Document of i _ The World Bank FOR OFFICIAL USE ONLY Report No. 4284 SECTOR SUPPORT STRATEGY PAPER ELECTRIC POWER January 1983 Energy Department Power Advisory Unit This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. ABBREVIATIONS AND ACRONYMS CIGRE - International Conference on Large High Voltage Electric Systems ECLA - Economic Commission for Latin America EEI - Edison Electric Institute EPRI - Electric Power Research Institute IAEA - International Atomic Energy Agency ICOLD - International Commission on Large Dams of the World TEA - International Energy Agency IEC - International Electrotechnical Committee IEE - Institution of Electrical Engineers (U .K.) IEEE - Institute of Electrical and Electronics Engineers (IU.S.) OLADE - Organization of Latin American Energy UNIDO - United Nations Industrial Development Organization UNIPEDE - European Union of Producers and Distributors of Electric Energy WEC - World Energy Conference - ii - Foreword The CPS Functional Management Task Force Report of November 1976 directed each CPS sector department to prepare, coordinate with the Regions, and present for senior management review a Sector Support Strategy Paper which would assess the Bank's activities in the sector over the preceding three years, the changes desired, the strategies and plans for bringing them about, and the resulting priorities and relationships among the various CPS "Other Output" activities. This paper is in response to that directive. The statistical data and analyses cover FY78-80; however, significant changes in policy or approach that have occurred subsequently are noted. The paper is not intended as a review of Bank policy in the power sector even though the last such review was made more than a decade ago. It does, however, allude to the need for such a review particularly in light of the changes that have taken place in the past decade. Furthermore, the paper is not intended to be critical of performance in the sector. Bank work in the power sector generally is highly professional and very efficient. The recommendations should serve to make good work even better. Neither is the paper a budget document, although budget-related conclusions are inescapable, partly because power ranks second in ithe volume of Bank lending but eighth in the staff and other resources devoted to it. The August 1980 policy paper, Energy in the Developing Countries proposed an increased lending program in energy,including electric power, but since long- range financing for this program has not been secured the implications that it will have for staffing and programming in electric power have not been taken into account. In accordance with the task force directive, this review is intended for internal use. It assumes that the reader is familiar with i:he Bank's organization and the nature of the policies and issues in the sector. The paper is organized in three sections: (i) power sector issues in developing countries and Bank lending in electric power and how it is programmed; (ii) lending objectives, strategies and issues for the Bank in the sector; and (iii) the resources that the Bank devotes to the power programs. Energy Department January 1983 - ili - Table of Contents Page No. SUMMARY AND RECOMMENDATIONS ....................................... v I. Power Issues in Developing Countries. v II. Issues in Bank Power Operations .vi III. Bank Resources in the Sector. x IV. Recommendations. xii SECTION I INTRODUCTION ............ ........................... 1 A. The Sector in Perspective. 1 B. Power Investment Requirements and Sources of Finance. 2 C. Volume of Bank Lending for Power. 3 SECTION II OBJECTIVES, STRATEGY AND ISSUES IN BANK POWER LENDING. 6 A. Institution Building. 7 B. Financial Objectives. 9 C. Efficiency Pricing .11 D. System Planning, Conservation and Efficiency 13 E. Legal Covenants Governing Bank Power Loans 15 F. Issues in Sector Work .18 G. Sector Lending ............................... 19 H. Issues in Project Justification and Implementation .19 I. Cofinancing .23 J. Rural Electrification .. 24 K. International Interconnection .26 L. Nuclear Power .26 M. Hydro and Other Renewable Sources .28 N. The Environment .31 SECTION III THE BANK'S RESOURCES IN THE POWER SECTOR .... ......... 32 A. Regional Power Staff ............................... 32 B. Regional Staff Time and the Work Program .... ....... 36 C. The Power Advisory Unit .38 - iv- Table or Contents (Cont'd) Page No. D. Policy Work and Research ........................... 39 E. External Relations ................................. 40 F. Professional Development ......................... 41 ANNEX I POWER SECTOR STATISTICS AND PROCEDITRES .44 ANNEX II COMPARISON OF ELECTRIC POWER LENDING AND POWER INVESTMENT .64 ANNEX III POWER SECTOR ISSUES IN DEVELOPING COUNTRIES 65 ANNEX IV GUIDELINES AND RESEARCH PUBLICATIONS .70 -v - SECTOR SUPPORT STRATEGY PAPER ELECTRIC POWER SUMMARY AND RECOMMENDATIONS I. Power Issues in Developing Countries 1. Electric power is indispensable to modern economies. It supplies one-fourth of the commercial energy used in developing countries. IrLdustry in particular depends on a reliable supply of electric power, and demand for electricity grows rapidly where industrialization and national income growth are rapid. In fact, in growing semi-industrialized countries, power growth rates of 20% a year, sustained over several years, have not been uncommon. 2. Even if the generating capacity of developing countries doubles between 1980 and 1990 (which implies commitments of US$400 billion in the decade), only one-fourth of the people in these countries--4% in some of them--will have access to electricity, and per capita annual consumption will be no more than a twentieth of that in industrial countries. 3. Since the rise in energy prices, concern has quite commonly been expressed that the developing countries cannot afford to keep pace with the growth of demand for electricity and therefore should not try. Additions to demand come not only from new connections, however, but also from existing consumers, and though there are various techniques for limiting demand to a certain extent, in practice it is impossible to limit consumers to a given amount of electricity. Once a power system is in operation, planning to meet less than the level of demand will inevitably lead to reductions in its effi- ciency and, eventually, to its collapse. Recognizing this, the Bank's policies towards power development are essentially based on the view that, if prices truly reflect scarcities, demands for power are justified and should be met. The task then is to identify the least-cost way of meeting the demand. 4. The planning of power development has become much more complex in the past decade. Real increases in the costs of construction, equipment and fuels have reversed a long-term decline in the real costs of electric power that lasted until the early 1970s. Changes in the structure of energy costs have rendered most power systems economically suboptimal. The returns on new investments in electric power are now usually lower than those from loss reduction, load management and the re-optimization of systems. The search for the least-cost sources of electricity supply now extends to hydro facilities (both large and small), coal-based plant, geothermal, and supplies from other countries through regional interconnection. Further, since the economic costs of electricity are linked to those of other energy sources available to generate power or to substitute for it, proposed investments in power and proposed power tariffs, must be considered in the context of the energy sector as a whole. It also needs to be borne in mind that since oil price increases - vi - have made electric power the most cost-effective "fuel" for many users, the potential for substitution of oil by electricity, if fully realized, could substantially alter past patterns of growth in the demand for power. 5. Finally, mobilizing the financial resources necessary to keep pace with demand is also more difficult than before, given the rise in investment costs and the higher cost of borrowing, and an even greater premium is placed on efficiency in implementing projects. 6. There are growing signs that power institutions in developing countries need help in dealing with these issues. Though they are far more capable and sophisticated than twenty years ago many of these institutions are having trouble finding enough financial resources and planning and implementing the increasingly complex projects that are required. Some have pursued policies which result in power shortages; and rationing. The economic cost of such policies is very high. These problems are acute not only in small and poor countries, but also in middle income countries with high growth rates and expanding opportunities for borrowing. II. Issues in Bank Power Operations 7. The Bank is the largest international financing institution active in the power sector of developing countries. It has been directly associated with about one-fifth of the total power investment in these countries in 1960- 80, and with half of their investment in hydro development. As total invest- ment has grown, the Bank's share has declined. Since 1956 Bank power lending has remained at around US$1.5 billion a year in 1981 dollars; while its lending made up about 10% of the total power investment in FY76-80, the program currently planned for FY82-86 will make up only about 3.5% of this investment. With the relative decline in its firnancial contribution, the Bank has put a growing emphasis on institution building, to create power enter- prises strong enough to raise funds for expansion without recourse to the Bank. 8. Increasingly, the Bank is being called upon to deal with the new and unique problems posed by the low income countries (Burma, Burundi, Upper Volta, the Yemens) while it continues to assist its mature customers. Cofinancing 9. Mobilizing project financing may be the largest single challenge faced by the Bank and its borrowers in the power sector in the next decade. Funds for power from other aid sources are projected to remain at about the same real level as in 1976-80. Funds generated internally by power enterprises are likely to supply at best a constant share of investment needs, because of the reluctance of governments to pass cost increases on to consumers. Increased use of export credits and larger borrowings from commercial sources are being counted on to fill the gap, but this implies amounts of perhaps ten times those of 1976-80. Concerns over creditworthiness and exposure are likely to limit access to these sources. - vii - 10. Before 1975 only 10% of Bank-financed projects were cofinanced. By 1978-80, two-thirds of them relied on cofinancing. Better mechanisms for identifying and coordinating project cofinancing are needed to avoid costly delays in completing financing plans. One current project has 23 prospective cofinanciers, and plans involving six to eight are not unusual. Paying more attention to cofinancing at the project preparation stage may require some changes in processing procedures. For example, project descriptions in project briefs may need to be expanded so that potential cofinanciers can easily identify the components which are attractive to them. Procurement, in particular, is being affected by the increased emphasis on joint financing, and with the appropriate policy and procedural changes, it can be made an avenue for attracting and allocating cofinancing. Institution Building 11. The Bank's principal objective of institution building in the power sector is to create both the environment and the ability necessary for power entities to manage their own affairs. This involves a variety of activities: planning projects that fit a coordinated development program; instituting well-focused financial policies; encouraging continuity and independence in management as well as advising on other organizational aspects, technology transfer, training and procurement. 12. Over the past few years, institution building has become more diffi- cult. The growth of power systems and their increasing technical complexity and need for public funds, uncertainty about future costs and energy supplies, and strong social and political pressures, in part arising from continuing inflation, have all exerted strains and invited more rigid government controls. In addition, efficiency has suffered because of a high turnover of managers, salary problems, and excessive system losses. In the! face of inflation and a reluctance to pass cost increases on to consumers, real prices of electric power declined in the late 1970s in many countries. Thus, not only are incorrect economic price signals being given to consumers, but power entities are generating less revenue than they should be, given their needs for expansion, while the continued failure to meet demand worsens poor performance. Experience has shown that recovery is inevitably slow and difficult. 13. These issues imply that the Bank should place new emphasis on rigorous and complete analysis of power sector policy, finances and operations, and that it insist on the needed changes which are revealed by this analysis by reasserting conditionality in all its forms. This necessarily implies more involvement in the pre-investment arnd policy formulation stages of the project cycle. More use of engineering loans and credits, technical assistance operations and other forms of pre-investment financing is required to bring utility policy, organization, and systems to the point where borrowers can make efficient use of the larger increments of financing associated with Bank project financing. - viii - Programming Power Lending 14. The increasing complexity of power development programs and the need to develop power sector lending strategies have required changes in the Bank's traditional approach to sector work. Program development work needs to consider alternative resources (e.g. small hydero and other renewables, coal and lignite) and the investment strategies being pursued in other sectors, which have implications for power demand. Considering the likely constraints on the Bank's administrative budget, it is necessary to explore the possi- bility of attracting additional resources for sector work, perhaps as part of the energy assessments program now being carried out by the Bank's Energy Department (EGY). Efficiency Pricing 15. The Bank's current philosophy on electricity pricing can be summarized as follows: prices should be set to cover the long run marginal costs of supply (LRMC) and all of the demand that materializes at these prices should be met, thus ensuring efficient allocation of resources. This implies a rejection of subsidies but at the same time discards overpricing as a rationing mechanism. Nevertheless, it recognizes that some adjustment to strict marginal cost price structures may be appropriate, for example, if social conditions induce a government to introduce life-line rates. 16. The Bank has helped to promote this approach in recent years in con- junction with conventional financial criteria for power pricing. At end-81 over 50 studies incorporating LRMC principles were underway or had been com- pleted, and LRMC pricing had been adopted fully or partially in some 20 coun- tries. For the next few years the most common scheme is perhaps likely to be one in which the overall average price level is determined by financial requirements (even if inconsistent with LRIMC), while the structure of tariffs reflects LRMC. 17. Financial viability is a principal objective in Bank power work and as a result the financial presentation in project appraisals often is confined almost exclusively to this analysis. Because of the significance of efficiency stressed in this paper, efficiency pricing should now receive full emphasis along with financial viability analysis. The operational implications of this change should not be underestimated. Aside from changes in presentation in appraisal reports, and necessary revisions of OMSs and staff guidelines, the staffing structure of the regional power divisions may need to be altered to reflect the corresponding increase in economic work, while financial analysts may need training (or retraining) to deal with this new balance of priorities. Legal Undertakings 18. Bank loans in the sector typically are highly conditioned. Financial covenants establishing revenue targets and setting limits on borrowing are included in virtually every loan. Compliance with covenants requiring specific action, particularly revenue covenants, has not been good, and - ix - disputes with borrowers over financial covenants are frequent. The rate of return test embodied in the Bank's standard revenue covenant is often per- ceived as reflecting a profit motive inconsistent with the concept of supply-ng power as a public service. This has led to frequent confrontations with borrowers over earnings requirements and other issues such as asset revaluation, and has prompted more use of cash-generation covenants, which are much less satisfactory. 19. Perfcrmance indicators and targets could be used more frequently in power loans to stress the importance of operational efficiency, rather than stressing financial results alone. This would require a broader approach to financial analysis which evaluates management and institutional performance as well as the narrower issues of earnings, cash flow and accounts. Financial covenants will continue to be important because they embody practical and testable performance measures. However, stressing their intent rather than their form in discussions with borrowers is a matter of urgency. 20. In add-ition, to reduce the Bank's role as policeman, every appraisal should seek agreement with borrowers on appropriate financial objectives and measlures (for example, incorporation into sector legislation) to ensure they will be pursued. Rural Electrification 21. In countries where rural electrification accounts for a large part of the sector program, as it does in Bangladesh or India, for example, and where the Bank can play an effective role in planning, or by coordinating cofinanc- ing, rural electrification projects are worthwhile. However, such projects are attractive to other donors and selection for Bank financing should be done with special care. Where rural electrification projects directly promote productivity they can be a powerful development tool, andl rural electrification plans should be reviewed as an integral part of sector analyses. Nuclear Power 22. The Bank's 1975 policy paper on nuclear power treated nuclear power as simply one option among those to be considered in developing least-cost prog ramrs. It advocated a more positive role for the Bank in developing training, safety and regulatory programs. This recommendation was not acted on, partly because many developing countries have deferred their nuclear programs but also because financing is commonly available from other sources. Meanwhile, there have been growing concerns about safety, reliability, and public objections to nuclear power. About 15 developing countries can now be considered "nuclear eligible," taking into account that the smallest economically justifiable unit is about 600 MW; should smaller units become economic in the near term, substantially more countries could consider installing nuclear plants. Even though it is very unlikely that the Bank would be called on to finance a nuclear plant, Bank sector work should include reviews of nluclear training and safety programs in nuclear eligible countries. The Bank is not well equipped to deal with these key issues and will likely have to continue to rely on IAEA for assistance. -x Hydro and Renewables 23. Hydro development is proceeding satisfactorily in most countries, at least on large projects. However, because of the large investments required and long construction times, it is becoming evident that many countries can more profitably pursue programs incorporating smaller projects that can be constructed quickly, although not necessarily at lower cost. Conventional hydro-project identification procedures often summarily exclude sites below some arbitrary capacity, usually 50 to 100 MWd. The Bank is conducting a country-by-country study of the status of hydro identification and feasibility work, but regional power divisions will need to play a more active role here by including small hydro studies in project identification and preparation activities. 24. "Mini" hydro (say, projects less than 5 MW) presents organizational problems. Construction programs of as many as several hundred subprojects will be needed in some countries if mini hydro is to contribute meaningfully to energy supplies, but successful strategies for dealing with site identifi- cation, construction and operation in programs this large have not yet been developed despite substantial work by the Bank and other development agencies. Moreover, the initial results of trial programs have not been encouraging. It now appears unlikely that mini hydro can be a panacea for rural energy problems; its more promising role is to supplement or supplant diesel generation in isolated systems, or as supplier to relatively large grids. Similar setbacks have hampered development of wind energy, dendro- thermal (forest and agricultural waste), solar, and other new technologies. Lack of site data is a constraint, and Bank projects could usefully include study components to collect such data. III. Bank Resources in the Sector Staffing 25. Although power is one of the largest sectors in lending terms in the Bank, it is one of the smallest in staff and resources. At end-FY8l only 65 regional operations positions (about 2% of 13ank professional staff) were assigned to power activities, and less than 4 staff years/year were devoted to advisory services, policy and research in the czentral staff complex over the FY78-80 period. The traditional practice in the sector of using small teams-- typically an engineer and a financial analyst---for all project work places a high premium on broad experience. When combirLed with language constraints, these requirements make the typical power division staff (5 engineers, 3 financial analysts, 1 economist) too small. The small number of economists is noteworthy. Despite the growing emphasis on efficiency pricing and other economic issues, the regional divisions do not yet have enough resources to permit economists to play a meaningful role in every operation. To support the growth in the operations program in power in FY82-84, staffing increases of about 20% would be required, but the analysis done for this paper indicates - xi - that doubling the complement of power staff over the next three years would probably be cost effective. Obviously, this is unlikely to happen, and not solely because of limitations on the Bank's administrative budget. There is a limit to the rate at which divisions can absorb new staff, and qualified candidates are scarce. Changes are likely to be evolutionary, through training, some increases in work coefficients as resources permit5 and changes in the allocation of staff. 26. Recruiting power staff is difficult principally because the broad experience requirements are met only by senior level mid-career individuals-- often managers who are not especially attracted to the Bank. Formal training programs in project and sector evaluation to "broaden' otherwise-qualified recruits, and the time to implement this training, will be necessary if future staffing needs are to be met. In the near term adding a central pool of a few specialized staff might increase efficiency, but real improvement will be possible only if additional resources are provided so that missions can consist of several members with relatively narrow technical skills, as in some other sectors. Staff Development 27. Aside from periodic state-of-the-art seminars and the annual audit review discussions, power staff have few, formal opportunities to meet. The Power Advisory Unit (PAU/EGY) has arranged additional seminars and taken other steps to remedy this but much more needs to be done. Also, while adequate opportunities exist for external training, conferences and seminars and other professional development activities, they are rarely attended by power staff since operational duties take priority. This is paralleled by a relative lack of participation in international conferences. Similarly, the number of subscriptions to popular technical journals should be increased so that staff have ready access to these, and the Bank should encourage membership in professional and technical societies by disseminating membership information and sharing costs. The Power Advisory Unit is developing programs to address these issues for power staff, but the issue of professional development needs to be dealt with in the context of a planned Bank-wide action program. The 1981 review of the Bank's training activities identified many of these shortcomings and the background work done for this sector paper reinforces the general conclusions and recommendations of that review. Power Advisory Unit 28. PATJ/EGY consists of two engineers assisted by the finanicial and economic advisers in the Energy Department Participation in the project process absorbs about half of the unit's resources and is necessarily quite selective. As in other OPS/ENI advisory functions, the unit provides functional support for the sector. Because of the relatively few staff in each division, some regions do not have technical review capability, so EGY,IPAU generally contributes to every project operation, often informally. Advice on policy issues, reviews of draft: reports and occasional seminars are principal, though informal, functions of the unit. Consultants, suppliers and general seekers of information on the Bank's power program usually contact - xii - this unit first. Meetings with them, which account for 15% of the unit's work, increased noticeably following the publicat-ion of the 1980 energy policy paper. In the past few years the Energy Department's resources have been directed to energy policy and as a result its work on power. including research, declined from about 7 staff years/year in the mid-70s to 4 staff years/year in FY78-80. Consequently, preparation of policy and guideline material, dissemination of research results, and training and "cross- fertilization" activities in electric power have slowed. Policy Work and Research 29. The Bank is one of the foremost authorities on efficiency pricing in electric power. Most of the small power reseach budget (only 2.5% of total Bank research in FY78-80) has been directed to this issue. Numerous publications and guidelines on this topic have been produced and several seminars have been conducted in borrower countries. In recent years, however, training for Bank staff on efficiency pricing has received insufficient attention and the economic evaluation guidelines are out of date. 30. Despite the volume of research material and staff guidelines on electric power available in the Bank, it is not comprehensive. There is little on the financial aspects of power operaticns and virtually nothing, for example, on techniques of system planning or the role of electricity in the development process. Nuclear safety and small hydro are illustrative of topical issues that deserve more attention. Even so, cataloging, organizing, revising and disseminating the large volume of statistical and guideline material on the sector that is available, is long overdue. The importance of the sector justifies a larger share of the Bank's research and policy formulation effort if conservation, public enterprise management efficiency, evaluation of rural electrification, and simila-r issues are to be addressed adequately. The problem here is not necessarily a shortage of research funds, but rather a shortage of staff time to manage the research. IV. Recommendations Sector Work New issues in the power sector require some expansion and refocus of the Bank's power sector work including: - evaluation of the impact of power sector programs on government fiscal policy and resources (2.15); - xiii - use of sector work to set priorities for Bank lending in the sector (2.44), and identification of opportunities for oil savings through electrification (2.28), training and safety programs in nuclear- eligible countries (2.75), and identification of new resources such as small-scale hydro (2.79). Programming and Preparation Project identification and preparation work should pursue: - opportunities for new style projects including maintenance and rehabilitation, conservation and efficiency (2.25); - a signifi-cant increase in technical assistance and engineering loans and credits to accelerate project identification and to address issues such as operational efficiency (2.44); - use of the preparation process, including project briefs, for "upstream" reemphasis of institutional autonomy (2.19), identification of specific legal undertakings (2.40), and identification and allocation of cofinancing, including prospects for increasing financing from private and commercial sources (2.61), and - innovative financing packages to organize co-financing, and formal incorporation of co-financing into the lending process (2.63). Pricing Policy Tariff analysis should emphasize economic efficiency as well as financial viability through: - relating tariff requirements to long-run marginal costs, supplementing the more traditional ways of ensuring financial viability (2.22); - taking the lead in reconciling differing pricing policies among the various development finance institutions (2.24), and - dialogues with bcrrowers, stressing the intent of the financial covenants rather than their form (2.37), and resisting anti- inflation excuses for not increasing prices (2.18). Procedures Changes in the Bank's approach to power operations make it desirable to: - xiv - modify the tariff analysis in appraisals to include emphasis on economic efficiency (2.22), and standardize the IERR calculation to permit meaningful comparisons across the sector (2.53); undertake more rigorous financial analysis to support innovative financing plans and more responsive measures to deal with shortfalls in financial performance (2.39); review the form of financial covenants and identify non- financial criteria for measuring management performance (2.41); introduce new alternatives into system planning including loss reduction and risk analysis at an earlier stage (2.46); give more attention to project implementation arrangements at the appraisal stage to minimize risk of delays (2.58); analyze rural electrification projects in accordance with more stringent criteria (2.70); add project components to quantify renewable energy resources (2.87); possibly, to review the power sector OMS and prepare a new sector policy paper (3.25) (3.26), as well as updating policy guidelines (3.27) and improving such matters as monitoring of dam safety (2.84) and dissemination of developments in international accounting (2.12), and integrate power sector research into the Bank's formal research program with emphasis on conservation and demand management, rural electrification, price elasticity, and similar issues (3.28). Staffing and Professional Development Coefficients for power sector operations are among the lowest in the Bank. Desirable adjustments include: - changing divisional staffing structures to reflect the increase in economic work (2.23); - flexibility in allocating posts, recruiting, and organization to achieve marginal efficiency improvements (3.07); - budget recognition of the wide variations in manpower coefficients in the sector (3.15); - xv - a cautious increase of joint activities with other development finance institutions (3.18); for the power advisory unit, setting realistic priorities for the additional tasks identified in this paper, and possible establishment of a core group of technical specialists (3.22) (3.23); obtaining division subscriptions to technical journals (3.32); improved participation in international technical activities (3.29), and establishing a "professional office" in PAS to deal with professional development policy and activities (3.37). I. INTRODUCTION A. The Sector in Perspective 1.01 Electric power is indispensable to modern economies. It supplies one fourth of the commercial energy used in developing countries. 1/ Industry in particular depends on a reliable supply of electric power, and demand for electricity grows rapidly where industrialization and national income growth are rapid. Over the past twenty years the generating capacity of the developing countries has expanded sixfold, or at about 9% a year. (This compares with rates of 6-7% in industrialized countries over the same period.) Particularly in fast growing semi-industrial countries, power growth rates of 20%, sustained over several years, have not been uncommon. 1.02 In the decade 1980-90, generating capacity in developing countries is likely to double, despite the high prices of energy and the world economic slowdown. Even after this growth there will still be tremendous potential for increases. Only one fourth of the people in these countries--less than 4% in some--will have access to electricity, and per capita annual consumption will be no more than one twentieth of that in industrial countries. 1.03 An almost universal feature of electric power systems is their persistent growth. Economic slowdowns, civil strife or natural disasters can reduce or halt this growth, but rarely for long, as the experience of Ethiopia, India, Nicaragua, Sri Lanka or Turkey illustrates. Since the price elasticity of demand for electricity is very low in developing countries, even quite large price rises have little sustained effect on the growth of demand. This growth comes not only from new connections bul: also from existing consumers--a fact which sometimes goes unrecognized by planners. Though changes in price and other techniques of load management can limit demand to a certain extent, in practice it is impossible to limit consumers to a given amount of electricity. If capacity is inadequate, the operators of the system can reduce voltage and/or frequency temporarily and reduce peak demand by 8-10%; another reduction of 10% or so can be made by sequentially disconnecting circuits, or shedding loads in other ways. But if the system cannot be managed with these measures the quality of service deteriorates rapidly, with severe consequences for the equipment both generating and using electicity. This means that once a power system is in operation, planning its expansion to meet less than the level of demand will inevitably lead to reductions in its efficiency, and eventually to its collapse. 1/ Taking account of the end-use efficiency of different fuels. The different criteria for comparing energy sources are explained in Annex I. -2- 1.04 Since the rise in energy prices, concern has quite commonly been expressed in the Bank and elsewhere that the developing countries cannot afford to keep pace with the growth of their demand for power and that therefore they should not try. But given the characteristics of power systems just outlined, it is better to take the view that provided prices truly reflect scarcities, demands for electricity are justified and should be met. The Bank's policies in the sector are based on this view. 1/ Thus they stress that where resources are limited, as they almost: always are, it is of first importance to make sure that electricity prices are "right" and that the cost of capital is not artificially low. The task then is to identify the cheapest way of meeting additional demand, ensuring, for example, that the opportunities are not overlooked for high returns on small investments to improve the efficiency of existing capacity. Bank policies also stress the use of electric power pricing to mobilize funds for further investment. 1.05 Since the economic costs of electricity are linked to those of other energy sources available to generate power or to substitute for it, it is necessary to consider proposed investments in power, and proposed power tariffs, in the context of the energy sector as a whole. Strategies for developing power supply also need to be considered together with those for industrialization. Industry consumes about 40% of the electric power in low income countries and up to 75% in middle income countries. As a rule of thumb, an investment in industrial capacity of $100 requires a complementary investment in power capacity of $30. Power investment costs have important implications for the type of industrialization strategy followed and the pace at which it can be implemented. The costs of providing electric power to one job in industry range from about US$3,000 in labor-intensive industries such as textiles and food processing to as much as US$200,000 in heavy industry (petrochemicals, steel, cement). Though a supply of power is obviously not a sufficient condition for industrial development, there are several documented cases (notably in Colombia and Thailand) where economic growth has been constrained by the lack of an adequate reliable supply of electricity. Linkages between the development of power supplies and industry need to receive more attention in the Bank's econom:ic work, particularly when forecasting public investment requirements, and in government planning, where power is still usually treated as a separate sector. B. Power Investment Requirements and Sources of Finance 1.06 As forecast for the 1980 energy policy paper, between 1981 and 1990 some US$400 billion will need to be committed for electric power investments 1/ All references to the Bank in this paper refer equally to IDA. -3- (see Annex I). 1/ Mobilizing resources on this scale is difficult for countries already burdened with high oil bills and inflation. Electric power now absorbs 10-20% of gross domestic investment in developing countries, almost all of it publicly financed. Though power has historically claimed 15- 20% of public sector investment, rising project costs and the expansion of systems are pushing this share higher--it has reached 50% or more in countries undertaking major industrial programs based on hydropower and other capital- intensive energy resources. 1.07 Assuming that financing from multilateral and bilateral sources remains at 1976-80 levels in real terms, one financing plan for the investment requirements in 1981-85 would be: US$bn % Multilateral 16 9 Bilateral/Official 24 14 Suppliers'/Buyers' Credits 30 18 Private Commercial 40 24 Own Sources 60 35 170 100 In this plan, suppliers'/buyers' credits and private commercial financing are assumed to be the balancing items, providing amounts several times larger than hitherto (US$6 billion in 1976-80). While there are good prospects for substantial increases in financing from these sources, concerns over credit- worthiness and exposure make it unlikely that the full amount shown here would be forthcoming. Funds, from domestic sources are also unlikely to reach the level shown, because governments are often reluctant to pass power cost increases on to consumers or indeed to encourage power utilities to make a profit. Making up for these shortfalls through private borrowirLg, now at harsher terms, will exacerbate the already serious liquidity problems. C. Volume of Bank Lending for Power 1.08 Electric power is probably the most "traditional" of the sectors in which the Bank lends. Over the past 30 years, power has received about 18% of 1/ Energy in Developing Countries, World Bank, August 1980. The distinction between commitments and disbursements is particularly important in power, because of the long disbursement periods. Because of growth rates, commitment requirements will exceed disbursements by one-third or more in any given year. This paper uses commitment estimates rather than disbursements because the former better reflect financing needs. -4- the Bank's lending, for some 380 projects in 83 countries. 1/ Except for a decline in FY72-75, when resources were being shifted into agriculture, average lending for electric power has been remarkably constant at about US$1.5 billion a year (1981 dollars) over the past 25 years, and is projected to remain so. 1.09 In FY78-80 63 electric power projects totalling US$4.9 billion were approved, accounting for 16% of Bank lending. Table 1.1 gives a breakdown of this lending by function. These totals exclude lending for electric power components of projects not managed by the regional power divisions; in 1978- 80, 24 other projects contained US$72 million in lending for such components in multi-purpose irrigation, rural development,, and tourism projects. 1.10 Although the reference period is short, Table 1.1 illustrates the longer-term trend of Bank lending for electric power, with continued emphasis on hydro, a movement away from oil- and gas-fired thermal generation towards coal, some involvement in new sources (geothermal), randomly distributed but significant activities in transmission and dist:ribution, and a small component for rural electrification. This distribution of lending roughly parallels the investment requirements of the developing countries except that distribution, which historically has absorbed about 40% of total investment, plays a smaller role in the Bank's lending program because it is more amenable to self- financing. 1.11 The number of projects dealt with (a measure of staff requirements) stayed relatively constant at 10-15 until 1976 when it increased to 20. There has been a gradual decrease in average loan size in real terms, while individual project size has increased substantially. During the review period, total project costs were four times the amount lent by the Bank, while cofinancing was equivalent to 60% of that amount. 1.12 The country mix of borrowers changed significantly in the period, from a rather stable group of old customers (Colombia, India, Thailand, Turkey) to one representative of countries at both ends of the income spectrum. Increasingly, the Bank is being called upon to deal with the new and unique problems of power development in low income countries (Burma, Burundi, Upper Volta, the Yemens) while it continues to assist its mature customers. 1/ A book on the Bank's experience in lending for electric power is being prepared by J.H. Collier. Publication is tentatively scheduled for 1983. Much of the statistical material in this section is from data collected by Mr. Collier. -5- Table 1.1: ELECTRIC POWER LENDING, FY78-80 1/ FY78 FY79 FY80 Total No. of Nof No.of lb.of Components $m GCponents $m Components $m CAoonents $m % Hydro Generation 6 348 4 183 10 783 20 1,314 27 Thermal Generation, Coal 3 305 3 495 4 837 10 1,637 34 Thermal Generation, Oil/Gas 2 97 4 243 1 52 7 392 8 Thermal Generation, Geothermal 0 - 1 9 1 40 2 49 1 Transmission and Distribution 4 256 8 191 12 588 24 1,035 21 Rural Electrification 5 140 3 234 2 92 10 466 9 Total Lending, Current $m 1,146 1,355 2,392 4,893 100 Total Lending, 1980 $n 1,339 1,458 2,392 5,189 No. of loans/projects 19 19 25 63 1/ Excludes US$72 million of powr lending nanaged outside the regional power divisions. Future Bank Lending for Power 1.13 An indicative 5-year lending program consisting of 135 power projects was developed during the preparation of the 1980 energy policy paper. The program would require commitments of US$11 billion (current dollars) in 1981- 85 (US$13 billion in 1982-86), in order to maintain the 1975-79 lending in real terms. As of mid-82, the level of commitments planned was US$8.1 billion in 1982-86. 1.14 Bank financing of sector investment needs would decline from about 10% in 1976-80 (already down from 30% in the 1960s and early 1970s) to 5% or less in 1981-85 under the expanded program, and to 3.5% under the current lending program. For the Bank to enlarge its role would probably mean increasing the number of power projects, or increasing its financial participation in individual projects or country programs, beyond what might be considered prudent. It should also be borne in mind that certain countries' reluctance to accept Bank policy recommendations has recently caused them to forgo or defer Bank financing. In any case, it is unrealistic to expect firm project-specific lending programs to be developed even 2-3 years in advance. 1.15 Annex II summarizes the Bank's recent and proposed amounts of power lending to individual countries, while Annex III shows the areas of emphasis in Bank power lending, country by country. The staffing implications of the lending program are dealt with in Section III. -6- II. OBJECTIVFS, STRATEGY AND ISSUES IN BANK POWER LENDING 2.01 OMS 3.72 identifies the principal objective of Bank involvement in energy, water and telecommunications projects as helping to provide the basic infrastructure required by directly productive sectors. In so doing, the Bank expects: (i) to achieve institutional improvements; (ii) to mobilize local resources for expansion, through appropriate pricing; (iii) to improve planning according to various least-cost methodologies; and (iv) to assist in organizing cofinancing. 2.02 In essence, these objectives in electric power development have been altered little over the years. However, certain changes in approach have been made in response to the developments of the 1970s. Some of these changes are only partially accomplished; the issues that they pose are discussed in the later sections of this chapter. 2.03 First, even more emphasis is now placed on institution building-- creating viable power enterprises--than on transferring resources. Efforts to strengthen power utilities stress the desirability of financial and managerial freedom of action, improvements in financial performance, the establishment and improvement of accounting systems and procedures, recruitment and training of local staff, and improvements in internal organization, project execution, long-term planning capability and management reporting systems. 2.04 Second, since the early 1970s, real increases in the costs of construction equipment and fuels, combined with the high cost of developing new hydro sites, have reversed the previous long-term decline in the cost of electric power. Revenues from price structures based on historical costs are adequate neither to meet the financial needs of power sector entities nor to give proper economic price signals to consumers. Recognizing this change, EGY has gradually sought to introduce tariff-making principles based on the long- run marginal costs of supply (LRMC) 1/, to supplement financial analysis designed to ensure that power entities are financially viable. The economic 1/ As used in the power sector, LRMC refers to the long-run marginal cost of supply from the power system, not the cost of supply from an incremental generating plant or project. Like historical costs, LRMC have a structure and vary by season and time of day. Thus, under a LRMC pricing system, average unit (kWh) costs will vary depending on the characteristics of the type of activity being supplied (e.g. irrigation pumping, steel production, residential consumption). - 7 - justification for LRMC-based prices is straightforward: if prices reflect true costs, the consumption that materializes at these prices automatically is economically justified and the resulting demand should be met. l/ This implies a rejection of subsidies but at the same time discards overpricing as a rationing mechanism. 2.05 Third, much more emphasis is now placed on "upstream" sector work. With the changes in the structure of energy costs, the planning Environment today is very different from ten years ago. Investment in power needs to be considered in the context of the other energy resources available to the economy, and of the prices which reflect the real economic costs of these resources. Not only have the prices of energy sources changed in relation to one another, so that it is now worth exploiting certain resources (e.g. certain hydro sites) previously considered uneconomic, but the escalation of construction costs has led to changes in the profitability of investments in new power capacity versus modifications to existing plant. Further, the increasing importance of conservation and of basing energy prices on real scarcity values makes it necessary for power planners to analyze energy demand --and particularly the implications for demand of the industrialization strategy being pursued--in much more detail than before. 2.06 Fourth, the increasing complexity of its power operations places new demands on the Bank's staff. Over the last ten years, planning and evaluating power operations has fast become more technically complex, as changes in relative prices have speeded the introduction of new technologies and new developments in conventional ones, and encouraged the exploitation of marginal hydro sites. This expansion in the range of options greatly increases the Bank's needs for staff with up-to-date technical competence. Further, the Bank has diversified into such new areas as rural electrification and modification of existing power capacity, which not only are technically demanding but also strongly emphasize the need to develop better tools of economic analysis. A. Institution Building 2.07 Perhaps the most important lesson the Bank has learned from its power lending is that the costs of inefficiency in the sector are high and the best means of ensuring efficiency is through responsible and reasonably autonomous institutions. The principal objective of institution building is to create both the environment and the ability necessary for power entities to manage their own affairs. 2.08 For many years, Bank policy has stressed the discipline of financial viability and self-management in order to build institutions that are 1/ See, for example, "Electric Power Pricing Policy," World Bank Staff Working Paper No. 340, July 1979. able to raise capital from ordinary commercial sources. In this sector, management should be delegated enough authority to hold it responsible for delivering electric power efficiently at an acceptable price. This is now more difficult than in the past, partly because governments increasingly view the power sector as simply another public service facirig the same budgetary and staffing constraints as other government functions. 2.09 For example, the Bank prefers to lend directly to the power entity which is to carry out a project; when this is not possible (e.g. for IDA credits), a direct contractual relation is established through a Project Agreement or similar vehicle. In the FY78-80 review period about half of the Bank's power loans were made directly to beneficiary agencies, compared with FY75-77 when 65% of the loans were made to such agencies. Though this shift is not statistically significant, it is consistent with the trend away from autonomy. Governments have occasionally objected to the Bank's lending directly to beneficiary agencies and in most of these cases the Bank has acquiesced. Provided the direct contractual relationship is maintained, there is little short-term disadvantage in this. It does, however, raise questions about the long-term objective of Bank involvement in the sector. A reaffirma- tion of the value of self-management--as well as increased efforts to deal with issues such as public-sector salaries on a country basis--is needed to restore efficiency and to help borrowers meet the large resource needs of the sector. 2.10 When the beneficiary agency is not the borrower it is necessary to agree on some kind of onlending arrangement. Public utilities are not considered "commercially-oriented project entities" in the context of OMS 3.81; therefore, unlike DFCs or similar entities, on lending to power entities at Bank rates usually is accepted without a risk or guarantee fee since the ultimate beneficiaries are. the power consumers and in most cases the revenue objective is set independently of borrowing costs (see OMS 3.72). If there are private shareholders who may benefit from the concessional element in Bank terms, or in accordance with Government policy in the sector, a lending fee or other premium may be charged. Alternatively, in cases where a government- owned utility is facing severe cash flow problems, onlending at low interest, or in rare cases transferring the funds as equity, may be justified as long as it does not result in hidden subsidies or prices less than LRMC. Responsibility for the exchange risk (it should rest with the beneficiary) is also relevant. 2.11 For a power entity to be viable, it needs both adequate finances and competent management and staff. A real challenge facing the Bank is to develop strategies for borrowers to achieve competitive salaries, incentives for efficient operation, and planning based on sound economic principles. 2.12 To improve the financial management of power entities it is often necessary to improve their standards of accounting, costing and audit. The accounting profession is not well developed in many borrowing countries, yet the financial test requirements associated with Bank loans demand a relatively high level of accounting sophistication. The issues that this poses are not peculiar to the power sector and they should be addressed, at least generally, -9- as part of a wider Bank effort to encourage professionalism in accounting. The International Accounting Standards Committee in which the Bank is an active participant, has made substantial progress in this area. As in the technical and engineering fields, enhanced efforts to keep Bank staff informed of developments in international accounting standards, and more atl:ention to disseminating lessons learned from case studies, still would be worthwhile. As part of this latter effort, the EGY financial adviser is attempting to keep track of Interesting lessons learned in Bank work and to ensure that power sector financial analysts are made aware of these. 1/ 2.13 The efficiency of public sector enterprises is the subject of a recent Bank study, but because this study focuses on non-revenue earning entities, it has not received much attention from power staff. However, lessons from these studies can be useful in power, especially in centrally- planned economies, (Burma, Syria, Algeria, China as well as Romania, Yugoslavia) and developing techniques to measure efficiency in such instances is an immediate task facing the advisory staff. In addition, a comparative data base is needed to permit meaningful use of the various monil:oring and performance indicators that have been used sporadically In the sector since about 1976. Although loans have conditions requiring collection of data on staffing, system losses, plant availability and similar nonfinancial indicators, there is no way of telling whether the performance is good or bad, other than observing the trends of the indicators. EGY/PAU is sponsoring a study in FY82/83 which is intended to develop criteria and establish compara- tive guidelines for common indicators, using data from a broad sample of power institutions in developed and developing countries. B. Financial Objectives 2.14 For many of its borrowers in the power sector the Bank plays a unique role in setting financial objectives and advising on tariffs. The criteria of revenue adequacy which the Bank uses to promote financial viabililty are not infrequently seen as inappropriate by borrowers, who feel these are more applicable to investor-owned companies than to government-owned entities providing a public service. Given the sector's needs for investment funds, however, it is essential that power utilities achieve sound financial management both to generate funds for expansion so as to reduce the burden on government budgets, and to give potential lenders confidence in them. 2.15 The Bank is also concerned with power pricing policies as they affect other sectors of the economy. When relevant, it is essential that the impact of projects on the government budget, fiscal policy and pricing be evaluated as a routine part of project appraisal and sector work. This is particularly important in cases where the power program accounts for a significant share of 1/ One planned study concerns accounting practices in centrally-planned economies and translation of accounting practices in such countries into formats that can be analyzed using conventional Bank financial presentation. The results would be incorporated in a sector-specific staff guideline. - 10 - public sector investment. At present this analysis is seldom done, since the regional power divisions do not have the appropriate staff and resources. Programs staff can assist; in fact, greater participation by country economists to relate sector financing to country macroeconomic policy is necessary to provide a better basis for discussing financial objectives. Whenever possible, borrowers/governments should be encouraged to participate in this work as a necessary foundation for a constructive dialogue on the sector's autonomy and financial objectives. This approach is essential to achieving a common understanding on these issues. 2.16 Coupled with the reduction in the autonomy of power utilities, governments have become increasingly reluctant to raise power prices to reflect rising costs. An extract from a recent appraisal report is illustrative: "Tariff-setting stopped working well in the early 1970s. While (the tariff) law was not repealed, scant attention was paid to it. The Electricity Law provided only vague guidelines on what the utilities' return was intended to provide... High (staff) turnover, much higher inflation than had been the case in the 1960s, and political pressure to keep the rates of the recently-nationalized utilities low all combined to produce unsatisfactory operating results." Power utilities in many developing countries let their prices decline in real terms in the late 1970s, even though energy costs were rising. Thus, not only are incorrect economic price signals still being given to many consumers, but these utilities are generating much less revenue than they should, given the resources needed for expansion. 2.17 Probably no single issue has created so mnuch dissension as pricing in the recent past, and the problems seem to be increasing. They are often made worse by the relative sophistication of the sector, which permits financial undettakings to be defined in precise, quantitative terms, to be reflected in the legal covenants, and later tested to see iwhether they have been ful- filled. Unfortunately, because of the usual format of the revenue covenants in Bank loan agreements (see Section E below), the underlying economic and resource mobilization objectives are often obscured by disputes with borrowers over arcane issues such as asset revaluation indexes. 2.18 Energy pricing is usually politically sensitive, and dealing with this sensitivity requires careful preparation including analysis of the structure of consumption and of the impact of price changes on various consumer classes. Often this will reveal that the major adjustments fall on industry or other government-controlled sectors, while the impact on household consumers can be mitigated. Because in most indlustries the cost of electric power is only a small percentage of the final product cost (3% or less) governments' fears about the ripple effects of electricity price increases are generally exaggerated. Furthermore, the alternatives to electricity price increases usually are inflationary. Stressing these facts is essential in Bank-borrower dialogues on this issue. - 1i - 2.19 Finally, given the urgency of improving efficiency in the power sector of most borrowing countries, and the Bank's declining financial contri- bution to the needs of the sector, it may be useful to reexamine the Bank's role in power lending--particularly its role in establishing financial objectives. Increased emphasis could well be placed, particularly during project preparation, on the promotion of appropriate policies through legislation and regulatory procedures, rather than on strict financial covenants, and on strategies to improve the management of agenacies in the sector. Here the intent should be to achieve agreement in advarnce among the government, the utility and the Bank on efficiency improvements, pricing and the mobilization of resources. Surh systems obviously require a strona government administrative structure to function properly--not always preset in developing countries--hence the need to be selective in identifyi- measures to be used to supplement the traditional financial performance objec- tives. In this regard, innovative approaches such as the French "Program Contract" system 1/ deserve more attention, and EGY/PAU will endeavor to alert staff to such opportunities in Bark projects. C. Efficiency Pricing 2.20 Despite its obvious relevance to economic decision making and resource allocation, the full adoption of LRMC pricing has been slow, due to practical problems of implementation as well as a reluctance to mcve away from prices based on actual costs. Other reasons include a legitimate concern ,about applying LRMC to extensions of non-optimal systems, and the sectoral distortions that can result when LRMC is not applied throughout an economy. The major obstacle, however, is that governments are wary of the price increases thought to accompany the introduction of such polIcies. This perception is not necessarily valid. For example, in most cases, provided assets are properly revalued and no major structural changes in the sector are taking place or are foreseen, the conventional rate of return objective can be a reasonable short-term substitute for LRMC price levels. Questions are often raised as to the distribution of the cash surpluses ("economic rent") which may arise if full LRMC pricing is adopted. Although in practice these surpluses seldom materialize because of the large financing needs of the sector, taxation measures or other methods may be required to capture these funds. 1/ A system whereby the public agency periodically agrees with the government on specific objectives to be attained and the resources to be used. The set of objectives then becomes a contractual obligation and the agency is free to function autonomously within the framework of the program in order to realize those objectives. See Accelerated Development in Sub-Saharan Africa, World Bank, 1981, p. 39, for a discussion of this mechanism in Senegal. - 12 - 2.21 By the end of 1981, over 50 tariff studies incorporating marginal cost principles had been completed or were underway in developing countries, many with Bank participation, and these had resulted in full or partial incorporation of LRMC principles in tariff structures in 20 countries, with reportedly beneficial results. Furthermore, in an increasing number of countries (at least six), the financial requirements of the sector dictate that prices be set above LRMC (Colombia, Malaysia). Although this is not an economically defensible course, limits on local borrowings and similar constraints may leave no option. Adjustments to strict LRMC prices are not precluded. For example, social conditions mnay induce a government to introduce life-line tariffs. In general, however, such measures should be avoided because power tariffs are poor vehicles for income redistribution. Where subsidies are considered necessary, it is preferable that they be visible and direct. For example, government support for an aluminum industry is better channeled through direct subsidies than through underpricing the electric power on which the industry depends. 2.22 The difficulties noted above have led most borrowers to adopt a com- bination of tariff levels based on revenue requirements and tariff structures reflecting LRMC. This may be a satisfactory short-run approach, but it is not an adequate response to the long-run changes taking place in the sector. Accordingly, marginal cost analysis should now receive full attention in appraisal report discussions of tariffs and their adequacy, to supplement the financial analysis which is needed to establish conditions needed for finan- cial viability. 2.23 The operational implications for the Bank of advocating efficiency pricing should not be underestimated. The discussion of power tariff issues in appraisal reports should be related to efficiency pricing objectives, rather just to financial analysis. OMSs will be reviewed to determine whether revisions are needed. Other necessary measures may include shifts in divi- sional staffing patterns to handle the corresponding increase in economic work, and supplemental training courses for financial analysts to bridge the gap between economics and finance. Since the LRMC concept is difficult to express in legally-testable covenants, the Bank will have to continue to rely on the traditional financial covenants to convey understandings on LRMC pricing. 2.24 In the development community the Bank has been the principal pro- ponent of economic pricing and sound financial olbjectives for project entities in developing countries, but recently the regionial banks and other development institutions have paid more attention to pricing and in some ca'ses have advoc- ated policies that are not consistent with those proposed by the Bank. The Bank should take the lead in reconciling these conflicting pricing policies and should promote the adoption of a uniform tariff-making philosophy in the developing countries. The proposed finance task force (para. 2.41) should decide how to approach this. In the interim, Bank financial analysts should continue to explain the Bank's position on financial objectives and pricing when meeting with their counterparts in other development institutions. In country-specific cases, special visits for this purpose are worthwhile and should be planned for. - 13 - D. System Planning, Conservation and Efficiency 2.25 The shift in the structure of energy costs over the last few years rather suddenly rendered most power systems economically suboptimal. Studies indicate that the highest rates of return on funds available for system investment are likely to be realized from expenditures on loss reduction and rehabilitation, load management, plant efficiency improvements, and better operating procedures. It is important to note that, because costs of constructing neT generating capacity have increased almost as rapidly as oil prices, the monetary savings due to a reduction in demand (kW) and hence in the capacity that needs to be built can be as large as the savings due to reduction in energy losses (kWh). Thus, the potential for improving effi- ciency should be investigated even in countries that have systems based on low-cost energy sources (hydro, geothermal, mine-mouth coal). 2.26 linfortunately, such investments fall outside the usual purview of system planners, and feasibility studies investigating alternatives for system expansion seldom fccus on the prospects for meeting new demands through more efficient use of existing facilities. 1/ Moreover, because such investments usually come under the operating budgets of the power entities, they do not gain a place in the portfolios of the planning ministries or other government core agencies and thus they rarely feature in the project list discussed with the Bank at the programming stage. Since programs staff are those most often involved in these discussions, they should take special care to ensure that such project opportunities are not overlooked. 2.27 Conservation projects are often difficult to prepare because they lnvolve changes to existing facilities, and lack the neatness and glamor of new development. Furthermore, they usually blend investment (rehabilitation, upgrading, retrofitting) with procedural changes and demand much from managers, who are often in short supply in developing countries. 2.28 Most power investment programs already have been adjusted to reflect the increases in oil prices and as a result the share of oil in the generation mix will decline from 25% in 1980 to less than 10% by 199C. Further reductions may not be worthwhile because much of the oil is used by peaking stations and to support hydro capacity in dry years. Substituting other sources for oil in such uses generally is not economic, even at today's prices. Furthermore, conversion of existing capacity to other fuels, such as coal, is not technically feasible unless the plants were originally designed 1/ For example, recent advances in long-term and short-term hydlro-meteoro- logical forecasting appear to offer good prospects for enhancing system dispatching and operating efficiencies in the mixed hydro/thermal systems typical of most developing countries. However, the utilization of these advances requires, first, an awareness that these new techniques exist and, second, the capacity to implement them. - 14 - to use coal--a rare situation in developing colntries. Opportunities for saving oil do exist in three areas, which deserve more attention in sector work and project preparation: (i) Most system planning is based on adding new capacity as demand grows, but it still may be economical to advance the construction of new coal and hydro plants to replace oil plants. Where a simulation model is available for planning, this option should be among those evaluated. (ii) Oil-fired generating plants can be made more efficient by improving operating procedures, and retrofitting with new controls, fan drives, and so forth can yield large annual savings. (iii) If sufficient hydro capacity or, in some cases, coal is available, certain oil-based equipment in industry can be replaced with electrical equipment. Bank staff should seek out opportunities for such replacement either directly or in conjunction with studies in other sectors. 2.29 To date, most of the emphasis on conservation in Bank power projects has been on reducing the proportion of energy generated that is "lost or unaccounted for". Little distinction has been made between reducing technical losses (which improves the system's efficiency and reduces capacity requirements) and reducing financial losses (through better metering, accurate billing and the elimination of theft). Although both present serious problems, only technical losses are considered true economic losses since it is assumed that the energy which is stolen or incorrectly metered is used pro- ductively, even though the utility may lose revenues as a result. The reduc- tion of technical losses should receive first priority in any program to improve a power system. Other losses should not be ignored, but procedures for dealing with these through improvements in metering and billing practices, criminal legislation, disconnection policies and other measures are relatively well developed and are often addressed through specific conditions attached to Bank loans. 2.30 The savings to an economy from technical loss reduction can be quite significant. For example, assuming that technical losses are 10% higher than optimal in Pakistan, the annual cost of these losses is about US$50 million. An estimated two-thirds of the developing countries have losses that are higher than optimal but very little is being done to deal with this problem. Although usual "rules of thumb" consider losses below 15% as being quite acceptable, the latest planning criteria in Europe, for example, call for losses of 3% or less in distribution systems. Investments to reduce technical losses to 8% or less are probably economic now in most systems. Current research sponsored by EGY/PAU aims to improve the methods for determining technical losses and to establish priorities among investment opportunities in various parts of the system (reconductoring, power factor correction, substa- tion equipment, MV feeders, etc.). -15 - 2._3 1 In FY78-80, 28 projects (45%) included components which will improve system efficiency, but in only five of these was efficiency improvement iden- tified as a specific objective. This indicates that this criterion had not yet taken its place alongside capacity additions and system expansion in the project identification and selection process. Bank power staff are now much more aware of the need for efficiency and such components are now appearing more frequently in projects. Even so, more formal procedures for identifica- tion and assessment of conservation opportunities, through efficiency studies as part of project preparation, and for the inclusion of operations audits in Bank projects, would be worthwhile. The Bank is executing agency for a UNDP program to do quick audits of power systems to identify loss-reduc:ion proj- ects. If successful, this program may be expanded to cover tup to 60 countries and could serve as a significant source of projects suitable for Banik financ- ing. For its part, EGY/PAU will place more emphasis on identifying and dis- seminating state-of-the-art information on demand management, retrofitting and similar techniques. The proposed research program (Annex I) includes compo- nents to support this effort. E. Legal Covenants Governing Bank Power Loans 2.32 Typical conditions attached to the Bank's loans for electric power are described in OMS 3.72. These are imposed through three cate!gories of legal covenants: standard, project-specific and financial. 2.33 Standard covenants include those on performance and records (due diligence, sound practice, plans and specifications, access to property, records of expenditure), insurance, procurement, land acquisition, etc. Although these covenants may vary slightly in substance from loan to loan, the undertakings reflected are too standard to be mentioned in most appraisal or president's reports. The content of the standard covenants continues to evolve but, aside from a few curious anomalies (e.g. the project insurance covenant covers goods, but not works), they provide the Bank with a set of undertakings that covers most exigencies. Project-specific covenants also tend to be quite standard (covering completion reports, dam safety, con- sultants, environment studies, training programs, management and accounting systems) but their importance or uniqueness usually warrants mention in the appraisal report. Financial Covenants 2.34 Virtually all loans for electric power include covenants pertaining to financial performance. A covenant specifying a rate of return on net revalued fixed assets is the most common--60% of FY78-80 projects had cove- nants of this type--and is the preferred form. Covenants specifying a con- tribution to investment or a level of cash generation (so that a specified percentage of capital expenditures can be met from internally generated funds) are also common (20% of FY78-80 projects). 2.35 The rate of return covenant remains technically the best available instrument provided that: (i) it is soundly constructed (e.g., the assumed - 16 - opportunity cost of capital is appropriate to the country concerned, and no allowance is built in for inflation when assets are to be regularly revalued), and (ii) the government, where it is not the borrower, is kept fully involved in the formulation of the tariffs necessary to sustain the covenant's oblec- tives. 2.36 Partly because some borrowers see rate of return covenants as embodying a profit motive, inconsistent with providing electric power as a public service, cash generation covenants are regaining favor. These covenants address more directly the need to raise funds for investment and are also seen as easier to justify to budget-conscious government officials, and (wrongly) as avoiding the messy issues associated with asset revaluation. Despite these reported advantages, cash generation covenants have many deficiencies: (i) they are vulnerable to small changes in operating costs, especially in mixed hydro/thermal systems; (ii) they result in large annual variations in revenue requirements in systems having large annual changes in investment (the usual case) and thus demand some form of averaging or shortfall recovery arrangement which is inherently difficult to monitor; and (iii) they may induce borrowers to defer needed investments, so as to make their revenue requirements appear smaller than they are. For these reasons the rate of return covenant should continue to be preferred. To some extent, the principal problem with the rate of return covenant is its name and a simple semantic change may provide a solution, especially considering the closer link to LRMC pricing which it permits. 2.37 Disagreements with governments over compliance with revenue covenants indicate, in most cases, a failure to evolve an agrreed pricing strategy before processing the loan, rather than the selection of an inappropriate form of covenant. It is important, in discussions with borrowers, to emphasize the principles of efficiency pricing and resource mobilization and to stress the intent of the revenue covenant rather than its form. 2.38 In addition to a revenue covenant the financial conditions of power loans usually include a debt limitation covenant. The preferred form requires the borrower to undertake a debt service test and to consult with the Bank before incurring any debt that would cause his forecasted debt service coverage in any future year to fall below some specified value (usually 1.5). This test is the most direct, comprehensive and effective means of ensuring that a borrower will be able to meet his long-term debt commitments. It is essential that it be enforced to bring to the mutual attention of the Bank and the borrower any weaknesses in the latter's debt service capacity and to enable any necessary corrective measures to be taken. 2.39 A 1977 study indicated non-compliance with financial covenants in half of the power projects observed and a similar survey at end-FY81 indicated that two-thirds of earnings covenants were not being met. Waivers are frequent, but formal modifications are not made as often as may be justified, given that conditions can change radically during the life of a project. The Bank's response to defaults understandably varies among countries and regions, but, in general, more rigorous financial analysis emphasizing resource mobilization, financial viability and economic pricing is necessary to - 17 - formulate responsive action plans in such cases. To the extent possible, financial analyses undertaken during appraisal should anticipate the need for modifications in the requirements by incorporating studies of sensitivity and risk. When there are large uncertainties these studies should identify alternative requirements and courses of action to be followed if the assumptions used in the base analysis do not materialize. 2.40 In power as in other sectors, the procedures for formulating legal covenants could be improved. In particular, project briefs and issues papers should devote more attention to the scope and intent of anticipated legal undertakings to ensure that these are not overlooked or misinterpreted in the later stages of loan processing, and negotiating strategies should focus on the purpose of the covenants rather than on their form. Typically, Bank lawyers become involved too late in the appraisal process to allow the regional staff or sector advisers a meaningful review of the draft cove- nants. A more serious issue is that the borrower is not well briefed in advance, at appraisal or before negotiations, on the scope of the covenants, the form in which these will be embodied, and the rationale behind the pro- cess. Ideally, many of the conditions to be required should be agreed upon and the implementation mechanism should be in motion prior to lending (the SAL approach). Significant movement in this direction is taking place. Next Steps 2.41 Provided additional support is made available to the EGY Financial Adviser in FY83 as is now planned, a joint task force would be formed to review the status of financial covenants and to deal with the other finance matters discussed in this section. The issues to be addressed would include: - reconciling revenue requirements with efficiency pricing objectives (including a consideration of training for financial analysts in LRMC pricing); - adopting agreed plans/programs of action (already used by some regions) to replace historical tests which now result in ex post defaults; - improving cash generation covenants; - developing specific strategies for country or regional improvements in accounting, costing and auditing; - adopting common policies with other development banks; - identifying needed sector-specific guidelines and research work; - reviewing accounting practices of utilities in centrally planned economies to formulate appropriate revenue convenants for power borrowers in those countries; - 18 - identifying appropriate power utility costing systems and encouraging their use by borrowers as efficiency indicators, and planning professional development activities for financial analysts, including participation in international accounting courses and seminars. F. Issues in Sector Work 2.42 Before 1978 only a few regions treated power lending in the context of the energy sector as a whole. Since then, the structure of energy sector work has changed substantially. Economic reports and CPPs have begun to focus on the economic efficiency, resource mobiliziation and investment issues associated with all forms of energy. The move into structural adjustment lending provided an operational focus for this work. As a result, energy related undertakings are becoming common features of structural adjustment loans. Nonetheless, in 1981, power activities accounted for only 2% of country economic and sector work. 2.43 Past sector work in power, and to a lesser extent in energy, has been largely supply-oriented with emphasis on resource assessment, pricing, and sector organization. The increasing importance of conservation and efficiency pricing has made necessary an added focus on energy demand. Data on petroleum product consumption and related issues, such as refinery product balance, are now mentioned frequently in power reports. Hlowever, aside from sectoral consumption data derived directly from billing records (agriculture, industry, residential, etc.), power reports commonly contain little functional analysis of electricity demand. 1/ The collection and analysis of demand data is time consuming, but it is essential to a proper treatment of demand-related issues such as efficiency pricing and conservation. An analysis of energy consumption in industry, and to a lesser extent in agriculture, is particularly important to deriving a consistent development strategy. For this reason, much closer integration and exchange of sector information between EGY, IND, and the Regions will be necessary. The programs divisions or country economists might usefully serve as the catalyst in this exchange. In this context the need for separate "power" sector reports is declining, except to review the institutional structure in the power subsector. 2.44 The Energy Department is undertaking a 60-country program of energy sector assessments, partially supported by a UNDP grant. This program will require more sector work participation from the regional power divisions than in FY78-80. Existing coefficients for energy sector work and project preparation, already the lowest of any major sector, are insufficient to deal 1/ Determining what portion of power consumption is used for industrial motive power, irrigation pumping, lighting, appliances, space condi- tioning, etc. -1a - with the increasing complexity of the sector. More deliberate efforts to identify priority countries (using "issue matrices" as in Annex III, or similar techniques), and to build lending programs for these countries through the country energy assessments and CPPs, would be worthwhile. Technical assistance and engineering loans deserve more prominence in lending programs to complement project lending and to address the many new issues in the sector, especially those dealing with operations and efficiency. Considering the likely constraints on the Bank's budget, additional resources for this work will be needed and a cooperative program with the UNDP to mobilize these resources is under consideration. G. Sector Lending 2.45 The need for flexibility in allocating Bank finance to supplement other sources is leading to more complex multi-component power "projects," and in a few cases to sector loans for power. W4hile sector lending, as defined in Annex A of OMS 1.19, would appear to have many advantages in power, the large scale of subprojects, and the relatively advanced state of preparation required before appraisal, generally preclude delegating appraisal to others, except perhaps for distribution subprojects. Therefore, moving toward sector lending or multi-component projects is likely to increase, rather than decrease, the demands on staff. Innovative approaches, such as involving the borrower in appraisal report preparation (Brazil), and making more use of post-award procurement review, may ease this burden somewhat. Periodic and detailed review of borrowers' investment programs, coupled with more intensive procurement reviews and less concern with specific project engineering (the strategy used for power loans to Mexico in the early 1970s), is a feasible alternative but would require much more supervision than is now usual in Bank power projects. H. Issues in Project Justification and Implementation 2.46 A proposed power project is considered justified if it is shown to be part of the least-cost development program required to meet demand at economically efficient price levels. Most project justification work must, of necessity, rely on feasibility studies prepared by the borrower or his con- sultant. Bank staff cannot always review the terms of reference or monitor the progress of these studies. As a result many studies suffer because (i) all practicable alternatives have not been identified, (ii) estimates of cost and risk for the alternatives have not been prepared to the same standard as those for the proposed project, (iii) the project and the alternatives have not been optimized using proper economic criteria, or (iv) loss reduction, load management and other techniques for demand management and system improve- ment have not been considered as an integral part of system planniLng. Con- straints on project execution, uncertainty about costs, and analysis of the risks associated with delays, also deserve a more prominent place in feasibil- ity studies and project appraisals. 20 - 2.47 As well as a deliberate effort to become more involved in the feasi- bility stage of project preparation, more specific guidelines for power pro- ject feasibility studies for use by Bank staff and borrowers may be helpful in alleviating some of the shortcomings just noted. EGY frequently receives requests from consultants and others for information and details on how the Bank "looks at" power projects. Resources permitting, such guidelines will be prepared in FY82/83 by EGY/PAU. 2.48 Project justification work does not suffer from a lack of methodology. Work in the Bank and by others has produced numerous models, both simulation and linear, capable of analyzing the most complex power system. 1/ The WASP III model, developed by IAEA from a model originally used by TVA, is available on the Bank's computer and enables Bank staff and borrowers to verify a development program proposed by a country (or consultant), or to test the implications of ef'ficiency and social prices for the merits of a program developed using financial prices. 2.49 Although the utility of analyses with such models is acknowledged, using these models in the Bank is generally not practical because the detailed engineering data needed is not available, and too much staff time is required to run them. Moreover, it is becoming evident that the current uncertainties of project cost, risk, and timing cast doubt on the validity of least-cost programs derived from models. Models need to be improved to permit more realistic treatment of planning in resource-constrained environments, and to take account of risk and uncertainty, including such diverse aspects as transmission constraints, reliability, and probability estimates for costs and construction time. On the other hand, adequate treatment of these complications would make it virtually impossible for Bank staff to undertake the actual planning exercises, even in a monitoring capacity, because of the amounts of data that would need to be collected and the time needed for the analysis. EGY/PAU is sponsoring a research project to catalog the models that are available and their applicability to developing countries, identify their advantages and weaknesses, and determine areas where additional work may be profitable. Economic Analysis of Power Projects 2.50 Revenues from incremental power sales are used as a measure of benefits to calculate an internal economic rate of return (IERR) for eacn power project. This calculation is relevant to the investment decision only inasmuch as the rate of return estimated is a useful lower bound, based on revenues that represent a minimum measure of economic benefits. A rate of return less than the opportunity cost of capital generally indicates that the 1/ e.g. see Dennis Anderson, Models for Determining Least-Cost Investments in Electricity Supply, World Bank Report Series, No. 2, 1972, and Herman van der Tak, "The Economic Choice between Hydroelectric and Thermal Power Development," IBRD Occasional Paper No. 1, 1966. - 21 - average selling price is less than the LRMC of supply, or the economic effi- ciency price. Thus when calculated in this way, the IERR should be used as a check on the average tariff level, rather than as a criterion for the invest- ment decision. (A low IERR, implying prices significantly below LRMC, does raise the issue of whether an increase in prices would choke off so much demand as to justify a delay in the project or other changes in the investment program; this question must be answered in the project justificat:ion analy- sis.) 2.51 Internal rates of return based on revenues from incremental power sales can be misleading. This is because revenues form only the directly observable part of the total economic benefits, which also includle consumer surplus. In theory, the full economic benefits can be measured by consumers' willingness to pay (WTP) for extra electricity. In practice, willingness to pay is very difficult to measure, except in certain special cases. Studies that have quantified WTP by determining the value of a kWh forgone typically yield high figures, compared with the average tariff in developing countries of about 4 USJ per kWh, indicating that consumer surplus is quite high. Hence the revenue-based IERR is probably significantly lower than the "true" IERR. 2.52 Some degree of standardization is needed to improve the usefulness of rate of return calculations in Bank power operations, for two main reasons. First, in attempts to capture some of the consumer surplus in these calcula- tions, various combinations of future tariffs and expenditures on alternatives (e.g. kerosene, batteries, diesel irrigation pumps, autogeneration), are frequently included in the estimates of benefits. This results in rates of return closer to the "true" one, though still of little relevance to the investment decision. Second, since projects almost never cover complete systems, revenues and costs must be allocated among the various system components to obtain a return on the specific project investment. There are several methods for doing this, each more or less acceptable but yielding different results. Because of the varying methods now used and the differing assumptions, comparisons of calculated returns among power projects, or with projects in other sectors, are not always meaningful. 2.53 To be consistent with the emphasis on efficiency pricing, the rate of return calculated should be the return on the entire investment program (in- cluding the project), using expected incremental revenues (in real terms) as benefits, and valuing all inputs at economic costs. This presumes that the investment program has been demonstrated to be the least-cost invesl'ment pro- gram required to meet the demand at economically efficient price levels, and it requires an explicit judgment that the price levels do not exceed a rea- sonable willingness to pay. If desired, a supplemental calculation may be made using demonstrated willingness to pay for alternatives which wi:Ll be sup- planted by the proposed program. Care should be taken to explain the basis of the calculations in the appraisal and president's reports to prevent misunder- standings over the implications of the rates of return. - 22 - Project Implementation 2.54 Because of the capital-intensive nature of power projects and the high economic cost of shortfalls in electric power capacity, the costs to an economy of project delays can be substantial, often exceeding the concessional element of Bank financing or other official aid. In today's inflationary environment delays also usually result in cost overruns which make mincemeat out of project financing plans. 2.55 Bank guidelines (e.g. OMS 3.55) encourage project officers to build implementation monitoring devices into projects and require the use of plan- ning and monitoring tools such as computer-based systems like CPM or PERT. The Bank's training program includes courses on construction project management and incentive contracting measures designed to improve project implementation efficiency. 2.56 Nonetheless, insufficient emphasis has been placed on project execu- tion measures in the power sector. This is largely because specific require- ments are considered unnecessary since projects are supervised by competent consultants or borrowers. 1/ Unfortunately, experience does not support such a sanguine view. Half of the project completion reports prepared in 1978-80 indicated avoidable delays in project completion and current experience is not encouraging. Implementation delays are becoming common, especially those associated with shortages of local currency. 2.57 The "Construction Manager" approach is seldom used in Bank-financed projects, partly because of borrowers' resistance to the delegation of procurement responsibility that is inherent in this approach. Even so, more use of this technique, "fast tracking" and similar contracting procedures which emphasize timeliness and shared risks and incentives between owner and contractor would seem desirable. EGY/PAU will endeavor to stress these points during preparation and appraisal reviews. 2.58 Better supervision is not necessarily the solution. The remedies proposed presume that supervision is adequate, and it probably is unrealistic to expect Bank staff to do more than is already being done at that late stage in the project cycle. At the appraisal stage, it would be desirable to give more attention to the proposed implementation arrangements, perhaps adding to appraisal reports a statement concerning the proposed contracting arrangements and indicating what incentive and/or penality provisions are proposed. Managers should encourage staff to take advantage of training opportunities in contracting and project management. 1/ None of the power projects approved in FY78-80 included specific legal clauses with respect to project implementation (other than the standard "due diligence" and monitoring clause), and cases were rare where the Bank intervened to seek changes in contract format or to take other measures to improve implementation efficiency. - 23 - I. Cofinancing 2.59 Before 1975 only about 10% of Bank-financed power projecl-s employed cofinancing. With the growth of regional banks and other maltilateral official sources, the increasing size of power projects, and the limitations on the Bank's resources, by 1978-80 two-thirds of the power projects approved were cofinanced. Cofinancing with export credit agencies (suppliers'/buyers' credits) and commercial sources has also grown rapidly. Since power projects are particularly attractive to cofinanciers it is likely that most if not all of those approved in 1982-86 will use this form of financing. 2.60 To date, most cofinancing arrangements have been organized on an ad hoc basis, often delaying projects considerably while financing was being sought or contracts allocated. The Bank has been quite flexible in allocating its financing to studies, civil works, and similar components less amenable to financing from private sources, but there is ample scope for staff, including projects staff, to be more active in identifying and allocating coflnancing. 2.61 Unfortunately, borrowers often delay arranging for cofinancing in hopes that the Bank might at the last minute increase its loan to cover any shortfall. Therefore, early and open discussions of the probable ceilings on loans are important. More effort to encourage borrowers to obtain precommit- ments for export credits (buyers' credits) and to explore the commercial markets is necessary in the early stages of project preparation. 1/ Bank projects staff can play a particularly important role by evaluating, at the project preparation stage, the options for contract packaging and t:he result- ing cofinancing possibilities in order to establish reasonable limits of Bank financing; these should be mentioned in the project brief. 2.62 Requests for information on cofinancing opportunities in power are increasing rapidly and some mechanism, other than the Monthly Operational Summary, is likely to be needed in the near future to respond to these inquiries. For example, a reformulation of the project brief to permit extraction of relevant information on project technical details, expected contracting arrangements and packaging, and financing needs, may be required. Presumably the EIS cofinancing coordinator will address these issues. 2.63 At present, most cofinancing arrangements with suppliers and commer- cial banks are "arms length" deals as far as the Bank is concerned. If cofinancing is to be accelerated, more use of joint financinig may be necessary. Innovative contract packaging and procurement procedures offer possibilities for achieving this and staff should be encouraged to propose such arrangements at an early stage. This may require that the Bank adjust 1/ See Bank Staff Working Paper 409, "The Changing Nature of Export Credit Finance," July 1980, for a discussion of the current status of the export credit market. - 24 its procurement policies to permit cofinancing to be organized during the bidding process so that Bank funds can be used to meet downpayments or to provide a specific (small) percentage of contract financing. EGY/PAU has be-an working closely with the office of the Senior Adviser for Cofinancing to explore such alternatives, but now it may be diesirable to establish a more formal system or procedure for early identification of project components that are suitable for cofinancing. 2.64 In earlier years the Bank was instrumental in helping power borrowers issue shares and bonds in local and world marklets. The need for these ser- vices diminished as governments acquired sector assets and tended to arrange external financing through export credits or from commercial banks. Although today the Bank would be hard-pressed to muster staff with expertise in this area, at least in the power sector, expanding sector or project-related commercial bank borrowing may be beneficial both to increase lenders' security and as a means of obtaining longer repayment terms. Use of mediura-term commercial credit to finance assets (hydro projects) with long economic lives is the cause of many of the cash flow problems faced by growing utilities today. Roll-overs, balloon payments and simiLar arrangements to deal with this problem can be more easily justified if ithe financing is related to a specific project, and the Bank should be in a position to advise borrowers on these matters. Adoption of efficiency pricing as proposed in this paper will improve financial performance in many cases, and may make access to private capital markets possible. Many developing countries' financial markets are small relative to sector needs, but both local and offshore sources might be tapped through innovative instruments such as bonds indexed to energy prices. Prospects should improve for sales of local share issues and similar equity paper. Bank staff should be sensitive to opportunities for including such arrangements in the creative financing plans that will be needed to mobilize resources of the magnitude indicated int Section I of this paper. J. Rural Electrification 2.65 Formal Bank lending for rural electrification (R.E.) started only in FY76 following publication of the rural electrification policy paper. 1/ Many earlier projects included electrification components to support rural productivity increases, but there was no specific policy for evaluating and approving such components. Today, most power sector studies include discussions of access to service and an evaluation of country polixCis concerning connection charges, grid extensions, and rural electrifJcation programs, while institutional and organizational obstacles to the expansion or service are identified and addressed in the course of lending. 2.66 Rural electrification projects are nc,t always appropriate for Bank lending in power and special care should be taken before including R.E. 1/ Rural Electrification, A World Bank Paper, October 1975. - 25 - projects in lending programs. Often R.E. projects are capable of attracting financing from a wide variety of sources, often on concessional terms. Also properly-prepared rural electrification projects can absorb two to three times as much staff time as projects in more conventional parts of the sector. However, if the central agencies are reasonably strong and efficient or can be made so, if rural electrification accounts for a large part of the sector program (India, Bangladesh), or if the Bank can usefully establish planning criteria for rural projects or coordinate cofinancing, then participation in rural programs is well justified. 2.67 Recent experience has shown that the Bank can play a needed role in R.E. projects by: (i) coordinating aid from other sources; (ii) using more rigorous economic analysis to justify projects and to help put subprojects in order of priority (a task usually subject to strong political intervention); and (iii) optimizing design standards. As the Bank's share of resources in the sector dwindles, the task of sorting out the politically-inspired from the truly worthy is becoming increasingly important. It is also necessary to ensure that good programs are not shelved simply because their justification cannot be proven. 2.68 More explicit guidelines for evaluating the social benefits of rural electrification projects are needed to establish a better analytical base for designing these projects. Though such projects appear to bring direct bene- fits to a disadvantaged part of the population, hard evidence of this is lacking. The FY83/84 EGY/PAU research program includes a study to review past Bank-financed R.E. projects with the objective of identifying better methodologies for preparing and justifying R.E. programs. 2.69 In their early years, until demand has increased to economic levels, R.E. projects rarely generate enough revenue to cover costs. Hence most Bank- financed rural electrification projects will continue to demand lenient treat- ment with respect to financial earnings covenants. This is acceptable pro- vided R.E. tariffs are set to cover LRMC. Subsidies in R.E. projects are common, but they should not exceed the difference between LR4C and the (higher) revenues necessary to achieve financial viability. Although social reasons are often cited for subsidizing R.E. programs, in fact the principal beneficiaries often are large landholders and subsidies may actually represent transfers from poorer urban consumers. For this reason, emphasis on LRMC- based prices is particularly important in R.E. projects, and "socially- mandated" prices should be not accepted without scrutiny. 2.70 Because of the marginal economic nature of most R.E. projects, it is especially important to investigate the basis on which they have been planned, and to ensure they meet the following criteria: - 26 - (i) the project should be part of an integrated rural development program; (ii) subproject priorities within the program should have an economic basis; (iii) a proven market with a developed cash economy should exist; (iv) project designs should follow least-cost practices appropriate to the market; and (v) either the entity responsible for operating the project should have a proven capability to maintain and expand the power system, or a mechanism for developing such a capaLbility should be included in the project. K. International Interconnection 2.71 As systems grow, international interconnection is becoming an increasingly viable alternative which deserves more study. This would be especially useful for several markets in western and eastern Africa, Central America, western Asia, the Indian subcontinent and central South America. Interconnections are not rare in developing countries, and they tend to exclude situations which create dependence by one party on the other. Thus, net-zero interchange" connections, border-town sales, marginal imports and similar arrangements are common, having survived wars and other political disturbances. In planning such connections a pragmatic approach has been taken which acknowledges that countries can legitimately consider security as an objective in planning power development, but at the same time should not ignore the possibilities for imports in association with appropriate back-up facilities. As a corollary, power developments intended primarily to supply an export market should not be encouraged unless it is likely that the exports can be sold at prices which will fully recover long-term costs and ensure that the supplier retains a fair share of the consumer benefits. This is akin to the policy which should be followed for enclave projects (e.g. aluminum smelters). These principles are straightforward, and the Bank is prepared to consider financing such international interconnection projects. L. Nuclear Power 2.72 The Bank's lending policy for nuclear power was set out in the 1975 policy paper, Nuclear Power: Its Significance for the Developing World. This report treated nuclear power as simply another option to be considered when searching for the least-cost way of meeting demand. Aside from a very early Italian project, the Bank has provided no direct financing for nuclear power and very little Bank funding has been used for nuclear training and studies. The coordinating and guiding role that was envisaged in the 1975 paper was not developed and the Bank no longer has expertise in nuclear power technology, or in areas such as training, safety and waste disposal. 27 - 2.73 Few developing countries have nuclear plants, partly because a relatively large system (about 5,000 MW) is required to make use of a nuclear unit of minimum economic size. For those countries that do have nuclear programs, enough financing has generally been available from bilateral sources and thus there has been no financing role for the Bank. 2.74 As in industrialized countries, growing concerns about safety and reliability, and public objections to nuclear power, have led to delays in the nuclear programs in developing countries. However, the systems have not stopped growing and it is reasonable to expect that more power dlevelopment programs will contain economically justified nuclear projects. Although bilateral sources will probably provide the needed financing for the nuclear steam supply components of nuclear projects, in conjunction with technology transfer and fuel processing agreements, the escalating cost of nuc:Lear plants makes it likely that other financing sources, including the Bank, will be approached to finance the "balance-of-plant." 1/ Export credits should be available to cover most of this expensive hardware and it would be preferable for the Bank to confine its financing to training and studies, or other parts of the power development program less likely to attract financing. 2.75 There have been no significant commercial developments since 1975 which would warrant a shift in Bank policy, but in case the Bank is called upon implicitly to endorse power programs containing nuclear plants, it would be useful to include assessments of nuclear training and safety programs in sector evaluations. Consultants can be used for this purpose if necessary. EGY/PAU is rebuilding the Bank's links with IAEA, which were recently weakened by staff changes in both institutions, because it is presumed IAEA will continue to be responsible for recommending policies towards trainirng, nuclear waste disposal, safety, siting, nonproliferation, and similar issuess. 2.76 At present, nuclear program planning assumes that the minimum economic size for nuclear units will remain at about 600 MW. Interest has been expressed recently in smaller units specifically for use in developing countries; France is marketing a 120-MW unit based on submarine technology and the UK has developed a 300-MW unit aimed at smaller markets. It is unlikely that units smaller than about 600 MW can be manufactured cheaply enough to compete with alternative plants (coal, for example). However, aggressive sales promotion and subsidized price structures may make their introduction attractive to some countries. Should this happen, nuclear prospects in developing countries will take on a completely new character. There is little the Bank can do at this stage to prepare for such an eventuality but EGY/PAU will continue to monitor this aspect of nuclear development to provide as much "early warning" to regional staff as possible. 1/ Romania requested Bank finance for a nuclear project in 1979 but subse- quently withdrew; Egypt is revising its nuclear program and may request Bank participation. - 28 - 2.77 A related issue concerns nuclear-eligiLble countries' proposals to start constructing nuclear plants for training purposes even though such facilities have not been shown to be part of a least-cost program, If given the opportunity, the Bank should resist such proposals since, in general, a combination of training in research-scale facilities and in plants overseas is both cheaper and quicker. M. Mydro and Other Renewable Sources 2.78 A recent review of hydropower programs in developing countries indicates that, with few exceptions, work on feasibility studies and imple- mentation of large-scale hydropower projects is proceeding satisfactorily and at a rate which takes account of resource costs and the potential for shifting away from oil. Additional efforts are needed to identify small-scale resources and to upgrade hydrological resource data. Bank staff and country planners alike are still seeking out prospects for international intercon- nections and large multinational hydro projects, even though few viable ones have been identified. 2.79 Traditional feasibility study approaches often dismiss small sites as uneconomic or outside their scope even though those which would yield 5 MW or above may be large enough to develop in their own right. Given today's high inflation rates and the long construction periods for large projects, a program of several small projects may be preferable even though unit cosl:s would be higher as a result. Bank engineers are expected to review each major hydro feasibility study critically, not only to look at the benefits for purposes other than power but also to ensure that prospects for smaller project alternatives have not been overlooked. 2.80 The prospects for smaller projects ("mini" hydro, say 50 kW to 2 MW) are not quite as good because their benefits are usually insufficient to justify site-specific feasibility studies: most countries lack the manpower to undertake the large number of such studies which are required if smaller hydro projects are to make a meaningful contribution to a country's energy supply. The Bank and other development agencies are financing demonstration- scale "mini" hydro projects in several countries but a satisfactory formula for dealing with these projects on a program scale has not yet evolved. Even though the sum of such projects is not likely to contribute much to the world energy supply, this activity is worthwhile. EGY/PAU has initiated discussions with other agencies in this field, and has commissioned work which could lead to feasibility studies for projects consisting of large numbers of mini-hydro schemes. Although its proponents see the main use of small-scale hydro as bringing electricity to rural areas not previously supplied, its principal role, at least initially, will probably be to supplant or supplement diesel- based generation in existing isolated systems. For this reason mini hydra schemes should not be planned solely in the context of rural development projects. - 29 - 2.81 At the lowest end of the scale, there seems to be little scope for including "micro" hydro (0.5 - 50 kW) in the Bank's power program. Instead, projects in this category should receive attention through its renewable energies program, rural development programs, and projects to supply individual consumers (e.g. households or small farms). 2.82 The TJNDP is funding a cooperative program with the Bank to identify pre-investment needs for new and renewable resources, including hydro, as an outgrowth of the 1981 Nairobi renewable energy conference. The Bank may be asked to identify and manage studies under this program. Safety of Dams 2.83 The failure of some European dams in the 1960s and the implementation difficulties experienced at the Tarbela project in Pakistan in the mid-1970s focused attention on the Bank's role in assuring the safety of projects it finances containing large dams. OMS 3.80, issued in June 1977, imposed three requirements: (i) review of the project concept and design by an independent panel of experts; (ii) a convenanted undertaking to monitor, inspect and maintain the dam; and (iii) a review of past projects financed by the Bank to determine whether proper monitoring procedures were being followed, and whether covenants should be included in repeater loans to require monitoring of past projects. Since issuance of the OMS, monitoring covenants have been included routinely in loans for projects containing large dams. The standard monitoring covenant requires only that a satisfactory monitoring program be submitted to the Bank one year before the project is completed. Thus, given the usual construction time, such programs are only just beginning to be received in the Bank. 2.84 A major problem is that the Bank lacks the capacity in the power sector, and to a lesser extent in AGR and TWD, to monitor these undertakings effectively. An OPN on dam safety is being prepared by AGR with input from EGY and TWD. This will provide guidelines for evaluating monitoring programs and suggest terms of reference for review panels. Even so, the monitoring of dam safety is a complex and highly technical subject which, given the responsibility implicitly assumed by the Bank in reviewing such programs, deserves more careful attention. It should also be noted that although an initial survey of past projects with large dams was undertaken in 1977/78, not all regions have taken the follow-up measures required by the OMS. To achieve the objectives of the dam safety program, it would be desirable to establish a central point within OPS/ENI to oversee the program and to provide specialized advice on the technical aspects of dam safety. Establishment of a standing expert panel on dam safety would be worthwhile. Until that time, regional divisions should continue to engage specialized consultants to assist in the review of the safety aspects of dam projects within their purview and to evaluate the proposed monitoring systems. - 30 - Other New and Renewable Sources 2.85 The 1980 Bank study on renewable energy potential 1/ indicated that there would be no radical change in Bank power operations with respect to renewable resources other than small hydro, since the available alternatives are still rarely cost effective for supplementing power grids, even small ones. Successful applications are more likely to be made for specific tasks, like water heating and crop drying using solar energy, neither of which tasks consumes much electricity in developing countries. 2.86 Though prospects for central-scale aLternatives using renewable resources are not bright, they are far from hopeless and deserve deliberate attention by Bank staff in formulating power development programs. If ready stocks of agricultural and wood wastes are availalble, thermal plants (steam or pyrolytic conversion) using them can be economic additions to a power system, and in this regard sugar-growing countries should make every effort to ensure that bagasse is being used completely and efficiently. Ocean thermal conver- sion (OTEC) is being promoted aggressively in some countries but is not yet a proven technology, except on a very small scale. Economic prospects for OTEC are brightest where open-cycle equipment can be used to produce fresh water along with electric power, and small prototype units are being installed where conditions are good, as in some Caribbean sites. Wind-powered induction gen- erators can be economic when used to augment oil-based systems, but the requi- site site surveys often have not been made. These surveys require relatively long record periods, as do those for hydro development. 2.87 Bank projects could include many more study components to collect the data needed to develop renewable resources. There seems to be ample awareness of this issue in most countries, often enhanced by the Bank's energy sector assessment missions, and some data collection systems are being put in place. Even so, Bank power staff still need to be alert to opportunities to include development of renewable energy sources as part of longer-term plans. 2.88 The cost of peripheral equipment (e.g. structural systems, batteries, transformers, and AC/DC converters) appears to pose an obstacle to the commer- cialization of large-scale solar power systems, including photovoltaic and thermal systems. However, household-scale solar systems to substitute for kerosene lighting and to provide modest power fcor radio and TV may be useful where demand centers are scattered or non-household demand is too small to justify the extension of the grid or the installation of decentralized sys- tems. Other technologies approaching the marketable stage may find a place in future programs. Examples are fuel cells, which run well on 50% hydrous alco- hol and therefore can be coupled synergistically with simple fermentation/dis- tillation processes, and low-Btu gas which can be produced from waste diges- ters and wood or coal gasifiers. EGY will make every effort to stay abreast 1/ Renewable Energy Resources in the Developing Countries, November 1980. 31 - of such developments as well as of chan ges In larger-scale technologies such as those for cogeneration, geothermal e'lectricity, and combined cycle, and bring them to the attention of interested staff. N. The Environment 2.89 Present policy monitoring procedures and staff awareness ensure that the Bank is unlikely to participate unvittingly in projects which would cause significant social or ecological harm. Indeed, in many cases the Bonk's emphasis on these issues seems to exceed that of its borrowers. Environmental considerations nonetheless raise some difficulties in its operations. Project evaluation and planning sometimes fail to take account of the costs and time constraints imposed by these considerations, while project implementation can be delayed by difficulties in resettling population in hydro reservoir areas or by borrowers' inattention to putting monitoring systems and environmental protection facilities into place. These problems seem to be particularly difficult in the power sector because the implementing agencies typically are not especially sensitive to these issues; even those with a long history of dealing with such matters as population resettlement may not be attuned to the increased awareness of environmental issues in the developing countries as well as in other parts of the world. 2.90 Environmental concerns have aided to the operational difficulties arising from energy price increases and the growth of demand that already face the sector. In addition, the project alternatives now being considered--more complex hydro projects and greater use of coal in particular--have potentially larger environmental consequences. Resettlement issues are dominant in many hydro projects, and accordingly loan conditions requiring submission of detailed resettlement plans are common features of such projects. Resettle- ment must be dealt with as an integral part of the construction program and of general project planning. To avoid the major problems that can arise during implementation, the Bank often needs to play a stronger coordinating role to ensure that the implementing agency receives adequate support from other parts of the government, especially those responsible for legal, agricultural, and social matters. The Office of the Envi-ronmental Adviser (PAS) should be con- tacted at the earliest possible stage to provide assistance in. such projects. 2.91 The environmental impact of coal-fired power stations may require the creation of pollution control agencies and related legislation at a very early stage in the project, and much monitoring of air and water quality prior to the project to support the work of these agencies. The Bank's role in the sector here is not to impose standards taken arbitrarily from elsewhere but rather to assist governments in identifying and costing policies appropriate to their circumstances. 2.92 In general, it can be said that Bank staff need to give more attention to potential environmental issues in projects where the threat to the environment is not immediately evident. Coal and lignite projects are obvious candidates but the environmental issues associated with other forms of generation, particularly geothermal, may be equally serious. - 32 - III. THE BANK'S RESOURCES IN THE POWER SECTOR A. Regional Power Staff 3.01 Whereas power is the second largest sector of Bank lending, and ranks fourth in number of projects, it ranks eighth in operational staff (see Table 6, Annex I). As a consequence, manpower coefficients for lending and super- vision, and OPS/ENI advisory input, are among the lowest of any sector. 3.02 As of September 30, 1980, there were 52 K-M level staff assigned to power in the seven regional divisions responsible for power activities. This small cadre (2% of total J-Q staff or 5.7% of regional projects staff) was responsible for 16% (or nearly US$5 billion) of total Bank lending in FY78-80. By August 31, 1981, as a result of filling vacancies and eight new authorized positions, the power staff complement was 65, but the increase was not sufficient to reflect the growth in operations. 3.03 The mix of technical specialists has not changed appreciably over the years. An engineer supported by a financial analyst (occasionally the other way round) forms the basic power project team, advised by an economist. A typical power division has five engineers, three financial analysts and one economist. As economic work has become increasingly important in the sector, the operations divisions have become increasingly understaffed in the economic field. Managers are attempting to adjust their staffing complements to recognize this but are bound by a budget system based on historical co- efficients in which the technical and financial inputs are already considered to be irreducible. 3.04 The average division is almost too smal:L when language abilities are considered. The staffing pattern, a legacy of the 1972 reorganization, offers little opportunity for specialization or flexibility, and the small EGY power advisory unit can offer little direct technical support. Adoption of matrix management 1/ would permit more flexible use of technical staff, but resources would still be too scarce to mount the larger miLssions which would be needed if missions were staffed with individuals having narrower technical specialties. Future Staffing 3.05 Given the scope of the sectoral changes outlined in this paper, historical staffing coefficients are not reliable guides to future staffing needs. Even using historical coefficients, the projected growth in the 1/ A form of organization used widely in technical and research firms wherein each individual is responsible to two supervisors: one for administration and one for technical matters. - 33 - operations program in FY 82-84 (Annex l) would require staffing increases of 22% in the power divisions. At present, the seven regional power divisions have a sustained processing capacity of about 22 projects a year. In order to adLdress the issues outlined in Chapter II, while expanding the lending program into some 14 new countries, particularly in Africa, a program approaching 30 projects a year would be appropriate. On this basis, doubling the complement of power staff over the next three years would probably be cost effective. 3.06 Obviously, this expansion of staff is unlikely, and not solely because of limfitations on the Bank's administrative budget. Even if there were no other constraints, the rate at which divisions can absorb new staff depends on the amount of training these staff require. Also, qualified candidates are scarce--recruiting power engineers to cope with the Bank's 30% turnover in this specialty in the last two years has proven to be difficult; neither is there a pool of qualified economists and financial analysts, and some vacancies have gone unfilled for more than a year. 3.07 Support and cooperation from all levels of management, as well as PMD, will be required if the Bank is to cope successfully with the staffing issues in the sector. More flexibility is needed in allocating posts. Recruitment efforts should be continuous regardless of the vacancies actually open, and a budget mechanism is needed to bring good candidates on board even if a specific vacancy does not yet exist (as of end-81 the Bank had ceased actively recruiting power economists, In spite of future sectoral needs, because there was no identified vacancy). Managers need to allow additional resources for on-the-job training. Although a sound knowledge of the electric utility business will continue to be vital in project work, some specializa- tion in technical fields can be accommodated if unit managers are assured of the resources needed to plan an adequate training program and to move toward the multi-disciplinary staffing of missions dealing with complex projects. Experience 3.08 Given the Bank's emphasis on institution building, in-depth experience in public utility engineering, management or finance has long been a prircipal criterion for employment in a power division. Although actual experience in a utility is considered desirable, staff have come from related organizations including consultant agencies, other development: banks and securities firms. All of the power engineers in the Bank have a direct utilities background and four of them are procurement speciialists with training in that field. Over half of the engineers had more than 15 years' experience in utility engineering and management before joining the Bank. Financial analysts are no longer required to have prior experience in public utility financing, and only half of the present complement have such experi- ence. (The requirement was relaxed in the past few years due to recruitment diffictulties and a deliberate effort to provide programs and economic staff with increased opportunities for interchange among the sectors.) Since power engineers with broad experience are becoming increasingly scarce, rmanagers are starting to accept candidates with narrower technical backgrounds and are teaming them with senior financial analysts who have operational experience in the sector. This has increased the training burden on EGY advisory staff, but - 34 - the overall outcome has been successful and will continue to be so provided appraisal teams can include at least one member who is thoroughly grounded in public utility management and operations. Recruitment 3.09 Difficulties in recruiting power engineers have been worsening since the early 1970s. Energy economists, and to a lesser extent financial analysts, are also becoming more and more difficult to recruit. The circumstances responsible for this are: (i) the disappearance from the "market" of the mobile broad-based engineer/manager; (ii) the employment by utilities in developing countries of indigenous managers who have no desire to work abroad; (iii) increasing specialization in utilities in industrialized countries, so that the only candidates attractive to the Bank are those in or near top management; and (iv) a decline in the Bank's competitive position. Since most power missions are already staffed at an irreducible minimum (two individuals for appraisal, often only one for supervision), the demand for even broader skills has increased with the Bank's movement into renewable energy sources and conservation, and with the increasing sophistication of the its borrowers in the sector. For example, the terms of reference for a recent supervision mission to a small country required that the engineer review the draft feasibility report on one hydro project and initiate studies on two others, determine progress on a program for reducing distribution losses, review the technical designs for a rural electrification project, discuss consultants' recommendations on organization and management, determine whether a project on oil-to-coal conversion was worthwhile, and supervise the ongoing thermal generation and transmission project. Such requirements are not unusual but one wonders whether even a paragon could do an adequate job on all. 3.10 Recent experience has shown that it is nearly impossible to attract qualified power staff from Part II countries, barring political instability or other abnormal circumstances (which in fact, have played significant roles in past recruitment). Qualified candidates from t'hese countries generally have senior management or high level government positions with substantial perquisites and the Bank cannot offer them competitive compensation pack- ages. Consulting firms occasionally provide candidates, and the IDB's resident representative training program has proven to be a good source of qualified staff. As for Part I candidates, the Bank is only marginally competitive with utilities and consulting firms in Western Europe, but it can and still does attract qualified individuals from US consulting firms, utilities and government organizations, though identification of these individuals is difficult. Advertising in US journals produces a large volume - 35 - of replies (300 plus in November 1980-January 1981) but few qualified candidates, and even when acceptable candidates are identified rejection of the Bank's salary offer is not uncommon. Personal contact through at:tendance at professional conferences and direct recruitment efforts have proven to be essential in recruiting high caliber staff, and the Bank should be more purposive in using such means. Staff referrals, an important source of qualified candidates in the past, are playing a declining role partly because staff do not perceive themselves as part of the recruitment mechanism. Training 3.11 Even organizations with more liberal resources than the Bank are faced with similar recruitment problems. To some extent this reflects a shift in the education patterns in the US and elsewhere. Few engineering schools now teach power engineering as a technical specialty. As a result, ut:ilities, consulting firms and others who need such skills have developed in-house training programs to meet their requirements for technical specialists. The Bank has neither the staff nor the money to undertake broad technical training beyond the state-of-the-art seminars that are presently arranged. However, it would be able to supplement its language training and the usual in-house operations courses with ad hoc training in system modelling and program development, distribution engineering, or other specialties. More planned use of the EDI power projects training course, which usually can accommodate two or three Bank staff, would be beneficial and thought should be given to expanding this into an in-house training program. The USDA graduate school power course is an option to be considered. An extended period of working with more experienced staff must also be allowed for as the present "one shot" training and familiarization mission for new staff is far from adequate. 3.12 To bridge these training gaps attempts have been made to arrange secondments from major overseas utilities, but generally without success except for the good working relationships achieved with Electricite de France and,the Irish Electricity Supply Board. Although initial contacts were dis- appointing, prospects for such arrangements with US utilities need to be explored more thoroughly as a mechanism for providing broader experience to otherwise qualified candidates. TVA has recently expressed interest in working with the Bank in energy development. Prospects for seconding Bank staff to utilities, investment banks, and similar institutions for sabbaticals or similar training and refresher programs also need to be explored, particu- larly for financial analysts. Although virtually no power staff have participated in the sabbatical program, managers should consider this possi- bility, or an equivalent substitute, as an integral part of the prof'essional development program. Many of these issues were identified in the 1981 Bankwide training review, and EGY/PAU is working with the training division of PMD to develop a program responsive to the needs. Reassignment 3.13 Previous attempts at arranging formal succession and rotation schedules among power staff have not all been successful, especially for power engineers. Recent experience with the rotation panel has been more favorable - 36 - and rotational reassignments are now quite routine. Vacancies and opportuni- ties are unpredictable, and the staff pool is so small that a formal rotational program is difficult to work out in practice, but this difficulty has not inhibited rotation. As of end 1981 only 19% of power staff had been in post over 5 years and prospective rotational assignments were being arranged for those individuals. Over half of power staff have been in post less than 3 years. B. Regional Staff Time and The Work Program 3.14 Table 3.1 shows an activity breakdown of the power work program managed by the regional divisions in FY78-80. The relatively large shares of staff time assigned to lending activities (44%) and supervision (22%) are typical of the sector, as are the very small shares used for economic/sector work and completion/evaluation. This division of effort varies little among regions. Practices in the sector regarding project identification and preparation, appraisal, and supervision are described in Annex I. Given the large lending program managed in FY80 (25 projects, compared with 19 each in FY78 and 79), the increase in pre-appraisal effort in that year is notable; it has enlarged the pipeline of new projects (see Annex I). Table 3.1: SUMMARY OF POWER STAFF TIME, INCLUDING CONSULTANTS, FY78-80 (Staff Years) FY78 FY79 FY80 (Avg.) Preappraisal 7.68 8.89 12.07 15.0 Appraisal/NGB 18.23 21.00 16.95 29.4 Supervision 13.21 14.54 14.96 22.4 Completion/Eval. 0.78 1.07 0.34 1.2 Econ/Sector 2.41 2.81 2.24 3.9 TAS & Other 1.74 1.02 1.00 2.0 Management 4.57 4.93 5.54 7.9 Training 1.06 1.42 1.09 1.9 General 2.99 2.98 3.46 4.9 Leave 6.43 7.92 7.45 11.4 TOTAL 59.10 66.58 65.10 100.0 3.15 TRS data for the 62 power projects approved in FY78-80 show that the amounts of staff time (in weeks) vary enormously among projects: -37 - Range Mean Std. fev. Preparation 4-104 22 20 Appraisal/NGB 14-162 61 30 Staff time requirements do not appear to depend on project type or size (Table 2, Annex I), though they tend to be higher in projects needing a resolution of sector issues (Turkey), and those involving first-time borrowers (Madagascar), difficult cofinancing arrangements (Honduras), or tests of new appraisal techniques (Morocco). Managers are usually able to anticipaite these differences and it would be desirable to make more *use of this t:alent at budget time. "Other" Resources 3.16 Only 5% of the time spent on the power work program (compared with 18% in agriculture) was provided by consultants and most of this was for supervision and sector work. Overall, it is difficult for managers to integrate consultants into the work program because of (i) the smnall-team concept prevalent in the sector and (ii) the need for staff continuity on appraisals and, to a lesser extent, supervision. In addition, since there is no UN agency specifically for power, unlike many other sectors in which the Bank lends, there is no possibility of drawing power staff from cooperative programs. There are few ad hoc tasks that can be assigned to consultants aside from occasional sector missions or specific-purpose supervision and completion report work. Moreover, the Bank's computerized file of individual power consultants is not easy to use because it is organized by discipline rather than by function. EGY/PAU is working with PMD to develop a new system based on codes used for the human resources study, but recoding the existing files will be a major undertaking. 3.17 Partly in an effort to overcome these constraints, ex-Bank staff are used as consultants whenever possible. A random sample indicates that nearly two-thirds of the power consulting time in FY78-80 was provided by former Bank staff. Since this source may well become inadequate, EGY/PAU will explore the possibility of providing voluntary training to interested consultants in the Bank style of project preparation and appraisal. 3.18 There are some prospects for saving Bank staff time through cooperation with other international financing agencies. In FY81, the preparation and appraisal work on two Bank projects was done largely by cofinanciers (IDB and ADB), saving the Bank 40% of the staff time that would otherwise have been needed. Such arrangements would not always yield savings, but there appears to be scope for further cooperation of this sort. Considering that the Bank must always contribute enough to cooperative projects to verify their soundness, perhaps 15% or more of its FY82-85 power program could consist of operations where cofinanciers, usually regional banks, contribute to preparation and appraisal. (Staff economies are possible only if the cooperating agency takes full responsibility for a portion of the appraisal; past "joint" appraisals where each cofinancier fielded full appraisal teams did not save Bank staff time.) Obviously, such arrangements - 38 - require the cooperating agencies to respect one another's capability and policies, and therefore should be undertaken judiciously. C. The Power Advisory tJnit 3.19 The FY78-80 period saw significant changes for CPS activities in electric power. The creation of the Energy Department (EGY) in July 1979 laid the basis for an integrated approach to energy sector planning, focusing on resource assessment and efficiency pricing. Electric power activities have benefited from these changes even though the resources originally assigned to them, like those of telecommunications, were drawn on to support diversifica- tion and increased activities in oil/gas and water supply. Staff Resources 3.20 The Power Advisory Unit (PAU) in EGY comprises two full-time authorized positions: the Power. Adviser (since mid FY80) and a Senior Power Engineer. In addition, the Financial Adviser (EGY) and a Senior Economist in the Policy and Economic Unit spend about two-thirds of their time on power- related activities, while the Senior Adviser for Operations, and the New and Renewable Energies Unit also contribute some of their time. As a whole, the Department is now able to supply about 3 staff years a year to the power program. These resources are augmented by about 1.5 staff years a year of consulting services, applied exclusively to the small research program. The power advisor position was added only in mid FY80. Formerly, this was a shared task of the Senior Adviser for Operations. 3.21 Power advisory work is fully integrated into the EGY policy and advisory function. Power-related work in CPS declined from 5 staff years in 1974 to an average of 2.7 staff years a year in FY78-80, as effort was shifted into general energy policy work; as one result, less time was spent on policy papers and guidelines. Though there are good reasons for this shift, it has meant that the staff available for work on power only have time for direct project-related advice, mandated policy work, and the minimum level of contact with consultants, manufacturers, aid agencies and the like that can be managed without being totally uncooperative. In FY78-80, lending-related activities accounted for half of the unit's work (two-thirds of the staff work load); research and policy work took up one fourth, and the remaining fourth was spent on training, personnel matters, external relations, and similar activities. 3.22 The zero-base-budget exercise of mid-FY78 proposed a "realistic" CPS power program of 6 staff years a year in FY81 and FY82 (excluding consulting assistance for research). This figure was based on a smaller FY81/82 power lending program than is actually being undertaken. EGY/PAU is trying to bridge the resource gap in several ways: the appraisal review process is being made more selective, with differing levels of reviews for engineering, finance and economics depending on the characteristics ofE the particular project; the computerization of sector statistical data and supervision files should improve efficiency; and some staff guidelines are being prepared under - 39 - research contracts, rather than in-house. But since these improveiments will not save very much time, it will still to be necessary to set priorities for the many tasks identified in conjunction with this paper. 3.23 One consultant position (probably a hydro specialist) had been requested for FY83 to increase direct support to operations (PAIJ provided virtually no direct operational support to the regions in FY78-81); however, approval for this position has not been received. In the medium term, establishing a technical core group of 3 to 6 specialists to provide direct technical support to regional operations would enhance project quality and might improve operational efficiency. This group could include professionals with in-depth experience in thermal generation, hydro-mechanical equipment, system operations, distribution engineering, rural electrification programs, system planning, ratemaking, utility finance, accounting and bil-ling, and perhaps nuclear power. A link to the energy assessments unit would permit early attention to technical issues as part of country sector reviews. 3.24 Dealing with enquiries from outside the Bank absorbs a large amount of the Power Advisory Unit's time (about 15%, on the basis of a 3-month sample). Given the importance of this task to the Bank as well as to those initiating the contact, a high degree of selectivity and tact is necessary to keep this task within manageable limits. Dissemination of the r~esults of these contacts is also a problem. In appropriate cases brochures and descrip- tive material (e.g. from consultants, contractors and suppliers interested in doing business with the Bank or participating in projects) are circulated to the regional divisions. Occasionally, short briefing sessions are arranged to describe advances in the state-of-the-art (15 such seminars in FY78-80), but more systematic circulation through a power newsletter or similar device could also be considered. D. Policy Work and Research 3.25 The principal policy guidelines for lending in the power sector are set out in OMS 3.72, which also covers Water Supply, Sanitation and Telecommunications. This OMS was last revised in 1978, and perhaps should be revised again to take account of power's separation from the other sectors. 3.26 No comprehensive sector policy document for power is currently in print and available for distribution outside the Bank. The last (only) such document is the 1971 Sector Working Paper, Electric Power. While this paper stressed the institution-building and resource-transfer objectives of power lending which were then dominant in the sector, current concerns span a much broader range. It would be desirable to update the sector paper, but for lack of resources, no such plans have been made. 3.27 Some 65 titles prepared since 1973 have been distributed to power staff in the form of Public IJtility Notes--now called Energy Notes (PUN/EGY)-- Research Papers (RES), and Guidelines and Standards (GAS), by the EWT/EGY departments. These are supplemented by Bank Staff Working Papers, sale - 40 - publications and Reprints (see Annex IV). The Energy Notes/GAS series include what is probably the most extensive list of titles available on the economic pricing of electricity, but they have little on the financial aspects of pricing and practically nothing on techniques of system planning and design, or on specific technologies such as small-scale hydro. Papers covering some of these subjects have been prepared but not incorporated into the series. As time is available the EGY power unit staff will summarize recent papers and research efforts. 1/ The EGY documentation system is to be reviewed to ensure that original studies and policy formulation documents are properly recorded and catalogued. Research 3.28 In 1979 the Bank's research in electric power was reviewed by a panel of outside experts. Annex I contains a summary of the findings of the panel, and recommendations for a future research program in power to deal with the many issues that have become prominent since the panel's review. These include system planning, conservation, loss reduction and demand management, and audit of organizational performance. The present level of research, 2.5 staff years in FY82 or about 2% of the Bank's total research effort, funded almost entirely from the EGY budget, is not adequate and more participation in the Bank's formal research program would be justified. This should be feasible, but in the past such participation has been limited by lack of staff time to manage the research. E. External Relations Conferences and Seminars 3.29 Given the Bank's role as the leading international financing institu- tion in the power field, its participation in international power activities is meagre. Although reasonably well represented at major conferences such as the World Energy Conference in Munich in 1980, the Bank's power sector is seldom represented at such major forums as ICOLD, CIGRE, UNIPEDE, OLADE, IAEA, etc. (see inside front cover for abbreviations). In the FY78-80 review period, power staff attended only 3 such activities, most of which were directed to specific-interest areas (e.g. Ecuador mini-hydro). The Bank is not an official member of any of these organizations (except the World Energy Conference) and routinely declines requests to lparticipate in most technical conferences. Since there is no single UN-sponsored agency dealing with electric power and energy, the channel usually used in other sectors for contacts with the international community is lacking. 1/ For example, on hydropower development in the developing countries, forecasts of power sector investment needs, summary of power sector country data sheets, and thermal generating plant efficiency improvement. - 41 - 3.30 Aside from partial isolation from the international technical community, which may have some impact on recruitment, it is difficult to identify any serious negative consequences for the Bank of its de facto noninvolvement policy. The Bank's international associates have a right to expect a greater degree of leadership and participation in activities related to the development process. Nevertheless, it is unrealistic to expect much change in this pol'icy with the present level of resources. EGY/PAU does expect to increase contacts with and participation in UN-sponsored programs in power and energy, such as the UN New and Renewable Energy Sources conference effort, the UN Centre for Natural Resources, Energy and Transport (UNCNRET), and the IAEA. In particular, closer ties with IAEA are necessary. 3.31 An attempt will be made to disseminate more information to regional staff on external professional and technical activities. Regional staff may participate on an individual basis as has been the custom, but this attendance will not be coordinated by EGY/PAU, although the unit would appreciate being informed of such attendance. Technical Literature 3.32 The main sources of power-related technical literature in the Bank are the Energy Library, 1/ periodicals circulation, and of course the Non- Regional Information Center (NRIC) and the Joint Bank/Fund Library. At present, the circulation of technical journals to EGY and the regional power divisions can take months, so that dated information (for example procurement advertising for Bank-financed projects, announcement of conferences and seminars) becomes virtually useless to the people at the end of the circulation list. While the divisions themselves can improve the circulation process somewhat, they should request their own subscriptions to principal technical journals. "Engineering News Record" (project advertising), "Public Utilities Fortnightly" (finance and accounting), "Electrical World", and "Water Power" are recommended publications, but each division may have its own preferences. The cost of this service would be small compared to the benefits. F. Professional Development Training and Short Courses 3.33 Professional development activities for power staff have received very little formal attention or encouragement. The Bank's stated policy on attendance at external seminars, short courses and professional meetings is quite liberal and approval for participation in such activities is usually obtained promptly from department directors, with registration fees paid by the training unit of PMD. Unfortunately, Bank procedures tacitly discourage 1/ Located in the Energy Department, D563. - 42 - professional development activities. Although internal technical seminars are held quite frequently, they are often sparsely attended because time spent on such activities merely increases staff overtime. Travel for external activities is generally approved only if it can he arranged in conjunction with other operational travel. Partly as a result, the training unit arranged only 13 such activities for power staff in FY 78-80, in which 9 staff (16%) participated, from four of the seven divisions. 3.34 Although these figures understate actual participation since they exclude activities undertaken by staff locally and on their own time, this level of participation is clearly below any reasonable standard of professional development. For example, some professional registration boards in the US are considering legislation which would require a minimum of about 3 days' participation a year in a related continuing education program, as a prerequisite for renewal of professional registration. Recently the Energy Library has attempted to collect and disseminate information on power-related short courses and seminars, but to be effective this should be formally incorporated in the Bank's training activities. Membership in Professional Organizations 3.35 No formal records of membership in professional associations are kept in the Bank and there is no explicit effort to encourage or recognize membership in professional societies, at least f'or power staff. Based on a survey of power staff, only half of the power engineers and virtually none of the financial analysts and economists are members of professional societies. Of those that are, most belong to at least two: typically the Institute of Electrical and Electronic Engineers (an international organization, with largely US membership), and often the TEE (UK) or an equivalent institution in the home country. Of the 20% of power engineers that were trained in specialties other than electrical engineering (civil, mechanical), about half maintain memberships in their respective technicaL societies. About one-third of the power engineers are registered professionaL engineers or the equivalent and belong to a professional association of engineers (e.g., UJS National Society of Professional Engineers). Less than one-third of the financial analysts are chartered accountants, certified public accountants (US), or the equivalent. 3.36 The staff survey also revealed that membership in professional organizations has declined over the years as staff members allow their memberships to lapse. This is understandable since annual fees in professional societies can be several hundred dollars. As one financial analyst said, "I feel 10 years behind, but I have other uses for the $400." Clearly, some effort by the Bank to encourage participation in professional societies would be desirable. This could range from simply encouraging members of local chapters of technical societies to publicize meetings and invite guest attendance, to keeping formal records and acknowledging the recognition of staff by professional societies, and, most importantly, to eimbursing part or all of the membership fees, as is common in industry in r.le US and elsewhere. The benefits of such participation are large. For the professional, it could mean a sense of being part of the professional - 43 - community, and for the Bank, a recognition of its standing by professional societies, including access to the recruiting systems sponsored by these societies. 3.37 Since these issues are common to all professional disciplines, PAS should establish a "professional office" to deal with professional development policy and activities, in cooperation with PMD. Department directors can authorize reimbursement of professional society membership fees, but this policy should be made explicit and uniform. - 44 - ANNEX I POWER SECTOR STATISTICS AND PROCEDURES A. INVESTMENT REQUIREMENTS AND BANK LENDING FOR ELECTRIC POWER 1. Energy in the Developing Countries (August 1980) included an estimate of the developing countries' investment requirement for electric power in 1981-90. This estimate is summarized in Table 1. Table 1. DEVELOPING COUNTRIES' ELECTRIC POWER INVESTMENT REQUIREMENTS, 1981-90 1/ (Commitments, 1980 price levels) - 1981-1985 1986-1990 GW $bn GW $bn Generating Capacity Thermal 62.4 47 81.7 69 Hydro 47.4 47 54.3 69 Nuclear 6.8 22 27.9 29 Geothermal 1.0 2 0.9 4 Sub-Total 117.6 118N 164.8 171 Associated Facilities Transmission - 17 - 24 Urban Distribution - 18 - 24 Rural Distribution - 16 - 26 Sub-Total 51 74 Total 117.6 169 164.8 245 1/ The distinction between commitments and disbursements in the power development program is particularly important because of the long construction times required for power projects and the consequently long disbursement periods. Because of growth rates, commitment requirements will exceed disbursements by one-third or more in any given year. This paper uses commitment estimates rather than disbursements because they better reflect financing needs. 2. Tables 2A, 2B and 2C show Bank lending for electric power in FY78, 79 and 80 respectively. Data are given for project components, loan amount, cofinancing, and staff input for project processing. The relatively large amount of cofinancing (60% of Bank lending) and the total project costs (four times the amount of Bank lending) are noteworthy. Table 2A FY 78 BANK LENDING FOR POWER BY SUBSECTOR, AND TIME INPUT Oil/Gas Total Nydro & Assoc. Thermal & Coal Thermal & T & D and Rural Total Bank/ Co-Financed Project Man Wks Man Wks Man wks Region Country Project Trans. Assoc.Trans. Assoc.Trans. General Electrif. IDA Loan Amount Cost Pre-App. App. Negot. --------------------------------------------------…11S Million…--------------------------------------------- EA Madagascar Andekaleka 33 33 67.8 116 103.6 57.2 39.0 Mauritius Power Transmission 15 15 15 - 18.9 21.1 28.2 10.7 WA Liberia Power IV 6 4 10 19.38 31.55 6 63.8 15.9 Sierra Leone Power III 1 7 8 5 15.2 14.6 74.3 26.6 EMENA Syria Regional Electrification 40 40 34.2 135.8 12.4 67.5 22.4 Yemen, P.D.R. Wadi Hadramout Power 5 5 - 28.3 27 70.5 11.0 Yugoslavia Middle Neretva Hydro 73 73 - 402.6 10.2 38 8.3 EAP Fiji Monsavu-Wailoa Hydro 15 15 34.9 54 20.1 53.0 15.6 Indonesia Power VII 84 25 109 - 161.53 6.8 61.6 15.2 Philippines Rural Electrification 60 60 - 160.5 1725 50.3 6.2 Thailand Pattani Hydro 43 7 50 - 144.6 25.9 54.9 30.9 Thailand Rural Electrification I 25 25 15 110 11.3 29 12.7 SA India Korba I Thermal 200 200 - 439 6.3 23.2 11.9 India Trombay Thermal 105 105 - 209.4 5.2 18.8 9.7 LAC Guatemala Chixoy Hydro 57 15 72 203 414.1 14.2 88.2 12.9 Brazil South-SE Distribution 120 10 130 - 2,346.8 36.3 56.3 20.0 Colombia San Carlos I Hydro 126 126 70 421.3 34.9 65.9 8.8 Colombia 500 KV Interconnection 50 50 40 167.2 75.7 33.0 7.7 Jamaica Power II 20 20 36.1 23.2 47.6 7.8 Total 348 97 305 256 140 1,146 489.28 5,412.88 - 46 - 001. ooc .r - q eO4 aCen0 z r O d.400 CC -444.O.4 4 < ~ ~ ~ ~ ~ - ene _ en _n _en___ y~~~4! C. le_n o1- -0o-41_4\ e 1vn K a . . .. . .-. 0<'n 0< -4 a--! ..C- c ¢ ° S E ^ r rr 4 en - m C- I v ~ ~ ~ e 0- rer .S UVI 6 S19V Cl '_ / ~~~~o o0 00 40 0 00 0 0:t h. so-. c~~ ~~~ a i -. C. 0- 0cIN oiIIct 0 C0, j r .0 0 0 O 0a0 00-0 ,00 M 00 - 0.00 0 000 .4C 04 0 0004 0j 000 r 00040 0: 0 ta' CF 0Ca 00> - 47 - a o_s s .10 o 0>0 N Eo.1.0.- _.C 150-a vo 0 o o0031 N N 5..' a sr eox 4aNNGNa<-fOl 'G I _a NO s sa-.s 4 ..4 a a 0-_ _ N 0 7 aN aa- 1-N C Ol a- a: I Na N al N N N a _N SN c c ! e o - ~~~~~~~~~~~~~~I ' < 'o. ~ ~ N c. a aol NosN .2 1 >~~~~~ Eao Na.as 0 >1 -0 _1 _ 0 rg - 48 - 3. Table 3 shows Bank lending for various periods from 1958 onward. Except for a drop in FY 72-75, average lending in real terms has been remarkably constant. The large increase from 1976 onward in the average number of projects processed each year should be noted. Table 3. ELECTRIC POWER LENDING, FY 58-86 FY58-67 FY68-71 FY72-75 FY76-81 FY82-86* Annual Average, current $m 288 420 529 1353 1681 Annual Average, 1981$m 1575 1522 1161 1634 1404 Annual Average, No. Projects 12 15 13 19 20 * Projected Programming Bank Power Lending 4. Table 4, which is based on a 5-year review, shows the expectations versus the actuality of power projects which were in the pipeline in mid- FY74. It points out that projects mature at different times and that there is no orderly progression from identification to preparation to appraisal. Table 4. PLANNED AND ACTUAL POWER LENDING, FY74-78 (No. of Operations) FY74 FY75 FY76 FY77 FY78 Total Planned for Year, as of mid-FY74 27 28 23 19 20 117 of which Executed in Year Planned 15 4 7 5 4 35 Executed Earlier than Planned 0 2 2 1 4 9 Executed Later than Planned 9 14 7 9 7 46 Dropped 3 8 7 4 5 27 Added in Period 1 8 8 9 7 33 Actual Operations 16 11 20 17 19 83 - 49 - 5. It is difficult to make firm forecasts of Bank power lending operations, even for one year ahead, but recently the forecasts have improved somewhat, as can be seen from Table 5. Table 5. PLANNED AND ACTUAL POWER LENDING OPERATIONS, FY79-82 FY79 FY80 FY81 FY82 No. Amount No. Amount No. Amount No. Amount ($m) ($m) ($m) ($m) Operations Program as of: 12/31/77 28 1674.0 12/31/78 24 1727.1 33 2311.0 12/31/79 30 2647.6 24 1744.1 12/31/80 25 1985.0 4:2 2658.0 Actual Lending 18 1354.9 25 2392.3 17 1323.0 22 2131.2 6. The large difference between the operations and lending programs is characteristic of the power sector as indicated by the large percentage of "standby" projects (typically 40% of the operations program). Alt:hough this appears to parallel experience in other sectors, unit budgets sometimes do not reflect this. The relatively large increase in the agreed operations program (29%) over the last four years is also significant because it is not backed by specific additional resources, implying that resources will need to be diverted to lending operations from other activities in the sector if this program is to be achieved. The large year-to-year variations in lending are also typical, even though average annual lending is quite stable. Bunching 7. Because of the high proportion of "standby" projects, the bunching experience in this sector is somewhat more severe than the Bank average, as indicated below (averages for FY 78-80)e On average, 70% of the power program goes to the Board in the last four months of the fiscal year. Projects Presented in Quarter I II III IV Power 5 18 19 58 Other 11 19 25 45 So far, regional managers have been able to cope with this situation by judicious scheduling of staff, but some effort to even out the flow would undoubtedly be worthwhile. The bunching problem is also apparent at earlier - 50 stages: 80% of the yellow cover reviews in FY81 were scheduled in two 3-week periods. Thus, what is widely perceived as the principal EGY/PAU advisory function (the yellow cover review) is, in fact, a marginal activity. Although there was somewhat less bunching in FY 80, the programs for FY81 and FY82 showed no change from the earlier pattern. B. ELECTRICITY AND ENERGY B3ALANCES 8. Electricity is both a primary commod-Lty and a secondary product (electricity produced from hydro, nuclear and geothermal sources is primary energy; that produced from fossil fuels plus l:hat from primary sources is secondary energy). Electricity is a premium "fuel" because of its conven- ience, and because it permits high efficiencies in end-use conversion. 1/ Therefore, for other than large-scale low-quality heating applications, electricity is not necessarily a less efficient route from primary fuel to end use. The percentage share of electricity in the commercial energy mix can be shown in three ways: 2/ Electricity Petroleum Coal, other (1) as Primary Energy: Developed countries 4 73 23 Developing countries 3 82 15. (2) as Secondary Energy: Developed countries 15 74 11 Developing countries 10 79 11 or (3) Adjusted for End-use Efficiency: Developed countries 35 56 9 Developing countries 25 66 9 1/ Low conversion efficiency (25-30%) is often cited as a major disadvantage of electrical energy from thermal stations. However, this conclusion may not be warranted when one looks at the end use of energy. Electric motors are 85% efficient compared with an in-service efficiency of 15-20% for diesel engines; electric cookers and water heaters are twice as efficient as oil, kerosene, or gas-fired units, and electric lighting is several times more efficient than competing lamps. Moreover, electricity is usually derived from lower quality fuels, and electric equipment (e.g. motors) is usually less expensive than utilization equipment for other "fuels". 2/ Data for 1978 from UN World Energy Supplies, Series J. 51 - Presentation in the third form, adjusted for end-use efficiencies, probably gives the most accurate reflection of the role played by electricity in the energy sector. It is partly for this reason that EGY/PAU recommends a conversion based on the amount of fuel needed to generate 1 kWh of electricity, rather than the energy content of that kWh (0.25 kgoe/kWh or 4,000 kWh/toe), when comparing electrical energy with other energy fonns. Presentation in this form also helps explain electricity's high share of total energy investment. C. OPERATIONS PROCEDURES IN THE POWER SECTOR 9. Bank power lending operations have quite low average coefficients, as Table 6 shows. This is derived historically from the Bank's practice of using small teams (typically, one engineer and one financial analyst) for project appraisal and supervision. Over the past six to ten years the Bank has introduced a substantial number of mandatory features into the project process. l/ While these innovations produce better reports and projects, there is growing concern among operations staff that the traditional planning and engineering aspects of projects are now receiving less than optimal attention and that there may be a corresponding decrease in the direct technical assistance content of power operations. 1/ Project briefs, issues/decision papers, identification of poverty-target beneficiaries, expanded economic analysis and mandatory IERR calculations, efficiency prices, environmental impact analysis, presentation of consultants' costs, semi-annual and quarterly disbursement estimates, efficiency and social price adjustments in economic analyses, energy resource assessments, formal post-completion evaluation, etc. Table 6: RANKING OF SECTORS FOR FY 78-80 BY AMOUNT OF LENDING, NO. OF PROJECTS, STAFF TIME AND STAFF NOS. Input to Lending Staff Operational Staff Amount of Lending Projects Operations Years/ as of 9/30/80 CPS "Other Output" US$Millions % No. % Staff Years % Project No. % Staff Years z Agriculture 9,250 30.9 256 35.0 959.8 45.7 3.7 301 32.8 77.4 41.6 POWER 4,893 16.4 62 8.5 108.1 5.1 1.7 52 5.7 9.9 5.3 ---------------------------------------------------------------------__------__--------------------------------------------------------------------__-- Transport 4,442 14.9 99 13.6 235.2 11.2 2.4 136 14.8 34.0 18.3 IDF 2,650 8.9 73 10.0 137.7 6.6 1.9 85 9.3 17.1 9.2 Water Supply 2,025 6.8 54 7.4 107.8 5.1 2.0 59 6.4 17.1 9.2 Industry 1,744 5.8 30 4.1 117.4 5.6 3.9 59 6.4 n.a - Education 1,288 4.3 51 7.0 157.3 7.5 3.1 73 7.9 30.8 16.4 PL/SAL 1,085 3.6 17 2.3 - - - - - n.a - Urban 1,027 3.4 31 4.2 133.9 6.4 4.3 56 6.1 n.a - Oil and Gas 497 1.7 17 2.3 52.1 2.5 3.1 32 3.5 n.a - Telecommunication 462 1.5 11 1.5 22.5 1.0 2.0 14 1.5 n.a - Population 315 1.0 10 1.4 50.2 2.4 5.0 37 4.0 n.a - Tourism 163 0.6 7 1.0 18.9 0.9 2.7 - - n.a - Technical Assistance 63 0.2 12 1.7 - 15 1.6 n.a - Total 29,904 100 730 100 2,100.9 100 919 100 186.3 100 Note: Data is aggregate for three years FY78, 79 and 80. - 53 - Identification and Preparation 10. The preparation of most projects is left to borrowers and their consultants with only minimum input or guidance from Bank staff. Project identification at the CPP stage is not very specific excepi: where rural electrification has been identified as socially desirable. Before FY81 the Bank was typically involved too late to make a meaningful contribution to feasibility studies or the formulation of development programs. Since FY81, however, the routine use of the Project Brief system has proven beneficial, and substantially more issues are being identified and resolved prior to appraisal. In some cases, this has led to delays in projects, and manpower ccefficients are likely to increase, but overall the change is perceived to be very worthwhile. 11. The increased emphasis on project preparation is leading to an earlier consideration of conditions before appraisal, and thus the timing of appraisal itself has become a useful tool in achieving project objectives. It is increasingly common to make appraisals conditional--for example on the completion of sector or project studies, identification of outline financing plans (cofinancing) or compliance with undertakings under previous loans/credits. Appraisal Practices 12. Project justification generally is well defined by the time the appraisal mission departs and therefore the emphasis during appraisal is on the institutional characteristics and especially the financial forecasts, since preliminary financial statements are seldom available in the Bank before appraisal. Still, the focus on program schedules occasionally leads to situations where appraisals are undertaken before it is entirely certain whether all required information (for example, financial forecasts) will be available. 13. Because the teams assigned to power operations are small, comprehensive sector reviews are rarely done in conjunction wilth appraisals and therefore, for example, an appraisal of a generation project will have little to say about distribution, and vice versa. Such coverage is further inhibited by the fragmentation of the sector in most countries, whereby different agencies are responsible for generation and distribution, or for separate geographical regions. Some increased emphasis on formal sector work and energy sector assessments has helped to fill this gap. However, a more deliberate effort to ensure full sector coverage either through judicious project selection or emphasis on "soft" areas by means of ad hoc reviews is necessary. 14. As in most other sectors, power appraisal reports are highly stylized documents, whose format and organization may discourage discussion of new subjects which do not fit the usual mold. For example, in projects whose justification depends critically on efficient implementation, something more than the usual short paragraph on project execution would be useful. This - 54 - would be especially true in cases where cofinancing may dictate how project procurement is to be packaged, or where a large part of the work is to be done by force account. In the latter, specific detaiLls on the amount of staff and equipment available and the scale of work compared with past activities are relevant. An updated checklist or guideline for power appraisal reports may be needed to assist in avoiding errors of omission. 15. There is some inconsistency among the regions with regard to the drafting of loan documents; some (especially those with a large complement of newer staff) do not play an active role in converting the recommendations of the appraisal report into legal undertakings. Given the significance of these documents (the only substantive residual element of the appraisal process) more attention to this phase of appraisal is warranted. Consistency of treatment of financial covenants in particular may warrant more input by EGY beyond the yellow cover stage than is provided by the present review procedures. Supervision Practices 16. Project supervision absorbed 23% of the power divisions' work in the review period--a share fairly typical in the Bank. The share increased slightly over the period but not enough to offset the increase in the number of projects being supervised (136 as of September 30, 1980). Overall, the average coefficient dropped slightly, from 6.9 staff weeks per project in FY78 to 6.5 in FY80, the lowest coefficient of any sector. 17. As with project appraisal, the supervision effort applied to individual projects varies widely from less than 1/2 staff week a year to over 30 depending on the stage of execution, project: issues, etc. Again as with appraisal, there is no clear correlation with type of project, region, cofinancing or other characteristics, but there is a country correlation. Power projects are not immune to problems in "difficult" countries despite their infrastructural character. 1/ 18. Supervision missions are usually combined with preparation or other activities, although rarely with appraisal. Exceptions to this rule usually involve "repeater" cases dealing with the same entities where supervision is an integral part of the appraisal; in such cases, separate supervision reports may or may not be issued. Multi-country missions and one-man missions are common. In fact, comprehensive supervisions are becoming rare. Most supervisions are addressed to specific issues or "fact-finding", with over half of the supervision effort being directed to financial issues, especially performance under pricing and earnings covenants. The result is that monitoring of ongoing projects in non-borrowing countries has sometimes been / See FY80 Fall Project Implementation Review for a discussion of this issue. - 55 - neglected (Ireland, Sri Lanka). In general, the tight schedules leave little scope for in-depth examinations to anticipate future financial or implementation problems. 19. In this respect, inclusion of monitoring indicators in loan agreements has not been particularly useful. There are two main reasons: either specific targets are lacking, so trend analysis is not meaningful, or data are not available for cross-correlation with other countries or borrowers. EGY/PAU is sponsoring a research project which is expected to fill these gaps. 20. Because of the issue-oriented nature of most power supervision missions, full supervision reports in accordance with OMS 3.50 are infrequent. Instead, most divisions use a streamlined format consisting of the Form 590, a page or two of explanatory text (sometimes following the "compliance with covenants" outline), and the usual draft letters as annexes. Aides-memoire are not seen frequently but those divisions that do use the procedure consider it useful. The uniformity of approach has helped managers (and EGY/PAU) to review of mission findings and focus promptly on issues raised by the missions. 21. Procurement activities account for most of the supervision workload of engineers and about half of the total supervision effort. Three divisions have procurement specialists assigned to this task and in general this system works well--regional consistency is assured and these procurement specialists tend to stay more up to date than their colleagues on current procurement issues and policy. The disadvantages include a variable workload (the procurement workload is not sufficient to justify a full-time position in most divisions) and some lack of coordination with the project officers. Resident representatives have occasionally helped with procurement work. The Inter- American Development Bank has power sector representatives in its major countries, using this supervision activity as a training program, and seems satisfied with the results, but proposals for a similar system for the Bank have met with a lukewarm response. If additional resources are to be provided, managers would prefer to augment their own staff. 22. Power procurement tends to be quite complex. Pre-qualification is used extensively for civil works. The long disbursement periods for many contracts make choice of proper escalation formulae mandatory, and the high premiums placed on efficiency demand precise specification of the requirements for bid evaluation, tests, and performance clauses. Given the complexity of this work and the wide range of technical and commercial issues faced in most power procurement activities, disputes and problems are surprisingLy rare and this in itself is a compliment to Bank staff dealing with this work. None- theless, recent contacts with consultants and others indicate a need for improvements in specifications and evaluation criteria for energy conversion equipment, since the capitalized cost or benefit associated with marginal differences in efficiency can be of the same order of magnitude as the first cost of the equipment. EGY/PAU is undertaking a small research program to provide staff guidelines for specifying power generating equipment. - 56 - 23. Power was not identified as one of the sectors with major problems in the Fall '80 and Fall '81 Project Implementation Reviews. Power projects with moderate and major problems were close to the average Bankwide, and inconsistencies due to Form 590 rating criteria were no better or worse than in other sectors. However, the trend of projects with moderate and major problems was upward, from 50% in FY78 to 60% in FY80, due primarily to difficulties in adjusting tariffs and prices to meet rising costs. Although the proportion dropped to 47% in the Fall '81 project implementation review, this seems to reflect more a shift in the perception of the seriousness of defaults under financial covenants rather than a real improvement since the trend of such defaults continued to rise--two thirds of the earnings covenants were not being met as of end-81. Economic and Sector Work 24. In FY78-80, 18 formal energy sector reports were produced; 12 by the regions and 6 by EGY. Of these, 7 were sector studies, 3 were sector reviews and 8 were sector memos. Each of these considered energy an integrated sector and provided a focus which was useful in dealing with the macroeconomic issues generated by the post-1974 oil crisis. Even so, economic and sector work has not played a significant role in power division work programs. In FY78-80, it accounted for only 4% of direct operational effort and half of this was spent on formal reports, mainly three large studies (Yugoslavia, Indonesia, Thailand). 25. This apparent lack of emphasis on sector work has several causes. First, electric power is not perceived as having had a large impact on development and it receives little emphasis in country economic reports and CPPs. Therefore, there is little demand from the Programs side for power sector work as an input to country economic planning. Secondly in power, unlike in some sectors, few countries have a defined lending strategy supported by a series of projects. Thus, there has been little direct operational benefit from sector work as a means of identifying issues to be treated through a series of lending opportunities. Finally, because of the Bank's sectoral approach to power project lending, where the emphasis is on least-cost program planning, institution building and adequate financial performance in the sector, the project is automatically placed in a sectoral context thus making separate "sector" work redundant. 26. Much of the appraisal work really deals with sector issues and is, in fact, sector work. One possible benefit of increased emphasis on upstream sector work, as recommended in this paper, would be a reduction in the sectoral aspects of appraisal, releasing resources for more work on the project issues. The formal sector work activitLes described in OMS. 1.13 are not given high priority by power division staff, and "informal" sector work (that which does not result in a report) plays a much larger role in the sector. Sector notes, not entered into the formal PAB code system, are frequently prepared as "pre-project briefs" or as inputs to CPPs. The sector work component of project briefs, recorded as project preparation effort, is large. - 57 - D. POWER ADVISORY UNIT PROCEDURES Project Review Process 27. The Power Advisory Unit contributes to the project review process through (i) technical advice and review, (ii) Bank policy and guidelines, and (iii) staff training. The major comments on project concept and related engineering questions are raised at very early stages in response to project briefs or, more often, during informal discussions with staff. 28. Procedural issues--such as the calculation of foreign exchange costs, disbursements, choice of borrower or onlending--usually surface at: the issues paper/decision stage. By this stage, it is too late to address major issues such as pricing and the sector development program since these should have been an integral part of the appraisal, yet it is too early t:o agree on specific financial targets since the analytical work on which to base them has yet to be completed. This latter problem can be managed during the later stages of appraisal, but the former requires more specific input as early as the CPP stage. Pre-mission issues meetings have sometimes helped staff to develop a strategy to deal with the issues expected to be encounlered during appraisal. 29. Advice concerning methodologies for calculating rates of return and similar issues is given during the preparation of yellow cover appraisal reports and at the preparation stage. Most of the advice on financial matters can be given only after the yellow cover report has been issued, since it is only at this stage that the detailed financial forecasts and analysis are available. Similarly, specific discussion of institution-building recommendations and project implementation arrangements is possible only once these have been spelled out in the SAR. Training needs, such as guidance for new staff on presentational format, and the position the Bank should take on such questions as treatment of customer contributions and shareholder equity, often become apparent only at the yellow cover stage. EGY/PAU also coordinates review comments from OPS and conveys these comments to the regional divisions. Quality Control 30. In accordance with current Bank policy, responsibility for project quality rests with the Regions. OPS/ENI acts in an advisory capacity as necessary to provide functional support. EGY/PAU gives this support selectively, but contributes to every project. An issue, however, is emerging as a result of the policy of selecting unit managers on the basis of their management capabilities rather than their technical expertise, and because Assistant Projects Directors are no longer expected to have technical knowledge of their assigned sectors. More specifically, only 4 of the 7 regional divisions have managers with pre-Bank experience in electric power. Thus, technical review of projects above the staff level is becoming increasingly less common; this situation may not be acceptable in view of the relative number of newer staff and the limited technical capacity in some subdisciplines. The present EGY/PAU review procedures provide some - 58 - backstopping in such cases and, pending the adoption of some system (e.g. matrix management) that would provide technical review capacity at the regional level, EGY/PAU will continue to provide such support. 31. As an interim step, EGY/PAU has offered to coordinate a "peer-panel" system to assist staff in identifying and dealing with technical issues in project preparation and appraisal. Meetings with colleagues, probably from other regions, knowledgeable about the particular issues (e.g. operations and dispatch, financial management systems, distribution design, power pools) would be arranged on a voluntary basis. This has not met wide acceptance, and a similar system initiated several years ago by TWT/Water Supply quickly fell into disuse because regional managers did not support it. Also, substantial effort to develop a "human resources file" is necessary since the personnel records of individuals commonly do not include data on areas of expertise. Nonetheless, the potential benefits of such a system would seem to justify more effort to gain managerial support and to develop the necessary data base and administrative procedures. Supervision and Audit 32. Supervision accounts for less than 5% of EGY/PAU effort. Most of this time is spent on the review of draft project completion reports and preparation for the annual audit-review seminar. Supervision reports (some 200 a year) are received in the unit and scanned for items of significance. When issues occur and as time permits, marked-up reports are returned to the region with requests for clarification or indication of intended action. In the few cases where this is done, no formal record of these comments is maintained nor is any attempt made to keep track of performance trends, monitoring indicators, etc. except when the same problem occurs repeatedly, (for example noncompliance with earnings covenants). The EGY/PAU overview function would benefit from a statistical analysis of supervision findings and hopefully this can be started once computer capability and the appropriate manpower is made available to the unit. 33. The annual review of OED power sector audits conducted by EGY/PAU in the past has attracted little attention from the regions for two reasons: (i) relatively few projects are reviewed each year--only 6 in 1980, and (ii) the issues raised are either perennial (failure to meet the earnings covenants) or are perceived as irrelevant to ongoing operations. (For example, cost overruns still appear as major problems in projects now being covered by completion reports, but the revised guidelines and practices on contingency allowances and preparation status (OMS 2.28) are considered to have long ago done as much as possible to anticipate or control unexpected cost increases.) 34. In order to make these reviews more issues-oriented, with the 1980 review EGY/PAU departed from the strict format of simply reviewing the issues raised in PPARs for projects completed in the calendar year. Instead, a concerted effort was made to address the issues that appeared in projects - 59 - presently under supervision. The results were worthwhile and the meeting was well attended. In the future, however, more involvement of the regional divisions in determining the agenda and schedule will be sought. 35. Ideally, an annual seminar for power staff similar to that conducted for water supply staff would be useful, both to transfer information and to provide a vehicle for lateral communication among staff of the regional divisions, but again resources for such an activity simply are not available. Given the large number of new staff, EGY/PAU has arranged a series of power economics seminars, and will expect to repeat these from time to time. The issues raised in this SSSP have provided a useful focus for these seminars. E. POWER STAFF CHARACTERISTICS 36. Engineers and financial analysts are the dominant technical specialists in the power divisions. At end-1981, the distribution was3: Power engineers 33 Financial Analysts 25 Economists 7 While all of the engineers have formal training in engineering, five of the financial analysts and six of the economists also have degrees in engineering, indicating that the technical knowledge is much deeper than might be inferred from their titles. 37. As of September 30, 1980, power engineers had an average of 5.2 years' of Bank experience. The upper limit was 11 years and the distribution fairly uniform. The average experience of financial analysts was about the same (4.6 years) but nearly three quarters of them had less than five years' experience. However, by mid-1981, after the addition of new power engineers is taken into account, average Bank experience drops to about four years, and one-third have less than two years' experience in the Bank. This uneven experience distribution has been taken into account in formulating the training activities proposed in this paper. 38. The average age of power engineers, 51 years, increased slightly over the FY78-80 review period. Only four engineers (15%) were under 45, and 12 (nearly half) were age 55 or over and thus eligible for retirement, although it is expected that most will elect to stay until their normal retirement date. Conversely, 11 financial analysts (58%) were under 45 and only two over 55. Their average age (43) declined slightly in the review period. This profile has not changed significantly over the years. Typically, power engineers are seasoned mid-career specialists while financial analysts are younger and economists are somewhere in between. 39. Only two engineers have moved elsewhere in the Bank (outside the sector) since the 1972 reorganization. No Bank staff from outside the sector have been appointed as power engineers. Conversely, at least six financial 60 - analysts have moved from power into other functions (notably to field offices and IFC); 5 of the current complement came fromL the YP program. The relative lack of mobility for technical specialists needs to be taken into account in formulating career development goals and guidelines for staff because, given the relatively advanced point in their career when power engineers join the Bank, it is unrealistic to engender career expectations outside the sector for most. For financial analysts on the other hand, contrary to the impression held by most, the prospects for transfer and promotion outside the sector are quite good. F. RESEARCH Research Review Panel 40. The Bank's research activity in Energy, Water and Telecommunications was reviewed by a panel of outside experts in October 78-March 79. The review panel recommended the following with regard to electric power: No significant change in resources applied to electric power research. Wider dissemination of the results of research activities in rural electrification and power pricing. Future research on public utility pricing under conditions of inflation, social pricing to determine implications for income redistribution, and institution building. -- No further work in valuing economic benefits of electric power. -- Further integration of power work with other energy work. -- More regional stimulation of research proposals; better coordination and collaboration with research activities of other departments. New Issues 41. These recommendations were made in the context of FY73-77 operations when lending for power was expected to stabilize at about $900m/yr (contrast $2400m in FY 80). Further, this was before the real implications of the energy crisis for Bank operations had been evaluated, and at a time when the small advisory staff was preoccupied with laying the groundwork for moving into oil/gas operations. Not surprisingly, many research topics which have now emerged were not considered by the panel. Among these are: conservation, loss reduction and load management in distribution and in thermal and hydro generation; valuing and pricing embodied energy in enclave projects for world- traded commodities (aluminum, fertilizer); 61 - -- implications of construction delays for project justification; -- impact of assumptions about price elasticity and oil substitution on load forecasts; -- optimizing multi-purpose projects; -- problems of long transmission lines with light loads; -- organizing and financing small-scale hydropower developmenlt; -- consistent criteria for evaluating hydroelectric resources; -- definitive methodologies for economic evaluation of rural electrification projects; -- reconciling economic and financial pricing objectives for public utilities; *- performance audit monitoring criteria; *- fiscal and budgetary constraints on the magnitude of power development programs; -- review of computer models available for power system planning and operations; -- improved methods of load management, hydrologic forecasting, and system operations, and -- role of electric power in industrial development. 42. Each of the above subjects features in one or more of the power projects in the prospective 5-year lending program. At present the resources available for power research (about 2.5 staff years in FY82, including consultants) are being applied to issues, principally loss reduction and small-scale hydro, which are believed to have the greatest scope for immediate application in the operations program. Next Steps 43. Past work on the supply side has stressed system plann:Lng and the economics of supply. The Department now has available a power system planning model (WASP, version III). This is an up-to-date and versatile analytical tool for testing the appropriateness of generation expansion programs proposed by borrowers and/or consultants. On the demand side, work on pricing issues has been consolidated through publications (e.g., the book Electricity Economics, Staff Working Paper 340 on "Electric Power pricing Policy", and the book, Electricity Pricing, Theory and Case Studies - and tariff seminars in four regions (South Asia, East Asia and Pacific, Latin America and 1l Munasinghe and Warford, World Bank, 1982. - 62 - Caribbean, and East Africa). Considerable success has been achieved, particularly in the developing countries, but the effort to apply research results in operational work must be maintained. A very modest EGY contribution is required to help the EMENA and West Africa regions hold tariff seminars. Otherwise, no new research or documentation is proposed in this field. 44. The results of ongoing research on rural electric networks and the reduction of losses from distribution systems will be incorporated in guidelines/checklists for operational staff. The final report of research project 671-86, Standards of Rural Electrification (R.E.) was completed in 1981. The study involves the review of existing R.E. institutional frameworks and engineering standards and practices in LDCs, the development of an economic-engineering model to optimize networlc design, and the practical implementation of the methodology in two case studies of R.E. networks in Costa Rica and India. Because of the increase in R.E. lending, the almost complete dependence on consultants for network design, and the very diverse (and sometimes inappropriate) practices adopted in different countries, the results of the research project should be quickly adapted for application in project work. 45. Research project R633, Distribution System Loss Reduction, was completed in mid 1982. It includes a review of the recent techniques for reducing distribution losses, the development of an economic-engineering model and computer programs for optimizing technical loss levels, and application of this methodology to several LDC distribution systems. The development of a checklist/guidelines based on this work will have a high priority. There appears to be good reason to suggest the adoption of a policy of routinely assessing system losses during any power lending operation and, if necessary, the setting of acceptable loss targets (akin to financial targets) for utilities. 46. Demand management and conservation should receive high priority in the research program. The benefits from more efficient use of generating capacity (e.g., a shift from peak to off-peak use) and lower consumption may warrant substantial investments in advanced solid state metering, microprocessor-based switching, etc. Through its contacts with suppliers and outside agencies (EPRI, EEI, USAID), EGY/PAU is attempting to identify state- of the art studies in this field, so far without success. The unit intends to develop policies, guidelines and checklists which Bank staff and consultants could use routinely to recommend to borrowing utilities, and for preparing outline TORs for energy audits and efficiency studies of large users. 47. One of the so-called "hard" tools of demand management is load control, which is particularly effective in limiting consumption in the short run. The use of load control techniques should be closely coordinated with the "soft" techniques of demand management such as pricing or financial incentives. Several of the Bank's studies offer guidance on the use of the latter methods, which are more useful in the longer run. EGY is now undertaking a study of load control techniques to review the state of the art, recommend promising areas for application in developing countries, and carry - 63 - out several pilot studies perhaps in conjunction with ongoing lending operations. The main potental for application would be in power supply to industry. To follow this project, it is hoped to study how industrial consumers use electricity, their sensitivity to changes in the level and structure of prices, and the flexibility of production processes with regard to substitution between electricity and other forms of energy as well as between energy, capital and labor, etc. 48. The research panel recommended that the Bank support the research needed to monitor and adapt new technologies including those for cogeneration and autogeneration, larger scale mini-hydro programs, wind and solar generation, and so on. At present there seems little need for the Bank to devote resources to the technology of alternative energy sources, but work on the application and economic evaluation of these sources is not well organized elsewhere and may be an appropriate Bank activity. Only work on small-scale hydro is being considered under the power research program; other technologies are being covered by the office of the New and Renewable Energy Sources Adviser. 0..-. 0. ooo 0. - '0 0.0 0 oo 0 0 0 o- Ko 0 0 00.0 00 0 00 0 o0.0000oO0.0. 0- 0-'  0.0.0000 - '1000.000.00.0 0.0.00.0.0000.0..Z.-00. 0<0.0.0000.00 0.000-'o-o 0000000000.0.0.000. .00.0-OOr 00.0.00.0 0. 0.000.001'000.000.0.000 flbj0.OO0O 000000 0.0.00 0. '-1 0 000.0.0.0.  00..o0'- Oooooo.ol 00.00oo000oo0.s0oo'10.0o000 00. - 0 0.0000< 0. 00.0000.000000.000000.000000000.0. 000.c0.000 0.000 0 0.0.  000.0.0.0000000.0.000 51. 00.000.0.00.0000.00.00 0.'lo..0000 0000 -0.000.000-00000- 000000.0.0.000.00 o-.0.ooooo 00 0.0.000.000.0000. C o00 00 000000000. 0 00 0000.00 0'10 0.0.0.00000000.0000] 0.0.0000.0. '10000.0.0. 00.000. 00.000.0.00.00 00.00'00.0.  000.0. 000 0.0.0000.0.0 - 00 00.0. 0000 0. 0.01 0. 0. 00000.00. 00.00.00.0.0' 000.00.0 0.00.00.0 0000 000.0.0-0 0 0' 00 00. 000. 000 000.0000 0.0 0. 0 0.0. 00.00 0 0.0. .0 0. <0-0 00 0. 00 00- 00000.0.000000000 1400. -00-0.0000 0 0 0 0 0.00 0 0. 0000.0. 00. 0.0.00.0000.0 0. 0. 0 0 0 'o 0 0-n. o 0.' 0. 000.0 00.0 0.  0.0 00.0000.0 01.0 0 000.0 .000. 0.0 0 00 0 0 0 00. 0. 0. 0 0. 0 00. 0 0 0.0 00 0. 00 0 0 0 0 00 0. '0 0.0 0 0 0 0. 0 0 0 0 0 0 0 00.0. 0. 0 0 0. 0 0 0. 05 0 - 00 0 0.00. 0 0 0 0 0 0 0 0 0 00 0-0. 0 0 0. 0 '0 0. 0 000 00. 0 0 0 0 - 0 000. 0 0. 10 0 0 0 0 00 0 0 0 0 00 0. 0. 0 I 0. 0. 00 I '0 II I 000000000000.00000.0030.0.0.0.0.0. '0.000000000000.0.0.0.0.0.0. 000.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.00.0.0.0 0.0.0.0.0.0.0.0.000.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0. o0. 00 10 0 0.0000010.0.0.00.00.000.000.0000000.0.0.0.0.0.0.0.0000 000.0.00.0.000.00.00.00.0.00.300. 000000.00000.0000.000.000000 0 0.0.000.000.0.0000.0.0000.0.0.000000000000.0.0.0.0 0000000.0.0.00.00.00.10'0.0.0. 000000 0.0(000 ii I0.0000000000000000000000000000000000000000.0000000000000000000 00000000000000000000000000000000000000.o0.o0..o 00. L 0. 00 oo 0-.. 1000. 0. I 0.000. 00 0. 0. 0. 0.0.0.0.0.0. 0. 0. 0.0.0.0.0.000.00.0.0. 0. - 0. 0. 0.0 0.0. 0. 0.0 00 0.00- [0.000.0.0.0.0.0.0.000.0.0.0.0.000.0.000.0.00 I 00.0. 0. 0. 0.000. 0.00.0.000.00000.0.000.00.00.0.00.V000 0.000.0.0.0.0.00.0.0.0.0.0.0.0.0.0. 00.0.0000.0.0.0.0.0.0.0.0.0.0.0.0. 0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.000.0.0.0. 0.0.0.0.00.0.0.0.0.0.0.0.0.0.00.0'' 0. 000.0.0.0.0.0.0.000.0.0.0.0.0.0.0.000.0.0.0.0.0.00. 000.0.0.0.0.0000 0 0.0.0.0000.0.000.0.0. 00.0.0.0.0.000.0..00.000.0.0000.0.0.0.0.000.0.0.0.0 00.0. 0.0.0. 0.0. 0. 0. 0. 0.0. 00 0. 0.0. 0. 0. 10. 0. 10.00. 0. 0 0. 0. . 0.00.0 0. 0.00.0.00.0.00. 0. 0.0.0.00.000.0000.0.0.0.0.0.0.00 0000.0.00.0.0.0.0. 10. 0.0. 0.0.0.0.0.0.000. 0. 0.0. 0. 0.0.110 0.000.000.0.0.0.000100000.0.000.0000.00.0.0.000000 l001l0100.0. 0000.0.0. 10. 10 00.00.000010.00. 10.00. 00110.0000.0.01011000.0100.0 0000. 00 0. j0.00 0. 0. 0. rI oo,'0 I' V 0000.0.00. 0 0. 0. 0. 0. 0.0. 00 0. 0.0. 00 0010.1 '0.0.000.0.110.1000100000.000.0.0.0.00.0.0.010.00.0.01 00 0.0.'I0.100.I0. II 0. 10.10.00110.10.0.10.10.0.0.0010.0000.0.000.0000.0000.0.0000 00.0. 0. 0 000 00.0.0. 000000 0 00.0.0.0000.000 0. 0. 00 0. 0.0. 0. 00.0 0. 0.000. 0 0.0.0 00 000 Ig'0 I. L Ion 0.0.0. 0.0.0.0.0.000.0.0.0.0 00.0.0.000000.0000.0.000010.0. 1000.10.0.010.10.0.0.0.0000. 100.000.0.00.0.0.0.0.000.01 0000 0.000.00.0.00.0.0.0.0.0.0.0.000000.0000.0.0.0.0.0.00.0.0. 0.0.0.0.0. [o 000.0. 0. 0. 00 0. 0. 0.0 V 0. 00000000.00.0.00. 000000.0.00.0.00. 000.0000.0.0.000000 Jo 0.0 00. 0 0. 0.0 0. 0000. 0.00 000.0.0.0.000000.0.0.0 II --- 0 000. 0.0. 0 0.00.0. 00 0. 0. 0.0. 0. 0 0. 0. 00 0000. - 00.0.0.0.0.0.0.0.0.0.0.000.0. 0000.0.0.0.0.0.0.00.000.0.0.0.0.0.0.0.0.0. 0.0. 000 000. 00.0.0.0.0. 0.0.0.0.0.0.0.00 0. 0.0.0.0. 0. 0. 00 0.00.  0000.0.0.0.0.000.0.00000. 0000.00.0.0000000.00.00000000000 000.0.0.0.000.0.0.00.0.0.00.0.0.00.0.0.0.0.0.00000. 000 0 0. 0.0.00.0.0.0.0.0.00.0.0.0.0.00.0.000.0.000.000.0.0. 0l0.0000.00000.001000.0.0.0000 00000.0. 0.00l0000l0.00.000.100.00 0 I 00.00000.00.00000.0.0.0.0.0.0 0. r 0.0.0.0.0.00.0.0.00.0.0. 00.0.0.00.0.00.00000 0. 00.0.0.0.0.0.00.0.0.0.0 0.0. 0.0.00.0. 00.000.0.0.0 0 "00 0. l0.0.00000000.0.00 00.0.0.0.0.0000 ANNEX III -65- POWER SECTOR ISSUES IN DEVELOPING COUNTRIES COUNTRY TNSTITUTION-AL FINANCIAL EAST AFRICA _ ; Botswana x |x x i Burundi x x x _ x x x . Cape Verde x x| _ _ = x= Comoros x x _ x x Djibouti x x _ Ethiopia x .x _x Kenya - x -x x Lesotho x x x x Madagascar _ x x = = = = x Malawi x x x x _ x __ Mauritius x x x x Mozambique = x ==x==== Rwanda x x x Seychelles _x x_ Somalia x = x x = x x Sudan x x x - x x _ Swaziland x x x Tanzania _x .x x = = x = Uganda x x _ x x x x Zaire x x x - x _ . Zabitwa x _x xx i.~ ~ ~~ _4-,-- _ - - __ _______ - 66 - POWER SECTOR ISSUES IN DEVELOPING COUNTRIES COUNTRY INSTITUTIONAL FINANCIAL WEST AFRICA. 0~~~~~~~~~~~~~~~~~~ Angola x _ _ x__ Benin x x x x x__ _ Cameroon x . x Central African. Republic x x x x Chad x x |Congo | Ix | x Guinea x x x .Gambia .x x x Ghana x x = = x =- x = = Guinea x x Guinea-Bissau x x Tvory Coast = x = x x =--x = = Liberia x x .- Mali x _x _ Mauritania x x _x _ Niger ix x _~~~~~~~~~ / Nigeria x x x_ x Sao Tome & Principe x x__ __ Senegal x = x _ =__ Sierra Leone ix x Togo x x x Upper Volta x x x - 67 - POWER SEC'TOR ISSTUES IN DEVELOPING COITNTRIES 1/0 .'-q ~~~~~~~~~~~~~~~~~~~~~- EAST & NORTH N AFRICAI Afghanistan x ,x x x x x__ Algeria x x x x . x x Cyprus x x x x x Egypt x x x x - x - x --- Jordan x x x Lebanon ,x. Morocco -. x .x x x x x__ Oman x x x x _ x x Portugal =x x X Romania i x _ _ Syria x x x |x x x x__ Tunisia x x x x x x__ Turkey x x x x x x x__ Yemen Arab Republic x x x x x x Yemen, People'sI Democratic Republic xx Yugoslavia _x x x x _I_ - 68 - POWER SECTOR ISSUES IN DEVELOPING COUNTRIES I X~~~~~~~~~~~~~~~- COUNTRY INSTITUTIONAL FINANCIAL SOUTH ASIA - Bangladesh x x x x x X X Burma x x x x x India x x x x Nepal x x x Pakistan x x x x x x Sri Lanka x x x x EAST ASIA & PACIFIC FiJi I x x x Indonesia x x x x x x x x South Korea x x x Laos x x x x x x Malaysia x x x x x Papua New Guinea x x x Philippines x x x Thailand x x x LATIN AMERICA & CARIBBEAN Argentina x x x x Barbados x x x x x x Bolivia x x x x x Brazil x x x x x x Chile ~~~x x xxx CFh i 1--.e- - - 69 - POWER SECTOR ISSUES IN DFVFLOPING COU.NTRIES 0 000 00 N ~~~~~~0 COUNTRY INSTITUTIONAL FINANCIAL LAC (continued) Colombia x x Costa Rica x x x x -Dominica x x x Dominican Republic x x x Ecuador x x x x x El Salvador x x x x Guatemala x x x x Guyana x x x x x x Haiti x x x x x x Honduras x x x x Jamaica x x x Mexico x x Nicaragua x x x x Panama x x x x Paraguay x x x Peru x x x Suriname x x x Uruguay x x x x - 70 - ANNEX IV GUIDELINES AND RESEARCH PUBLICATIONS ENERGY AND ELECTRIC POWER (From 1981 "CPS Catalog") Energy - General Alternative Fuels for Use in Internal Combustion Engines An Approach to Planning and Implementing Rural Energy Projects (Report on a Pilot Exercise in Colombia) Critical Factors in Economic Evaluation of Small Decentralized Energy Projects Energy and Development Energy Demand Forecasting: A Review Energy in the Developing Countries Energy Options and Policy Issues in Developing Countries Energy Pricing in Developing Countries: A Review of the Literature Energy Pricing in Developing Countries: Lessons from the Egypt Study Energy Pricing Studies: Criteria for Preparing Terms of Reference Financing Energy Developments in the Developing Countries Guidelines for the Presentation of Energy Data in Bank Reports Heavy Oil Occurrence and Development An Integrated Framework for Energy Pricing in Developing Countries Integrated National Energy Planning in Developing Countries Organizational Aspects of Energy Policies in Developing Countries Petroleum and Gas in Non-OPEC Developing Countries: 1976-1985 Petroleum and Gas in Non-OPEC Developing Countries: Projections through 1985: Production,. Consumption and Investment Requirements (Background Paper to Bank's Five-Year Lending Program FY78-82 in Oil and Gas) Proceedings of the South-East Asian Workshop on Energy Policy and Management Pricing Energy Products - Issues and Work Program A Program to Accelerate Petroleum Production in the Developing Countries Prospects for Traditional and Non-Conventional Energy Sources in Developing Countries (A Background Study for World Develop- ment Report, 1979) The Role of Natural Gas in Developing Countries Status and Outlook of Geothermal Energy Bangladesh: Rural and Renewable Energy Issues and Prospects Mobilizing Renewable Energy Technology in Deve'Loping Countries: Strengthening Local Capabilities and Research Renewable Energy Resources in the Developing Countries Electric Power - General Guidelines for Project Monitoring System for Public Utilities Projects Report of the Research Review Panel: Energy, Water and Telecommunications - 71 - A Report on Public Utility Insurance and Practice Diesel Plant Performance Study Guidelines for Estimating Costs of Tunnel Construction Guidelines for Sector Work in the Power Sector Nuclear Power: Its Significance for the Developing World Electric Power - Pricing & Economics Shadow Pricing and Evaluation of Public Utility Projects Economic Evaluation of Public Utilities Projects Costs Incurred by Residential Electricity Consumers Due to Power Failures Cost of Electric Power Outages: A Review Economic Criteria for Optimizing Power System Reliability Levels The Economics of Power System Reliability and Planning Energy Efficiency: Optimization of Electric Power Distribution System Losses Illustrative Audit Report for a Power Company The Leisure Cost of Electric Power Failures A New Approach to Power System Planning Optimum Economic Power Supply Reliability Optimal Electricity Supply: Reliability, Pricing, and System Planning Planning for Electrical Power: Costs and Technologies Thermal Generation Efficiency Study Phase I Workshop in Institutional Development in Electric Power Projects Economic Analysis of Electricity Pricing Policies: An Introduction The Economic Cost of Power Outages: Methodology and Application to Jamaica Electric Power Pricing Policy Electricity Economics: Essays and Case Studies Kenya Electricity Tariff Study Principles of Modern Electricity Pricing Shadow Pricing and Power Tariff Policy Rural Electrification Appraisal of Village Electrification Projects Costs & Benefits of Rural Electrification - A Case Study in El Salvador Electrification Rural Rural Electrification