Nepal - Marsyangdi Hydroelectric Power Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 15653 IMPLEMENTATION COMPLETION REPORT NEPAL MARSYANGDI HYDROELECTRIC POWER PROJECT (CREDIT 1478-NEP) May 20, 1996 Energy and Infrastructure Operations Division Country Department II South Asia Region 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 AND EQUIVALENTS 1 Nepalese Rupee (NR) = 100 Nepalese Paise Fiscal Year Averages: US$ 1 00 (FY84) = 15.6 NRs USS 1 00 (FY86) = 19.9 NRs US$ 1.00 (FY88) = 22.8 NRs US$ 1.00 (FY90) = 28.4 NRs US$ 1.00 (FY92) = 40.2 NRs US$ 1.00 (FY94) 47.9 NRs MEASURES 1 kilovolt (kV) = 1,000 volts I kilowatt (kW) 1,000 watts I megawatt (MW) = 1,000 kilowatts I gigawatt-hour 1,000,000 kilowatt-hours ABBREVIATIONS ADB - Asian Development Bank BPC - Butwal Power Company CfDA - Canadian International Development Agency DCA - Development Credit Agreement ED - Electricity Department EdF - Electricit6 de France International GLOF - Glacier Lake Outburst Flood GTZ - Geselischaft fur Technische Zusammenarbeit IERR - Internal Economic Rate of Return HEP - Hydroelectric Project HMG - His Majesty's Government KFAED - Kuwait Fund for Arab Economic Development KfW - Kreditanstalt ffr Wiederaufbau LDC - Load Dispatch Center LI - Lahmever International LRMC - - Long Run Marginal Cost MIDB - - Marsvangdi Hydroelectric Development Board MI-IPP - Marsvangdi Hydroelectric Power Plant MOWR - Ministry of Water Resources NEA - Nepal Electricity Authority NEC - Nepal Electricity Corporation OD - Operational Directive OMS - Operational Manual Statement PoE - Panel of Experts PSEP - Power Sector Efficiency Prolect SCADA - System Control and Data Acquisition Ssytem SFD - Saudi Fund for Development SLA - Subsidiary Loan Agreement SMEC - Snowy Mountains Engineering Corporation SPAF - Seriously Project Affected Family TA - Technical Assistance FISCAL YEAR Ending July 15 FOR OFFICIAL USE ONLY NEPAL MARSYANGDI HYDROELECTRIC POWER PROJECT (CREDIT 1478-NEP) IMPLEMENTATION COMPLETION REPORT Table of Contents PREFACE EVALUATION SUMMARY.........................................................i The Project and its Objectives.................... ...................... Project Outcome.................................................. Sustainability of Benefits..................... ....................... Findings and Lessons ...................................... ........ iii PART I: PROJECT IMPLEMENTATION ASSESSMENT.............. ..........I A. Evaluation of Project objectives...................... .......................1 Project and Sectoral Context.................... ......................1 Objective No. 1 - Meeting the Demand.......................... ..............3 Objective No. 2 - Strengthening the Sector Institutions.........................4 The Objectives in Perspective..........................................5 B: Achievement of Objectives.................................................5 Objective No. 1 - Meeting Demand......................................5 Objective No. 2 - Strengthening Sector Institutions...................... ..........6 Summary Assessment of Achievement of Objectives................... ..........8 C: Major Factors Affecting the Project..........................................8 Factors Generally not Under Government Control............................8 Factors Generally Under Government Control 9 Factors in General Subject to Implementing Agency Control 9 Cost Changes ........9 Implementation Delays.............................................10 D: Sustainability............................................................10 E: IDA Performance............................................................. F: Borrower Performance..................................................12 G: Assessment of Outcome ................................................12 H: Future Operation.............................................................13 : Lessons .............................................................. 13 Thsdocument 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 wh&hout World Bank authorization. I PART II: STATISTICAL TABLES Table 1 - Summary of Assessments Table 2 - Related Bank Loans/Credits Table 3 - Project Timetable Table 4 - Loan/Credit Disbursements; Cumulative Table 5 - Key Indicators for Project Implementation Table 6 - Key Indicators for Project Operation Table 7 - Studies Included in the Project Table 8A - Project Costs Table 8B - Project Financing Table 9 - Economic Costs and Benefits Table 10 - Status of Legal Covenants Table 11 - Compliance with Operational Manual Statements Table 12 - Bank Resources: Staff Inputs Table 13 - Bank Resources: Missions ANNEXES Annex I - Project Costs and Completion Dates Annex 2 - Submission of Audited Accounts Annex 3 - NEA Receivables from MHG Annex 4 - Reduction of Systems Losses Annex 5 - Land Acquisition and Resettlement Annex 6 - Project and Actual Values of NEA's Principal Operational and Financial Indictors Annex 7 - Aide Memoire Annex 8 - Operation Plan (Provided by NEA) Annex 9 - Comments Provided by Kreditanstalt fur Wiederaufbau NEPAL MARSYANGDI HYDROELECTRIC POWER PROJECT (CREDIT 1478-NEP) IMPLEMENTATION COMPLETION REPORT Preface This is the Implementation Completion Report (ICR) for the Marsyangdi Hydroelectric Power Project (Marsyangdi HEP) in Nepal for which Credit 1478-NEP in the amount of SDR 100.6 million was approved on May 22, 1984 and made effective on January 29, 1986. The credit was closed on December 31, 1994 compared to the original closing date of June 30, 1990. The Marsyangdi hydroelectric power plant was commissioned in February 1990 and has since been in successful operation. However, the closing date of the Credit was extended four times for a total of four-and-a-half years in order to allow His Majesty's Government (HMG) to complete the other components of the Project (loss reduction program, training program, catchment management study, etc.). Final disbursement took place on June 6, 1995, at which time SDR25,288,906.56 was canceled. Cofinancing was provided by Kreditanstalt fur Wiederaufbau (KfW), the Asian Development Bank (ADB - as part of ADB's Sixth Power Project), the Saudi Fund for Development (SFD), and the Kuwait Fund for Arab Economic Development (KFAED). The ICR was prepared by Mr. R. A. Ribi (Consultant) and Mr. Argun Ceyhan (Task Manager) of the Energy and Infrastructure Operations Division (SA2EI) of Country Department II, South Asia Region. It was reviewed by Jean-Francois Bauer, Division Chief, SA2EI and Kazuko Uchimura, Project Adviser, SA2. The present ICR is based on material in the project files, interviews with IDA personnel involved in project implementation and the results of discussions with the Borrower and the Beneficiary that took place during the November 1995 ICR mission to Nepal and are reflected in the aide-m6moire attached to the present document'. Comments received from KfW have been incorporated in the ICR. The operation plan provided by NEA is given in Annex 8.  NEPAL MARSYANGDI HYDROELECTRIC POWER PROJECT (CREDIT 1478-NEP) IMPLEMENTATION COMPLETION REPORT Evaluation Summary The Project and its Objectives i. The Marsyangdi Hydroelectric Power Project (the Project) aimed at helping the new national electricity utility, Nepal Electricity Authority (NEA) to: (i) meet the power demand in the first half of the 1990s at least economic cost to the country through the construction of the 69 MW run-of-river Marsyangdi hydro powerplant some 100 km west of Kathmandu; and, (ii) strengthen the power sector in Nepal, through, inter alia, the implementation of technical assistance in the areas of finances especially accounting and auditing, plant maintenance, reduction of system losses, and training. In parallel, to assure NEA's viability His Majesty's Government (HMG) and NEA were to comply with covenants, in particular on revenues and accounts receivable. The Borrower of the IDA Credit 1478-NEP of SDR 100.6 million (US$107.0 equivalent at appraisal) was the Kingdom of Nepal. As was the case with earlier projects, the Marsyangdi hydro power project was executed by the Marsyangdi Hydro Development Board (MHDB - the Executing Agency). Upon completion the facilities were transferred to the Nepal Electricity Authority (NEA - the Beneficiary) for operation and maintenance. The debt incurred in building the plant was also passed to NEA for servicing. The institutional and financial covenants aimed at developing NEA into a mature Government-owned utility, thus focused on NEA's performance. Project Outcome ii. NEA, with support from HMG achieved the physical objectives of the project with a high degree of success. Though the plant and its associated facilities were completed in February 1990, seven months late with respect to the schedule in the Staff Appraisal Report (SAR) their cost was some 20% lower than set forth in the appraisal estimate.(para. 13, Table 8A, and Annex 1). Since 1991, the installation's output was consistently in excess of 400 GWh per year, i.e. more than 10% higher than the expected average generation. The Internal Economic Rate of Return as recalculated after project completion was about 6.7%, at about the same level with the rate estimated at project appraisal (5.9%). iii. The closing date of the Credit was extended four times for a total of four-and-a-half years in order to allow His Majesty's Government (HMG) to complete the other components of the Project (loss reduction program, training program, catchment management study, etc.). Therefore, the Credit was closed on December 31, 1994 compared to the original closing date of June 30, 1990. - ii - iv. The implementation of the institutional measures included in the project was less successful than that of the project's physical component. NEA's institutional strengthening was slower than expected. The sector finances remained weak throughout project implementation (para. 17). However, in the context of follow-on projects revenues dramatically increased, accounting and auditing improved both in quality and timeliness (Annex 2), and accounts receivable from Government entities fell substantially (Annex 3). Most of the major maintenance of existing plant which was planned to be carried out during project implementation had to be delayed until after its completion (para. 18). The loss reduction program failed to achieve the targets agreed at appraisal and was taken up in follow-on projects (Para. 19 and Annex 4) Training developed only slowly under the project; however, progress improved substantially in the early 1990s, once the main disagreements between Nepal and IDA on the one hand and the consultants on the other were eliminated (para. 20). The subsequent Third Technical Assistance (Pancheswar) Project (TA III - Credit 1902-NEP) and the Power Sector Efficiency Project (PSEP - Credit 2347-NEP) included further institutional technical assistance in line with the previous programs. v. Furthermore, HMG's policy to compensate people to be resettled only in cash impeded the execution of the limited resettlements required for the Project in accordance with IDA's policies (para. 31 and Annex 5). Satisfactory rehabilitation of 41 Seriously Project Affected Families (SPAFs) still remain. This issue was taken up, once again, with IDA's letter dated May 7, 1996, addressed to HMG. vi. The emphasis of the Project clearly was on providing Nepal with a generation facility to meet its internal power demand. The limited success of the institutional component was due to HMG's and IDA's optimism about the development and delivery of the sector and NEA. This was addressed in the context of the succeeding TA III and PSEP projects which started in the middle and toward the end of Project implementation. Establishment of NEA and relative improvement in the coordination between the planning, construction and operating functions were steps in the direction of strengthening the power sector. On balance the outcome of the Project is rated as satisfactory. Sustainability of Benefits vii The sustainability of the benefits of the plant is rated as "likely", as operation and maintenance is well planned and properly executed, which is likely to continue provided NEA does not let the plant deteriorate by substantially reducing maintenance calling for stopping the units in order to increase in the short term plant availability. The sustainability of the limited institutional gains achieved under the Project is also likely mainly thanks to the follow-up carried out under subsequent IDA projects (paras. 28 - 30). - Ill - Findings and Lessons viii. The major findings and lessons are summarized as follow: a) The successful operation of the Marsyangdi plant suggests that the detailed planning, execution and monitoring of operation and maintenance carried out in this plant should be applied in other facilities in NEA's system. This might involve the continuation, for a limited time, of the services of at least one specialized expatriate. b) Project outcome confirms the lesson learned in other projects and which already for several years has been reflected in Bank policy: Projects with large, lumpy physical components should be submitted to Board consideration only after bidding for the main components has taken place. c) The Project, as many before it, illustrates the fact that borrowers and beneficiaries tend to own physical project components of IDA projects more clearly than institutional ones, because they usually get involved with the IDA primarily to obtain financing of capital intensive facilities such as power plants. To improve the chances of success of institutional project components associated with these projects, IDA should ensure the borrowers' and beneficiaries' ownership and use its leverage in a delicate and balanced way to lead in a reasonably short time to ownership by borrowers and beneficiaries of the objectives and measures considered. The measure of leverage applied in connection with this and follow-on projects (i.e., TA III and PSEP), was adequate, as e.g. in mid-1990s, NEA is in a better financial shape than it was in the late 1980s. d) Establishment of NEA and relative improvement in the coordination between the planning, construction and operating functions were steps in the direction of strengthening the power sector. However, the overall performance level is not sufficient to tackle the challenges Nepal power sector faces in the second half of the 1990s. A new strategy in line with the current power sector policy guidelines of the Bank Group needs to be developed. e) IDA created a precedent that when HMG and NEA were not in compliance with their commitments, these covenants were shifted to the next projects. f) In its deliberation on the then proposed Arun III Hydroelectric Project, the Inspection Panel observed that "the Marsyangdi Project was completed without compliance on resettlement." IDA should ensure that the necessity to undertake rehabilitation and restoration of livelihood of the project affected families be accepted as a principle by HMG.  NEPAL MARSYANGDI HYDROELECTRIC POWER PROJECT IMPLEMENTATION COMPLETION REPORT PART I. Project Implementation Assessment A. Evaluation of Project Objectives Project and Sectoral Context I. The two principal sources of energy in Nepal are forests and river systems. The former are already overused. In contrast, the development of the large hydropower potential is still at an early stage, though, for a long time it has been one of the main objectives of His Majesty's Government (HMG). In the short run, it was aimed at meeting internal demand, but, in time, it was also seen as a basis for future large scale exports providing revenues in foreign currency. In the long run it may also contribute to stop forest depletion. The Marsyangdi hydroelectric project, which formed the core of the IDA operation (the Project) discussed here was well in line with the first part of this strategy, as it aimed at meeting internal demand. IDA's parallel Karnali Preparation Project (Credit 1452-NEP) was oriented towards developing Nepal's power export capability. 2. At the time of Project preparation, in the early 1980s, the sector was completing the 60 MW Kulekhani hydroplant (supported by Credit 600-NEP), which suffered massive time and cost overruns (some two years and about 80% in current terms, respectively). The new facility brought the total capacity installed for public supply in the country to 133 MW (111 MW hydro and 22 MW diesel) and ended the load shedding that had hampered sector operations for several years. The corresponding total available energy reached 349 GWh including 62 GWh of net imports from India and an estimated 8 GWh generated by captive plant. This represented an exceedingly low 23 kWh per capita, which, at the time, was lower than the corresponding figure for Bangladesh (26 kWh) and very much less than that for India (140 kWh). 3. IDA's power sector program in Nepal, included the 1983 Second Technical Assistance Project supported by Credit 1379-NEP which helped HMG identify and prepare high priority projects in the sector for international financing. Investigations of various potential hydroelectric sites on the Marsyangdi River were begun in early 1960s. The 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. High dam alternatives too were investigated in several additional studies. In 1979, Gesellschaft fur Technische Zusammenarbeit (GTZ) and Kreditanstalt ffir Wiederaufbau (KfW), both of Germany provided funds for a feasibility study and detailed engineering, respectively, by a consortium formed by Lahmeyer International (LI), also of Germany, and Snowy Mountains Engineering Corporation (SMEC) of Australia. In 1980, the Water and Energy Commission of the Ministry of Water Resources (MOWR) compared 13 development -2- 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 with an installed capacity of 69 MW was selected as the next project. The same year, IDA started the preparation of the Project by sending its first pre- appraisal mission. It took three further preparation and pre-appraisal missions and nearly three years to bring the Project to a level permitting IDA to appraise it, and a further fifteen months to make it adequate for Board presentation, in May 1984. The main reason for this lengthy procedure was that IDA wanted to avoid cost and time overruns of the type experienced in the context of the Kulekhani project. Therefore, it called for further investigations which Nepal carried out under the guidance of a Panel of Experts (PoE). How justified this cautious approach was is demonstrated by the fact that between 1980 and 1983 the estimated project costs increased from US$ 150 million (only plant) to US$ 338 million (including technical assistance - TA -). These anticipated high costs, in turn, called for the identification of additional funds from both internal and external sources. Providing reasonable assurances for the availability of these resources contributed to slow credit processing. As was the case with earlier projects,' IDA agreed to project implementation by a separate entity, the Marsyangdi Hydroelectric Development Board (MHDB) established in 1981, which, after project completion and transfer of the plant to NEA's assets was to be dissolved. 4. At the time, a substantial number of entities were involved in power sector operations, the main utility being Nepal Electricity Corporation (NEC). The other entities were the Electricity Department of the Ministry of Water Resources (MOWR), Eastern Electricity Corporation (merged with NEC in 1982), Butwal Power Company (BPC), and several development boards, including those for the Kulekhani I and Marsyangdi hydroelectric projects, transmission line projects and small hydro projects. 5. The ED was a department within the MOWR. It was responsible for planning, designing and constructing new projects which, after commissioning, were handed over for operation to NEC. However, projects in the Mid- and Far-Western Regions were retained and operated by the Department. The Department was headed by a Chief Engineer appointed by a Cabinet Committee of HMG; he was directly accountable to the Secretary of the MOWR. In end-1982, there were some 900 staff, including 200 engineers of which about 150 were on secondment to other organizations in the power sector. All of the revenues of the ED were transferred to the Ministry of Finance, while funds for capital and operating expenditures were provided through the annual budget. 6. NEC was established in 1962 under the Nepal Electricity Corporation Act of the same year, to enable the management of the important power system in northern part of the Central Region to be freed form the constraints of government bureaucracy. Later, NEC also took over the operations in the Eastern and Near-Western Regions and in the remaining parts of the Central Region. In 1982, NEC had a staff of about 3,200 including about 100 engineers. It was the largest entity in the power sector, responsible for generation, transmission and distribution of electricity in the said regions, transferred to its ownership by the ED and the development boards. 1 The Kulekhani hydroelectric project was executed by the Kulekhani Hydroelectric Development Board. -3- It financed, designed and constructed modest reinforcements to its distribution network and extended supplies to new customers. It was also responsible for billing and collection of revenue. NEC had little contact with the ED or the development boards during planning, design and construction of new works even though the facilities were subsequently transferred to it for operation and maintenance. 7. The development boards were established under the Development Boards Act to execute large construction projects, which upon commissioning were handed over to other entities for operation and maintenance. Board Members, in the case of the Kulekhani I and the proposed Marsyangdi projects, included the Minister of Water Resources as Chairman, and civil servants from various Ministries, including the Secretary of MOWR and the Chief Engineer, ED. The executive head of each Board was the Project Manager who usually was a senior engineer seconded from the ED. He was paid by the Board, was accountable to the Board and was also a member and Secretary of the Board. The Boards were staffed by engineers on secondment from the ED and by others specially selected. A few engineers were 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 of the Nepalese with large projects, the Boards relied heavily on consultants to supervise the execution of the projects. 8. During the preparation and appraisal of the Project, it was clear from experience that this multiplicity of bodies to operate the Nepal electricity system contributed to a sector performance which was not satisfactory. HMG recognized 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. Consultants financed by the Asian Development Bank (ADB) developed the concept for a new sector organization around an autonomous corporation, the Nepal Electricity Authority (NEA), which would become responsible for the power sector on a national basis except for: (i) large complex projects involving neighboring countries; and (ii) small projects carried out under the jurisdiction of panchayats (districts). Conditions under ADB's fourth and fifth power projects in the country called for establishing NEA by April 1985 along the lines established in the above studies. IDA integrated the same condition in the agreements concerning the Project. 9. The Project's broad objectives were (SAR, para. 3.04): (i) to meet demand in Nepal up to FY93 and (ii) to strengthen the power sector. Objective No. I -- Meeting the Demand - The Marsyangdi Project 10. This objective was to be pursued by constructing the Marsyangdi hydroelectric project, a run-of-river plant 100 km west of Kathmandu, with an installed capacity of 3 x 23 MW and an expected average annual output of 464 GWh (209 GWh firm, reflecting the fact that, on average, the capacity permanently available over the dry season does not exceed 25 MW). This plant had been identified as the next step in Nepal's least cost power generation development program. It was to be completed by August 1989 at a cost of US$ 294 million, i.e. at a very high US$ 4,260 per kW installed for the plant alone. Taking into account this large expenditure for additional capacity, the Project also aimed at reducing system losses (para. 16). -4- Objective No. 2 - Strengthening the Sector Institutions 11. The institution building component of the Project included measures in five main areas: (i) sector restructuring, (ii) finances, (iii) plant maintenance, (iv) the already mentioned reduction of system losses, and (v) training. 12. NEA was created in August 1985, as the principal instrument of the sector restructuring, through the amalgamation of a number of public sector organizations, principally NEC, ED and the Small Hydro Development Board. The NEA Act of 1985 established NEA as a commercial entity with responsibilities for generation, transmission and distribution throughout Nepal. However, the Act interfered with NEA's capacity to realize commercial objective through limits on the Authority's autonomy, as well as in the lack of mechanisms for accountability. The 1988 Power Sector Review identified a number of impediments to NEA's efficient operation. With IDA assistance (Cr. 1902-NEP), NEA followed up on the Review's recommendations by entering into a twinning arrangement with a mature utility, Electricite de France International (EdF), in October 1989. The agreement provided for expertise in all aspects of modem power utility practices, with special emphasis on assisting NEA to develop a corporate plan and improve operations management. NEA and EdF developed a performance improvement plan which was incorporated into a Performance Agreement (PA) between the Government and NEA. The PA was signed by HMG and NEA in October 1992, as a condition of effectiveness of Cr. 2347-NEP. The NEA Act was amended in 1992 to address all the shortcoming identified by then, with the objective of establishing an institutional framework which would enable NEA to operate as a commercial entity. 13. Although NEA did not deliver all the expectations by IDA and ADB (the most important shortcoming being that HMG did not grant to NEA the necessary autonomy), it nevertheless consolidated coordination between planning, construction and operating functions, thereby strengthening the sector's overall performance. However, the overall performance level is not sufficient to tackle the challenges Nepal power sector faces in the second half of 1990s. Therefore, a new strategy in line with the current power sector policy guidelines of the Bank Group needs to be developed. 14. The IDA operation also aimed at strengthening the sector finances by covenanting (i) a reduction of the arrears of Government entities with NEC (and - once established - NEA) to two months' billings. By FY83, these had reached nearly two years' billings. Under previous credits, NEC was to generate internally sufficient funds to produce, from FY83 on, a 6% return on net fixed historically valued assets. To achieve this, tariffs would have had to increase by 110% in this same FY83. In connection with the preparation of the Project, Government and IDA acknowledging the political difficulties to implement such a steep adjustment, agreed on a two tranche approach with 65% steps in December 1982 and December 1983, respectively. Actually, a 58% increase (later reduced to 56%) took place in May 1983, and the second step, a 65% adjustment, became a condition of the new credit. 15. Until the commissioning of the Kulekhani plant in 1982, the shortage of available capacity had induced NEC to postpone major repair and maintenance of its generating plant, which -5- therefore was deteriorating fast. To reverse this trend, the sector needed a substantial rehabilitation program, which, at the same time, would lay the ground for systematic maintenance in the future. The Project came to include the preparation and initial implementation of such a program. 16. A study carried out with ADB financing showed that, in spite of earlier efforts to reduce system losses, in FY83, these were still in excess of 30% of energy sent out (Annex 4) and especially high in the central sub-system. In the early 1980s, an ambitious program supported by a Japanese grant and an ADB loan was aiming at developing Nepal's distribution system especially in the central and eastern regions. These measures were essentially helping reduce technical losses but did not directly address the energy theft issue. Nevertheless, the corresponding system improvements were expected to lead to a reduction of the losses by some 6 percentage points by FY86 and in excess of 10 percentage points by FY91. However, in 1983, it was already evident that the sector would not meet these goals. Therefore, the Project also came to include measures leading to reductions in both technical and non-technical losses. 17. In the early 1980s, there was no formalized trainin in the sector. Therefore, the Project covered the establishment of an adequate training function, in particular the provision of (i) a facility for practical training and (ii) specific training in connection with the consulting engineers' and the contractors' work, in particular in assessing feasibility studies of hydro projects, designing hydraulic structures, managing construction, erecting and installing equipment, and operating hydroplants. The Objectives in Perspective 18. The objectives were well in line with IDA's general policies for the power sector (OMS 2.25 and 3.72). They also fitted into the agreed broad goals for the country. However, as discussed in paras. 16 to 21 below, the detailed sub-objectives for the sector's institutional development were unrealistic, in particular with respect to the timeframe foreseen for achieving them, the more so as their ownership by the Nepalese authorities was, in part, only at the intellectual level and turned out weak whenever concrete action was required (e.g. tariff adjustments and loss reduction program). B. Achievement of Objectives Objective No. 1 - Meeting the Demand 19. The Marsyangdi plant was commissioned in 02/90, seven months late with respect to the timetable set forth in the SAR, but -- due to the long preparation time --, much later than planned in the early 1980s. Therefore, in the late 1980s, there were limitations of supply, as demand had grown about as anticipated at appraisal. The shortfall was the more severe, as losses did not lessen and NEA felt unable to carry out the rehabilitation of its existing plant. The cost of the Marsyangdi plant and associated transmission facilities was USS 244 million, which, though less than 80% of the appraisal estimate, still corresponds to a high 3,600 USS per kW. (Annex 1) -6- 20. For a time, large claims from the civil works contractor carrying out the main conveyor tunnel represented a major problem in particular after the contractor had filed two cases for arbitration. However, assisted by a claims expert, the PoE, and the engineering consultants, the executing agency (MHDB) was able to reach an out-of-court settlement by agreeing to pay US$ 3.5 million, a small fraction of the original amount claimed. Thus, the Project substantially achieved its objective of contributing in a major way to the sector's ability to meet domestic demand. It achieved this, to a large extent, because the lessons learned in the context of the disappointing outcome of the Kulekhani project were applied. However, as discussed below (para. 25), until the mid-i 990s, the loss reduction program failed to contribute in a substantial way to this same objective. The Internal Economic Rate of Return (IERR) calculated in the same way as in the SAR came out at some 6.7% which, though low, is at about the same level with the appraisal estimate (5.9%). 2 3 Objective No. 2 -- Strengthening the Sector Institutions 21. Most institutional measures started late, developed slowly over the second half of the 1980s and somewhat faster in the early 1990s. Some of these (e.g. those concerning maintenance and training) proved to lack specificity in order to reach the objectives adopted. They had to be further defined and, in part, included in a modified form in follow-on projects especially the 1988 Third Technical Assistance Project (TA III) supported by Credit 1902-NEP and the 1992 Power Sector Efficiency Project (PSEP) supported by Credit 2347-NEP. The origin of these shortcomings of the Project was, on HMG's and the sector's side, deficient implementation capability and weak ownership of several of the politically unsavory actions envisaged, and on IDA's side, an overly optimistic view of the sector's ability to strengthen its institutional capability. Factors that should have induced IDA to an even more cautious approach were: (i) the fact that, at Project start, the sector was to begin working under a new setup, which, even in a much further developed environment, invariably involves a slowdown of many activities; (ii) cumulating the construction of the largest power plant in the country with a host of institutional measures, as justified as they were, put an exceedingly heavy burden on sector management which was weak to start with; (iii) the experience gathered during the long preparation time should have made clear that whereas HMG and the sector strongly owned the physical measures (i.e. essentially plant construction), their ownership of the institutional measures was much weaker. 2 In the SAR the IERR was computed using the retail electricity tariffs as a proxy for the benefits of the Project. The tariffs in effect at appraisal were used for the base case. Tariff levels estimated based on adjustments agreed under the Project were used in a sensitivity run. The IERR for the base case and sensitivity run were computed as 3.9% and 5.9% respectively. It was mentioned in the SAR, that the true IERR would be significantly higher if all the benefits, such as willingness to pay off all types of consumers, increased system reliability, etc. could be quantified. No sufficient base data was available for calculating the above benefits. In the ICR, the tariffs in effect since the commissioning of the project were used. After commissioning of the Marsyangdi plant, IDA helped the sector by financing part of the improvement of the Marsyangdi-Mugling road, the main access to the plant, from funds remaining at the time in Credit 1478-NEP. -7- 22. The sector started operating under the new organizational setup at the beginning of FY86, i.e. at about the time agreed. But NEA as its centerpiece did not develop as hoped for (paras. 17 to 20). 23. The sector finances remained weak throughout project implementation. NEA was never able to meet the revenues covenant agreed with IDA and, from FY89 to FY93, the operating income was negative. These shortfalls were the consequence of inadequate tariff adjustments and cost containment. A 35% tariff increase in FY 85 and one of 22% in FY86 allowed the signing of the Development Credit Agreement (DCA) 17 months after the Board had approved the Project. An 18% adjustment in FY88 and one of 61% in FY 92 in connection with the processing of the PSEP /4 were the only steps taken until the most recent change in HMG's approach, related to the 1992 Performance Agreement between HMG and NEA on the one hand, and the preparation of the Arun III project, on the other. The FY88 and 92 increases were what HMG felt in a position to concede when much larger boosts would have been needed to meet the covenant in the subsequent year. During the period FY86-94 cash operating costs which are essentially those that NEA can to a reasonable degree control (except for fuel which depends on HMG decisions) have not decreased in constant terms and related to the kWh sold, although sales doubled over the same period. Government's accounts receivable improved only in the 1990s, when, however, they still were substantially in excess of the three months' billings covenanted in connection with the PSEP (Annex 3). During the May 1988 Paris Aid Group Meeting, donors endorsed the Government's plan for improving the power sector's performance in general and NEA's financial performance in particular. This has formed a basis for the revenue covenants of the PSEP. It is important to note that early in the Project, monitoring of the returns and the receivables was unreliable because of the weakness of NEA's accounting. But IDA's persistent efforts supported by those of other financiers, in particular ADB, contributed in a major way to improve the accounting in general to a level that makes recent results reasonably credible. In particular, after NEA had started operations with an opening balance sheet including a highly suspect valuation of the assets, it had consultants properly revalue these, which, in FY90, led to a 300% increase in the valuation of fixed assets and to the installation of a method for their future periodic revaluation. At IDA's insistence during the implementation of the Project, NEA's debt obligations to the Government was established and finalization of all subsidiary loan agreements (SLAs - for all donor funded projects) was initiated. In absence of SLAs, NEA's debt service liability to HMG could not be correctly established. This was thus rectified. A further result of the above efforts was that, in FY93, NEA was able to provide audited accounts of significantly improved quality within the time frame agreed with IDA (Annex 2), but still with numerous qualifications by the auditors. 24. Financial constraints and the lack of preparation made it impossible for the sector to avail itself of the opportunity created by the commissioning of Kulekhani to start carrying out postponed maintenance at the older plants. This, together with demand growth about in the order of magnitude expected at project appraisal soon led to severe capacity constraints which further 4 In connection with the PSEP two revenue covenants were formulated, one calling from FY96 on for a minimum rate of return and the other establishing a minimum contribution from internal cash generation to the local cost of NEA's investment program. -8- postponed the implementation of the rehabilitation program whose preparation and implementation was included in the Project. In 1992, with Marsyangdi in operation, NEA carried out the refurbishing of Kulekhani which, in the meantime, had also deteriorated. Further work was included in the PSEP. 25. For many of the reasons set forth in para. 15 above, the losses reduction program failed to achieve the targets envisaged at appraisal. By FY93, when losses should have been below 20% of energy sent out, they were still about 25%. Most recently, in connection with the preparation of the follow-on project, NEA and IDA envisaged on a new target of 20% by FY 2000 (Annex 4) 26. Training suffered under similar shortcomings. In the early 1980s, the training needs had been defined on the basis of a study financed by the Canadian International Development Agency (CIDA). But the training coordinator appointed in 1984 resigned and follow-up action stalled. In 1987, NEA appointed Electricite de France (EdF) as training consultants. But it took until 1989 to bring the program into full gear in the context of the twinning arrangement between NEA and EdF supported under IDA's TA III project. Subsequently, the program was repeatedly hampered by the lack of local funds and slow procurement of training equipment. Summary Assessment of Achievement of Objectives 27. The Project did not include macroeconomic policy, poverty reduction and gender concerns objectives. Achievement of financial objectives for NEA and sector policy and public sector management objectives was "partial" (paras 21-26). The physical objectives were achieved "substantially" with the satisfactory completion of the Marsyangdi hydropower project. The hydro project was constructed in accordance with IDA's environmental guidelines in effect at approval. However, during the implementation of the Project, it was determined that there was need to study further and then implement a management plan for the Marsyangdi catchment. This was included in Cr. 2347-NEP. Over the last ten years a glacier lake has developed upstream of the Marsyangdi dam and the risk of a major glacier lake outburst flood (GLOF) has become a potential threat (Para.36). Therefore achievement of environmental objectives is assessed as "negligible". Completion of a satisfactory rehabilitation of the Seriously Project Affected Families (SPAFs) is still outstanding. Achievement of other social objectives is assessed as "negligible". C. Major Factors Affecting the Project Factors Generally not Under Government Control 28. Start of project implementation was substantially postponed by the long preparation time, which, however, was unavoidable in view of (i) the complexity of the physical part of the project to be executed by a sector that was weak to start with and was to be reorganized immediately after project start, (ii) the need to heed the lessons learned in connection with the earlier Kulekhani project, (iii) the desirability for IDA to avail itself of the leverage provided by the sector's eagerness to secure substantial financing and its associated improved amenability to take substantial and difficult steps towards a sound financial development, and (iv) the fact that project cost estimates increased from US$ 150 million in 1980 to over US$ 300 million in 1982-83, -9- which called for the mobilization of ever more co-financing, which, in turn made project setup more complicated. 29. From 1989 on, the trade and transit controversy with India seriously affected Nepal's economy. However, it did not lead to major delays in the physical implementation, because HMG, giving Marsyangdi high priority, went so far as to have fuel flown from Bangladesh to the work site, of course, at the corresponding high cost. 30. The good performance of the engineering consultants and the valuable advice of the PoE was a major factor in the relatively smooth construction implementation. The PoE and the claims advisor were also very helpful in resolving the impasse that the large claims submitted by the tunneling contractor could have created (para. 20). Factors Generally Under Government Control 31. The main such factor was HMG's unwillingness to provide NEA the degree of autonomy that would have been a necessary (but not sufficient) condition to allow the utility to develop into an efficient company. Especially, HNMG did not allow NEA to adjust adequately its tariffs, which made NEA even further dependent upon government financial support. This, in turn, gave HMG apparent justification for continued interference. The lack of local funds also contributed to the delays suffered in the loss reduction, training, and plant maintenance/rehabilitation programs. Further, HMG's policy of compensating resettled people only in cash on the basis of their land's market value, impeded the execution of the limited resettlements required for the Project in accordance with IDA's more recent policies (Annex 5). Satisfactory rehabilitation of 41 Seriously Project Affected Families (SPAFs) still remain. This issue was taken up, once again, with IDA's letter dated May 7, 1996, addressed to HNMG. Factors in General Subject to Implementing Agency Control 32. NEA's managerial and operational weakness was the main factor (closely tied to the previously mentioned ones) impeding a satisfactory institutional development of the utility. It was re-enforced by the fact that, at the early implementation phase, the sector had to operate in a new setup. Cost changes 33. The power plant with the associated transmission facilities came to cost some US$ 244 million, i.e. in current terms some 20% less than estimated at appraisal. In constant terms the cost savings were about 18 %. Whereas the equipment cost was substantially higher than anticipated, civil works costs came out more than 30% less than estimated. This outcome is certainly less disappointing than a cost overrun; but it still is not fully satisfactory, as (i) the deviation from the estimates was large; and, (ii) the large effort by HMG, NEA, IDA and the cofinanciers for the mobilization of additional funding which was necessary because of the high overall estimate might have been avoided if a better estimate had been available. The latter could have been achieved by - 10- waiting with Board submission until after bids for the main contracts were opened. Since then, this procedure has become standard for this type of Bank Group projects. Implementation delays 34. The power plant was commissioned seven months late with respect to the schedule set forth in the SAR. The delay was largely incurred at the start when (i) processing of the project was delayed (18 months between appraisal and Board presentation and 16 months between Board approval and credit document signature), mostly related to slow compliance by HMG with agreed measures especially in the field of tariffs, (ii) the late mobilization of the tunnel contractor in charge of the main tunnel which was on the critical path of the project. Two years after its start, the project was well over a year late, but the engineering consultants, the PoE, and the contractor were able to devise and implement measures which allowed to make up for part of the delay. D. Sustainability 35. The operation plan (Attachment 2) reflecting the presently implemented operational procedures provides an adequate basis for the sustainability of the benefits to be reaped from the Marsyangdi plant. However, in the wake of IDA's decision not to proceed with plans to support the construction of the Arun III Hydroelectric Project, substantial risks persist which could lead to NEA's inability to sustain these benefits. Major delays in the construction of the next power plant and NEA's failure to improve its operational capability could lead NEA to assign low priority to maintenance work in its plants, as it happened once again in the early 1990s. HMG and NEA are following a two-track approach for the construction of the next generation projects. While they continue their dialogue with ADB and bilateral donors for the execution of hydro power projects by NEA, they take actions to induce private investors to Nepal power sector. HMG's dialogue with IDA is currently focused on the establishment of a Power Development Fund which could catalyze investments in generation, transmission and distribution, by private developers and public entities, under the condition that the latter meet PDF's creditworthiness criteria. 36. Another risk to the sustainability of benefits from the Project is the possibility that increased sedimentation at the Marsyangdi hydropower station intake. In this regard, IDA and MHDB (the Executing Agency) commissioned, in 1988, a catchment management plan study for the watershed of the Marsyangdi project. The plan was revised and updated during the appraisal of PSEP, taking into account watershed management experience in other areas in Nepal, local natural resource management traditions and the socio-economic dynamics of the region. A pilot project, funded by GTZ of Germany is being implemented under PSEP'. About 80 km upstream from the weir, a glacier lake (Thulagi lake) has developed over the last 10 years. As a result, the risk of a major glacier lake outburst flood (GLOF) became a potential threat to the power plant. Even though a GLOF may not have a direct impact on the weir, serious operational problems may result from such an event, owing to the increased bed load arriving at the intake over many The pilot project attempts to incorporate land management practices that improve and lead to a greater sustainable productivity, derived from successful innovations in other ongoing projects or from local practices that are encouraged. The strategy is to develop approaches for improved land management in the village by the villagers. - 11 - months after the overflow. Therefore, monitoring of the development of the lake, and, if necessary and feasible, appropriate measures to lower the water level should be given due consideration. KfW, upon request of the German Government, will commission a study on this issue, which will be financed under the existing Studies and Expert Fund. In addition, KfW is initiating a feasibility study on a new Load Dispatch Center (LDC) at Kathmandu which shall be equipped with system control and data acquisition (SCADA) system. The LDC should also contribute to an improved capacity utilization of the power generated at Marsyangdi. In view of the actions taken, the ICR considers that sustainability of the plant benefits as 'likely" 37. As discussed earlier, the institutional components of the project failed to achieve their objectives in the framework of the Project itself, but most were taken up, usually in a more specific and promising form, in the context of follow-on projects. In this framework, they have good prospects to be completed with a reasonable measure of success, which should make the sustainability of the limited institutional gains achieved under the Project likely, too. E. IDA Performance 38. Despite the urgency of constructing a new generating facility, IDA steadfastly linked the processing of the operation to the development of a technically sound project that could be executed at a reasonably firm price. It also did not proceed with the signing of the DCA before the Borrower substantially complied with financial conditions, in particular tariff increases, agreed earlier. Indeed, insisting on the fulfillment of these conditions before Board presentation might have been preferable, as this would not have significantly further postponed the start of large scale project implementation but it would have avoided having an approved operation in abeyance for 16 months. In addition, if IDA had waited with Board presentation until after the bids for the civil works and equipment were opened, it would have avoided the large effort made by HMG, NEA, the cofinanciers and IDA itself to mobilize additional funding. Despite these shortcomings, IDA's performance in identification, preparation assistance and appraisal of the physical component of the Project is assessed as 'tatisfactory" 39. As seen from Part II - Table 10, HMG and NEA did not comply with some of their commitments under the Project, in particular with those commitments relating to institution building in the power sector6 and land acquisition and resettlement (Annex 5). The institutional component of the project suffered inter alia: (i) from IDA working on the basis of an optimistic view of the speed at which the sector and in particular NEA could develop; and, (ii) from an overrating of the sector's capability to implement the measures envisaged on the basis of a TA program that was only defined in its outline. However, as the sector was being moved toward a new structure, expected to be more rational and efficient, IDA's above optimism should not be held against IDA's performance in the identification/ preparation and appraisal stages of the Project. Therefore, IDA's performance in these stages are rated as satisfactory. Establishment of NEA and relative improvement in the coordination between the planning, construction and operating functions were steps in the direction of strengthening the power sector. 6 Reorganization of the sector (paras. 21 and 22), financial strengthening (paras. 23 and 24; and, Annexes 2 and 3), reduction of system losses (para. 25; and Annex 4), and training (para. 26). - 12 - 40. IDA became quickly aware of this deficiency and, together with ADB, sponsored the 1987 Power Subsector Review which, in turn, provided a basis for IDA's TA III project. This latter operation took up in detail the loose ends of the institutional component envisaged under the Project. It was completed, and the supporting credit closed in 1994. However, this created a precedent that when HMG and NEA were not in compliance of their commitments, IDA shifted the covenants to the next project and did not use its legal remedies available under the Project. The process continued under PSEP and during appraisal and negotiations of the Arun III Hydroelectric Project (HEP)7. With regard to land acquisition and resettlement, the Inspection Panel of IBRD and IDA, in its deliberations on the then proposed Arun III HEP reviewed HMG, NEA and IDA's actions on this matter and observed that 'the Marsyangdi project was completed without compliance on resettlement"8. Although IDA's performance in supervising the implementation of the physical component was satisfactory, in view of the shortcomings in the institutional component and the observation of the Inspection Panel, on balance, IDA's performance during the supervision of the overall Project cannot be judged satisfactory. F. Borrower Performance 41. In the context of preparation and implementation of the physical component of the Project, HMG (the Borrower) and MHDB (the implementing agency) strongly assisted by IDA, ADB, KfW, the PoE, and the engineering consultants, performed satisfactorily. However, in 1984/85, HAMG did not feel in a position to adjust tariffs as agreed and, thus delayed to an extent Project implementation. Since its completion by MHDB and transfer to NEA, the hydro power project has been operated and maintained satisfactorily. However, NEA is being assisted by an engineer from LI whose service is funded under a KfW grant. 42. In carrying out the institution building component of the Project, both HMG and NEA performed poorly, in part because of a lack of capability, but also to an extent for their lack of ownership of the institutional measures agreed with IDA. This less than satisfactory performance is also evident from the Borrower's failure to comply with a number of legal covenants (Table 10). On August 1, 1995, IDA informed HMG of its decision not to proceed with plans to support the Arun III Hydroelectric Project. Memorandum to the President of IDA, dated June 21, 1995, from the Inspection Panel - Annex I to the said Memorandum; paras. I to 8. - 13 - G. Assessment of Outcome 43. The emphasis of the Project clearly was on providing Nepal with a generation facility to meet its internal power demand. The limited success of the institutional component was due to HMG's and IDA's optimism about the development and delivery of the sector and NEA. This was addressed in the context of the succeeding TA III and PSEP projects which started in the middle and toward the end of Project implementation. Establishment of NEA and relative improvement in the coordination between the planning, construction and operating functions were steps in the direction of strengthening the power sector. On balance the outcome of the Project is rated as satisfactory. Hf. Future Operation 44. NEA's operation plan for Marsyangdi is based on a computerized system of planning, implementing, and monitoring plant operation and maintenance, that covers, besides the broad outlines of these aspects several hundred individual activities defined in detail. The ICR mission was able to confirm that the system (that, inter alia, includes performance related incentives for the plant's personnel) is actually implemented and that personnel seems to identify with it. I. Lessons 45. The successful operation of the Marsyangdi plant suggests that the detailed planning, execution and monitoring of operation and maintenance carried out in this plant should be applied in other facilities in NEA's system. This might involve the continuation, for a limited time, of the services of at least one specialized expatriate. 46. Project outcome confirms the lesson learned in other projects and which already for several years has been reflected in Bank policy: Projects with large, lumpy physical components should be submitted to Board consideration only after bidding for the main components has taken place. 47. The Project, as many before it, illustrates the fact that borrowers and beneficiaries tend to own physical project components of IDA projects more clearly than institutional ones, because they usually get involved with the IDA primarily to obtain financing of capital intensive facilities such as power plants. To improve the chances of success of institutional project components associated with these projects, IDA should ensure the borrowers' and beneficiaries' ownership and use its leverage in a delicate and balanced way to lead in a reasonably short time to ownership by borrowers and beneficiaries of the objectives and measures considered. The measure of leverage applied in connection with this and follow-on projects (i.e., TA III and PSEP), was adequate, as e.g. in mid-1990s, NEA is in a better financial shape than it was in the late 1980s Establishment of NEA and relative improvement in the coordination between the planning, construction and operating functions were steps in the direction of strengthening the power sector. However, the overall performance level is not sufficient to tackle the challenges Nepal - 14 - power sector faces in the second half of the 1990s. A new strategy in line with the current power sector policy guidelines of the Bank Group needs to be developed. 48. IDA created a precedent that when HMG and NEA were not in compliance with their commitments, these covenants were shifted to the next projects. 49. In its deliberation on the then proposed Arun III Hydroelectric Project, the Inspection Panel observed that "the Marsyangdi Project was completed without compliance on resettlement." IDA should ensure that the necessity to undertake rehabilitation and restoration of livelihood of the project affected families be accepted as a principle by HMG. - 15 - H: Statistical Tables Table 1: Summary of Assessments A. Achievement of Objectives 1/ Substantial Partial Negligible Not Applicable Macroeconomic policies Sector policies O O Financial objectives O U O O Physical objectives O Poverty reduction O O U Gender concerns O O O Other social objectives O O U O Environmental objectives O 1g Public sector management 0 U O O B. Project sustainability Likly Unlikely Uncertain g 0 O C. Bank Performance Highly Satisfactory Deficient Satisfactory Identification O U 2/ O Preparation assistance O U 2/ 0 Appraisal O U 2/ 0 Supervision O O3/ D. Borrower performance Highly Satisfactory Deficient Satisfactory Preparation O U O Implementation O0 Covenant compliance O O Operation (if applicable) O W O D. Assessment of Outcome Highly Satisfactory Unsatisfactory Highly satisfactory unsatisfactory O *4/ O 0 1/ Para. 27. 2/ Paras. 38 and 39. 3/ Para. 40. 4/ Para. 43. - 16- Table 2: Related Bank Loans/Credits Loan/Credit Title and Nr. Purpose Year of Status Approval Preceding Operations: 1. Cr. 600-NEP and 600-1 -NEP: Support construction of 1975 Completed in Kulekhani Hydroelectric next generation plant on 1982; audited by least cost sequence. OED 2. Cr. 1379-NEP Assistance to HMG in 1983 Completed in Technical Assistance 1I/1 identifying and preparing 1990; PCR high priority projects processed suitable for IDA or other external financing 3. Cr. 1452-NEP: Feasibility study of Karnali 1984 Completed in Karnali Preparation (Chisapani) Multi-purpose 1990; PCR Project with an ultimate processed capacity in excess of 10,000 MW Following Operations 1. Cr. 1902-NEP: Support preparatory work 1988 Completed in Third Technical Assistance for further development of 1994; ICR in (Pancheswar) least cost supply for preparation domestic needs and export. 2. Cr. 2029-NEP: Preparatory work for Arun 1989 Canceled in Arun III Access Road III hydro project. 1995; ICR under preparation 3. Cr. 2347-NEP: Improving NEA's technical 1992 Under Power Sector Efficiency and operational efficiency, implementation upgrading existing generating plant. There was an earlier project called "Technical Assistance Project" and supported by Credit 659-NEP of US$ equiv. 9.0 million approved in 1976, which ultimately did not include measures in the power sector. - 17- Table 3: Project Timetable Steps in Project Cycle Date Planned Actual Date Identification 06/79 Preparation 07/79-10/82 Appraisal 01/82 11/82 Negotiations 09/82 04/84 Board presentation 12/82 05/84 Signing 01/83 09/85 Effectiveness 12/85 01/86 Project completion /1 07/89 01/90 Credit closing 06/90 12/94 /1 Marsyangdi plant Table 4: Loan/Credit Disbursements; Cumulative (Estimated and Actual) IDA FY 85 86 87 88 89 90 91 92 93 94 95 Est.(million US$) 14.7 40.7 66.7 85.9 96.1 107.0 107.0 107.0 107.0 107.0 107.0 Act.(million US$) 0.0 6.6 13.5 53.4 65.6 70.2 75.3 85.4 90.9 96.9 98.5 Act. in%ofEst. 0% 16% 20% 41% 68% 66% 70% 80% 85% 91% 92% Act.(million SDR) 0.0 5.8 11.2 27.6 51.2 51.8 58.4 65.9 69.9 74.1 75.2 Canceled (million 23.0 SDR) Date of Final Disbursement: June 6, 1995 - 18- Table 5: Key Indicators for Project Implementation Not applicable as the Project predates the standard requirement for key indicators. However, see Annex 6 Table 6: Key Indicators for Project Operation NEA's operation plan for Marsyangdi (presently generating about 50% of the energy sent out by NEA) aims at optimally using the water available under the restrictions imposed by the role assigned to the plant in NEA's system and by proper maintenance of the facilities. It includes the preparation of: (i) yearly generation objectives based on broad assumptions about availability of water and a general program of maintenance; (ii) monthly programs based on the yearly plan but using already better defined data on demand, water availability and maintenance requirements; and (iii) the detailed daily operation program. Several hundred activities related to operation and maintenance are defined in detail, as well as planned and monitored in the context of a computerized system, which also helps define the performance driven incentives to be provided to the personnel. Table 7: Studies Included in Project The Project did not include studies. Such technical assistance was covered in the TA II and III Projects, implemented in parallel with the Marsyangdi Project. - 19 - Table 8 A: Project Costs Item Appraisal Estimate Actual/Latest Estimate (Million US$) (Million US$) Local Foreign Total Local Foreign Total Costs Costs Costs Costs Preliminary Cost and 5.6 3.0 8.6 5.6 3.0 8.6 Administration Civil Works (Lot I): 13.6 45.0 58.6 5.3 21.3 26.6 Diversion Works, Intake etc. Civil Works (Lot II): 25.0 73.7 98.7 6.1 76.8 82.9 Tunnels, Power Station, etc. Electro-mechanical 6.9 40.8 47.7 3.9 71.2 75.1 Equipment Transmission Works, 5.0 13.9 18.9 3.4 25.6 29.0 Local Distribution Construction 0.0 7.5 7.5 0.0 21.9 21.9 Supervision Technical Assistance 0.0 4.1 4.1 1.7 4.0 5.7 Mugling-Marsyangdi /1 /1 /1 2.0 0.5 2.5 Road Sub-Total 56.1 188.0 244.1 28.0 224.3 252.3 Physical Conting. 7.8 25.8 33.6 Price Conting. 11.0 34.6 45.6 Grand Total 74.9 248.4 323.3 28.0 224.3 252.3 /1 Not included in appraisal estimate. The improvement of this road was financed through the part of Credit 1478-NEP that was not disbursed after completion of the plant. - 20 - Table 8B: Project Financing Source Appraisal Estimate Actual (million US$) (million US$) Local Foreign Total Local Foreign Total Costs Costs Costs Costs fBRD/IDA: 0.0 107.0 107.0 0.0 98.5 98.5 Cofinancing institutions:(*) 1. KfW 4.0 70.5 74.5 5.2 83.4 88.6 2. Saudi Fund 0.0 25.0 25.0 0.0 21.2 21.2 3. Kuwait Fund 0.0 21.0 21.0 0.0 21.2 21.2 Domestic contribution: 1. HMG 70.9 24.9 95.8 22.8 0.0 22.8 Total: 74.9 248.4 323.3 28.0 224.3 252.3 (*) ADB's financing contribution to the Project was part of ADB's Sixth Power Project. -21 - Table 9: Economic Costs and Benefits I. The ICR carried out a computation of the IERR (or, more precisely the IFRR) using the same methodology as the SAR (Annex 24), i.e. equating the revenues with the economic benefits, which tends to underestimate these benefits. 2. The main assumptions used were as follows: - The capital costs appear in accordance with the actual expenditures. They further include renewal costs of US$ 35 million in the FYs 2015 to 2018. - Incremental generation corresponds to the actual generation until FY 1995; from then on it was assumed 400 GWh until FY 2010 and 380 GWh thereafter. These figures compare with actual generation between 400 and 450 GWh per year in the first half of the 1990s. - System losses are assumed at actual levels (i.e. about 25% of energy sent out until FY 1995, diminishing to 20% by FY 2000 (as assumed in the SAR for Arun III), and to 18% by FY 2005. - Operating costs are estimated at US$ 1.5 million per year, which, though lower than the SAR estimate (0.8% of capital costs) still represents a high 12 to 15% of NEA's total cash operating costs; thus, the computation should still be on the conservative side. - Tariffs used are: Case A: Actual average revenue per kWh sold as of the first year of operation (FY 1991), which corresponds to the assumption made in the SAR (Case 1), i.e. 1994 NRp. 2.07 or 1994 USc 4.1 per kWh. Case B: Actual average revenue per kWh sold as of FY 1995, i.e. 1994 NRp. 3.55 or 1994 USc 7.1 per kWh. The resulting IERRs are: Under Basic Operating Costs Tariffs 10% lower Tariffs 10% higher Assumptions 50% higher from FY 1995 on from FY 1995 on Case A 2.8% 2.2% 2.0% 3.3% Case B 6.7% 6.3% 6.0% 7.2% 4. Case B reflects more likely figures than Case A, as the higher revenues assumed under Case B have actually been achieved and are unlikely to suffer reductions in future. The resulting IERRs favorably compare with those determined at appraisal (6.4% for a comparable case). 5. The above sensitivity analysis demonstrates that the IERR is insensitive to substantial modifications of the operating costs, but, as could be expected, quite sensitive to variations in the average tariff level. - 22 - Table 10: Status of Legal Covenants Agreement Section Present Original Revised Description of Comments status fulfillment fulfillment covenant date date DCA 3.02(a) met n.a. n.a. HMG to employ Consultants were consultant employed. DCA 3.02(b) met n.a. n.a. HMG to employ a Claims advisor was claims advisor employed DCA 3.03(b) met n.a. n.a. MG to maintain MHDB was MHDB maintained until commissioning DCA 3.07 partly 01/86 06/86 HMG to adopt and Still pending. By 1994 met implement a plan outcome not yet for resettling satisfactory, albeit population in the under conditions immediate project established in a 1989 area review and more stringent than agreed during negotiations DCA 3.08 plan 03/86 06/89* HMG to prepare, Plan implementation prep. adopt, and became part of the but not implement a PSEP impl. catchment management plan DCA 3.10(a) met n.a. n.a. HMG to maintain Panel was maintained existing Panel of Experts DCA 3.10(b) met n.a. n.a. HMG to arrange for NEA's consultants periodic inspection carry out such of project facilities inspections under the operation plan DCA 4.01(a) met in n.a. n.a. HMG to assure that Throughout the 1980s form proper project quality of accounts only accounts are unsatisfactory maintained DCA 4.01(b) not met n.a. n.a. Audited project Performance accounts to be improved in the mid- submitted not later 1990s (see Annex 2) than 6 months after the end of the FY DCA 4.02(a) met in n.a. n.a. HMG to cause NEA Accounting deficient form to maintain proper until project only accounts completion -23 - DCA 4.02(b) not met n.a. n.a. Audited of NEA Performance accounts to be improved in the mid- submitted not later 1990s (see Annex 2) than 12 months of the end of the FY DCA 4.03(a) not met n.a. n.a. NEA to conduct its Improvements in affairs in 1990s in connection accordance with with twinning sound public utility agreement, but still practices undue HMG intervention DCA 4.03(b) met in 01/86 05/85 NEA to submit and implementation of form (broad plan) implement a plan only in the mid- maintenance plan 1990s DCA 4.03(c) substanti 01/86 06/88 NEA to develop completion of ally met and implement an implementation insurance program under PSEP DCA 4.06(a) not met n.a. n.a. NEA to achieve a (see Annex 3) since return on assets of 1986 5.5% in FY86 and 6% thereafter DCA 4.07 met in 07/86 07/86 NEA to revalue its 1986 revaluation of form assets periodically poor quality DCA 4,08 met n.a. n.a. debt service No case of limitation consultation with IDA recorded DCA 4.09 met n.a. n'a. NEA not to declare No dividends dividends declared DCA 4. 10(a) not met 01/86 09/86 HMG to submit and The issue was taken submission submission implement a plan up again in the PSEP of plan of plan for reducing (see also Annex 3) government arrears DCA 4.11(a) met in 03/86 01/86 HMG to submit a plan submitted substance plan for reducing systems losses DCA 4. 11(b) not met n.a. n.a. NEA to implement implementation loss reduction delayed; losses still unduly high (see Annex 4) DCA 4.12 met 6 months 4 months transfer of assets transferred to after after Marsvangdi plant to NEA commission. commission. NEA's assets (06/90) - 24 - Table 11: Compliance with Operational Manual Statements The ICR did not identify any deviation of substance from the relevant OMS. Table 12: Bank Resources; Staff Inputs Stage of Planned Revised Actual Project cycle /2 /2 Weeks 1000US$ Weeks 1000US$ Weeks 1000US$ Through appraisal 104.7 200.8 Appraisal - Board 89.8 188.8 Board - effectiveness /1 /1 Supervision 143.9 394.0 Completion 16.1 55.1 Total 354.5 838.7 /I included in supervision /2 not available (old project) - 25 - Table 13: Bank Resources: Missions Stage of Month/ Number of Days Specialized Performance rating Types of project cycle Year persons in staff skills problems Field represented Implementat Developmen ion status t impact Through 01/80 2 14 PE, FA -- - appraisal 10-11/80 2 12 PE, FA 01-02/82 2 18 PE, FA 07/82 2 10 PE, FA (Appr.) 11/82 4 29 PE, FA, EC Appraisal 07/83 1 6 PE -- -- through Board approval Board 07/84 1 5 PE approval 03/85 1 7 PE I I through effective-ness Super-vision 02/86 1 7 PE 2 1 06/86 3 14 PE, FA, EC 2 1 10/86 1 7 PE 2 1 02/87 1 5 FA 2 1 03/87 2 5 PE, FA 2 1 02/88 6 20* DC, PE, FA, 2 1 RS, ES, TS 11/88 5 20* SP, PE, FA, 2 1 RS, ES 02/89 1 16* SP 2 1 05/89 5 16* PE, SP, FA, ES, 2 1 RS 11/89 5 11* PE, SP, FA, 2 1 ES, RS 05/90 2 5* PE, FA 3 1 Claims, finances, losses 11/90 2 13* PE, FA 3 2 (See above) 03/91 2 15* PE, FA 3 2 (See above) 12/92 3 13* SP, PE, FA 3 2 NEA manag. and finances 05/92 2 5 PE, FA 2 2 Completion 11/95 1 ll* PE -- -- Specialized Staff: DC (Division Chief), EC (Economist), ES (Environmental Specialist, FA (Financial Analyst), PE (Power Engineer), RS (Resettlement Specialist), SP (Systems Planner), TS (Training Specialist). * Missions also providing substantial contributions to other power projects in Nepal. -26- Annex 1 NEPAL MARSYANGDI HYDROELECTRIC POWER PROJECT (CREDIT 1478-NEP) PROJECT COSTS AND COMPLETION DATES Evolution of Project Costs 1. One of the main reasons it took IDA more than four years to process the project was that it wanted to avoid the type of cost and time overruns it had experienced in the context of the Kulekhani project (75% cost overrun and 26 months delay -- Cr. 600-NEP and Cr. 600-1 -NEP), which it had supported with Credit 0600 approved in 1975. Table 1 shows how estimated costs and the commissioning date for the Marsyangdi project evolved over time. Table 1: Evolution of Project Costs and Commissioning Dates Date of Estimate Project Costs (Million US$) Commissioning Date 04/80 150 12/84 07/81 170 n.a. 03/82 213 n.a. 08/82 303 06/87 01/83 328 12/87 04/83 338 n.a. 05/84 (SAR) 323 07/89 03/86 207 12/89 03/88 250 06/90 Actual 244 03/90 2. The table clearly justifies IDA's cautious approach to the project and shows the growth of estimated costs related to the deepening understanding of the technical difficulties of the project. The probing of the Panel of Experts was a major factor in this development. By mid-1983 the costs reached a high point and induced HMG to doubt the realism of the estimates. Though actual costs came out some 20% below appraisal estimates (which was at about the same level as the maximum estimate), they still were some 70% higher than those determined in connection with the feasibility study. Actual Costs in Current Terms Table 2 sets forth the actual costs determined by NEA supported by their consultants. -27- Annex I Actual Project Costs in Million Current USS Nr. Item Local Foreign Total I Preliminary Cost 2.9 2.5 5.4 2 Administration 2.7 0.5 3.2 Subtotal 1+2 5.6 3.0 8.6 3 Civil Works (headworks) 5.2 21.1 26.3 4 Weir Control Building /1 0.1 0.2 0.3 Subtotal 3+4 5.3 21.3 26.6 5 Civil Works (headrace, power house etc.) Settlement of contractor claims 6.1 73.3 79.4 6 3.5 3.5 Subtotal 5+6 6.1 76.8 82.9 7 Hydraulic Steel Structures 1.1 18.4 19.5 8 Mechanical Equipment 0.5 13.0 13.5 9 Electrical Equipment 2.1 35.6 37.7 10 O&M Equipment /2 0.1 4.0 4.1 11 Gate Control Equipment /1 0.1 0.2 0.3 Subtotal 7+8+9+10+11 3.9 71.2 75.1 12 Transmission Lines 1.3 3.1 4.4 13 Substations etc. 2.0 22.1 24.1 14 Communications Link 0.1 0.4 0.5 Subtotal 12+13+14 3.4 25.6 29.0 15 Construction Supervision -- 21.9 21.9 16 Panel of Experts, Claims Specialist Supervision of Loss Reduction Program /13 -- 0.7 0.7 17 Training 0.1 2.3 2.4 18 1.6 1.0 2.6 Subtotal 16+17+18 1.7 4.0 5.7 Grand Total 26.0 223.8 249.8 /1 Contracted in 1995 /2 Not explicitly included in SAR estimate /3 Estimated at about 20% of total TA for loss reduction. -28- Annex 1 Costs in Constant Terms, Quality of Cost Estimates Using actual exchange rates for the conversion of local costs into NRs, the GDP deflator to adjust convert local current costs into constant costs, and the MUV index to achieve the analogous conversion of foreign costs, leads to the results summarized in Table 3 below. Table 3. Comparison of Costs in Constant Terms Local Foreign Total Estimated Base Costs 56 188 244 in million 1984 US$ Actual Costs 24 176 200 in million 1984 US$ Increase in Terms of -63% -6% -18% the Base Costs The weakness of the above calculation is the distribution of the costs between local and foreign ones and the assumed distribution of the costs over time. However, a sensitivity analysis involving once the doubling of the assumed local costs and once a grossly different distribution over time lead to total costs in 1984 terms that differ from the above by less than 2%. Therefore, the conclusion is that the total costs in constant terms are valid in the measure that the actual costs in current terms are reliable. It appears further that - total costs in current and in constant terms are in the order of 23% and 18% lower than those estimated at project appraisal; - the civil works were substantially overestimated at appraisal whereas actual costs of equipment were in the order and partly (especially electrical equipment) even higher than the corresponding estimates; and - the construction supervision generated expenditures that were nearly three times those expected at appraisal; they represented some 12% of the total cost directly attributable to the plant construction and installation, which is high but not extraordinarily so; the reason for this seems to be the long preparation time and probably a low estimate to start with. -29- Annex 2 NEPAL MARSYANGDI HYDROELECTRIC POWER PROJECT (CREDIT 1478-NEP) IMPLEMENTATION COMPLETION REPORT SUBMISSION OF AUDITED ACCOUNTS 1. Para. 2.10 of the DCA called for the submission of (i) audited project accounts and full reporting for MHDB within six months of the end of each FY, and (ii) audited financial statements and full report for NEA within 12 months of the end of the FY. There also was an agreement that NEA's unaudited financial statements would be submitted within six months of the end of a FY. 2. The table below sets forth expected and actual dates of submission of the accounts as they appear in supervision reports. There often are discrepancies among successive reports, as audits appear as submitted in one report, obviously when submission seemed imminent, and in the next report the same statements appear as still not available. There also is a substantial delay between the completion of the audit and its certification by the Auditor General, which contributes to the uncertainty about the definition of the date of completion of an audit. Anticipated and Actual Dates of Audit Submission FY Project Accounts NEA Accounts Due Actual Due Actual 86 01/87 02/87** 07/87 12/88 87 01/88 04/90 07/88 04/90 88 01/89 04/90 07/89 11/90* 89 01/90 03/91 07/90 03/91* 90 01/91 11/91 07/91 05/92 92 *** 07/93 10/93 94 *** *** 07/95 08/95 * Audited but not yet certified by Auditor General ** Accounts not in accordance with standard accounting practices ***Project completed, assets transferred to NEA. 3. The above table shows that, though performance improved over time, by project completion, NEA supplied audited statements at least six months late, and this does not even include the further delay related to the certification of the audits by the Auditor General. However, more recently, NEA's performance with respect to the timely submission of audited (but not yet certified by the Auditor General) financial statements has become satisfactory as the accounts for FY93 were submitted before the limit date and those for FY94 within six months of the end of the FY. The quality of the accounts has also substantially improved and are less the subject of observations by the auditors. -30- Annex 3 NEPAL MARSYANGDI HYDROELECTRIC POWER PROJECT (CREDIT 1478-NEP) NEA RECEIVABLES FROM HMG 1. The Borrower was expected to submit by 01/86 a plan for reducing NEA's Government receivables. By 03/86 the submission date was postponed to 05/86. In 07/86, IDA threatened remedial action unless a proposal was available to IDA before 09/30/86. The submission occurred at this limit date. 2. Supervision missions consistently checked on the development of these accounts receivable, but the resulting figures appearing in successive supervision reports are to an extent contradictory as set forth in the table below. Presumably, the main reasons for this were, that the accounts were unreliable in the first place and that various values for a given FY refer to different points in time in the course of that year. It is further likely that the figures appearing under sales to Government cover the entire public sector and not only Government proper. 3. Even taking into account the uncertainty reflected in the table, it is evident that Government and NEA have, until now never complied with the covenant. However, in FY89 a substantial improvement took place, when NEA implemented an action plan agreed with IDA. In connection with the 1992 PSEP, the parties agreed on a new action plan implying a methodical approach to ensure that within one year: (i) all consumer ledgers would be updated, (ii) correct consumer accounting procedures rigorously enforced, (iii) ageing analysis done, and (iv) collection performance monitored and maintained at satisfactory levels. At the same time, the parties agreed that Government receivables would not exceed three months billings instead of the two months' billings limit set in connection with the Project (SAR PSEP, paras. 4.13 to 4.15). In FY93 NEA nearly achieved this goal. -31- Annex 3 Government Receivables Arrears Billings Arrears/Billings NRs.Million NRs.Million Mnths of Billings FY87 Total 129 452 3.4 (Sup. Rep. 11/88) Gov. 41 72 6.8 non-Gov. 88 380 2.8 FY87 Total 179 452 4.8 (Sup. Rep. 06/89) Gov. 39 15 31.2 non-Gov. 140 437 3.8 FY88 Total 218 541 4.8 (Sup. Rep. 06/89) Gov. 58 20 34.8 non-Gov. 160 521 3.7 FY88 Total 256 551 5.6 (Sup. Rep. 06/90) Gov. 58 19 36.6 (Sup. Rep. 12/89) non-Gov. 198 532 4.5 FY89 Total 264 694 4.6 (Sup. Rep. 06/90) Gov. 51 24 25.5 non-Gov. 213 670 3.8 FY89 Total 218 707 3.7 (Sup. Rep. 12/90) Gov. 38 69 6.6 non-Gov. 180 638 3.4 FY89 Total 309 707 5.2 (Sup. Rep. 01/92) Gov. 38 69 6.6 nor-Gov. 271 638 5.1 FY90 Total Not given in the supervision report (Sup. Rep. 0 1/92) Gov. non-Gov. FY91 Total 477 961 6.0 (Sup. Rep. 01/92) Gov. 31 76 4.9 non-Gov. 446 885 6.0 FY91 Total 437 961 5.5 (Sup. Rep. 07/92) Gov. 37 76 5.8 non-Gov. 400 885 5.4 FY92 Total 593 1441 4.9 (NEA statement Gov. 168 132 15.3 received in 11/95) non-Gov. 425 1309 3.9 FY93 Total 485 1794 3.2 (NEA statement Gov. 49 164 3.6 received in 11/95) non-Gov. 436 1630 3.2 FY94 Total 538 2437 2.6 (NEA statement Gov. 83 201 5.0 received in 11/95) non-Gov. 455 2236 2.4 -32- Annex 4 NEPAL MARSYANGDI HYDROELECTRIC POWER PROJECT (CREDIT 1478-NEP) REDUCTION OF SYSTEMS LOSSES 1. After the sector's first, unsuccessful, attempts at reducing losses in the first half of the 1980s, ADB, in the context of its second power project in Nepal, provided funds for TA to study the losses and the concrete measures to reduce these from a level in excess of 30% to some 15% in 1987. Before the start of the Project, NEC had already created in its own organization a department responsible for the reduction of system losses and was said to enact by-laws dealing with electricity theft. Under the Project, IDA made available funds for studying a second phase program with less ambitious targets. British Electricity International (BEI), the consultants hired under the ADB project submitted a draft program, which allowed IDA to consider that NEA had met the initial part of Section 4.11 (a) of the DCA calling for NEA to submit such a draft program by 03/31/86. The contract for the detailed design and the supervision of the implementation of the second phase was signed in 08/88 and work started in 03/89. Therefore, as set forth in further detail in the table below, NEA was unable to comply with several targets agreed with IDA. 2. The main reasons for the low rate of meeting the targets seem to be: (i) the sector's re- organization that slowed down the sector's operations until the new setup was reasonably functional, (ii) NEA's shortage of local funds, (iii) low priority accorded by NEA to the losses reduction program, due to weak management, limited ownership of the program by NEA management, and slow procurement. -33- Annex 4 Losses Reduction, Targets and Achievements Targets Achieved Observations Overall Losses: Achieved in - FY86: 24% 30% FY93: 26% - FY91: 18% 26% FY94: 27% FY95: 26% Losses in Central Achieved in System: FY92: 32% - FY86: 25% n.a. FY94: 28% - FY91: 20% 32% In FY95 the target was revised to 20% by FY 2000 (SAR Arun III). Metering all n.a. At the time, NEA's records were not reliable customers by enough to substantiate the percentage of metered 09/30/86 connections. Indications are that target was substantially met (Sup.R. 11/86). Phase II of the program was to address shortcomings in metering management and recording). NEA actively pursues the program. Checking and sealing n.a. In 06/89, targets were changed to achieving under of the meters of all Phase II of the program: large customers by - inspection, control, and calibration 06/88, and rectifying of 40,000 connections; and recalibrating all - replacement of 25,000 meters; and meters by 06/90. - sealing all meters. The program is continuing. Indications are that the set goals were substantially achieved, though later than anticipated. -34- Annex 5 NEPAL MARSYANGDI HYDROELECTRIC POWER PROJECT (CREDIT 1478-NEP) LAND ACQUISITION AND RESETTLEMENT Background 1. The SAR (paras. 3.22 and 3.23) sets forth that the Project would require the acquisition of about 69 ha of land from some I10 land owners /1. An additional 5 ha were needed for the quarry. By the presentation of the Project to IDA's Board of Executive Directors, MHDB had purchased 7 houses and 49 ha of land and already compensated the owners of 36 ha. It further envisaged to provide employment for those whose livelihood would suffer most by the Project. The SAR further states that no tribal people would be affected. 2. To cover the above aspects of the project, the DCA (Section 3.07) states that the Borrower, by 01/31/86, would adopt and thereafter implement a "rehabilitation program satisfactory to the Association for those persons dislocated as a result of' the project. A side letter to the DCA specifies that the Borrower would adequately compensate the 110 /2 land owners mentioned above and about 8 former house owners such as to allow these to continue in their work and profession. It was understood at the time that about 65 of the affected land owners would retain land in excess of I ha (considered to be the minimum required for sustainable farming), about 42 would remain with less than one ha, and some three families would be left landless, but MHDB would provide employment for them /3. Thus, strictly speaking, the Credit documents did not covenant the resettlement and the development of a resettlement plan: Only a rehabilitation program for the affected people was legally called for. This, of course, does not mean that IDA understood that a resettlement plan was unnecessary. But, it seems that, before credit approval, IDA was not particularly forceful in conveying this message to the Nepalese side. It is worth mentioning that, during preparation and implementation of the Project, IDA's codified policies on environmental issues evolved. Indeed, OMS 2.36 on environmental aspects of Bank work was issued in 05/84 (i.e. about the time IDA's Board approved the Project) and OD 4.30 on involuntary resettlement in 06/90. Both directives replaced less explicit and weaker ones. 3. At the time of the Project's preparation and early implementation, HMG's policy on resettlement and land acquisition for major projects included the following main points: - The project entity carries out a survey which is sent to MWR and, after decision, to the Chief District Officer of the Ministry, who publishes a notice informing landowners about the land acquisition. /1 The figure became 186. /2 Actually 186. /3 According to an 11/14/86 internal memo on resettlement issues, at that time, the Project left about 37 families landless or with plots of clearly uneconomic size. For as much as 50% of the affected families the sale of the land needed for the Project threatened to reduce them to marginality. -35 - Annex 5 - A price assessment committee including district and project representatives is constituted to fix the compensation rate. In case of disagreement, there is a possibility to appeal to the Zonal Commissioner. (By mid-1986, 51 persons had appealed). - HMG compensates in cash and lets people resettle in the place of their choice. 4. The Project consultants in their environmental impact assessment of 1981 had proposed that the affected families benefit of 15% disturbance allowance on top of the market value of the land, special assistance in hardship cases, priority for employment on the project work force, and assistance for increasing production on the land situated in the immediate vicinity of the project but not affected by construction. Action Taken During Project Implementation 5. The lack of precision in defining the issue at the preparatory stage and the low priority that the land acquisition issue had as well as other factors that delayed the Project resulted in HMG's inability to submit a program for the "rehabilitation" within the agreed time frame. Nevertheless, in mid-1986, MHDB produced a preliminary proposal, which lacked an action plan, but made clear that a total of 226 families would be affected and suggested that there would be no difficulties with claims for compensation. 6. In 11/86, IDA started a major effort to push this issue into the right direction. In particular, it made clear that IDA did not consider cash compensation adequate for achieving the objectives of resettlement in the context of an IDA Project, as it fails to assure that the affected people can continue to make their livelihood through the activities they pursued before the Project started. IDA, taking a more active stance, suggested the preparation of a resettlement plan based on a strategy of agricultural development, watershed management programs (also to be studied in the context of the Project) and the creation of employment opportunities. To achieve this IDA proposed the preparation of a plan including (i) a census of the affected and host populations, (ii) an inventory of public and common infrastructure and resources, (iii) identification and assessment of new land and housing sites, (iv) survey of potential host population, (v) definition of resettlement policy to be applied specifically in the context of the Marsyangdi development; (vi) setting up and training the organization required for the specific resettlement, (vii) defining development objectives and policies for the new sites and the possible new production systems, (viii) elaboration of programs for both agricultural of non-farm employment in the affected areas and the new sites, (ix) planning of infrastructure needed for the above development, and (x) definition of time tables coordinated with those for the hydroplant, of the costs, and of the funding. /4 In view of the links between the resettlement and the watershed development plan, IDA suggested that the consultants for the latter area should also prepare the resettlement plan. HMG delayed the decision on this point and ultimately refused to proceed as it did not wish to depart from its established policy and consider compensation other than in cash for the people affected by the project. 7. The controversy continued through 1988, when the Borrower agreed to have a Nepalese consultant carry out a socio-economic study on site. This analysis became available in 1989. The consultants were able to contact 78% of the affected households, which suggests that only a few families had left the area in the meantime. / 4 Internal note of 11/16/86. -36- Annex 5 8. Compensation had been given in cash only, though a few landowners had asked for land. Rates paid were below market prices, and most sellers were dissatisfied since they were unable to purchase alternate land with the compensation provided. Most people whose land was acquired for the project had no legal knowledge, and, in particular, no knowledge of their right to appeal the land deals. There is also evidence that the later acquisitions were at better prices than the earlier ones, which produced further dissatisfaction. 9. The study identified 42 hardship cases out of the 174 PAPs found on site, and recommended a series of rehabilitation measures including redistribution of land acquired but no longer required by MHP, and income restoration programs. Although IDA follow-up missions continued to ask for a detailed plan for carrying out the recommended scheme, the only outcome was grudging recognition that perhaps some hardship cases still needed to be dealt with. The Borrower thinned the number of hardship cases down to 27 but until 1995 had still not undertaken any rehabilitation activities other than the initial cash compensation paid between 1978 and 1985 when the land was acquired. Further Action 10. An IDA mission undertook field investigations to review the situation of project affectees in March 1996, and held discussions with NEA to determine appropriate remedial measures. The findings of that mission are that (1) NEA commissioned an internal study in 1995 to develop a remedial plan but the study only covered the 27 households accepted by NEA from the 1989 study as being hardship cases; (2) the plans being proposed by NEA were inadequate to address outstanding resettlement issues; (3) all affected households interviewed confirmed that land values paid to affectees were inadequate for them to purchase alternate land; (4) field investigations validate the basic data provided in the 1989 consultant's report but suggest that, if anything, their estimate of hardship cases was an underestimate; (5) the actual number of seriously affected households is 48, of which 7 have been rehabilitated through employment of one person from each family on the project, but 41 still need rehabilitation; (6) MHP has 9.5 ha of land (8 ha agricultural, 1.5 ha land adjacent to the road) acquired for MHP but no longer needed by the project; (7) NEA has been advised to re-design a manageable rehabilitation plan using the 1989 survey data as baseline information, with three basic rehabilitation packages: re-allocation of the 8 ha of agricultural land, development and redistribution of the 1.5 ha roadside plot as a commercial area for redistribution, and provision of livestock loans, to seriously affected households, leaving the choice of rehabilitation package to individual households. NEA has expressed an interest in implementing the rehabilitation program, preferably through an NGO, to bring MHP to closure. The issue was taken up, once again, with IDA's letter dated May 7, 1996, addressed to HMG. -37- Annex 6 NEEL PROJECTED AND ACTUAL VALUES OF NEAS PRINCIPAL OPERATIONAL AND FINANCIAL INDICATORS FY 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 Total Enery Sent Out (GWh) -Projected 471 544 610 676 722 770 Actual 489 571 628 672 774 906 981 963 1031 117 Sales (GWh) -Projected 349 413 469 525 569 615 - Actual 342 403 465 496 548 669 737 709 757 830 Salesroth - Projected 12.4% 183% 13.6% 119% 8.4% 8.1% 94% - Actual 148% 178% 15.4% 67% 105% 22.1% 102% 3 8% 6.9% 9.6% Peak LoadM -Projected 124 144 159 170 182 198 210 231 -Actual 110 126 141 150 176 204 216 214 231 244 0.00% 0 00% Peak Load Growth -Projected 308% 15.6% 106% 69% 6.8% 92% 5.8% -Actual 37.5% 145% II 9% 64% 17.3% 159% 5.9% -09% 7.9% 56% Losses in Terms of Energy Sent Out -Projected 257% 24.1% 23 1% 22.3% 212% 201% - Actual 30.0% 29.4% 26 0. 26.2% 29.2% 26.2% 24.9% 26.4% 26.6% 25.7% Load Factor - Projected 43 3% 43.2% 43 8% 45.4% 454% 44.3% 47.3% 47.3% -Actual 50.7% 51.7% 50.8% 51.1% 50.2% 507% 51.9% 514% 50.9% 52.3% Averte Revenue per kWh Sold (Current NRp) -Projected 1 35 135 135 1.35 1 59 1.76 -Actual Ill 112 118 1.39 139 1.44 1.96 2.53 3.27 3.96 Averae Revenue or kWh Sold (Constant 1994 NRp, -Projected 2.16 202 1.91 180 188 2.10 - Actual 188 185 1.91 2.19 2.13 2.07 2.57 2.90 3.27 3.56 Personnel Cost per kWh Sold (Constant 1994 NRD.) -Projected 0.20 0.17 0.16 017 0 15 013 -Actual 045 046 049 0.54 054 0.60 0.53 0.52 0.46 Operation & Administration Cost per kWh Sold (Constant 1994 NRp.) -Projected 0.26 026 022 019 0.34 030 - Actual 0.26 027 0.22 027 026 021 0.21 026 0.28 Total Cash Oneratin Cost ncr kWh Sold (Constant 1994 NRp. - Projected 050 0.46 041 055 0.49 044 -Actual 081 0.78 0.95 1 33 129 092 109 1 29 Ill Rate of Return on Net Fixed Assets in Operation - Projected 54% 60% 6.0% 6.0% 6.0% 6.0% -Actual 5.8% 4.4% 3.0% 3.2% 2.9% 19% 0.5% 0.2% 1.6% Operatmg Ratio (Operatna expenditures includina taxes/Total operating revenues3) -Projected 60.7% 56.8% 57.7% 63.3% 51.2% 48.4% -Actual 63.3% 689% 785% 1455% 159.4% 137 7% 106.9% 102.6% 82.7/. Debt Service Coveran" Ratio -Projected I9 16 1 8 1 8 1.0 14 -Actual 1 6 1.2 16 1 4 1.6 3.3 6.3 12 1.7 Debp/Equity Ratio - Projected 40/60 47/53 46/54 45/55 52/48 52/48 - Actual 46/54 34/66 29f11 11189 13/87 33/67 34/66 37/63 39161 Current Ratio -Projected 1 8 2.1 1 7 1 8 5 4 5.0 -Actual 09 25 13 12 11 13 23 20 47 Accounts Receivable (months of billings -Projected 3 3 3.3 3.3 3 3 37 4.3 -Actual 38 44 5.2 5.0 56 89 6.0 3.0 3.0 Accounts Payable (months of billings) -Projected 09 0.8 0.7 1.1 09 0.9 -Actual 103 40 3.9 6.5 4.2 2.8 2.0 3.0 30 * The indicators which use balance sheet data are estuiates for FY93 and FY94 and need confirmation thorugh analysts ofNEA's audited financial statements. -38- Annex 7 NEPAL MARSYANGDI HYDROELECTRIC POWER PROJECT (CREDIT 1478-NEP) THIRD TECHNICAL ASSISTANCE PROJECT (CREDIT 1902-NEP) IMPLEMENTATION COMPLETION REPORT (MISSION (NOVEMBER 1995) AIDE-MEMOIRE 1. Between November 8 and November 17, 1995, Mr. Rene A. Ribi, Consultant to the Energy and Infrastructure Division of the World Bank's South Asia II Department, stayed in Nepal to carry out the field work related to the preparation of IDA's ICRs for the two above projects. The corresponding credits were closed in December 1994 and June 1994, respectively. Mr. Ribi worked under the guidance of the Task Manager, Mr. Argun Ceyhan, who was in Nepal on a different mission. Mr. Ribi would like to express his appreciation to the Electricity Development Center (EDC) of the Ministry of Water Resources (MOWR) and Nepal Electricity Authority (NEA) for their assistance and cooperation during the various discussions and the visit to the Marsyangdi plant that took place during his stay in Nepal. IDA's final comments will be sent by Management following its review of the present Aide-memoire 2. The objectives of the mission were: - to complement the data base for the report; - to discuss with NEA (a) the operation plan for Marsyangdi which NEA still has to provide and (b) the measure to which the goals set in the projects were achieved, and - to provide advice and support to the entities of the Borrower in preparing their own contribution to the ICR; Data Base 3. NEA has provided most of the data requested in IDA's letter of April 4, 1995. EDC and the concerned departments of NEA have agreed to make available, by the end of November 1995, the remaining information called for. /1 /1 On November 15. only (i) the breakdown of the Pancheswar component of the TA III project, (ii) the final costs and financing of the various components of TA III for which NEA was responsible, and (iii) the breakdown of NEA's operating costs for the past three fiscal years and a statement of the development of Government receivables were still outstanding. Part of this information is likely to become available before Mr. Ribi's departure from Kathmandu. -39- Annex 7 Operation Plan NEA has submitted a draft operation plan which the mission discussed with the main author. It was agreed that the final plan, which NEA is to provide by mid December 1995, would outline in some more detail than the draft the sophisticated arrangements directing operation and maintenance of the plant, as these arrangements should be a model for the management of other facilities in the system. Outcome of the Marsyangdi Project (Credit 1478) 4. IDA's ICR for this project, which was co-financed by His Majesty's Government (HMG), Kreditanstalt ffr Wiederaufbau (KfW), the Saudi Fund, the Kuwait Fund, and IDA, will substantiate IDA's following main findings that crystallized before and during the mission: - NEA, with support from HMG achieved the physical objectives of the project with a high degree of success: (i) though the plant and its associated facilities was completed seven months late with respect to the schedule in the Staff Appraisal Report (SAR) their cost was some 20% lower than set forth in the appraisal estimate. (ii) the plant's output grew from 331 GWh in the first full year of operation (1990) to 438 GWh in 1994 which corresponds to a high plant utilization of 72% and represents about half the total generation in NEA's system. At the time of the mission's visit, the installation was running at over 75 MW, the name plate capacity being 69 MW. - The implementation of the institutional measures included in the project was less successful than that of the project's physical component: (i) NEA institutional strengthening was slower than expected; (ii) the sector finances remained weak throughout project implementation; however, in the context of follow-on projects revenues dramatically increased, accounting and auditing improved both in quality and timeliness, and accounts receivable from Government fell substantially; (iii) most of the postponed maintenance which was planned to be carried out during project implementation had to be delayed until after its completion; (iv) the loss reduction program failed to achieve the targets agreed at appraisal and was taken up in follow-on projects; and (v) training developed only slowly under the project; however, progress improved substantially once the main disagreements between Nepal and IDA on the one hand and the consultants on the other were eliminated, which occurred in later stages of the Third Technical Assistance Project (Pancheswar) - TA III - . The subsequent Power Sector Efficiency Project (PSEP) included further training in line with the previous programs. - Major complementary work at Marsyangdi, mostly financed or expected to be financed by KfW, includes: (i) the modifications needed at the manifold where hydraulic instability -40- Annex 7 occurs and creates unplanned fluctuations in the turbine output; this work is presently under negotiations with a member of the consortium that supplied the manifold; the cost is likely to be substantially higher than the DM I million still retained from the contractor; completion is expected for the first semester of 1997; (ii) at the weir site, NEA is constructing facilities that will house central control equipment for all the gates in the weir/sedimentation pond/intake area and the laboratory for water and sediment analysis, and (iii) as a glacier lake in the upper reaches of the Marsyangdi catchment is growing and could become a threat to the plant, a study of the problem is planned for 1996. Outcome of the TA III Project (Credit 1902) 5. The ICR for the TA III project co-financed by HMG, KfW, and IDA will discuss the following main findings by IDA: The first project objective was to assist HMG in further ascertaining the least cost way to develop electricity supply to meet expected local and export demand. It was achieved through the conclusion, about one year later than anticipated at project start, of the field investigations (on the Nepalese side of the Nepal/India border) for the Pancheswar development. The second broad project goal was to advance the design of Arun III which was achieved through the timely completion of the engineering for the facilities and of the power evacuation study. The technical assistance measures supporting the sector's institutional development were consolidated in a package that was executed in the context of the twinning arrangement between NEA and Electricite de France (EdF) concluded to follow up on recommendations set forth in the 1987 Power Sector Review sponsored by ADB and IDA. At first, progress was disappointing. Only after the TA III project was essentially completed, did the measures started under the project have a measure of success in the context of follow-on projects. The 10-year transmission and distribution plan was successfully completed and fulfills the purpose for which it was prepared. The same applies to the long-run marginal cost and tariff study which is one of the main bases used by NEA for its planning. The rural electrification plan, in turn, led to the inclusion of the implementation of a test scheme in the Marsyangdi catchment under the PSEP; all these studies were somewhat hampered by the limited response of NEA to their early results. The design of the Trisuli-Devighat upgrading was the only item not specified at project start. However, it fit the concept of the project and formed the basis for the implementation of the improvements under the PSEP. -41- Annex 7 Borrower's and Beneficiary's Own Assessments 6. Section 3.03 of the Development Credit Agreement (DCA) for Credit 1478 requires NEA to prepare its own assessment and analysis of the project's implementation and outcome. The essential content of such a report was discussed with NEA's project manager and it was agreed that NEA would complete its contribution to the ex-post evaluation of the Marsyangdi project by mid December 1995. 7. The DCA for Credit 1902 does not include the section on the preparation of an ICR, but the mission considers that it would be valuable to prepare such an assessment, the more so as the responsibility for the various project parts were widely disseminated within the sector and most of the persons who, at the time, were responsible for the components are either in new positions or have left the organization. Thus, the institutional memory for this project is vanishing fast. -42- Annex 8 NEPAL MARSYANGDI HYDRO POWER PLANT (CREDIT 1478-NEP) OPERATION PLAN 1. Operation Plan for Future Operation - The plant will be operated in the future based on the following operation plan. Operation Plan 2. The plan is in operation since 1989 and has generated energy as follows (Annex 3): Year MWh 1990 331,360 1991 417,929 1992 401,614 1993 441,343 1994 438,332 1995 453,000 (estimated) During the last six years of operation many operational problems have been faced (refer Section 3) and based on this experience suitable a operation and maintenance plan has been established to ensure sustainability of its operation. The established operation and maintenance plan can be summarized as follows: Operation 3. Water Level -During the rainy season the reservoir will be operated at its minimum level of 333.00 meter with continuous sluicing to maintain the bed load as low as possible. During the dry season and whenever the sediment content is insignificant the reservoir will be operated at its maximum level of 337.5 meters. 4. Flushing. During the rainy season the flushing of sediments from the settling basin will be done as per requirement without delay. About 20 flushings are required during one rainy season. The time required for one flushing operation has been optimized to about 2.5 hours by proper coordination of the flushing procedure. The time estimated during design phase was about 5 hours. Total time of flushing in a year has also been reduced from 254 hours design period to about 50 hours in actual operation. Flushing requirement is determined through daily echo sounding. Regular dredging of the compensating pondage shall be carried out. -43- Annex 8 5. Debris. The removal of debris shall be carried out regularly with the operation of the Intake Rating machine. Debris is also being partially retained in the floating boom installed at the entrance to the compensating pondage. This retained debris is flushed out at the time of each flushing. 6. Unit Operation. Output of the individual unit shall not be below 16 MW for long period of time. Monitoring Activities 7. Sediment. Major effort has been invested in the accurate measurement of sediment concentration, deposits and effects to the structure and equipment. A comprehensive program including daily samplings echo sounding, and their analysis is being executed by the regular plant personnel. 8. Structures. Yearly measurements are taken for anchors fixed on powerhouse slope for analysis of their settlements. This year only these anchors have been retensioned to their original values. Now slopes are stabilized and measurements shall be taken in long intervals. Maintenance and Spare Parts 9. Maintenance System- A computerized preventive maintenance system has been established. 1,774 (at the end of September 1995 )work orders (detailed instruction on procedure for execution of the maintenance work) covering all the important equipment have been prepared for easy, timely and reliable execution. The execution work is constantly monitored and up to date records are maintained. During the period of April 95 to September 95 following numbers of maintenance activities have been executed by different sections: civil maintenance - 463; mechanical maintenance 1,532; electric maintenance - 1,745; operation 32. Maintenance activities to be performed by different sections are distributed to the section's chief regularly and after their completion reports are entered to the computer by the administrator of the program. Status of the maintenance work can be received by the computer any time. This system is quite new for NEA. It is very effective and efficient. Because of its effectiveness NEA has initiated to use it in another powerhouse also. Major Preventive Activities 10. Turbines. Based on our experience it has been found appropriate to carry out the preventive overhauling of a unit every 3 years. In this overhauling the critical parts eroded and damaged are repaired and protected extending its useful life. The complete overhaul -44- Annex 8 work is carried out by the plant personnel. Special high resistance materials (which have been selected after trail and run of many other products) are used in repair. 11. Sluice. The ordinary steel lining of the sluice was suffering very high degree of erosion due to high velocity of sediment and presence of boulders resulting in a major maintenance every year. Special high resistance materials has been introduced for lining and it is expected that frequency of maintenance will be increased. 12. Concrete Ogees. Special material has also been used on ogees for their easy repair. 13. Tailrace River Training. Accumulation of boulders and sediments on the tailrace outlet affects and available head specially during the dry season. Tailrace river training was done by the plant personnel with available equipment during the year 1994. It was very effective and now regular river training shall be carried out. 14. Availability of Spare Parts. A computerized Inventory Management system has been established. Replacement of consumed part with procurement as per minimum stocks established and new requirements determined is being regularly made to ensure their availability. 15. Staffing of the Plant. The plant is being operated by NEA staff There are 120 in number out of which approximately 60 are technical. Almost all the technical staff have been trained for operation and maintenance. A performance oriented incentive scheme has been introduced in this plant. This scheme is contributing a lot to motivate the staff for work. Several operations and maintenance parameters including generation target, timely preventive maintenance are monitored through this scheme. It is the intention of NEA to continue this scheme of reward in fiture also. 16. Operation and Maintenance Budget. An yearly budget is allocated to the plant based upon the proposal of station manager. The station manager is solely responsible for the administration of the approved budget and in case of unforeseen requirements additional budget is provided to the plant. 17. Performance Indicators. For monitoring the performance of the project in future NEA maintains the following indictors: i) Plant load factor ii) Generation factor (Actual/planned) -45- Annex 8 18. Additionally at the plant level the following indicators will be maintained: i) Productivity factor (KWh/person) ii) Direct cost of generation Rs/KWh iii) Forced outage hours iv) Planned outage hours v) Reliability factor (no of days-no of trip/no of days) - 46 - Annex 9 NEPAL MARSYANGDI HYDRO POWER PLANT (CREDIT 1478-NEP) COMMENTS PROVIDED BY KREDITANSTALT FUR WIEDERAUFBAU (KfW) KfW's letter dated February 8. 1996, addressed to the World Bank Subject Financial Cooperation with Nepal (i) Marsyangdi Hydro Power Project (Cr. 1478-NEP) (ii) Third Technical Assistance Project (Cr. 1902-IN) Draft Implementation Completion Reports Your letter dated January 25, 1996 Dear Sirs: This is to acknowledge with thanks the receipt of the Draft Implementation Completion Reports in caption. While we find the reports in general very comprehensive and well conceived we would like to provide a short comment on the issue specified below. With regard to "sustainability" of Marsyangdi Hydro Power Project (p.8, para 28 of the Report) it may be worthwhile to mention that about 80 km upstream from the weir a glacier lake (Thulagi lake) has developed over the last 10 years. As a result, the risk of a major glacier lake outburst flood (GLOF) became a potential threat to the power plant. Even though a GLOF may not have a direct impact on the weir, serious operational problems may result from such an event, owing to the increased bed load arriving at the intake over many months after the overflow. Therefore, monitoring of the development of the lake, and, if necessary and feasible, appropriate measures to lower the water level should be given due consideration. As you may already know, KfW, upon request of the German Government, shall commission a study on this issue, which will be financed under the existing Studies and Expert Fund. In addition we would like to point out that KfW is initiating a feasibility study on a new Load Dispatch Center at Kathmandu which shall be equipped with SCADA. The LDC should also contribute to an improved capacity utilization of the power generated at Marsyangdi. (Signed by Mr. Heidt and Dr. Mohnhaupt)    IMAGING Report No: 15653 Type: ICR