumers that value it most). 22 ··competitive markets" are generally assumed to have the advantage in that these types of efficiency are achieved simultaneously. For regulated activities, incentives should be carefully structured so that the outcome is similar to the competitive outcome. Regulation must foster contracts, tariffs and trading arrangements that encourage efficient operation of the generators, network operators and customers. Ideally, these incentives would be provided with the absolute minimum • of regulatory intervention. As the structure of the industry is unbund1ed into its separate components, more commercial agreements between companies {re required to allow them to function as an integrated network industry. These commercial agreements must be designed to encourage companies to collaborate efficiently. 3.4 Social Policy Obligations Social policy obligations are such things as demand side management (DSM) and conservation programmes, low-income assistance, ..fuel diversity (which may include subsidies to supplying industries), environmental issues, high local taxes and economic development. These can be categorised as above-market costs which can be divided into two sets: those that are connected with generation (DSM, fuel diversity and environmental issues) and those that are not (low-income assistance, economic development). Whatever the market model under which the industry operates, the ability to impose and collect above-market costs depends upon the ease with which the customers can choose alternatives that do not have such costs attached. It will therefore be difficult to force the competitive functions to absorb above-market costs. However, since the regulated sector is a monopoly these above-market costs can more easily be collected as distribution charges, for customers normally cannot bypass the delivery system. In each of the successive models, the regulated sector gets smaller and the competitive sector gets larger, reducing the scope for social policy obligations. 23 Economic development activities are local functions, and can be paid for locally, through a charge on the delivery of electricity. The same applies to low-income assistance, If the legislature wants, it can mandate the distribution business to pick up the costs. However, the costs of fuel diversity (windmills, nuclear, etc.) have usually been seen as simply high-cost generation. It is unlikely that these will be built under a competitive regime unless they are subsidised. There are non-distorting ways to subsidise these activities, but in a more competitive world it will need to be made explicit. j 3.5 Treatment of "Stranded Costs" Stranded costs are above-market costs, usually of generation, but also potentially of transmission and distribution, which cannot be recouped in a fully competitive market. They are usually costs which the customer is already paying, these become most apparent in Model4. ... 3.6 Pressures for Change Each model has its own forces for stability and its own internal pressures for change. 24 Pa ge 2 5 M O D EL 1 -M O N O PO L Y V er tic al in te gr at io n Sy st em A G en er at or ~ W ho le sa le r/ Tr an sm itt er . . . . . F In te r- U til . tt y sa te s a Po ol Sy ste m B G en er at or • .. W ho le sa le r/ .. . Tr an '-J !li tte r t I i t I D is tc o I I D is tc o t t I I C us to m er I I C us to m er CHAPTER4 MODEL 1- MONOPOLY 4.1 Description of the model Model l is a monopoly model, typically characterised by a vertically integrated system. In any area, one utility owns and operates all of the generating, -plants and the transmission and distribution wires used to transport the electricity! and is responsible for retailing the electricity to the final customer. The utility has a monopoly over production and over retailing in its service area. These service areas may cover a whole country, as in France, a single region or even a town. The model is shown diagramatically in Figure 1. Optional features are separately owned "distribution companies·' that own the low voltage wires and have a monopoly in retailing in their..service territory, but which can only purchase from a single generating/transmission company. The UK was like this before 1990. This is sometimes called "vertical integration by contract". In the case of monopoly the utility generally has an obligation to serve customers, i.e. to provide energy to everyone in the service area, at a tariff price which is regulated to the cost of service, somehow defined. The monopoly over generation may be enforced rigorously, so that literally no one else may generate, or it may permit self-generation, with very limited sales of excess energy to the utility at regulated "buy-back rates". 4.2 Trading arrangements \tfodel 1 does permit trading between similar vertically integrated utilities across an interconnecter, and Model 1 utilities often coordinate their dispatch through pooling arrangements. This can provide back-up, increased security and help reduce costs by dispatching cheaper plant first. However, these Model 1 pools are generally short-term 26 trading arrangements, based on comparisons of very short-run marginal cost. Model 1 utilities typically buy and sell to each other at prices which split the gains from trade. The prices reflect the presumption that trading will be reciprocal, and are not intended to cover the full cost. The utilities own franchise customers pay for the rest of the costs of generation. There is no competition to generate or to build plant, each utility is expected to meet its own load needs. The agreements set up under a Model 1 system include elaborate arrangements to prevent free riding. This is the most usual type of arrangement found in current systems. . ' I 4.3 Transmission access In Model I the question of transmission access only arises as the question of access for traders to cross the network to get to the other side. For example, if France, Spain and Portugal all have Model 1 systems, but France wishes to sell to Portugal across Spain, then the conditions and price for access must be agreed. In the US, this would be called ''wheeling across" a utility's transmission system, ia Europe it is called "transit''. (fn Model I there is no wheeling or transit into a service area, since customers have no choice but the local utility, and no wheeling out, since there are no independent producers.) The issues here are about the traders' responsibility for the overhead costs and whether the transporting utility can capture the rents of the transactions (by charging a price equal to the difference between the cost of the power and its value in the receiving zone). 4.4 Should the Distcos be seperate? In a Model 1 structure, the most usual arrangement is the vertically integrated company. This company owns and controls the generating plant and the transmission and distribution systems, and reaps economies of scale by building bigger plant and covering the territory efficiently. It also can take important advantage of the economics of coordination, especially the coordination of the dispatch of the generating plant. The 27 transmission system operator can command and control the operation of the plant. This ensures not only that the transmission system remains stable, but that the plants are dispatched economically, i.e. they are run in "merit order". from lowest to highest marginal costs. This is the most economic method of dispatching plant since it minimises cost. The structural question which most often arises in Model I is whether distribution should he separated from generation and transmission, and if so; ·what is the optimal j area for the Distco to cover. These questions arise, even in the absence of restructuring, as questions of internal organization, and many companies go through cycles of de- centralisation and re-centralisation as they evaluate these issues. The answer to these questions usually lies in consideration of "economics of scale and scope". Economies of scale and scope are reasons for having a single firm rather than several firms which contract with each other. Their presence would suggest consolidation, but lack of these economies does not necessarily require separating companies, or even creating separate businesses within an existing company. There is ne universal answer to the question of separate distribution companies in Model I, although the economies of scale in distribution seem to run out at a relatively small size. Electricite de France, a Model 1 company which has evaluated this issue more than most, runs its distribution business in a highly decentralised fashion, which would confirm this impression. If a later move to Model 3 is contemplated (where Distcos compete as purchasers of electricity), distribution companies may be separated from generation as a transition mechanism. Several smaller sized distribution companies ·might also be created from large ones. 4.5 Achieving efficiency In Model 1. minimum-cost construction of generation is achieved through a planning process carried out by the utility, the outcome of which generally needs to be approved 28 by the regulatory body or the government. The utility owns and operates the plant, although it may contract out the building construction. The costs of approved plant are passed to the franchise customers through the retail tariffs. Incentives for efficiency mainly arise from the regulatory lag between price settings. In Model I. most risk is usually passed through to customers under cost-of-service regulation. The customer pays for mistakes in investment, changes in demand, unanticipated technological obsolescence, and indeed virtually everything. This reduces the risk borne by the investors in the integrated company, whicfin tum may lead to a reduced cost of capital for investments by the company. However, this can also induce errors into the Construction decision, since the cost of capital for investments as a whole is seen as low when the risk of any given project, and therefore the cost of capital appropriate for that project, may be high. Passing on all costs, so that prices rise when costs rise, gives bad incentives to reduce costs. Various steps, generically known as ··incesative regulation", can improve the incentives by shifting some risk to the owners or operators of the assets. The general notion is that prices should be, at least partially, unhooked from costs. so that there is an incentive to reduce costs. Even in cost-based forms of regulation, slowness of the regulatory process to adjust prices to costs (regulatory lag) can unhook prices sufficiently from costs. Other methods include explicit limits on the ability to pass increased costs on to customers, indexing prices to an independent measure of costs (for example an index of retail prices) or setting a price path in advance. The cost of putting the plant in the ground is a large element in final cost. The decision to build, and its accomplishment on time and to budget, has therefore been the area where there has been most pressure to substitute market mechanisms for the "planning process". This has led to bidding systems under Model 2, and a competitive market under Models 3 and 4. 29 4.6 Social policy obligations One attraction of Model 1, which is carried over into Model 2, but is greatly reduced in Models 3 and 4, is the ability to accommodate social policy obligations. These are outcomes wanted by the government that would not appear in fully competitive markets. These can be divided into two groups. Those related to generation, and those that are not. The former group includes "obligations to supply", environmental regulation of emissions, diversity of fuel sources and subsidies to th~ -coal industry and to nuclear power. Social policy obligations not related to generaton include uniform pricing across areas with unequal costs, rural electrification, discounts for customers who use a large amount of electri city, "lifeline rates" for poor people, conservation programmes and high local taxes. In Model I all can be achieved, but in the later models the generation-related policies come under serious threat. The ability to achieve these objectives is made possible by the monopoly of the utility over its customers, which enables the utility to charge..them the excess costs. High input costs and excess capacity can only be sustained if the customers have no other choice. Discrimination among customers is also feasible, since the tariff can be structured to sell to large users at a different price from small users. Model I utilities are often the agents of so many social policies that they become effectively a tax collector. These policies are supported by above-market prices that the industry can charge because of its monopoly and because demand is so strong for a product which is close to a necessity. 4.7 Implications for asset values and stranded costs Revenues and asset values in a Model I system bear a close relation to accounting concepts of cost-of-service rather than market valuations. In Model I, tariffs determine the utility's revenue. The regulatory body or the government will regulate tariffs to 30 provide a return on assets and to keep prices in line with costs. Provided the tariffs are set at an adequate level, and provided the revenues are collected, the generators will be adequately remunerated. Model I relies upon the franchise customers to pay the capital costs of the plants and ensure an adequate level of profits. Certainly, regulators do not always ensure adequate returns: sometimes they may permit too generous a return. "Regulatory lag·· (slowness to adjust prices to costs) provides some deviation from full cost recovery. However, the regulatory bargain is that the utility gets reimbursed for prudent expenditures. .I Customers in a Model 1 system not only pay all the costs of the utility, they also take the risk of changes in technology which render existing plant obsolete. The customer takes the risk for mistakes made by the utility, if made in good faith. Often the customer also pays for social policy objectives that regulators or the governments deem wise, but that have little to do with provision of low-cost electricity. In return, regulators give the utility an obligation to supply that guarantees customers a supply of power. .... In a well-regulated and well-run Model I utility, the prices are set to deliver an adequate rate of return to stockholders, whether stockholders are private individuals or governments. The asset values (as measured in the stock market - if the company is privately owned) will approximate to the asset values recorded in the books of the company. This will be true even with past mistakes or government imposed social objectives. As long as the form of price control permits the company to recoup an adequate return from customers, it can maintain the value of its assets. 4.8 Pressure for change Model I (monopoly and monopoly service) begins to break down most spectacularly when the marginal cost of competitive generation, or the price that new entrants could charge, is less than the price charged by the utility. 31 !his price may be higher than under competition because: • the depreciation policies of the regulatory regime do not adequately capture technical progress; • past capacity acquisitions have been poor (nuclear plants have often been excessively expensive): • the incumbent's choice of plant must meet social policy objectives but similar requirements do not apply to competitors; . j • large quantities of a fuel (such as gas) may become available at low prices, making a different technology a cheaper option. (This appears to have been a major factor in many cases.) Model 1 utilities also create dissatisfaction by refusing to offer reasonable terms when customers install their own equipment and need back-up provisions. Utilities have sometimes refused to purchase excess energy from self-generators, and have refused access to their wires. The UK had a law requiring <>pen transmission access for independent generators for six years before it implemented a radical restructuring. There were no takers because of the terms offered for transmission by the incumbent Model I utility, the CEGB. Tn the US litigation over access has been extensive and costly. 4.9 Examples of Model l Most countries start with monopolies covering their entire electricity supply industry. Sometimes there is one monopoly for the whole country, sometimes local monopolies. Almost all countries had this form of organisation up to 1980, and most still do. Electricite de France (EDF) owns the entire industry in France. The UK was an example of Model 1 until 1990; the CEGB owned all generation and transmission and there were separate monopoly distribution companies. Italy, Malaya and Japan all follow Model 1. 32 The US had almost complete monopoly until PURP A was passed in 1978. Investor- owned utilities served most of the country and had a monopoly from generation to the final customer. (There were also some publicly owned power companies eligible for tax-free financing, and some separate retail I distribution-only companies that bought from a single supplier.) Despite having multiple generating and distribution companies, some privately owned and some publicly owned. Venezuela also has a Model I system. Oniy utilities may generate, and no one has any choice over whom they buy from an.f sell to. The list of Model 1 countries is almost as long as the list of countries, since the movement to competition is quite recent, and other models are the exception. 4.10 Conclusions Model 1 has been the paradigm for a century, for good reason. This form of vertically integrated organisation has enabled the development of large-scale transmission systems and has enabled introduction of larger plants. These "economies of scale" arguments that were persuasive for many years, and still apply in some developing countries, justified monopoly arrangements. The total monopoly has also allowed subsidies for poor areas, rural electrification, development of indigenous fuels, and other government policy objectives. These objectives may continue to require a monopoly in electric production and retailing. 33 Pa ge 34 M O D EL 2 - C O M PE TI TI O N IN G EN ER A TI O N (a) D is ag gr eg at ed v er si on (b) In te gr at ed v er si on O w n i IP P IP P IP P IP P IP P I G en er at or s / ~ X ~ ' I . ,, • . . W ho le sa le . . . W ho le sa le Pu rc ha si ng . . . . . Pu rc ha si ng A ge nt . . . . I A ge nt 7 1 F ~ , ' · . D is tc o D is tc o D is tc o D is tc o . . , 1F 1 F , 1F I Cu sto m er ! Cu sto m er Cu st om er Cu st om er C us to m er En er gy sa le s F IG U R E 2 CHAPTERS MODEL 2 - PURCHASING AGENCY. 5.1 Description of the model l In Model 2, shown diagramatically in Figure 2, Independent Power Producers (IPPs) are allowed. These may be created from existing utilities by divestiture, or they may be new producers who enter the market when new plant is needed. The IPPs compete to construct and operate plant and carry the construction and operating risk. (This distinguishes this model from a Model I utility that may contract for new plant if it does not have a construction division to construct its own plant.) IPPs sell their output to a purchasing agency. In turn, the agency sells the output on to Distcos that have a ... monopoly over their customers. While Model 2 allows competition in generation, all power must be sold to a purchasing agency, so the purchasing agency is a monopoly, buying the output from the generators. Generators compete to sell to the purchasing agency. This introduces competition at the level of new construction and for generation operation. Generators wi ll typically compete for contracts to supply the purchasing agency. The purchasing agency can in principle discriminate between generators. Either bidding procedures or some other provision will be needed to prevent this. Lower prices will be offered to low cost generators so that they will not get windfall gains. By this way the purchasing agency can pay appropriate ''economic rents" for low cost sources or sources for which consumers have already paid. 35 5.2 Transition mechanisms When a government-owned industry moves to Model 2. existing plant may be sold by tender to private buyers, complete with contracts for power sales to the purchasing agency. Alternatively, a regulator could order a Model 1 utility to purchase new power requirements by competitive bid from IPPs. This method is fraught with potential conflicts. In the US a Model 2 system was introduced by PURP A, which required utilities to purchase at "avoided cost". i 5.3 Trading arrangements In Model 2. the purchasing agency model, generators typically have contracts with the purchasing agency, known as power purchase agreements or PPAs. Normally, these contracts have an availability payment. designed to cover fixed costs, and an energy payment, set to cover the variable costs of generation in order to dispatch the plant. The contracts are called in order of their variable costs of generation to achieve short-term efficiency in dispatch. Economic dispatch requires that the energy payment be designed to match, as accurately as possible, the marginal cost of running the plant. However, setting energy payments to actual costs incurred gives the generators poor incentives to reduce those costs. One solution to this is to track costs closely, but independently of actual costs, by linking the energy payment to an index of costs. In practice there are many PP As with clauses that link energy payments to fuel price indices. Full payment of costs requires that the overheads also are paid, and this is usually done through an availability payment (usually paid for each kilowatt of generating capacity). If the plant is to have an incentive to be available to generate. this payment needs to be linked to the actual availability of the plant. llowever. the availability of a plant is difficult to monitor directly if the plant is not actually running. As a result, we usually 36 see availability payments together with "penalties" for not being available. Plants incur these penalties if the dispatcher calls them to generate, but they are unable to do so. Ideally, these penalties would relate to the market value of energy, so that there arc incentives to be available at the times of highest value. Design of the availability penalty is a major consideration in IPP contracts. ln Model 2, sales from the purchasing agency to Distcos often take place at preset wholesale tariff prices. Efficiency considerations suggest that this tar:iff should follow the marginal costs of the system. Also, the tariff should cover ttfe. total costs to the purchasing agency of purchasing power. Multi-part tariffs, with fixed and variable charge elements are often used to meet these objectives. Variable elements of the tariff can be set to mirror system marginal costs. Fixed components can then be set to recover any remaining costs. The tariffs should then be differentiated appropriately by time of day or season. Retail tariffs would inevitably reflect the cost of purchasing at the purchasing agency's wholesale tariff. .... Such a wholesale tariff allows the introduction of interruptible rates, allowing the purchasing agency to cut off demand, usually from large industrial customers, at times of system stress. These customers usually provide this service in return for lower rates. Interruptible rates offer more opportunity to adjust the system to demand and supply conditions. It is possible to calculate a spot price for this system at the wholesale level, broadly similar to that in the England and Wales Pool price, to provide spot incentives for load management and plant availability. 5.4 Transmission access The question of crossing the system discussed in Model 1 is still an issue here, and continues to be in all the models. The additional issue in transmission access in Model 2 is the question of how to reflect transmission costs in the location of generation and the dispatch of plant. The bidding process for obtaining new plant must allow for the actual 37 and potential transmission constraints and losses and handicap the bids accordingly. Clear terms and prices for transmission access must also be laid down. These terms of access will determine how independent generators are treated if they cannot run because of transmission constraints. For instance, the purchasing agency may guarantee a generator access to the system. If the generator was then unable to deliver power because of a transmission constraint, the purchasing agency would have to compensate the generator. j 5.5 Should the purchasing agency be separate? A structural question in Model 2 is the identity of the purchasing agency. The purchasing agency has to make long-term contracts with generators, so it needs to be credit-worthy. The government or a well-established uti lity is therefore a primary candidate, but either of these creates other conflicts. ... The purchas ing agency should in principle be independent of the owners of the generation, or conflicts will inevitably arise. The agency needs to be seen not to discriminate in favour of its own resources, either in procurement or in operation. In procurement, it might seem a simple matter to