Report No. 4314-YAR Yemen Arab Republic L. Power Subsector Review June 23, 1983 Projects Department Europe, Middle East and North Africa Regional Office FOR OFFICIAL USE ONLY Document of the World Bank 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. CURRENCY EQUIVALENTS (as of- January, 1983) Currency unit = Yemen Rial (YRl) Official rate 1 Yemen Rial (YR1 1) = US$o.22 YRls 4.54 = US$1.00 YR1 1 = 100 fils WEIGHTS AND MEASURES 1 kilowatt (kW) = 1,000 watts (103W) 1 Megawatt (MW) = 1,000 kilowatts (103kW) 1 kilowatt hour (kWh) = 1,000 watt hours (103Wh) 1 Megawatt hour (MWh) = 1,000 kWh (103kWh) 1 Gigawatt hour (GWh) = 1,000,000 kWh (106kWh) 1 kilovolt (kV) = 1,000 volt (103V) 1 kilovolt-ampere (kVA) = 1,000 volt-ampere (103VA) 1 Megavolt-ampere (MVA = 1,000 kilovolt-amperes (103kVA) 1 kilocalorie (kcal) = 3,968 British thermal units (Btu) 1 kilogram (kg) = 2,206 pounds (lb) I ton (metric ton) = 1000 kg = 2205 lb = 1.102 short ton = 0.984 long torn 1 meter (m) = 3.281 feet (ft.) I millimeter (mm) = 0.001 m 1 kilometer (km) = 1,000 m = 3,281 ft GLOSSARY OF ABBREVIATIONS BEI British Electricity International Ltd. CPO Central Planning Organization EDF Electricit4 de France IDA International Development Association GDP Gross Domestic Product K&D Kennedy and Donkin LDA Local development association LRAIC Long run average incremental cost LRMC Long run marginal cost LV Low voltage (less than lkV) MEW Ministry of Electricity and Water MV Medium voltage (less than 132kV, greater than 1kV) NRECA National Rural Electric Cooperative Association ODA Overseas Development Administration (UK) YAR Yemen Arab Republic YGEC Yemen General Electricity Corporation YOMINCO Yemen Oil and Mineral Resources Corporation YPC Yemen Petroleum Company YEMEN GENERAL ]ELECTRICITY CORPORATION FISCAL YEAR July 1 to June 30 (until 1980) January 1 to December 31 (from 1981) FOR OFFICIAL USE ONLY ABSTRACT The development of the Yemen Arab Republic (YAR) has been constrained by inadequate infrastructure which it inherited following a decade of civil war and the rule of the Imams. The 5-Year Plan for 1976-1980 had the strengthening of the physical infrastructure as one of its main objectives. During this period The Yemen General Electricity Corporation (YGEC) embarked on an ambitious program to electrify the main cities and major towns, which had suffered from prolonged shortages of electric power. Parallel to this, village entrepreneurs and a few co-operatives established private utilities, while some households and enterprises installed their own generators. This report concentrates on identifying the major issues in the power subsector and the options open to the authorities for dealing with them. The report is essentially a policy orientated document that addresses these issues which, in the view of the Bank, deserve attention. Following a brief review of the place of the power subsector in the energy sector, the report examines subsector institutions and recommends that the Government should consider alternative institutions on a pilot basis that would coexist with YGEC. The report reviews past and future electricity demand and supply and recommends improvements to power system planning, steps to reduce losses and unaccounted for electricity and measures to ensure a better balance between investment in distribution and generation and transmission that would be consistent with YGEC's financial and manpower constraints. The report shows that present YGEC electricity tariffs, while above economic cost, fail to generate sufficient income to cover YGEC's operation costs. It recommends a financial recovery plan, which would lead to operation, cost savings, plus tariff increases to achieve financial viability by 1986, as well as a lower tariff to attract autoproducers to YGEC supply. The report also recommends that the Government should place ceilings on YGEC operating subsidies and take steps to improve the safety and technical standards of the private power subsector. 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. I YEMEN ARAB REPUBLIC POWER SUBSECTOR REVIEW Table of Contents Page I. INTRODUCTION 1 II. THE ENERGY SECTOR 7 A. Institutions 7 B. Energy Resources 7 C. Petroleum Demand, Supply and Pricing 8 III. POWER SUBSECTOR ORGANIZATION AND MANPOWER 11 A. Subsector Organization 11 B. YGEC Manpower and Training 14 IV. HISTORICAL TREND IN THE CONSUMPTION AND SUPPLY OF ELECTRICITY 17 A. Past Electricity Consumption 17 B. Past Electricity Supply 21 V. FORECAST CONSUMPTION AND SUPPLY OF ELECTRICITY 26 A. Projected Demand of Electricity 26 B. Future YGEC Supply 32 C. Autoproducers and Private Power Systems 37 VI. ELECTRICITY PRICING 39 A. Institutional Responsibility for Tariffs 39 B. Historical Review 39 C. Tariffs and Economic Costs 41 D. Electricity Tariff Policy 43 VII. INVESTHENT AND FINANCING 46 A. Past Investment and Financing 46 B. Financial Recovery Plan 48 C. Future Investment and Financing 49 ANNEXES ANNEX 1 Manpower and YGEC Organization Attachments 1. Surplus/Shortfalls by Major Occupational Category Up to 1986 2. Projected Manpower Demand and Supply for Selected Public Sector Occupations This report was prepared by A. McKechnie (Economist), I. Elwan (Economist), C. Christofides (Engineer) and P. Craun (Secretary) - ii - YEMEN ARAB REPUBLIC POWER SUBS'ECTOR REVIEW Table of Contents Page ANNEX 2 Electricity Demand, Past & Projected A. Available Data on Electricity Demand 1 Availability of Electricity Consumption Data 1 B. Past Electricity Demand 2 Growth of Overall Consumption 2 Electricity Supplied by YGEC 4 Electricity Consumption by Sector 5 Regional Consumption of Electricity 6 Electricity Supply to Households 7 Loan Characteristics 7 C. Projected Demand for Electricity 8 Economic Development 8 Forecasts of YGEC Electricity Demand 9 Problems in Forecasting YGEC Demand 10 Projected Demand for Electricity 12 Future Improvements to Demand 16 Attachments 1. Electricity Sold and Consumers, YGEC, 1970-81 2. Past Growth In Consumers and Electricity Sold, YGEC 3. YGEC Load Forecast for 1982-1992 4. Load Forecast Sana'a (1982-1992;) 5. Load Forecast Hodeidah (1982-1992) 6. Load Forecast Taiz (1982-1992) 7. Demand Forecasts for Dhamar, Ibib and Hajjah 8. Load Forecast of Isolated Systems 9. Load Forecast of Isolated Systems after their Connection to the National Grid 10. Demand Forecast for Bulk Consumers 11. Assumptions for Load Forecast 12. Basis of Forecasts of YGEC Demand 13. Accuracy of Demand Forecasts Made in 1977 14. Sources of Demand Forecast Error - lii - YEMEN ARAB REPUBLIC POWER SUBSECTOR REVI,EW Table of Contents Pag ANNEX 3 Electricity Supply A. Past Electricity Supply 1 Generating Capacity 1 Fuel Consumption 3 Distribution 4 Losses 5 B. Power System Planning 6 Institutional Responsibility 6 Generation 7 Transmission 8 Distribution 8 Future Power System Planning 9 C. Future Electricity Supply 11 Generation 11 Fuel Supplies 12 Transmission 12 Distribution and Rural Electrification 13 Autoproducers and Private Power Systems 13 Attachment s 1. YGEC Plant Capacity (1975-1982) 2. Balance of Capacity and Demand, YGEC Main Systems 1975-81 3. Size and Age Distribution of YGEC Generating Plant 4. Fuel Consumed by YGEC 1975-81 5. Balance of Supply and Demand, YGEC 1981-92 6. Forecast YGEC Generation and Actual Consumption, 1981-92 ANNEX 4 Electricity Pricing Institutional Responsibility for Tariffs 1 Existing YGEC Tariffs 1 Tariffs Charged by Private Suppliers 3 Connection Charges 4 Economic Costs of Supply 5 Demand Related Costs 5 Energy Related Costs 9 Comparison of Existing Tariffs to LRMC 9 Comparison of Present Tariffs to Costs of Private Generation 10 YGEC Financial Constraints on Tariffs 12 Electricity Tariff Policy 13 - iv - YEMEN ARAB REPUBLIC POWER SUBSECTOR REVIEW Table of Contents Page ANNEX 4 Attachments 1. Calculation of LRAIC of Generation and Transmission 2. Calculation of LRMC of Generation *ad Transmission 3. Calculation of LRAIC's of Distribution 4. Marginal Costs of Supplying Consumer Groups ANNEX 5 A Financial Recovery Plan for the Yemen General Electricity Corporation Terms of Reference 1 Background I Objectives 1 Scope of Services 2 Organization and Manpower 2 Planning 3 Finance 3 Project Management 5 ANNEX 6 Tne Energy Sector Institutions 1 Energy Resources 2 Petroleum Demand, Supply and Pricing 4 Fuel Consumption 4 Fuel Demand Projections 5 Fuel Supply Facilities 5 Fuel Pricing 6 MAP, IBRD 17306 Yemen Arab Republic Power Subsector Review I. INTRODUCTION 1.01 The development of the Yemen Arab Republic (YAR) has been constrained by inadequate infrastructure which it inherited following a decade of civil war and the rule of the Imams. The 5-Year Plan for 1976-1980 had the strengthening of the physical infrastructure as one of its main objectives. During this period The Yemen General Electricity Corporation (YGEC) embarked on an ambitious program to electrify the main cities and major towns, which had suffered from prolonged shortages of electric power. Parallel to this, village entrepreneurs and a few co-operatives established private utilities, while some households and enterprises installed their own generators. 1.02 This report concentrates on identifying the major issues in the power subsector and the options open to the authorities for dealing with them. A summary of the issues and recommendations is presented in Table 1.1 below. This table classifies recommendations according to either first, second and third priority. The priorities shown are a combination of the time scale by which the recommendations should be implemented, as well as the importance of the recommendation. For example, while it is essential to strengthen YGEC's power system planning, it is not necessary to do so immediately since no major investment decisions need be taken for several years, although the nucleus of a planning department could be formed now. The report is essentially a policy oriented document that addresses those issues which, in the view of the Bank, deserve attention. 1.03 The report is divided into two parts. The first part provides a brief review of the background to each issue and the recommendations of the mission. These are contained in six chapters. The second part contains six Annexes which provide more detailed support for the analysis in the first part. 1.04 Chapter II gives a brief review of the energy sector in YAR; its organization, energy resources, petroleum demand and petroleum pricing. YGEC is the largest consumer of petroleum fuel in YAR. The interrelationships between the petroleum and power subsectors are discussed briefly in this chapter. However, this is intended only to enable the power subsector to be placed in its sectoral setting. Therefore, no issues are developed in this chapter since these would be identified and developed in an Energy Assessment. 1.05 Chapter III presents the institutional setting of the power subsector and examines the manpower, organizational and institutional difficulties which continue to constrain its development. This chapter also provides a first attempt at addressing the position of producers of electricity other than YGEC and their likely role in the future development of the subsector. 1.06 Chapter IV examines past electricity demand and supply and Chapter V presents forecasts of demand and supply. Chapter V concentrates on identifying the difficulties involved in forecasting electricity demand in YAR and proposes measures to ensure the balanced growth of electricity supply. 1.07 Chapter VI discusses the issues in electricity pricing and proposes measures to achieve a compromise between the objectives of promoting a financially viable utility and assures the least cost provison of electricity supply. 1.08 Finally, Chapter VII discusses the financing of power subsector investment and the financial viability of YGEC. -2 - Table 1.1 Proposed Strategies for the tevelopment of the Power Subsector Issues Objectives Recommendations Studies PriorLty III. POWER SUBSECTOR OXGANIZATION AND MANPOWER A. Subsector Organization Private and other small power (a) Provide electricity supply (a) On a pilot basis, MEW Proposed private power section utilities charge prices much in remote locations at a should provide technical in MEW responsible for non-YGEC higher than their costs of price near its cost, with- assistance to selected producers to monitor pilot supply (para. 6.06). out reducing its availabilty. local development associa- projects. tions wishing to establish (b) Provide information on tne cooperatives (para. 3.06). long term optimai organiza- tion of the subsector. (b) MEW should allow YGEC to provide bulk supply to selected private utilities on an experi- mental basis and subject to agree- ment being reached on technical and safety standards and retail prices (para. 3.06). Coordination between tne power (a) Ensure that suc supply (a) Strengthen the role of CPO Third subsector and other subsectors facilitles for YGEC and by estabiishing a committee needs strengthening. major industries are least of representatives from energy cost from the national sector and related organizations, There ha6 been little coorui- viewpoint (para 3.07). which would, inter alta, assist in nation of the fuel require- coordinating YGEC's fuel supply ment or YGEC in overall facilities with those of YOIINCO national fuei supply (para 3.06). (paras. 3.07-3.08) (a) Establish the institutional (b) Consider supplying the Sajil Desk study by YOMINCO of local First means by which tne power and Amman ceaent works with tuel oil transport costs for subsector can participate fuel oil unloaded using GP0 or- Council of Ministers. fully in national investment YGEC's facilities at Ras and manipower planning. Katenib (para. 3.08). Responsibilities between MEW btrengtien authority of YGEC Government snould clarify res- First and YGEC line mangement need managers, to improve day-to-day ponsibility of MEW and YGEC to be clearly defined in efficiency and to speed up management, set objectives for practice (para. 3.10). procurement of minor items YGEC and ensure these are met (para 3.10). (para 3.0o). B. YGEC Manpo)wer and Training YGEC suffers from Train and use existing man- Take necessary measures imme- To be included in consultancy First shortages of skilled man- power effectively, plan to diately to improve staff pro- for Financial Recovery Plan power, waich is a national economize on future manpower ductivity. Formulate medium (para 7.07). problem (pare 3.12). Pro- requirements, identify to long-term manpower and ductivity of non-professional training and expatriate needs training plan (para 3.18). staff low (para 3.13). well in advance to minimize costs (para. 3.18) . -3- Table 1.1 Proposed Strategy for the Development of the Power Subsector Issues Objectives Recommendations Studies Priorities IV. HISTORICAL TRENDS IN THE CONSUMPTION AND SUPPLY OF ELECTRICITY A. Past Electricity Supply System losses are nigh (27%) Reduce technical losses and Implement measures to Consultants engaged for First and nave been increasing theft of electricity identify and correct Financial Recovery Pian (para. 4.20) (para. 4.21) losses and establish (para 7.07) could recom- procedures for reduc- mend a strategy tor reduc- ing them (para. 4.21) ing; losses. B. Future Electricity Supply Skillea manpower scarce neces- Least cost system operation YGOC should examine options Analysis of alternatives First sitating expatriates to man within constraints imposed for reducing manpower by YGEC advisor proposed power stations by supply of Yemeni man- requirements per unit of (para. 5.22) power output (para. 5.18) There may oe savings in fuel Savings in fuel cost from YOMINCO or YGEC snould not Fuel pipeline feasibility Tnird transport costs from building large shipments procede with a pipeline study a pipeline from Salif to without a feasibility Ras Katenib (para. 5.20) study (para. 5.20) Economic alternatives of Least cost transmission YGEC to include economic Third transmission development development analysis of transmission need more analysis (para. 5.23) in Terms of Reference for future planning studies (para. 5.23) Investment in generation, Establishing investment YGEC should formulate a 10- Should be included in First transmission, distribution priorities after taking year investment plan Financial Recovery Plan and rural electrification committed investment and (para. 5.27) (para 7.u7) has been unbalanced and other constraints into uncoordinated (para. 5.27) account Investment planning needs Improve medium term invest- YGEC should formulate a Annual review of investment Second to respond flexibly to ment programming medium term investment plan, manpower and training changed circumstances and memorandum each year plan and financial projections io take account of financial (para. 5.29) and manpower considerations (para. 5.28) C. Autoproducers and Private Power Systems Private and other small uti- (a) National resource savings (a) MEW whould draw up standard Second lities need incentives to from least cost investment terms for providing bulk construct networks to high in private power supply supply or for taking over standards (para. 5.32) (b) Savings in construction private networks (para. 5.31) costs when private networks (b) MEW should set up division to integrated with national provide technical assistance system to non-YGEC suppliers (para. 5.32) -4 - Table 1.1 Proposed Strategy for the lktvelopment of the Power Subsector Issues Objectives Lecommendations Studies Priorities V. FORECAST CONSUMPTICIN AND SUPPLY OF ELECTRICITY A. Projected Demand for Electricity Comprehensive data on pri- Obtain national data on Include questions on YGEC's advisers and consult- second vate electricity supply and electricity consumption electricity supply in ants to assist in framing autoproducers is inadequate aod supply (para. 5.05) forthcoming census and questions, for census and (para. 5.05) in surveys of economic econcamic surveys sectors (para. 5.05) (paret. 5.05) Customer categories ill- Classify consumers according Reclassify consumers Should be included in Second defined and there are to their economic sector and and verify their the establishment of new discrepancies in tne costs of supply and improve existence when billing system. number of consumers the accuracy of consumer branch accounts are (para. 5.05) records (para. 5.05) computerized (para. 5.05) Existing YGEC data is in- Establish a database for plan- Engage consultants Coul be included in proposed Second sufficient for power system ning and a management informa- to formulate a Financial Recovery Plan planning and managenent tion system (para. 5.Ub) strate6y for setting (para. 7.u7) (para. 5.06) up an information and data processing system (para. 5.05) B. Future YGEC Supply (a) YGEC power system (a) Improve planning capability strengthen planning department Training component could be Second planning capaoilities (b) Make economic choice be- in techniques suitable for plan- included in proposed financial needs strengthening tween local generation and ning isolated diesel systems recovery plan (para 7.u7) (b) YGEC will have to redis- supply from main systems and their interconnection to tribute surplus diesel (para. 5.15) the main system (para 5.15) plant (para. 5.15) -5- Table 1.1 Proposed Strategy for the Development of the Power Subsector VI. ELECTRICITY PRICING issues Objectives Recomasendations Studies Priorities Financial viability of YGEC To adjust tariff levels quickly The Government should First is vulnerable to fuel price when fuel prices increase undertake measures for increases (para. 6.05) (para. 6.05) quickly adjusting power tariffs in reponse to increases in fuel and other prices (para. 6.06) The economic cost of YGEC is Encourage autoproducers to (a) Restructure YGEC tariff Tariff restructuring study First below the economic cost of take YGEt supply where this for MV consumers to be carried out by expatriate autoproduction. The cost of is available (para. 6.09) (b) Refuse licenses for auto- adviser (para. 6.01) Second autoproduction is below the production under Art. 9, YGEC tarLff (para. u.U9) Law 12 (1975) where YGEC supply is available (para 6.09) financial position of YGEC is (a) Mobilize resources to (a) Maintain tariff levels above First weak, yet present tariffs are finance new investment the economic cost of supply if in excess of economLc cost. (b) Ensure ultimate financial necessary to achieve financial Aie willingness to pay is viability of YGEC viability of YGEC. greater than YGEC's tariffs, (para. 6.12) as snown by even higher (b) Ensure tariff levels are set tariffs charged by private during the next five to seven suppliers (paras. 6.12) years to assure gradual progress to financial viability. T'ariff structure does not Improve accuracy of economic (a) Delay restructuring until reflect economic costs. costs. Gradually bring tariffs tariff increases required T'iere is cross-subsidization in line with economic costs (b) Carry out load research to Second oetweel, groups of consumers as fast as financial constraints improve estimates of (para. 6.13) permit (para. 6.17) economic costs (c) Reclassify consumers (para 6.14) -6- Table! 1.1 Proposed Strategies for the Development of the Power Subsector Issues Objectives Recosmmendations Studies Priority Potential MV consumers dis- (a) Encourage existing auto- Imple:aent a program to Tariff restructuring study First couraged from taking YGEC producers to take public restructure the existing to be carried out by expatri- supply, which is more economic supply immediately, to im- tariff by introducing a ate adviser (para. 6.01) from the national viewpoint, prove YGEC's finances and new tariff. by a YGEC tariff which is to achieve national fuel above both the cost of auto- savings production and the economic (b) Restructure MV tariffs cost of YGEC supply (para. 6.15) VII. INVESTMENT AND FINANCING A. Past Investment and Financing There have been frequent (a) Strengthen YGEC's project (a) Develop procedures and Part of Financial Medium delays in project imple- management and financial techniques for project Recovery Plan mentation (paras 7.02-7.03) planning management and financial Financial Recovery Plan (b) Improve coordination of planning and train staff (para 7.07) contracts within an overall in their use. investment plan (b) Program project activities, (para 7.03) financing and coordinate con- tracts through the proposed Annual Planning Memorandum (para 7.03) (a) YGEC's organization and (a) Save costs by improving Engage consultant to prepare a Financial Recovery Plan First management needs strength- organizational efficiency Financial Recovery Plan, then (para. 7.07) ening (b) Achieve financial targets implement their findings after (b) System project, financial that return YGEC to a sound discussing with IDA (para. 7.07) and manpower planning needs financial position improvement (para. 7.07) (c) YGEC's financial position needs to be gradually improved (para. 7.07) YGEC is not forecast to cover (a) Restore YGEC to a sound Fix ceilings for operating First its operating costs with financial position subsidies for YGEC in advance present tariffs (para. 7.10) (b) Provide incentives for and propose phasing them out operational efficiency by 1985, with any operating (para. 7.12) deficit being met by increased tariffs (para. 7.12) -7- II. THE ENERGY SECTOR A. Institutions 2.01 There is no organization formally responsible for the energy sector in YAR. The Yemen Oil and Mineral Resources Corporation (YOMINCO) is responsible for petroleum distribution through one of its operating subsidiairies, Yemen Petroleum Corporation (YPC). YOMINCO is also responsible for exploration for hydrocarbons and geothermal steam. The Ministry of Electricity and Water (MEW) oversees the operation of the Yemen General Electricity Corporation (YGEC); the only publicly owned electric utility in the power subsector. Traditional fuels such as firewood, crop residues and animal wastes are within the domain of the Ministry of Agriculture. The Central Planning Organizaion (CPO) has a coordinating role through its economic planning function and its activities in arranging foreign assistance. B. Energy Resources 2.02 At present, all conventional energy is imported, mostly in the form of petroleum products. Although there are indications of of natural gas, coal, oil shale and geothermal energy, so far little has been done in systematically exploring for these resources. YAR has significant renewable energy resources mainly in the form of solar and wind energy; however, to this day their commercial exploitation has not been attempted. The indigenous resources which have the greatest potential as fuels for the next major power station are natural gas and geothermal steam. 2.03 Shell Oil Company (UK) recently carried out exploration for oil and gas in the Tihama and off-shore in the Red Sea. Two holes were drilled in the Tihama. Gas, mainly methane, and condensates were found in one and the other was dry. The gas was trapped in non-porous sandstone which could make its extraction fairly difficult. YOMINCO has requested Bank assistance in financing the consulting'services needed to evaluate Shell's revsults. In addition, YOMINCO is negotiating with oil companies an arrangement for further exploration either on a contractor or production sharing basis. 2.04 Geothermal Energy. There are indications of geothermal activity in several places in YAR. Only the Dhamar-Rada'a area has been studied intensively. During 1981 ELC-Electroconsult (Italy) carried out scientific investigations near Dhamar, financed jointly by an Italian Government grant and the Government of YAR . The report indicated two areas and recommended that there was sufficient potential to justify an exploratory drilling program. The ELC investigations were reviewed by the UN Department of Technical Co-operation for Development and also by Geothermex Inc. (US) for the World Bank. 2.05 Both these reports reviewed the uncertainty surrounding the adequacy of the steam temperature for power generation. They both recommended that the issue be resolved by exploratory drilling. Geothermex concluded that there was a 50% probability of finding an exploitable geothermal field near Dhamar. - 8 - If found the potential should be in the range 125-560 MW. The capital cost of a geothermal power plant would typically range between 60-170% of the cost of a conventional oil fired plant, depending on design criteria which cannot be fixed until the exploration drilling program is completed and fully evaluated. However, site conditions near Dhamar would indicate a capital cost in the higher part of this range. 2.06 YOMINCO has decided to encourage private risk capitaL to explore the Dhamar prospect. Geothermex will complete the preparation of a promotion report tor distribution to investors in early 1983. This is being partly financed through IDA. If sufficient interest is shown, an agreement between YOMINCO and an investor would be reached by the end of 1983. This approach offers low risk and low exploration cost to YAR. However, the likelihood of an investor willing to embark on geothermal exploration program may not be high. If efforts to attract commercially financed exploration are unsuccessful, the Government would consider embarking on its own exploration program, possibly with IDA assistance. C. Petroleum Demand, Supply and Pricing Fuel Consumption 2.07 Recorded imports of petroleum fuels are shown in TabLe 1. These figures exclude gasoline and gas oil/diesel smuggled from Saudi Arabia to evade duty. Illegal imports of these products by road tankers are said to amount to as much as 40% of total consumption. The Government has recently started to enforce the collection of duty on fuel imports across the border with Saudi Arabia. Table 2.1 Petroleum Product: Imports 1973-79 (ton '000) Aviation Year Gasoline Gas Oil Kerosene Fuel Oil Fuels Total 1973 38 78 29 8 2.1 155.1 1974 39 79 30 9 4.2 161.2 1975 48 92 38 10 6.2 194.2 1976 70 128 49 10 9.1 266.1 1977 114 178 64 10 16.0 382.0 1978 171 238 78 10 13.8 510.8 1979 193 300 84 11 20.4 608.4 1980 226 306 90 11 20.4 608.4 1981 174 347 76 6 24 627 Growth Rate 1973-81 (% p.a.) 20.9 20.5 12.8 -3.5 35.6 19.1 Source: YOMINCO Note; 1/ 1981 figures are preliminary. - 9 - 16. Gas oil consumed by YGEC for power generation (see Annex 3, Attachment 4) amounted to 51,730 tons in 1981 or 15% of total consumption. The fuel consumption of the private power sector is unknown but it probably takes a further 15% to 18% of gas oil, 2-5% of gasoline and 5-10% of kerosene. On a weight basis, the entire power subsector probably consumes about 20% of the national recorded petroleum demand. In the future there will be a change in the demand for petroleum products by the power subsector. Fuel oil will increasingly substitute for the gas oil consumed in both public and private power generation (Annex 3, para. 33). Fuel Supply Facilities 2.08 Oil products are imported into the port of Hodeidah, except for fuel oil which is imported at Al Mukha because of congestion at Hodeidah. YAR pays high freight charges for small shipments in tankers in size up to 15,000 dwt. Presently, road tankers are the only available means for the inland transport of petroleum products. The Government has carried out a study of an oil product pipeline from Salif to Sana'a. The project would involve a marine terminal at Salif and distribution facilities at Al-Turbah (between Salif and Bajil) and Sana'a. An IDA mission has been examining this project, which is currently expected to be completed by 1987. 2.09 Fuel will be imported by small 5000 dwt tankers directly to the Ras Katenib and Al Mukha power stations. Consideration is being given to a fuel oil pipeline from the proposed oil terminal at Salif to Ras Katenib. Fuel supplies for power generation are discussed further in Annex 3, para. 34. Fuel Pricing 2.10 The development of the prices of petroleum products since 1976 is shown in Table 3. Prices for all products were roughly doubled in 1979. This increase, together with the adjustments in November 1980 and January 1981 eliminated the Government fuel subsidies and brought domestic prices more in line with border prices. Prices of gas oil supplied to YGEC have been lower than the retail price. In 1982 gas oil supplied to YGEC was priced at 1.36 YRls/liter, less a 5% discount, resulting in a net price of 1.292 YRls/liter. Table 3 Petroleum Product Retail Prices (YRls per liter) Date Effective Product 1976 1977 1978 Nov. 79 Nov. 80 Jan. 81 1/ Gasoline 0.93 1.10 1.11 2.30 2.30 2.60 Gas Oil 0.48 0.70 0.495 0.90 1.70 2/ 1.50 Kerosene 0.62 0.62 0.72 1.50 1.90 2.06 Fuel Oil 0.56 0.62 0.63 1.23 1.34 1.67 Source: YOMINCO, IMF 1/ January 1981 prices were in effect in late 1982. 2/ Price to YGEC in November 1980 was 0.76 YRls/liter. - 10 - 2.11 A preliminary analysis of the relation between fuel prices and costs is shown in Table 2.2 and described more fully in Annex 6. A more detailed analysis of fuel supply costs will be carried out in the forthcoming Energy Assessment. The economic costs are below the selling prices, except for gas oil, mainly because of taxes and duties. In financial terms, the price of gas oil appears to be cross-subsidized by the other fuels, but its retail price is close to its economic cost. Table 2.2 Economic Cost of Fuel (US$/ton, 3rd Quarter 1982) Gas Oil Kerosene Gasoline C & F price 327 371 350 Delivery costs 69 75 74 Economic cost 396 446 424 Taxes, duties, overheads & residual items -4 126 357 Selling price 392 572 781 Source: Annex 6, Table 4 2.12 Fuel oil is sold presently by YOMINCO at nearly $400/ton compared to its C & F price of $219/ton. This larg;e difference between price and C&F price reflects the small size and special circumstances of the current fuel oil market. YOMINCO has stated that it will supply the larger quantities of fuel oil required by YGEC at the C&F price, which it believes to be $200/ton. - 11 - III. POWER SUBSECTOR ORGANIZATION AND MANPOWER A. Subsector Organization Background and History 3.01 Electricity production in YAR began by Government agencies and private individuals installing small diesel powered generating sets to produce electricity for their own use in the main cities. Some of the private producers supplied their own homes and those of their immediate neighbors. These small generating and distribution schemes were gradually combined; and in i963 the General Electricity Company of Sana'a was established as an independent private company, whlich was owned by city government (75%), the Yemen National Bank (15%), and private shareholders (10%). Other companies were soon established in Hodeida and Taiz. The Taiz Electric Company was established in 1963 as a private company owned by the local government (17%), the Yeien iational Bank (23%) and private citizens (60%). In 1965, the Hodeidah Electricity and Water Company was established as a wholly-owned subsidiary of the Yenen National Bank. hlese companies developed along completely independent lines, with electricity distribution at 15kV in Sana'a and Taiz and lOkV in Hodeida. All companies suffered from the lack of qualified technical and managerial staff resulting in poor construction, mainitenance and planning. The shortage of generating capacity resulted in stagnation of load growth. 3.02 After the civil war it became obvious that the individual electricity companies could not continue pursuing their activities independently and meet the national objectives of development. The shareholders of the companies requested the Government to buy their holdings and in 1975 Law 12 was passed, which established the Yemen General Electricity Corporation (YGEC), with the former three companies as its nucleus. 3.03 In 1981, the Ministry of Electricity and Water (MEW) was created to direct the Government's plans for the electrification and the supply of water anld sewerage systems throughout the country. Before that time, the chairman of YGEC reported to the Minister of Economy; however, in 1981 YGEC's organizational structure was changed and the chairman of the Corporation was appointed the Deputy Minister of MEW and a Managing Director was appointed. irl 1982 YGEG's organizational structure was revised with the intention of strengthening the executive management. Power Suosector Organization 3.04 MEW supervises the operation of YGEC. The Corporation is entrusted witn the generation and distribution of electricity to the three main cities of Sania'a, Hodeidah and Taiz, as well as a further 28 towns and rural centers. Although the law establishing YGEC (Law 12 of 1975) grants the Corporation ttie sole responsibility of generating and distributing electricity, autoproduction is permitted with a license from the Ministry (Article 9). Moreover, despite a rapid expansion of YGEC's facilities, the public system has been unable to meet the tremendous growth in electricity demand in rural areas, because of shortages of manpower and finance. During the economic boom of the 1970's, a large number of small generators were - 12 - imported for autoproduction, small private utilities and for a few cooperatives. The mission estimates that about 33% of the population had electricity supply in 1981, compared to only 4% in 1975. Of those supplied in 1981, about 9% were supplied by YGEC and the remaining 24% either generated their own electricity, or were supplied from private utilities. YAR has had one of the most rapid rates of electrification in the world. To a considerable extent this was due to local village level initiatives financed by money remitted from relatives working abroad, as well as assistance given by the Government and the Confederation of Yemeni Development: Associations (CYDA). YGEC also expanded considerabLy, despite having to rehabilitate or replace most of the plant it began with. Electricity produced by YGEC grew at an average annual rate of 32% p.a. betiween 1975 and 1981. Private and Semi-Public Power Utilities- 3.05 Technical and safety standards of private and semi-public power utilities are generally poor. Furthermore, prices charged by these suppliers are in the order of 2 to 3 times higher than YGEC's tariffs, which are in turn among the highest in the region. From a resource allocation viewpoint, YGEC supply is more economic than autoproduction in those areas with access to the interconnected system (para. 6.06). MEW should therefore act to discourage autoproduction in areas with access to YGEC supply. MEW has not yet formulated a formal policy towards small power utilities, since the Ministry has lacked resources and was established as recently as June 1981, but an informal policy of allowing it to coexist with YGEC has been arrived at, which is both realistic and desirable in terms of meeting the national demand for electricity. However, there is a clear need for MEW to form a section to monitor the private power subsector, since, in the medium term, private suppliers and small autoproducers are likely to continue to account for a substantial portion of electricity production. This section would provide input to overall policy for the power subsector, but in particular would be responsible for improving the technical standards of private utilities (para. 5.32); collecting information on private suppliers and autoproducers (para. 5.05); and licensing installations where private supply is economically justified from the national viewpoint, or where public supply will be unavailable. 3.06 It is doubtful whether a policy by MEW to regulate the prices charged by private suppliers would be effective, since the Ministry lacks the resources to enforce price regulation. The formation of cooperatives in areas unlikely to be supplied by YGEC might lead to lower prices being charged. However, this form of organization for electricity supply is by no means widespread in YAR. Local development associations (LDA's) have successfully implemented road and water supply projects on a large scale. LDA's demands for electricity are usually met by the Government, subject to sufficient resources being available. However, to the mission's knowledge, attempts to encourage LDA's to set up their own utilities or cooperatives have been limited. YGEC appears likely to remain the major insititution in the power subsector, since it alone can capture the economies of scale and fuel savings from steam generation. Any stengthening of private power supply should not be at the expense of YGEC. Nevertheless, MEW should provide technical assistance to a few LDA's to enable them to establish local power utilities, possibly operating only at low voltage. Another institutional option is for YGEC to supply bulk electricity to well run private systems. Such an arrangement - 13 - would require agreement on, inter alia, retail prices to be charged, obligations as to who should be supplied and technical and safety standards. Supplying private utilities might lead to the benefits of YGEC generated electricity being made available with a saving in the human and financial needed by YGEC to extend its network. MEW should therefore consider allowing YGEC to supply bulk electricity to some private utilities on a pilot basis. Experience gained from these LDA and private schemes would be invaluable in deciding on the form of organization and technology needed to supply power to remote or mountainous areas, where YGEC is unlikely to be able to supply for some time and where demand is too low to utilize the surplus large diesel units which YGEC will have available. It is, therefore, recommended that a strengthened MEW provide technical assistance to selected LDA's to establish power co-operatives, on a pilot basis. A criterion for selecting the villages is that these should be in villages where YGEC would be expected to be unable to supply from the national interconnected system for at least ten years. MEW should also allow YGEC to supply bulk electricity to private utilities meeting the required technical and safety standards and which agree to regulated retail prices. Relations with Other Sectors 3.07 The main interaction between the power subsector and the rest of the energy sector concerns fuel supplies for power generation by YGEC. A second interaction arises from the substitution of electricity from private producers which will influence the demand for petroleum products, since fuel oil used by YGEC will indirectly be substituted for gas oil, gasoline and kerosene used by private producers (para. 4.16). Fuel supplies for YGEC have been planned independently of YOMINCO. Similarly, YOMINCO has not taken much account of YGEC's requirements, which might have involved advancing fuel oil facilities at Salif or upgrading fuel handling at Al Mukha. There has been little coordination of power station fuel supplies with the requirements of other industries. For example, it may be cheaper to import the fuel oil requirements of the northern cement works, at Bajil and Amran through YGEC facilities at Ras Katenib, rather than transport it by road tanker from the southern port of Al Mukha because of port congestion at Hodeidah. YGEC is considering using imported coal as a fuel for the next power station after Al Mukha. If coal is imported for a power station at Salif, the possibility arises of substituting coal for oil in other industries. Particular possibilities include the conversion of the Bajil cement works kilns and the use of coal for steam raising by small to medium size industries in the Hodeidah area. Before a decision to build a coal fired station is taken, a coal utilization study should be carried out, to ensure the optimum use of coal in the overall energy sector. 3.08 Furthermore, some of the problems facing the power subsector, such as shortages of manpower and finance and delays to project implementation are common to other sectors. There is a need for the power subsector to participate fully in national manpower planning, the allocation of finance and in preparing proposals to eliminate some of the obstacles to project implementation. It is therefore recommended that the role of the CPO in coordinating the energy sector be strengthened by the formation of a committee of representatives from relevant organizations to review investment and operations in tne sector and to provide inputs to national economic planning. In addition, it is recommended that further consideration by given to saving fuel transport costs by supplying the Bajil and Amran cement works with fuel oil from Ras Katenib rather than from Al Mukha port. - 14 - Organization of Major Power Subsector Institutions 3.09 Ministry of Electricity and Water. MEW was established by Law 26 on June 23, 1981, to formulate policies and plans for the development of the power subsector, and the development and exploitation of water resources in Yemen. Policies for the Government-owrned electric and water utilities are controlled by the MEW through a Government appointed comb-ined Board of Directors for water and power consisting of a Chairman (the Minister of MEW), a Vice-Chairman (the Deputy Minister of MEW), and 9 members. In regulating thie power subsector, the Minister is assisted by the Deputy Minister, who is the former Chairman of YGEC and two technical advisers for electricity, one of wnom is an expatriate provided by UK ODA. The Ministry is still in its infancy and technical and executive staff are virtually nonexistent. 3.10 The objectives and functions of the Ministry are still evolving. However, at present the Ministry serves as senior management for YGEC, effectively acting as the office of executive directors. The Minister is Chairman of tne Board of Directors of YGEC and the Deputy Minister is Vice Chairman. The relative responsibilites of the Ministry and YGEC line management do not appear to be well defined in practice. This is partly due to a shortage of qualified staff at YGEC. As the capability of YGEC line management grows their role and authority should be strengthened to enable them to ensure the efficient day-to-day operation of the utility and to speed up the procurement of minor items. It is recommended that the Government ciarify the lines of responsibility between MEW and YGEC, provide realistic objectives for YGEC management and ensure that these are achieved. 3.11 YGEC. Under the general supervision of the Board o.- Directors, a Director General is responsible for managing YGEC supported by a Deputy Director General, and a Board of Executive Directors consisti'ng of a Chairman (the Director General), a Vice-Chairman (the Deputy Director General) and 11 members. This Board does not formulate policy, and its function is effectively that of a management committee. It is responsible for operational matters and consists of six Directors from the head-office in Sana'a and five Regional Managers to whom the 26 Branch Managers report. Alt-hough each region has its own technical, administrative, accounting billing ancd purchasing department, all of them depend on the head-office for support in system planning and project preparation. Branch managers, on the other hand, depend onl the regions for support in maintenance and supply of materials. Details of YGEC's organizational structure are given in Annex 1. B. YGEC Manpower and Training YGEC's Manpower and Training 3.12 Currently, YGEC has 2,060 employees of which 54 are expatriates, 52 are engineers, 728 administrative, 84 foremen and the remaining 1,226 skilled anid unskilled staff. Of YGEC's total labor force, 215 are at the head-office in Sana'a and the remaining are divided among the regions as follows: (a) Sana'a 714; (b) Taiz 403; (c) Hodeidah 407; (d) Dhamar 185; and (e) Hiajjah 79. Since its creation YGEC has had staffing difficulties, which have led to dependence on expatriate staff and consultants for the highly technical -- 15 - operational matters. Overall, the services provided by the expatriates and the consulting finns are good, but they impose a heavy financial burden on YGEC. However, taking into consideration the slow process involved in training specialized staff, YGEC's dependency on expatriate expertise is expected to continue for some years to come. 3.13 At present, the Corporation has to employ a disproportionately large number of administrative, semiskilled and unskilled staff as indicated by the volume of electricity sales per employee (ratio of sales per employee is 97 MWh compared to 184 MWh for Syria and 273 for Egypt) and the low number of customers per employee of about 60. YGEC's low ratios of sales per employee and customers per employee are a typical case of diseconomies of scale accentuated by the Corporation's need to provide on-the-job training by assigning counterpart staff to contractors engaged in construction of various facilities, the duplication of staff necessary to man the isolated systems, ttue total dependence on manual practices for the preparation of electricity bills, and the very low level of sales to industrial and commercial enterprises. As YGEC's interconnected system evolves, sales to industrial consumers increase, and computerized billing is introduced, the ratio of sales per employee would improve contributing, thus, towards the reduction of YGEC's operating cost. However, it will take a few years before any noticeable change comes about and intensive training will be required. National Demand and Supply for Manpower 3.14 Shortages of manpower have been one of the greatest constraints to economic development in YAR. Manpower shortages have extended to unskilled as well as skilled labor, although unskilled labor shortages have eased recently. Labor stiortages have been exacerbated by the rapid pace of economic development, and migration of Yemenis to oil-producing countries. Skilled labor has been especially scarce, despite an impressive expansion of schools. Adult literacy has climbed from 3% in 1960 to an estimated 20% in 1980. The time taken for scnooling, university education and training has limited the supply of professional and managerial manpower and will continue to do so for the next 10 years. 3.15 The most recent exercise in manpower planning was carried out by CPO for the 1982-86 Five Year Plan. Annex A, Attachment 1 shows the aggregate demand and supply for broad occupational categories during 1982-86. These projections show shortages in all categories, except for unskilled labor where there is a large surplus. The most acute shortages are forecast for technicians, skilled office workers and semi-skilled workers requiring post-primary school job training. 3.ib Attachment 2 in Annex A shows the CPO public sector manpower supply-demand projections for occupations found in the electricity supply industry. The forecast deficit of electrical engineers during 1982-8' is 109, or 41% of demand. The deficit for accountants is projected to be 892, or 61% of demand. However, these projections may include underestimates of the manpower required for the power subsector. CPO forecast YGEC's manpower to increase by only 208 during 1982-86, yet the Ras Katenib power station alone will employ more than 250, of whom more than 20 will be engineers and 50 will be senior technicians. However, reductions in the manpower requirements of other sectors due to the slowdown in economic growth, may compensate partially for the extra requirements of YGEC. This indicates the need to coordinate power subsector plarnning with other sectors. - 16 - Future YGEC Manpower 3.17 In an effort to overcome the shortage of qualified lpersonnel, YGEC introduced in 1976 a program for the training, at various levels, of management, administrative and technical staff. Training was to take place, both abroad and in Yemen. This program was supported by financial assistance provided by bilateral aid agencies (French Cooperative Program and UK Overseas Development Administration), IDA (Credit 837-YAR), and the Arab Fund. dowever, training abroad had a slow start because of the difificulty in identifying candidates with the required academic background, a problem common in the public sector. Up to 1979 only 17 persons were sent for technical training (five for 18 months, and 12 for five to six years). In the last tihree years (1979-1982) an additional 51 candidates have been sent for administrative and technical training varying from 2 to 6 years. In addition, 23 candidates were sent to Italy for training on the operation of the new Ras Katenib power plant. Training in Yemen was also delayed a few years untill January 1, 1981, because no facilities were available, while the construction and manning of the training center at Dhahban, financed mainly by the French Cooperative program and marginally by ][DA, was being completed. Currently, the center provides training for 84 resident trainees per year. The programs offered vary from 6 months to 3 years. In addition, the facilities of the center are used for on the job training provided by various contractors who are constructing facilities associated with distribution projects and financed under a buyers credit and Credit 1102-YAR. On-the-job training is proving to be more effective in meeting YGEC's inmmediate needs of skilled and semiskilled staff, although the training center has been underutilized because of the lack of suitably qualified candidates. 3.18 YGEC is likely to continue to suffer from acute shortages of skilled manpower for the remainder of the decadle. Consequently, the Corporation has no option other than to use expatriates or contractors to bridge the gap, which in turn is likly to result in high staffing costs continuing. However, the Corporation can take some steps to minimize its staffing problems, such as: a) using existing manpower effectively, by for example, providing incentives to staff, evaluating and rewarding staff performance, improving morale, introducing a job costing system, better training of supervisors and the removal of bottlenecks such as shortages of transport, tools and materials; b) economize on future manpower requirements, through for example, the choice of investment, not permanently manning surplus plant and the use of mobile maintenance teams for large specialized items of plant; and c) planning manpower requirements in advance so as to identify areas where training produces the greatest benefits and to engage expatriates in time and at minimum cost. It is recommended that YGEC should immediately take all the necessary measures to improve the productivity of its staff. Policies to overcome medium to long term manpower problems should be tormulated in the proposed Financial Recovery Plan (para. 7.07). - 17 - IV. HISTORICAL TRENDS IN THE OONSUMPTION AND SUPPLY OF ELECTRICITY A. Past Electricity Consumption Data Availability 4.01 Establishing past trends in electricity consumption in YAR is more difficult than in most countries because of two main reasons. First, data on private electricity suppliers, which account for the greater part of electricity production is limited to a few limited surveys. Second, most of tihe older data available for public electricity supply are highly aggregated or unreliable. This is mainly because YGEC has inherited a billing system and consumer classification from the previous companies that is now inadequate for a utility of its size. Data requirements for planning are discussed further in para. 5.05. However, limitations to the extent and accuracy of the data available probably distort the picture of the growth of electricity consumption in YAR. Growth of Overall Consumption 4.02 Electricity consumption in the areas supplied by YGEC was constrained by supply shortages for most of the 1970's. These shortages affected consumption by: (a) reducing the number of consumers that could be connected; (b) shedding peak load; and (c) low voltage and frequency. Nevertheless, consumption grew rapidly. Electricity sold by YGEC grew at an average rate of 27.7% p.a. during 1970-81. Precise figures for the past consumption of electricity by the private supply sector and autoproducers are impossible to obtain. Electricity generated by the private sectors probably grew at an average rate of around 40% p.a. during the same period. Total national production (gross consumption) probably grew at about 35% p.a. during 1970-81 (Table 4.1). However, the rate of growth in generation for both YGEC and the private producers increased in the second half of the decade. This increase in the growth of electricity consumption was a consequence of the economic boom brought about by remittances from Yemenis working in oil producing countries and Arab and other foreign assistance. 4.03 Whilst the growth in consumption was rapid, it did start from a very low base. In 1975 annual generation per capita by YGEC was 7 kWh and the 1975 national generation per capita was about 17 kWh. In 1981 these figures had increased to 31 kWh and 92 kWh respectively. Consumption per capita is low compared to other countries in the region. For example, in 1980 kWh/capita was Egypt 493, Ethiopia 17, Jordan 330, Morocco 260, Somalia 28, Sudan 67, Syria 426 and Tunisia 440. - 18 - Table 4.1 Growth in Electricity Consumption 1970-81 (Gwh) Annual Rate of Growth (%) 1970 1975 1981 1970-1975 1976-1981 YGEC Energy sold 11.2 33.7 161.2 24.6 29.8 Energy generated 14.2 43.3 228.0 2.5.0 31.9 Private suppliers and Autoproducers I/ Energy generated Mean 10 60 440 43 39 High 20 70 595 29 43 Low 5 50 285 5 34 National Consumption 2/ Total generated 24 100 670 33 37 YGEC share of national generation (X) 59 ,43 34 Electricity Intensity (kWh/$'000 of GDP, 1975/76 prices) YGEC 23 42 156 12.8 24.4 Total 39 97 459 20 30 GDP Elasticity YGEC 2.2 4.8 Total 2.8 5.5 Source: YGEC, mission estimates 1/ Mean, high and low estimates for private sector are based on alternative interpretations of NRECA survey results. The high estimates assume all villages in YAR are similar to those surveyed by NREC, which had already been selected as candidates for connection to the YGEC system. Low estimates were based on the least e:Lectrified areas surveyed being typical of the whole country. 2/ National demand is based on the mean estimates for the private sector. - 19 - Electricity Supplied by YGEC 4.04 Total electricity sold by YGEC increased at an average of 25.2% p.a. during the period 1970-1975, 32.7% p.a. for the period 1976-81. The difference between the rates of growth during the two periods reflects the creation of YGiC at the end of the first period and its efforts in extending public supply of electricity to a larger proportion of the population than was originally served by the private utilities. As a result of YGEC's efforts the growth in the number of consumers increased from 13.0% p.a. during the period 1970-75 then almost doubled to 25.2% p.a. between 1976 amd 1981. In 1981 alone 25,700 consumers were connected, most of whom were in private systems taken over by YGEC. The rate of growth in the average consumption per consumer fell from around 11% p.a. during 1970-75 to 6% p.a. during 1976-81. This was probably caused by the average size of new consumers falling as lower income households were connected and as public supply was extended to rural areas (Table 4.2). 4.05 Electricity generated grew faster than energy sold reflecting the increase in distribution losses and unaccounted for electricity. Maximum demand, however, grew slower than energy generated not only because of shortages in capacity leading to load shedding, but also because a few large commercial and government consumers with high load factors took supply and there was some increase in air conditioning (Table 4.2). Table 4.2 Electricity Demand, YGEC Systems Annual Rate of Growth (%) 1970 1975 1981 1970-1975 1976-1981 No. consumers (end year) 22,500 2/ 36,200 116,000 13.0 25.2 Electricity sold (GWh) 11.2 33.7 161.3 25.2 32.7 Eiectricity generated (GWh) 14.2 43.3 228.0 25.4 34.6 Maximum demand (MW)31 6.54 10.6 54.3 17.2 33.9 Losses (X) 21.0 22.2 29.3 Load factor (%) 40.7 4/ 46.6 47.9 Source: Annex 2, Attachment 2 I/ Growth rates estimated by method of least squares 2/ Data for 1971 3/ Sum of maximum demands of individual systems, excluding diversity. 4/ Data for 1972. Electricity Consumption By Sector 4.06. YGEC classifies its consumers as Private (houses and shops), Industry, Government (including street lighting) and Diplomatic. Households and shops were the fastest growing sector, increasing at 34.4% -p.a. during - 20 - 1975-81, followed by Government (38.8% p.a.) (Table 4.3). Sales to industry grew at 24% p.a. and its share declined from 11.6% of total sales in 1975 to 7.2% in 1981. TablE 4.3 Electricity Sales by Sector, YGEC, 1975-81 (Mwlh) Annual Growth Rate (% p.a.) Sector 1975 1976 1977 1978 1979 1980 1981 1975-1981 Private 22,100 26,800 34,467 46,707 69,362 97,682 116,658 34.4 Industry 3,900 4,800 6,162 8,824 13,467 13,137 11,623 24.2 Government 3,400 6,800 8,664 11,685 13,419 18,708 25,168 28.8 Diplomatic 2,320 3,130 4,030 5,440 4,260 6,212 7,777 19.8 Total 33,720 41,530 53,323 72,653 100,508 135,739 161,226 31.6 Source; YGEC Note: Annual growth rate calculated by method of least squares. 4.07 The proportion of electricity sold to industry is small compared to most countries. There are three reasons for this. First, until recently public electricity supply has been consistently unreliable. Even now there remain restrictions on the consumption of electricity by industry during peak periods. Second, electricity tariffs for industry have been the same as, for other consumers and are above the cost of private generation (see para. 6.08). Consequently, private electricity generation has been more attractive financially, especially when diesel fuel was heavily subsidized. Third, manufacturing accounts for only 7% of GDP and most industries in YAR are not electricity intensive. However the sales by YGEC to industry will increase substantially when the cement works at Bajil and AmraLn take public supply (para. 5.09). Regional Consumption of Electricity 4.08 Sana'a region consistently accounted for about half YGEC's energy sold during 1970-1981. After 1975, consumption grew at 31.2% p.a. in Sana'a, but growth was slower in Hodeidah (23.7% p.a.) and Taiz (26.8% p.a.), partly because of greater capacity shortages outside of Sana'a. Until 1976 public supply was virtually limited to the three main cities. However, during the late 1970's public supply was extended to other regions, so that in 1981 the three main cities had only 73% of all YGEC consumers and 86% cf consumption (Table 4.4). - 21 - Table 4.4 Proportion of Electricity Sold By Region, YGEC (% of total YGEC sales) Region 1970 1973 1975 1978 1981 Sana'a 55.7 46.5 53.7 56.1 52.5 Hodeidah 20.1 34.0 30.8 24.3 20.9 Taiz 24.2 19'.5 15.5 12.8 12.6 Dhamar - -- - 3.0 2.8 Hajjah - -- - 1.1 1.3 Ibb - -- - 1.6 3.1 Branches - - - 1.1 6.8 Total 100.0 100.0 100.0 100.0 100.0 Source; Annex 2, Attachment 1. Electricity Supply to Households 4.09 According to the 1975 census 47. of houses had access to public or private supply, or generated their own electricity. The mission has estimated that in 1981 33% of households had access to electricity supply, of which nearly 9% were supplied by YGEC. If these figures are correct, they imply that almost 30% of households would have gained access to supply in 6 years, which would be a remarkable achievement. B. Past Electricity Supply Generating Capacity 4.10 The growth of installed capacity in YAR is not known because detailed data on private suppliers and autoproducers are not readily available. YGEC generating capacity increased from a 15MW in 1975 to 121 MW in 1982; representing an average annual increase of about 35%. The growth of installed capacity maintained by the private suppliers and autoproducers can not be determined exactly, but the total capacity in 1981 probably amounted to between 200MW to 60ORW and had grown from between 1OMW and 30MW in 1975 (Table 4.5). This range of uncertainty is too high to enable sound policies to be formulated for this part of the power subsector. However, much improved estimates would be obtained from more comprehensive surveys of the private power subsector (para. 5.03). - 22 - Table 4.5 Growth of Generating Capacity, YAR, 1975-81 (NW) Annual RAte of Growth (%) 1975 1981 1975-81 YGEC Sana'a 7.0 41.5 34.5 Hodeidah 3.3 25.9 41.0 Taiz 2.8 18.5 37.0 Others 1.6 23.7 56.7 Total 14.7 109.6 39.8 Autoproducers & private 10-30 200-600 Total YAR 25-45 310-710 Sources: YGEC, mission estimates. 4.11 Both the public and private powger systems comprise isolated systems fed by diesel generators. YGEC is a new utility formed in 1975 where development in its initial years was determined by the unsatisfied demand for electricity in the main cities and its establishment as a Government institution, rather than as a result of a systematic plan for development. The private power subsector is large in relation to YGEC. This is a consequence of both the constraints on YGEC which prevented it from keeping abreast of demand as well as the physically scattered nature of demand. The rate at which YGEC could expand was constrained by the national shortage of skilled manpower, shortages of finance and problems in project implementation. 4.12 Until the late 1970's demand in YGEC's systems was constrained by shortages of generating capacity. From 1979 onwards the quality of supply improved as a program was implemented to install diesel units at Sana'a, liodeidah and Taiz as an interim measure before Ras Katenib is commissioned in 1983. However, the reinforcement and expansion of distribution did not keep in step with the increase in generating capacity and as a result, the new diesel power plants are presently underutilized, especially in Hodeidah and Taiz. Parallel to this, YGEC embarked on a program of electrification in provincial capitals and major towns. 4.13 Trade statistics and anecdotal evidence suggest that most private generating plant is less than five years; old. Most of this consists of high speed units with capacities between 5-10 kVA that have not been adequately maintained. Consequently, most private generating plant must be reaching the end of its economic life and will need replacement during the next five years. Small units may need replacement every 12-18 months. 4.14 YGEC's existing generating plant is a legacy of the period of the rapid growth in electricity demand which necessitated a quick unplanned response by YGEC. The demand which YGEC could not satisfy was met by the private suppliers and autoproducers. However, much of the interim plant purchased by YGEC will remain in the future. The large diesel units would be used for standby and to provide spinning reserve on the interconnected system. The smaller plant (500-1,200 kVA) would be available for use in remote systems, where private generating plant is nearing the end of its I nomic life. - 23 - Fuel Consumption 4.15 Gas oil (diesel) is the sole fuel used by YGEC for power generation. Between 1975 and 1981, consumption of gas oil increased at an average annual rate of 28.3%, from 21.7 thousand tons in 1975 to 51.7 in 1981. In 1981, this represented 15% of the total consumption of gas oil in YAR. However, the larger diesel units will operate on heavy fuel oil when supplies become available from the Ras Katenib power station. In 1981 average thermal efficiency was for the YGEC systems 36.4%. Efficiency increased by about 5 percentage points during 1975-1981, mainly because new medium speed diesel units were commissioned and old inefficient, non-turbocharged low speed machines and small high speed units were either retired or operated for fewer hours. 4.16 Autoproducers and small utilities consume gas oil, gasoline and kerosene for power generation. Many small units, designed to run on gasoline, are fuelled by kerosene because of its low price relative to gasoline. No reliable data on the fuel consumed by private producers exist. A crude estimate based on rough estimates of installed capacity would put gas oil consumption of power producers other than YGEC at between 79,000 to 156,000 tons and gasoline and kerosene each at between 5,500-19,000 tons. The entire power sector in 1981 would therefore have consumed between 131-208 thousand tons of gas oil, or 38-60% of recorded national consumption. Distribution 4.17 YGEC inherited runr-down distribution networks when it was formed in 1975. The previous companies had suffered from lacK of staff and finance. Maintenance had been inadequate and network expansion was insufficient to meet demand, which had started to grow rapidly. Moreover, each company had developed along separate lines so that technical standards varied widely. The primary distribution voltage was 15kV in Sana's and lOkV in Hodeidah and Taiz. Low voltage networks required rehabilitation or replacement. The best distribution networks had small conductor sizes (35mm2) which were incapable of delivering increased loads and led to high losses and voltage drop. 4.18 After its formation YGEC initiated a program of distribution rehabilitation and reinforcement, with assistance from the Arab Fund, IDA and bilateral donors. YGEC carried out a nationwide study of rural electrification with assistance from IDA, which proposed a program of development and established technical standards. YGEC managed to make steady progress in improving its distribution systems, installing 180km of MV line, 194km of LV line and about 230km of cable, (Table 4.6). Despite this progress, distribution investment was insufficient, as indicated by the high losses in 1981 (para. 4.21). The distribution system tended to expand linearly in roughly equal annual increments, whereas demand was growing exponentially at over 30% p.a. during 1976-81. - 24 - Table 4.6 YGEC Distribution Reinforcement (Km installed, cumulative) 1975 1976 1977 1978 1979 1980 1981 Uverhead Lines 33kV - - - - - 17.5 24.0 lOkV, llkV, 15kV - 15.5 43.2 81.6 115.0 138.0 156.5 LV - 10.6 34.6 65.0 123.2 151.0 194.4 Underground Cables 33kV - - - - - - 3.0 lOkV, llkV, 15kV - 12.0 35.3 61.4 90.8 111.5 135.9 LV - - 4.7 9.8 19.1 25.5 93.2 Source: YGEC 4.19 Distribution facilities of private suppliers are generally of a low technical standard and are often little more than housewiring cable strung from buildings, trees, etc. Some of these networks are a hazard to public safety. However, some suppliers have erected lines attached to insulators on steel poles, although conductor sizes and distances between conductors are usually inadequate. Savings in terms of national resources would result if private networks were designed and constructed to higher standards. In view of the apparent importance of private producers in overall electricity supply, MEW should initiate policies to improve their technical and safety standards (para. 5.32). Losses 4.20 In 1981 total losses for all YGEC systems amounted to 29.3% of energy generated. Moreover, losses have been increasing. In 1970 losses for all YGEC systems were 21.5% which increased only slightly to 22.2% in 1975 before rising steeply to the present levels (Annex 2, Attachment 1). 4.21 No investigations of YGEC's losses have been undertaken yet, but it is believed that a proportion of the losses comprises theft and billing errors. A further proportion can probably be accounted for by inaccurate metering and some unmetered consumption of YGEC employees. The remaining would be technical losses arising from an overloaded distribution system. These losses are high by any standards, but are a consequence of demand growing more rapidly than the system could expand. The lack of an effective Consumers Department also contributes to the large amount of unaccounted for electricity. Future losses will appear to fall as MV consumers such as the cement works are connected, since losses are lower at MV. The trend towards lower losses arising from a greater proportion of MV sales should not obscure the need to take action to reduce losses elsewhere. In view of the adverse impact of losses on YGEC's revenue generation and planning, it is recommended that the Government strengthen YGEC s Consumer Department and consider a two pronged strategy for reducing losses. The first would aim at reducing the - 251 - losses attributed to existing facilities and the other relates to the reduction of theft. These would involve: a) inspections of LV networks and the rehabilitation of inadequate service connections; b) improving procedures for meter reading and billing; c) introducing an effective sealing method for meters together with procedures for meter inspection; d) strengthening the meter testing, calibration and repair workshop and formulating a program for the regular testing of all meters, starting with old meters supplying large consumers; e) metering all consumption of employees and showing this as consumption, rather than losses; f) carrying out surveys to identify the main parts of the networks where technical losses occur; g) instituting immediate distribution reinforcements to reduce losses in those parts of the networks where losses are greatest; and h) establishing a system to regularly measure and record transformer and feeder loadings to identify distribution reinforcement projects for inclusion in the proposed annual planning memorandum (para. 5.28). Steps to reduce losses and theft should be implemented immediately, but longer term policies should be formulated by the consultants engaged in the Financial Recovery Plan (para. 7.07). - :26 V. FORECAST CONSUMPTION AND SUPPLY OF ELECTRICITY A. Projected Demand for Electricity Economic Development 5.01 GDP increased at an average annual rate of 7.6% during the decade 1971-1981. As Table 7 shows, economic growth was more rapid during the first half of the decade. GDP increased at about 9.4% p.a. between 1971 and 1977, then on average declined to 5.6% p.a. thereafter. This decline has been due to a recent fall in the rate of economic growth resulting from the decline in workers remittances. In 1980 GDP increased by 3.8% followed by 2.5% for 1981. The 1975 to 1980 First Five-Year Plan forecast GDP to grow by 8.2% p.a. A growth rate of 5.9% p.a. was achieved. Table 5.1 Growth in GDP by Sector (YRls million, constant 1976) An^nual Rate of Growth Fiscal Year (% p.a.) Sector 1972 1974 1981 1972-1977 1977-1981 Agriculture, Mining 1577 1892 2252 3.7 3.9 Manufacturing 159 274 460 11.5 12.2 Electricity and Water 8 17 46 16.3 24.8 Construction 170 373 481 1L7.0 5.8 Trade 475 944 1032 1.4.7 2.0 Transport and Communication 116 160 220 6.6 7.3 Government services 340 451 885 5.8 16.2 Other services 299 549 896 12.9 11.5 Others 171 526 363 25.2 -7.9 Total GDP (market prices) 3315 5186 6635 9.4 5.6 Source: CPO Remittances from Yemenis working abroaci and net private and official transfers have been the main influence on the rate of economic growth. Investment during the 1975/76 to 1980/81 5-Year Plan was impressive. Thre ratio of gross fixed capital formation to GNP averaged around one third, more than twice its level during the preceding Three Year Development Program (1974-1976). The decelleration in growth since around 1978 has been mainly due to remittances declining in both nominal and real terms and some development assistance has been delayed. The balance of payments,, which registered considerable surpluses before 1979, showed an estimated $330 million deficit in 1981 (11% of GDP). The budgetary position has also deteriorated rapidly, and in 1981, the deficit amounted to about US$500 million, or 17% of GDP. - 27 - 5.02 Given the decline in private transfers in recent years and the changing demand for Yemeni labor in the oil exporting countries, the Second 5-Year Plan GDP growth target of 7.0% p.a. looks optimistic, since, the greater part of capital formation during the plan period, is projected to come from workers' remittances. GDP growth during the Plan is unlikely to exceed 5.5% p.a. and could even fall as low as 4% p.a. YGEC Electricity Demand Forecasting 5.03 Sales during the period 1979-1981 were substantially lower than forecast in 1977. Actual sales fell short of the forecast by as much as 87%. There were four main reasons for this; a) the forecast was based on optimistic projections of connections of new consumers and improvements to the quality of supply sufficient to induce new consumers; b) the optimistic plans of the Goverment which projected a large number of industrial projects coming on stream than was reasonably possible and at dates that, in retrospect, could not have been achieved; c) a simple tariff structure whicht did not give the correct incentive for autoproducers to take public supply. However, this will cease to be an issue if the proposed restructuring of MV tariffs is implemented (para. 6.15); d) a poor database which did not provide an adequate base to forecast from. 5.04 These past sources of load forecast error still need to be considered when projecting electricity sold by YGEC. As the connection of households without supply in the main cities becomes completed, the major source of growth in consumers will be in rural areas. The extension of public supply to these areas will continue to be constrained by shortages of finance and manpower. A substantial part of the non-industrial demand for electricity will therefore remain supply constrained. The dates at which large industries require electricity supply are also dependent on shortages of manpower and finance and the institutional problems that affect most sector of the YAR economy. Uncertainty in the timing of large electricity consuming projects and indeed, whether the project will go ahead at all, translates into electricity demand forecast uncertainty and leads to potential errors in inevestment by YGEC. There is little YGEC can do in terms of improvement to techniques and procedures to avoid load forecast errors arising fom its own supply constraints or in changes to industrial investment plans. Apart from exercising sound judgement, the Corporation can do little more than to improve the integration of its forecasting and distribution planning to allow for supply constraints and to identify the major sources of uncertainty arising from industrial projects. For example, if the forecast is particularly sensitive to say, the timing of cement works, and the timing of generation investment is in turn sensitive to the cement works' load, then the two investments should be considered jointly. This is more an issue of the coordination of investment between sectors (para. 3.08) than electricity demand forecasting. - 28 - 5.05 Despite the appointment of an expatriate statistics officer and improvements to its database, a number of data problems continue to plague YGEC demand forecasting. These are partly inadequacies in the data itself and partly limitations to the old procedures for collecting and processing information which have been strained by the rapid growth of the Corporation. First, despite some recent surveys of large autoproducers, there is still a shortage of information on autoproducers. Furthermore, the only information available on privately supplied electricity is from the surveys carried out with finance from the IDA Power II Credit, of villages considered suitable for YGEC supply, which may not be typical of the entire country. Together, autoproducers and private suppliers probably account for the greater part of electricity supply in YAR (Table 4.1). The size, composition and physical distribution of this market is an indication of the potential market for YGEC supply and is therefore useful for demand forecasting. Improved information on private electricity suppliers would also assist MEW in forming policies towards autoproducers and private producers, and improve the planning of petroleum distribution facilities. However, a nationwide survey of private power supply could cost as much as US$2 million and it is doubtful whether such a cost would be justified, especially since the number of private power suppliers will probably contract as the YGEC system expands and autoproducers take public supply (para. 5.29). There are cheaper alternatives for obtaining information on private electricity supply and autoproduction. These include placing questions on electricity supply in the forthcoming popoulation census and in economic surveys of industry, trade and agriculture; and including an electricity section in any survey of energy consumption directed primarily at fuelwood or petroleum issues. It is recommended that CPO include questions n electricity supply in the forthcoming population census and economic surveys and the MEW assist in the framing of these questions, using advice from its advisors or consultants as necessary. Second, YGEC's consumer classification is too broad to serve as a base for demand forecasting, e.g. the private tariff category lumps shops and residences together. Tariff categories are neither clearly defined by economic sector nor in terms of the costs of supply. Such broad consumer definitions do not enable the factors which determine demand, e.g. household formation, to be linked to consumption. Third, rapid growth in consumers has prevented YGEC from maintaining accurate consumer records with its old manual billing system. Since YGEC should ultimately restructure its tariff (para. 6.17) and at present is planning to introduce a computerized billing system, it is recommended that YGEC reclassify its consumers and verify their existence when branch accounts are computerized. 5.06 Fourth, YGEC's information gathering, analysis and reporting procedures have reached a stage where they need rationalization. There are problems in collecting data in the many scattered branches and forwarding it to the head office. There are gaps in the information collected. Reporting has grown to meet ad hoc demands for information both within and outside the Corporation. YGEC has a full time statistics officer and the nucleus of a data base which could, if strengthened, provide valuable information for management and operations, as well as for load forecasting and planning. It is recommended that YGEC recruit consultants to formulate a strategy for constructing a database,.including the specifications of a system for the collection, analysis, storage, retrieval and reporting of data, as well as the training of local manpower in its operation. This could be included in the management consultancy services for the proposed Financial Recovery Plan (para. 7.07). - 29 - Projected Demand for Electricity 5.07 The mission reviewed YGEC's detailed forecast, prepared with assistance from consultants, and considered it optimistic. The mission therefore revised the forecast for YGEC's systems and the updated forecast was discussed with the Corporation and agreed in principle. Details of the forecast and assumptions used are shown in Annex 2. 5.08 It is extremely difficult to forecast the demand by private electricity producers because of the lack of adequate information about this part of the subsector. A rough estimate of the future consumption from private producers was made by the mission on the basis of available information. The estimate assumes that aLll large autoproducers in 1981 would gradually be absorbed into YGEC's system by 1987 because of tariff restructuring. During the next five years generators in villages will be near the end of their lives. Replacement of them is unlikely in areas where YGEC supply becomes available since the YGEC tariff is already several times lower than tariffs charged by private producers. YGEC supply would be more economic in comparison to private supply or autoproduction, primarily because of the high cost of diesel fuel in comparison to the fuel oil used by YGEC. Villages in or around existing or planned YGEC networks are therefore assumed to be supplied by YGEC. Consequently, electricity produced by autoproducers and private suppliers is projected to be half its 1981 level by 1987 and to decline slowly thereafter as YGEC supply slowly penetrates the more inaccessible areas. National total YGEC and private power generation would increase at about 9.4% p.a. in 1981-87, slowing down to 8.9% p.a. in 1987-92 (Table 5.2). Generation by YGEC would rise from 31 kWh/capita in 1981 to 165 Table 5.2 Forecast National Electricity Consumption Actual Projected 1970 1975 1981 1987 1992 Electricity Generated (GWh) YGEC 14 43 228 925 1552 Autoproducers & private 10 60 440 220 200 Total YAR 24 103 668 1145 1752 Population ('000) 5258 6075 7264 8375 9429 GDP (US$million, 1975/76 prices) 617 1032 1460 2013 2631 Generation per capita (kWh) YGEC 3 7 31 110 165 Total 5 17 92 137 186 Intensity (kWh/$'000 of GDP) YGEC 23 42 156 459 590 Total 39 97 459 569 666 GDP Elasticity YGEC 2.2 4.8 4.4 1.9 Total 2.8 5.5 1.7 1.6 - 30 - kWh/capita in 1982, increasing sharply, by 1987 because of new cement and other bulk consumers, such as existing autoproducers. Between 1981 and 1987, the increase in consumption from these new consumers is expected to lead to an apparently high elasticity with respect to GDP for YGEC sales. However, the national elasticity is expected to decline to 1.7 in 1981-87 and remain almost constant at 1.6 in 1987-92. This is indicative of the transfer of generation from autoproducers to YGEC. 5.09 Total electricity sales by YGEC are forecast by the mission to grow at an average annual rate of 37.0% between 1981-85 then decline to 12.1% during the period 1985-90. The decline in growth would be pirimarily due to the approaching saturation in the number of household consumers, especially in the cities and the completion of the transfer of autoproducers to public supply, as summarized in Table 5.4 below. Much of the growth in electricity consumption up to 1987 is accounted for by the expansion of YGEC supply to industry and the connection of the cement works at Bajil and Amran. Such extension of public supply to industry would depend upon the implementation of a new tariff for MV consumers which would provide incentives to discontinue autoproduction (para. 6.11). If the recommended changes in tariffs are introduced, the share of the industrial categories in total YGEC consumption would increase from 10% in 1981, to 50% in 1987 and then decline gradually to 44% in 1990. Consequently, sales to industry would increase at 38% p.a. between 1981-92, before declining to 8.5% p.a. during the second half of the forecast period, 1987-92. Table 5.4 Projected Electricity Sold by Sector, YGEC (GWh) Annual Growth Actual Forecast % Share Rate (%) Sector 1981 1987 1992 1981 1987 1992 1981-87 1985-92 Domestic 76.3 214.0 356.7 47.3 29.8 29.1 18.8 10.8 Commercial 18.1 25.9 34.8 11.2 3.6 2.8 6.2 6.1 Government 19.1 30.4 44.6 11.8 4.2 3.6 8.1 8.0 Street Lighting 1.8 3.4 5.7 1.1 0.5 0.5 11.2 10.9 Diplomatic 7.8 9.3 10.8 4.8 1.3 0.9 3.0 3.0 Small Industry 9.5 13.6 18.2 5.9 1.9 1.5 6.2 6.0 Bulk 6.2 162.1 263.0 3.8 22.5 21.4 72.3 10.2 Cement - 185.0 260.0 - 25.7 21.2 - 7.0 Rural 22.5 75.3 233.6 13.9 10.5 19.0 22.3 25.4 Total 161.3 719.0 1227.4 100.0 100.0 100.0 28.8 11.3 Source: Annex 2, Attachment 3 Domestic consumption in Table 5.3 refers to households in the three main cities of Sana'a, Hodeidah and Taiz. Most of the growth in sales to these consumers is accounted for by new connections. Connections are forecast to - 31 - increase at 16.5% p.a. in 1981-85 and 6.61% p.a. between 1985-92. Consequently, the percentage of households connected rises from 68% in 1981 to 95.0% in 1985 and remains more or less constant thereafter. 5.10 Rural consumption is expected to grow at 22.3% p.a. between 1981 and 1987 and 25.4% p.a. during the period 1985-92. The average consumption per rural consumer is projected to increase by only 3.1% p.a. from 65OkWh/a in 1981 to 10OkWh in 1992. The growth in rural consumption is mainly due to the connection of households without YGEC supply. During the next 10 years, extensions to YGEC's networks will be constrained because of shortages of finance and skilled manpower. Most of the growth up to 1986 occurs in areas where YGEC already supplies some consumers. By 1992, 55% of the rural consumption would be in projects that have not yet been formally identified. 5.11 Distribution losses, station use, and theft in Sana'a, Hodeidah and Taiz are forecast to fall from their current level of around 30% of gross generation to 22% in 1987. This is predicated upon effective measures to reduce losses being taken (para. 4.21). Some reduction in average losses will arise from the rapid growth in sales at MV to bulk consumers, but this will be offset to some extent by the greater station own use of steam power stations, increased losses on the 132kV system and extensions to LV networks. Total energy generated is projected to grow at 26.3% p.a. during 1981-87 and 10.9% p.a. during 1987-92 (Table 5.5). Table ).5 Generation and Maximum Demand, YGEC Actual Forecast Annual Rate of Growth (%) 1981 1987 1992 1981-87 1985-92 Isolated Systems Sales 161.3 19.5 20.0 -29.7 0.5 Gross generation (GWh) 228.0 25.3 26.0 -35.6 0.5 Maximum demand (MW) 53.9 6.7 7.4 -29.4 2.0 Losses (X) 29.3 22.9 23.1 Load factor (X) 48.3 43.1 40.1 Interconnected Systems Sales (GWh) - 699.6 1,207.4 - 11.5 Gross generation (GWh) - 900.1 1,526.4 - 11.1 Maximum demand (MW) 2/ - 172.8 289.7 - 10.9 Losses (X) - 22.3 20.9 - Load factor (x) - 59.5 60.1 - Total YGEC Sales (GWh) 161.3 719.1 1,227.4 28.3 11.3 Gross generation (GWh) 228.0 925.4 1,552.4 26.3 10.9 Maximum demand (MW) 1/ 53.9 198.7 329.3 24.3 10.6 Losses (X) 29.3 22.3 20.9 Load factor (%) 48.3 53.2 53.8 1/ Sum of maximum demands of each load centre. 2/ Coincident maximum demands of interconnected load centres. - :32 - 5.12 Allowing for diversity, the maximum demand on the interconnected system is forecast to be 133 MW in 1985, increasing to 290 MW in 1992. The cement works are projected to have a coincident demand of 24 MW or 18% of the total peak demand in 1985 and account for 24% of energy generated on the interconnected system. Other bulk consumers are forecast to contribute a further 20% to peak demand. Therefore, nearly 40% of projected maximum demand in 1985 is due to cement works and other bulk consumers. Any change in their plans will consequently have a major influence on system demand. B. Future YGEC Supply Institutional Responsibility 5.13 YGEC is responsible for planning the development of its system. In practice, because of shortages of professional staff, the Corporation has been almost totally dependent on expatriates or consultants. The Corporation relies on its own staff to plan some of the small rural electrification and distribution projects, especially those requiring local generation. All projects must be approved by MEW which consults with CPO over project financing plans. However, major decisions involving generation and transmission to the main cities are firnalized at cabinet level. A certain amount of project identification, such as the location of major power stations and the choice of villages to be electrified by YGEC, is carried out at the highest levels of Government. YGEC hasi been improving its project identification capability, especially in connection with the national rural electrification study financed by IDA uLnder the Power II Credlit. However, this capability needs considerable strengthening. Generation Planning 5.14 The techniques adopted by YGEC's consultants for generation planning were reviewed by the mission and found appropriate. A disadvantage in relying on consultants for all generation planning has been that the GoverrLment cannot get quick economic analysis of options which arise from a changing environment. Because of YGEC's shortage of professional staff, the Government has had to take major decisions on the size and timing of generating units, together with decisions on financial packages, without the benefit of economic or financial analysis of the proposals being offered to it. However, since no major new generation or transmission investment is now likely, until the 1990's, it is not a high priority for YGEC to develop its own capabiliby for planning major generation or transmission investments at present. Nevertheless, when the time arises for major investment decisions to be taken, YGEC should arrange to have consulting services or expatriate advisors on hand to provide economic and financial analysis of supliers' proposals. 5.15 Generation planning for areas outside of the three main cities of Sana'a, Hodeidah and Taiz has been carried out by YGEC itself. Because of the urgent need for public supply and the Eshortage of professionaEl staff, such planning has been of an ad hoc nature, generally considering a standard size generator (typically 530 kW) to meet an immediate requirement. This approach was not unreasonable, given that YGEC was given the responsibility for providing reliable supply in face of rapidly growing demand in electrified areas and unsatisfied demand elsewhere, so that the Corporation did not have - 33, - time to carry out economic analysis of these projects. YGEC will have surplus diesel generators once the interconnected system is completed to supply the main rural centers by 1986. There will also be surplus generating capacity on the interconnected system. YGEC will haLve to decide whether to supply villages from the main system by constructing sub-transmission, or use surplus diesels. Such decisions should be basedi on economic analysis, which would enable the Government to set priorities for rural electrification. These projects should give YGEC staff the opportunity of developing their skills in investment appraisal. Since YGEC is unlikely to require major generation investment in the medium term, it is recommended that YGEC concentrate on strengthening its capability planning small for isolated rural systems. Proposed Generation Development 5.16 YGEC is committed to two major generation projects. These comprise oil fired steam stations at Ras Katenib (165MW) and at Al Mukha (160MW). The Ras Katenib station is almost complete. The contract for Al Mukha reportedly envisages the first unit being commissioned by 1986. Based on the time taken to complete the Ras Katenib and other projects, the mission believes that the first two of Al Mukha units are not likely to be in full operatioa before January 1987 and the remaining two before 1988. Table 5.6 shows forecast electricity supply for the public system up to 1992. It appears that further generation is required in 1992. No decision on the type, fuel or size of this need be taKen for several years. However, it is likely that a 50MW of combustion turbine capacity would be commissioned in 1992, since diesel peaking plant would reach the end of its economic life in the early 1990's. However, if the geothermal exploration program is successful (para. 2.U6), a geothermal power station could be commissioned in the early 1990's. In 1984, after Ras Katenib is commission, 60% of YGEC's installed capacity will be steam. This will rise to 74X in 1988 after Al M4ukha is completed and some diesel units retired. In 1992, the projportion of steam plant would fall to 66% of total installed capacity if a gas turbine were installed. Table 5.6 Projected YGEG Electricity Supply Ac tua 1 Forecast 1981 1987 1992 Interconnec ted System Installed Capacity Steams: Ras Katenib - 165 165 Al Mukha - 80 160 Total - 245 325 Diesel - 72 67 Other - 50 Total T-37 442 Available capacity 238 332 Maximum demand 173 290 Margin (% demand) 38 14 isolated Systems Installed capacity (diesel) 109 42 49 Available capacity 78 31 36 Maximum demand 53.9 6.7 7.4 Margin (% demand) 51 360 387 Source: Annex 3, Attachment 5 - 34 - 5.17 YGEC is likely to have considerable excess capacity for the rest of the decade. Af ter allowing for scheduled maintenance and breakdowns, the margin of spare capacity will be above 10 in every year and as high as 40% when Ras Katenib and Al Mukha are brought into commercial operation (Annex 3, Attachment 5). This is partly a consequence of the lumpiness of unit sizes needed to capture economies of scale. Partly, it is due to a downward revision to the load forecast (para. 5.08). However, it is also a consequence of ordering units in advance of need. This may have been justified if non-fungible soft credit were available for these projects, although any savings in capital cost need to be set against the earlier commitment to higher staffing costs. 5.18 High staffing costs are one of the main reasons why YGEC at present makes financial losses while having some of the highest electricity tariffs in the region. Commissioning any capacity at Ras Katenib and Al Mukha in advance of need would exacerbate YGEC's financial problems, particularly since the stations would have to rely on expatriate staff, at least initially. High manpower costs are largely unavoidable, but nevertheless YGEC should continue to search for marginal operating cost savings in its power stations and transmission system. Potential areas of saving might include placing temporarily surplus steam units in rotational storage, not fuLly manning steam stations with temporary surplus capacity, use of mobile maintenance teams and unmanned operation of substations. It is recommended that YGEC examine such options for saving skilled manpower and costs per unit of o!5J)ut; and assign one of its advisors with the responsibility of identifying operational savings. Fuel supplies 5.19 By 1985 steam plant is forecast to produce over 80% of electricity generated by YGEC. This proportion is likely to increase to above 85X for the remainder of the decade. There will be a corresponding change in YGEC's fuel consumption from gas oil to fuel oil (Annex 3, Attachment 6). YGEC's consumption of gas oil would fall from 52,000 tons in 1981 to 27,000 tons in 1987 and, if a gas turbine were commissioned, rise slightly to 41,000 tons in 1992. Fuel oil consumption is projected to be 227,000 tons in 1987 and 385,000 tons in 1992. There is considerable uncertainty about: the existing fuel consumption of norr-YGEC producers aLnd their future consumption is even more uncertain. Allowing for YGEC to extend its service as forecast (para. 5.08) and taking the midpoint of the range of estimates, gas oil consumption of private producers would decline from 118,000 tons in 1981 to 54,000 tons in 1987 and 50,000 tons in 1992. Their consumption of both gasoline and kerosene would decline from 24,000 tons in 1981 to 19,000 tons in 1987 and 18,000 tons in 1992. 5.20 A pipeline has been proposed from the proposed oil terminal at Salif to Ras Katenib, which would be separate from the other project for petroleum product pipelines (para. 2.08) . The pipeline to Ras Katenib from Salif would enable YGEC to obtain a lower oil price from importing fuel in larger tankers. However, this project needs careful justification. YGEC is likely to be able to procure medium fuel oil (1500sec Redwood) at a price of US$200 per ton, (October 1982 prices) which is below the present C & F border price for small shipments of $219 per ton. A reduction of more than $10 from $200 per ton arising from larger shipments iS unlikely. Furthermore, the Al Mukha power station, will come on stream before demand has grown to the point where - 35 - Ras Katenib will reach full output. Al lMukha will burn slightly cheaper 3500 sec oil and will have a higher thermal efficiency. It will therefore be operated in preference to Ras Katenib so that Ras Katenib may not need to operate continuously at full load. The economic case for a fuel pipeline from Salif to Ras Katenib is therefore uncertain. It is recommended that YOMINCO or YGEC carry out a feasibility study of the fuel pipeline from Salif to Ras Katenib before taking a decision to procede with it. Transmission 5.21 Transmission lines have been constructed from Ras Katenib to Hodeidah, Sana'a and Taiz. The lines to Hodeidah and Sana'a will be brought into service in 1983 and the Taiz line in 1984. Delays in completing 132/33kV substations and 33/llkV distribution substations have delayed the commissioning of the interconnected system. 5.22 YGEC is committed to building a further 132kV line to connect Al Mukha to Taiz. A branch from this line at Al Barh is proposed to cross the Tihama to Bajil. The Tihama line is intended to supply Zabid and other towns in the region. It will also improve the security of supply when demand grows in Sana'a and other highland load centers. Other 132kV lines are planned to Hajjah and Sada'a. It is doubtful whether the 132kV lines to Zabid, Hajjah and Sada'a are economically justified at present, especially when YGEC will have surplus diesel capacity available for local generation. The Tihama line could possibly be delayed until justifiedl on the grounds of system power flows and security. Hajjah could be connected using the transmission line from a future steam station located at Salif, although such a station is unlikely before the early 1990's. 5.23 Transmission planning studies have concentrated on the technical feasibility of the planned development of the system. Even though some transmission requirements are dictated by power station location, a number of alternatives exist that should be subjeci: to economic analysis. These alternatives mainly concern the timing of projects rather than their economic viability per se. These include extending the 132kV transmission system to Sa'ada and Hajjah, compared to local generation, the timing of reinforcements such as the Al Barh to Bajil 132kV line and the choice between local peaking plant and delaying the development of steam generation and 132kV transmission. It is recommended that the planning of transmission be strengthened to place a greater emphasis on economic as well as technical issues. YGEC should prepare terms of reference for future transmission studies that ensure a thorough examination of the economics of alternatives. Distribution and Rural Electrification 5.24 YGEC plans further reinforcemenit and extension of its urban distribution networks and to expand distribution facilities into rural areas. During the years 1982 to 1986 it plans to complete 830km of overhead MV line, 90km of LV line, 220km of MV cables and 78km of LV cables. The plan focuses on MV distribution which should lead to :Lower losses. However, 700km of the 830km of MV overhead lines is at 33kV and most of this will be to connect towns and villages to 132kV bulk supply points. 5.25 Distribution planning for the main cities has generally been carried out by consultants, although YGEC plans and executes some projects intended to - 36 - overcome immediate problems. A nationwide rural electrification study carried out by NRECA, with financial assistance from IDA, established the design philosophy for rural distribution. YGEC has designed and executed a number of rural electrification schemes itself. Consultants have been retained for large programs, such as that financed under the Power II Credit. 5.26 YGEC's general problems of shcortages of manpower, financial constraints and difficulties in project implementation have had the greatest impact on distribution. The extension of distribution networks in both urban and rural areas has fallen behind schedule and has not developed in step with generation. Delays to some apparently minor distribution contracts, caused by shortages of finance or contractual problems, have delayed the commissioning of major projects. YGEC has found finance for generation and transmission projects more readily available than for distribution. Furthermore, distribution, rural electrification, generation and transmission projects have generally been considered one at a time in isolation from one another. This is not surprising in view of the scattered systems which YGEC consisted of and the lack of finance. However, distribution investment has lagged behind generation investment to the extent that only about 40% of the output of the Ras Katenib power station could be delivered when the station is commissioned. As the system develops, there is a clear need for the plan for urban and rural distribution to be brought together with those for generation and transmission into an integrated program for YGEC. Integrated Power System Planning 5.27 The overriding issue in power system planning is the need to formulate a balanced investment program so that generation, transmission and distribution investment can achieve better balance and coordination. This would bring together projects that have already been identified into a coherent, feasible stragegy. The priority at present is to expand the distribution system in order to utilize excess generation and transmission capacity. The immediate action required should be to formulate an overall medium term investment plan for the next ten years which would focus on distribution. Such a plan would establish priorities, ensure the balanced growth of distribution and rural electrification with generation and transmission, and be consistent with the resources available to YGEC. This plan would represent the least cost development of the system, given that major generation and transmission projects are committed. Furthermore, it would enable modifications to the investment program to be balanced among each element of the system and be consistent with YGEC's overall economic objectives. Such modifications could be brought about by, for example, unforseen changes in the availability of finance, project delays, or divergence between projected and actual demand. It is therefore recommended that YGEC formulate an investment strategy that would bring together projects identified in existing consultants' studies, establish priorities among them and take a realistic view as to what can be achieved during the next 10 years. It should be formulated to enable balanced changes to be made to the strategy when unforseen circumstances eventuate. 5.28 It is now clear that the financial, manpower and other constraints on the ability of YGEC to implement projects were greater than realized when the original plans to expand the system were formulated. Projects have been delayed, necessitating expensive emergency measures, lower quality of supply, - 37 - or restrictions on new connections. A Lesson from the recent past is the need for flexibility in planning to avoid commitment too early to major investment decisions that may require adjustment later in response to changes in demand and delays to other projects. The need for flexibility, particularly the timing of projects, should continue since future demand remains highly uncertain and the choice of fuel and location of the next generation project will be influenced by the results of the geothermal exploration drilling. YGEC should continue to formulate a long term generation program to guide the design of immediate projects, but avoid taking decisions on particular projects within the program until such decisions become necessary. The long term program should be reviewed annually. This could be achieved by means of a Planning Memorandum which would be up,dated each year. The Planning Memorandum would identify decisions relating to each step in the project planning and execution sequence and set dates by which they should be taken, to be reviewed annually. It would examine the economics of tactical decisions such as whether to place firm orders for only the first two generating units in a new power station with options for purchasing additional units later, which might increase the total cost of the station, or whether to order a complete station at once, which might have a higher present value cost. The planning memorandum would be an annual updating of the proposed financial recovery plan. It should be short and policy orientated with technical annexes describing the analysis underpinning major recommendations. It is recommended that each year YGEC should prepare a medium term planning memorandum which should include a coordinated program for timely investment in generation, transmission, distribution and rural electrification; a financial plan and projections of accounting statements; and an updated manpower and training plan. C. Autoproducers and Private Power Systems Autoproducers 5.29 Most major industrial autoproducers are likely to take public supply, providing suitable tariffs are offered by YGEC. New industries are also likely to take YGEC supply unless they are located in remote areas where connection to the grid is uneconomic. Existing generating plant will presumably be retained for standby when YGEC supply fails. Some of the larger installations, e.g. cement works or the Taiz biscuit factory, could be used in future for peak lopping, or even to provide power to YGEC. However, this would involve extra costs of synchronizing equipment and relays, which would need to be evaluated. 5.30 Many, if not most of the small (less than 5kW) units used to supply households, shops and workshops will require replacement during the next 2 to 5 years. These are likely to be replaced by similar units unless YGEC supply is provided in the village or private utilities or cooperatives are formed. Many of the private distribution systems are in poor condition and will require replacement to maintain continuity of supply. Improving Technical and Safety Standards 5.31 Some savings in national resources could be made if private distribution systems were constructed to an adequate standard to enable them - 38 - to be integrated with the YGEC system. ]From the national viewpoint, improvement in the standard of private networks is probably economically justified in terms of reduced losses and longer physical life, even if they remain as private systems or are not physically connected to the YGEC system. Moreover, there would be advantages in improved public safety if private networks met minimum safety standards. Private networks invariably operate at LV where the technology is simple. Savings would be made if private LV networks could be ultimately connected to the YGEC MV distribution system without being rebuilt or rehabilitated. Private producers are unlikely to construct LV networks to adequate standa;rds unless first, it is financially worthwhile to them to do so; and second, they know what safety and technical standards are acceptable. It is therefore recommended that MEW draw up standard terms for YGEC to use in either purchasing private networks, or supplying bulk electricity to them. These terms should link compensation to the condition, age and design of the network and provide incentives to the private producer to connect a high proportion of households. 5.32 Encouraging the construction of private networks to adequate standards is more difficult. Manpower and Government finance available for electrical supply that is not already committed is likely to be extremely scarce during the 1980's. Assistance to the private power subsector should not result in delaying the development ofE YGEC into a strong institution that is financially viable. YGEC appears likesly to and will continue to be the major institution in the subsector, particularly since it is the only organization capable of capturing the economies of scale from central power generation and the fuel savings from steam plant. There are several ways that MEW could provide technical assistance to the private power subsector. Such measures include: a) providing drawings and technica:L handbooks to private electricity suppliers, merchants and workshops; b) commissioning television and radio programs to explain how networks should be constructed. Other m:inistries use the mass media for similar purposes; c) persuade merchants to stock distribution equipment that meets YGEC's standards; d) use YGEC facilities for short training courses in LV distribution; and e) allow YGEC to sell limited specialized technical serviices, such as meter testing. In view of the importance of private suppliers in the subsector, it is recommended that MEW create a division responsible for private electricity affairs. This would formulate and implement measures to improve the efficiency of the subsector, monitor its performance and represent its interests within the Government. InitiaLly, such a division might need to consist of an engineer, an administrator responsible for licensing and purchasing networks and at least three teschnician field assistants, all fluent in Arabic. - 39 - VI. ELECTRICITY PRICING A. Institutional Responsibility for Tariffs 6.01 YGEC is responsible for designing its own tariffs. Tariff proposals from YGEC have to be approved by the Minister of Electricity and Water. He in turn obtains the agreement of the cabinet. As in many countries, electricity tariffs of the public supplier are politically sensitive and may in some cases require the approval of the President. YGEC has a commercial adviser, financed by UK ODA, whose responsbilities include tariff design. The adviser is currently preparing a proposal for tariffs for industrial consumers. This proposal is expected to be completed by April 1983. If approved by YGEC and the MEW, it would be implemented in late 1983. 6.02 In principle, private suppliers can charge what the market would bear. Their prices are constrained only by; (a) the costs of competing small scale autoproduction for individual houses and shops; (b) the general level of private electricity prices in the area; and (c) local leaders and villagers who might influence tariffs informally. B. Histori.cal Review Existing Tariffs 6.03 YGEC has two tariffs, one for sales to consumers in the urban centers of Sana'a, Hodeidah, Taiz and Dhamar anid the other for sales to consumers in the rest of the country, which are mainly at rural branches. All categories of consumer in either urban or rural areas are charged a uniform tariff, i.e. the same rate irrespective of the kWhs consumed. These tariffs are: Urban 110 fils/kWh (US$0.24/kWh) Other 150 fils/kWh (US$0.33/kWh) In 1982, the rural tariff was 30% higher than the urban tariff. YGEC has been able to sustain higher tariffs in rural areas because the opportunity cost to the consumer of private electricity supply is still considerably greater than YGEC's tariff. 6.04 During the period 1975-1982, urban electricity tariffs fell in real terms by about 20%. On the other hand, rural electricity tariffs increased by nearly 9% during the same period. - 40 - Table 6.1 Electricity, Fuel and Retail Price Indices (1976 - 100) Fiscal Retail Prices Urban Electricity Rural Electricity Diesel Fuel Year Sana'a Current Prices Current Prices Current Prices 1976 100 100 100 100 1977 126 100 100 97 1978 147 100 100 99 1979 179 106 111 118 1980 198 144 194 162 1981 212 174 250 293 1982 230 183 250 286 Rate of increase 1975/81- (% p.a.) 14.9 10.6 16.5 19.1 In general, the increases in electricity tariffs fell short in compensating YGEC for the increases in its cost of operation. The cost of fuel represents 41% of total operating expenses and wages and salaries account: for a further 32%. During the period 1978-1982 the price of fuel increased by 189% and wages per kWh increased by 48% over the same period while tariffs were increased by only 83%. As a result, the financial position of- YGEC deteriorated to the extent that the Corporation covered its operating costs during 1979-1981 only by arbitrarily capitalizing much of its head office expenses (see Chapter VII). 6.05 The Power II Credit contained a covenant for YGEC to implement a fuel price adjustment clause. This was intended to translate the gradual removal of petroleum subsidies into higher electricity tariffs. However, the Government decided to remove petroleum subsidies at a stroke. The implied rise in electricity tariffs would have been politically unacceptable so that IDA agreed to relax the Covenant, in view of the importance of- removing petroleum subsidies. Fuel prices generally are not expected to increase in 1983, but might increase faster than the rate of inflation thereafter. It is therefore unlikely that any increase in tariffs to compensate for increases in fuel prices will be required in 1983. From 1984 onwards, the possibility exists that increases in fuel prices could upset any plan intended to bring the finances of YGEC to a sounder footing. Automatic fuel price adjustment of tariffs is the most simple means of compensating quickly for fuel price increases with minimum administrative effort. However, to date the Government has wished to avoid complicating tariffs to the consumer in this way. An alternative would be for an interministerial committee to review the level of YGEC's tariffs every 3 months and then propose tariff increases to the Council of Ministers when fuel prices increases. It is recommended that the Government consider undertaking all necessary measures to ensure that all increases in the costs YGEC's operations, particularly fuel, are reflected in the tariffs as quickly as possible to guarantee the continued viability of the Utility . - 41 - 6.06 Tariffs charged by private producers vary considerably, but generally are about three times higher than those charged by YGEC. However, despite the vast difference between the private and YGEC's tariffs, many households in rural Yemen subscribe to private suppliers because supply from YGEC is not available. This is a clear indication that the consumers' willingness to pay for electricity is substantially higher than the present level of YGEC's tariffs which justifies the maintenance of YGEC's tariffs at their present level (para. 6.12). C. Tariffs and Economic Costs Tariff Levels 6.07 Existing tariff levels are compared to the economic cost of supply in Table 6.2. The assumptions used in estimating the economic cost of supply are described in Annex 4. The present tariff, on average, is almost 40% above the economic cost of supply. When applied to MV consumers it is between 2.7 to 3.1 times above the economic cost. However, YGEC supplies very few MV consumers, partly because of the past unreliability of supply and partly because private generation has been cheaper than public supply. The tariff is 1.3 times above the LRMC of supplying households, 1.4 times for shops and 1.9 times for workshops. Rural consumers are probably paying close to the economic cost. The gap between tariff levels needed to meet financial targets Table 6.2 Comparison between Existing Tariff Levels and Economic Cost Existing as % Existin Tariff Economic Cost of Economic Cost (YR/kWh) (uS/kh) YR/kWh) (US/kWh) Consumer Type Bulk Supply Cement .. 0.299 6.6 MV Bulk Industry 1.10 24.2 0.387 8.5 284 Offices 1.10 24.2 0.361 7.9 305 Hotels & Shops 1.10 24.2 0.405 8.9 272 Low Voltage Workshops 1.10 24.2 0.567 12.5 194 Shops 1.10 24.2 0.766 16.9 143 Households 1.10 24.2 0.832 18.3 132 Rural 1.50 33.0 1.440 31.7 104 Average Tariff 1.17 25.7 0.842 18.5 139 and the economic cost of supply should gradually narrow as the system approaches equilibrium. The large difference between the existing tariffs, based on accounting costs and economic cost is a consequence of the temporary - 42 - excess capacity at Ras Katenib and in 132 kV transmission, as well as the decision to build the Al Mukha power station in advance of need (para. 5.17). If economic costs were calculated assuming that the timing of the entire Al Mukha station and associated transmission could change in response to changes in electricity demand, the tariff for households would be only 9% above economic costs and the rural tariff would be 10% below economic cost. This estimate of economic cost is overstated, even if the timing of Al Mukha is flexible, since the true economic cost of the project is lower than the figures used in the calculation because of the concessional, non-fungible funds used to f inance the project. However, since the Government appears irrevocably committed to building Al ?kikha according to contract, it is incorrect to caculate economic costs assuming that the timing of the project is flexible. Since, the present tariff is roughly equal to economic cost calculated on the basis of Al Mukha's itiming being flexible, when the urban and rural tariffs are averaged, there is little justification for lowering existing tariffs to economic cost calculated on the other basis when YGEC is already needing operating subsidies from the Government (para. 6.12). 6.08 Despite the high level of tariffs in relation to, economic cost, YGEC has made losses in fiscal 1980 and in the six months to Decermber 1981. Unless tariffs are increased, these losses are projected to continue until 1988 (see Chapter VII). Tariffs are high because of the high cost of Labor, which is partly due to the cost of employing expatriates to run the utility, and the high cost of diesel fuel delivered to YGEC power stations. Average costs are likely to remain high in the future because: a) underutilized plant will result in high depreciation and other fixed costs; b) YGEC will continue to rely on expatriates; c) fuel prices ar,e likely to rise in the medium term; d) debt servicing costs are likely to rise when grace periods expire on existing loans. 6.09 The existing YGEC tariff is above the costs of private generation. Ihese are about YRIs 0.88/kWh (US¢19.4/kWh) for a new installation and YRls 0.69/kWh (USO15.2/kWh) for operating an existing installation. These are typical costs faced by an enterprise, but the economic costs in terms of national resources are almost identical. A potential autoproducer has little incentive to choose public supply. For potential MV consumers, the economic cost of autoproduction is roughly twice the economic cost of public supply. Therefore, from a resource allocation viewpoint, public supply is more economic and represents the least cost option for meeting the future demand of potential MV consumers. In view of the substantial differences between autogeneration and public supply in the economic cost of MV supply, it is recommended that the Government move to restrain the wide use of autogenerators in areas where YGEC supply is available primarily by restructuring the YGEC MV tariff (para. 6.15); and (b) also generating licenses for autogeneration under Article 9 of Law 12 of 1975 only where this is clearly in the interests of the national economy and the firm, such as when YGEC distribution facilities are inadequate, or when the firm requires a high standard of supply reliability. - 43 - D. Electricity Tariff Policy Background and Objective 6.10 Electricity pricing for YGEC needs to be set against the background described above, the main elements of which are; a) YGEC tariffs are above economic costs for all classes of consumer. Economic costs are likely to rise in real terms as demand grows to approach optimal balance with supply; b) YGEC tariffs are also above the cost of autogeneration. The cost of autogeneration, in turn, is greater than the economic cost of YGEC supply; c) tariffs charged by private suppliers are several times greater than YGEC's tariffs. Moreover, tariffs charged by private producers are greatly in excess of the economic cost of private supply and probably well above the financial costs perceived by the producers. The fact that many households in rural Yemen pay these high tariffs demonstrates substantial willingness and ability to pay for electricity; and d) YGEC is likely to continue making losses in the medium term, if tariffs are held at l:heir present level. 5.11 The environment in which electricity is priced in YAR suggests two main objectives for electricity pricing: a) to ensure the financial viability and to mobilize resources to enable YGEC to reduce its burden on public finance; b) to encourage the least cost supply of electricity from the national viewpoint. Tariff Levels 6.12 On economic efficiency grounds, tariffs should be set at levels that cover the economic cost of all the resources used in the generation of electricity. These tariffs should then be adjusted to ensure that the revenues generated from the sale of electricity would provide the internal resources needed to finance the investment plan of the utilities. YGEC's financial position has been deteriorating mainly because of two reasons: a) the lumpy investment, traditionally needed to establish a new utility such as YGEC, which would not be optimally used for the first five to seven years of operation (transmission, distribution substations, etc.); and b) the substantial increases in the cost of inputs used in the generation of some of electricity; namely, fuel and expatriate technical staff, some of which are expected to be needed for another five to ten years. These factors would require the maintenance of tariffs above economic cost if YGEC is to ultimately become financially viable. Moreover, judging by the consumers' willingness to pay and the fact that remittances from abroad accrue to individuals rather than the Government, which is expected to face shortage of resources in the future, there are strong grounds for maintaining the tariff level at, and above, its present level to mobilize resources for the power subsector. There is, however, a substantial potential for YGEC to mobilize more resources by restructuring its tariff by lowering it for sale to large consumers whose connection to YGEC's system in the near future would improve the utilization of existing facilities. The new industrial tariff would attract large consumers and provide disincentives for autogeneration. However, such restructuring should be undertaken while ensuring that the average revenue per kWh would, as a minimum be at its level of 1983. Therefore, from a resources mobilization point of view, the maintenance of the - 44 - YGEC's tariffs above economic cost of supply is justified in economic terms and the Government should consider all measures to ensure that tariff levels over the five to seven years are set so as to assure the gradual progress of the utility to financial viability. However, in view of the fact that tariffs are already high and given the political sensitivities involved, the Government could elect to postpone any action to raise the taLriffs across the board until after 1983. Tariff Structure 6.13 The present tariff structure does not provide incentives to shift peak demand to off-peak periods. Moreover, there is considerable cross- subsidization between different categories of consumer (Table 6.2). Restructuring tariffs to encourage off-peak consumption is likely to be economic only for MV consumers, at least initially, since the! low level of consumption of most LV consumers would not justify the additional metering costs. Excess generating and transmission capacity means that, at present, encouraging peak load to shift off-peak is not an urgent priority, although it is desirable to encourage patterns of consumption that are economic to supply in the future. A strategy for implementing restructured tariffs for MV consumers is shown in para. 6.15 below. 6.14 Any restructuring of LV tariffs, to reflect the different costs of supplying each consumer group without changing the total income of YGEC, would lead to higher domestic tariffs and lorer tariffs for commerce and small industry. This probably would be politically unacceptable. However, if a general increase in tariffs were to take place, some restructuring of LV tariffs could take place to take account of the different economic costs of supply for each group of consumers. Nevertheless, any such restructuring should be implemented slowly so as to continue to capture the willingness to pay for electricity and be based on improved estimates of economic costs obtained from load research. As a firslt step, consumers should be reclassified into homogeneous groups when YGEC's accounts are computerized (para. 4.02), and small differences in tariffs introduced after the following general tariff increase. Such work would require a strengthened Consumer Department. It is recommended that an overall restructuring of tariffs based on LRMC's be delayed until an increase in tariff levels is required, but meanwhile YGEC should strengthen its Consumer Department and carry out load research and consumer reclassification. 6.15 Restructured tariffs for bulk consumers should be introduced immediately. This is to achieve efficient resource allocation by encouraging autoproducers to switch to cheaper public supply and to generate income for YGEC from additional sales. Because ol the time taken to implement a restructured tariff and the present excess capacity, which will lead to the economic cost increasing over time, it is recommended that restructured tariffs for bulk consumers be introduced in the following stages: a) YGEC should immediately offer tariffs above YGEC's e-conomic cost, but below the variable cost of private generation, to attract existing autoproducers to public supply and to ensure that new industries are also supplied by YGEC. A flat: rate tariff of around YRlsO.75/kWh should be implemented immediately, although the precise rate would - 45 - depend on what price would induce! autoproducers to take supply from YGEC. This tariff should be regarded as an interim measure and should last for about 3 years; b) YGEC should carry out load research on bulk consumers to improve the estimates of the economic cost of supply, particularly the costs they impose on the distribution system; c) since economic costs will rise as excess capacity diminishes there will be a greater incentive to enicourage peak demand to shift off-peak. YGEC should gradually introduce more elaborate tariffs for MV consumers to encourage demand to move off-peak, concentrating most on the largest consumers and those most able to respond to such a tariff. This could be achieved by time-of-day tariffs. Procurement of meters for these tariffs should be scheduled accordingly; d) time-of-day tariffs for MV consumners should eventually evolve into tariffs based on the economic cost of a balanced system, once YGEC achieves financiai viability and excess capacity vanishes; and e) time-of-day tariffs are likely to encourage industries to operate their generating plant to supply their own needs during the peak. Some of these installations may have surplus capacity. It is recommended that YGEC consider entering into agreements with the large autoproducres, e.g. cement works, to supply power to the public system in the future. This might enable YGEC to delay some generation and improve the reliability of supply. Since excess capacity is likely to persist throughout the decade, co-generation is not a high priority at present. - 46 VII. INVESTMENT AND FINANCING A. Past Investment and Financing Past Investment 7.01 Between July 1976 and December 1981, investment by YGEC amounted to YRls 1140 million (US$251 million) at current prices. As Table 7.1 shows, YGEC's investment was about 5% of total gross fixed capital formation in YAR during that period and this proportion had tended to rise. YGEC's level of investment increased because of expenditure on large lumpy investments towards the end of the 1970's, namely the steam power station at Ras Katenib and associated transmission to the main cities. Most YGEC investment has consequently been in generation and transmission. Trade data suggest that the investment by the private power producers during 1976 to 1981 was probably in the order of YRls 250 million (US$55 million). Table 7.1 Investment by YGEC 1976 to 1981 (YRls million, current prices) Fiscal YGEC Total Gross Fixed YGEC Year Investment Capital Formation (YAR) (X) 1977 69.0 1496 4.6 1978 69.8 3160 2.2 1979 97.0 4445 2.2 1980 259.4 4882 5.3 July-Dec. 80 169.0 (2573) 6.6 1981 475.4 5600 8.5 Total 1139.6 22,156 5.1 Sources: YGEC, CPO Project Implementation 7.02 Delays have occurred in the implementation of some projects. The Ras Katenib power station contract envisage!d the first unit being; commissioned in January 1982, whereas the feasibility study had recommended that the first unit be commissioned in January 1981. The station should enter commercial operation in mid 1983, about 30 months later than originally planned. Similarly, the project for reinforcing urban distribution is about 24 months behind schedule. Major projects have been delayed because of delays to relatively minor contracts. For example, part of the delay to commercial operation of Ras Katenib has been caused by the failure to complete relatively minor distribution contracts. 7.03 The main sources of delays to project completion have been the time taken to arrange local and foreign finance; administrative delays during bid evaluation, contract negotiations and the opening of letters of credit; and - 47 - complete4 plant being unable to be brought into service because of delays to complementary investment. However, despite the time taken to finalize a financial plan, YGEC has had considerable success in financing an ambitious program. The Corporation should aim to reduce delays in financing projects by identitying its financial needs well in advance, so that new finance can be co-ordinated better with investment needs. YGEC should examine ways to reduce pre-contract delays by preparing realistic programs for project implementation and carefully monitoring their progress. Co-ordination of contracts needs some improvement so that the importance of each can be judged against an overall investment plan. Moreover, there is scope for strengthening YGEC's project management skills, through the adoption of modern techniques for project management. It is recommended that YGEC engage the assistance of consultants or recommend suitable procedures and techniques for strengthening groject management and financial planning. The consultants should also identify training needs. This consultancy could be included in the proposed financial recovery plan. In the future, it is recommended that the programming of pre-contract.activities and the identification of project financial requirements and the co-ordination of contracts within an overall plan be achieved through the proposed Annual Planning Memorandum. Financing of Past Investment 7.04 YGEC's investment during 1976 to 1981 was financed mainly by long term loans, supplemented by injections of Government equity. However, the Government contribution to YGEC financing, as well as the amount of total investment itself, is understated by an uanknown amount of contractor finance serviced by the Government and not enteried into YGEC's accounts. The change in YGEC's capital structure is shown in Table 7.2. The ratio of debt to equity increased from 0.6 in 1976 to 1.4 in 1981. The proportion of debt finance, in 1981 would have been even higher if loans from the Saudi Fund and Arab Fund, amounting to nearly YRls 80 million, had not been turned into grants. Table 7.2 Financing of YGEC Investment (YRls million, current prices) Contribution Fiscal Years Increase to Total 1976 1981 1976-1971 Increase Equity & retained earnings 1/ 33.1 451.7 418.6 (38.2%) Long term debt 20.4 612.8 592.4 (54.0%) Consumer contributions & deposits 2.8 104.4 101.6 (9.3%) Net current liabilities -14.7 -31.2 -16.5 -1.5% Total 41.6 1,137.7 1,096.1 (100.0%) 1/ Includes employees' retirement fund. Source: YGEC Balance Sheets - 48 - 7.05 No net contribution to investment arose from retained earnings during the period 1976 to 1981. Despite having some of the highest tariffs in the region YGEC made operating losses in 1980 and 1981, if some Ihead office expenses which were arbitrarily capitalized are returned to operating expenses. This poor financial performance has been due to high staffing costs arising partly from the need to employ expatriates to run the utility, the high cost of delivered diesel fuel, increasing losses and inadequate cost management. YGEC has also been unable to increase its sales as fast as demand would indicate because of limitations to its distribution systems. The effect of lower than anticipated sales has been compounded by tariffs falling in real terms for most YGEC consumers. 7.06 Staffing costs should ultimately rise more slowly as the Corporation trains more Yemenis to fill positions occupied by expatriates, but this is dependent on an effective training program (para. 3.18). Average fuel costs should fall as fuel oil replaces diesel as the main fuel for power generation (para. 5.19). However, YGEC's manpower problems are not likely to be solved quickly, plant underutilized initially will impose depreciation and other fixed costs and debt service will rise when the grace periods on existing loans expire. Measures open to the corporation to improve self financing include improving efficiency by controlling operating costs, intensifying the utilization of existing facilities through tariff restructuring (para. 6.5) and by extending distribution networks to connect new consumers. However, without adjustments to the structure and level of tariffs, these measures are unlikely to produce a quick improvement in the financial position of YGEC. B. Financial Recovery Plan 7.07 Although many of the factors influencing YGEC's present financial position are now effectively outside of the Corporation's control, it is the view of the mission that some progress towards lowering costs could be made by improving the efficiency of YGEC through strengthening its management. There is a need to analyze YGEC's organization difficulties in detail on a systematic basis, design procedures for overcoming them and then implement these procedures. The activities whichn need the greatest strengthening are (a) investment planning; (b) finance and accounting procedures; (c) manpower planning, personnel policy and training; (d) project management; and (e) organization and administrative procedures. YGEC has reports prepared by its engineering consultants covering projects in urban distribution, grid connection of small towns, rural electrification, generation and transmission. To date these have not been brought together into a coherent feasible investment plan. YGEC has an accounting system designed by consultants, but this is being administered very poorly because of the lack of experienced and qualified staff. The Corporation has to rely on expensive expatriates yet employs a disproportionately large number of administrative, semi-skilled and unskilled staff. There is a need to plan future manpower requirements, use existing staff effectively and rationalize training. Cost saving procedures in maintenance, inventory control, along with improved technical operating and safety standards should be introduced. These issues should be addressed by employing management consultants who iwould (a) provide - 49 - assistance on organization, general management, finance and accounting; (b) make an extensive study of operating costs; (c) prepare a manpower plan and plan more permanent training programs; and (d) 'prepare a financial recovery plan and train YGEC personnel in accounting, inventory control and adminstrative procedures. These services are outlined in Annex 5. It is recommended that YGEC engage consultants to assist the Corporation to prepare a financial recovery plan and agree to arrangements for the prompt implementation of this plan. C. Future Investment and Financing Future Investment 7.08 Table 7.3 shows the projected investment plan for YGEC. Total investment is anticipated to fall after 1987 when the Ras Katenib power station and associated transmission is completed and the major part of expenditure on the Al Mukha power station has taken place. Major projects and their expanditure during 1982-92 include: completion of Ras Katenib, $36 million; Al Mukha power station, $205 million; and transmission from Al Mukha to Taiz and Bajil, $131 million. 7.09 The program shows distribution taking the largest share of investment (45 X during 1983-87 and 58% 1988-92). The focus on distribution is to compensate for under investment in the past and to enable more electricity to be sold from the committed generation and transmission facilities, thus improving YGEC's financial position. Table 7.3 Investment Plan for YGEC 1/ 1983 - 1987 1988 - 1992 (US$jR) (x WUSW (X) Generation 198.6 29.6 91.0 31.3 Transmission 117.3 17.5 16.5 5.7 Distribution 302.4 45.0 169.6 58.4 Other 53.2 7.9 13.2 4.5 671.5 100.0 290.3 100.0 1/ Excludes interest charged to construction. YGEC Operating Subsidies 7.10 With present LV tariffs and the introduction of a tariff of YRls 0.75/kWh (US016.5/kWh) for bulk consumers, YGEC would continue to make operating deficits. The deficits after interest payments would amount YRls 34 million or (US$7.5 million) in 1983 and YRls36 million (US$ 7.9 million) in 1984 and then fall to YRlslO million (US$ 2.2 million) in 1985 as Ras Katenib replaced diesel generation and sales increased. The Government therefore faces the choice of providing operating subsidies amounting to YRls80 million (US$17.7 million) during 1983 to 1985, in addition to the equity it provides for capital expansion, or raising tariffs. - 50 - 7.11 However, if the proposed Financial Recovery Plan were implemented to achieve a 3% rate of return on revalued average net fixed assets from 1986 to 1988, YGEC could begin to generate finance for'new investment: from its operations. Average tariff increases of about 16% in 1986, L4% in 1987 and 7% in 1988 would be required to meet this target. Despite such tariff increases, total self financing during 1983-1987 would be negative, amotnting to YRlsl2O million (US$26 million) which would need to be off-set by other sources of finance, probably from the Government. 7.12 Operating subsidies, in principle, can reduce the incentives to management to operate at maximum efficiency, especially if the annual amounts are not fixed in advance. Furthermore,, there is ample willingness to pay the existing YGEC tariffs. Moreover, a 23% tariff increase in 1986 would be substantially less than the projected 43% increase in local iinflation between the last tariff increase in 1981 and the proposed increase in 1986. The tariff level during the period 1983-85, before the financial Recovery Plan is implemented will depend on the degree to which the Government: wishes to subsidize YGEC, in addition to contributing to its investment. This in turn depends on the Government's budget priorities and the overall macroeconomic situation. Therefore, it is recommended that the Government fix ceilings for operating subsidies to cover current costs, depreciation and interest on long term loans for YGEC in advance and propose a program for phasing them out by 1985, with any operating deficit met by increases in tariffs. Financing of Future Investment 7.13 Assuming average tariff increases of 16% in 1986 and 14% in 1987, the financing plan for YGEC is as shown in Table 7.4. Government: finance including operating subsidies, is expected to account for about 90% of YGEC's investment during 1983-1987. Most of this would be in equity, but 21% of investment is expected to be loans and contractor finance raiLsed and serviced by the Government on behalf of YGEC. Long term loans serviced by YGEC, which include the first and second IDA credits and the proposed third power credit, are projected to amount to only 9% of total investment. The high proportion of Government finance is a consequence of the inability at present for YGEC to generate self finance. Despite the willingness of households, to pay the high prices charged by private suppliers, the Government considers that it would be inopportune politically to raise YGEC tariffs at present. Self financing would gradually improve if tariffs were increased each year in 1986-88 and the proposed Financial Recovery Plan were implemented. Greater self financing and less budgetary support would eventuate if the Government were to consider interim tariff increases before 1985. Table 7.4 Financing of New Investment 1983-87 $ million X Long term debt 61.7 9.2 Loans passed on to YGEC as equity 143.1 21.3 Government equity 452.8 67.4 Net available from operations 1/ 13.9 2.1 Total Finance 671.3 100.0 1/ After provision for working capital. - 51 - 7.14 To date, the private electricity sector has financed its investment from remittances from migrants. However, falling remittances might lead to other sources of finance being sought, especially since many private generators will require replacement during the next few years. The proposed section in MEW responsible for monitoring private power producers should review the financing of their investment and recommend policies to remove any institutional problems that might arise. June 1983 (1161P) - 52 - ANNEX 1 Attachment 1 YEMEN ARAB REPUBLIC POWER SUBSECTOR REVIEW Surplus/Shortfalls by Major Occupational Category Up to 1986 Need to Surplus(+) Demand from Replace Tota]L Total or Category Sector 1982-86 Foreigners Need Supply Deficit(-) A-1 Professional Science Based Public 1857 427 2284 1708 - 576 Private 497 295 792 1187 + 395 Total 2354 722 3076 2895 - 181 A-2 Professional Arts Based Public 7850 842 8692 2002 - 6690 Private 300 592 892 1226 + 334 Total 8150 1434 9584 3228 - 6356 B-1 Technician Science Based Public 2567 1255 3822 1219 - 2603 Private 889 358 1247 721 - 526 Total 3456 1613 5069 1940 - 3129 B-2 Sub Professional Arts Based Public 2501 1403 3904 2302 - 1602 Private 1753 72 1825 122 - 1703 Total 4254 1475 5729 2424 - 3305 C-1 Skilled Office Based Highschool Education Public 30186 8644 38830 8102 - 30728 Private 40264 358 40622 - - 40622 Total 70450 9002 79452 8102 - 71350 C-2 Skilled Prof. Preparatory School Education Public 4245 104 4349 2162 - 2187 Private 10978 2425 13403 2471 - 10932 Total 15223 2529 17752 4633 - 13119 D Semi Skilled Job Training Public 2706 63 28869 2706 - 63 Private 38515 2425 40940 2006 - 38934 Total 41221 2438 43709 4712 - 38997 E Unskilled Public 8154 48 8282 8154 - 48 Private 10938 2425 111563 251371 +139808 Total 117292 2473 119765 259525 +139760 GRAND TOTAL 262400 21736 284136 287459 + 3323 Source: CPU - 53- AIR= 1 Attachment YEMEN ARAB REPUBLIC POWER SUBSECTOR REVIEW Projected Manpower Demand and Supply for Selected Public Sector Occupations ToLal Manpower Increase in Manpower Requirement Yemen Supply Economic Attrition and Retireiment of Deficit as to 1981 1986 Growth Retirement Expatriates l/Total Local Training Foreign Training Total Deficit of Requirement Al Professional-Science Based University Degree Structural Engineers 17 66 49 - 13 62 2 25 27 35 56 Civil Engineers 118 316 198 3 45 246 10 236 246 - Electrical 6 electronic engineers 117 345 228 - 36 264 5 150 155 109 41 Mechanical engineers 67 191 124 2 36 162 5 128 133 29 18 Cnemists 14 25 11 - 6 17 - 17 17 - - Statisticians 22 210 188 - 10 198 - - - 198 1on Others 875 1916 1041 13 281 1334 182 1039 1130 204 15 Total 1230 3069 1839 18 427 2283 204 1595 1708 575 9c A, Protessional-Arts Based University Degree Econo:msts 80 227 247 2 22 271 176 53 229 42 15 Accountants 1054 2349 1295 23 153 1468 280 296 576 892 61 Department Heads 1374 2202 828 126 47 ID01 - - - 1001 100 Public Adm.n. Managers 142 221 79 1 .3 83 - - - 83 0nn Otners 2650 9058 4677 572 620 5869 1030 167 1197 4672 8n3 Total 6931 14057 7126 724 842 8692 1486 516 2002 6690 77 51 Sub-rofessional & Technician, Service Based Electrical and Electronic 122 152 30 - 9 39 39 - 39 - Mecianical 36 55 19 - 2 21 21 - 21 - Computer Programmer 31 75 54 - 5 59 - - - -5 10t) Transport Supervisors 57 101 44 - 6 50 - - - 50 100 C SK.lied-Hish School or Technical Training Uffice Supervisors 289 1124 835 6 3 844 - _ _ 844 100 Typists & Secretaries 1185 3339 1854 18 176 2048 256 - 256 2686 76 book-keepers 2286 5686 3400 90 65 3555 869 - 869 2686h 76 Storekeepers 655 1036 381 144 28 553 32 - 32 917 93 Correspondence Clerks 5052 10630 5588 591 158 6337 - - - -6337 100 dlacksmiths 14 23 9 - - 9 - - - q 100 Vehicle mecnanics 17 92 75 - 3 80 75 - 75 5 6 Electrical Maintenance 131 304 173 1 10 184 174 - 174 10 5 Electrical Appliance Repair 323 505 263 16 17 296 296 - 296 - 296 D Semiskilled Electrical & Linesmen 1050 2135 1085 39 - 1124 Sourca: CPO Nutes 1/ Expatriates shown are the total existing in 1981. - 54 - ANNEX 2 Page 1 of 17 YEMEN ARA3 REPUBLIC POWER SUBSECTOR REVIEW Electricity Demand, Past and Projected A. Available Data on Electricity Demand Availability of Electricity Consumption Data 1. Establishing past trends in electricity consumption in YAR is more difficult than in most countries. First, data on private electricity supply, which accounts for the greater part of electricity production, are restricted to a few limited surveys and other scraps of information. Second, some of the older data available for the public electricity supplier (YGEC) are incon- sistent or unreliable, although YGEC is steadily improving its database. YGEC appreciates the need for such data for effective day to day management of the utility, as well as planning for the future. An improved management information system, which should result from the proposed financial recovery plan and the adoption of computerized billing should improve the quality of data available for demand forecasting. 2. Some problems exist in using YGEC's present commercial data for demand forecasting. Consumer group definitions are too broad to serve as a base for demand forecasting, e.g. the private tariff category lumps shops and residences together. Tariff categories are not clearly defined by economic sector or in terms of the costs of supply. There are discrepancies in the number of consumers in some regions. Eurthermore, some of the generation metering is inaccurate at urban and rural branches. Some improvements could be made in presenting information to maLnagement systematically in regular reports. However, YGEC has made some recent improvements in reporting, particularly for information on generation and system demand. Future improve- ments could be made in the data on consumers and their consumption. To over- come these remaining deficiencies it is, recommended that: a) YGEC designate part time or full time counterparts to the Head Office statistics officer, responsible for local data collection in each region and branch; b) the consultants engaged for the financial recovery plan, design a management information system that would generate most of the data required for demand forecasting; c) monthly reports be produced that, in addition to financial and manpower data, would cover for each branch and region: - 55 - ANNEX 2 Page 2 of 17 (i) a detailed breakdown of consumer numbers; (ii) the corresponding sales; (iii) generation by unit; (iv) energy sent out, station use and losses; (v) maximum demand; and (vi) fuel consumption by unit; d) consumers be re-classified and their existence verified when regional and branch accounts are computerized; and e) in view of the number of branches and load centers, the database should be gradually computerized, starting with commercial data extracted from the new billing system. 3. The quality of the information on private electricity suppliers cannot be improved without a nationwide survey. Better information on private power generation would assist MEW in forming policies towards autoproducers and private producers, enable better forecasts of electricity demand and improve the planning of petroleum distribution facilities. There are three main options for obtaining better data on the private electricity subsector: a) a nationwide survey of electricity supply; b) as part of a nationwide survey of rural energy use, which would also examine fuels used for cooking, e.g. fuelwood; and c) CPO could consider including questions on electricity supply in the forthcoming population census and in surveys of industry, trade and agriculture. It is recommended that MEW, YOMINCO and CPO agree on a policy to improve the information available on the energy sector in general, and private power generation in particular, and that questions on electricity supply be included in the forthcoming population census and other surveys of production. B. Past Electricity Demand Growth of Overall Consumption 4. Electricity consumption in the areas supplied by YGEC was constrained by supply shortages for most of the 1970's. These shortages affected demand by: (a) reducing the number of consumers that could be connected; (b) shedding peak load; (c) lowering consumption through low voltage and frequency. Nevertheless, consumption grew rapidly. Electricity sold by YGEC grew at an average rate of 27.7% p.a. during 1970-81. Precise figures for the past consumption of electricity by the private supply sector and autoproducers are impossible to obtain. Electricity generated by the private sectors probably grew at an average rate of around 40% p.a. during the same period. National electricity demand is shown in Table 1. Total national production probably grew at about 35% p.a. during 1970-81. However, the rate of growth in generation for both YGEC and the private producers increased in the second half of the decade. This increase in the growth of electricity demand was a consequence of the economic boom brought about by remittances from Yemenis working in oil producing countries and Arab and other foreign assistance. - 5i5 - ANNEX 2 Page 3 of 17 5. Whilst the growth in demand was rapid, it did start from a very low base. In 1975 annual generation per capita by YGEC was 7 kWh and the 1975 national generation per capita was about 17 kWh. In 1981 these figures had increased to 31 kWh and 92 kWh respectively. Consumption per capita is low compared to other countries in the region. For example, in 1980 kWh/capita was Egypt 493, Ethiopia 17, Jordan 330, Morocco 260, Somalia 28, Sudan 67, Syria 426 and Tunisia 440. Table 1 Growth in Electriciity Demand, 1970-81 (Gwh) Annual Rate of Growth (Z) 1970 1975 1981 1970-1975 1976-1981 YGEC Energy sold 11.2 33.7 161.2 24.6 29.8 Energy generated 14.2 43.3 228.0 25.0 31.9 Private suppliers and Autoproducers I/ Energy generated Mean 10 60 440 43 39 High 20 70 595 29 43 Low 5 50 285 59 34 National Demand 2/ Total generated 24 100 670 33 37 YGEC share of national generation (x) 59 43 34 Electricity Intensity (kWh/$'000 of GDP, 1975/76 prices) YGEC 23 42 156 12.8 24.4 Total 39 97 459 20 30 GDP Elasticity YGEC 2.2 4.8 Total 2.8 5.5 Source; YGEC, mission estimates 1/ Mean, high and low estimates for private sector are based on alternative interpretations of NRECA survey results. The high estimates assume all villages in YAR are similar to those surveyed by NREC, which had already been selected as candidates for connection to the YGEC system. Low estimates were based on the least electrified areas surveyed being typical of the whole country. 2/ National demand is based on the mean estimates for the private sector. - 57 - ANNEX 2 Page 4 of 17 Electricity Supplied by YGEC 6. The growth of electricity demand on the principal YGEC systems is shown in Attachment 1 and summarized in 'Table 2 below. Table 2 Electricity Demand, YGEC Systems Annual Rate of Growth (X) 1970 1975 1981 1970-1975 1976-1981 No. consumers (end year) 22,500 2/ 36,200 116,000 13.0 25.2 Electricity sold (GWh) 11.2 33.7 161.3 25.2 32.7 Electricity generated (GWh) 14.2 43.3 228.0 25.4 34.6 Maximum demand (MW)3' 6.54 10.6 54.3 17.2 33.9 Losses (%) 21.0 :22.2 29.3 Load factor (x) 40.7 4/ 46.6 47.9 Source: Annex 2, Attachment 2 1/ Growth rates estimated by method of least squares 2/ Data for 1971 3/ Sum of maximum demands of individual systems, excluding diversity. 4/ Data for 1972. Total electricity sold by YGEC grew at 25.2% p.a. during 1970-1975, increasing to 32.7% p.a. during 1976-81. Most of this growth was attributable to the connection of new consumers. Total consumers grew by 13.0% p.a. during 1970-75 and the growth rate almost doubled to 25.2% p.a. during 1976-81. This growth occurred mostly after 1977 when the Stage II electrification program got underway. In 1981 alone 25,700 consumers were connected, nearly half of which (11,300) were connected with the assistance of contractors outside the three main cities. Within the main cities YGEC connected an average 140 consumers per week in Sana'a, 58 in Hodeidah and 83 in Taiz. The rate of growth in the average consumption per consumer fell from around 11% p.a. during 1970-75 to 6% p.a. during 1976-81. This was probably caused by the average size of new consumers falling as lower income households were connected and the distribution system being extended to rural areas. 4. Electricity generated grew faster than energy sold especially during the second half of the decade. This reflects the increase in distribution losses and unaccounted for electricity. Maximum demand, however, grew slower than energy generated. This implied rises in load factor was mainly due to shortages in capacity leading to load shedding, especially in the mid 1970's. However, the increase in load factor was also influenced by a few large commercial and government consumers taking supply. -8 - ANNEX 2 Page 5 of 17 Electricity Consumption By Sector 8. YGEC classifies its consumers as Private (houses and shops), Industry, Government (including street lighting) and Diplomatic. Electricity sold to each of these categories is shown in Table 3. Table 3 Electricity Sales by Sector, YGEC, 1975-81 (MWh) Annual Growth Rate (Z p.a.) Sector 1975 1976 1977 1978 1979 1980 1981 1975-1981 Private 22,100 26,800 34,467 46,707 69,362 97,682 116,658 34.4 Industry 3,900 4,800 6,162 8,824 13,467 13,137 11,623 24.2 Government 3,400 6,800 8,664 11,685 13,419 18,708 25,168 28.8 Diplomatic 2,320 3,130 4,030 5,440 4,260 6,212 7,777 19.8 Total 33,720 41,530 53,323 72,653 100,508 135,739 161,226 31.6 Source: YGEC Note: Annual growth rate calculated by method of least squares. Table 3 shows that households and shops were the fastest growing sector, increasing at 34.4% p.a. during 1975-81, followed by Government (38.8% p.a.). Sales to industry grew at 24% p.a. and its share declined from 11.6% of total sales in 1975 to 7.2% in 1981 as shown in Table 4. Part of this fall in share is accounted for by the absolute decline in electricity sold to industry in 1981. This was probably because of consumer reclassification and illustrates the difficulty of interpreting some of YGEC's commercial data. Table 4 Share of Electricity Sold by Sector, YGEC (') Sector 1975 1981 Private 65.6 72.4 Industry 11.6 7.2 Government 16.0 15.6 Diplomatic 6.9 4.8 Total 100.0 100.0 Source: Table 3. - 59 - ANNEX 2 Page 6 of 17 9. The proportion of electricity sold to industry is small compared to most countries. There are three reasons for this. First, until recently public electricity supply has been consistently unreliable. Even now there remains a ban on the consumption of electricity by industry during the peak. Second, electricity tariffs for industry have been the same as, for other con- sumers and are above the cost of private generation (see Annex 4, para. 24). Consequently, private electricity generation has been more attractive financially, especially when diesel fuel was heavily subsidized. Third, manufacturing accounts for only 7% of GDP and most industries in YAR are not electricity intensive. 10. Rough estimates by the mission based on NRECA's surveys and K&D's fieldwork indicate that the composition of the private electricity suppliers in 1981 was as shown in Table 5. This breakdown shows tht industry probably accounts for a high proportion of the generation of private suppliers and autoproducers, although households could equally be the major consumers or autoproducers under alternative assumptions. Table 5 Private Suppliers' and Autoproducers' Consumption of Electricity by Sector, 1981 Electricity Generated (GWh) Households 50-295 Shops 30-50 Workshops 45-90 Large autoproducers 160 285-595 Source: Mission estimates Regional Consumption of Electricity 11. Consumption of electricity supplied by YGEC by region is shown in detail in Attachment 1, and summarized in Table 6. Sana'a region consistently accounted for about half YGEC's energy sold during 1970-1981. After 1975, consumption grew at 31.2% p.a. in Sana'a, but growth was slower in Hodeidah (23.7% p.a.) and Taiz (26.8% p.a.), partly because of greater capacity shortages outside of Sana'a. Until 1976 public supply was virtually limited to the three main cities. However, during the late 1970's public supply was extended to other regions, so that in 1981 the three main cities had only 73% of all YGEC consumers and 86% of consumption. This is shown graphically in Attachment 2. - .60o - ANNEX 2 Page 7 of 17 Table 6 Proportion of Electricity Sold By Region, YGEC (% of total YGEC sales) Region 1970 1973 1975 1978 1981 Sana'a 55.7 46.5 53.7 56.1 52.5 Hodeidah 20.1 34.0 30.8 24.3 20.9 Taiz 24.2 19.5 15.5 12.8 12.6 Dhamar - - - 3.0 2.8 Hajjah - - - 1.1 1.3 Ibb - - - 1.6 3.1 Branches - - - 1.1 6.8 Total 100.0 100.0 100.0 100.0 100.0 Source: Attachment 1. Electricity Supply to Households 12. According to the 1975 census :37,700 houses had electric lighting. Although precise figures do not exist, YGEC probably supplied about 25,000 of these. Therefore, of the total 904,100 houses determined in the census, 4% had access to public or private supply, or generated their own electricity. The mission has estimated that YGEC supplied about 73,000 houses in 1981 and a further 155,000 to 435,000 had supply had from other sources. Taken with the 1981 CYDA census figure for the number of families of 1,097,800, these figures imply that between 20% to 45% of households had access to electricity supply. Taking the midpoint of this range (33%) implies that almost 30% of households would have gained access to supply in 6 years, which would be a remarkable achievement. 13. Households use electricity pri.marily for lighting, television and small appliances, although air conditioning fans are common iin Hodeidah and, to a lesser extent, in'Taiz. Because of data deficiencies the average household consumption of electricity isi not known precisely, but is likely to be about 1300 kWh p.a. in Sana'a, 2000 KWh p.a. in Hodeidah and 850 kWh in Taiz. A sample of 52 from a total 391 household billing records in the YGEC rural branch of At Turbah showed that the mean annual consumption was 604 kWh. Load Characteristics 14. The underlying seasonal variation in demand is masked by growth during the year and the influence of past supply shortages. However, it is apparent that demand in Sana'a peaks during the winter in December, whereas demand peaks in September in Hodeidah and usually in July in Taiz. The summer peaks in Hodeidah and Taiz are almost certainly due to air conditioning. - 61 - ANNEX 2 Page 8 of 17 15. Daily maximum demand occurs around 1900h in each of the main YGEC load centers. Minimum demand is about 60% of the maximum and occurs at around 0300. Demand does not increase much until the evening. During Ramadan the peak demand is about one hour later and demand remains high throughout the night, so that minimum demand occurs about 0900. C. Projected Demand for Electricity Economic Development 16. GDP grew at an average annual rate of 7.6% p.a. during the decade 1971/72 to 1981. Economic growth was more rapid during the first half of the decade. As Table 7 shows, GDP grew at 9.4% p.a. during 1971/72 to 1976/77 and 5.6% p.a. thereafter. Recently, there has been a marked slowdown in economic growth. In 1980 GDP grew by 3.8% and this declined to about 2.5% in 1981. The 1975/76 to 1980/81 First Five-Year Plan forecast GDP to grow by 8.2% p.a. A growth rate of 5.9% p.a. was achieved. Table 7 Growth in GDP by Sector (YRls million, constant 1975/76) Annual Rate of Growth Fiscal Year (% p.a.) Sector 1971 1976 1981 1971-1976 1976-1981 Agriculture, forestry/fishing 1550 1850 2177 3.6 3.7 Mining/Quarrying 27 42 75 9.2 13.8 Manufacturing 159 274 460 11.5 12.2 Electricity and Water 8 17 46 16.3 24.8 Construction 170 373 481 17.0 5.8 Trade 475 944 1032 14.7 2.0 Restaurants and Hotels 44 68 84 9.1 4.8 Transport and Communication 116 160 220 6.6 7.3 Finance 47 227 452 37.0 16.5 Real estate and Business services 170 193 272 2.6 7.9 Other services 28 44 64 9.5 8.7 Government services 340 451 885 5.8 16.2 Private non-profit services 10 17 24 11.2 8.0 Import duties 222 737 785 27.1 1.4 Bank services charge -51 -211 -422 32.8 16.7 Total GDP (market prices) 3315 5186 6635 9.4 5.6 Source: CPO - 62 - ANNEX 2 Page 9 of 17 17. Foreign savings have been the! main influence on the rate of economic growth. These comprise remittances from Yemenis working abroad and net private and official transfers. Remittances in 1981 amounted to slightly less than US$1 billion. Official transfers and development loans; were about US$0.5 billion. Investment during the 1975/76 to 1980/81 Five-Year Plan was impres- sive. The ratio of gross fixed capital formation to GNP avetraged around one third, more than twice its level during the preceding Three Year Development Program (1974-1976). 18. The decelleration in growth since around 1978 has been mainly due to remittances and private tranfers declining in nominal as well as real terms, as shown below: Year Private Transfers (Receipts) (YR million, current prices) 1977/78 6351 1978/79 5595 1979/80 6118 1980 6034 1981 4444 1982 Ql 1174 Furthermore, some development assistance has been delayed as a consequence of the effect of falling oil prices, the global recession and some donor countries being faced with emergency demands from their recipients. The balance of payments, which registered considerable surpltuses before 1978/79, showed an estimated $330 million deficit in 1981 (11% of GDP). The budgetary position has also deteriorated rapidly, despite an increase in revenues because of increases in capital and defense spending. The deficit amounted to about US$500 million in 1981, or 17% of GDP. 19. The Second Five-Year Plan (1982-1986) forecasts GDP to grow at 7.0% p.a. Of the total capital formation during the plan period, 87% is projected to come from net current tranfers and factor income from abroad, i.e. mainly from workers' remittances. This is projected to increase slightly in real terms. Given the decline in private transfers in recent years and the changing demand for Yemeni labor in the oil exporting countries, the Second Five-Year Plan GDP growth target of 7.0% p.a. looks optimistic. GDP growth during the Plan is unlikely to exceed 5.5% p.a. and could even fall as low as 4% p.a. Forecasts of YGEC Electricity Demand 20. The most recent demand forecasts for YGEC were prepared in 1982 by the Corporation, assisted by K & D. The YGEC forecast for rural load centers was based on earlier projections prepared by NRECA during their rural electrification study. NRECA based their forecast on information obtained from village surveys. YGEC based its March 1982 forecast on the consultants' work, but revised this forecast in October 1982 to take account of more recent information, particularly concerning the implementation of projects by YGEC - 63 - ANNEX 2 Page 10 of 17 and its potential consumers. A description of the consultants' load forecast- ing methodology is given in Attachment 12. The mission considered that the earlier (March 1982) forecast was optimistic. The revised (October 1982) forecast appeared plausible but lacked detail and only projected demand for five years up to 1986. It was therefore necessary for the mission to revise the YGEC forecast. The three forecasts are summarized in Table 8. The mission's forecast has been used in this report as the most likely scenario of future demand supplied by YGEC. There is reasonable agreement between the revised YGEC and mission forecasts. Differences arise mainly from the timing of large industrial projects. However, there are considerable differences between these forecasts and the earlier forecast, the reasons for which are discussed later (para. 22). Table 8 Comparison of Forecasts of Energy Sold All YGEC Systems (-GW-h) YGEC Forecast Year March 1982 October 1982 Mission Forecast 1981 (actual) 161 161 161 1982 203 202 201 1983 278 250 249 1984 537 454 407 1985 676 555 556 1986 853 691 617 1987 1025 719 1988 1278 845 1989 1450 930 1990 1629 1030 1991 1807 1130 1992 2010 1227 Growth rate (% p.a.) 1981-86 39.6 33.8 30.8 1986-92 15.4 12.1 1/ March 1982 YGEC forecast excludes enesrgy sold by rural branches before they are connected to main system. Figures for 1982 in the March 1982 YGEC forecast are for Sana'a, Hodeidah, Taiz, Ibb and Dhamar. Problems in Forecasting YGEC Demand 21. There are a number of major problems involved in electricity demand forecasting for YGEC: - 64 - ANNEX 2 Page 11 of 17 a) demand has been constrained by supply deficiencies in the past. This has resulted in periods of sluggish growth, followed by very rapid growth in recent years when supply constraints have been overcome. Past trends, therefore, are not a reliable indicator of what will happen in the future; b) much of the past growth in electricity demand has been accounted for by connecting households in t'he main urban centers. The number of households with public supply in these areas appears to be approach- ing saturation, particularly in Sana'a. Future growth in new connec- tions will therefore have to come primarily from rural areas. There the potential demand for connecting to cheaper, more reliable YGEC supply is high, but consumption per consumer is low; c) demand is nevertheless likely to remain supply-constrained in the foreseeable future. This will result from consumers willing to take supply at the prevailing YGEC prices not having access to public supply because of the manpower, financial and institutional limita- tions which the Corporation faces. Demand forecasting, therefore, cannot be undertaken in isolation from projections of supply, i.e. the forecaster has to project what distribution and rural electrifi- cation program is feasible with regard to YGEC's availability of finance, skilled manpower and ability to implement projects; d) YGEC demand is only part of an electricity market for which little accurate information exists. This lack of information makes it particularly difficult to project electricity demand in relation to economic development; e) the base data from which demand is projected is generally unreliable. Considerable improvements need to be made in YGEC's data by region on consumers, consumption and elsectricity production (para. 2); f) forecasts of YGEC demand are essentially population driven, yet the base population in areas with access to supply has not been estab- lished satisfactorally. Furthermore, projecting urban population growth rates is complicated by uncertain rural-urban and inter- national migration; g) economic development in YAR is particularly sensitive to external circumstances, viz. remittances from Yemenis working abroad and foreign assistance (para. 18). This uncertainty in turn influences the ability of consumers to purchase electrical app'Liances and the implementation of electricity consuming projects; and h) overall demand is sensitive to the magnitude, timing and connection of particular projects, particularly the cement workcs at Bajil, Amran and Al Barh (Mufraq). These problems lead to considerable uncertainty in forecasts of YGEC demand. As is shown in the following paragraphs, the likelihood is greater that actual demand will be below the March 1982 YGEC forecast, rather than above it. - 65 - ANNEX 2 Page 12 of 17 Projected Demand for Electricity 22. Analysis of the most recent information available to the mission, some of which was not available to the consultants, suggests that the March 1982 YGEC forecast is probably too high. Major reasons for this assessment are: a) greater delays in regional interconnection and the connection of new consumers is now likely; b) the ability of YGEC to expand the distribution system will be severely constrained by shortges of finance and skilled manpower, as well as institutional problems. These constraints will persist during the 1980's; c) revisions to population and household data, especially the larger number of persons per household registered in the 1981 census, compared to the 1975 census; d) revised population and census data suggest that the proportion of households presently with supply in Sana'a is 72%, Hodeidah 54% and Tiaz 73%. The scope for connecting households without supply is therefore limited, except in Hodeidah. However, there is more impermanent housing, which may not take supply, in Hodeidah than the other cities; e) the consultants projections of urban population growth rates are generally near the high projections of the Bank urban sector report (Report No. 2799-YAR). Central population estimates would be more suitable for electricity demand forecasting; f) recent data for average consumption per consumer have been biased upwards by a few large consumers, e.g. a hotel being recently connected. Extrapolating these trends could lead to demand being overestimated. Actual consumptions of residential and small consumers are low (para. 13) and are likely to remain so given the high price of electricity in YAR; g) the YGEC March 1982 forecast was based on the draft 1982-86 Five Year Plan GDP target of 7% p.a. This now looks to be at the upper range of likely outcomes (para. 19). 23. The mission had to produce its own revision to the forecast given these reservations about the detailed (March 1982) forecast provided by YGEC and the limited detail and horizon of its revision (October 1982). This is summarized in Attachment 3. The forecast was based on a similar methodology to that used by the consultants. Detailed forecasts for the main urban load centers are shown in Attachment 4 to 7. Attachments 8 and 9 show projections for the rural load centres before and after interconnection. These were based on the YGEC forecasts, but the timing of. interconnection has been delayed to allow for the most likely program. Similar adjustments were made for the timing of the connection of the loads of large industry and of the inter- connection of Sana'a, Hodeidah and Taiz. Projections for bulk (MV) consumers - 66 - ANNEX 2 Page 13 of 17 are shown in Attachment 10 and 11. Assumption's used to make the forecast are shown in Attachment 11. Domestic consumer forecasts were based on population projections. Domestic specific consumption growth rates were estimated using a more likely projection of GDP per capita. The total consumption for other consumer groups was projected in relation to the likely growth in GDP of the corresponding economic sector. 24. Table 9 summarizes the demand forecast for YGEC. Total electricity sold by YGEC is forecast to grow at an average annual rate of 37.0% during 1981-85 then fall to 12.1 during 1985-90. This fall in growith can be explained by the changing sectoral distribution of demand shown in Table 9. Table 9 Projected Electricity Sold by Sector, YGEC (GWh) Annual Growth Actual Forecast % Share Rate (X) Sector 1981 1987 1992 1981 1987 1992 1981-87 1985-92 Domestic 76.3 214.0 356.7 47.3 29.8 29.1L 18.8 10.8 Commercial 18.1 25.9 34.8 11.2 3.6 2.8 6.2 6.1 Government 19.1 30.4 44.6 11.8 4.2 3.6 8.1 8.0 Street Lighting 1.8 3.4 5.7 1.1 0.5 0.5 11.2 10.9 Diplomatic 7.8 9.3 10.8 4.8 1.3 0.9 3.0 3.0 Small Industry 9.5 13.6 18.2 5.9 1.9 1.'5 6.2 6.0 Bulk 6.2 162.1 263.0 3.8 22.5 21.4 72.3 10.2 Cement - 185.0 260.0 - 25.7 21.2 - 7.0 Rural 22.5 75.3 233.6 13.9 10.5 19.( 22.3 25.4 Total 161.3 719.0 1227.4 100.0 100.0 100.() 28.8 11.3 Sourcet Attachment 3 Much of the growth in electricity consumption up to 1985 is accounted for by the expansion of YGEC supply to industry and the connection of the cement works at Bajil and Amran. Such extension of public supply to industry is dependant upon restructuring the tariff for these consumers (see Annex 4, para. 34). The share of the industrial categories in total YGEC consumption is projected to increase from 10% in 1981, to 50% in 1987 and then decline gradually to 44% in 1990. Their consumption is forecast to grow at 38% p.a. during 1981-92, although this growth slows down to 8.5% p.a. during the second half of the forecast period, 1987-92. 25. Domestic consumption in Table 9 refers to households in the three main cities of Sana'a, Hodeidah and Taiz. Most of the growth, from these consumers is accounted for by new connections. Connections are forecast to increase at 16.5% p.a. during 1981-85 and 6.61% p.a. during 1985-92. Popula- tion in the three cities is projected to grow at 6.8% p.a. during 1981-85 and 6.5% p.a. during 1985-92. Consequently, the percentage of households con- nected rises from 68% in 1981 to 95.0% in 1985 and remains mcre or less - 67 - ANNEX 2 Page 14 of 17 constant thereafter. An average of 270 connections per week will need to be made during 1983, which should be feasibLe. 26. Rural consumption is expected to grow at 22.3% p.a. during 1981-87 and 25.4% p.a. during 1985-92. The average consumption per rural consumer is projected to increase by only 3.1 p.a. The growth in rural consumption is mainly due to the connection of househol,ds without YGEC supply. In the medium term rural electrification will be constrained because of shortages of finance and skilled manpower, as well as limitations to the rate at which YGEC can implement projects. Details of the rural demand forecasts are shown in Attachments 8 and 9. Most of the growth up to 1986 occurs in areas where YGEC already supplies some consumers. By 1992, 55% of the rural consumption is in projects that have not yet been formally identified. 27. Distribution losses, station use, and unaccounted for energy in Sana'a, Hodeidah and Taiz are forecast to fall from their current level of around 30% of gross generation to 22% in 1992. This is predicated upon effective measures to reduce losses being taken (see Annex 3, para. 15). Some reduction in average losses will arise from the rapid growth in sales at MV to bulk consumers, but this will be offset to some extent by the greater station own use of steam power stations. Total energy generated is projected to grow at 26.3% p.a. during 1981-87 and 10.9% p.a. during 1987-92. Table 10 shows generation losses and maximum demand. Table 10 Generation and Maximum Demand, YGEC Actual Forecast Annual Rate of Growth (%) 1981 1987 1992 1981-87 1985-92 Isolated Systems Sales 161.3 19.5 20.0 -29.7 0.5 Gross generation (GWh) 228.0 25.3 26.0 -35.6 0.5 Maximum demand (MW) 53.9 6.7 7.4 -29.4 2.0 Losses (X) 29.3 22.9 23.1 Load factor (X) 48.3 43.1 40.1 Interconnected Systems Sales (GWh) - 699.6 1,207.4 - 11.5 Gross generation (GWh) - 900.1 1,526.4 - 11.1 Maximum demand (MW) 2/ - 172.8 289.7 - 10.9 Losses (X) - 22.3 20.9 - Load factor (X) - 59.5 60.1 - Total YGEC Sales (GWh) 161.3 719.1 1,227.4 28.3 11.3 Gross generation (GWh) 228.0 925.4 1,552.4 26.3 10.9 Maximum demand (MW) 1/ 53.9 198.7 329.3 24.3 10.6 Losses (X) 29.3 22.3 20.9 Load factor (X) 48.3 53.2 53.8 Notes: 1/ Sum of maximum demands of each load centre. 2/ Coincident maximum demands of interconnected load centres. - 658 - ANNEX 2 Page 15 of 17 28. The sum of the maximum demands at all YGEC load centres is forecast to increase from 54 MW in 1981 to 199 MW in 1987 and 329 MW iin 1992. Sana'a and Hodeidah will be partially intercornected with Ras Katenib in 1983. Taiz, Ibb and Dhamar will be connected in 1984. Allowing for diversity, the maximum demand on the interconnected system is forecast to be 133 MW in 1985, increas- ing to 290 MW in 1992. The cement works are projected to have a coincident demand of 24 MW or 18% of the total peak demand in 1985 and account for 24% of energy generated on the interconnected system. Other bulk consumers are fore- cast to contribute a further 20% to peak demand. Therefore, nearly 40% of projected maximum demand in 1985 is due to cement works and other bulk con- sumers. Any change in their plans will. consequently have a major influence on system demand. 29. It is extremely difficult to forecast the demand for electricity by autoproducers and other private producers because of the lack of information about this part of the subsector. However, if it is assumed that all the large autoproducers take supply from YGEC, there are few village consumers able to pay for private supplies who are not already supplied and that YGEC will gradually take over private systems, non-YGEC generation would fall to about 220 GWh by 1987 and decline slowly thereafter. The assumed growth in national electricity consumption is shown in Table 11. National power genera- tion would therefore increase at about 9.4% p.a. during 1981-87, slowing down to 8.9% p.a. during 1987-92. Table 11 shows that generation per capita would rise from 31 kWh/capita for YGEC in 1981 to 165 kWh/capita in 1982, increasing sharply by 1987 because of new cement and other bulk consumers. These new consumers are expected to lead to a high elasticity with respect to GDP for YGEC sales. However, the national elasticity is expected to decline to 1.7 during 1981-87 and 1.6 during 1987-92. This is indicative of autoproducers taking supply from YGEC and the assumed tendency towards saturation in electricity supply in the villages not supplied by YGEC. Table 11 Forecast National Electricity Consumption Actual Projected 1970 1975 1981 1987 1992 Electricity Generated (GWh) YGEC 14 43 228 925 1552 Autoproducers & private 10 60 440 220 200 Total YAR 24 103 668 1145 1752 Population ('000) 5258 6075 7264 8375 9429 GDP (US$million, 1975/76 prices) 617 1032 1460 2013 2631 Generation per capita (kWh) YGEC 3 7 31 110 165 Total 5 17 92 137 186 Intensity (kWh/$'O00 of GDP) YGEC 23 42 156 459 590 Total 39 97 459 569 666 GDP Elasticity YGEC 2.2 4.8 4.4 1.9 Total 2.8 5.5 1.7 1.6 - 69 - ANNEX 2 Page 16 of 17 Future Improvements to Demand Forecasting& 30. YGEC's demand forecasts for 1977-81, which were made in 1977, are compared to the actual outcomes in Attachments 12 and 13. Forecast total sales for the three main cities were substantially above the actuals for all three cities for 1979-81 by between 46% to 87% during the three years. However, the forecast errors for 1979-81 for Sana'a were much lower (5% to 10%), than for Hodeidah (97% to 135%), or Taiz (90% to 113%). System losses were underestimated so that the error for total generation was less, ranging from -12.9% in 1978 to +23% in 1981. Load factors were also underestimated so that forecast errors for maximum demand were lower and in the order of +15%. Attachment 13 shows a breakdown of forecast and actual sales in 1981 by consumer category. The forecasts for sailes to households and commerce (Private), Government and Diplomatic tended to be less than the actuals for each of the three cities by as much as 38%. Overall, total sales to these consumers in the three cities were underestimated by 20.3%. However, the forecast errors for the bulk consumers asnd large industry were very large. Forecasts for these consumers were more that 10 times greater than the actual sales to these consumers. Some of this apparent error is almost certainly due to bulk consumers being included in the actual data for sales to Private and Government tariff categories. Revising these data would alter the apparent under-forecasting of demand for Private and Government consumers. This highlights the need for collection of consumer sales data on a systematic basis (para. 2). 31. Nevertheless, overall there were large errors in demand forecasting which suggest that there is considerable scope for improving the methodology. More accurate load forecasting should enable improved investment decisions to be taken. For this reason it is recommended that measures to improve load forecasting be linked to preparation of the next major generation project. The immediate requirement is for YGEC to improve its database along the lines recommended in para. 2. 32. Differences between actual and projected demands of bulk and industrial consumers appear to have been the major source of past forecasting errors. It is essential to collect more information on the present industrial electricity market. Because autoproducers and the private sector dominate this market, it will be necessary to carry out detailed surveys of these enterprises, either as a special study or as part of a wider industrial survey. Projections of the overall industrial market should be based on the most likely scenario of economic growth,, preferably broken down by industrial subsector. Plans of individual enterprises should be adjusted to allow for likely constraints on their implementation. A realistic view on the likelihood of enterprises taking supply from YGEC should be taken and the electricity prices needed to achieve this spelled out. In this respect, allowance should also be made for private generating plant to be used for peak lopping or cogeneration; especially large installations such as the cement works (see Annex 3, para. 28). - 70 - ANNEX 2 Page 17 of 17 33. Although the forecasting performance for Private and Government consumers appears better, errors are still high and the picture may be obscured by changes to consumer classification. Nevertheless, it does not necessarily follow that methods which worked reasonably in the past will continue to be satisfactory in the future. Again, there is an immediate need to improve the base data by a more detailed breakdown of consumer data. The population in areas to be served has been ill-defined geographically. Reliable information is lacking on the relationships between population, households, houses and electricity consumers both at present and in the future. Little is known about appliance ownership and use in urban areas. A survey of urban electricity consumers, similar to that carried out by NRECA, should ideally be carried out before the next major forecast revision, but this is not a high priority at present. This should also collect socio-economic data which could be related to appliance ownership. Furthermore, YGEC billing data should be used to examine changes in specific consumption for cohorts of consumers. 34. As the focus of system expansion will move to rural areas further information on actual experience after electrification is required. It is recommended that YGEC carry out period reviews of villages that have been electrified to determine how closely actual population, numbers of consumers and demand compared to forecasts. This will provide a valuable guide to the outcome of future projects. YGEC is required to collect data on the implementation of the village electrification projects financed with assistance from IDA and it is recommended that this reporting system by extended to all villages. Improvements in defining the population in points to be served should be made by detailed analysis of census maps and data, possibly supplemented by interpretation of aerial photographs that are taken periodically for mapping and surveying purposes. - 71 ANNEX 2 Attachment 1 YEMEN ARAS RJIPUBLIC POWER SUBSECTOR REVIEW Electricity Sold and Consulmers, YGEC, 1970-81 Growth Rate (7 9.5.1 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1970-75 lq78-8l Electricity Sold (MWh) bas a 6240 7235 7966 10594 13827 18067 23365 29732 40681 55308 74601 84610 24.1 31.2 Hodeidan 2246 3392 5619 7727 9239 10381 11383 14644 17629 20656 27103 33763 36.8 21.7 'raiz 2713 3171 3905 4441 4820 5219 6730 7730 9240 13193 17327 20392 131. 26.8 Dhamar - - - - - - - 1000 2200 3300 4100 4550 44.1 Hajjan - - 130 820 740 1500 2031 8b.1 lbb - - - - - - - - 1168 2675 3799 4923 59.9 drancnes 163 825 4636 7337 10957 18R Total 11199 13898 17490 22762 27886 33667 41478 53399 72,563 100,508 135,766 161,226 29.2 12.7 Number of Consumers (end year '000) Sana's 11.4 12.4 14.2 16;.6 18.8 20.3 21.7 23.9 28.1 33.4 41.2 48.3 13.0 18.0 HoBiedah 4.3 5.2 6.5 7.2 7.4 7.7 7.9 8.5 9.9 12.3 15.3 15.2 1I.1 Taiz 4.6 5.8 6.5 7.7 7.9 8.5 9.2 9.6 10.3 13.4 17.3 21.6 12., 1q.7 Dhamar - - - - - - - 2.2 3.0 3.8 4.3 5.3 23.6 Hajjah - - - - - - - 0.8 1.0 1.2 1.3 I.8 22.0 Ibb - - - - - - - - 2.2 3.1 4.12 6.8 44.1 Branches - - - - - 0.44 2.76 8.23 9.91 16.94 13.8 Total 22.5 25.9 30.8 33.9 36.2 38.6 44.8 55.9 73.0 90.4 116.1 13.0 25.2 EneryE Generaed M dana'a 8324 9630 10755 14385 18482 24203 29649 38503 54440 77144 100,563 119,700 24.2 33.8 hodeiaah 2660 4044 6991 9426 10478 11780 15180 29648 24726 29319 38904 47,700 35.4 29.5 TalZ 3199 4158 5467 6275 6747 7315 10157 11058 12091 19978 25557 30,000 17.8 27.2 Dhamar - - - - - - - 1200 2700 4100 5363 6200 48.7 Hajjah - - - - - - - 200 1100 1400 2200 2400 76.7 Ibb - - - - - - - - 1600 3300 4700 6000 54.0 branones - - - - - - - 200 1100 6300 10700 16000 202 Total 14183 17832 23213 30086 35707 43298 54986 70809 97757 141,541 187,987 228,000 29.4 34.6 Maximum Demand (MW generated) =ansa 2.10 2.60 3.00 4.55 5.70 5.95 7.75 10.12 14.53 18.85 23.75 28.30 25.6 30.4 HNdeidah 1.20 1.70 2.68 2.66 2.65 2.66 3.68 4.58 5.70 8.00 10.10 22.5 30.1 Taiz 1.80 1.80 1.80 2.00 2.38 2.89 3.47 5.18 6.45 7.45 3.2 27.5 lihamar -- - - - - 0.76 0.90 1.30 1.30 1.80 23.3 Hajjan - - - - - - - 0.35 0.35 0.35 0.40 0.50 R.R lbb - - - - - - - - 0.90 1.10 1.20 1.67 21.4 :Iranches - - - - - - - 0.114 0.717 2.21 2.68 4.50 138 TOtal 6.59 9.03 10.16 10.60 12.79 17.91 25.45 34.69 43.78 54.32 17.2 13.3 Losses (5) Sana'a 25.0 24.9 25.9 26.4 25.2 25.4 21.2 22.8 25.3 28.3 25.8 29.3 Hodeidah 15.6 16.1 19.6 18.0 11.8 11.9 25.0 25.5 28.7 29.6 30.3 29.2 'faiz 15.2 21.3 28.4 29.2 28.6 28,7 33.7 30.1 23.6 34.0 32.2 32.0 Dhamar - - - - - - - 16.7 18.5 19.5 23.6 26.6 Hajjah - - - - - - - 35.0 25.5 47.1 31.8 15.4 Ibb - - - - - - - - 27.0 18.9 19.2 18.0 Branches - - - - - - - 18.5 25.0 26.4 31.4 31.5 fotal 21.0 22.1 24.7 24.3 21.9 22,2 24.6 24.6 25.8 29.0 27.8 29.3 Load Factor (7) sana'a 45.2 42.3 40.8 36.1 37.0 46.4 43.6 43.4 42.8 46.7 48.2 48.3 Hodeidah 38.5 46.8 40.2 45.0 50.7 65.0 60.9 61.6 58.7 55.4 53.9 Taiz 34.6. 39.8 42.8 41.8 48.6 43.7 39.8 44,.0 45.1 46.0 Deamar - - - - - - - 18.0 34.2 36.0 47.0 39.3 Hajjan - - - - - - - 6.5 35.9 45.7 62.6 94.8 lbo - - - - - - - - 20.3 34.2 44.6 41.0 Brancnes - - - - - - - 20.0 17.5 32.5 45.5 40.6 Total 40.7 38.0 40.1 46.6 48.9 45.1 43.8 46.6 48.9 47.0 Source: YGEC Note: Growth races estimated by method of least squares and are calculated for the years for which data are available within the period shown. 72 - ANNEX 2 Attachment 2 YEMEN ARAB REPUBLIC POWER SUBSECTOR REVIEW PAST GROWTH IN CONSUMERS AND ELECTRICITY 'SOLD. YGEC 200- 180- 160- 140- TOTAL H~ ~ ~ ~ D 1200 Z-1 0x U,~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~l ioe &Z 1230 ! H I I I 1 2~~10 z 80 10t 90 > z 60 ~~~~~~~~~~~~~~80 ' TOTAL -70 40-~~~~~~~~~~~~~~~~~~~~~~~~~~~0 20- 40 30 U 0-1 ~~~~~~~~------- 20 z 1970 1972 1974 1976 1Q78 t980 -73- ANNEX 2 ttachEment 3 YEMEN ARAB REPUBLIC POWER SUBSECTOR REVIEW YGEC Load Forecast for 1982-1992 ACTUAL ------------------------------------------ FORECAST ------------------------------------------- 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 Isolated Systems Sales: Domestic (GWh) 76.3 99.6 58.5 20.1 - - Commercial (GWh) 18.1 19.3 9.7 5.2 - - Government (GWh) 19.1 20.7 8.4 2.6 - - Diplomatic (GWh) 7.8 8.0 - - - - Small industrial (GWh) 9.5 10.1 3.4 2.6 - Rural branches (GWh) 22.5 25.0 27.9 29.7 19.0 11.8 19.5 21.9 13.8 17.1 18.5 20.0 sulk (GWh) 6.2 16.0 13.4 15.3 - - - - - - - - Streetlighting (GWh) 1.8 1.9 0.8 0.6 - - - - - - TOTAL (GWh) 161.3 200.6 122.1 76.1 19.0 11.8 19.5 21.9 13.8 17.1 18.5 20.0 Gross generation (GWh) 228.0 285.2 172.1 106.7 25.8 15.0 25.3 30.6 18.7 22.8 24.3 26.0 Peak demand (MW) /1 53.9 67.2 40.2 27.4 7.0 3.8 6.7 8.3 5.3 6.5 6.9 7.4 Losses (X) 29.3 29.7 29.0 29.0 26.4 21.3 22.9 28.5 26.2 25.0 23.9 23.1 Load factor (C) 48.3 48.4 48.9 44.5 42.1 45.1 43.1 42.1 40.3 40.0 40.2 40.1 Interconnected S stem bales: Domestic (GWh) - - 66.8 130.6 170.6 191.4 214.0 238.3 264.6 294.5 325.8 356.7 Commercial (GWh) - - 10.8 16.5 23.0 24.4 25.9 27.4 29.1 30.8 32.7 34.8 Government (GWb) - - 13.9 21.4 26.0 28.0 30.4 32.7 35.4 38.2 41.3 44.6 Diplomatic (GWh) - - 8.3 8.5 8.8 9.1 9.3 9.6 9.9 10.2 10.5 10.8 Small industrial (GWh) - - 7.3 8.8 12.0 12.8 13.6 14.3 15.2 16.2 17.2 18.2 Rural branches (GWh) - - - 2.0 17.5 30.3 55.9 67.2 100.8 141.7 180.2 211.6 Sulk (GWh) - - 18.0 71.5 126.3 146.3 162.1 180.2 197.3 216.4 238.7 263.0 Streetlighting (GWh) _ _ 1.5 1.9 2.8 3.0 3.4 3.8 4.2 4.6 5.2 5.7 SUBTOTAL (GWh) - - 126.6 261.2 387.0 445.3 514.6 573.5 656.5 752.6 851.6 '947.4 Cement plants - - - 70.0 150.0 160.0 185.0 250.0 260.0 260.0 260.0 260.0 TOTAL (GWh) - - 126.6 331.2 537.0 605.3 699.6 823.5 916.5 1,012.6 1,111.6 1,207.4 Gross generation (GWh) - - 175.8 435.7 698.8 772.1 900.1 1,043.7 1,159.4 1,279.9 1,403.3 1,526.4 Non-coinciaental peak demand (MW) - - 41.0 95.1 147.4 163.0 192.0 216.0 240.6 267.1 294.6 321.Q Coincidental peak demand (MW) /2 - - 36.9 85.6 132.6 146.7 172.8 194.4 216.4 240.4 265.1 289.7 Losses (C) - - 28.0 23.9 23.2 21.6 22.3 21.1 21.0 20.9 20.8 20.9 Load factor (2) - - 48.9 58.1 60.2 60.1 59.5 61.3 61.2 60.8 60.4 60.1 Isolated & Interconnected System Installed capacity (MW) 109.0 109.0 175.0 274.0 274.0 274.2 354.0 434.0 434.0 434.0 434.0 434.0 Available capacity (MW) 78.0 78.0 117.6 166.3 183.0 189.6 265.3 301.2 315.2 319.2 319.2 319.2 Gross generation (GWh) 228.0 285.2 347.9 542.4 724.6 787.1 925.4 1,052.0 1,178.1 1,302.7 1,427.6 1,552.4 Non-coincidental peak demand (MW) 53.9 67.2 81.2 122.5 154.4 166.8 198.7 224.3 245.9 273.6 301.5 329.3 Coincidental peak demand (MW) 2/ 51.2 63.8 73.1 110.3 139.0 150.1 178.8 201.9 221.3 246.2 271.4 296.4 Losses (C) 29.3 29.7 28.5 24.9 23.3 21.6 22.3 22.3 21.0 21.0 20.8 20.9 Load factor k%) 50.8 51.0 54.3 56.1 60.0 59.9 59.1 59.5 60.1 60.4 60.0 59.8 TOTAL SALES (GWh) 161.3 200.6 248.7 407.3 556.0 617.1 719.1 845.4 930.3 1,029.7 1,130.1 1,227.4 1/ Diversity - 0.95 2/ Diversity - 0.90 74 - ANNEX 2 Attachment 4 YEMEN AAB REPUBLIC POWER SUBSECTOR REVIEW Load Forecast Sana'a (1982-1992) 1981 1982 1983 1984 1985 1986 19587 1988 1989 1990 A. Town Total Population ('000) 319.4 342.6 367.4 394.0 422.6 448.0 474.8 503.3 533.5 56S.S Population Under Supply ('000) 231.2 253.9 280.8 310.5 342.2 373.3 4tl6.4 442.3 480.9 523.0 Peak Demand (MW) 28.0 34.8 41.0 53.8 58.8 63.9 70.4 75.9 82.Q 9o.9 Gross Generation (GWh) 119.7 148.9 175.8 231.2 253.6 275.8 304.6 329.2 360.1 395.6 Total Sales (GWh) 84.6 104.2 126.6 171.1 190.2 209.6 231.5 253.5 277.3 304.6 Losses (Y) 29.3 30.0 28.0 26.0 25.0 24.0 24.0 23.0 23.0 23.0 Loan Factor (X) 48.8 48.8 49.0 49.1 49.2 49.3 49.4 49.5 49.6 49.7 b. Domestic Consumers Potential Consumers ('000) 46.3 49.6 53.2 57.1 61.2 64.9 68.8 72.9 77.3 82.0 Actual Consumers ('000) 33.5 39.5 47.5 54.5 58.6 62.3 66.2 70.3 74.7 79.4 Specxfic Consumption (kWh/customer) 1,300.0 1,352.0 1,406.0 1,462.0 1,520.0 1,582.0 1,649.0 1,710.0 1,780.0 1,850.0 Sales (GWh) 43.6 53.4 66.8 79.7 89.1 98.6 108.9 120.2 132.3 146.9 C. Commercial Consumers Actual Consumers 11,700.0 11,934.0 12,173.0 12,416.0 12,644.0 12,918.0 13,176.0 13,440.0 13,708.0 13,983.( Specific Consumption (kWh/customer) 820.0 853.0 887.0 922.0 959.0 998.0 1,038.0 1,079.0 1,122.0 1,167.0 Sales (GWn) 9.6 10.2 10.8 11.4 12.1 12.9 13.7 14.5 15.4 16.3 B. Small Industrial Consumers Actual Consumers 2,290.0 2,335.0 2,381.0 2,429.0 2,478.0 2,527.0 2,578.0 2,629.0 2,682.0 2,736.0 Specific Consumption (kWh/customer) 2,840.0 2,954.0 3,072.0 3,195.0 3,322.0 3,455.0 3,594.0 3,737.0 3,887.0 4,042.0 Sales (GWh) 6.5 6.9 7.3 7.8 8.2 8.7 9.3 9.8 10.4 11.1 E. Government Consumers Actual Consumers 800.0 832.0 865.0 900.0 935.0 973.0 1,012.0 1,053.0 1,095.0 1,139.0 Specific Consumption (kWh/customer) 14,880.0 15,475.0 16,094.0 16,738.0 17,407.0 18,104.0 18,828.0 19,581.0 20,364.0 21,179.0 Sales (GWh) 11.9 12.9 13.9 15.1 16.3 17.6 1.9.1 20.6 22.3 24.1 F. Diplomatic Consumers Actual Consumers 350.0 353.0 357.0 360.0 364.0 367.0 371.0 375.0 378.0 182.0 Specific Consumption (kWh/customer) 22,290.0 22,732.0 23,186.0 23,650.0 24,123.0 24,605.0 25,097.0 25,599.0 26,111.0 26,634.0 Sales (Gwn) 7.8 8.0 8.3 8.5 8.8 9.1 9.3 9.6 9.9 10.2 G. Large Industrial Consumers 1/ Sales (GWh) 4.0 11.5 18.0 47.0 53.9 60.8 69.1 76.5 84.4 93.2 H. Street Lighting .aiea (CUb) 1.2 1.3 1.5 1.6 1.8 1.9 2.1 2.3 2.6 2.8 1/ From Attachment 10. - 75 - ANEX 2 Attachment 5 YEMEN ARAB REPUBLIC POWER SUBSECTOR REVIEW Load Forecast Hodeidah (1982-1992) 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 A. Town Total Population ('000) 126.4 137.5 149.5 162.6 176.8 192.3 209.1 227.4 247.3 269.0 Population Under Supply ('000) 67.8 77.4 87.7 99.2 112.0 126.0 142.7 160.0 179.8 202.2 Peak Demand (MW) 10.0 13.6 19.3 25.5 31.2 35.0 40.9 45,5 49.5 53.9 Gross Generation (GWh) 47.5 64.8 92.4 122.4 150.4 168.8 198.2 220.8 240.9 263.0 Total Sales (GWh) 33.8 46.0 65.6 88.1 108.3 130.0 144.7 163.4 180.7 194.9 Losses (X) 28.9 29.0 29.0 28.0 28.0 27.0 27.0 26.0 25.0 24.0 Load Factor (X) 54.4 54.5 54.7 54.8 55.0 55.1 55.3 55.4 55.6 55.7 S. Domaestic Consumers Potential Consumers ('000) 19.8 21.4 23.4 27.6 30.0 32.7 35.5 38.6 42.0 45.7 Actual Consumers ('000) 10.6 14.6 18.6 22.6 25.4 28.1 30.9 33.6 36.4 39.2 Specific Consumption (kWh/customer) 2,000.0 2,080.0 2,163.0 2,250.0 2,340.0 2,433.0 2,531.0 2,632.0 2,737.0 2,847.0 Sales (GWb) 21.2 30.4 40.2 50.9 59.4 68.4 78.2 88.4 99.6 111.6 C. Coercial Consumers Actual Consumers 4,100.0 4,182.0 4,266.0 4,351.0 4,438.0 4,527.0 4,617.0 4,709.0 4,803.0 4,998.0 Specific Consumption (kWh/customer) 1,034.0 1,075.0 1,118.0 1,163.0 1,210.0 1,258.0 1,308.0 1,361.0 1,415.0 1,472.0 Sales (GWh) 4.2 4.5 4.8 5.1 5.4 5.7 6.0 6.4 6.8 7.2 D. SWml Industrial Consumers Actual Consumers 397.0 405.0 413.0 421.0 430.0 447.0 456.0 465.0 474.0 484.0 Specific Consumption (kWh/customer) 2,170.0 2,257.0 2,347.0 2,441.0 2,539.0 2,640.0 2,746.0 2,856.0 2,970.0 3,089.0 Sales (GWh) 0.9 0.9 1.0 1.0 1.1 1.2 1.2 1.3 1.4 1.5 E. Government Consumers Actual Consumers 220.0 227.0 233.0 240.0 248.0 255.0 263.0 271.0 279.0 287.0 Specific Consumption (kWh/customer) 23,110.0 24,150.0 25,237.0 26,372.0 27,559.0 28,799.0 30,095.0 31,450.0 32,865.0 34,344.0 Sales (GWh) 5.1 5.5 5.9 6.3 6.8 7.3 7.9 8.5 9.2 9.9 F. Large Industrial Consumers 1/ Sales (GWh) 2.2 4.5 13.4 24.5 35.2 47.0 50.9 58.2 63.1 69.0 G. Street Lighting Saies (GWh) 0.2 0.2 0.3 0.3 0.4 0.4 0.5 0.6 0.6 0.7 1/ From Attachment 10. - 76 - ANNEX 2 Attachment 6 YEMEN ARAB REPUBLIC POWER SUBSECTOR REVIEW Load Forecast Taiz (1982-1992) 1981 1982 1983 1984 1985 1986 _1987 1988 1989 1990 A. Town Total Population ('GOD) 136.4 143.9 151.8 160.2 169.0 178.7 189.0 199.9 211.4 223.5 Population Under Supply ('000) 99.2 131.0 145.8 153.9 162.8 172.4 182.8 193.9 205.7 217.6 Peak Deanad (MW) 7.4 9.3 10.4 16.5 24.8 25.9 27.8 29.6 31.8 34.0 Gross Generation (GWh) 30.0 37.4 42.1 67.2 101.4 106.2 114.6 122.5 132.0 141.9 Total Sales (GWh) 20.4 25.4 28.6 46.4 71.0 75.4 82.5 89.4 97.7 106.4 Losses (S) 32.0 32.0 32.0 31.0 30.0 29.0 28.0 27.0 26.0 25.0 Load Factor (1) 46.0 46.0 46.2 46.4 46.6 46.8 47.3 47.2 47.4 47.6 .B. Domestic Consumers Pocential Consumers ('000) 18.4 19.4 20.5 21.6 22.8 24.1 25.5 27.0 28.6 30.2 Actual Consumers ('000) 13.4 17.7 19.7 20.8 22.0 23.3 24.7 26.2 27.8 29.4 Specific Consumption (kWh/customer) 860.0 894.0 930.0 967.0 1,006.0 1,046.0 1,088.0 1,132.0 1,177.0 1,224.0 Sales (GWh) 11.5 15.8 18.3 20.1 22.1 24.4 26.9 29.7 32.7 36.0 C. Commercial Consumers Actual Consumers 7,200.0 7,344.0 7,491.0 7,641.0 7,794.0 7,949.0 8,108.0 8,271.0 8,436.0 8,605.0 Specific Consumption (kWh/customer) 600.0 624.0 649.0 675.0 702.0 730.0 759.0 790.0 821.0 854.0 Sales (GWh) 4.3 4.6 4.9 5.2 5.5 5.8 6.2 6.5 6.9 7.3 U. Small Industrial Consumers Actual Consumers 794.0 810.0 826.0 843.0 859.0 877.0 894.0 912.0 930.0 949.0 Specific Consumption (kWh/customer) 2,700.0 2,808.0 2,920.0 3,037.0 3,159.0 3,285.0 3,416.0 3,553.0 3,695.0 3,843.0 Sales (GWh) 2.1 2.3 2.4 2.6 2.7 2.9 3.1 3.2 3.4 3.6 E. Covernment Consumers Actual Consumers 180.0 187.0 195.0 202.0 211.0 219.0 228.0 237.0 246.0 256.0 Specific Consumption (kWh/customer) 11,650.0 12,116.0 12,600.0 13,105.0 13,105.0 13,629.0 14,174.0 14,174.0 15,331.0 15,944.0 Sales (GWh) 2.1 2.3 2.5 2.6 2.9 3.1 3.4 3.6 3.9 4.2 F. Large Industrial Consumers 1/ Sales (GOn) - - - 15.3 37.2 38.5 42.1 45.5 49.8 54.2 G. Street Lighting Sales (GWh) 0.4 0.4 0.5 0.6 0.6 0.7 0.8 0.9 1.0 1.1 i/ trom Attachment 10. -77- ANNEX 2 Attachment 7 YEMEN ARAB REPUBLIC POWER SUBSECTOR REVIEW Demand Forecasts for Dhauar, Ibb and Haijah 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 Dhaner Potential Consumers 8,640.0 9,090.0 9,570.0 10,070.0 10,600.0 11,240.0 11,910.0 12,630.0 13,390.0 14,190.0 15,040.0 15,940.0 X Connected 61.3 62.7 64.0 65.4 - 72.0 75.3 78.7 82.2 86.0 89.9 94.0- No. Consumers 5,300.0 5,700.0 6,120.0 6,590.0 7,300.0 8,090.0 8,970.0 9,940.0 11,010.0 12,200.0 13,520.0 14,980.0 Specific Consumption (kWh) 849.0 883.0 918.0 955.0 993.0 1,033.0 1,074.0 1,117.0 1,162.0 1,208.0 1,257.0 1,307.0 Energy Sold (KWh) 4,550.0 5,030.0 5,620.0 6,290.0 7,250.0 8,360.0 9,630.0 11,100.0 12,790.0 14,740.0 16,990.O lq,580.0 Losses (Z) 27.0 27.0 27.0 27.0 25.0 23.0 23.0 23.0 23.0 23.0 23.0 23.0 Energy Sent Out (MWh) 6,230.0 6,890.0 7,700.0 8,620.0 9,760.0 10,857.0 12,506.0 14,416.0 16,600.0 19,143.0 22,065.0 25,429.0 Maximum Demand (MW) 1.8 2.0 2.2 2.4 2.6 - 3.0 3.4 3.9 4.5 5.2 6.0 6.9 Load Factor (X) 39.0 39.0 40.0 41.0 42.0 42.0 42.0 42.0 42.0 42.0 42.0 42.0 Ibb Potential Consumers 11,330.0 12,240.0 13,220.0 14,270.0 15,410.0 16,650.0 17,980.0 19,420.0 20,970.0 22,650.0 24,460.0 26,420.0 Z Connected 60.0 59.2 58.1 56.9 59.8 62.9 66.1 69.5 73.2 76.9 80.9 85.0 No. Consumers 6,800.0 7,240.0 7,b80.0 8,120.0 9,220.0 10,470.0 11,890.0 13,500.0 15,340.0 17,420.0 19,780.0 22,460.0 Specific Consumption (kWh) 724.0 753.0 783.0 814.0 847.0 881.0 916.0 953.0 991.0 1,030.0 1,072.0 1,115.0 Energy Sold (MWh) 4,923.0 5,450.0 6,010.0 6,610.0 7,810.0 9,220.0 10,890.0 12,870.0 15,200.0 17,942.0 21,200.0 25,040.0 Losses (X) 18.0 18.0 18.0 18.0 23.0 23.0 23.0 23.0 23.0 23.0 23.0 23.0 Energy Sent Out (MWh) 6,000.0 6,650.0 7,330.0 8,060.0 10,143.0 11,974.0 14,143.0 16,714.0 19,740.0 23,301.0 27,532.0 32,529.0 Kaximum Demand (MW) 1.7 1.9 2.1 2.3 2.9 3.3 3.8 4.5 5.4 6.3 7.5 8.8 Load Factor (X) 40.0 40.0 40.0 40.0 40.0 41.0 42.0 42.0 42.0 42.0 42.0 42.0 Haiiah Potential'Consumers 5,610.0 5,890.0 6,190.0 6,490.0 6,820.0 7,160.0 7,520.0 7,890.0 8,290.0 8,700.0q 140.0 9,600.0 X Connected 33.9 42.4 50.1 - 64.5 - 80.5 82.4 84.2 86.1 88.1 90.0 No. Consumers 1,900.0 2,500.0 3,100.0 3,700.0 4,400.0 5,200.0 6,050.0 6,500.0 6,980.0 7,490.0 8,050.0 9,640.0 Specific Consumption (kWh) 1,069.0 1,111.0 1,156.0 1,202.0 1,250.0 1,301.0 1,353.0 1,407.0 1,463.0 1,522.0 1,582.0 1,646.0 Energy bold (MWh) 2,031.0 2,778.0 3,584.0 4,447.0 5,500.0 6,765.0 8,186.0 9,146.0 10,212.0 11,400.0 12,735.0 14,221.0 Losses (X) 15.4 15.0 15.0 15.0 15.0 15.0 15.0 15.0 23.0 23.0 23.0 23.0 Energy Sent Out (MWh) 2,400.0 3,268.0 4,216.0 5,232.0 6,470.0 7,959.0 9,631.0 10,760.0 13,262.0 14,805.0 16,539.0 18,469.0 Maximum Demand (M4W) 0.5 0.7 0.9 1.1 1.5 1.8 2.2 2.6 3.2 3.7 4.2 4.8 Load Factor (X) 55.0 54.0 53.0 52.0 51.0 50.0 49.0 48.0 47.0 46.0 45.0 44.0 78 - ANNEX 2 Attachment 8 Y1MEN ARAB REPUBLIC POWER SUBSECTOR REVIEW Load Foreicast of Isolated Systems 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 Sales of Isolated System (MWh) qa'ataba 353 380 410 442 477 514 554 597 644 694 748 806 Dawran Anis 272 293 316 341 367 396 427 460 496 535 576 621 Sa' ada 217 234 252 272 293 316 341 367 396 427 460 496 Al Mahwit 502 541 583 629 678 731 788 849 916 987 1,064 1,147 Rada'a 1,032 1,113 1,199 1,439 1,727 - - - - - 964 1,040 Marakha 455 491 529 570 614 662 714 770 830 895 576 621 Kuhlan 272 293 316 431 367 396 427 460 496 535 835 qOO Al Turbah 394 425 458 494 532 574 618 667 719 775 721 777 Abs 340 367 395 426 459 495 534 575 620 668 721 777 Harad 340 367 395 426 459 495 534 575 620 668 576 621 Midi 272 293 316 341 367 396 427 460 406 535 - - Thula 346 373 402 433 467 - - - - - - Shibam 160 173 186 200 216 - - - - - - - Al Mukha 543 585 631 680 733 - - - - - - - Al Baydah 1,369 1,464 1,578 1,701 1,834 - - - - - - - Al Marawiah 1,014 1,082 1,167 1,258 1,356 - - - - - - - Yarim 1,195 1,288 1,389 - - - - - - - - - Khamer 557 600 647 698 752 - - - - - - - dababa 380 410 442 476 513 - - - - - - Zabio 679 732 789 851 917 - - - - - - - Raydah 265 286 308 332 358 - - - - - 0.6 0.6 Khrif - - 0.4 0.4 0.4 0.5 0.5 0.5 0.5 0.6 0.4 0.4 Kawkaban - - 0.3 0.3 0.3 0.3 0.4 0.4 0.4 0.4 4.0 4.2 Al Luhayan - - 2.7 2.8 3.0 3.1 3.3 3.4 3.6 3.8 4.0 4.2 Al Zuhrah - - 2.7 2.8 3.0 3.1 3.3 3.4 3.6 3.8 0.9 0.9 Hazm Al Jawf - - 0.6 0.6 0.7 0.7 0.7 0.8 0.8 0.8 0.9 0.9 Marib - - 0.6 0.6 0.7 0.7 0.7 0.8 0.8 0.8 Other (unidentified) 1/ _ _ __ - 5,921 7,000 8,561 10,402 11,202 Total Sales (GWh) 10,957 11,790 12,715 12,358 13,494 4,983 11,294 12,789 13,804 17,131 18,454 20,019 Losses I 31.5 31.5 31.0 30.0 30.0 29.0 28.0 27.0 26.0 25.0 24.0 23.0 Gross Generation (MWh) 15,996 17,212 18,428 17,654 19,277 7,108 15,686 19,882 18,654 22,841 24,282 25,999 Load Factor 40.6 40.0 40.0 40.0 40.0 40.0 40.0 40.0 40.0 40.0 40.0 40.0 Peak Load .5 4.9 5.3 5.1 5.5 2.0 4.5 5.7 5.3 6.5 6.9 7.4 Specific Consumption (kWh/customer) 650 676 703 731 760 790 822 855 889 925 962 1,000 No. of Consumers 16,940 17,440 18,087 16,906 17,755 6,307 13,740 14,958 15,528 18,520 19,183 20,019 1/ Displaced generating capacity of 7.5 MW would be installed in other isolated areas. Initial loading of the units is assumed to be 30%, increasing to 50% over 4 years and operating for 6 hours per day. - 79 - ANNEX 2 ittachten t YEMEN ARAB REPUBLIC POWER SU8SECTOR REVIEW Load Forecast of Isolated Syetems After Their Connection to the National Grid 1983 1984 1985 1986 19S7 1988 1989 1990 1991 199? Sales (Kwh) Thula - - - 644 742 853 980 1,128 1,297 1,402 Shibam - - - 299 343 395 454 522 600 690 Al Mukna - - - 1,012 1,164 1,338 1,540 1,770 2,035 2,341 Al Laydah - - 2,531 2,910 3,347 3,850 4,427 5,090 5,854 Al Maraweah - - - 1,871 2,152 2,474 2,845 3,272 3,763 4,328 Yarc, - 1,999 2,400 2,880 3,456 4,147 4,976 5,971 7,165 8,598 Khamer - - - 1,038 1,194 1,373 1,579 1,816 2,088 7,497 Hababa - - - 708 815 937 1,078 1,238 1,424 1,638 4abid - - - 1,266 1,456 1,674 1,925 2,213 2,545 2,427 Kaydah - - - 494 568 653 751 864 994 1,142 MaDar - 4 5 6 7 8 10 12 15 18 Bani Hushaysh - 2 2 3 9 12 15 18 21 27 Al Turbah - - - 7 8 9 9 10 11 12 Sajil - - - 8 9 10 11 12 13 14 Amran - - - 2 3 3 3 3 4 4 Al Barn _ - - - 1 1 1 1 1 1 Sui Al JarraNi - - - I I I 1 I gait al Fagih - - - - 4 5 5 6 6 7 dadan - - - 2 2 2 2 2 1 3 Al Misrakh - - -1 1 1 1 1 1 1 JcDlan - - - 1 1 1 1 1 1 1 Naua'a - - - 2,072 2,487 2,984 3,581 4,297 5,157 6,188 Suototal Sales (MWh) - 2,005 2,407 12,769 17,333 20,228 23,618 27,585 32,235 37,784 Sales-New Unidentified Projects (MWh) - - - - ,000 23,000 39,000 70,000 97,000 117,000 Total Sales (MWh) 2,005 2,407 12,769 35,333 43,228 62,618 97,585 129,235 154,784 Losses X 23 23 23 23 23 23 23 23 23 23 Gross Generation (KWh) - 2,604 3,126 16,583 45,887 56,140 81,322 126,734 167,838 201,108 Load Factor X - 47 47 40 48 50 - 50 51 51 51 oaxtmum Demand - 0.6 0.8 3.9 11.0 12.8 18.6 28.4 37.6 45.0 Specific Conau,sption (kWh/customer) - 731 760 790 822 855 889 925 962 1,000 No. ot Consumers - 2,743 3,167 16,163 21,086 23,659 26,567 29,822 13,508 37,784 - 80 - A(NNEX 2 Attachment 10 YEMEN ARAB REPUBLIC POWER SUBSECTOR REVIEW Demand Forecast for Bulk Consumers 1981 1982 1983 1984 :1985 1986 1987 1988 1989 1990 1991 1992 Sana'a Water and Sewerage - 8.9 15.4 19.2 21.7 23.6 26.0 28.0 30.4 33.1 36.1 39.2 Hotels - 2.6 2.6 5.7 5.7 5.7 5.7 5.7 5.7 5.7 5.7 5.7 Textile Factory - - - 3.1 3.1 3.1 3.1 3.1 3.1 3.1 3.1 3.1 Soft Drinks Factories - - - 6.3 6.6 6.9 7.3 7.6 8.0 8.3 8.5 8.6 Hadua Road Complex - - - 4.4 4.7 4.9 5.1 5.4 5.7 5.9 6.1 6.2 Airport - - - 2.0 2.5 3.0 3.5 4.0 4.1 4.2 4.3 4.4 Conference Centre - - - - - 2.6 2.6 2.6 2.6 2.6 2.6 2.6 Military Housing - - - 3.5 4.8 6.0 7.5 6.5 9.0 9.4 9.7 9.9 Glass Factory - - - 2.8 4.8 5.0 5.3 5.6 5.8 5.9 6.0 6.1 Unidentified ____ 3.0 6.0 10.0 15.0 21.0 28.0 lotal Energy Sold - 11.5 18.0 47.0 53.9 60.8 69.1 76.5 84.4 98.2 103.1 113.8 Energy Generated 1/ - 13.1 20.5 53.4 61.3 69.1 78.5 86.9 95.9 105.9 112.2 128.5 Maximum Demand (MW) 2/ - 2.6 4.1 10.6 12.2 13.7 15.6 17.3 19.0 21.0 23.3 25.5 Hodeidah Existing bulk Consumers 2.2 4.5 4.5 4.5 4.5 4.5 4.5 4.5 4.5 4.5 4.5 4.5 Cotton Factory - - - - - 2.5 2.5 2.5 2.5 2.5 2.5 2.5 Cold Stores & Ice Plant - - 1.6 3.5 6.5 15.0 15.0 15.0 15.0 15.0 15.0 15.0 Hodeidah Port - - 7.3 14.7 15.4 15.5 15.6 15.7 15.8 15.9 16.0 16.1 W4ater and Sewerage - - - 1.8 2.1 2.5 3.0 3.3 3.6 4.0 4.4 4.8 Salif Port - - - - 3.9 4.1 4.3 4.6 4.8 5.0 5.3 5.7 oil Pipeline - - - - 1.3 1.4 1.5 5.1 5.4 5.6 5.8 6.0 Salt Factory - - - - 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 Unidentified _____3.0 6.0 10.0 15.0 Total Energy Sold 2.2 4.5 13.4 24.5 35.2 47.0 50.9 58.2 63.1 69.0 76.0 84.1 Eniergy Generated i/ 2.5 5.1 15.2 27.8 40.0 53.4 57.8 66.1 71.7 78.4 86.4 95.6 Maximum Demand (MW) 3/ 0.5 1.1 3.2 5.9 8.5 11.3 12.2 14.0 15.2 16.6 18.3 20.2 Taiz Water and Sewerage - - - 1.3 2.5 2.8 3.0 3.4 3.7 4.1 4.5 5.0 Biscuit Factory - - - 10.0 20.0 20.0 20.0 20.0 20.0 20.0 20.0 20.0 Gnee & Soap Factory - - - - 6.7 7.3 7.3 7.3 7.3 7.3 7.3 7.3 Plastics Factory - - - 2.9 5.8 6.1 6.4 6.4 6.4 6.4 6.4 6.4 Airport - - - 1.1 2.2 2.3 2.4 2.4 2.4 2.4 2.4 2.4 Unidentified _ 3.0 6.0 10.0 14.0 19.0 24.0 Total Energy Sold - - - 15.3 37.2 38.5 42.1 45.5 49.8 54.2 59.6 65.1 Energy Generated 1/ - - - 17.4 42.3 43.8 47.8 51.7 56.6 61.6 67.7 74.0 Maximum Demand (MW) 4/ - - - 17.4 42.3 43.8 47.8 51.7 56.6 61.6 12.9 14.1 Cement Plants Sales (GWh) Bajil - - - 40.0 90.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Amran - - - 30.0 60.0 60.0 60.0 60.0 60.0 60.0 60.0 60.0 Mufraq - - - 25.0 90.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Total Energy Sold - - - 70.0 150.0 160.0 185.0 250.0 260.0 260.0 260.0 260.0 Gross Energy Generated 1/ - - - 79.5 170.5 181.8 210.2 284.0 295.5 295.5 12 12 Load Factor (X) - - - 60.0 74.0 74.0 69.2 74.0 75.5 75.5 295.5 295.5 Maximum Peak Demand (MW) - - - 15.2 26.3 28.0 34.7 43.8 44.7 44.7 75.5 75.5 1/ Assuming 12% losses. B/ Assuming a load factor of 57.5% (K&D) 3/ Asuming a load factor of 54.0% (K&D) 4/ Assuming a load factor of 60.0% (K6D) - 81 - ANNEX 2 Attachment 11 YEMEN ARAB REPUBLIC POWER SUBSECTOR REVIEW Assumptions for Load Forecast Sana'a Taiz Hodeidah Dhamar Ibb Hajjah Country Town Town & Suburbs Town Town & Suburbs Town Town Town Town Population 1981 8,500,000 279,400 319,400 116,400 136,400 126,400 Population growth 2.4X Populstaon gruwtn (1982-1985) 7.25% 5,50% 8.75% Population growth (1986-1992) 6.00% 5.75% 8.75Z Population with elec. service (1982) 33.0X 72.00X 73.00% 54.00% 61% 60% 34% Family size (persons/family) 6.9 7.4 6.4 Population with elec. service (1982) 97.00X 95.001 90.00% 94% 85% 90% Number of Customers (1981) - Domestic 33,500 13,400 10,600 5,300 /4 6,800 /4 1,900 /4 - Commercial 11,700 7,200 4,100 - Small industrial 2,290 794 394 - Government 800 180 220 - Diplomatic 350 - - Growth in Number of Consumers - Domestic /l /2 /3 /5 /6 /7 - Commercial 2% 2% 2% - Small industrial 2% 2% 21 - Government 4X 4X 3% - Diplomatic 1X - - Specific Consumption (kWh/year) - Domestic 1,300 850 2,000 850 /8 720 /8 1,070 /8 - Commercial 820 600 1,030 - Small industrial 2,840 2,700 2,170 - Government 14,880 11,650 23,110 - Diplomatic 22,290 - - Growth in Specific Consumption - Domestic 4X 41 4% 4Z /8 4Z /8 4X /8 - Commercial 42 4% 4Z - Small industrial 41 4% 4% - Government 4% 41 4.5% - Diplomatic 2% - - /1 Average growth is 2.75X p.a., but connection rate is assumed to be faster during initial years due to implementation of IDA's power projects. /2 Average growth is 2.42% p.a., but connection rate is assumed to be faster during initial years due to implementation of IDA's power projects. 7T Average growth is 4.75% p.a., but connection rate is asssumed to be fastetr during initial years due to implementation of IDA's power projects. /4 All customer classes. 7; Average growth is assumed to be 9.91 p.a., but connection rate is assumed to be slower during initial years. /Tb Average growth is assumed to be 11.51 p.a., but connection rate is assumed to be slower during initial years. /7 Average growth is assumed to be 14.81 p.a., but connection rate is assumed to be slower during initial years. /8 Average of all customer classes. - 82; - ANNEX 2 Attachment 12 Page 1 of 4 YEMEN ARAB REPUBLIC POWER SUBSECTOR REVIEW Basis of Forecasts of YGEC Demand Load Forecasting 1. YGEC's forcasts for urban areas and the interconnected system have been prepared with assistance from K & D. NREC established a demand forecasting methodology for rural areas based on their village surveys and this has been used by YGEC and its consultants to produce forecasts. YGEC itself has revised a forecast, prepared in March 1982 with assistance from K & D based on recently obtained information. This revised forecast, prepared in October 1982, is substantially lower than the K & D forecast, but close to the revised forecast by the mission. Urban Load Forecasting 2. Detailed urban forecasts were prepared in March 1982. Residential and commercial loads were forecast based on projections of population. A combination of trend extrapolation and consideration of individual large project loads was used to forecast the electricity demand of industrial, government and other consumers. The number of residential and commercial consumers (the Private tariff category) were forecast in relation to the growth in population. Preliminary results for the 1981 CYDA census were used to establish the number of potential consumers in 1981, taking the number of persons per household from the 1975 Census results. Adjustments were made to allow for more consumers than households, i.e. to take account of shops and one meter serving more than one consumer, (1.08 consumers/household in Sana'a and Hodeidah, 1.29 in Taiz). Urban population was assumed to grow during 1981-85 at 8% p.a. in Sana'a, 7% p.a. in Hodeidah and 6% p.a. in Taiz. After 1985 the population in all three cities was forecast to grow at 5% p.a. The number of persons per household was projected to fall because of young former emigrant workers establishing new households. The rate of growth of households was therefore projected to grow 2-5 percentage points per annum above the population growth rate. Further adjustments were made to allow for households without electricity supply to be connected. Projected growth rates for the number of potential and connected residential and commercial consumers are summarized below: Actual ----Forecast---- Annual Growth Rate (% p.a.) 1981 1985 1990 1981-85 1985-90 Potential Consumers Sana'a 49,332 79,886 120,065 12.8 8.5 Hodeidah 21,176 32,627 48,304 11.4 8.2 Taiz 23,143 33,137 49,425 9.4 8.3 TOTAL 93,651 145,650 217,794 11.7 8.4 Connected Consumers Sana'a 45,332 77,241 118,516 14.3 8.9 Hodeidah 14,343 28,635 46,533 18.9 10.2 Taiz 21,043 31,908 48,880 11.0 8.9 TOTAL 80,718 137,785 213,929 14.3 9.2 ANNEX 2 Attachment 12 Page 2 of 4 The projection assumes that 86% of potential consumers are already connected and that this would rise to 98% by 1990. During 1981-1986 274 consumers per week would have to be connected, compared to 277 for all types of consumers in 1981, which is feasible. Average consumption per consumer was assumed to grow at 9% p.a. for Sana'a, 10% p.a. for Hodeidah and 7% p.a. for Taiz. Overall rates of growth in consumption were therefore: Growth in Electricity Sold to Households and Commerce (% p.a.) 1981-85 1985-90 Sana'a 24.5 17.0 Hodeidah 30.8 21.2 Taiz 18.7 16.5 TOTAL 25.3 18.3 3. The consumption of large consumers, defined as having an ultimate load greater than 1 MW, was forecast from the plans of individual projects. The most important of these were the cement works at Bajil, Amran and Mufraq, urban water supply and sewerage, a proposed steel plant, ports and oil products pipelines from Salif to Sana'a and Al Mukha to Taiz. Rural loads were forecast using the NRECA methodology and assumptions. Consumption of remaining consumer types, small industry, government, diplomatic and street lighting were projected by extrapolating past trends of consumers and consumption per consumer, modified by judgement. In the case of small industry, the potential market was estimated from a survey and extrapolated at 15% p.a. Assumptions were then made on he share of the potential market which YGEC would supply. 4. The forecast is summarized in the table below. Total electricity sold by YGEC is projected to grow at 43.1% p.a. during 1981-85 and 19.2% p.a. during 1985-90. Most of this growth comes from the industrial sectors, which the consultants project to account for 51% of total sales in 1985 and 1990. The cement works alone amount to 22% in 1985 and 16% in 1990. Sales to rural consumers are also projected to grow rapidly (49.3% p.a. during 1981-85, 29.2% p.a. during 1985-90) but their share in total electricity does not increase greatly. ANNEX 2 Attachment 12 Page 3 of 4 Table 9 March 1982 Demand Forecast (YGEC Systems) Consumer % Share Annual Growth Rate (%) Group 1981 1985 1990 1981 1985 1990 1981-85 1985-90 Private 105.7 230.4 533.0 65 34 33 21.5 18.3 Governmentl/ 25.2 31.6 50.8 16 5 3 5.8 10.0 Diplomatic 7.8 10.2 13.6 5 2 1 6.9 5.9 Small Industry 11.6 70.2 216.5 7 10 13 56.8 25.3 Bulk - 129.0 357.8 - 19 22 22.6 Cement - 150.0 260.0 - 22 16 1.3 Rural 11.0 54.7 196.8 7 8 12 49.3 19.2 TOTAL 161.2 676.2 1628.7 100 100 100 43.1 19.2 Energy delivered (GWh) (219) 763.8 1840.6 36.7 19.2 Total maximum demands (MW) 53.9 167.9 394.8 32.9 18.6 Distribution losses (%) 26.4 11.5 11.5 Interconnected System 2/ Energy generated(GWh) - 813 1958 19.2 Maximum demand(MW) - 170 401 18.7 NOTES 1/ Government included Streetlighting 2/ Figures for interconnected system take account of a progriam for regional interconnection and include a mission estimate of 6% station use. Rural Load Forecasting 5. Rural demand forecasts were based on surveys conducted by NRECA. Projections were prepared for six sectors; households, street]Lighting, public buildings, potable water projects, shops and industry. In addition, agricultural water pumping and large industry (greater than 75 kVA) were considered in areas where such projects were likely. Numbers of consumers were estimated from population projections. Total rural population was projected to grow at 2.3% p.a. and the number of persons per household to decline because of changes to the age structure of the population. Forecasts were based on NRECA's design concept of "full area coverage" where all potential consumers in a district were assumed to ultimately take supply within an unspecified period. - 85 - ANNEX 2 Attachment 12 Page 4 of 4 6. Forecasts of consumption per consumer were based on estimates of consumers' applicance stocks. Initial stocks were obtained from surveys. Households were stratified into four income groups. Different levels of applicance ownership were projected for each, based on the initial stock and professed intention to acquire appliances obtained from the surveys. An example of appliance ownership levels projected for Ma'bar is shown in the table below. Household Appliance Ownership Levels, Ma'bar Consumption per Appliances per 100 Electrified Households Appliance (kWh/a) Survey Year 9 Saturation Light 1327 538 538 84 TV, radio, cassette recorder 127 86 86 220 Refrigerator 18 33 45 329 Washing machine 18 33 55 80 Electric cooking (fry pans, kettles) 36 50 73 168 Electric water heating (immersion coils) 45 98 98 64 Electric space heating - 30 30 234 Sewing machine 18 71 71 10 Forecasts of appliance stocks were also used to project the demand of commercial, government building and industrial consumers. - 86 - ANNEX 2 Attachment 13 YEMEN ARAB REPUBLIC POWER SUBSECI'OR REVIEW Accuracy of Demand Forecasts Made in 1977 1977 1978 1979 1980 1981 Sana'a Sales (Gwh)-Actuai 29.7 40.7 55.3 74.6 84.6 -Forecast 29.0 34.7 59.8 78.3 93.2 -% Error -2.4 -14.7 8.1 5.0 10.2 Generation -Actual 38.5 54.4 76.8 100.6 119.7 (Gwh) -Forecast 36.2 42.3 70.4 88.9 103.6 -X Error -6.0 -22.2 -8.3 -11.6 -13.5 Maximum -Actual 10.1 15.0 18.9 23.8 28.0 Demand -Forecast 9.9 11.0 16.3 20.0 23.1 (MW) -D Error -2.0 -26.7 -13.8 -16.0 -17.5 Hodeidah Sales (Gwh)-Actual 14.6 17.7 20.7 27.1 33.8 .-Forecast 14.5 18.9 40.8 60.2 79.5 -X Error -0.7 5.8 97.1 122.1 135.2 Generation -Actual 19.6 24.7 29.3 38.9 47.7 (Gwh) -Forecast 18.2 23.6 48.0 66.9 88.3 -x Error -7.1 -4.5 63.8 72.0 85.1 Maximum -Actual 3.7 4.6 5.7 8.0 10.0 Demand -Forecast 3.7 5.2 10.2 14.0 18.2 (MW) -% Error - 13.0 78.9 75.0 82.2 Taiz Sales -Actual 7.7 9.2 13.2 17.3 20.4 (Gwh) -Forecast 8.0 10.2 25.1 35.4 43.4 -% Error 3.9 10.9 90.2 104.6 112.7 Generation -Actual 11.1 12.1 20.0 25.6 30.0 (Gwh) -Forecast 10.6 13.6 33.5 44.2 51.1 -- Error -4.5 12.4 67.5 72.7 70.3 Maximum -Actual 2.9 3.5 5.2 6.5 7.5 Demand -Forecast 2.7 3.4 7.6 9.5 10.9 (MW) -% Error -6.9 -2.9 46.2 46.2 45.3 Total Sales -Actual 52.0 67.6 67.4 119.0 138.8 (Gwh) -Forecast 51.5 63.7 125.7 173.8 216.1 -Z Error -1.0 -5.8 86.5 46.1 55.7 Generation -Actual 69.2 91.2 126.1 165.1 197.4 (Gwh) -Forecast 65.1 79.4 151.9 200.0 242.9 -% Error -5.9 -12.9 20.5 21.1 23.0 Maximum -Actual 15.8 21.9 28.3 36.3 43.2 Demand -Forecast 15.5 18.6 32.4 41.3 49.6 (MW) % Error -1.9 -15.1 14.5 13.8 14.8 - 87 - ANNEX 2 Attachment 14 YEMEN ARAB REPUBLIC POWER SUBSECTOR REVIEW Sources of Demand Forecast Error (Forecasts for 1981 made in 1977) Sector Sana'a Hodeidah Taiz Total Private -Actual 55,531 25,916 15,749 97,196 -Forecast 34,409 27,594 14,530 76,538 -% Error -38.0 6.5 -7.7 -21.3 Government -Actual 14,722 5,984 2,497 23,203 -Forecast 12,981 5,147 2,329 20,457 : Error -11.8 -14.0 -6.7 -11.8 Diplomatic,-Actual 7,777 - - 7,777 -Forecast 5,100 - - 5,100 -% Error -34.4 -34.4 Industry -Actual 6,580 1,863 2,146 10,589 + Bulk -Forecast 40,725 46,708 26,498 113,931 -% Error 518.9 2407.1 1134.8 975.9 Total -Actual 84,610 33,763 20,392 138,765 -Forecast 93,215 79,454 43,357 215,026 -X Error 10.2 135.3 112.6 55.7 - 88 - ANNEX 3 Page 1 of 14 YEMEN ARUB REPUBLIC POWER SUBS]ECTOR REVIEW Electricity Supply A. Past Electricity Supply Generating Capacity 1. The growth of installed capacity in YAR is not known because detailed data on private suppliers and autoproducers are not readily available. YGEC generating capacity increased from a 15MW in 1975 to 121 MW iin 1982; representing an average annual increase of about 35%. The growth of installed capacity maintained by the private suppliers and autoproducers can not be determined exactly. According to surveys conducted by NRECA and K&D, together with import statistics, the number of sets in 1981 amounted to between 80,000 to 100,000 with a total capacity of between 200MW to 600MW. Trade data suggest that this has grown from between 1OMW and 30MW in 1975. This range of uncertainty is too high to enable sound policies to be formulated for this part of the power subsector. However, much improved estimates would be obtained frQm more comprehensive surveys of the private power subsector (Annex 2, para. 3). 'the growth in generating capacity is shown in detail for YGEC in Attachment 1 and summarized in Table 1. Table 1 Growth of Generating Capacity, YAR, 1975-81 Annual Rate of Growth (%) 1975 1981 1975-81 YGEC Sana'a 7.0 41.5 34.5 Hodeidah 3.3 25.9 41.0 Taiz 2.8 18.5 37.0 Others 1.6 23.7 56.7 Total 14.7 109.6 39.8 Autoproducers & private 10-30 200-600 Total YAR 25-45 310-710 Sources: YGEC, mission estimates. 2. Both the public and private power systems comprise isolated systems fed by diesel generators. YGEC is a new utility formed in 1975 where development in its initial years was determined by the unsatisfied demand for electricity in the main cities and its establishment as a Government institution, rather than as a result of a systematic plan for development. - 89 - ANNEX 3 Page 2 of 14 The private power subsector is large in relation to YGEC. This is a consequence of both the constraints on YGEC which prevented it from keeping abreast of demand as well as the physically scattered nature of demand. The rate at which YGEC could expand was constrained by the national shortage of skilled manpower, shortages of finance and problems in project implementation. 3. Attachment 2 shows the balance of supply and demand for the main YGEC systems. Until the late 1970's demand was constrained by shortages of generating capacity. Reserve margins in Sana'a, for example, were below 15% of demand until 1980. From 1979 onwards the quality of supply improved as a program was implemented to install diesel units at Sana'a, Hodeidah and Taiz as an interim measure before Ras Katenib is commissioned in 1983. However, the reinforcement and expansion of distriLbution did not keep in step with the increase in generating capacity and as a result, the new diesel power plants are presently underutilized, especially in Hodeidah and Taiz. Parallel to this, YGEC embarked on a program of elecitrification in provincial capitals and major towns. In some of these towns uniit sizes were large in relation to demand, necessitating either a large amount of spare capacity, or the risk of load shedding when the sole generating unit breaks down. (para. 19) 4. YGEC generating plant tends to be almost entirely medium speed diesel units. Transportable high speed diesel lplants were introduced to alleviate severe shortages in the public supply of public electricity and to this date about 3 MW still in service. Private generating plant is also diesel driven, but very small units are powered by gasoLine engines. Most private diesel sets are of the high speed type. 5. The size distribution of YGEC generation plant is shown in Attachment 3. Over half of YGEC's units have a capacity between 400kW and 700kW. Most of these units are located in provincial capitals or towns. Some were temporarily installed in the three main cities to meet severe shortages of power supply and later moved to other locations. YGEC has 18 large sets amounting to 77MW in Sana'a, Hodeidah and Taiz, ranging in unit capacities between 2.5 MW and 6.4 MW. Furthermore, there are another 10 units with capacities between 1.2MW and 1.3MW. All of these, with the exception of one 1.2 MW unit at Sa'ada, are located at centers to be interconnected by 1985. 6. The age distribution of YGEC generating plant is also shown in Attachment 3. Over 80% of YGEC's units are less than seven years old. The condition of plant commissioned since 1975 is generally good, although some units need refurbishment because normal maintenance could not be carried out when there were shortages in generating capacity. - 90 - ANNEX 3 Page 3 of 14 7. Trade statistics and anecdotal evidence suggest that most private generating plant is less than five years old. Most of this consists of high speed units that have not been adequat:ely maintained. Consequently, most private generating plant must be reaching the end of its economic life and will need replacement during the next five years, or earlier. 8. Existing generating plant is a legacy of the period of the rapid growth in electricity demand which necessitated a quick unplanned response by YGEC. The demand which YGEC could not: satisfy was met by the private suppliers and autoproducers. However, much of the interim plant purchased by YGEC will remain in the future. The large diesel units would be used for standby and to provide spinning reserve on the interconnected system in the medium term. Much of the smaller plarLt would be available fior use in remote systems, where private generating,plant is nearing the end of its economic life. Fuel Consumption 9. Gas oil (diesel) is the sole fuel used by YGEC for power generation. Between 1975 and 1981, consumption of gas oil increased at an average annual rate of 28.3%, from 21.7 million tons in 1975 to 51.7 in 1981. (Attachment 4) In 1981, this represented 15% of the total consumption of gas oil in YAR. Average thermal efficiencies for the YGEC systems are shown in Table 2. In 1981 average efficiency was 36.4%, based on a net fuel heat value of 10,400 kcallkg. Thermal efficiency was highest in Sana'a and Taiz (37.6%) and lowest in Ibb and Hajjah (29.2X). Efficiency increased by about 5 percentage points during 1975-1981, mainly because new medium speed units were commissioned and old inefficient, or high speed units were either retired or operated for fewer hours. All existing medium speed engines would be converted to burn heavy fuel oil. 10. Autoproducers and private utilities consume gas oil, gasoline and kerosene for power generation. Many small units, designed to run on gasoline, are fuelled by kerosene because of its low price relative to gasoline. No reliable data on the fuel consumed by private producers exist. A crude estimate based on rough estimates of installed capacity would put gas oil consumption of power producers other than YGEC at between 79-156 thousand tons and gasoline and kerosene at between 6-19 thousand tons, each. The entire power sector in 1981 would therefore have consumed between 131-208 thousand tons of gas oil, or 38-60% of recorded national consumption. - 91 - ANNEX 3 Page 4 of 14 Table 2 Average Thermal Efficiency 1975 1981 Sana'a 29.9 37.6 Hodeidah 31.4 36.5 Taiz 33.5 37.6 Dhamar - 34.2 Hajjah - 29.2 Ibb - 29.2 Branches - 31.9 Total 30.8 36.4 Source: Mission estimates based on YGEC data. Note: Diesel fuel assumed to have a net heat value of 10,400 kcal/kg or 12.1 thermal kWh/kg. Distribution 11. YGEC inherited run-down distribution networks when it was formed in 1975. The previous companies had suffered from lack of staff and finance so that maintenance had been inadequate and network expansion was insufficient to meet demand, which had started to grow rapidly. Moreover, each company had developed along separate lines so that technical standards varied widely. The primary distribution voltage was 15kV in Sana's and lOkV in Hodeidah and Taiz. Low voltage networks required rehabilitation or replacement. The best distribution networks had small conductor sizes (35mm2) which were incapable of delivering increased loads and led to high losses and voltage drop. 12. After its formation YGEC initiated a program of distribution rehabilitation and reinforcement, with assistance from the Arab Fund, IDA and bilateral donrs. YGEC carried out a nationwide study of rural electrification with assistance from IDA, which proposed a program of development and established technical standards. No inventory of distribution lines in existence when YGEC was formed is available, but most of these lines required replacement. However, YGEC managed to make steady progress in improving its distribution systems, installing 180km of MV line, 194km of LV line and about 230km of cable, (Table 3). Nevertheless, distribution investment was insufficient, despite this progress, as indicated by the high losses in 1981 (para. 14). The distribution system tended to expand linearly in roughly equal annual increments, whereas demand was growing exponentially at over 30% p.a. during 1976-81. Increasing the rate of distribution investment is a major problem faced by YGEC (para. 24). - 92 - ANNEX 3 Page 5 of 14 Talble 3 YGEC DistributLon Reinforcement (Km installed, cumulative) 1975 1976 1977 1978 1979, 1980 1981 Overhead Lines 33kV - - - - - 17.5 24.0 lOkV, llkV, 15kV - 15.5 43.2 81.6 115.0 138.0 156.5 LV - 10.6 34.6 65.0 123.2 151.0 194.4 Underground Cables 33kV - - - - - - 3.0 10kV, 11kV, 15kV - 12.0 35.3 61.4 90.8 111.5 135.9 LV - - 4.7 9.8 19.1 25.5 93.2 Source: YGEC 13. Distribution facilities of private suppliers are generally of a low technical standard and are often little more than housewiring cable strung from buildings, trees, etc. However, some suppliers have erected lines attached to insulators on steel poles, although conductor sizes and distances between conductors are usually inadequate. Savings in terms of national resources would result if private networks were designed and constructed to higher standards. In view of the apparent importance of private producers in overall electricity supply, MEW should initiate policies to improve their technical standards (para. 43). Losses 14. In 1981 losses for all YGEC systems amounted to 29.3% of energy generated. In the three main systems losses were: Sana'a 29.3%, Hodeidah 29.2% and Taiz 32.0%. Moreover, losses have been increasing. In 1970 losses for all YGEC systems were 21.5% which increased only slightly to 22.2% in 1975 before rising steeply to the present level (Annex 2, Attachment 1). 15. No investigations of YGEC's losses have been undertaken yet, but it is believed that a proportion of the losses comprises theft and billing errors. A further proportion can probably be accounted for by inaccurate metering and some unmetered consumption of YGEC employees. The remaining would be technical losses arising from an overloaded distribution system. Technical losses represent wasted fuel,, supply capacity that cannot service consumers and potential loss of income to YGEC. These losses are high by any standards, but are a consequence of demaand growing more rapidly than the system could expand. The lack of an effective consumer department also contributes to the large amount of unaccounted for electricity. Future losses will appear to fall as MV consumers such as the cement works are connected, since losses are lower at MV. The trend towards lower losses arising from a greater proportion of MV sales should not obscure the need to take action to - 93 - ANNEX 3 Page 6 of 14 reduce losses elsewhere. In view of the adverse impact of losses on YGEC's revenue generation and planning, it is recommended that the Government strengthen YGEC's consumer department and consider a two pronged strategy for reducing losses, the longer term elements of which would be formulated by the consultants engaged to the Financial Recovery Plan (para. 7.07) The first would aim at reducing the losses attributed to existing facilities and the other relates to the reduction of theft. These would involve: a) inspections of LV networks and tihe rehabilitation of inadequate service connections; b) improving procedures for meter reading and billing; c) improving procedures for meter reading and billing; d) introducing an effective sealing method for meters together with procedures for meter inspection; d) strengthening the meter testing, calibration and repair workshop and formulating a program for the regular testing of all meters, starting with old meters supplying large consumers; e) metering all consumption of employees and showing this as consumption rather than losses; f) carrying out surveys to identify the main parts of the networks where technical losses occur; g) instituting immediate distribution reinforcements to reduce losses in those parts of the networks where losses are greatest; and h) establishing a system to regularly measure and record transformer and feeder loadings to identify distribution reinforcement projects for inclusion in the annual planning memorandum (para. 26). B. Power System Planning Institutional Responsibility 16. YGEC is responsible for planning the development of its system. In practice, most planning is carried out by consultants because of shortages of professional staff. The Corporation doeEs plan some rural electrification and distribution projects, especially those requiring local generation. All projects of any size are approved by MEW. MEW consults with CPO over project financing plans. Major decisions involving generation and transmission to the main cities are finalized at cabinet levetl. A certain amount of project identification, such as the location of major power stations and the choice of villages to be electrified by YGEC, is carried out at the highest levels of Government. YGEC has been improving its project identification ability, especially by the national rural electrification study financed by IDA under the Power II Credit. However, YGEC's project identification and planning functions need considerable strengthening. This will be addressed in the financial recovery plan (see Chapter VII). - 94 - ANNEX 3 Page 7 of 14 Generation 17. Generation planning for the interconnected system has been carried out by consultants. Consultants also prepared the program of interim diesel generation before the Ras Katenib steam station could be completed. A simulation approach was used to determine the least cost: program to meet projected demand on the interconnected system. The consultants assumed that all diesel units with a capacity below 2MW would be scrapped or transferred to isolated systems. No direct or indirect contribution to the system from large autoproducers, e.g. the cement works, was assumed. The most recent report on generation planning, prepared in 1980, set out to establish the optimum unit size and phasing for Al Mukha power station. A study currently being completed will establish YGEC's generation requirements until the end of the century. This study will also examine the choice between g.as turbine and steam plant and the options for using imported coal instead of oil, as well as the optimum unit size and phasing. 18. The techniques adopted by YGEC's consultants for gleneration planning were reviewed by the mission and found appropriate to the size of the system and range of alternatives to be considered. A disadvantage in relying on consultants for all generation planning has been that the Government can not get quick economic analysis of options which arise from a changing environment. However, since YGEC is unlikely to require major generation investment in the medium term, it is not recommended that YGEC develop its own generating planning capability at present, except for isolated rural systems. Distribution planning has a much greater priority. (para. 26). 19. Generation planning for areas outside of the three main cities of Sana'a, Hodeidah and Taiz has been carried out by YGEC itself. Because of the urgent need for public supply and the shortage of professional staff, such planning has been of an ad hoc nature, generally considering a standard size generator (typically 530 kW) to meet an immediate requirement. YGEC has responded flexibility to the changing balances of supply and demand in its systems by moving small generating plant when large units are commissioned to areas where capacity has been in short supply. Providing reliable supply in face of rapidly growing demand in electrified areas and unsatisfied demand elsewhere did not give YGEC time to carry out economic analysis of these projects. Generating units sizes selected for rural areas aLre large in relation to demand. In some cases supply rests on a single unit and demand during the day may be insufficient to enable continuous operation. However, such quality of supply may be acceptable to rural consumers, since it is usually better than supply from private producers. Large unit sizes are to some extent a consequence of the policy of using these units to supply the main cities when supply was short before moving them to the branches. The medium speed 530 kW units could also ultimately be converted to burn heavy fuel oil. 20. YGEC will have surplus diesel generators once the interconnected system is completed to supply the main rural centers. There will also be surplus generating capacity on the interconnected system. YGEC will have to decide whether to supply villages from the main system by constructing sub-transmission, or use surplus diesels. Such decisions should be based on - 95 - ANNEX 3 Page 8 of 14 economic analysis using discounted cash flow techniques. It is recommended that YGEC carry out this analysis itself in order to assure that the least cost method is chosen for supplying rural projects, and to enable its staff to gain experience in generation planning. Transmission 21. Transmission planning has concentrated on analysis of power flows, fault levels and transient stability. These studies have established the technical feasibility of the planned development of the system. Transmission requirements are dictated to a large extent by power station location. However, a number of alternatives exist that should be subject to economic analysis. These include extending the 132kV transmission system to Sa'ada and Hajjah, compared to local generation, the timing of reinforcements such as the Al Barh to Bajil 132kV line and the choice between local peaking plant and delaying the development of steam generation and 132kV transmission. It is recommended that the planning of transmission be strengthened to place a greater emphasis on economic as well as technical issues. YGEC should prepare terns of reference for future transmission studies that ensure a thorough examination of the economics of alternatives. Distribution 22. Distribution planning for the main cities has generally been carried out by consultants, although YGEC plans and executes some projects intended to overcome immediate problems. Most of the urban distribution networks were in poor condition when YGEC was established in 1972. A major urban distribution design study was carried out following a recommendation by IDA in 1978 using optimization techniques. This study led to major potential savings compared to designs using traditional techniques and established design standards for subsequent urban network reinforcement. 23. A nationwide rural electrification study carried out by NRECA, with financial assistance from IDA, established the design philosophy for rural distribution. YGEC has designed and executed a number of rural electrification schemes itself. Consultants have been retained for large programs, such as that financed under the Power II Credit. A list of rural electrification projects has been prepared by consultants. Cost-benefit analysis of each of these schemes has been carried out on the assumption that grid connection is preferred to local generation, which is probably true for most, if not all, of the cases considered. Distribution, and rural electrification generation, transmission projects have generally been considered one at a time in isolation from one another. This is not surprising in view of the scattered systems which YGEC consisted of. However, as the system develops, there is a clear need for the plan for urban and rural distribution to be brought together with those for generation and transmission into an integrated Program for YGEC. - 96 - ANNEX 3 Page 9 of 14 24. YGEC's general problems of shortages of manpower, financial constraints and difficulties in project implementation have had the greatest impact on distribution. The extension of distribution networks in both urban and rural areas has fallen behind schedule and has not developed in step with generation. Delays to some apparently minor contracts, caused by shortages of finance or contractual problems, have delayed the commissioning of major projects. A striking example of this is the inability to supply Sana'a with power from the completed Ras Katenib power station and 132kV transmission system because of delays to a relatively minor 33/llkV substation contract. With the benefit of hindsight, it is clear that manpower, financial and administrative constraints on project implementation were greater than envisaged when YGEC's distribution program was prepared. Future Power System Planning 25. The overriding issue in power system planning is the need to formulate a balanced investment program so that generation, transmission and distribution investment can achieve better balance and coordination. This would bring together projects that have already been identified into a coherent, feasible stragegy. However, the highest priority at present is to expand the distribution system in order to utilize excess generation and transmission capacity. The immediate planning requirement is the formulation of an overall medium term investment plan which would focus on distribution. Such a plan would establish priorities, ensure the balanced growth of distribution and rural electrification with generation and transmission, and be consisted with the resources available to YGEC. Furthermore, it would enable modifications to the investment program to be balanced among each element of the system and be consistent with YGEC's overall economic objectives. Such modifications could be brought about by, for example, unforseen changes in the availability of finance, project delays, or divergence between projected and actual demand. It is thereFore recommended that YGEC formulate an investment strategy that would bring together projects identified in existing consultant's studies, establish priorities among them and take a realistic view as to what can be achieved during ithe next 10 years. It should be formulated to enable balanced changes to be made to the strategy when unforseen circumstances eventuate. It is recommended that this investment strategy be prepared as part of the proposed financial recovery plan (see Chapter VII. 26. It is now clear that the financial, manpower and other constraints on the ability of YGEC to implement projects were greater than realized when the original plans to expand the system were formulated. Projects have been delayed, necessitating expensive emergency measures, lower quality of supply, or restrictions on new connections. Two measures to improve YGEC's planning are recommended. First, the analysis of major projects should be broadened to include an examination of the effect of the project on YGEC's financial accounts and overall manpower requirements. Financial analysis and manpower planning should be an integral part of YGEC's project appraisal. Second, each year YGEC should prepare a medium term planning memorandum which should include a coordinated investment program for generation, transmission, distribution and rural electrification; a financial plan and projections of accounting statements; and an updated manpower and training plan. The - 97 - ANNEX 3 Page 10 of 14 planning memorandum would be an annual updating of the proposed financial recovery plan. It should be short and policy orientated with technical annexes describing the analysis underpinning major recommendations. 27. A lesson from the recent past is the need for flexibility in planning to avoid commitment too early to major investment decisions that may require adjustment later in response to changes in demand and delays to other projects. The need for flexibility should continue since future demand remains highly uncertain and the choice of fuel and location of the next generation project will be influenced by the results of the geothermal exploration drilling. It is recommended that YGEC continue to formulate a long term generation program to guide the design of immediate projects, but avoid taking decisions on particular projects within the program until such decisions become necessary. The long term program should be reviewed annually as part of the medium term planning memorandum. The planning memorandum should identify decisions relating to each step in the project planning and execution sequence and set dates by which they should be taken, to be reviewed annually. Furthermore, YGEC is recommended to negotiate contracts with suppliers that have options for increasing or reducing the scope of supply. Particular attention could be given to placing firm orders for only the best two generating units in a new power station with options for purchasing additional units later. 28. A further issue is the range of alternatives that should be examined in long term planning and project design. Some feasible alternatives could be given more attention. These include: a) The optimum use of existing diesel generating plant; b) The potential for co-generation or peak-lopping with private generating plant, such as that owned by cement works and other large autoproducers; c) Power station location, particularly in relation to fuel unloading and in the case of peaking plant, system security and fuel transport costs; d) phasing power station development, i.e. proccuring units singly or in pairs, or negotiating a contract for say, the first two units, with an option for ordering others later; and e) phasing transmission development, e.g. delaying subsidiary transmission lines or bulk supply points until they are economically justified. It is recommended that, in drawing up terms of reference for future planning studies, YGEC ensure that adequate economic analysis is carried out for all technically feasible alternatives. - 98 - ANNEX 3 Page 11 of 14 C. Future Electricity Supply Generation 29. YGEC is committed to two major generation projects. These comprise oil fired steam stations at Ras Katenib (165MW) and at Al Mukha (160MW). The Ras Katenib station is almost complete. Two units 33 MW wilL be commissioned in early 1983 and the remaining three by 1984. YGEC has recetntly signed a contract for the 4 x 4OMW station at Al Mukha. The contract reportedly envisages the first unit being commissioned in 1985. Based on the time taken to complete the Ras Katenib and other p)rojects, the mission believes that the first two of Al Mukha units are not likely to be in full operation before January 1987 and the remaining two before 1988. 30. Attachment 5 shows forecast electricity supply for the public system up to 1992. It has been assumed that all diesel units with a capacity less than 2.5 MW will be retired or removed from the interconnected system, although this policy needs to be coordinated with plans for rural electrification. Attachment 5 also describes further assumptions made concerning the expected capacity availaLble to meet peak demand after allowing for planned maintenance and plant breakdowns. On the basis of the demand forecast (Annex 2, Attachment 3) and the above assumptions, further generation is required in 1992. No decision on the type, fuel or size of this need be taken for several years. However, it is likely that a 50MW of combustion turbine capacity would be commissioned in 1992. In 1984, after Ras Katenib is commission, 60% of YGEC's installed capacity will be steam. This will rise to 74% in 1988 after Al Mukha is completed and some diesel units retired. In 1992, the proportion of steam plant would fall to 66% of total installed capacity if a gas turbine were installed. The decision on the next generation project will depend on the results of geothermal exploration (para. 2.06). 31. Attachment 5 shows that YGEC is likely to have considerable excess capacity for the rest of the decade. After allowing for scheduled maintenance and breakdowns, the margin of spare capacity will be above 10% in every year and as high as 40% when Ras Katenib and Al Mukha are brought into commercial operation. This is partly a consequence of the lumpiness of unit sizes needed to capture medium term economies of scale. However, it is also a consequence of ordering units in advance of need. This may have been just'ified if non-fungible soft credit were available for these projects, although any savings in capital cost need to be set against the earlier cormmitment to higher staffing costs. 32. High staffing costs are one of the main reasons why YGEC at present makes losses while having some of the highest electricity tariffs in the region. Staffing costs are high becauses the utility needs to employ expatriates because of the extreme shortage of Yemeni professional and technical manpower. Commissioning excess capacity at Ras Katenib and Al Mukha will exacerbate YGEC's financial problems, particularly since the stations will have to rely on expatriate staff, at least initially. This is now largely unavoidable, but nevertheless YGEC should continue to search for operating cost savings in its power stations and transmission system. Potential areas of saving might include placing temporarily surplus steam units in rotational storage, not fully manning steam stations with temporary - 99 - ANNEX 3 Page 12 of 14 surplus capacity use of mobile maintenance teams and unmanned operation of substations. It is recommended that YGEC examine such options for saving skilled manpower and costs and assign one of its advisors with the responsibility of identifying further operational savings. 33. A projection of generation and fuel consumption is given in Attachment 6. By 1985 steam plant is forecast to produce over 80% of electricity generated by YGEC. This proportion is likely to increase to above 85% for the remainder of the decade. There will be a corresponding change in YGEC's fuel consumption from gas oil to fuel oil. Fuel supplies 34. Fuel oil will be imported directly to the Ras Katenib and Al Mukha power stations. Each station will have an off-shore mooring buoy. Because of the shallow draught, only 5000 dwt tankers will be able to supply the power stations. Rough seas preclude unloading at Ras Katenib for three consecutive months of the year. With three month's storage capacity, there is a small risk of fuel shortage, if the station is operating at full load. 35. A pipeline has been proposed from the proposed oil terminal at Salif to Ras Katenib. This would enable YGEC to obtain a lower oil price from importing fuel in larger tankers. However, this project needs careful justification. YGEC is likely to be able to procure medium fuel oil (1500sec Redwood) at a price of US$200 per ton, (October 1982 prices) which is below the present C & F border price for small shipments of $219 per ton. A reduction of more than $10 from $200 per ton arising from larger shipments is unlikely. Furthermore, the Al Mukha power station, will come on stream before demand has grown to the point where Ras Klatenib will reach full output. Al Mukha will burn slightly cheaper 3500 sec oil and will have a higher thermal efficiency. It will therefore bes operated in preference to Ras Katenib so that Ras Katenib may not need to operate continuously at full load. The economic case for a fuel pipeline from Salif to Ras Katenib is therefore uncertain. It is recommended that YGEC carry out a feasibility study of the fuel pipeline from Salif to Ras Katenib before taking a decision to procede with it. 36. There is similar potential for a fuel pipeline from Al Mukha port to be the power station. Al Mukha port is limited to small ships, but 15,000 dwt tankers possibly could unload to a bouy in the harbor approaches. Again such a project would need careful justification. Transmission 37. Transmission lines have been constructed from Ras Katenib to Hodeidah, Sana'a and Taiz. The lines to Hodeidah and Sana'a will be brought into service in 1983 and the Taiz line in 1984. Delays in completing 132/33kV substations and 33/llkV distribution substations have delayed the commissioning of the interconnected system. 38. YGEC is committed to building a further 132kV line to connect Al Mukha to Taiz. A branch from this line at Al Barh is proposed to cross the - 100 - ANNEX 3 Page 13 of 14 Tihama to Bajil. The Tihama line is iintended to supply Zabid and other towns in the region. It will also improve the security of supply when demand grows in Sana'a and other highland load centers. Other 132kV lines are planned to Hajjah and Sana'a. It is doubtful whether the 132kV lines to Zabid, Hajjah and Sada'a are economically justified at present, especially when YGEC will have surplus diesel capacity available for local generation. The Tihama line could possibly be delayed until justified on the grounds of system power flows and security. Hajjah could be connected using the transmission line from a future steam station located at Salif, although such a station is unlikely before the early 1990's. Distribution and Rural Electrification 39. YGEC plans further reinforcement and extension of its urban distribution networks and to expand distribution facilities into rural areas. During the years 1982 to 1986 it plans to complete 830km of overhead MV line, 90km of LV line, 220km of MV cables and 78km of LV cables. The plan focuses on MV distribution which should lead to lower losses. However, 700km of the 830km of MV overhead lines is at 33kV and most of this will be to connect towns and villages to 132kV bulk supply points. Autoproducers and Private Power Systems 40. Most major industrial autoproclucers are likely to take public supply, providing suitable tariffs are offered by YGEC (see Annex 4, para. 33). New industries are also likely to take YGEC: supply unless thay are located in remote areas where connection to the grid is uneconomic. Existing generating plant will presumably be retained for standby when YGEC supply fails. Some of the larger installations, e.g. cement wrorks or the Taiz biscuit factory, could be used in future for peak lopping, or even to provide power to YGEC. 41. Many, if not most of the small (less than 5kW) units used to supply households, shops and workshops will require replacement during the next five years. These are likely to be replaced by similar units unless YGEC supply is provided in the village or private utilities or cooperatives are formed. Many of the private distribution systems are in poor condition and will require rehabilitation or replacement to maintain continuity of supply. 42. Some savings in national resources could be made if private distribution systems were constructed to an adequate standard to enable them to be integrated with the YGEC system; From the national viewpoint, improvement in the standard of private networks is probably economically justified in terms of losses and longer physical life, even if they remain as private systems or are not physically connected to the YGEC system. Moreover, there would be advantages from improved public safety if private systems met minimum safety standards. Private networks invariably operat,e at LV where the technology is simple. Savings would be made if private LV networks could be ultimately connected to the YGEC MV distribution system without being - 101 - ANNEX 3 Page 14 of 14 rebuilt or rehabilitated. Private producers are unlikely to construct LV networks to adequate standards unless first, it is financially worthwhile to them to do so; and second, they know whalt technical standards are acceptable. It is therefore recommended that MEW dralw up standard terms for YGEC to use in purchasing private networks supplying bulk electricity to them. These terms should link compensation to the condition, age and design of the network and provide incentives to the private produc,er to connect a high proportion of households. 43. Encouraging the construction of private networks to adequate standards is more difficult. Manpower and Government finance available for electrical supply that is not already committed is likely to be extremely scarce during the 1980's. Assistance to the private power subsector should not result in delaying the development of YGEC into a strong institution that is financially viable. YGEC appears likely to continue to be the major institution in the subsector, particularly since it is the only organization capable of capturing the economies of scale from central power generation and the fuel savings from steam plant. There are several ways that MEW could provide technical assistance to the private power subsector. Such measures include: a) providing drawings and technical handbooks to private suppliers, merchants and workshops; b) commissioning television and radio programs to explain how networks should be constructed. Other ministries use the mass media for similar purposes; c) persuade merchants to stock distribution equipment that meets YGEC's standards; d) use YGEC facilities for short training courses in LV distribution; and e) allow YGEC to sell limited specialized technical services, such as meter testing. In view of the importance of private suppliers in the subsector, it is recommended that MEW create a division responsible for private electricity affairs. This would formulate and implement measures to improve the efficiency of the subsector, monitor its performance and represent its interests within the Government. Initially, such a division might need to consist of an engineer, an administrator responsible for licensing and purchasing networks at least three techniician field assistants, all fluent in Arabic. - 102 ANNEX 3 Attachment 1 YEMEN ARAB REPUBLIC POWER SUBSECTOR REVIEW Yemen General Electricity Corporation Plant Capacity (1975-1982) (MW) 1975 1976 1977 1978 1979 1980 1981 1982 Sana'a 7 7 11 16.7 21.7 35.5 41.4 44.8 Hodeidah 3.3 3.3 4.7 7 11.5 11.5 25.9 25.9 Taiz 2.8 2.8 3.2 4.2 8.1 9.1 18.5 22.2 Dhamer 1 1 1 1 1.5 2.7 2.7 2.9 Hajjah 0.4 0.4 0.4 0.8 0.8 2.6 2.18 1.9 Ibb - - - 1.1 1.5 1.6 2.52 4.5 Thula 0.2 0.2 0.2 0.2 0.3 0.3 0.79 0.7 Shibam - - 0.3 0.3 0.3 0.3 1.1 1.1 Al-Mukha - - - 1.1 1.1 1 . 1.1 1.1 Qa'ataba - - 0.1 0.1 0.1 0.2 0.28 0.26 Al-Baydah - - - 1.3 1.3 1.3 1.2 1.1 Al-Marawiah - - - 0.1 0.1 0.4 0.56 0.56 Dawran Anis - - - 0.2 0.2 0.2 0.165 0.165 Sa'ada - - - 0.8 0.8 0.8 2.0 2.0 Yarim - - - 0.3 1.1 1.1 1.1 1.1 Khamer - - - - 1.1 1.1 1.1 1.1 Al-Mahwit - - - - 0.4 0.5 0.55 1.6 Hababah - - - 0.3 0.3 0.3 0.47 0.57 Rada'a - - - - 1.1 1.1 1.l 1.1 Maniakha - - - - 1.1 1.1 1.1 1.1 Kuhlan - - - 0.2 0.2 0.2 0.3 0.2 Al-Turbah - - - - - 0.5 0.96 0.53 Zabid - - - - - - 0.94 0.94 Abs - - - - - - 0.4 0.4 Harad - - - - - - 0.56 0.6 Midi - - - - - - 0.56 0.6 Raydah - - - - - - - 0.53 Khrif - - - - - - - 0.2 Kawkaban - - - - - - - 0.2 Ai Luhayah - - - - - - - 0.78 Az Zuhrah - - - - - - - 0.78 TOTAL 14.7 14.7 20.9 35.7 54.6 73.4 109.6 121.4 SOURCE: YGEC - 103 - ANNEX 3 Attachment 2 YEMEN ARAB REPUBLIC POWER SUBSECTOR REVIEW Balance of Capacity and Demand, YGEC Main Systems 1975-81 1975 1976 19,77 1979 1979 1980 1981 Sana'a Capacity 7 7 l.l 16.7 21.7 35.5 41.5 Maximum Demand 6.0 7.8 1.0.1 15.0 18.9 23.8 28.0 Margin (Z) LI 17 -9 9 11 15 49 48 Hodeidah Capacity 3.3 3.3 4.7 7.0 11.5 11.5 25.9 Maximum demand 2.7 2.7 3.7 4.6 5.7 8.0 10.0 Margin (X) 22 22 27 52 102 44 159 Taiz Capacity 2.8 2.8 3.2 4.2 8.1 9.1 18.5 Maximum demand 2.0 2.4 2.9 3.5 5.2 6.5 7.5 Margin (x) 40 17 l0 20 56 40 147 Dhamar Capacity 1.0 1.0 1.0 1.0 1.5 2.7 2.7 Maximum demand .. .. 0.76 0.9 1.3 1.3 1.8 Margin (x) 32 11 15 108 50 Hajjah Capacity 0.4 0.4 0.4 0.8 0.8 2.6 2.2 Maximum demand .. *- 0.35 0.35 0.35 0.4 0.5 Margin (1) 14 129 129 550 340 Ibb Capacity - - - 1.1 1.5 1.6 2.5 Maximum demand - - - 0.9 1.1 1.2 1.7 Margin (x) 22 36 33 47 SOURCE: YGEC 1/ Margin defined as the difference between capacity and demand, expressed as a X of demand. ANNE 3 - 104 - Attackment 3 YEMEN ARAB REPUBLIC Power Subsector Review Size and Age Distribution of YGEC Generatin& Plant 324 28- 24- . 3 *-W 20 , 44 Figure 1 0 Size Distribution of YGEC X 16 , Generating Plant 12 , _ 100 1W 500 SW Intervals Intervals 4- ol .5 li 200 400 600 Boo 1000 1200 1400 2000 3000 4000 5000 6000 Unit Capacity (MW) Figure 2 24 Age Distribution of YGEC Generating Plant 20 X AU AAIAT,ow n YEMEN ARAB REPUBLIC 0 9 UNEC Arb g ,t} g Ns.a-POWER SUBSECTOR REVIEW *JJDA~~N> / a UEN ~ p/ " (\ \ V/. ~*, . 0 r S \ EAM APOIRI P ANT Pl SONPE On - EThIUE AP ; nD = - T ~) t t 0 t ~ i Dl SIELIT P0ER STAT ANSEXISTING 'P 0 if < C 3 1 ThANSPOIMER S'ATIOUS FAIXTINS Iota,, 11" N¸0 HIN T=9ANSMISSION L NESA INTING *ITJ r * -' .: C f 1 11 1 t a XI Hor INTERNA1 ONAL CAI TAL 0 ;3 r :- t-?t , 8 la -- X at l v-' o~~~~~~~~~~~~~~ 20 40 60 an io 1-600 q )J ;,/ r>/ )M|L 16' I: 0'0. _0 TOWNS .5.. ,:Y i; S :... .,,.,..,--- A .M., - - - - j C. N ~Z WASIl I' I~~~~~~~~~~~~~ E THI 0 PIA A H INNAl AAL BA,SIEr D K I j A TO AT T A ( I N S j I A A OD L b o b f '. . 4 ,~ ~ ~ ~~ ~~ 0 N ' eS VCIETiS ta E OPIAT I. AS.X ..< A~MoroyP- OA t,N ,OM ./ D J I BO UTI X X X : z EE~~~~~~~~~AotA'LIU /~~~~~ ,o _ \pa,man. AP