Rwanda - Power Project

Document of The World Bank FOR OMCIAL USE ONLY MICROFICpE COPY Report No. 10492 Report No. 10492-RW Type: (PCR) MAUPRIVEZ,/ X31709 / T9 069/ OEDD3 PROJECT COMPLETION REPORT RWANDA POWER PROJECT (CREDIT 1495-RW) APRIL 3, 1992 Industry and Energy Operations Division South-Central and Indian Ocean Department Africa Regional Office This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENTS 1983 US$1 = RwF 97 (Appraisal) 1984 US$1 = RwF 100.2 1985 US$1 = RwF l01.3 1986 US$1 = RwF 87.6 1987 US$l = RwF 79.7 1988 US$1 = RwF 76.4 1989 US$1 = RwF 79.9 MEASUREMENTS k'' Kilovolt = 1,000 volts kW Kilowatt = 1,000 watts MW Megawatt = 1,000 kilowatts TOE Ton of Oil Equivalent = 10,500,000 kilocalories ACRONYMS AND ABBREVIATIONS Bank World Bank CCCE Caisse Centrale de Coop6ration Economique DM DeutBch mark EDF Electricite de France Electrogaz Etablissement Public de Production de Transport et de Distribution de 1'Eau, d'Electricit6 et de Gaz Government Government of Rwanda HV High Voltage MPWE Ministry of Public Works and Energy MV Medium Voltage PCR Project Completion Report Project Rwanda Power Project (Credit 1495-RW) PPF Project Preparation Facility SAR Staff Appraisal Report SAUR Societ6 d'Amenagement Urbain et Rural SDR Special Drawing Rights UNDP United Nations Development Program BORROWER'S FISCAL YEAR January 1 to December 31  THE WORLD BANK FOR OMCIAL USE ONLY Washngton, D.C. 20433 U.S.A. Office of Diretot-GeieraI Operatons Evaluation April 3, 1992 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Completion Report on Rwanda Power Proiect (Credit 1495-RW) Attached, for information, is a copy of a report entitled "Project Completion Report on Rwanda - Power Project (Credit 1495-RW)" prepared by the Africa Regional Off ice, No audit of this project has been made by the Operations Evaluation Department at this time. Attachment This document has a restticted distribution and may be used by recipients oniy in the performance| of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.|  FOR OFFICIAL USE ONLY PROJECT COMPLETION REPORT RWAND>A POWER PROJECT (LOAN 1495-CH) TABLE OF CONTENTS Page No. PREFACE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .i EVALTIATION SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . ii PART I: PROJECT REVIEW FROM THE BANK'S PERSPECTIVE Project Identity ........ ... .. ... .. ... . .. 1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1 Project Origin, Design and Organization . . . . . . . . . . . . . . . 2 Project Objectives and Description . . . . . . . . . . . . . . . . . 3 Project Implementation . . . . . . . . . . . . . . . . . . . . . . . 4 Project Results. . . 5 Project Sustainability .. 9 Bank Performance ......................... . 10 Borrower Performance . . . . . . . . . . . . . . . . . . . . . . . 10 Project Relationship .............. . 10 consulting Services ............... . 11 Conclusions and Lessons Learned ............ . 11 PART II: PROJECT REVIEW FROM THE BORROWER'S PERSPECTIVE Introduction ........................... . 13 Ntaruka Hydroelectric Generating Station. . . . . . . . . . . . . . . 13 Restoration of Reliability of the National Electricity Network . . . 15 Revaluation of Project Costs ............ . 16 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 PART III: STATISTICAL INFORMATION 1. Related IDA Credits .19 2. Project Timetable .19 3. Credit Disbursements . 20 4. Project Implementation Indicators . . . . . . . . . . . . . . . 20 5. Project Costs and Financing . . . . . . . . . . . . . . . . . . 21 6. Project Results Economic Impact .22 Financial Impact . . . . . . . . . . . . . . . . . . . . . 23 7. Status of Covenants . . . . . . . . . . . . . . . . . . . . . . 24 8. Use of Bank Resources .27 9. Mission Data .27 ANNEXES Annex 1 Evolution of Faults in the Transmission Network . . . . . . 28 Annex 2 ELECTROGAZ Financial Statements . . . . . . . . . . . . . . 30 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 3ank authorization.  PROJECT COMPLETION REPORT RWAIDA POWER PROJECT (LOAN 1495-CH) PREFACE This is the Project Completion Report (PCR) for the Rwanda Power Project, for which Credit 1495-RW in the amount of SDR 8.5 million (US$ 9.0 million) was approved on June 12, 1984. The Credit was closed on June 30, 1988, as scheduled. It was fully disbursed and the last disbursement was on October 31, 1987. This report was jointly prepared by the Industry and Energy Division of the South-Central and Indian Ocean Department, Africa Regional Office (Preface, Evaluation Summary, Parts I and III) and the Borrower's implementing agency for this project, Electrogaz (Part II). No comments were received from the project cofinancier. The preparation of Parts I and III of this report is based on the Staff Appraisal Report (SAR), the President's Report, the Credit and Project Agreements, supervision reports, correspondence between the Bank and the Borrower, consultants' reports and other information available in the project files.  ii PROJECT COMPLETION REPORT RWANDA POWER PROJECT (LOAN 1495-CM) EVALUATION SUMMARY Background At the time of preparation of the Rwanda Power Project (Credit No. 1495-RW) in 1983, only 2% of Rwanda's 6.5 million population had access to electricity. Electrogaz, the Government-owned utility for electricity, gas, and water, had a total installed electricity generating capacity of 27.4 MW, the source of which was mostly hydropower. In addition to this capacity, the Ruzizi II regional power project (Burundi, Rwanda and Zaire) was to make a considerable amount of power available to Rwanda in order that electricity service could be extended to unserved areas. Much of the equipment in Rwanda's own power stations and transmission system was old, maintenance was inadequats and plant failures, outages and shutdowns of the entire system were high (para. 3). Rwanda had no long-term power sector develcpmant plan and there was considerable scope for technical assistance to improve the utility's capability in planning, operating and maintaining the system. Furthermore, accounting and financial management of Electrogaz required strengthening. Project Obiectives and Description The Project's main objectives .. re to: (a) improve reliability of pi-duction, transmission and distribut,on of electricity; (b' serve additional electricity consumers ,c) help sustain the contribution to economic growth by the industrial and sommet.cial sectors of the country; and (d) strengthen Electrogaz in its planning, management, organization of its electricity department, general operations and maintenance. The physical components consisted of (a) rehabilitation at the Ntaruka power station (Rwanda's second largest), (b) the rehabilitation of the transmission network; (c) the construction of two new transmission lines; (d) the supply of maintenance equipment and spare parts; and (e) the provision of vehicles. The *-oject also provided experts' services to complement the institutional improvement program which has been initiated under the IDA and CCCE supported Water Supply Project (Cr. 1345-RW) (para. 10). This complementary assistance consisted of advisers to the General Manager, for electricity sector investment planning and technical improvements and for training of electricity service staff. Proiect Imolementation Experience The Project's implementation experienced major problems in the areas of project management (para. 14) and the availability of funds (para. 15) to fulfill contracts under the Project. There was a six-month delay in the completion of the project, which ended in June 1988 instead of December 1987 as estimated in the SAR. The major physical works of the Project -- the rehabilitation of the Ntaruka power plant, the rehabilitation of existing transmission lines and the construction of new transmission lines -- were completed, but there are still some outstanding technical problems related to the rehabilitation of Ntaruka (paras. 17-20), and there is a need for improvement in the quality of distribution services (para. 37). Concerning improvements to Electrogaz' operations: a planning unit was created, but de factor planning responsibility remained with the Ministry of Public Workds and Energy (para. 32); recommendations of the Engineering Adviser concerning network maintenance were only partially implemented (para. 33); and few of the financial performance targets were met owing to stagnant sales revenues, creation of non-productive assets and accumulation of Government receivables (para. 36).  iii Sustainability In order to fully attain the Project's benefits and maintain them, considerable remedial work is needed, including technical assistance for power system maintenance and the development of operating practices appropriate for the new facilities installed under the Project. Furthermore, Electrogaz needs to make significant improvements to the quality of distrJbution service in order for the benefits of the Project to reach the consumer. Electrogaz agrees that this is a priority and has requested that such works be included in the proposed Energy Sector project presently under preparation. Additional technical assistance will be required to sustain progress made in strengthening Electrogaz' commercial activities and accounting and to implement organizational changes, improved financial management and power sector planning all aimed at greater autonomy for the enterprise. Institutional reform will also be inne-ded in the focus of the aforementioned Energy project. Conclusions and Lessons Learned The Project was an important first step in power system rehabilitation but additional work is required to complete some tasks started during the Project, resolve problems with the Ntaruka power station rehabilitation and improve operation of the distribution system. The major lesson to learn from the Project is to design future projects to ensure the proper completion of work under the highest priority components in order to avoid situations such as the one discussed in paragraph 42, concerning the extension of the consulAing engineer's contract. A prudent measure would be to carry out rehabilitation works in phases, programming essential components in the first phase, and components which are economically justified but could be postponed in a second phase. Contracts would be written accordingly. A second lesson is the need to establish and strictly enforce procedures to improve the performance of consultants and contractors. These procedures should ensure that (i) records of past performance are adequately taken into account at the time of qualification for bidding, (ii) performance security fully reflects the risk inherent in obtaining full performance and (iii) undue delays in paying consultants and contractors are eliminated. A third lesson is that basic policy changes related to institutional reform should be implemented as upfront conditions. Finally, the project has shown the need for agreement with the Government and the implementing agency on a framework for investment planning in the power sector which would specify sector priorities. The lack of such a framework is partly responsible for concerns about the sustainability of the Project benefits. The proposed follow- up Energy project will incorporate the above lessons.  PROJECT COMPLETION REPORT RWANDA POWER PROJECT (LOAN 1495-CH) PART I: PROJECT REVIEW FROM THE BANK'S PERSPECTIVE A. Proiect Identity Name Power Project Credit Number 1495-RW RVP Unit Africa Region Country Rwanda Sector Energy Cofinancier Caisse Centrale de Coop6ration Economique B. Background 1. Rwanda, a country of 26,300 =m2 in the eastern part of central Africa, had a population of 5.5 million at the time of the Rwanda Power Project appraisal in 1983, and only two percent of this population had access to electricity. Electrogaz, the Government-owned electric power, gas and water utility, established in 1976, had a total electricity installed generating capacity of 27.4 MW, most of which (89%) came from iydropower, with the remainder supplied by diesel power. Prior to 1982, the main sources of power had been the Ruzizi hydroelectric generating facility in Zaire and the Ntaruka hydroelectric station in northern Rwanda. In 1982, with the completion of the Mukungwa generating station 12 x 6 MW), the country's dependence on electricity importo declined considerably. A 110-kV transmission line connected the Ruzizi, Ntaruka and Mukungwa power stations to Kigali, the capital of Rwanda. The power system was under the pressure of a fairly high annual growth in electricity demand, averaging 10.6 percent during 1977-82. In 1982, electricity sales were 73.4 GWh, of which 21 percent went to domestic users, 60 percent went to industry and 19 percent to Government users and public lighting. 2. Until the late 1970s, the Ministry of Planning had been in charge of electricity planning. In the early 1980s, the Rwanda Government had requested assistance from the Government of Switzerland in establishing an energy planning unit under the Director General of Energy within the Ministry of Public Works and Energy (MPWE). While the MPWE wished to retain the responsibility for electricity planning and investment programming, leaving network operating responsibility and project implementation work to Electrogaz, the Bank believed that Electrogaz should also have planning responsibility since it had the technical and financial responsibility for electricity supply. The Power Project therefore included a technical assistance component in the Project to help build this capability. 3. Rwanda's power stations had a considerable amount of old equipment, and maintenance was inadequate. Plant failures were high and during the ten months prior to project appraisal there were 113 unplanned outages and 27 complete shutdowns of the entire system. Most of the interruption occurred during the rainy season when severe lightning and frequent storms caused voltage fluctuations. Because equipment was old or not properly serviced to correct the faults automatically, it was necessary to reclose circuits manually, and this extended the length of time without power. Total losses on the system were about 20 percent of energy generated. Most of the transmission system was antiquated. For example, the Ntaruka  power station was designed as an isolated plant and additional equipment was required for it to function effectively as part of an integrated network. Concerning substations, their design and layout did not allow for convenient maiatenance and servicing. There was some circuit overloading, and a lack of adequate over-voltage protection and automatic switch reclosing equipment. These specific problems, combined with general system deterioration, had significantly diminished the reliability of electricity service to consumers. 4. Accounting and financial management of Electrogaz needed considerable improvement at the time of project appraisal. The accounts were unreliable and receivables were at the level of twelve months of revenue. There was poor inventory control, and unreliable accounting of long-term interest and debt due. A low debt equity ratio (27:73) resulted from the Government's treating grants which it received as a contribution to Electrogaz' capital and there was no complete long-term debt account. The satisfactory current ratio of 1:7 was mainly due to the high level of receivables. To correct this situation, the Project included a financial recovery program aimed at reducing receivables, improving inventory/ procuremert management, reducing operating costs and preparing separate financial statements for electricity, gas and water. 5. The Staff Appraisal Report (SAR) for the Rwanda Power Project (Credit 1495-RW, hereafter referred to as the "Project") noted that there was no power development program in the country but added that, through participation in the Ruzizi II hydroelectric project (27 MW in the first phase), Rwanda would have an adequate supply of electricity for several years. The Government's priorities for the power sector were: (a) mobilizing domestic power resources to meet growing electricity demand; (b) increasing the accessibility of rural areas to electricity; and (c) improving the reliability of the existing system. The Bank's objectives, in support of the Government's priorities in the sector, were: (a) creating a sound, financially viable electric utility; (b) developing a least-cost power expansion plan for the country; and (c) reducing the economic losses associated with frequent power outages and voltage fluctuations by improving system reliability. C. Prolect Origin. Desian and Oraanization 6. Originally, the Ruzizi II Regional Hydroelectric Project (Credits 1419-BU, 1420-RW, 1421-ZR), in addition to constructing a hydroelectric plant, was to cover transmission and distribution facilities for the three countries involved (Burundi, Rwanda and Zaire). But because of project preparation constraints, it was necessary to separace the transmission and distribution components from Ruzizi II and implement them as national power projects of which one was the Rwanda Power Project. Later, during the design phase of the Rwanda Power Ptoject, there was a further modification. The Bank's issues paper on the Project noted "... that because of time constraints in preparation and the urgent need to press on with the badly needed system rehabilitation, the Rwanda power project would be split into two projects: the proposed Power Network Rehabilitation Project, leaving the new transmission lines extending service to new consumers to the following years." The paper also recommended that a second project be included in the lending program if that were not possible, that the Government be advised immediately so that they could seek alternative financing sources. Finally, however, the Caisse Centrale de Coop6ration Economique (CCCE) made additional funds available to rehabilitate the medium voltage transmission network, freeing IDA funds for construction of two new MV Iines of high economic return. Also, in 1982, the Joint UNDP/Bank Energy Assessment Program had prepared  - 3 - an energy assessment report on Rwanda.1/ This report raised two issues which the Bank took into account in the design of the Project: the need to improve the efficiercy of the power system and to strengthen Electrogaz as an institution. 7. A Project Preparation Facility advar.ce (PPF), amounting to US$440,000 was extendr4 to the Government in 1982, covering engineering studies for the rehabilitation of the Ntaruka plant, the transmission system requirements and the revaluation of Flectrogaz' assets. Work under the PPF advance provided useful information for project preparation. Engineering consultants inspected the Ntaruka power plant and concluded that the tunnel h..d the penstock were in good shape, needing (nly minor repair. However, the turbines required considerable repairs. Concerning the transmission system, the consultant noted that work on the transmission system would require changes to the facilities of the Jorpina and Gikondo sub-stations which supply power to Kigali. New transformers with corresponding disconnects were required to allow for convenient and uninterrupted service. There was considerable discussion about the need for telemetering and remote control equipment in order to facilitate power dispatching. The revaluation of assets under the PPF advance lead to a 22 percent increase in the book value of Electrogaz' assets at the end of 1983. 8. During project preparation, the question arose whether there shoul.d be an increase in electricity rates, but finally there was a decision to postpone any increase until the reorganization of Electrogaz, supported by the Water Supply Project financed by IDA and the CCCE2/, had achieved anticipated reductions in operating costs. In the meantime, however, there was a need to include financial covenants covering minimum requirements for cash generation, and the hiring of technical specialists and Rwandese counterparts along with adopting procedures relating to Electrogaz' financing and administration. D. Project Objectives and Description 9. The Project's main objectives were to: (a) improve reliability of the production, transmission and distribution of electricity; (b) serve additional electricity consumers (c) help sustain the contribution to economnic growth by the industrial and commercial sectors of the country; and (d) strengthen Electrogaz in its planning, management, organization of its electricity department, general operations and maintenance. 10. The physical components to meet these objectives consisted of (a) the rehabilitation of the Ntaruka power station (Rwanda's second largest), (b) the rehabilitation of the transmission network; (c) the construction of twe 4w transmissioc lines; (d) the supply of maintenance equipment and s parts; and (e) he provision of vehicles. The Project also provided - ts, services to complement the aforementioned institutional :ovement program (para. 8). Three types of complementary advisory _._vices were to be provided by the Power Project. 1/ Rwanda: Issues and Options in the Energy Sector (Report No. 3779-RW). 2/ A component of the Water Supply Project (Cr. 1345-RW) provided a management assistance team from the Societe d'Amenagement Urbain et Rural (SAUR), financed by the CCCE. This assistance focussed on accounting, budgeting procedures, billing, receivables, procurement, warehouses/inventory, personnel management, revision of rates for water and electricity and the establishment of a computerized accounting system.  (a) Adviser to the General Manager. The manager had requested an adviser to assist in the organization and supervision of the company. The CCCE was to fund the cost of such an adviser for not less than three years; lb) Engineering Adviser. The main responsibility of this adviser was to assist Electrogaz in preparing a five-year power soctor investment plan. In addition, the adviser was to set up an engineering unit, assist in project design and implementation, provide advice on operation and maintenance of the system and prepare an action plan to investigate the high losses in the system. The adviser was also to be funded by the CCCE. (c) Training Advisers. The Projict was to provide for two specialists to ti.-n staff in system protection, communication an intenance. 11. The estimated total cost of the Project was US$19.3 million, excluding interest during construction. The Bank' s Credit of US$ 9 million (SDR 8.5 million) was to account fo- 47% and was to finance (a) repayment of the PPF (US$ 1,000,000)3/ .or preliminary engineering studies, detailed design, preparation of contract documents and project supervisors and (b) 100% of the foreign component for the rehabilitation of the Ntaruka power station, new transmission lines, maintenance equipment and spare parts; vehicles and consulting engineering. The Goverrunent of France, through the CCCE, was to provide the equivalent of US$ 9.2 million to finance the services of the two advisers mentioned above and the rehabilitation of the medium voltage transmission system. Electrogaz and the Government were to provide the remaining 1.1 million. E. Project Implementation 12. Overview. During most of its implementation period, the Project experienced major problems in the areas of project management (para. 14) and the availability of funds (para. 15) to fulfill contracts under tha project. '.ere was a six-month delay in the completion of the project which was June 1988 instead of December 1987 as estimated in the SAR. The major physical works of the Project -- the rehabilitation of the Ntaruka power plant, the rehabilitation of existing transmission lines and the construction of new transmission lines -- are completed but there are some outstanding technical problems surrounding the rehabilitation of Ntaruka, and a need for improvements in distribution services (paras. 17-18, 37). Concerning technical assistance under the project, the late arrival of the Engineering Adviser to Electrogaz caused a delay in the preparation of the five-year investment plan (para. 32). 13. In addition to reporting on specific Project components, supervision .ports on the Project also reviewed progress made in technical assistance -.r institutional improvements to Electrogaz under Water Supply Credit No. 1345-RW, since the technical assistance under the Project was complementary to some of this work. The institutional studies under the Water Supply Credit were completed by mid-1986, but the Government refrained from restructuring Electrogaz pending anticipated measures for cverall public enterprise reform. T1. tariff study financed by 3/ During negotiations, Electrogaz requested an increase in the PPF from US$ 440,000 to US$1,000,000 to cover engineering costs until the effectiveness of the credit, i.e. the preparation of detailed design and tender documents, as well as the call for and analysis of tenders.  Electricit6 de France (EDF) was completed in 1987, but the Government did not implement the reco.mmerded rate adjustments (rhanges -.o the peak energy charges and a 26% increase to the MV category). 14. Manaaement Problems. There were three major problems in the management of the Project. First, there was a considerable delay in the arrival of the Engineering Adviser (late 1985) and as a result, there was a delay in preparing the five-year investment and loss reduction plans. Second, the completion of the rehabilitation work at Ntaruka was not satisfactory. Before informing the Bank, the company accepted the rehaoilitation work as complete when there were serious reservations about the performance of the rehabilitated units (para. 19). Furthermore, they did not heed the Bank's recommendations to extend the consulting engineer's contract or to seek a ..'eplacement for him after the completion of the Project in order to resolve several technical issues at the Ntaruke power plant (para. 30). Third, there were major delays in the payment of the consulting engineer. 15. Funding Problem. There was a shortfall Ji Project financing after full disbursement of the IDA Credit in October 1987. The shortfall amounted to US$ 2.1 million on contracts the Bank had approved and it resulted mainly from currency fluctuations, except in the case of consulting services, for which costs increased due to extension of the consulting engineer's contract to cover (i) additional transmission system rehabilitation work, (ii) a fire at the Ntaruka power station (covered by the contractor's insurance) and (iii) completion of the Project six-months later than planned in the SAR. German funding was utilized to pay outstanding balances on contracts for vehicles, spares and maintenance equipment; it is unclear how overruns in DM for consulting services were financed. Overruns on workshop and warehouse construction and erection of transmission lines were financed from the Rwandese Government Development Budget. The Government apparently had requested funds from the Swiss Government to finance the shortfall on the Ntaruka rehabilitation contract but it is not clear that such an agreement was concluded. CCCE financed network rehabilitation costs were considerably higher in dollar terms than the appraisal estimate (final cost of US$16.4 million compared to an estimated US$7.6 million) mainly due to revaluation of the French franc against US dollar during the project execution period. 16. Performance Indicators. The SAR established performance targets for the level of losses in the power system, the number of connections per employee, the rate of return on revalued assets for electricity operations, and receivables, in terms of months of sales. The losses in the system decreased from about 22% in 1983 to 17% in 1988 compared to an ultimate target of 15%. tne number of connections per employee in 1988 was 30, which was double the target in the SAR, as electricity connections increased and the number of Electrogaz personnel declined. Concerning the rate of return on fixed assets, Electrogaz did not prepare separate accounts for its electricity, gas and water activities making a calculation of a return on electricity operations alone highly subjective and of little value. The situation of receivables, in terms of months of sales, fluctuated during the project implementation period owing to periodic compensations between debt due to the State by Electrogaz and Government receivables. There was, however, little overall improvemant at the end of the implementation period compared with the receivable situation at Project start-up (9.7 and 10.5 months respectively). F. Proiect Results 17. Ntaruka Rehabilitation. The Ntaruka hydro-electric power station has three turbine-generator groups of 3.75 MW each. Prior to the Project, the original manufacturer had replaced the turbine rotor of Group I. The Project included the rehabilitation of the other two 3.75 MW groups and completion of the earlier rehabilitation work on Group I. In accordance  with the SAR, the bidding documents specified that the original manufacturer was to supply replacement rotors for all units. The lowest bidder did not comply with that specification so the consulting engineer telexed the Bank recommending rejection of the bid and requesting the Bank's comments. The Bank replied that it would comment after receiving the Borrower's recommendation. In subsequent correspondence, Electrogaz stated that compliance with that specification was not essential and that the contractor had guaranteed to restore the groups to their performance as new. Therefore, the consulting engineer telexed that, in view of this guarantee, he withdrew his objection. 18. The Bank accepted Electrogaz' recommendation to award the contract for the rehabilitation works to the lowest bidder, who used turbine rotors not made by the original manufacturer. After rehabilitation of the Ntaruka station, cavitation limited its output to 3.6 MW for Group I, and to 3.2 MW for the other two groups, about 11 percent less than the station's nominal rating. The greater output of Group I seemed to indicate the advantage of using a rotor made by the original manufacturer. 19. Before the end of his contractual obligations, the consulting engineer had recommended imposing contract penalties on the firm responsible for the rehabilitation work because of the shortfall in output. However, in February 1989, Electrogaz informed a Bank mission that rather than collect damages, it was exploring, with the assistance of experts provided by German aid, how the contractor could increase the output of the turbines. It is likely that German aid will finance the corrective works. The February mission's aide memoire noted that satisfactory resolution of the Ntaruka issue was a matter of great urgency and that Electrogaz should inform the Bank of any developments in this area. Howeveir, in late 1939, Electrogaz accapted the rehabilitation of Ntaruka as complete, without informing the Bank and without imposing penalties, although the output remained at 10 MW compared to a nominal capacity of 11.25 MW. Electrogaz explained to the Bank that it did not have a legal case for collecting damages for two reasons. First, Electrogaz had lost the records of the actual output performance for Groups II and III, when the units in these groups were new; these records were essential for making a claim. Second, in the case of Group I, the power shortfall was less than the five percent tolerance allowed in the contract hefore the imposition of penalties. 20. Another technical problem arose during 1989, when the contractor damaged a connection on the alternator of Group I. The contractor made a repair which did not hold and Electrogaz took the group out of service in October 1989 but was unable to repair it. The contractor subsequently denied responsibility and the group remained out of service during much of 1990. 21. Maintenance eauinment. snares and vehicles. Although Electrogaz has stated that the delivery of these items has been satisfactory, the distribution system, including some medium-voltage facilities, 4/ remains in poor shape due to lack of spares and overloading. Apparently there has been some replacement of broken protective devices by untested makeshift arraxvgements. Furthermore, according to Electrogaz, some of the spare parts were used for system extension rather than rehabilitation. 22. New Transmission Lines. The Project completed two 30-kV transmission lines, one between Ntaruka and Cyanika (15 km) and the other between Gifurwe and Gakenke (24 km). Line construction work was generally satisfactory. There were delays in constructing associated distribution 4/ These facilities were not covered by the rehabilitation works included in the Project.  facilities but some progress took place in 1989, with German aid financing 230 connections for the two lines. 23. Transmission System Rehabilitation. Virtually all now transmission equipment provided under the Project is in place, but interface problems arose between equipment installed by the two contractors involved in the rehabilitation works. Furthermore, the settings of protective devices were not correct at the end of the project implementation period, and the protection system could not teil real trouble from false alarms, resulting in unnecessary disconnections of parts of the network. 24. Load DisDatctiina. At the Gikondo dispatch center, the meter readings corresponding to some of the substations were wrong because there was a change in the settings of current transformers during the installation of protective devices. Gikondo'e staff apparently did not know the new settings and therefore cannot interpret the readings. Additional dispatching problems also remain related to substandard or inappropriate equipment. 25. Power Outages. A major factor in the economic justification of the transmission system rehabilitation was the elimination of outages which were extremely high at the time of project appraisal (para. 3). Annex 1 provides a profile of electricity service interruptions during 1983-89 in terms of fault frequency, total fault duration, and average fault duration. These interruptions are additional to scheduled interruptions required to install and service equipment. The data provided in Annex 1 separate outage information for the high-voltage network (70 and 110 kV) and the medium-voltage network (15 and 30 kV). For the medium-voltage system, there is also information on outage durations per km, which are notable since the length of this network nearly doubled during the Project period, from 507 km to 957 km. 26. Figures 1 and 2 of Annex 1 show the number and duration of system outages in the high-voltage (HV) system. There was an appreciable decline in the number of outages from nearly 400 in 1983 to about 300 in 1989. However, the duration of outages rose substantially, from 85 hours in 1983 to about 175 hours in 1988. Correspondingly, the average outage time, which had fluctuated in the range of 5 to 13 minutes during 1983-1987, suddenly rose sharply in 1988 to about 36 minutes and then tapered off slightly to around 32 minutes in 1989. 27. Figures 3 and 4 of Annex 1 diagram the outage profile of the medium- voltage (MV) system. In this system, the number of outages increased from about 325 in 1983 to a high of nearly 675 in 1988, followed by a decrease to about 400 in 1989. There was a substantial corresponding increase in outage duration from about 58 hours to nearly 260 hours during 1983-88, declining to about 160 hours in 1989. Average outage duration increased from about 10 minutes to 24 minutes in 1983-88 and this high level remained unchanged in 1989. At the same time, the duration of outages per km increased from around 6 minutes to about 16 minutes during 1989-88, but declined to the 1983 level in 1989. 28. Overall, the outage profile shows a surge in outages due apparently to problems in adjusting to new transmission system equipment and extensions of the medium-voltage system. The trend since 1988 reflects increasing reliability of equipment but longer time required to locate and correct outages due to the lack of system discrimination and more complex equipment. Outages still remain high, however and Electrogaz maintains that the contractors have not responded to its requests for correcting the system deficiencies relating to the outages. In the absence of the supervising consulting engineer, it is difficult to determine who is responsible for coordinating the work. During 1989, Electrogaz started some remedial work, using its own staff, Led by a technician financed by EDF. There are indications that the situation is beginning to improve, since the daration of outages has started to decline. However, it is too  early to determine whether the rehabilitation of the transmission system will achieve the results expected at appraisal. 29. Mururu Substation Issue. An issue also arose at the end of the implementation period over the configuration of Rwanda's Mururu I substation, through which Electrogaz receives power from Zaire's Ruzizi power plant. After Project rehabilitation work on Mururu I, some of the substation's equipment was dismantled and transferred to a new substation, Mururu II, in order that Rwanda could also receive power from the Ruzizi II plant. The Soci6t6 Internationale d'Electricit6 des Pays des Grands Lacs (SINELAC) the regional power cou..pany created to construct and operate the Ruzizi II plant, operates Mururu II and, as a result of a design change in the Ruzizi II Project5/, now controls part of the distribution network of Electrogaz. This arrangement is unacceptable to Electrogaz, which would like assistance to modify tl.e Mururu I- Ruzizi II connection to regain control of its network. The Bank has indicated that it would have no objection to such a change. 30. Consulting Engineer for Plant Rehabilitation. In June 1988, Electrogaz decided not to extend the contract for the consulting engineer supervising the rehabilitation of the Ntaruka power station, despite the fact that the commissioning of major equipment was still in progress and some serious technical issues had arisen about the station's operation after the rehabilitation work. There was no follow-up on repeated Bank recommendations to extend the contract, or to appoint another consultant. 31. Management Improvements. Technical assistance under the Project was to include an Adviser to the Director and an Engineering Adviser for electricity operations, both to be financed by CCCE (para. 10) Their employment by Electrogaz was a condition in the Project Agreement. In 1986, a Bank supervision report stated that SAUR, the consulting team for management and water studies, was providing a part-time adviser to the Director (on a bi-monthly basis) but it does not seem that such part-time arrangement could have been sufficient considering the management needs of the company. Electrogaz informed the Bank that they had not appointed an adviser on their own because they believed that confusion would have arisen between his advisory and executive functions. However, there were precedents for such an arrangement and a similar arrangement for an Electrogaz team working on the transmission system seems to have been satisfactory. 32. Engineering Adviser. There was a delay in the appointment of the Engineering Adviser, who did not arrive until October 1985. He was supposed to have helped Electrogaz prepare, by June 1985, a five-year investment program and an action plan to reduce electricity losses. Although Electrogaz had put together a provisional investment program for projecting electricity rates, it was later decided that Rwanda should engage the services of a firm specialized in power planning to prepare a long-term, least-cost, Power Master Plan. A planning unit was created within Electrogaz, but de facto planning responsibility has remained with MPWE. Electrogaz subsequently produced a plan for reducing losses to the 10% level and took constructive measures which resulted in some reduction of losses (para. 16). 33. The Engineering Adviser's most important contribution has been a comprehensive proposal for a unit to oversee the maintenance of the transmission network, covering organization, staffing and equipment, and a list of spare parts needed to complete the rehabilitation of transmission facilities. Unfortunately, Electrogaz has implemented recommendations only partially. Furthermore, because the Adviser was not 5I IDA Credits 1419-BU, 1420-RW and 1421-ZR supported the Project to construct the Ruzizi II regional hydroelectric plant which began operation in July 1989.  fully integrated into the management team, he lacked the necessary overview and authority to make a substantial impact and his work was confined to discr-te tasks. However, Electrogaz made much more effective use of the Adv' r's assistant, a technician also financed by the CCCE. The assistant nias been supervising an Electrogaz team in charge of correcting the transmission system deficiencies left by the contractors. 34. Trainina. Training advisers were not appointed during the project execution period. However, training in system protection and maintenance is foreseen under a German Technical Cooperation program. Financing for communication training is yet to be identified. 35. Economic Impact. The SAR calculated economic rates of return at 12% for the rehabilitation of the Ntaruka plant, 17% for the rehabilitation of the existing transmission system and 10-12% for the two new transmission lines constructed under the Project. For rehabilitation of the Ntaruka plant, the ERR at the end of the Project has been estimated at 11.3% (base case), slightly below that in the SAR but higher than the opportunity cost of cap4.tal in Rwanda. Part III section six provides the details on this ERR calculation. No rate of return has been calculated for now transmission lines, as data on the number of new connections planned or energy to be delivered to new clients was not adequate. Owing to the absence of complete information on the CCCE financed rehabilitation of the MV network, no rate of return was calculated for this component either. 36. Financial Performance of Electroaaz. Concerning the financial performance of Electrogaz, the Project monitored the company's contribution to investment, the financial rate of return on revalued assets, the level of receivables in terms of months's due, and settlement of Government arrears in payments for its electricity and water consumption. The percentage contribution to investment was satisfactory only in 1984. For 1985-86, the percentage dropped below the targetted 25%. In 1987 and 1988, owing mainly to large increases in receivables and stagnant sales revenues, the contribution to construction from internally generated funds became negative; Electrogaz continued to receive investment subsidies from the Government in spite of agreement to end this practice. The company's rate of return has been around 2%, which is not far from the projected rate but is still very low, due mostly to excessive non-productive assets, mainly staff housing. Furthermore, Electrogaz has not been re-valuing its assetE on an annual basis, as recommended by the asset evaluation study financed by a PPF advance for the Project. Audited financial statements for Electrogaz (covering the 1984-88 implementatiorn period) are attached as Annex 2. G. Proiect Sustainability 37. In order to fully attain the Project's benefits and maintain them, considerable remedial work is needed, including technical assistance for power system maintenance and the development of operating practices appropriate for the new facilities installed under the Project. Furthermore, Electrogaz needs to make significant improvements to the quality of distribution services in order for the benefits of the Project to reach consumers. Electrogaz agrees that this is a priority and has requested that such works be included in the proposed Energy Sector project under preparation. Additional technical assistance will be required to sustain progress made in strengthening Electrogaz' commercial activities and accounting and to implement organizational changes, improved financial management and power sector planning all aimed at greater autonomy for the enterprise. Institutional reform will also be included in the focus of the aforementioned Energy project.  - 10 - H. Bank Performance 38. The Bank found itself in a difficult position over the Ntaruka contract issue (paras. 17-19). The consulting engineer's telex on the rejection of the lowest bidder may wll have been a request for Bank support in resisting pressures from the Borrower to take advantage of a low bid. If this were the case, the consultant should have persisted in his objection. One could argue that the Bank could have delved more deeply into the situation. However, the Bank could not have responded before receiving the Borrower's recommendation. To do otherwise, would have been tantamount to participating in the Borrower's bidder selection. Concerning the contract of the consulting engineer for the Ntaruka rehabilitation work, it was appropriate for the Bank to insist on its extension after June 1988. But unfortunately, the Bank's leverage in the matter had largely disappeared with the full disbursement of the Credit in October 1987. The Bank should have given greater attention in the design of the Project to the distribution works needed to realize the benefits of the new MV transmission lines; a component for new connections should have been incorporated in the project. Finally, the Bank could have invoked the appropriate remedies when financial covenants were not met, but did not do so. This was mainly due to the rapid disbursement of the IDA credit which lessened the leverage which could have been applied, and to the realization that excessive Government intervention in Electrogaz' operations often prevented the latter from taking the actions necessary to meet its obligations. I. Borrower Performance 39. Electrogaz was the Project's implementing agency for the Borrower (the Government of Rwanda), responsible for the execution of the Project which the credit supported. Electrogaz, along with the Bank, found itself in a difficult position in the bidding process for the Ntaruka rehabilitation. It is understandable that Electrogaz wanted to take advantage of the lowest bid, but subsequent events showed the merit of the original bid document requirement which specified that new rotors should come from the original manufacturer (para. 17). The reduced output of the plant has resulted, for Groups II and III, from not adhering to these requirements. since Electrogaz did not have sufficient qualified management staff, the company's rationale for not appointing an Adviser to the General Manager (para. 31) does not seem justified. A clear definition of the Adviser's responsibilities and authority could have eliminated the perceived confusion which Electrogaz seems to have feared. Furthermore, Electrogaz' record keeping for the Project has not been fully satisfactory, as evidenced by the loss of records, which were critical to the Ntaruka plant's contractual issues (para. 19). Maintaining good records should receive attention from Electrogaz in future management improvements. The Government could have made a greater contribution to improving Electrogaz' financial performance by promptly settling its electricity and water bills, and by setting power rates in line with sound economic and financial pricing principles (paras. 36 and 13). J. Proiect Relationships 40. There does not seem to have been sufficient understanding between the Government and the Bank on priorities for network investment. At appraisal, the Government expressed its interest in the construction of medium-voltage lines totalling 343 km, but there was sufficient economic justification for only two lines, totalling 39 km, and the Project included these. However, the MV network increased by 450 km during the period of project execution. The extensions of the MV system outside the Project, when extension of the distribution network to make adequate use of the two new lines constructed by one Project and general distribution network rehabilitation should have received greater priority, exemplifies the need for rational investment planning and increased coordination among  - 11 - donors. Furthermore, the technical problems associated with this project are not unique. Similar problems of varying severity have affected Rwanda's Mukungwa and Gihira power plants, which would suggest that there is a pattern of unsatisfactory performance by consultants and/or contractors, which appears to have been accepted as unavoidable by the Government and Electrogaz. 41. Several joint supervision missions were undertaken with Bank water supply sector and CCCE staff with emphasis on general institutional strengthening and improved financial management for Electrogaz. Through good cooperation, a unified position on major issues was presented to Electrogaz and the Government. K. Consulting Services 42. The same consultant who prepared the Project also supervised it. There were no unexpected major works excepc that towards the end of the Project's execution, it became apparent that Ntaruka's cooling system needed rehabilitating. The need for additional work is not unusual in a rehabilitation project because it often is possible to identify the full extent of wear only after dismantling equipment for overhaul. The consulting engineer's initial recommendation on the award of the Ntaruka contract turned out to be justified and he should have resisted pressures to change it (para. 18). Although Electrogaz verbally has expressed its disappointment with the consulting engineer's performance, the company has not been willing to document its apparent dissatisfaction. The main complaint concerns the consultant's apparent lack of specialized expertise needed to handle some of the technical problems that arose following commissioning oi the plant. It is true that the consulting engineer did not resolve an issue on the design of turbirA shear-pins, on which he had protracted discussions with the contractor. However, an expert, provided by German aid, ultimately helped to resolve the issue. Furthermore, the consulting engineer did not react sufficiently to problems in the transmission system but, in terms of system performance, the average duration of outages did not exceed 1983 values until 1988, after his contract had ended. Finally, there were payment arrears due the consultant, amounting to $550,000, and the consultant perhaps would have been more motivated to provide additional specialists if payment had been timely, thus enhancing overall performance under the consulting contract. Electrogaz should appoint a new Engineering Adviser, and in the meantime, should enhance the effectiveness of the expatriate advisers already working for the company, by integrating them more closely into the Electrogaz management team. L. Conclusions and Lessons Learned 43. The Project was among the early rehabilitation projects in the Bank and the first Bank Group supported power project in Rwanda. Some difficulties in Project execution and shortcomings in the rehabilitation work were, therefore, understandable. Although the physical works are complete, the sustainability of the Project's benefits hinges on (a) resolution of the outstanding technical issues in the operation of the rehabilitated Ntaruxa units (paras. 17-20) and (b) remedial work, including technical assistance for improving maintenance and operational practices (paras. 19 and 36). There is considerable scope for a follow-up project focussing on improvements to the distribution system and restructuring of Electrogaz as an autonomous public utility. The Rwandese Government and Electrogaz have expressed an interest in such a project which is presently under preparation by the Bank Group. 44. The major lesson to learn from this Project is that future projects in Rwanda should be designed to ensure the proper completion of work under highest priority components to avoid the situation, discussed in para. 42,  - 12 - concerning the extension of the consulting engineer's contract. A prudent measure would be to program rehabilitation works in phases, designing the first phase to include essential components and the second phase to include components which, though economically justified, could be postponed. Contracts would be written accordingly. 45. A second lesson is the need to establish and enforce procedures to obtain satisfactory performance from consultants and contractors. Three sets of actions appear necessary. First, procedures should ensure that records of past performance are adequately taken into account at the time of qualification for bidding. Second, performance security should fully reflect the risk inherent in obtaining full performance. The proposal to not use the original manufacturer's replacement rotors as originally specified in the bid documents should have been reflected in the contractor's performance guarantee. Third, undue delays in paying consultants and contractors should be avoided. 46. The third lesson is that basic policy changes related to institutional reform should be implemented as upfront conditions. Electrogaz would probably have managed the physical works of the Project better had its management and financial situation been strengthened early in the project implementation period. 47. Finally, the Project has shown the need for agreement, with the Government and the implementing agency, on an overall framework for investment planning in the power sector, which would specify sector priorities. The lack of a planning framework is partly responsible for concerns about the sustainability of the Project's benefits.  - 13 - PART II: PROJECT REVIEW FROM THE BORROWER'S PERSPECTIVE A. Introduction 48. Until 1978 the energy consumed in Rwanda was produced by the Ntaruka and Gisenyi power stations as well as several generating sets located in different areas of the country. 45. Around 1974, economic development and electricity supply problems had prompted the establishment of an interconnected grid that initially linked the Ntaruka power station with Ruzisi I, and later with Mukungwa, which was to be commissioned in December 1981. 50. Operating since 1959 and supplying a high-voltage network 150 km long, the Ntaruka plant was not rehabilitated until April 1984, when a Credit Agreement was signed with the International Development Association for US$9 million. A credit of FF 120 million was additionally negotiated with the Caisae Centrale de Coop6ration Economique (CCCE) to restore the reliability of the national electricity network. 51. As the rehabilitation of the Ntaruka plant and restoration of reliability of the national electricity network were to improve the services provided by Electrogaz, this component of special interest tc Electrogaz was included in the Rwanda Power Project. To achieve the project objective, a number of activities were awarded to eight firms selected in accordance with procurement procedures in effect in Rwanda. The works, eivided into eight lots, were briefly: - Rehabilitation of the Ntaruka generating station; - Restoration of the reliability of the national electricity network; - Improvement of maintenance infrastructure (logietical support); - Expansion of spare parts inventory and maintenance equipment; - Construction of a warehouse and three repair and maintenance shops; - Expansion of the medium-voltage (30 kV) distribution network through the construction of two 30-kV (Ntaruka-Cyanika and Gifurwe-Gakenke), including 30/0.4 kV distribution substations in the centers crossed by the lines; - Support for management and training of technicatl personnel at Electrogaz; - Study of the project and monitoring of implementation. B. Ntaruka Hydroelectric Generating Station 52. The rehabilitated Ntaruka plant is located between Bulera and Ruhondo Lakes. A supply tunnel and penstock supply the generating sets from Lake Bulera over a drop in elevation of about 100 m. 53. The power plant has three generating sets aligned horizontally, consisting of a (double) Francis turbine and a synchronous generator, the excitation of which is of the parallel static type fed by offtake from the generator.  - 14 - 54. Rehabilitation Works at the Plant. The works execution contract was concluded on June 1, 1985 for an 18-month period, and all rehabilitated machinery at the plant was to be in service by no lator than November 30, 1986. 55. The rehabilitation program had been initially set as follows: - Group II in rehabilitation while I and III were in service; - Group I in rehabilitation while II and III remained in service. 56. Unfortunately, following an incident involving Group III on April 26, 1986, the timetable for the works an a whole was readjusted so that by mutual agreement the end of the testing period (marc2e industrlelle) for all groups was postponed to April 30, 1988. 57. Thus, Group III preceded Group I, which should have e-on rehabilitated in the second phase. Following the incident involving Group III, responsibility was attributed to the contractor, which bore all additional costs related to supplies and works made necessary by that incident, namely: - replacement of the turbine wheel and shaft; - replacement of the stator; - replacement of bearings and bearing supports; - replacement of suction elbow; - repair of rotor shaft and rotor poles; - repair/replacement of oil rings and seals; - repair of the damaged civil works. 58. Most of the anomalies and defects observed during the testing period of the generating sets were resolved prior to their provisional transfer to Electrogaz on May 5, 1987 for Group II, June 7, 1987 for Group III and November 26, 1987 for Group I. The remaining items on the list were to be resolved during the warranty period. 59. An acceptance commission was set up by Rwandas Contract Award Council and visited Ntaruka on November 29, 1988 to observe the quality of the rehabilitation works carried out in accordance with the recommendations of the Council and the specifications. 60. During the inspection visit, a number of problems were noted and the contractor was requested to resolve them before the Chairman of the Award Council issued a provisional acceptance. 61. The problems noted by the members of the Commission included cavitation, since in accordance with the specifications of the machinery, generator output was to reach 3.75 MW without risk of damage. However, the rehabilitated machinery could not achieve that power. 62. The contractor noted, however, that the cavitation problem was tied to the hydraulic structure of the whole installation and could not be avoided %ithout changing that structure, a change that the consulting engineer rejected. 63. After the acceptance commission reviewed the available documentation, which showed that for Group I cavitation was experienced at 3.47 NM, and after observing that the operating conditions in 1959, when the plant was  - 15 - commissioned, had changed significantly, in particular as regards the change in elevation, which fell from 103 m to 100 m, and that, in addition, the measurements taken during the trials were recognized as being in conformity because they were identical in tolerance to within 2% of those for Group I in 1959. The commission felt that the contractor had met its obligations with regard to output. 64. In view of the foregoing, provisional acceptance was given on June 2, 1989 and final acceptance for all works on January 16, 1991 by the Chairman of the Award Council on the basis of the report given him by a committee set up for that purpose, which inspected the plant on December 24, 1990. 65. Regarding the subject of cavitation, the experts from the German Technical Cooperation who were contacted found that the defect was not conventional cavitation and that the problem could consequently be readily corrected by aligning the wheel and blowing air into the suction pipe and elbow. 66. As the same German cooperation officials had promised to contact the wheel manufacturer mnd provide financing for the neceesary works, performance should i ave. 67. Apart from that p,oblem of yield, the Ntaruka hydroelectric plant is operating to the satisfaction of the Borrower, which uses documentation furnished by the contractor to provide routine maintenance and upkeep. C. Restoration of Reliability of the National Electricity Network 68. Rehabilitation of the National Network. Rehabilitation of the national network was to eliminate the weak points in the transmission and distribution system to ensure reliability and enhance protection of equipment, facilities, operating personnel and users, on the one hand, and to meet the demand for electricity on the other. To achieve that objective, the contractor was to carry out the following tasks: manufacture of equipment and materials, factory trials, delivery, handling, transportation to the site, assembly of equipment and materials at the worksite, commissioning of facilities, operating trials and facility maintenance during the warranty period as well as training of Electrogaz personnel involved in running the rehabilitated network. 69. The statistics on disturbances for the year 1983 served as the baseline data justifying substation rehabilitation of stations and improvement of the protection system. 70. The frequency of network disturbances was due to unplanned outages attributable to the condition of equipment, relays, circuit breakers and their settings. 71. Following the rehabilitation works, network discrimination was as mediocre as before, but Electrogaz, with valuable assistance from EDF Technical Cooperation and eager to improve service to clientele, carried out certain works in the stead of the contract awardee, who had not bothered to remedy the problems observed, claiming that such works would be carried out during the warranty period. The works carried out by Electrogaz falling within the purview of the network rehabilitation contract would include adjustment of distance relay zones and ground fault coefficient settings, function changes and zone extensions in stage acceleration mode, construction of phase selectors, increase in time delays of all directional/land relays using a phasing system, adaptation of pole discrepancy relays to the single phase reclosing time-delays, adjustment of telecommand wiring, etc. Following the works, system discrimination is now achieved.  - 16 - 72. Apart from these finishing works that the contractor did not carry out, we feel that the network rehabilitation was done in technical conformity with the bidding documents. 73. As regards the overall teletranamission uyste:i; consisting of powerline-carrier linkages using the high-voltage aerial lines for the transmission of protection signals, telephone communications and telemetering, the rehabilitation works were carried out to the satisfaction of Electrogaz. 74. Nevertheless, the mimic diagram prepared by the consulting engineer would have been more useful had it used darkened conditions rather than an illuminated situation. To the extent possible, corrections nee: to be made. According to the estimate prepared by CGEE Alathom on July 7, 1989, that work would cost FF 323,000. 75. As regards remote transmission and information messaging, the system used requires human intervention whenever there is an interruption in network voltage or disturbance in supply. Since the substations and power plants are located throughout the national territory, rapid operation is not possible and hence automatic reset of remote terminal unite (RTUs) is required. To that end, the existing RTU volatile memory software must be switched from RAM to resident EEPROM memory so that the RTU does not have to be downloaded except when the memory board is replaced or the data base modified. 76. This change in software to ensure better system operation would cost FF 425,000, according to the 1989 Alsthom CGEE estimate. 77. The construction of two 30-kV power transmission lines with feeders was done to state-of-the-art standards and the full satisfaction of the Borrower. 78. It is nevertheless highly unfortunate that the project had not provided for any expansion in low-voltage or client connections, but thanks to German financing, the low-voltage network could be completed. Two hundred and thirty subscribers are now served by the Ntaruka-Cyanika and Gifurwe-Gakenke lines in question. 79. With a view to maximizing the return on the project, Electrogaz installed a line between the Gifurwe substation to Kirambo subprefecture, which made it possible to connect an additional 40 subscribers. D. Revaluation of Proiect Costs 80. T'- total project cost was RwF 2,143,955,214, provided by three sources, namely the Government of Rwanda (RwF 275,787,389), World Bank/IDA Credit 1495-RW (SDR 8,500,000 (US$9 million)) and the Caisse Centrale de Cooperation Economique (FF 120,000,000). 81. The World Bank credit financed the rehabilitation of the Ntaruka power station (SDR 2,460,000), construction of the new 30-kV Ntaruka- Cyanika and Gifurwe-Gakenke lines (SDR 940,000), equipment for upkeep and spare parts (SDR 2,080,000), purchase of vehicles (SDR 850,000), consultant fees for works supervision (SDR 290,000), and repayment of the PPF advance (SDR 940,000). 82. The CCCE loan financed the restoration of reliability of the national electricity network (FF 108,000,000) and support for Electrogaz management and miscellaneous studies (FF 12,000,000). 83. The agreement concluded between the World Bank and the Government of Rwanda specified that the latter would finance the portion in local currency and any cost overrun. The project ran into difficulties  - 17 - throughout most of its execution, .wost often because the necessary funds were not available to cover project contracts. Accordingly, US$2,100,000 had to be shouldered by the Government after total disbursement of the credit in October 1987, owing to fluctuations in the exchange rate. 84. For the rehabilitation of the Ntaruka plant, total commitments were SwF 7,028,077 and RwF 42,854,792. 85. For the works on the transfcrmation substations and lines, FF 53,405,636 and RwF 189,028,118 were committed. For teletr&nsmission, total commitments were FF 29,500,473 and RwF 34,000,882. 86. For vehicles, total expenditure in foreign exchange was DM 1,493,944, the balance of DM 62,560 to be paid from the KfW counterpart fund in RwF. 87. For spare parts and maintenance equipment, the total cost was DM 1,760,334.20. The balance of DM 199,094.60 was paid by Electrogaz from the KfW counterpart fund in RwF (RwF 9,344,227). 88. The total cost of workshop and warehouse construction was US$805,291.28 and RwF 40,879,951. The balance of RwF 11,893,737 was paid by the Government from the Development Budget. The construction of the 30-kV Ntaruka-Cyanika and Gifurwe-Gakenke lines cost BF 69,455,366 a-d RwF 90,129,716. The balance of RwF 39,814,438 was paid by the Government from the Development Budget. 89. The total cost of the project study and implementation monitoring was DM 4,097,943.70 and RwF 16,475,433.60. 90. The following table shows overall project costs (all eight lots), for the foreign exchange and local currency portions. COMPANY ESTIMATED AMOUNT AMOUNT PAID in foreign in RwF in foreign in RwF exchange exchange GRICH & V SF 8,132,362 4,000,741 SF 7,028,077.06 42,854,794 GTME FF 55,469,266 157,281,856 FF 53,405,635 189,028,118 SOFRA-TEP FF 30,348,352 33,502,188 FF 29,500,473.3 34,000,882.7 DAIMLER BENZ DM 1,553,019 - DM 1,493,944 (DM 62,5603* SAG DM 1,959,429 - DM 1,760,334.2 9,344,227 ASTALDI US991,053 28,986,214 US805,291.3 40,879,951 INTERN. [11,893,7371* ABAY BF 84,729,082 50,314,280 BF 69,455,366 90,129,716 [39,814,438]* FICHTNER DM 5,392,156 18,795,760 DM 4,097,943.7 64,155,558** * Corrected per text above. ** Paragraph 89 gives this amount as RwF 16,475,433.6.  - 18 - E. Conclusions 91. The Rwanda Power Project was completad with generally satisfactory results, despite a number of problems that arose after the commissioning of the Ntaruka plant and interconnected grid. These problems due to shortcomings of the studies and inexperience of the representatives of the consulting engineer who were providing work supervision were resolved as summarized below: 92. The cavitation of the machinery at the Ntaruka plant prior to achieving the expected power will be corrected by the manufacturer of the wheel through alignment. This should improve turbine perforrnance and will be financed by German Technical Cooperation to Electroqaz with a view to increasing the production of the Electrogaz units. 93. Thanks to the first-rate work of the Central Maintenance Office (Bureau Central de Maintenance -- BCM) established by the team providing support to Electrogaz technical management financed by the CCCE under the Rwanda Power Project, certain services were done on behalf of GTME, which did not see the urgency in regulating the protective relays that caused unplanned interruptions and undermined the reliability of electricity supply. To remedy the situation, Electrogaz carried out the following tasks through its safety engineer and the maintenance unit: - replacement of relays found to be unsuitable; - calculation of network stability; - regulation of relays in accordance with the above calculations; - improvement of grcunding connections following a measurement program; - replacement of grounding devices with more suitable ones; - replacement of normal control cables by noninduction cables for certain transformer substations; - assembly of frequency-based automatic load-shedding equipment; - completion and connection of control circuit cables for the Ruhengeri, Gihira and Kabarondo substations, which was not covered by the Rwanda Power Project; - installation of a 70-kV circuit breaker at the Rwinkwavu station to allow for remote control. 94. Given the sharp decrease in the frequency of outages in the high- voltage/medium voltage network and the fact that temporary defects are automatically eliminated (automatic mono- and tri-phase opening and closing), Electrogaz considers that the Power Project achieved its objective and expresses its deep gratitude to all those who contributed to the success of the project in general and in particular the members of the World Bank missions that visited Rwanda in connection with that project for their valuable advice that proved of great value to Electrogaz personnel.  - 19 - PART III: STATISTICAL INFORMATION 1. BRelated IDA Credits Credit/Number Year of Purpose of Project StatuB Project Title Approval Credit 1345-RW 1983 To improve Electrogaz' Completed in Water Supply Project capability in the areas April 1987 of planning & admini- stration, and finance as well as operation and maintenance of the water supply network; improve the level of service in five secondary centers; and strengthen the insti- tutional aid financial arrangements for the rural sub-sector. Credit 1420-RW 1983 Construction of Completed in Ruzizi II Regional 27-MW hydroelectric June 1989 Hydroelectric Project plant to supply intes- connected network of Burundi, Rwanda & Zaire 2. Proiect Timetable Item Planned Actual Date Date - Identification 10/82 - Preappraisal 07/83 - Appraisal 10/83 - Credit Negotiations 04/16/84 04/23/84 - Board Approval 06/14/84 06/12/84 - Credit Signature 08/10/84 - Credit Effectiveness 11/12/84 05/01/85a/ - Credit Closing 06/30/88 06/30/88 - Project Completion 12/31/87 06/88 6/ a/The limit date for credit effectiveness was postponed three times. 6/ Major worics were largely completed by June 1988, but additional rehabilitation of the transmission network, financed by the CCCE, and residual works continued into 1990.  - 20 - 3. Credit Disbursement Bank Fiscal Year and Disbursement (USs million) Quarter Credit 1495-RW Estimated Actual Actual as % of Est. Cumulative Cumulative Estimates 1985 4 0.40 0.41 102 1986 1 0.50 0.50 100 2 2.50 2.57 103 3 4.00 2.86 71 4 6.00 3.21 53 1987 1 6.50 6.01 92 2 7.00 6.44 92 3 7.50 7.47 99 4 8.00 8.74 109 1988 1 8.30 10.05 121 2 8.60 10.05 117 3 8.80 10.05 114 4 9.00a/ 10.05a/ 112 a/ The difference between estimated and actual amounts is due to exchange rate movements between the SDR, the currency of the original Credit, and the US$ equivalent used in the Bank's disbursement profile. 4. Project Implementation Indicators 1983* 1984 1985 1986 1987 1988 System losses.% Target 19 19 19 18 17 15 Actual 22.4 20.0 15.1 17.5 16.8 17.0 Number of connections Der elec-.riciv employee Target Actual 11 11 11 13 14 15 NA 18 21 24 25 30 ROR on revalued assets Target 7/ 2.7 1.3 0.6 2.4 2.0 1.6 Actual 8/ 5.0 3.0 2.2 1.1 1.6 2.0 Receivables, in months of sales Target 6.0 4.0 3.0 3.0 3.0 3.0 Actual 12.7 10.5 8.5 5.7 10.7 9.7 *Project appraisal, which took place in 1983, was based on 1982 data. Performance objectives were agreed for 1983, but results were not available before Board presentation. 7/ The SAR set this objective in terms of electricity fixed assets, however, Electrogaz did not prepare accounts for its different activities, making a calculation for electricity operations alone highly subjective and of little value. 8/ Rate of return shown covers electricity, water and gas operations for reasons given in the preceding note.  - 21 - 5. Proiect Costs and Financing A. Project Costs (in millions of dollars) - Appraisal ------ -------Actual-------- Foreign Local Total Foreign Local Total Ntaruka Rehabilitation 2.88 0.32 3.20 3.83 0.48 4.31 Network Rehabilitation 6.92 0.68 7.60 13.52 2.84 16.36 New Transmission Line 1.23 0.27 1.50 1.66 1.02 2.68 Vehicles 1.00 - 1.00 0.72 - 0.72 Spares & Equipment 1.35 0.15 1.50 0.65 - 0.65 Warehouses 1.10 - 1.10 1.05 0.46 1.51 TA/Engineering 2.88 0.52 3.40 1.87 0.73 2.60 17.36 1.94 19.30 23.30 5.53 28.83 Sources: Staff Appraisal Report No. 4691-RW, ELECTROGAZ Note: Total costs for individual project components were furnished to IDA in their respective contract currencies (Belgian francs, French francs, Rwandese Francs, Swiss francs, Deutsch marks, US dollars). Final project costs were then estimated by Bank staff based on hypothetical disbursement patterns and average annual exchange rates between the various currencies and the US dollar. The resulting costs are therefore indicative and should be interpreted with caution. B. Project Financing (in millions of US dollars) -- Appraisal ------ -------Actual-------- Foreign Local Total Foreign Local Total IDA 9.0 - 9.0 9.42 0.63 10.05 CCCE 8.2 1.0 9.2 13.52 2.84 16.36 ELECTROGAZ/ Govt. of Rwanda - 1.1 1.1 0.36 2.06 2.42 17.2 2.1 19.3 23.00 5.53 28.83 Source: ELECTROGAZ  - 22 - 6. Proiect Results A. Economic lmoact Re-Evaluation of the Rate of Return on the Rehabilitation of Ntaruka Appraisal Estimate Estimated Actual Economic Rate of Return 17.3% 11.3% Assumotions 1. The re-evaluation was made in Swiss francs, the currency of the rehabilitation contract. 2. Cost were deflated to 1985 values. Works not executed as of 1988 were treated as if they had been executed in 1989, the year in which Electrogaz accepted the rehabilitation as complete. 500,000 Swiss Francs were added in 1989 for the overhaul of the cooling circuit: which was deemed necessary in 1988, but were to be executed subsquently. The damage to the alternator of Group I was ignored except in the sensitivity analysis, for its value was taken as its replacement cost. 3. Benefits were taken as the avoided replacement of the units. The SAR estimate of $500,OO per kW was used: in 1985 prices, this was equivalent to 1200 Swiss Francs per kW, or 12 million Swiss Francs for 10 MW. 4. As the SAR stated that in the absence of rehabilitation, new groups would be needed in 1990, their cost was split between 1989 and 1990. The result.ng cost and benefit streams are sh-wn below, in thousands of Swiss Francs. Year 1985 1986 1987 1988 1989 1990 Costs 2202 3358 1543 0 1573 Benefits 6000 6000 The resulting rate of return was 11.3%.  - 23 - 5. Sensitivitv Analysis Rate of return Base case 11.3% Overhaul of cooling circuit postponed to 1991 11.6% Group I out of service for one year (out 1988, in 1989), valued at replacement cost 10.0% Benefits postponed by one year (i.e., without project, new groups needed in 1991 instead of 1990) 8.5% Group I out of service for one year, and benefits postponed by 1 year 7.7% B. Financial ImDact -----------------------~ ~~ ~ ~~~~ . . .................................................................... ......... . ........ 1983 1984 1985 1986 1987 1988 Eat Act Est Act Est Act Est Act Est Act Est Act ~~~~~~~~~~~~~~~~~~~~~~~. ............................... ........................................................................ ..... ....... ... ate of Return (X) 2.7 5.0 1.3 3.0 0.6 2.2 2.4 1.1 2.0 1.6 1.6 2.0 3erating Ratio 80.4 103 89.7 89 95.5 91 83.3 130 85.3 114 87.7 109 it. gen. furds C() 54.6 - 54 51 32.8 17 38 19 46.5 -47 47.9 -50 ebt Serv. Coverage 8.3 3.6 2.7 3.4 4.6 3.7 1.9 6.4 2.2 8.0 1.4 ebt/Equity 12/88 18/82 19/81 24/76 23/77 36/64 26/74 44/56 27/73 48/52 29/71 47/53 irrent Ratio 3.0 1.8 2.5 2.3 2.3 1.3 2.4 1.9 2.5 1.7 2.6 1.4 iceivables (mos) 7.0 12.7 4.0 10.5 3.0 8.5 3.0 5.7 3.0 10.7 3.0 9.7  - 24 - 7. Status of Covenants Brief Description Applicable Compli- of Covenants Section ance Remarks Four-month limit on total CA 3.02 No Both limits were far exceeded. receivables for gas, electricity and water; three-month limit on government receivables. Borrower to permit ELECTRO- CA 3.03 No Instead of a needed rate GAZ to adjust electricity increase, the Government decided tariffs based on revalued to decrease rates in January assets and available tariff 1988. and cost reduction studies Availability of foreign CA 3.04 No Foreign exchange was not made exchange available to settle unpaid contracts following exhaustion of the IDA Credit. Borrower ensures that all CA 3.05 Par- ELECTROGAZ still has little appointments to the staff tial effective role in the of ELECTROGAZ be approved recruitment of its personnel who by its Personnel Manager are civil servants. Further investment in CA 3.06 No ELECTROGAZ continued to receive ELECTROGAZ by the Borrower subsidies from the Government's shall be in the form of investment budget throughout the loans rather than project execution period. contributions to capital Employment of consultants PA 2.02 Par- However, the consulting for project supervision. tial engineer's contract was allowed to expire prematurely before the end of the project when commissioning was in progress and serious issues had arisen. Goods and services to be PA 2.04(b) Yes Although ELECTROGAZ reported used for project purposes that it had used some spares only. for system expansion instead of maintenance. Performance of obligations PA 2.06 No Due to accumulation of under the Subsidiary Loan Government receivables, Agreement ELECTROGAZ has not always been able to meet its debt service obligations. Exchange of views with IDA PA 2.07 Par- However, despite IDA on progress of the project tial recommendation to the contrary, as well as on conditions and without informing IDA, which interfere with its rehabilitation work on the purpose. Ntaruka units was accepted even though performance by the contractor was unsatisfactory.  -25 - (i) Employment of a person- PA 2.08 Par- A personnel director was nel manager and an adviser tial appointed. A part-time adviser to the general manager; rather than full-time adviser to preparation of a statement one director was appointed; a of personnel policies and draft statement was completed in procedures; 1985. (ii) employment of a planning director and an expert in electrical Par- Both were named with some delay. engineering; preparation of tial The decision was taken to forego a 5-year investment plan the 5-year study and prepare a and a plan of action for long-term Master Plan; work on reduction of electrical the Plan began in 1991. energy losses by June 30, Measures were taken to reduce 1985. both non-technical and technical losses, but a formal plan of action was not completed. Appointment of two training PA 2.09 Par- The training officers and one officers and employment of tial agent were employed, but no two training experts who formal training plan was would establish a training submitted to IDA. plan for electricity operations by June 30, 1985. ELECTROGAZ to establish PA 3.07 Yes Guidelines were issued in 1986. guidelines for procurement and for inventory management by March 31, 1985. Beginning in FY85, PA 4.01(b) No ELECTROGAZ to prepare separate accounts for its water, gas and electricity operations. Audit accounts and furnish PA 4.02 Yes Accounts were audited annually them to IDA by Price-Waterhouse, but often submitted with several months' delay. Debt service requirement of PA 4.03 Yes 1.5 ELECTROGAZ to generate PA 4.04 No 1984 was the only year in which funds from internal sources the target was achieved. of at least 25% of its annual capital expenditure program averaged over three years; tariffs to be reviewed June 30 in each year to ensure their adequacy in meeting aforementioned contribution. Beginning in 1986, PA 4.05 No ELECTROGAZ's tariffs remained ELECTROGAZ to fix average unchanged until January 1988 tariffs for electricity at when the Goverrnment decided to levels determined on the decrease rates by 14% in spite basis of revalued assets of analysis showing the need for and available tariff and increase. cost-reduction studies.  - 26 - Receivables not to exceed PA 4.06(a) No ELECTROGAZ was not able to meet four months' billings at this condition owing to a end of 1984 and three chronic problem of Government months' billings at the end receivables for works and of 1985. electricity and water consumption. For contracted works and PA 4.06(b) Par- ELECTROGAZ did implement a connections, ELECTROGAZ to tial special payment plan, but only require at least 60% on a limited basis. advance payment and balance upon completion, or for low income customers, 30% in advance and the remainder over a period not exceeding 12 months. Annual revaluation of PA 4.07 No assets commencing in 1984. Preparation by March 31, PA 4.08(a) Par- A formal plan was never 1985 of a detailed plan for tial submitted to the Bank although the reduction of operating acceptable actions to reduce costs and exchange of views personnel, vehicle and materials with IDA expenditures were put into effect in 1985; results were limited. Freeze on hiring non- PA 4.08(b) No Although ELECTROGAZ's overall professional staff until permanent staff declined, hiring cost reduction plan in of excessive temporary labor effect continued. Consultation with IDA prior PA 4.08(c) Yes No investments of this size were to undertaking any capital undertaken in the power sector investment above US$10 during the project million. implementation period.  - 27 - 8. Use of Bank Resources Staffweeks Through Appraisal 40.8 Appraisal through Board approval 8.4 Supervision 45.7 Total 99.1 9. Mission Data Stage of Month/ Number of Days SpecialIzation Performance Types of Project Cycle Year Persons In Field Represented !/ Rating Status k/ Problem c/ Throuch Appraisal Pro-appraisal 7.83 2 S F, E Appraisal 10.83 a 12 F, E, T(C)/ Appraisal throuch Board Post-appraisal I 12.83 2 1 F, E Post appraisal II / 3.84 1 a F Board through effectiveness SPN g/ 7.84 2 1 F, E F; G Supervision SPN 1 7-8.86 3 4 D, E, F 2 PU, D SPN 2 1.86 2 7 E, F 2 PM, D SPN 3 10.88 2 6 E, F a PF, PM, D SPN 4 10.87 1 6 F 3 E, PF, PU, D SPN 6 11-12.87 2 10 F, E(C) a F, E, PF, PM, 0 SPN 6 6-8.88 3 17 F, Ec, E(C) a F, E, PF, PM, D SPN/PCR f/ 2.89 2 7 F, E(C) 3 E, Pf, PM, D SPN/PCR update g/ 1.89 2 3 F, E(C) 3 E, PF, PM, D a/ 0 = division chief; E = engineer; Ec a economist; F = financial analyst; T training specialist; (C) = consultant 9 From forms 590: 2 = moderate problems; 8 = severe problems / D = development impact; E = engineering; F = financial; C = government interference with management; F a financial problems; PF = funds available for project execution; PM = project management; J The training specialist was only 7 days in the field p/ No report on file. Data from terms of reference. f Combined with identification of proposed energy project. No formal rating of problems, but no Improvement. g/ No form 690.  - 28 - Annex 1 Page 1 of 2 Evolution of Faults in the Transmission Network Rwanda - 11 0 kV and 70 kV - Number and duration of system faults 400. 350 300 250 Number Emb or hours ur 1 5 0 50- 0- 1983 19841985 19861987 19881989 year Fia. I Evolution of faults on the H.V. network. The length of the network was unchanged Rwanda - 11 0 kV and 70 kV - average fault duration 40 35- minutes 20 . Io-jM~ ~~ 5. l 0 1983198419851986198719881989 year Fic 2. Evolution of the averaae duratiorn of faults on the H.V. network  - 29 - Annex 1 Page 2 of 2 Rwanda - 30 kV and 15 kV - number and duration of system faults 700 600_ 500- Number 400- hours or hours 300 number 200 4 - 100 198319841985 19861987 19881989 year Fio. 3 Evolution of faults on the M.V. network During 1983-1989, network length increased from 507.5 to 957.5 km. Rwanda - 30 kV and 15 kV - average fault duration, and duration per km 25 20 1s minutes/fault minutes I3 I F minutes 10 M; Eminutes/km 0* 1983 1984 1985 1986 1987 1988 1989 year Fic 4 Evolution of the averaae duration of faults. and of fault duration ner km of line, on the M.V. network  Annex 2 - 30 - Page 1 of 4 ETABLISSEMENT PUBLIC DE PRODUCTION DE TRANSPORT ET DE DISTRIBUTION D'ELECTRICITE, D'EAU ET DE GAZ Balance Sheet as of December 31: Audited 1983, 1984, 1985, 1986, 1987, 1988 (RwF Millions) 1983 1984 1985 1986 1987 1988 -ASSETS get Fixed Assets 8 ,2 9,154 10,015 9,847 9,602 13,026 lork in Progress 999 1,273 2,497 4,813 6,657 3,188 Intangible 54 - 176 164 207 636 4edium and L/T claims - - 48 57 55 54 )ther 10 27 - 93 100 Investments in Company - - 10 95 95 157 3hares Iurrent Assets lash and Bank Deposit 121 158 262 402 326 392 kccounts Receivable 1,399 1,317 1,165 812 1,461 1,310 Enventory 435 505 654 853 823 904 ?ersonnel Advances 30 41 20 - 4 13 Mtate - - 3 6 68 465 )repaid Expenses - - 58 65 69 49 )ther 17 14 3 7 - 4 !otal Current Assets 2,002 2,035 1,511 2,146 2,751 3,137 ?otal Assets 11.157 12.472 14,938 17,122 19.460 20,298 .IABILITIES :quity Capital 430 430 430 430 430 430 Govt.Subsidy for Invest. - - 3,858 3,986 4,014 4,088  Annex 2 Page 2 of 4 1983 1984 1985 1986 1987 1988 Provision for renewal of 3rd Party financed - - 185 313 583 722 assets Retained Earnings 197 148 462 441 512 653 Asset Renewal Reserve 266 440 267 267 267 267 Revaluation Reserve 82 324 1,322 1,322 1,322 1,322 Other Reserves - - 81 81 81 81 Equipment Subsidy 7,2e9 7,427 2,235 2,118 2,027 1,931 Total Equity 8,264 8,769 8,841 8,958 9,236 9,495 Long-Term Debt 1,777 2,818 4,890 7,046 8,582 8,577 Current Liabilities Accounts Payable 320 282 477 354 423 464 State 176 393 486 288 759 1,293 Other S/T Debt 280 104 87 152 174 233 Advances for Works 255 - - 141 63 - Consumer Deposits 83 106 144 171 210 226 Misc. Provisions 1 - 13 12 12 li Total Current 1,115 885 1,207 1,118 1,641 2,226 Liabilities Total Liabilities 11,157 12,472 14.938 17,122 19.460 20.298 Current Ratio 1.8 2.3 1.3 1.9 1.7 1.4 Receivables (months) 12.7 10.5 8.5 5.7 10.7 9.7 Debt: Equity 18/82 24/76 36/64 44/56 48/52 47/53  Annex 2 32 Page 3 of 4 ETABLISSEMENT PUBTIC DE PRODUCTION DE TRANSPORT ET DE DISTRIBUTION D'ELECTRICITE, D'EAU ET DE GAZ Income Statement as of Decewber 31: Audited 1983, 1984, 1985, 1986, 1987, 1988 (RwF Millions) 1983 1984 1985 1986 1987 1988 REVENUES ales 1,173 1,225 1,325 1,404 1,459 1,409 =nnections 118 253 299 270 146 193 -ter Charges 13 15 - - - - Lher 23 22 202 163 246 186 Lher Non-operating Income 321 139 104 492 520 524 umcrtization of External - - 528 173 183 197 ;ubsidies Total Revenues 1.544 1,654 2,458 2.502 2,554 2,509 OPERATING EXPENSES Salaries 95 282 389 453 538 424 Materials and Services 776 509 646 998 922 712 consumed Transportation 3 21 60 26 1 - Depreciation 385 540 573 613 647 815 -ovision for debt and iventory write off - - 425 29 48 71 other - 40 152 271 194 226 Total Expenses 1,259 1,S2 2.245 2,390 2.350 2.248 operating Income 389 262 213 112 204 261 Interest Charges 104 114 96 132 133 120 Net Income 285 148 117 (20) (71) 141 erating Ratio 1.03 0.89 0.91 1.30 1.14 1.09 perating Ratio (excl. epreciation) 0.71 0.54 0.60 0.80 0.79 0.64 Rate of return (%) 5.0 3.0 2.2 1.1 1.6 2.0  Annex 2 33 P Tage4 of 4 ETABLISSEMENT PUBLIC DE PRODUCTION DE TRANSPORT ET DE DISTRIBUTION D'ELECTRICITE, D'EAU ET DE GAZ Sources and Applications of Funds: Audited 1984, 1985, 1986, 1987, 1988 (RwF Millions) --------------------------------------------------------------__-------------__---------- 1983 1984 1985 1986 1987 1988 SOURCES OF FUNDS Generated from Operations N.A. Operating Income 262 213 112 204 261 Depreciation 540 573 613 647 815 Amortization of Equipment Subsidies - - (139) (183) (197) Total 802 786 586 668 879 Investment Subsidy 138 - 150 79 122 Third-Party Financing of - 186 127 311 192 Assets Other - 1,313 3 7 14 Loans 1,185 2,037 2,213 1,628 330 Total Sources 2.125 4.322 3.079 2,623 1,537 APPLICATIONS OF FUNDS Construction/Renewal 1,580 2,834 2,724 2,290 1,201 Reserves 24 173 - - - Debt Service Repayment 144 63 59 92 334 Interest 114 96 132 133 120 Other - 1,334 155 97 80 Change in Working Capital 263 (178) 68 81 (198) Total Applications 2.125 4.322 3.079 2.693 1.537 Debt Service coverage 2.7 4.6 1.9 2.2 1.4 Net Cash Contr. to 51% 17% 22% -47% -50% Construct.