Nepal - Marsyangdi Hydroelectric Power Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 4422A-NEP NEPAL MARSYANGDI HYDROELECTRIC POWER PROJECT STAFF APPRAISAL REPORT MAY 1, 1984 South Asia Projects Department Power and Transportation Division 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 1 Nepalese Rupee (NRs) = US$ .0625 1 Nepalese Rupee = 100 Nqepalese Paise 16.0 Nepalese Rupees = US$1,.00 HMG AND MHDB Fiscal Year (FY) Beginning July 16 and Ending July 15 WEIGHTS AND MEASURES kW = Kilowatt (=1.341 horse power) MW = Megawatt (thousand kilowatts) kWh = Kilowatt hours (=860.42 K cals) MWh = Megawatt hours (thousand kilowatt hours) GWh = Gigawatt hours (million kilowatt hours) kV = Kilovolts (thousand volts) MVA = Megavolt-ampere (thousand kilovolt-ampere) KOE = Kilogram oil equivalent (ten thousand K cals) TOE = Tons oil equivalent (thousand kilogram oil equivalent) km = Kilometer (0.6214 mile) m = Meters (3.2808 feet) cu m/sec = Cubic Meters Per Sec (61022 cu in/sec) Mm = Million Cubic Meters (35.3147 million cubic feet) PRINCIPAL ABBREVIATONS AND ACRONYMS USED ADB - Asian Development Bank CIDA - Canadian International Development Agency CY - Calendar Year ED - Electricity Department EEC - European Economic Communilty FY - Fiscal Year GTZ - Deutsche Gesselschaft fuer Technische Zusammenarbeit GmbH HMG - His Majesty's Government IRR - Internal Rate of Return KFAED - Kuwait Fund for Arab Economic Development KfW - Kreditanstalt fuer Wiederaufbau LI - Lahmeyer International MASL - Meter Above Sea Level MHDB - Marsyangdi Hydroelectric Development Board NEC - Nepal Electricity Corporat:ion OECF - Overseas Economic Cooperation Fund (Japan) OPEC - Organization of Petroleum Exporting Countries SATA - Swiss Association of Techrical Assistance SFD - Saudi Fund for Development SHDB - Small Hydro Development Board SCF - Standard Conversion Factor SMEC - Snowy Mountains Engineering Corporation UNDP - United Nations Development Program  FOR OMCIAL USE ONLY NEPAL MARSYANGDI HYDROELECTRIC POWER PROJECT STAFF APPRAISAL REPORT Table of Contents Page No. CREDIT AND PROJECT SUMMARY ................................... iv-v I. THE ENERGY SECTOR ...............* 1.............. ....... 1 Background ............................1................ Energy Resources ........ ............................. . 1 Hydro Resources .......... ............................. 2 Small Hydro Schemes ................................... 2 Petroleum ................................... 3 Coal .................................................. 3 A. The Power Subsector ................................... 4 Evolution and Present Structure .................... O.. 4 Planning in the Power Sector ................. i........ 5 Previous Bank Group Involvement in the Power Sector ... 6 The Nepal Power System ............. 6 Status of Supply and Access to Electricity .... ........ 7 Rural Electrification . ......................... ....... . 8 Power Exchange with India ..... ........................ 8 Load Forecast .............. 9 Demand Forecast and Capacity Margin ................... 10 Maintenance of Plant ........... .. .. 10 System Losses .................. ..... 0........ *0.... ............ 11 II. EXECUTING AGENCY AND THE BENEFICIARY ...................... 12 A. Executing Agency ................. . .................... 12 B The Beneficiary ...... .............................. .. 13 Audit ................................................. 14 Government Arrears ............................ ........ 15 Dividends .. ............................. .............. 15 Insurance ............................................ 15 Training .............................................. 16 This report is based on information obtained during an appraisal mission in November 1982 by Messrs. J.M. Vance, M.P. Manrai and Ms. L. Villa, and on updated information provided by the Electricity Department of HMG and the Nepal Electricity Corporation. 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.  - ii- Page No. III. THE PROJECT . ... .......................................... 16 A. Project Description and Objectives ..................... 16 Project Components ........ .. .......................... 17 Project Objectives .. .............. ... ................. . 18 B. Background of Project Investigation and Formulation ... 18 C. Cost Estimates ............. ..6.0 .... 60.... 19 D. Project Financing ........ o... ..... ... ....... 22 E. Procurement o......*.**.**...*.*..*.... * . .............. ** ..... ... 23 Contract Review .... .............. . . ......... .. .. ...... 25 Fo Disbursement ..... o........ ............o ... 25 G. Implementation * .......... ..... #........**...Os...s#.... 26 Engineering Services ........ . . . . . . . . . . . . . ...........o..o .. . 26 Ecology and Resettlement o . ...................- 27 Local Distrtibution in the Kathmandu Valley ........... 28 Training Facility ... ......... ............... 28 Risks ..... o_ ..... o_ ...........o ........ 30 Completion Report ...... * ....... ...... .. -..-.-.....o... . 30 IV. FINANCES ....... ............ __ ................... 30 Tariffs ........ o ..............-....... .-.-...... 31 Earnings Covenant .. ............. ... .... .. * ... ....... 31 Revaluation of Assets .. .........._ _ .....o ............. . 33 Past Operations .... *.. .. ...... ... . . . . .. . .. ........... .... .. . 33 Financing Plan .... .* ...... ................... .......... 35 Future Finances ... .. ....... .. *. . .. . . .............._... ...... . .... ... . 37 V. JUSTIFICATION .. ..... * . .... ................... ,. . .. 38 Least Cost-Solution ... . .... ............ . ....., .. 39 Cost/Benefit Analysis ........... ...... 42 VI. AGREEMENTS REACHED AND RECOMMENDATION ..o. .................., 43  -i ii- ANNEXES 1. Prospective Hydro Sites 2. Small Hydro Projects 3. Generation Capacity 4. Transmission Lines and Distribution Lines 5. Grid Sub-Stations and Local Distribution S'ub-stations 6. Past Sales and Generation 7. Captive Plants 8. Power Exchange with India 9. Load Forecast - Assumptions TABLE I - Load Forecast TABLE 2 - New Industrial Loads TABLE 3 - New Irrigation Loads 10. Maximum Demand and Installed Capacity 11. Project General Layout 12 Project Details Table 1 - Important Basic Data 13. Hydrology and Sedimentation 14. Geology and Seismicity 15. Project Cost Estimate Table 1 - Project Cost Summary Table 2 - Summary Account By Time Table 3 - Detailed Cost Table Table 4 - Detailed Cost Table - Civil Works Table 5 - Summary Account By Project Component 16. Implementation Schedule 17. Training Facility - Cost Estimate 18. Summary of Tariffs 19. Actual and Forecast Income Statements (1978/79 - 1990/91) 20. Actual and Forecast Balance Sheets (1978/79 - 1990/91) 21. Forecast Sources and Application of Funds Statements (1983/84 - 1990/91) 22. Notes and Assumptions for Financial Forecasts 23. Economic Comparison Table 1 - Capacity Balance Table 2 - Energy Balance Table 3 - Assumptions for Comparison Table 4 - Energy Availability Table 5 - Cash Flow Statement and Present Values 24. Internal Rate of Return 25. Data and Documents Available in the Project File CHARTS 1. Proposed Organization Chart of New Electricity Authority 2. Organization Chart of Marsyangdi Hydro Development Board MAP 1. IBRD Map 16928R   -iv- NEPAL MARSYANGDI HYDROELECTRIC POWER PROJECT CREDIT AND PROJECT SUMMARY Borrower: Kingdom of Nepal Beneficiary: Nepal Electricity Corporation Amount: SDR 100.6 million (US$107.0 million equivalent) Terms: Standard Terms to Beneficiary: 30 years with interest at 12 percent per annum Project The project seeks to meet the forecasted demand for Description: electricity in Nepal for the medium term and to strengthen the Borrower's power sector. This would be a run-of-river hydroelectric power project, with an installed capacity of 69 MW, located on the Marsyangdi River. Maior project components include diversion works and diversion weir, headrace and tailrace tunnels, powerhouse, transmission lines, substations and equipment. The project also provides for consulting engineers, technical assistance and training. A primary benefit would be derived from the generation of additional electricity for industrial, commercial and domestic consumption, and for irrigation purposes. Benefit would also be received through improvement of the distribution system, enhancement of maintenance capacity and the training of manpower. Some risk relates to possible deficient project design and construction, poor management and cost overruns. Appointment of a panel of experts to review design and construction, adequate provision for supervision in close cooperation with the consulting engineer, and assignment of a claims advisor should help minimize these risks.  _v Estimated Costs: ------US$ millions------ Item Foreign Local Total 1. Preliminary Cost 2.5 2.9 5.4 2. Administrative Expenditure 0.5 2.7 3.2 3. Civil Works (Lot I) 45.0 13.6 58.6 4. Civil Works (Lot II) 73.7 25.0 98.7 5. Equipment 40.8 6.9 47.7 6. Substations and Local Distribution System 8.4 2.2 10.6 7. Transmission Lines 5.5 2.8 8.3 8. Consulting Engineer 7.5 - 7.5 9. Technical Assistance 4.1 - 4.1 Total Base Costs 188.0 56.1 244.1 Physical Contingencies 25.8 7.8 33.6 Price Contingencies 34.6 11.0 45.6 TOTAL 248.4 74.9 323.3 a/ Financing Plan: -----US$ millions----- Foreign Local Total IDA 107.0 - 107.0 Saudi Fund b/ 25.0 - 25.0 Kuwait Fund b/ 21.0 - 21.0 KfW b/ 69.3 5.2 74.5 Government 26.1 69.7 95.8 Total 248.4 74.9 323.3 Estimated Disbursements: -----------------US$ millions-------------- Bank FY FY85 FY86 FY87 FY88 FY89 FY90 Annual 14.7 26.0 26.0 19.2 10.2 10.9 Cumulative 14.7 40.7 66.7 85.9 96.1 107.0 Staff Appraisal Report: Report No. 4422a-NEP Rate of Return: 5.9% Map: IBRD 16928R a! Includes taxes and duties of US$15.81 million equivalent. b/ Amounts are expressed in US dollar equivalents only for ease of comparison.  NEPAL MARSYANGDI HYDROELECTRIC POWER PROJECT STAFF APPRAISAL REPORT I. THE ENERGY SECTOR Background 1.01 The two principal known sources of energy in Nepal are forests and extensive river systems. Though exploration for fossil fuel is continuing, thus far there have been no discoveries. Nepal realizes that unless its water resources are exploited for energy production, the depletion of its forests, which have supplied a major portion of its energy needs, will con- tinue. One of the main obiectives, therefore, outlined in the Sixth Plan (1980-85) is the development of hydro potential. 1.02 There are three ministries dealing with the energy sector in Nepal: the Ministry of Water Resources, the Ministry of Commerce and Industry, and the Ministry of Forests. The Electricity Department of the Ministry of Water Resources is responsible for power development; the Ministry of Commerce and Industry is responsible for oil, gas and coal exploration efforts through its Department of Mines and Geology; and the Ministry of Forests looks after forest resources. Realizing the importance of water resources, a Water and Energy Commission was created in 1976 under the Ministry of Water Resources to investigate and prepare plans for the development of water and energy resources. Energy Resources 1.03 Of total energy consumed in FY83, 5.5% was from commercial sources consisting of petroleum (3.8%), coal (1.1%) and electric power (0.6%). Non-commercial sources provided the remaining 94.5%, with fuelwood being the primary source (92.2%). Other non-commercial sources such as charcoal, crop and animal waste accounted for the remaining 2.3%. The volume of commercial energy consumption is very small. In FY83 it totalled about 196,170 TOE. Per capita consumption is 9.8 KOE; this compares to 33.4 KOE in Bangladesh, 424.9 KOE in China and 144.2 KOE in India. 1/ 1.04 During FY71-83, the annual growth rate of commercial energy consump- tion was 5.2%, compared to 2.7% for total energy. It is, however, interest- ing to note that the growth rate of electricity sales during the same period was about 15.5%. 1/ The comparison is on 1980 basis and is based on 1983 World Bank Publica- tion titled "The Energy Transition in Developing Countries."  -2- Hydro Resources 1.05 Nepal is endowed with vast hydro resources. The estimated theoreti- cal hydroelectric potential is about 83,000 MW, of which about 25,000 MW has been investigated for development. This potential is estimated to be dis- tributed in various river systems of Nepal as follows: Sapt Kosi basin in Eastern Region /a /b : 22,000 MW Sapt Gandaki basin in Central and Western Regions : 21,000 MW Karnali and Mahakali basins in Far Western Region : 36,000 MW /a Master Plan of Hydroelectric Poweir Development in Nepal - Japan International Cooperation Algency, September 1974. /b Nepal is administratively divided into 5 regions - Central, Eastern, Western, Mid Western and Far Western. Another 4,000 MW has been identified in other rivers, including the Kankai, Kamla, West Rapti, Bagmati and Babai. A list of prospective sites is attached as Annex 1. 1.06 Some hydro projects would be of such size that export to the Indian market would be required to economically justify them. A project in which Nepal and India have shown common interest is the Karnali (Chisapani multi- purpose) project. A joint committee of representatives of the two countries called the Karnali Committee has been formed to approve, inter alia, terms of reference and scope of such prolect and the short list of consultants to carry out a study of which the main objective is to assess the economic justification and viability. The study is to be financed by the Bank Group under a technical assistance credit to Nepal approved by the Board in March 1984. Saldro Schmes 1.07 Nepal is faced with a difficult situation. Rugged terrain, difficult access, massive deforestation, high costs associated with large civil works and long transmission lines have forced HMG to focus its attention on many attractive sites where small hydros may be installed. Nepal has embarked on a program for developing these sites so as to supply electricity to isolated towns and villages. The size of these small generating units varies from 25 kW to 500 kW. So far 11 power stations with a total installed capacity of 2,179 kW have been commissioned, another 21 stations with an installed capacity of 9,750 kW are under construction, and 16 schemes with a potential of 1,786 kW are under investigation. A list of these projects is in Annex 2. Although many projects referred to have been investigated or are under various stages of implementation, the costs of investigation, construction and maintenance are high. Under financing by USAID, HMG has now engaged the services of the National Rural Electrification Cooperative Association (NRECA) of USA to advise it on such matters as possible steps in reducing the capital and maintenance cost, use of indigenous material and manpower as far as possible, use of cost effective methodology for collecting stream flow and rainfall data for the site and improvement in the load factor so as to make each project viable, etc. With the help of consultants, HMG hopes that small hydros will play a significant role in the future power development of Nepal.  -3- Most of these small hydro stations are presently being constructed and operated by the Small Hydro Development Board (SHDB) established in 1976. 1.08 There is appreciable interest by private individuals in using water power for milling and sawing purposes, and HMG's policy supports this type of activity. About 200 locally manufactured turbines 1/ varying from 4.5 to 30.0 horse power have been installed by various individuals. These installa- tions, unlike small hydros are cheaper and are running successfully because the spare parts of the locally manufactured equipment are easily available. Petroleum 1.09 To date neither oil nor gas reserves have been discovered. In 1980, an IDA-financed airborne survey of the Terai area was carried out, which indicated evidence of geological structures that might be suitable for trapping oil. In June 1982, IDA agreed to finance a seismic survey and geomechanical and geological studies in the areas of interest with a credit of US$9.7 million. The total cost of the studies is estimated at US$10.9 million. The completion date of this survey project is June 30, 1985. 1.10 The Nepal Oil Corporation is responsible for importing and distribut- ing petroleum products throughout Nepal. During the last ten years (FY73-83), consumption grew at an average annual rate of 5.5%, totalling 147,580 metric tons in FY83. Projections for the next 20 years show an average annual growth rate of about 7%. High-speed diesel, used mostly in the transport sector, accounts for the major share, about 42% of total imports, followed by kerosene, used for cooking and lighting (24%), and aviation fuel (15%). The remaining 19% consists mainly of gasoline, low-speed diesel and furnace oil. The retail and border prices of selected petroleum products (December 1983) were as follows: Retail Price Border Price Gasoline 877 per ton 467 per ton Kerosene 424 per ton 359 per ton Diesel 458 per ton 344 per ton Coal 1.11 There are no known coal resources. Consumption of coal, which is estimated at about 100,000 tons per year, is met entirely by import from India. 1/ These turbines are of a cross-flow type. The local manufacturers, under the guidance of United Mission to Nepal (UMN) and Swiss Association of Technical Assistance (SATA), have developed designs suitable for heads from 2.5 meters to 45.0 meters and rated discharge from 20 litres/second to 2,000 litres/second.  -4- A. The Power Subsector Evolution and Present Structure 1 12 Electricity supply was introduced in Nepal in 1911 with a 500 kW hydro power station biilt to supply a few selected consumers in Kathmandu. It was not until 1934 that this supply was augmented with a 640 kW hydro station, followed by a 1,700 kW diesel generating station 20 years later. Thle management of this rudimentary system was expatriate. As the supply capacity started to grow, the Government took over responsibility for public electricity supply. 1.13 The Nepal Electricity Corporation (NEC) was established in 1962 under the Nepal Electricity Corporation Act of 19652. It was established to enable the management of the important power systemn in northern part of the Central Region to be freed from the constraints characteristic of the government bureaucracy. Later, NEC also took over the operations in the remaining part of the Central Region and in some parts of ithe Western Region. The opera- tions throughout the rest of Nepal, plus the major part of planning, design and construction were left to the Electricilty Department of the Ministry of Water Resources. In 1974, the Eastern Elect:ricity Corporation (EEC) was established to manage the Eastern Region. A more recent development has been the establishment of separate development boards, each with its own chairman and board members, to be responsible for the construction of larger projects. 1.14 The following bodies have been directly involved in operating the power sector: (a) Electricity Department (ED) of the Ministry of Water Resources; (b) Nepal Electricity Corporation (NEC); (c) Eastern Electricity Corporation (EEC) (merged with NEC in June 1982); (d) Butwal Power Company (BPC) (merged with NEC in 1981); and (e) Several development boards, including those for the Kulekhani I and Marsyangdi hydroelectric projects, transmission line projects and the small hydro projects. 1.15 ED is a department within the Ministry of Water Resources. It is responsible for planning, designing and constructing new projects which, after commissioning, are handed over for operation to NEC. In the Mid and Far Western Regions, the projects are retained and operated by the ED. In practice the larger construction works are executed by autonomous, develop- ment boards set up under statute for this purpose (para 1.18). 1.16 The ED is headed by a Chief Engineer, who is appointed by a Cabinet Committee of HMG; he is directly accountable for the work of his Department to the Secretary of the Ministry. There are currently some 900 staff, including 200 engineers of which about 150 are on secondment to other organizations in the power sector. All of its revenues are transferred to  -5- the Ministry of Finance, while funds for capital and operating expenditures are obtained through the annual budget. 1.17 NEC is a government-owned corporation with a staff of about 3,200 including about 100 engineers. It is the largest entity in the power sector, responsible for generation, transmission and distribution of electricity in the country. NEC operates and maintains power stations, transmission lines and other facilities transferred to its ownership by the ED and the develop- ment boards. It finances, designs and constructs modest reinforcements to its distribution network and extends supplies to new customers. It is also responsible for billing and collection of revenue. NEC has little contact with the ED or the development boards during planning, design and construc- tion of new works even though the facilities are subsequently transferred to it for operation and maintenance. 1.18 The development boards are established under the Development Boards Act to execute large construction projects, which upon commissioning are handed over to other entities for operation and maintenance. Board Members, in the case of the Kulekhani I and the proposed Marsyangdi projects, include the Minister of Water Resources as Chairman, and civil servants from various ministries, including the Secretary, Ministry of Water Resources and the Chief Engineer, ED. The Project Manager is a senior engineer seconded from the ED. He is paid by the Board, is accountable to the Board and is also a member and Secretary of the Board. The boards are staffed by engineers on secondment from the ED and by others specially selected. A few engineers are attached to the staff of the consultants and contractors for the projects. Considering the early stage of power development in Nepal, and the lack of experience by the Nepalese with large projects, the Boards have relied heavily on consultants to supervise execution of the projects. 1.19 It is clear from experience that this multiplicity of bodies to operate the Nepal electricity system has not been satisfactory. HMG recog- nizes that in order to meet the management and technical tasks associated with the planned expansion and operation of the system, the power sector's institutional structure needed streamlining. Therefore, with the assistance of consultants, HMG has decided to reorganize the power sector into one entity (see paras 2.05-2.08 for a discussion of the reorganization.) Planning in the Power Sector 1.20 Planning in the power sector is the responsibility of the ED and the Water and Energy Commission. The ED is responsible for system studies, load forecasting, assessing the new requirements for generation schemes, and transmission and distribution systems, and also planning for rural elec- trification. The Water and Energy Commission investigates new hydro resources for development of power. 1.21 The two agencies are assisted in their power planning function by a Canadian International Development Agency (CIDA) team of about twelve profes- sionals, which has been domiciled in Nepal for the last six years. 1.22 In 1974, under Japanese technical assistance, Nippon Koei carried out a comprehensive Master Plan for hydroelectric power development. In 1979, the Gandaki river basin was studied in detail by Snowy Mountains Engineering  -6- Corporation of Australia under technical assistance from UNDP. Other sites have also been studied by various consulting engineers (Annex 1). Recently arrangements have been made for Japan Interniational Cooperation Agency to finance the reconnaissance of the Kosi basin to identify projects. In addi- tion, under the finances provided by USAID, NRECA of USA is studying the role of small hydros in the power sector (para 1.07). The Water and Energy Com- mission intends to carry out detailed basin studies of other major rivers with funds provided by various donors. At present the sites for which feasibility/prefeasibility studies are avaiLable are limited. The new studies will increase the choice and will enable the Government to carry out a ranking study of the country's potential power projects on a wider basis. 1.23 During negotiations, an agreement was reached that reports of such studies would be made available to IDA. Thereafter, views will be exchanged between IDA and the Government to identify a future power development program in Nepal. Previous Bank Group Involvement in the Power Sector 1.24 Bank Group involvement in the power sector has been in respect of the 60 NW Kulekhani Hydroelectric Power Project (Credit 600-NEP and Sup- plemental Credit 600-1-NEP, totalling US$40.8 million). This project, cost- ing about US$120 million equivalent, was commissioned in April 1982, and to date is the largest project ever undertaken in Nepal. It approximately doubled the total generating capacity of the country. The financing was arranged through a group of colenders consisting of IDA, OECF, Kuwait Fund, OPEC, EEC, and UNDP. The original cost of the project was estimated at US$68.0 million. The reasons for cost incretases were (a) increased cost of civil works, (b) imposing of local taxes by the Government at a later stage, (c) increase in the consulting engineers cost due to increase in manmonths required for supervision, (d) appreciation cf the Japanese Yen against the US dollar, and (e) increase in the scope of substations and transmission lines. As a result of the lessons learned during the execution of the Kulekhani Project, adequate steps have been taken to prevent reoccurrence. A panel of experts has been appointed in the early design stages of the project, neces- sary site investigation has been done and adequate physical contingencies have been provided. The Nepal Power System 1.25 The strategy of Government has been to build hydro and diesel power stations mainly in the Central Region because of the concentration of major loads in the Kathmandu valley. The present total firm plant capacity in Nepal is 133 MW, of which 115 MW is in the Central Region. Of the total 133 MW, hydro plant capacity accounts for 111 MW, and diesel plant for 22 MW, with the largest hydro power station (60 MW) located at Kulekhani and the largest diesel power station (16 MW) located at Hetauda (Annex 3). All hydro plants are run-of-river types except Kulekhani which has a live storage capacity of 73.3 million cubic meters, a catchment area of 126 sq km and a rated head of 550 meters. During the wet season Kulekhani is operated only during peak time. (For month wise energy output of various stations refer to Table 4 of Annex 23.) In the event of a shortage of energy caused by abnor- mally low river flows, NEC and ED rely on the regulation of stored water in Kulekhani reservoir and on full use of diesel plants.  -7- 1.26 The grid system presently is confined to the Central and part of the Western Regions and consists of 239 km of 132 kV lines and 227 km of 66 kV lines. The firm plant capacity in the grid is 130 MW, of which 111 MW is hydro. The maximum demand of the grid system was 69.1 MW in FY83, compared to the total Nepal demand of 83.2 MW (Annex 6). 1.27 The Eastern Region registered a maximum demand of 11.0 MW in FY83. This region does not form part of the grid and depends heavily on energy import from India. For example, in FY83 only 6% of the energy was generated within the region, with the remaining 94% being imported from India. The isolated network in this area consists of 188 km of 33 kV and 76 km of 11 kV lines connected to small hydros, diesels and to supply points from India. 1.28 The Mid and Far Western Regions, which registered a combined demand of 3.1 MW in FY83, are managed by the ED. This isolated network consists of 85 km of 33 kV and 52 km of 11 kV lines, also connected to small hydros, diesels and to supply points from India. 1.29 Extension of the grid to interconnect the Eastern Region through a 283 km 132 kV line from Hetauda to Biratnagar via Janakpur is being financed by ADB (completion in FY86). There are also plans to extend the grid westward by erecting 45 km of 132 kV line from Dumkibas to Butwal under French assistance, and 225 km of 132 kV line from Butwal to Nepalguni via Shivpur and Lamahi under an ADB loan. This interconnection is likely to be commissioned in FY87 (IBRD Drawing 16928R). Details of existing and proposed transmission lines and substations in various regions are shown in Annexes 4 and 5. Status of Supply and Access to Electricity 1.30 The electric energy consumption in Nepal is characterized by seasonal variations. Due to increased heating and lighting requirements in winter, the average monthly consumption in the Central Region, where domestic load is predominant, is more in winter than in summer. The overall annual load factor in Nepal is low at about 45% (Annex 6). Of total annual consumption, about 55% is consumed during the winter months of December through May and about 45% during the summer months of June through November. The maximum demand recorded in summer is about 80% of the maximum demand recorded in winter. 1.31 The total energy generated in Nepal in FY83 was 292 GWh (Annex 6), of which 279 GWh was hydro, 5 GWh diesel and 8 GWh was generated from captive plant (Annex 7); in addition, 63 GWh was imported from India. About 98% of hydro and 42% of diesel energy was generated in the Central Region. Energy sales within Nepal totalled 232 GWh and export to India was 6 GWh. The Central Region accounted for the largest amount of sales (72%) followed by the Eastern Region (15%), the Western Region (8%), and the Mid and Far West- ern Regions (5%). Of the total energy imported from India, 68% was imported into the Eastern Region (Annex 6). 1.32 The grid system in the Central and Western Regions has been beset with systematic load shedding since FY78. During the winter of 1981, when load shedding was at its worst, about 14 MW and 2 MW of load was shed during  -8- peak time in the Central and Western Regions, respectively. It is estimated that the unserved detand for energy in FY82 was about 15 GWh, equivalent to 10% of the comlined - .nergy sold in the Cenl:ral and Western Regions. Since the commissioning of Kulekhani I in April 1982, there has been no necessity for shedding load. 1233 Applications for new connections have been accumulating since load shedding was introduced in 1978. At present, about 20,000 applications are pending, out of which 10,000 are in the Western Region and 8,000 in Central Region. All but 300 applications are for new domestic connections. Plans call for connecting about 6700 new domestic consumers annually beginning in FY84, increasing to 10,000 in FY92 (Annex 9). 1.34 At the end of FY83, the number of consumers totalled 133,672 of which 130,056 were domestic. Based on a total population of 15.4 (1982) million and assuming six persons per household, 5.0% of the population are served by electricity. Per capita production of electricity is about 16 kWh; this compares to 164 kWh in India, 26 kWh in Bangladesh and 307 kWh in China. 1/ Rural Electrification 1.35 The five administrative regions of Nepal are further divided into 14 zones and 75 districts. Each district has a headquarters. The towns and villages in the districts are grouped together and each group is either a town panchayat or a village panchayat. The total number of town panchayats is 29 and village panchayats is 2,905. To date 39 district headquarters (52%), 28 town panchayats (97%) and 137 village panchayats (5%) have been electrified. Out of a population of 15.4 million, 1.77 million are living in the electrified towns and villages. In the next 5 years, HMG plans to elec- trify an additional 21 district headquarters, one town panchayat and 241 village panchayats. On this basis, a total of 3.6 million people would be living in electrified towns and villages. ADB has shown an interest in rural electrification and HMG intends to carry out its program, mainly with ADB assistance. Power Exchange with India 1.36 Over the years, the import of power from India has been increasing. In FY83 about 63 GWh was imported, equivalent to about 22% of energy gener- ated in Nepal. The power exchange began in April 1954 when India and Nepal signed an agreement 2/ to supply up to 6.8 MW of power to Nepal for use in the Biratnagar area in the Eastern Region. In October 1970, another agree- ment was signed for a two-way exchange of power of up to 5,000 kW along the border. Recently, in April 1982, a third agreement was signed for Nepal to export upto 25 MVA (approximately 20 MW) of secondary power to India at Raxaul and Gandak and to import about 18 MVA from India for consumption in 1/ The comparison is on 1980 basis and is based on 1983 World Bank Publica- tion titled "The Energy Transition in Developing Countries." 2/ Kosi Agreement of April 1954 amended in December 1966.  -9- the Eastern, Mid-Western and Far-Western Regions. The actual exchange of power during FY76-83 at various points is shown in Annex 8. Load Forecast 1.37 The load growth has been hampered because (a) the existing hydro generating power stations have not been fully utilized due to operational problems (para 1.45), and (b) there is no interconnected grid system for the whole country. In spite of these constraints, the total sales in Nepal grew at an average annual rate of 18.2% during FY71-78 and 11.7% during FY76-83. From FY78 onward, load shedding was introduced and growth rates are therefore distorted. The growth rates during FY71-78 were higher because there were no restrictions and the system was developing from a small base. 1.38 A load forecast was prepared in 1981 by the Water and Energy Commis- sion, assisted by the consultants. This forecast was based on an estimate of the sales requirement for each major category of load in each region. The generation requirement was then worked out by aggregating the sales and system losses in each region. The maximum demand of each region and of the system was calculated by assuming certain load and diversity factors. Other methods, such as extrapolation of historical trends and the use of economic indicators, were used in the above forecast as checks to verify the validity of the load forecast. During appraisal, this load forecast was reviewed and updated through FY92 in consultation with the ED and NEC. Overall, the demand is expected to increase by an average of 12.8% per annum. For details see Annex 9. A summary discussion of the load forecast in major areas is given below. 1.39 Industrial. Most existing industries are small scale, involved with the processing of agricultural commodities and to some degree also with textiles, leather and wood. However, present plans are to set up larger scale industrial plants for cement, steel foundry, paper, electrical acces- sories under joint enterprise with countries in the Region. In view of this, an appreciable increase in industrial load is expected (Table 2 of Annex 9). 1.40 It is expected that the industrial load would grow at an average annual rate of 15.9% during FY83-92, compared to the past growth rate of 25.4% during FY71-78 and 12.6% during FY76-83. It is expected that in FY92 the share of industrial load would be 41% of total sales in Nepal compared to 32% in FY83. 1.41 Domestic. Keeping in view the past trend and considering that the local distribution system reinforcement is already taking place in the Central and Eastern Regions (para 1.50), it is expected that a steady growth in domestic load would continue. An average annual growth rate of 6.1% during FY83-92 is expected, compared to the past growth rate of 16.2% during FY71-78 and 9.8% during FY76-83. It is estimated that the share of domestic sales would be 30% of total sales in FY92 compared to 52% in FY83. 1.42 Commercial. At present the commercial load is not being properly classified; only non-industrial load in excess of 50 kW is classified as commercial. Small hotels, lodges, offices, shops and other commercial estab- lishments having a load of 50 kW or less are classified as domestic users.  -10- NEC is in the process of correcting this. It is estimated that the commer- cial load would grow at an average annual rate of 14.8% during FY83-92 com- pared to the past growth rate of 16.1% during FY71-78 and 18.5% during FY76-83, It is expected that the share of commercial sales would be 11.0% in FY92 compared to 10.0% in FY83. 1.43 Irrigation and Water Supply. At present only three irrigation projects, Batter (lift irrigation), Birganj and Lumbini (tubewells) are in operation in the Central and Western Regions. HMG has embarked on a number of lift irrigation and groundwater pumped irrigation schemes. Eight of these projects, for which finance has been arranged, have been included in the load forecast (Table 3 of Annex 9). For other prospective irrigation and water supp7ly loads, a 5% annual growth rate has been assumed. It is expected that in FY92, the share of irrigation and water supply sales would be 16.0%, compared to 3.0% in FY83. Demand Forecast and Capacity Margin 1e44 As stated in para 1.38, the maximum demand is calculated by assuming certain annual load factors. The present load factor is about 47.5%. It is assumed that it would improve to 50% in FY86 as new tariffs are expected to levy extra demand charges. The system maximum demand, proposed installed capacity and available capacity margin is shown year-wise in Annex 10. Except for the winter of 1988 one largest unit (30 MW) would be available as rese-ve. Maintenance of major hydro units is planned during the summer when maximum demand is low and water flows are high. Additionally those units in a power station which cannot be fully utilized during the winter could be servi-ced during that time. Maintenance of Plant 1345 Until the recent commissioning of Kulekhani I, which alleviated the shortage of generating capacity, NEC has been postponing malor repairs and maintenance of its hydro power stations. A typical example is the Trisuli Power Station, manufacturers of whose equipment have recommended the closing of the power station for immediate repairs and proper maintenance. Further problems at Trisuli include an outage of one unit for the last two years due to a burnt-out stator, and a reduction in the station's daily peaking capacity because the pondage has been filled with silt. NEC has been aware of these problems and of the urgent need to carry out plant maintenance. Realizing that this maintenance is essential for the system to have a proper capacity margin, it has recently ordered essential spares. Also, for routine maintenance of civil works at hydro power stations, the Electricity Depart- ment has purchased maintenance equipment with funds provided by OECF. Main- tenance programs are essential and to ensure that this work is carried out, an agreement was reached that a plant maintenance program for rehabilitating the existing hydro plants, satisfactory to :DA, would be furnished to IDA by March 31, 1985 and thereafter taking into account the Association's com- ments, if any, the said program would be implemented.  -11- System Losses 1.46 System losses in Nepal are high. In FY83 total Nepal losses 1/ amounted to 30.5% of total generation. Following is a breakdown of losses by region: Central Region - 31.8% Western Region - 31.0% Eastern Region - 21.3% Mid & Far Western Regions - 19.0% An analysis of losses in the NEC system is as follows: Self Consumption - 4.0% (including station use) Transmission - 5.0% Distribution - 22.5% Total 31.5% 1.47 ADB, in its second power loan to Nepal, provided technical assistance for a comprehensive study by a specialist, a distribution engineer from British Electricity International, UK, to analyze the incidence of high distribution loss. He has visited Nepal, completed his study and has sub- mitted recommendations to NEC and ADB. 1.48 According to the study, most of the losses occur in the distribution system of the Bagmati Zone, which with the Naryani and Janakpur Zones form the Central Region. In terms of consumption, the Bagmati Zone accounts for 75% of the Central Region's energy, 60% of NEC's energy, and 55% of Nepal's total energy; thus, it is the maior area consuming electricity. A breakdown of system losses in each zone of the Central Region operated by NEC is given below: Bagmati Zone - 36.0% Naryani Zone - 13.5% Janakpur Zone - 23.0% 1.49 The high losses in the distribution system of the Bagmati Zone are mainly due to improper metering and theft of electricity. Based on the recommendations of the specialist, NEC has recruited 46 appropriate staff, purchased two vehicles and 7,200 meters. Also, enactment of by-laws dealing with theft is being undertaken. NEC's program calls for metering all unmetered consumers by January 31, 1985, checking and sealing meters of all large consumers and all domestic consumers by June 30, 1988 and rectifying and recalibrating all meters by June 30, 1990. NEC has started with the Bagmati zone. During negotiations, targets for reducing losses were agreed: in the Central Region losses to be reduced to 25% by FY86 and to 20% by FY91; the overall losses in Nepal to be reduced to 24% by FY86 and to 18% by FY91. The loss reduction program would include the above items of work. It was also agreed that the above specialist would be appointed by December 1984 1/ The definition for losses is given in Annex 6.  -12- and, after his review, a revised comprehensive program will be submitted to IDA by July 15, _185. Funds have been provided for this specialist under the proposed project. 1.50 In order to reduce technical losses in the distribution system, NEC is strengthening and upgrading its local distribution system. Under a Japanese Government grant of 1,500 million Yen, substantial improvements have been made in the Kathmandu distribution system during FY82. Another grant of 2,100 million Yen for the second stage of improvements has been negotiated and these works are planned to be commissioned by the end of 1984 (para 3.23). Apart from the Kathmandu valley, disitribution facilities are also being strengthened in other areas. ADB's Third Power Project has provided a loan of US$2.6 million for the Biratnagar (Eastern Region) distribution system and US$1.92 million for the Birganj and Janakpur areas (Central Region). These works are likely to be compLeted by June 1986. The strengthening of the local distribution sysl:ems of the major load centers should enable technical losses to also be reduced, making it possible to realize the targets stated above. II. EXECUTING AGENCY AND THE BENEFICIARY A. Executing Agency 2.01 It has been the practice of HMG to establish separate organizations, known as development boards, to execute large, complex projects requiring close coordination with contractors and engineering consultants. The Minis- ter, Ministry Water Resources, is usually Chairman of the Board which has served to facilitate the overall execution cf the project, including being able to expedite decisions through Government and to gather a cohesive project team to focus full time on the project. 2.02 The Marsyangdi Hydroelectric Development Board (MHDB) would be the executing agency responsible for constructing the proposed project. Upon completion, the facilities would be transferred to the Beneficiary (para 2.05) for operation and maintenance. The debt incurred in building the plant would also be passed on to the Beneficiary for servicing (para 4.21). This arrangement would be the same as that used in constructing the Kulekhani Hydroelectric Power Project where the Kulekhani Hydroelectric Development Board was the executing agency. 2.03 MHDB has been officially established by Gazzette Notification No. 32, dated December 8, 1981. The Board members consist of the Minister and Assistant Minister, Ministry of Water Resources as Chairman and Vice Chair- man, respectively. The other seven members are representatives of the Minis- tries of Water Resources, Finance, and Law and Justice; a member of the National Planning Commission; the General Manager, NEC; Chief Engineer, ED and the Project Manager, MHDB as Member Secretary. Once the NEC and ED are absorbed by the new Authority (para 2.05), the General Manager, NEC and Chief Engineer, ED will be replaced in the MHDB Board by appropriate officers from the new Authority.  -13- 2.04 A Project Manager has been appointed and the staffing of MHDB is near completion. The staff will total about 280, of which 170 will be involved with the technical aspects of the work and the remainder with administration, accounts and support work. The technical staff will include 35 engineers, many of whom will be seconded from the ED or, after the reorganization, from the new entity. The organizational structure includes five divisions/ sections: (a) Project Evaluation and Planning Division; (b) Civil Works Division; (c) Electromechanical Division; (d) Accounts Section and (e) Administration, Stores and Procurement Section (see organization chart). MHDB has already started some preliminary work (para 3.20). The size and com- plexity of the project requires engineering consultants to be heavily involved in its execution (para 3.18). During negotiations, assurances were obtained that HMG would maintain the MHDB with adequate staff, funds, powers and responsibility in order to carry out the works under the project. Such obligation will terminate six months after the works are handed over to the new Authority. B. The Beneficiary 2.05 The Beneficiary would be a new Authority which is in the process of being established under a reorganization of the Nepal power sector (para 1.19). It will be known as the Nepal Electricity Authority (NEA). Draft legislation calls for it to be an autonomous and corporate body. During negotiations, agreement was reached that it will absorb the operations of NEC and ED, as well as the work of the development boards, except that those Boards presently executing projects will continue with the execution through to completion, after which future work of this nature would be carried out by the new Authority. Except for large, complex power projects involving neigh- boring countries and those small power projects carried out under the juris- diction of the local panchayats, the new Authority will be responsible for the power sector on a national basis and will plan, construct, operate and maintain generating stations, transmission lines, distribution systems, and all associated facilities. This consolidation will serve to improve coor- dination between the planning, construction and operating functions, and thereby strengthen the sector's overall operations. 2.06 Under ADB's Fourth Power Project, a Memorandum of Understanding with HMG outlines the basic organizational structure of the new Authority. It shows four functional units: (a) generation and transmission; (b) distribu- tion and customer service; (c) rural electrification (including isolated mini-hydro projects) and (d) planning, evaluation and finance. The Board would consist of eleven members, including a Chairman on a part-time basis, supported by a full-time Vice Chairman. Four members of the Board would be the officers in charge of each of the four operating units. Other members would be determined by HMG. A condition of effectiveness of ADB's Fifth Power Project requires HMG's approving the legislation for establishing the new entity. Also, under ADB's Fifth Project HMG has agreed that NEA would commence operations by April 30, 1985. 2.07 The reorganization along the lines indicated is a sound decision for Nepal. IDA has fully supported it and has been urging HMG to move ahead with it. There has been close coordination between ADB and IDA in all aspects of  -14- this effort. In parallel with ADB's covenant requiring that the new Authority commence operations by April 30, [985, during negotiations agree- ment was reached that the new Authority wilL be established with the powers and functions, and organizational structure (paras 2.05 and 2.06) by such a date to enable it to commence operations by April 30, 1985. 2.08 Consultants have been assisting HMG in implementing the reorganiza- tion. Coopers and Lybrand (financed by ADB), under the first stage of their assignment, have completed the institutionaL and organizational review. Under the second stage, C & L have been concentrating on the internal organizational structure and the design of Taanagement and accounting systems and procedures for which operating manuals are being prepared. A team of four experts (an institutional/management specialist, two accounting/ financial experts and a valuation engineer), involving 56 manmonths, has been carrying out the first and second stages of the work, assisted from time to time by specialists from the consultant's headquarters. The institutional/ management specialist worked with HMG in formulating a comprehensive schedule for implementing the reorganization, including the preparation of draft legislation for the new authority. The accounting/financial specialists have been assisting in developing accounting syst:ems appropriate for the new utility. The valuation engineer has been stLpervising a physical inventory of all fixed assets, materials and stocks of the NEC and ED so that an asset accounting system may be established. This will be the basis for developing appropriate depreciation schedules and also for formulating a policy for a rational valuation and, as appropriate, reva,luation of assets. The second stage is now nearing completion. The third stage of C & L's work, which has been contracted and which will also be funded by ADB, consists of 22 man- months of work in helping to implement the new systems and in training of staff prior to the new entity commencing operations. Finally, under ADB's Fifth Power Project, an additional 31 manmonths of consulting services are being made available for expatriate adviser(s) to assist the new utility at the senior level. 2.09 A program for strengthening the technical operations has been iden- tified through a study carried out by the CIDA team in Nepal. The program is in coniunction with establishing a training center for the power sector, which is to be financed under the proposed project (paras 3.24-3.28). Audit 2.10 Nepal's constitution stipulates that the Auditor General, who is appointed by his Majesty, is responsible for auditing the accounts of Govern- ment corporations and authorities. Under the Kulekhani project, the AG carried out the audit for the project. With respect to the MHDB project accounts, this arrangement will continue. With respect to NEC, the AG has been delegating the work of performing the auadit to an independent auditing firm. This arrangement is satisfactory and is expected to continue with the new Authority once it is established. This was confirmed during nego- tiations. NEC completes the preparation of its accounts in about four months. The AG however takes an additional eight months, including the three months it takes for the outside independent audit, before finally approving the audited accounts. The reason given for the delay is a shortage of AG staff. During negotiations, agreement was reached that (a) project accounts of MHDB will be audited for each fiscal year by an independent auditor  -15- acceptable to IDA, and a report of such audit furnished to IDA not later than six months of the end of the fiscal year, and (b) the audited financial statements of the new Authority together with the full audit report will be submitted to IDA within twelve months of the end of the fiscal year. In order to enable IDA to monitor the financial performance within a reasonable period, an understanding was also reached that the unaudited financial state- ments of the new Authority will be submitted to IDA within six months of the end of the fiscal year. These arrangements are satisfactory. Government Arrears 2.11 Private consumers have been settling their electricity bills satis- factorily, while the Government has been delinquent in settling its accounts. Based on NEC's records, as of the end of FY83 government arrears, including those for street lighting, amounted to NRs 12.8 million, which compares to government billings for that year of about NRs 6.8 million. The government arrears amounted to 39% of NEC's total arrears though government billings are only about 8% of total billings. 2.12 The Ministry of Finance agrees that in order to solve this problem, arrangements need to be made for ensuring that government agencies and departments pay on time, including treating them like any other consumer and disconnecting service for non-payment. During negotiations, agreement was reached that by March 31, 1985, HMG will furnish to IDA for its review and comments a proposal of methods/procedures for ensuring that Government users pay their electricity bills by no later than two months after receiving the bill and, taking into account IDA's comments, commence implementing such methods and procedures in a manner satisfactory to IDA, so as to ensure that the bills will be paid within two months of receiving them. Regarding government arrears owing NEC, those arrears pertaining to government agencies and departments have now been settled, and those pertaining to street light- ing will be settled by August 15, 1984. Dividends 2.13 The new Authority will require all available funds in order to be able to meet its operating expenses, service its debt and finance a portion its capital expansion. Although in the immediate future it will not be constructing generation and transmission plant (the existing development boards will continue with this construction until their respective projects are completed--para 2.05), it should nevertheless be expected to begin con- tributing to the financing of such plant. These funding requirements should be met before paying any dividends. During negotiations, agreement was reached that there will be no declaration of dividends before the completion of the Marsyangdi project and then only after the aforementioned requirements for internal funding are met, including the new Authority financing not less than 30% of its capital expansion. Insurance 2.14 An appropriate insurance program for NEC that would be consistent with sound public utility practice was discussed with the NEC General Manager. He advised that he would request the National Insurance Corporation  -16- of Nepal to review the matter in order that a recommendation for an accept- able insurance program could be presented. Should this not prove satisfac- tory, the General Manager would seek outside assistance. During nego- tiations, agreement was reached that the new Authority will undertake to develop an insurance program consistent with sound public utility practice and that a proposal of the program will be submitted to IDA for review and comment by June 30, 1985, and thereafter, tsking into account IDA's comments, implement the insurance program in a timely manner satisfactory to IDA. Training 2.15 There is no formalized training program in the power sector. The training that does take place is more on an ad hoc basis with counterpart staff being attached to consultants and contractors working on projects, a few staff being attached to the CIDA team working in an advisory capacity to the Ministry of Water Resources, and the occasional participant selected to attend seminars abroad. HMG recognizes the need for establishing training programs to upgrade staff essential for operating and maintaining the expand- ing power system in Nepal. In order to establish a training function as an integral part of the power sector's operations, a facility for practical training in technologies involved in an electric power system is required. This is being provided under the project (paras 3.24-3.27). 2.16 Other training under the project would be in conjunction with the work of the consulting engineer and contractor. The draft agreement of the consulting engineer provides for ten engineers to receive training in Germany for periods of from 1 to 6 months in such sulbjects as assessing the feasibility of hydroelectric projects, design of hydraulic structures and civil works, construction management of hydroelectric projects, erection and installation of equipment and operation of hydroelectric projects. The contractor's contract will include a provision for training local staff at the construction site in accordance with a pre-agreed program. Provision will also be made for the training of engineers in the workshops of contrac- tors during the manufacturing and testing of equipment. Finally, training will also be carried out in conjunction with the work of the Coopers and Lybrand consultants (para 2.08), whose assignment includes the establishing of a training program concurrently with their developing improved systems and procedures. III. THE PROJECT A. Project Description aid Oblectives 3.01 The Marsyangdi Hydroelectric Power Project is a run-of-river project, located on the Marsyangdi River, a tributory of the Trisuli River which drains the Gandaki basin in Central Nepal (IBRD map 16928R). The project is about 110 km west of Kathmandu and is accessible either from the Kathmandu- Mugling-Pokhara highway or from the Hetauda-Bharatpur-Mugling road. The project is designed to develop the potential of the Marsyangdi River on a 13 km stretch where a gross head of about 90 m is available.  -17- 3.02 The project would divert the waters of the Marsyangdi River into a tunnel for the generation of power. Being a run-of-river scheme, the project will not alter the daily volume of water flow to India. The Association is satisfied that the project would not adversely affect the interests of India. The installed generating capacity consisting of three units, each rated at 23 MW, has been designed keeping in view the low flows during dry winter months (para 7 of Annex 13). The total annual energy based on monthly average flows that would be available from the project is estimated at 462.5 GWh (Table 4 of Annex 23), out of which 357 GWh is estimated to be consumed in the system (Table 2 of Annex 23). The firm annual energy based on 95% probability would be 209 GWh. The project would meet additional peak demand of the grid system up to 69 MW during winter 1/ and 66 MW during summer. The river flow data is available only for the last 10 years. It is therefore difficult to calculate the reduction in energy on occurrence of a dry year. However, it is estimated that within a return period of 10 years, minimum 1 day duration low flow could be 33.6 cu m/sec as compared to 35.6 cu m/sec on 95% probability (Annex 13). The average annual flow could go down to 186 cu m/sec in place of normal estimated flow of 209 cu m/sec. Energy output could be reduced by about 6% in winter months. On the occurrence of a drier year, there would be further loss in energy production. Should such a situation arise, existing diesels or regulated release of stored water from Kulekhani reservoir would be able to meet the shortfall. Alternatively, load shedding could be resorted to. The extent of shortage would depend upon the drought severity. For details of the project see Annexes 11 and 12. For project details of hydrology and sedimentation, geology and seismicity related to the project, refer to Annexes 13 and 14, respectively. Project Components 3.03 The principal components of the project are: (a) diversion works including the settling basin which will be used as diversion channel during construction but will be required for settling of suspended sediments later on; (b) a 98 meter wide gated overflow type diversion weir with 5 radial gates; (c) a 44 meter wide flushing structure with 2 radial gates for flushing sediment from the settling basin; (d) an intake structure for the headrace tunnel; 1/ Sixty-nine MW is based on 95% probability flow available during month of December and the 1.5 million cubic meters available for daily pondage. Under one unit operation, the maximum output would be 26 MW. Under three unit operation, the maximum output would be 69 MW during winter and 66 MW during summer, the output of each unit being 23 and 22 MW, respectively. The variation is due to different head losses and different tailrace levels under varying river flow conditions.  -18- (e) a concrete-lined circular headrace tunnel 7,100 meters long with a diameter of 6.4 meters; (f) a surge tank; (g) a 75 meter steel-lined pressure shaft with a 5 meter diameter; (h) three tailrace tunnels of 30, 35 and 40 meters; (i) a semi-underground power station with 3 units of 23 MW each; (j) an outdoor 132 kV switchyard at Marsyangdi, extension of 132 kV substations at Bharatpur, and upgrading of the 66 kV Balaju substation to 132 kV; (k) two 132 kV transmission lines from (i) Marsyangdi to Balaju (90 km) and (ii) from Marsyangdi to Bharatpur (40 kmi); (1) extension of local distribution system at Kathmandu; and (m) consulting engineer and technical assistance. Project Objectives 3.04 The obiectives are: (a) to meet the forecasted demand for eliectricity in Nepal up to FY93; and (b) to strengthen the power sector in Nepal. B. Background of Project Investigation and Formulation 3.05 With the assistance of various donors, there have been investigations of various potential hydroelectric sites on the Marsyangdi River since 1966. The present site was first identified by a team of Chinese engineers in 1966 who proposed a run-of-river project with a power station of about 40 MW capacity using a canal of about 6 km and a 57 meter head. This proposal was slightly modified by the ED in 1974 which proposed a power station of 36 MW using a head of 51 meters through a 4.5 km long water conveyor system con- sisting of canal, tunnel and pipeline. Also, there have been several addi- tional studies in which the high dam alternatives on the Marsyangdi River were investigated. 3.06 In 1980, the Water and Energy Commission compared 13 development programs to meet load demands up to FY91. These programs included 6 possible hydro projects and also various thermal (coal, gas, diesel) options. Based on this economic comparison, the Marsyangdi run-of-river project was selected as the next project.  -19- 3.07 The feasibility report of the proiect was prepared by consultants LI and SMEC in 1979 (para 3.18). In 1981, HMG entrusted the task of prepara- tion of detailed engineering designs and tender documents to the same con- sultants. The funds for the feasibility study were provided by GTZ, and KfW provided funds for preparing detailed engineering design and tender documents. 3.08 Before finalizing the present design, various possible sites were inspected by the consulting engineer. Three alternative weir sites and six power station sites were examined in the area of the present site. As an alternative to a tunnel, a canal was also considered. In order to be able to get additional water from Darondi Khola (a small tributory of Marsyangdi) during dry winter months, the possibility of shifting the weir to a site downstream of the confluence of the Marsyangdi River with Darondi Khola (about 10 km downstream of the present site) was a:Lso examined. The geological conditions of all sites were found to be less favorable than the site finally selected for the proposed project. The power potential at other places was also found to be lower than the proposed project site because of smaller head. 3.09 The number of generating units chosen is three, keeping in view the following considerations: limits of the heaviest single weight for transport, requirement of capacity margin for forced outage and planned maintenance, efficiency of the units under low water discharge and, finally the increase in the cost of the project with the increase in the number of units. C. Cost Estimates 3.10 The project is estimated to cost NRs 5,173.1 million (US$323.3 mil- lion equivalent), including a foreign exchange component of US$248.4 million. Details of estimated costs are shown in Annex 15, including supporting tables, and are summarized below:  -20- Foreign Description Foreign Local Toltal Foreign Local Total Exchange ---ONRs mlllion)-= ---(US$ million)-- Preliminary Cost & Administrative Expenditure 1. Preliminary works including land acquisition etc. 40.3 47.0 87.3 2.5 2.9 5.4 46.1 2. Administrative ex- penditure by MHDB 8.0 43.4 51.4 0.5 2.7 3.2 15.5 Civil Works (Lot I) 3. Site installation I 101.6 18.0 119.6 6.3 1.2 7.5 84.9 4. Diversion works 109.1 48.8 157.9 6.8 3.1 9.9 69.1 5. Diversion weir, flush- ing structure and intake structure 510.4 148.6 659.0 31.9 9.3 41.2 77.5 Civil Works (Lot II) 6. Site Installation II 308.6 54.8 363.4 19.3 3.4 22.7 84.9 7. Head race tunnel, surge tank 700.9 265.5 966.4 43.8 16.6 60.4 72.5 8. Pressure shaft, tail race, power station, switch yard 159.7 74.7 234.4 10.0 4.6 14.6 68.1 9. Permanent roads and outdoor works 9.4 5.9 15.3 0.6 0.4 1.0 61.8 Equipment 10. Hydraulic steel structures 232.5 30.1 262.6 14.6 1.9 16.5 88.5 11. Mechanical equipment 120.6 24.0 144.6 7.6 1.5 9.1 83.4 12. Electrical equipment 297.3 55.2 352.5 18.6 3.5 22.1 84.3 Transmission Works 13. Transmission lines 88.0 44.8 132.8 5.5 2.8 8.3 66.3 14. Substations & Local 135.2 36.7 171.9 8.4 2.2 10.6 79.2 distribution system  -21- /a /a Foreign Description Foreign Local Total Foreign Local Total Exchange -(N Rs million) (US$ million)% Consulting Engineer 1D. Consulting Engineer during construction supervision 120.4 - 120.4 7.5 - 7.5 100.0 Technical Assistance 16. Panel of Experts & Claims Specialist 17.2 - 17.2 1.1 - 1.1 100.0 17. Supervision of system loss reduction program 1.7 - 1.7 0.1 - 0.1 100.0 19. Catchment management plan 3.4 - 3.4 0.2 - 0.2 100.0 20. Training for power sector 43.0 - 43.0 2.7 - 2.7 100.0 Total Base Cost 3,007.3 897.5 3,904.8 188.0 56.1 244.1 Contin encies Physical b 413.4 124.8 538.2 25.8 7.8 33.6 Price Ic 553.8 176.3 730.1 34.6 11.0 45.6 Total 3,974.5 1,198.6 5,173.1 248.4 74.9 323.3 /a 'Duties and taxes are included in local costs and amount to NRs 253.1 million equivalent to US$15.81 million. These are based on contract taxes at 5.0% of total contract price and custom duty at 1% of CIF price of imported equipment and material. /b Physical contingencies as a percentage of base cost are provided at 20% for head race tunnel, 15% for other civil works, 10% for equipment, transmission lines, local distribution, substations and 5% for consulting engineer. /c Price contingencies are provided as follows: Year 1984 1985 1986 1987 1988 1989 1980 Foreign (%) 7.7 7.2 6.5 6.0 6.0 6.0 6.0 Local (%) 9.0 8.0 7.0 6.0 6.0 6.0 6.0 3.11 Cost estimates are based on mid-1984 prices. The estimates have been prepared by the consulting engineer after detailed site investigations and have been reviewed by the panel of experts. The transmission system for connecting the proposed Marsyangdi power station with the grid has been designed after detailed system studies. Keeping in view the uncertainties involved in the geology (para 6 of Annex 14), physical contingencies of 20% for the headrace tunnel have been provided as compared to 15% for other civil works. Total physical contingencies amount to 13.8% of the base cost. Price contingencies amount to 16.4% of the base cost, including physical contingen- cies. Exchange rate of US$ 1 = NRs 16.0 has been assumed.  -22- D. Project Financing 3.12 The finances arranged are shown in the following table: Source Amount Purpose (million) Equivalent US$(million) IDA US$ 101.19 i) Civil works, Lot II (foreign cost) 1.0 ii) Consulting Engineer (see para 3.14 for details) 4.81 iii) Technical Assistance (see para 3.14 for details) Sub-total 107.00 KFW DM 186.3 54.57 /a i) Equipment (foreign & part of local cost) 12.65 ii) Substations and local dis- tribution.system (foreign and local cost) 7.30 iii) Consulting Engineer (see para 3.14 for details) Sub-total 74.52 SFD SRls 86.0 24.5 i) Civil works, Lot I (part of foreign cost) 0.5 ii) Consulting Engineer (see para 3.14 for details) Sub-total 25.0 KFAED KD 6.0 20.5 i) Civil works, Lot I (part of foreign cost) 0.5 ii) Consulting Engineer (see para 3.14 for details) Sub-total 21.0 /a Based on exchange rate of US$1 - DM 2.5 - SRls 3.44 - KD 0.286  -23- Source Amount Purpose (million) Equivalent US$(million) HMG NRs 28.05 i) Civil works, Lot II (local cost) 29.77 ii) C:Lvil works, Lot I (local cost & part of foreign cost) 2.08 iii) Equipment and local distribu- tion (part of local cost) 10.54 iv) Transmission lines (foreign and local cost) 5.73 v) Preliminary works (foreign and local cost) 3.81 vi) Administrative expenditure (foreign and local cost) 15.81 vii) Local duties and taxes Sub-total 95.79 Total 323.31 E. Procurement 3.13 The following table shows the procurement arrangements: Project Elements Procurement Method ICB LCB Others /a N.A. Total - - US$ Million Preliminary works (including land acquisition) - 5.33 0.4 0.22 /b 5.95 Administrative expenditure - - - 3.81 3.81 Civil works, Lot I 74.77 - - 3.96 /b 78.73 Civil works, Lot II 129.24 - - 6.84 /b 136.08 (101.19)/c Equipment - - 56.65 3.40 /b 60.05 Substations and local - - 12.65 0.77 /b 13.42 distribution Transmission lines - - 10.54 0.62 /b 11.16 Consulting engineer - 9.3 - 9.3 (1.0)/c Technical assistance - - 4.81 4.81 (4.81)/c 323.31 /a Tied procurement and consulting services. /b Local duties and taxes payable by HMG. /c Figures in parenthesis show IDA funds.  -24- 3.14 Preliminary works consisting of offices, residences, water supply and electricity at the proiect site (US$5.73 million) are already under execution with the funds provided by HMG. Tihis includes US$0.40 million which is the estimated compensation cost for land acquisition. Administrative expenditure consists of the salary and other incidental expenses of staff of MHDB and will be financed by HMG. Civil works have been split in 2 lots. The costs of Lot I, consisting of the diversLon weir, flushing structure, intake structure, diversion works and site installation (required for Lot I) is estimated at US$74.77 million, out of which foreign portion is US$60.48 million. US$45 million will be cofinanced by SFD, KFAED. Cofinancing for remaining portion will be arranged by HMG. ISFD, KFAED will consider increas- ing their cofinancing after opening of bids. The contract will be awarded on the basis of international competitive bidding (ICB) as per guidelines of SFD, KFAED. The foreign cost of Lot II, consisting mainly of the headrace tunnel, pressure shaft and permanent roads, estimated at US$101.19 million out of total cost of US$129.24 million will be financed by IDA. The contract will be awarded on the basis of ICB in accorcdance with the Bank's procurement guidelines. The foreign and local cost of equipment (hydraulic steel struc- tures, mechanical equipment and electrical equipment) is estimated at US$56.65 million, of which the foreign portion is US$51.37 million. Whole of the foreign portion and part of local cost, amounting to US$54.57 million would be financed by KfW. KfW would also finance the total cost of the substations and local distribution system estimated at US$12.65 million. The contracts will be awarded after inviting bids from suppliers within the Federal Republic of Germany. HMG is arrangirng funds from outside sources, including ADB, for the transmission lines estimated at a total cost of US$10.54 million. During negotiations, it was agreed that HMG will obtain necessary financing for these works by not later than June 30, 1986. The fees of the consulting engineer, amounting to US$9.3 million, will be cofinanced by KfW (US$7.3 million), IDA (US$1.0 million), SFD (US$0.5 mil- lion) and KFAED (US$0.5 million). The components of technical assistance consisting of the panel of experts fees (US$1.27 million), the cost of the Marsyangdi catchment management plan (US$0.25 million), a power sector train- ing program (US$3.16 million) and supervision of loss reduction program (US$0.13 million) will be financed by IDA in accordance with IDA guidelines. The bidders for civil works are in the process of being prequalified. A price preference of 7.5% will be given to local contractors if they apply individually or as a joint venture. The design and technical specifications to be incorporated in the bidding documents have been prepared by the con- sulting engineer and reviewed by the panel of experts. In addition to two bidding package for civil works, there probably will be one bidding package for each of the following items: 1. Hydraulic steel structure 2. Mechanical equipment 3. Electrical equipment 4. Substations and local distribution system 5. Transmission lines It is estimated that the bidding documents for civil works will be ready for issue by June 1984.  -25- Contract Review 3.15 There would be one bidding package for Lot II of the civil works, which will be financed by IDA and is estimated to to cost US$129.24 million equivalent (including contingencies). This package would be reviewed by IDA prior to award of contract. Therefore, there would be 100% review of the Lot II civil works contract. Copies of the contract for equipment, transmission lines and substations, financed out of non-IDA funds, would be sent to IDA after contract award for information. F. Disbursement 3.16 Disbursement under the proposed IDA Credit: would be in respect of Lot II of civil works, consulting engineer and for technical assistance as shown below: Category % of Expenditure to be Financed 1. Civil Works (Lot II) 100% of foreign expenditures 2. Consulting Engineer 10.75% of foreign expenditure 3. Technical Assistance 100% of foreign expenditure 3.17 The project implementation program prepared by the consulting engineer (Annex 16) was discussed in detail with the panel of experts, KfW and MHDB during appraisal. The Consulting Engineer suggested an implementa- tion period of 42 months. This was considered as optimistic; therefore, a period of 52 months was adopted after detailed analysis. The most critical item in the project is a tunnel of about 7 km length for which 37 months have been allowed. Keeping in view various types of rocks likely to be encountered during tunnelling, the panel of experts had recommended a period of 763 working day (32.5 months after allowing for Sundays and holidays). This was based on an average rate of tunnelling achievable. The project schedule however provides for period of 37 months allowing for unforseen contingencies. The overall schedules of 52 months is therefore realistic. This compares with the recently commissioned Kulekhani Proiect where civil work contract was awarded in August 1977 and the power station was commis- sioned in April 1982 involving a period of 55 months. Moreover, for the Marsyangdi Project a panel of experts was appointed in the early stages of project design and substantial preliminary work has been completed at the site. The disbursement schedule is based on 52 months implementation plan. A comparison of disbursements has been made between an average profile of eighteen IDA- financed hydroelectric power projects (commissioned between FY73 to FY83) and the proposed project. This comparison is shown below:  -26- Cumulative Cumulative Cumulative Disbursement Disbursement Disbursement (ProJect) (Project) (Average Profile) (US$ million) % % 1984/85 December 31, 1984 June 30, 1985 14.7 13.7 1.0 1985/86 December 31, 1985 27.7 25.9 9.0 June 30, 1986 40.7 38.1 21.0 1986/87 December 31, 1986 53.7 50.2 34.0 June 30, 1987 66.7 62.4 48.0 1987/87 December 31, 1987 76.3 71.3 62.0 June 30, 1988 85.9 80.0 74.0 1988/89 December 31, 1988 91.0 85.0 84.0 June 30, 1989 96.1 90.0 92.0 1989/90 December 31, 1989 101.6 95.0 97.0 June 30, 1990 107.0 100.0 100.0 For the first 5 semesters, the proposed disbursements are expected to exceed those of the profile. Preliminary work is already taking place, including prequalification of the civil works contractor, so that payments would be made soon after credit effectiveness for such items as the civil contractor's advance payment, mobilization fee, constructLon of the contractor's camp. Later, the disbursements are expected to slow down and follow the profile more closely. It is expected that by about ithe end of 1988, when the first unit of the power station is near commissioning, about 85.0% of the Credit would have been disbursed, about 90% would bes disbursed by June 1989 when all three units are commissioned. The remaining 10.0%, representing such pay- ments as claims, payments for remaining work, outstanding payments and reten- tion fees is likely to be disbursed by June 1990. Thus, 100% of the Credit is expected to be disbursed in eleven semesters, which compares to the profile. There have been only three IDA financed hydroelectric projects completed in South Asia. This data base is insufficient and therefore a comparison has not been made with profiles of projects in South Asia. G. Implementation Engineering Services 3.18 MHDB has decided that the joint venture of Lahmeyer International of the Federal Republic of Germany (LI) and Snowy Mountains Engineering Corpora- tion (SMEC) of Australia would be the consulting engineers for construction  -27- supervision. This loint venture carried out the feasibility studies and detailed engineering for the project, and were financed by KfW after being selected from consultants within Germany. This selection is acceptable and meets IDA's guidelines. A provision of US$7.5 million (ex-luding contingen- cies) has been made in the project estimate to cover 701 manmonths of work, of which 128 are for local consultants and 573 are for expatriates. The costs are based on a draft agreement submitted by the consulting engineer to HMG in September 1982. This agreement is now being revised. 3.19 A panel of four experts consisting of (a) a hydrologist, (b) an engineering geologist, (c) a geo-technical engineer, and (d) a civil engineer has also been established by MHDB to review the design and safety aspects of the project and to make periodic reviews during construction. This panel has met four times , twice at Kathmandu during July 1982 and Ocvtober 1982 to review the design, method of construction, implementation schedule and cost estimates prepared by LI and twice at Karlsruhe Universit': (West Germany) during May 1983 and August 1983 to review the hydraulic mudel test results. In order to protect itself against claims from the contractors, MHDB intends to employ a claims advisor. A provision of US$1.1 million has been made to cover the cost of the panel and the claims advisor. Agreement was reached that the claims advisor would be appointed by January 31, 1985, in accordance with IDA's procedure so that he could assist HMG during contract nego- tiations, and that the services of the panel and claims advisor would be retained until project completion or such period thereafter as mutually agreed. 3.20 The Marsyangdi Hydroelectric Development Board (MHDB) (paras 2.03 and 2.04) has already undertaken considerable preliminary work (land acquisition at the project site, construction of houses and offices, access roads, water supply, electricity supply, etc.) using HMG funds amounting to US$5.7 mil- lion. About 50% of this amount is budgeted to be spent by the end of FY85, when it is planned to complete the essential portion of these works before contractors begin mobilizing at the site. Ecology and Resettlemert 3.21 Various ecological aspects, such as erosion in the catchment area, inundation of agricultural land, dislocation of huiman population and affects on fish life cycle, have been studied in detail. It is expected that about 69 ha of land and 8 houses will have to be acquired affecting about 110 land owners. About two ha of forest area will be submerged. To date seven houses and 49 ha of land have been acquired. Compensation has already been paid to land owners of 36 ha. MHDB has earmarked a sum of NRs 6.5 million (US$0.4 million) to meet compensation expenses. MHDB has also undertaken to offer employment to those displaced on a priority basis. It has been confirmed that tribal people are not being dislocated. During negotiations, HMG fur- nished a rehabilitation program for the dislocated persons. The project also provides for US$200,000 for preparing a report for implementation of a Mar- syangdi catchment management plan. This plan would outline various steps to be taken, such as training of river banks and planting of trees and grass required to reduce the erosion and transport of sediment to the project site. The report would be prepared by consultants to be retained by HMG in accord- ance with IDA guidelines. An agreement was reached that the terms of reference, short list of consultants and procedure for selection would be  -28- furnished to the Association so as to appoint the consultants by December 1984. Based on the consultant's report and. with the approval of IDA, HMG will adopt and commence to implement the plan by December 31, 1985. 3.22 In addition to land required at the power station site, land will also be needed for extension of substations; and for transmission line towers. An agreement was reached that the land acquisition process will be started as and when needed and that HMG will furnish to the Association, promptly after such acquisition, evidence satisfactory to the Association that such land and rights are available. Local Distribution in the Kathmandu Valley 3.23 The present maximum demand in Kathmandu is 51 MW compared to 83 MW in the whole of Nepal. It is expected that Kathmandu will continue to be a maior load centre after commissioning of the Marsyangdi project. Substantial improvements have been made in the Kathmandu distribution system during FY81 and FY82 and are likely to be made in FY84 with the help of grants provided by OECF of Japan (para 1.50). The details of these works are given below: Works Already Executed Works Planned in Next Program (1981-82) (1983-84) 1. 11 kV lines 53.0 km 1. 11 kV lines 47.5 km 2. L.T. lines 119.0 km 2. L.T. lines 116.0 km 3. 11 kV line reinforcement 33.0 km 3. 11 kV reinforcement 3.0 km 4. Distribution transformer 4. Distribution trans- capacity 26.0 mvA former capacity 28.0 mvA 5. Cables (11 kV & L.T.) 11.0 km 5. Cables(11 kV & L.T.) 25.0 km It is foreseen that additional augmentation of the Kathmandu local distribu- tion substations, not included in the OECF program of 1983-84, will be needed by 1989, when the proposed project is commissioned. This will consist of upgrading the existing 11 kV Balaju-Lainchaur line to 66 kV with a new 66 kV substation at Lainchaur. This has been included in the project. Training Facility 3.24 Under technical assistance, the project provides for establishing such a facility along the lines recommended in a recent study carried out by the CIDA team. (A copy of the study is part of the project documentation file.) In determining the training needs, the study examined both the exist- ing electric power system and its expected expansion over the next decade and the required staffing to operate and maintain it. The study shows, and senior NEC management agrees, that the most serious manpower problems lie in the technician category and that a training program for this category is especially needed. The technician category is that body of technical specialists whose training and deployment lie between that of the tradesman and the professional engineer, who directly supports engineers, can analyze operations and maintenance problems and carry out the work. According to the study, this category of technicians in the power sector numbers some 1,600 non-officer grades I-IV (supervisor-I, foreman-II, journeyman-III, hel- per-IV). Of the 1,600, about 1,100 are in operations and maintenance and the remainder in the design and construction divisions. While the plan is for  -29- the proposed training program to focus on the technician categorv of staff. it would not be wholly restricted to this group. There also would be scope for training of financial and commercial staff as well as specialized and refresher courses for engineers. 3.25 The objective of the training component is to establish in the power sector a comprehensive, practical training facility as a center for formal- ized, continuous :raining programs that will serve to upgrade the staff and enable an overall system approach to be developed in dealing with Nepal's growing and increasingly complex power system. 3.26 The cost for establishing the training center and operating it for a period of three ye~ars is estimated at US$3.6 million, including a foreign exchange component of US$2.7 million (mid-1984 prices). The foreign cost, which would be financed under the project, would cover five specialists to be domiciled in Neprl for three years (US$1.9 million), two vehicles, including spare parts, (USi30,000) and training equipment (IJS$750,000). The local costs, to be financed by HMG, amount to US$0.9 million and would finance the building and housekeeping expenses for the three years. A project cost estimate is shown at Annex 17. 3.27 The plan of action together with a timetable for establishing the Center is: (a) HMG to appoint a training coordinator by not later than two months from date of Credit effectiveness; (b) HMG to prepare a short list of firms (about four), from which to call proposals, and to prepare terms of reference. The short list, the invitations for proposals and terms of reference to be submitted to IDA for approval by not later than four months from date of Credit effectiveness; (c) invitation by HMG to submit proposals to be issued by not later than one month following the approval of Association of short list under (b) above; (d) HMG to submit its recommendation for selection of consultants by not later than six months after issue of invitations under (c) above; (e) the contract to be signed by not later than one month from date of approval of Association; (f) the consultants to arrive in Nepal and commence their work by not later than two months after signing the contract. At this time the training center will commence operations in an existing building supplied by HMG. A regular building designed for the training center will be built and completed by not later than twelve months from the time of arrival of consultants; and  -30- (g) HMG to appoint counterpart staff to the consultants team to be ready to join the team upon its arrival in Nepal. The counterpart staff will take over the training center's operations from the consultants team upon the latter's departure. During negotiations agreement was reached that the plan of action for estab- lishing the training center will be carried out in the time frame indicated in (a) through (g) above. Risks 3.28 The risks involved in the execution of the project could be in respect of design, construction, project management, delays in supply of essential items such as fuel, cement, steel, excessive claims and cost over- runs. Steps taken to prevent these risks include the appointment of a panel of experts to review the design and make periodic reviews during construction (para 3.19), appointment of a claims advisor (para 3.19), provision of appropriate physical contingencies (para 3.19), a well staffed executing agency (MHDB) which, with the close assistance of the consulting engineer, will manage the project and supervise the contractors (para 3.20). MHDB will also draw upon the experience gained in executing the Kulekhani Project. The most critical item in the project is the 7.1 km tunnel. Should exceptionally poor geology be encountered, the project could be delayed. The panel of experts, therefore, includes one geologist and one tunnelling specialist. Further steps include the erection of a 33 kV line to the project site to avoid dependency on diesel power generation, and construction of approach roads to the construction camp and offices. The contractors, suppliers and consultants will insure against physical risks to equipment during transport, handling and erection and against third party risks. The individual works during execution will also be insured against physical hazards. The insurance will cover design defects in the civil works as well as in the equipment. In view of these steps, HMG has decided that it would not appoint a special consultant to carry out a risk ana:Lysis of the project. This decision is acceptable. Completion Report 3.29 HMG should undertake not later than six months after the closing date, or such later date as may be agreed for this purpose, to prepare and furnish to the Association a Project Completion Report of such scope and in such detail as the Association shall reasonably request, on the execution and initial operation of the project, its cost and the benefit derived from it, the performance by the Beneficiary and the Association of their respective obligations under the draft Development Credit Agreement and the accomplish- ment of the objectives of the Credit. IV. FINANCES 4.01 The following discussion on the financial aspects is confined to the operations of NEC and would be applicable to the new Authority once it is officially established.  -31- Tariffs 4.02 There has been a reluctance in Nepal to raise tariffs even in the face of increasing costs. Consequently, the present tariff level, averaging 81 paise/kWh (5.1 US cents), which was increased in May 1983 for the first time in three years, is still on the low side and has been a major factor in NEC's depressed earnings (para 4.14). This average tariff level compares to the estimated average long-run marginal cost of NRs 2.50 per kWh. 4.03 The tariff schedule consists of seven categories: domestic, commer- cial, industrial, irrigation, transport, street lig'hting and bulk supply. The tariff has a two-part structure consisting of a demand (kW) charge and an energy (kWh) charge for commercial and industrial consumers; there is an increasing block tariff for domestic consumers, and bulk electricity export to India is charged at 14 Indian paisa per kWh under a special agreement. (A summary of tariffs is at Annex 18.) 4.04 In order to achieve financial targets and objectives, tariff levels need to be raised and the schedules carefully restructured taking into account the marginal cost within consumer categories and the willingness and ability of consumers to pay. The basis for classifying consumers needs to be rationalized and, where possible, off-peak supply should be encouraged. Also, a distinction is required between wet and dry periods to reflect dif- ferences in cost of supply. It was in pursuance of these objectives that a tariff study was undertaken by HMG's Energy and Water Commission, assisted by consultants. The study's report presents an estimate of the marginal cost of providing electric power by season and time of day, and applies marginal cost to structuring tariffs for each customer class. In early 1983, NEC presented a tariff proposal along these lines to HMG; however, the tariff as finally approved addressed the level but not the differential for seasonal and time of day costs. Tariffs were increased on average by 58% in May 1983 and then three months later reduced by 2%, to 56% (para 4.07). In future tariff adjustments, HMG should be urged to reflect the cost differentials. Earnings Covenant 4.05 The earnings covenant under Credit 600-NEP (Kulekhani I) stipulates that revenues of the Nepal Electricity Corporation (NEC) be sufficient to earn the following annual rates of return on average net fixed assets in operation: FY76-77 - 2.5%, FY78-82 - 4.0%, FY83 and thereafter - 6.0%. After the 6% target is reached, the position is to be jointly reviewed by HMG and IDA with a view to reaching 8% as early as possible thereafter. Except for FY80, NEC has not been meeting the earnings covenant (para 4.15). 4.06 In order for NEC to have reached a 6% rate of return in FY83 (FY83: July 16, 1982 - July 15, 1983), tariffs would need to have been increased by about 110%. In discussions between senior HMG and Bank officials, it was recognized that a single increase of this magnitude was not possible, and it was informally agreed that, as an alternative, tariffs could be raised in two tranches of 65% each, with the first increase to be implemented by December 16, 1982, to be followed by the second within nine months, i.e., by September 16, 1983. However, considering that it would likely prove too difficult for HMG to effect a second 65% increase within nine months, it is recommended that the interval be extended to eighteen months, at the latest.  -32- 4.07 Tariffs were eventually increased i'n May 1983 by 58%, and then reduced three months later to 56%. 1/ Based on an interval of 18 months, the second tranche of the increase would be int:roduced in November 1984 and would be 65%, the increase required to enable a 6% rate of return to begin to be achieved. Forecasts show that the 65% tariff increase would enable the rate of return to rise from 1.5% in FY83 to between 5.1% and 5.7% during FY84-86, and to 6% during FY87-89. 2/ A 65% increase, while substantial, is feasible with a rationally structured tariff, and is required to enable the electric utility operating the power sector to begin achieving financial viability. 4.08 The 56% tariff increase raised the average charge per kWh of 52 paise (3.3 US cents) to 81 paise (5.1 US cents); the 65% increase would raise it to NRs 1.34 (8.4 US cents at the present exchange rate). The reason for the substantial charge per kWh required to earn even a 6% rate of return 3/ is the higher than normal cost of building power infrastructure in Nepal. This stems from the remoteness of the country and its difficult terrain and geological conditions, poor communications, scattered load centers which are widely distributed, lengthy high voltage transmission lines relative to the size of loads and high cost of fossil fuel supplies for diesel generation. 4.09 During negotiations, agreement was reached that the second tranche of the tariff increase, amounting to 65%, would become effective by not later than November 30, 1984. Also, agreement wals reached that (i) tariffs would continue to be maintained at a level to produce revenues sufficient to earn a rate of return on average net plant in service of not less than 5.0% in FY85, 5.4% in FY86 and 6% thereafter, and (ii) before April 16 in each fiscal year the operating authority will, on the basis of forecasts prepared, satisfac- tory to IDA, review the adequacy of its rates to produce an annual rate of return as set forth above in respect of the current year and the following fiscal year, and furnish IDA the results of' the review. If the review shows that the entity does not earn the required return, it will promptly take all action to do so, including adjusting the level and structure of its tariffs. In calculating the rate of return, the amount of income tax at the rate of 50% will be included as an operating expense. Should HMG in the future lower the rate of income tax, the required rate of return to be earned would be adjusted upward in the same proportion as though the 50% rate of income tax were in force. (See para 4.12 for a statement on that part of the earnings 1/ Under the previous Government, a tariff' increase of 58% became effective in May 1983; but, under the new Government, it was reduced to 56% in August 1983. 2/ FY means Borrower's fiscal year. 3/ It should be noted that while a 50% income tax is included as an operat- ing expense, the tax is calculated on ret profit after interest charges. Forecasts show that with the incurrence of debt, interest charges will become sizeable so that though there is a tax effect it is not as sub- stantial as it might appear. Calculating the rate of return net of the tax would increase only by between 0.1% - 1.9% during the forecast period.  -33- covenant related to revaluation of assets.) This proposed earnings covenant would supersede the existing agreement. Revaluation of Assets 4.10 NEC does not revalue its assets nor is there a method presently estab- lished for doing so. For the time being, the effect of not valuing the assets at current prices would not be substantial, since expansion of the power sector to any degree has only recently commenced. For example, under the capital expansion programs, the value of gross fixed assets is shown to increase from NRs 914 million in FY83 to NRs 10,205 at the end of FY91, or eleven times, with 27% of the increase taking place in fiscal years 1984 and 1985, and 56% in FY89. 4.11 In considering revaluation, it should be borne in mind that to achieve a 6% rate of return on historically valued assets, tariffs already will need to be raised from the present average charge per kWh of 81 paisa (5.1 US cents) to NRs 1.34 (8.4 US cents). This is considered a reasonable step in a long-term strategy of increasing tariffs. In terms of a roughly revalued rate base, this would be equivalent to about 4.5% rate of return. One of the most important obiectives of the covenant is to maximize earnings, and the present covenant provides for this. 4.12 While the principle of preserving values would require the earnings covenant to be based on revalued assets, it is too early to consider a revalued rate base until the new system of asset accounting and an agreed method for revaluation is established. The consultants, which are assisting HMG in designing the new accounting systems (para 2.08), will also be developing a method for asset revaluation. 4.13 Under the circumstances, a requirement to revalue assets was not included in the earnings covenant at this time. Instead, the revaluation concept will be introduced through the consultants' work and implemented at a later time. During negotiations, agreement was reached that assets would be revalued, using a method satisfactory to IDA, by JulY 16, 1986. Once the revaluation has been made, IDA will review it with HmG with the aim of agree- ing on a rate of return based on revalued assets. Until such time as the rate of return on revalued assets would be introduced the proposed covenant would continue to be operative. Past Operations 4.14 NEC income statements for FY79-83 are shown in Annex 19. The follow- ing table is a summary of past operating results:  -34- Year ending July 15: 1978/79 1979/80 1980/81 1981/82 1982/83 GWh generated and purchased 182.0 195.0 201.2 228.5 333.9 GWh sold 119.3 137.2 133.5 148.2 224.1 Station and office use (GWh) 3.4 3.7 3.6 3.6 4.1 System losses 32% 28% 32% 34% 32% ----------------NRs Million-------------- Operating revenues 49.3 63.0 74.0 80.8 126.8 Operating expenses (incl. income tax 55.8 50.3 69.6 79.7 116.3 Net operating income after tax and before interest (6.5) 12.7 4.4 1.1 10.5 Rate base 369.7 365.5 399.6 551.3 710.2 Rate of return (1.8%) 3.5% 1.1% 0.2% 1.5% 4.15 NEC's past operating results have been depressed. As shown in the foregoing table, NEC sustained an operating loss in FY79 and earned returns of 3.5%, 1.1%, 0.2% and 1.5% in FYs '80, '81, '82 and '83 respectively. The improvement in FY80 was a result of HMG discontinuing the royalty surcharge 1/ and a 30% tariff increase, and satisfied the covenanted 4% rate of return when, in accordance with the covenant, revenues were annualized to reflect a tariff increase in that year. The generally depressed financial results have been due to low tariffs, high system losses and, before it was discontinued, the royalty surcharge. 4.16 Despite weak operating results, NEC's finances have been manageable because of its rather unique financial position of very little debt, a cash flow just sufficient to finance modest distribution expansion, and no call on earnings to finance generation and transmission plant. The background to this unusual situation for a public utility is that the few additions for generation and transmission plant have been provided through bilateral grants and, when the asset has been transferred to NEC, it has been recorded as equity. However this is now changing as credits/loans are being used to finance capital expansion. Beginning with FY81, NEC/the new Authority will increasingly incur debt in connection with assets transferred to it for operation (para 4.20). 4.17 NEC Balance Sheets for FY79-83 are shown in Annex 20. Following is a summary of NEC's capitalization as of July 15, 1983: 1/ A royalty surcharge of 7.5 paise for each kWh sold was levied. It was discontinued in FY80.  -35- US$ Million Equity NRs Million Equivalent % Government investment 755.5 47.2 Consumer contributions 10.5 .7 Retained earnings (3.4) (0.2) Total Equity 762.6 47.7 95 Long-Term Debt 41.0 2.6 5 Total Capital 803.6 50.3 100 As mentioned, this very conservative capital structure, with debt amounting to only 5% of total capital, reflects a financing pattern where capital expansion has been financed with grants which were then passed on in the form of equity to NEC. The negative figure for retained earnings reflects the past depressed operating results. Financing Plan 4.18 A forecast of sources and applications of funds for the period FYB4-91 is shown in Annex 21. The assumptions used are outlined in Annex 22 and include upward tariff adjustments and execution of the construction program as planned. Following is a summary of the financing plan for the Beneficiary as a whole covering the period FY85-90, during which the proposed project would be implemented.  -36- FINANCING PLAN Period of Marsyangdi 'roject FY85-90 US$ Million NRs Million Equivalent % Application Capital assets transferred to NEC 7,271 454 91 Capital expenditures 290 18 4 Advances for construction 243 15 3 Working capital increase 157 10 2 Total requirements 7,961 497 100 Sources Internal cash generation 2,228 139 28 Less: Debt service 1,526 95 19 Net internal cash generation 702 44 9 Government equity contributions 3,172 198 40 Debt to be incurred, incl. the proposed IDA credit of US$107 million 4,087 255 51 Total sources 7,961 497 100 4.19 The amount of funds shown in the financing plan is large and illustrates the characteristic capital intensiveness of power infrastructure. During the six-year period under review, fund requirements are expected to total NRs 7,961 million (US$497 million), consisting of NRs 7,804 (US$487 million) for capital investment and NRs 157 million (US$10 million) for increase in working capital. The proposed project, estimated to cost NRs 5,173 million (US$323 million), including a foreign exchange component of US$248 million, represents 66% of the total capital investment during this six-year period. 4.20 Sources of funds include debt to be incurred by the Beneficiary in respect of assets to be transferred. This debt amounts to NRs 4,087 million (US$255 million), including the proposed IDA credit of US$107 million which would provide 43% of the project's foreign exchange component of US$248 million. The remaining US$148 million of debt is in respect of the debt portion of the capital cost of the proposed project and other projects being executed which upon completion will be triansferred to the Beneficiary during the six-year period under review. Those iworks outside the proposed project, other than modest extensions to the distribution system by the Beneficiary, are being executed by Boards similar to the Marsyangdi Board or by donor organizations. All funds for these projects have been arranged.  -37- 4.21 The ownership of the Marsyangdi project upon its completion and commissioning would pass to the new Authority which would operate and maintain it. During negotiations, it was agreed that the ownership of the Marsyangdi plant shall be transferred to the new Authority within not more than six months of commissioning and that the Borrower shall inform the association within sixty days of the transfer. Such information shall include the date of transfer, the amount of assets transferred as equity and debt and in the latter case, the terms and conditions of such debt. The debt recorded would be less any amount that the new Authority would have con- tributed towards the cost of the project. It was further agreed that the part of the project financed with IDA funds would be transferred as debt with repayment over 30 years at 12% annual interest. 1/ A grace period is not involved since the Authority would assume ownership of the assets only after commissioning. The Government would bear the foreign exchange risk. It was also agreed that the Borrower shall cause NEC to increase its authorized share capital a&and when needed to issue shares for assets transferred as equity. 4.22 The financing plan shows HMG providing f-unds to the Beneficiary as equity amounting to NRs 3,172 million over the six-year period. However, there is also a transfer of funds from the Beneficiary to HMG amounting to about NRs 1,550 million. This transfer consists of debt service payments to HMG at or near commercial terms on concessionary loans/credits extended to HMG, and payments for income tax and customs duties. The total of these payments is equivalent to about 20% of the total capital investment during the six-year period. 4.23 The Beneficiary is shown to finance 9% (12% before income tax) of capital requirements from internal cash generation, after debt service; government equity would provide 40% and borrowings 51%, bf which 42% would be the proposed IDA credit. The Beneficiary's modest contribution of 9% of capital investment reflects a situation where the present system is small (151 MW installed generating capacity and 127,000 consumers) in relation to a substantial capital program. In these circumstances and at the early stage of power development in Nepal, it is not expected that internal sources would finance more than a modest portion of capital investment. Future Finances 4.24 Forecast income statements, balance sheets, and sources and applica- tions of funds statements for FY84-91 are shown in Annexes 19, 20 and 21 respectively; notes and assumptions are in Annex 22. A summary of forecast operating results for the eight-year period is as follows: 1/ During the five-year period FY79-83, the average annual increase in the National Consumer Price Index in Nepal was 9.7%. Forecasts show an inflation rate of between 7% and 9%. The current interest rate for government loans for working capital range between 11% and 15%, and for investment purposes between 10% and 11%.  -38- Year ending July 15: 1984 1985 1986 1987 1988 1989 1990 1991 Average revenue per kWh sold (NRs) 0.82 1.17 1.35 1.35 1.35 1.35 1.49 1.76 Revenue (NRs Million) 221 348 451 537 613 690 827 1058 Net Operating Income (after tax) 90 140 177 232 260 253 404 547 (NRs million) Times debt service covered by internal cash generation 1.2 1.5 1.9 1.6 1.8 1.8 1.0 1.4 Debt/equity ratio 47/53 36/64 40/60 47/53 46/54 45/55 52/48 52/48 Rate of return (historically valued assets)(%) 5.7 5.1 5.4 6.0 6.0 6.0 6.0 6.0 4.25 Revenue projections are based on increases in generation of elec- tricity, tariff increases and a gradual reduction in system losses. Operat- ing expenses are forecast based on system expansion and planned thermal generation. Depreciation is taken at a 2.7% composite rate. The eight-year forecast shows operating revenues and expentses increasing at an average annual rate of about 29% and 22%, respectively. The number of kWhs available for sale during this period is shown to increase by 131% and kWhs sold by 174%, reflecting a decrease in system losses. For the period through FY91, this improvement together with the tariff increase would enable the Beneficiary to begin earning a rate of return of 6%, cover its debt service and contribute 9% of the cost of the capital investment program. Forecasts show that in fiscal years 1990 and 1991 addLitional tariff increases of about 11% and 18% would be required in order to continue achieving a 6% rate of return. 4.26 Forecast balance sheets (Annex 20) show the Beneficiary's financial position during the eight-year period FY84--91. The balance sheets reflect the substantial growth expected in gross fixed assets, from NRs 914 million in FY83 to NRs 10,205 at the end of FY91. In financing this expansion, long-term debt is expected to increase from a negligible NRs 41 million in FY83 to NRs 4,986 million in FY91, resulting in a debt/equity ratio of 52/48 in FY91, normal for an electric utility. Equity would increase from NRs 763 million to NRs 4,645 million as a result of substantial Government equity contributions. Other than debt service coverage by internal cash generation reaching unity in FY90, debt service coverage is satisfactory throughout the forecast period, as is the ratio of current assets to current liabilities. Nevertheless, with the new financing pattern emerging with capital assets being increasingly financed with debt, the debt servicing position should be monitored. During negotiations agreement was reached regarding a limitation on the new Authority's incurring debt, based on the Bank's debt limitation covenant, with debt service coverage stipulated at 1.2 times. V. JUSTIFICATION 5.01 A comparison of the load requirement of the integrated system with the available generating capacity reveals that unless a new gienerating sta- tion is added to the system by FY88/89, the forecast load demand would not be  -39- met. Without any additional generation, the installed available capacity would be 142 MW compared to the maximum system demand of 159 MW; the energy requirement would be about 697 GWh compared to available hydro energy of about 614 CWh (Tables 1 and 2 of Annex 23). To meet these projected load demands, it is planned to construct the Marsyangdi Hydroelectric Power Project. As the first unit of Marsyangdi is not likely to be commissioned until March 1989, a capacity and energy shortage may be experienced in the winter of 1988 (para 1.44). Least-Cost Solution 5.02 In 1982, the consulting engineer used a computer program (EXSIM) 1/ to determine the least-cost power expansion program out of four possible programs consisting of a mix of hydro projects (for which feasibility/ prefeasibility reports were available), 2/ coal-fired thermal and low-speed diesel. All programs were designed to meet a system demard of 495 MW and 1995 GWh by FY95. It was concluded that even up to a discount rate of 15%, it would be more economical to build Marsyangdi in comparison to alternative thermals. 5.03 Further economic comparison was carried out which established that the Marsyangdi Project is still the least-cost solution. Four alternative programs have been considered for the development of the power system to meet system demand of 391 MW and 1862 GWh up to FY2000. Alternative I (a) Eastern Interconnection consisting of 283 km of 132 kV transmission line from Hetauda to Biratnagar (to be operational by June 1986); (b) Western Interconnection consisting of 270 km of 132 kV transmission line from Dumkibas to Nepalganj (to be operational by June 1987); (c) Kulekhani II Power Station (hydro) consisting of 2 units of 16 MW each (to be commissioned by October 1986); (d) Marsyangdi Power Station (hydro) consisting of 3 units of 23 MW each (to be commissioned between March-July 1989); and (e) Sapt Gandaki Power Station (Hydro) consisting of 3 units of 75 MW each (to be commissioned during FY94, 95, 96). 1/ Interim report of the consulting engineer establishing the least-cost sequence for power generation expansion will be included in the project document file. 2/ Other prolects for which studies are not available yet could not be considered. To enlarge the scope of selection for future power develop- ment program, additional studies are being carried out (para 1.23).  -40- Alternative II 1/ Same as Alternative I except that t:he Marsyangdi Power Station is replaced by a coal-fired therma:L station. Alternative III I/ Same as Alternative I except that the Marsyangdi Power Station is replaced by a low-speed diesel station. Alternative IV 1/ Same as Alternative I except that the Marsyangdi Power Station is replaced by a gas turbine staticin using diesel oil. 5.04 In all alternatives, Sapt Gandaki is assumed to be the next hydro prolect because HMG has made detailed investigations of this project. The location of all thermal power stations has been assumed to be at Birganj, the terminus of the railway system from India. The Nepal Oil Corporation (NOC) has an agreement with the Indian Oil Corporation (IOC) for supplying fuel. All fuel is transported by NOC across the border by trucks. The total fuel consumption in Nepal in FY83 was about 147,600 tons and is estimated to grow to about 290,000 tons by FY93, a 7% per annum growth rate (para 1.10). If gas turbines are used, the requirement for of fuel would rise by an addi- tional 117,000 tons in FY93 (peak period of thermal generation), or a 40% increase over the 290,000 tons. In order to handle this sharp increase, there would be two options. The first option would be to negotiate with IOC for an increase in supply and extend their existing road haulage facilities. However, this would not be reliable because of frequent breakdowns of trucks and likely interruptions due to land slides. The second option would be to use Indian Railways from Calcutta to Birganj via Baruni. This option would ensure better firm supplies of fuel if satisfactory transport arrangements are made. This option has been used in the economic analysis. The cost of transporting fuel has been estimated on the basis of existing tariff. (For details refer to Table 3 of Annex 23.) 5.05 In all alternatives, the thermal energy has been generated only when hydro energy is not available. The calculations are based on the assumption that 357 GWH produced by Marsyangdi annually would be used in the system. In case of drought, usable hydro energy would be reduced resulting in increased fuel cost required for extra thermal energy. However, these considerations have not been included in economic calculations because (a) it is difficult to quantify the shortage of energy in any particular dry year as hydrological data is short; and (b) the effect is not lilkely to be significant (para 3.02). The costs are based on mid-1984 prices and do not include price 1/ It is not possible to replace the Marsyangdi project with any other hydro proiect because there has been no detailed engineering for any other hydro proiect. The only practical way of meeting the system load demand by 1988/89 is either to implement the Marsyangdi project or to replace it with one of the thermal alternatives referred to above.  -41- escalation, nor local taxes and duties. Local costs have been converted to border costs. The planning period is up to FY2000. The comparison has b-en made between the present value of various alternative programs at discoun rates of between 8% and 16%. and assuming the life of the proposed proaec'> t be 50 years. 5.06 The supporting energy balance calculations are shown in Table 4 of Annex 23. It will be seen from the cash flow and present value statements (Table 5) that the program which includes the Marsyangdi proiect, is the least-cost solution up to a discount rate of 11.9%, with the next best solu- tion being Alternative IV (gas turbine). Nepal, being landlocked, HMG is strongly opposed to the idea of developing further power generation programs which depend upon fuel oil. It is felt that the continuous import of fuel oil would be a a drain on foreign exchange resources and also that in the light of past experience and tremendous transport constraints which exist cne supply itself courd be unreliable. This reasoning is sound, particularly in light of the abundant hydro resources in Nepal. The Marsyangdi project being the least-cost solution up to discount rate of 11.,9% is therefore the co'lracc choice. 5.07 The following sensitivity analysis was carried out in order to check the influence of a lower load growth, a variation of fuel costs and the cost of civil works of the Marsyangdi project: Case 1: Investment of all alternatives has been postponed by one year on the assumption that load demand projected for FY89 will materialize only by FY90. This would mean that instsad of overall annual growth rate of 14.7% in sales during FY83-FY89, it will be 12.5%; Case 2: Instead of 3% annual increase in real prices of fuel from 1985 onwards, assumed in the calculations, a 4% increase has been considered for Alternatives III and IV; Case 3: An annual increase in real prices of fuel has been assumed by 2% from FY85 to FY95. No increase has been considered beyond FY95. Case 4: The cost of civil works of Marsyangdi project has been increased by 10%; Case 5: The cost of civil works of Marsyangdi project has been decreased by 10%. The present values of all alternatives and the switching values of discount rates up to which Alternative I is the least cost: solution under various conditions are given in Table 5 of Annex 23. It will be seen that whereas postponement of the project has no effect (Case 1), an increase in the annual increase of 1% in real prices of fuel (4% as compared to 3%) makes the project the least-cost up to discount rate of 12.7% (Case 2). By assuming an annual increase of 2% upto FY95 and no increase thereafter, the project is the least cost upto discount rate of 10.4% (Case 3). Ten percent increase in civil works cost changes the discount rate to about 10.9% (Case 4).  -42- 5.08 The consulting engineer has used a computer program (EXSIM) to check if it would be economical to design Marsyangdi as 50 or 100 MW instead of 69 MW. For each different version, designs were optimized by the consultant and the costs were estimated. When these versions were used to replace 69 MW project in the expansion program referred to in para 5.02, it was concluded that 69 MW version is the most economical. This is a correct conclusion. Primarily because of low flows during winter, increase in installed capacity beyond 69 MW would not be justified. In addition, the pondage available for daily peak would also not permit a peak as high as 100 MW. Cost/Benefit Analysis 5.09 The benefits of Marsyangdi project are: (a) increased generation of primary and secondary energy making it possible to sell more electricity to consumers within Nepal or to India until the power market in Nepal is developed; (b) conservation of imported fuel which would have been required to run the existing diesel stations to meet essential loads in the absence of Marsyangdi project; and (c) strengthening of the grid transmission system of Nepal and local distribution system of Kathmandu. The internal rate of return on the proiect is the discount rate which equal- izes the economic costs and benefits. For the sale of energy to consumers within Nepal, there is consumers' surplus beyond the tariff. The benefits should, therefore, not only include the observed revenues paid by the con- sumers according to the prevailing tariff but also should include consumers' surplus. In the absence of electricity, some consumers would be willing to use alternative energy sources, even if the cost were higher than that for electricity. No sufficient bases are available for calculating benefits due to consumers' surplus; it is also not possible to calculate the benefits attributable to better transmission system reliability and to improved Kathmandu local distribution. 5.10 Benefits have therefore been calculated on the tariffs alone. Both the existing as well as the proposed tariff have been considered. All costs have been converted to border costs and are based on mid 1984 prices. Cost and benefit streams up to CY 2040 are shown in Annex 24. In Case 1 where the existing tariff is considered, the IRR is 3.9% and in Case 2 where the proposed tariff is considered, the IRR is 5.9%. 5.11 The true economic rate of return would be significantly higher if all benefits could be quantified, such as willingness to pay of all types of consumers, increased system reliability. Additional effect would be the conservation of foreign exchange due to reduced oil imports. 5.12 A sensitivity analysis has been made in order to check the impact of the following assumptions on the IRR: (a) Benefits are increased by 10%;  -43- (b) Benefits are decreased by 10%; (c) Output of Marsyangdi (benefits) is deferred by one year; and (d) Secondary energy is sold to India at the existing tariff of IRs 0.14/KWh. The results are given in Annex 24. It will be seen that in all cases the sensitivity analysis has little effect on the internal rate of return. 5.13 The IRR of 5.9% is based on tariffs alone and shows that tariffs are low in relation to the cost of supplying electricity in Nepal. A start has been made to increase tariffs. An increase of 56% was made in May 1983 and another increase of 65% would be made under the project. Additional increases would be introduced in the future as part of the earnings covenant (para 4.25). VI. AGREEMENTS REACHED AND RECOMMENDATION 6.01 The following matters were raised during negotiations on which satis- factory agreement was reached: (a) IDA to receive reports of studies for future power projects in order to identify a future power development program (para 1.23); (b) NEA establishing a plant maintenance program (para 1.45); (c) A system loss reduction program to be implemented and a loss elimination specialist to be retained (para 1.49); (d) Maintain MHDB with adequate staff, funds, powers and responsi- bilities to carry out the project (para 2.04); (e) Except in special circumstances, the discontinuance of the formation of new development boards (para 2.05); (f) The new Authority (NEA) commencing operations by April 30, 1985 (para 2.07); (g) Submission of audited project accounts and full report for MHDB and audited financial statements and full report for the new Authority (para 2.10); (h) HMG instituting effective methods/procedures for ensuring that government users pay their electricity bills on time and settling government arrears (para 2.12); (i) Restriction on declaration of dividends (para 2.13); (j) Development of an insurance program (para 2.14); (k) Arranging funds for transmission lines (para 3.14);  -44- (1) Maintain the existing panel of experts and appoint a claims advisor (para 3.19); (m) A rehabilitation program for dislocated persons (para 3.21); (n) Commencement of implementation of recommendations on the Marsyangdi catchment management plan (para 3.21); (o) Land acquisition (para 3.22); (p) Plan of action for establishing a training center (para 3.27) (q) Financial performance covenant (para 4.09); (r) Revaluation of assets (para 4.13); (s) Transfer of the Marsyangdi plant to the new Authority, terms for repayment of the IDA funds to HMG and increases in share capital (para 4.21); and (t) Limitation on incurrence of debt (para 4.26). 6.02 The following are conditions of effectiveness of the proposed credit: (a) meeting of all conditions of effectiveness of (i) Kuwait Fund Loan Agreement, (ii) Saudi Fund Loan Agreement, and (iii) KfW Grant (paras 3.12 and 3.14); and (b) securing of additional financing required to meet the total cost of Lot I of the civil works from sources outside Nepal (para 3.14). 6.03 With the above agreements reached, the proposed project forms an appropriate basis for an IDA Credit of US$107 million equivalent.  o " N 00000 0 - CON C "'0 '0 C  U -  w 0000CC No N - 0 0.03 o' N - 00 CC 0 0' 0' - N N 0' - 0 N 0 0' C' IC o N 0 - - 0 - 0 *O 0 COCONCON NOON - 0 C IC 0 ON N  C  0 0 0 0 0 '-" " " 0 -- 0' NO. CC - S - - - 0' - - 0 0' * - 0 OC 0 00 - 0 o - '0 - - N N ON 100 0' CO - - N go SoNoos CC CC OOLCZN 0000 CC N N N N 0 0O  0 0 0 B B 0 0 0 000 * 0' - - -C - - 0. ' ' 0 0000 NO ON OCCO CCO ON ""CO N CO N 00000 CC CO  Br C0 DCC 0 0- N 0 '1 000 00 00 00 o'0 00 'I N 0 NIN 0"' '0 00. 00 ON 0'- 0001000 0 C 010 NO oNooo B B B B" o 0 0.0 0 - 0 0 CO '0 00 - ON o' N 0' IC 0 00' - - - - - C- - - - 00 00 0 CC CO N N N N N N Co a CCC 0 C . . - COO COO CC" 0 0 N N 0 0 0 00 COCOCO CCC CO 0 CO N 00.0 NC CO N CO N CO N N N N N N N N N too - C'C 0 - - - - 000 CO - N '0 C" CO 00 0 0 NC" C N iOO N N'0 0 N N ON - - - NOC 0 0' 0 0 0 0 N 0 CO N '0 0' CO '. C N 0' 0 N OC 0' '0 0 ' '0 N 00 0 CO N C 0 00 0 0 N 0 CC 0 0 0 0 N N CO N 0' 0 '0 0 '0 0' CC 0 0 tO 0' 0 0 N C  0 " N 0 0 '0000 0 0 0 0 0-00 I,, 0 N C N N '0 N ON N '0 N N C '.C - CC 0 0 '0 0 CON 000 ON 0' 0' N - 0 0' N 0 N 0 0 0 '0 ' 0 N a 0 O NC 0' - - - 0' 0 ' 0' N ..o0'NNN N 0' -* - N NN0000' N - '1 0 N '0 N CO N CO 0' 0' '0 '0 N 0' - '0 COON C C 00 0 N N N CO 0 N CO N N '0 - N '0 N 0' N C - 0' CO '0 0' C N O 0' - 0' - C - - - - - 0' 0' 0' 0'NN .0 N -00 B00 N C - N CO CO N N C CO '0 '0 0 0 '-C C N 0' N N C N C N 0' '0 N N 0 C CO o' C C-C C C N C CC C 0 '0 C C N CO C CO N N CO -C'OCO '0 N - 0 - N 0 0C''0'' - CO N '0 '0 0 -. 0' 0' 0' C C N 0 C 0 N 0' '0 N N 000 CO DCC N 0' N N C N CC 0' CO N N N '0 0 N CO N N '-'N N 0' CC CO C 0 0' N CO C CO N C N '0 C CO 0' '0 0 '0