Colombia - Mesitas Hydroelectric Power Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 6638 PROJECT PERFORMANCE AUDIT REPORT COLOMBIA: LAS MESITAS HYDROELECTRIC PROJECT (LOAN 1628-CO) February 13, 1987 Operations Evaluation Department This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contenth may not otherwise be disclosed without World Bank authorization. THE WORLD BANK FOR OFCIAL USE ONLY Washington. OC 20433 USA 01ce of Dector-Cenval February 13, 1987 MEMCRANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Performance Audit Report - Colombia: Las Mesitas Hydroelectric Project (Loan 1628-CO) Attached, for information, is a copy of a report entitled "Project Performance Audit Report on Colombia: Las Mesitas Hydro- electric Project (Loan 1628-CO)", prepared by the Operations Evaluation Department. Attachment 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. COUNTRY EXCHANGE RATE Name of Currency (Abbreviation): Peso Colombiano (Col$) Exchange Rate in Col$/US$ Appraisal Years: Yearly Average End of Year 1977 36.320 /a 37.960 1978 39.095 41.000 1979 42.550 44.000 1980 47.280 50.920 1981 54.491 59.070 1982 64.102 -<.290 1983 78.857 88.770 1984 100.800 113.890 1985 143.045 172.200 1986 191.150 /b 210.100 /b /a Exchange rate used in the project cost estimate. /b Estimated. FOR ICIAL USE ONLY ABBREVIATIONS AND ACRONYMS CHEC - Central Hidroelectrica de Caldas (Caldas Hydro Power Plant) CHIDRAL - Central idrutlectrica del Rio Anchicaya (Anchicaya Hydro Power Plant) CONPES - Consejo Nacional de Planeacion Economica y Social (National Council for Economic and Social Planning) CORELCA - Corporacion Electrica ie la Costa Atlantica (Atlantic Coast Power Corporation) CVC - Corporacion del Valle del Cauca (Cauca Valley Corporation) DNP - Departamento Nacional de Planeacion (National Planning Department) EEEB - Npresa de Energia Electrica de Bogota (epresa) (Bogota Power Company) EAAB - EApress de Acueductos y Alcantarillado de Bogota (Bogota Enterprise for Water Supply and Sewerage) ELECTRIBOL - Electrificadora de Bolivar (Bolivar Power Company) EPM - Empresas Publicas de Medellin (Medellin Public Enterprises) FEN - Financiera Electrica Nacional (Financial Institution for the Power Sector) ICEL - Instituto Colombiano do Energia Electrica (Colombian Power Institute) INCOMEX - Instituto de Comercio Exterior (Institute of Foreign Trade) ISA - Interconexion Electrica S.A. (Electric Interonnection Corporation) JNT - Junta Nacional de Tarifas (National Tariff Board) LEBRIJA - Central Hidroelectrica del Rio Lebrija (Lebrija Hydro Power Plant) This document has a restrkted distribution and may be used by sciplents only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authoratlon. PROJECT PERFORMANCE AUDIT REPORT COLOMBIA: LAS MESITAS HYDROELECTRIC PROJECT (LOAN 1628-CO) TABLE OF CONTENTS Page No. Preface ...... ... .. .. . ... .. ... .. ... . .i Basic Data Sheet ................................................ iii Evaluation Summary ......... ........*....... .............. v PROJECT PERFORMANCE AUDIT MEMORANDUM I. BACKGROUND ................................................ 1 The Structure of the Sector ...................... 1 The Atmosphere during Appraisal ........................ 2 II. THE PRINCIPAL DIFFICULTIES ............................... 5 III. THE GRANADA TUNNEL PROBLEM ............................. 8 IV. TH7 '-LNK'S ROLE IN ADMINISTRATION ......................... 10 V. REALIZATION VS. EXPECTATION .......................... 11 VI. CONCLUSIONS .......................................... 12 Attachment 1 - Comments from the Borrowerts Design Engineers **** 15 Attachment 2 - Comments from the Borrower ....................... 33 PROJECT COMPLETION REPORT I, Introduction .... .............. ................ 35 II. Project Preparation and Appraisal ........................ 36 III. Project Implementation and Costs .......................... 40 IV. Project Justification ................................... 43 V. Financial Performance ................................. 45 VI. Institutional Performance ................................ 47 VII. Lessons to Be Learned ..................................... 49 Annexes 1. Summary of Bank Lending to the Sector ..................... 51 2. Major Covenants of the Loan and Guarantee Agreements ...... 53 3. Project Components and Forecast and Actual Completion Dates ...o................................ 55 4. Schedule of Disbursements ................................. 56 5. Project Cost Summary and Sources of Financing ............ 57 TABLE OF CONTENTS (continued) Page No. Annexes (cont'd.) 6* Electricity Supply and Consumption ........................ 58 7, Actual and Forecast Sales Growth Rates (%) an! Percent (%) of Sales by Customer Category, 1977-82 ...... 59 8. Comparison of Actual and Forseast Incomde Statements ....... 60 9. Cmparlson of Actual and Forecast Sources Applications of Funds ...................... ...... .. 61 10. Comparison of Actual and Forecast Balance Sheets ......... 62 11. Rate of Return Based on Actual Project Cost ............ 63 12. Comparison of Actual and Forecast Performance Indicators (1978-1984) 0.0.............................. 64 13. Map of the Project ......o........................ 65 PROJECT PERFORMANCE AUDIT REPORT COLOMBIA: LAS MESITAS HYDROELECTRIC PROJECT (LOAN 1628-CO) PREFACE This report presents the results of a performance audit of Las Mesitas Hydroelectric Project for which Loau 1628-CO of US$84 million was approved in November 1978. This Loan was, at the time, the twenty-first to Colombia for electric power development, and the fourth to the borrower which is executing the project, Empresa de Energia Electrica de Bogota (Empresa), with which the Bank has enjoyed a close and cooperative relationship for over 25 years. Siuce 1978, the Bank has made three more loans to Empresa: two for distribution, 1807-CO (US$87 million) and 2634-CO (US$171 million) of March 1980 and November 1985, respectively, and 2008-CO (US$359 million) of May 1981 for the very large Guavio Hydroelectric Project now under construction.1/ The Loan under review here was, as noted, approved November 1978, signed April 1979, and the original closing date of December 1982 extended to December 1984. It is completely disbursed. The project, however, as brought out in the Project Completion Report (PCR) of June 1986, is not expected to be completed and pla.:ed in operation until December 1986. The Project Performance Audit Report (PPAR) consists of a Project Performance Audit Memorandum (PPAM) prepared by OED, and the PCR issued by the Latin America and Caribbean Regional Office. The PCR was prepared by staff actively associated with the sector, the borrower, and the project. In preparing the PPAM, OED has reviewed the Appraisal and President's Reports, the legal documents, the relevant operational files, the transcript of the Board meeting at which the Loan was approved, and held discussions with Bank staff associated with the project's conception and implementation. Finally, an OED mission visited Colombia, in September 1986, for discussions with Lhe Government, Empresa, and its consultants and contractors. 1/ These last loans which aggregate US$617 million, were made over a 4-1/2- year period during which the problems of Las Aesitas were emerging, and represent an amount, which when stated in currency of constant value, is something like two times all prior lending to Empresa, including the Loan under review. Loan 2634-CO was not yet effective as this Report was being prepared. (One agency of the Government indicated in September 1986 that consideration was being given to requesting the Bank to renegotiate its terms. As with the Loan for Guavio, 2008-CO, amorti- zation payments are required before the agreed construction period elapses.) - ii - This audit finds that, in most major respects, the PCR faithfully analyzes the project and rmpresa, together with the atmosphere which pre- vailed In Colombia during project implementation, and that its conclusions and suggested lessons are valid and relevant. On the other hand, the finding that the project has been successfully ccmpleted is premature, because it cannot be placed in operation until certain remedial works now underway are successfully carried out. The PPAM avoids presenting details which are covered fully in the PCR, including a very substantial amount of financial data, engineering cost estimates, various chronologies, etc., but rather refers to them where appropriate. The PPAM, however, expresses views on several aspects of the project's execution, and, in particular, the Bank's role in administering Loan 1628-CO. Following standard procedures, OED sent copies of the draft PPAR to the government and borrower for comments. The comments received are repro- duced in full text as Attachments 1 and 2 to the report. - iii - PROJECT PERPORMANCE AUDIT BASIC DATA SHEET COLOMBIA: LAS MESITAS HYDROELECTRIC PROJECT (LOAN 1628-CO) KBY PROJECT DATA ------US$ million- --- Appraisal Actual/Current Item Expectation Estimate Total Project Cost, excluding finance charges (% overrun) 260.8 413.7 (58.6) Loan Amount 84.0 84.0 Disbursed 84.0 84.0 Cancelled 0 0.0001 Repaid to February 28, 1986 19.38 19.38 Outstanding, February 28, 1986 64.62 64.62 Rate of Return on Project (2) 10.4 10.0 Financial Performance Required to EEEB's rate of obtain a return has 1been return of 12% 6.5% in 1978, in 1978 and 8.5% in 1979, each year 8.4% in 1980, thereafter./a 10.4% in 1981, 11.1% in 1982, 12.8% in 1983 and 13.2% in 1984. Institutional Performance Very good. Laproving. Cumulative Estimated and Actual Disbursements (US$ Million) FY: 1979 1980 1981 1982 983 1984 1985 (i) Estimated 25.0 56.5 75.7 84.0 84.0 84.0 84.0 (ii) Actu.al - 16.4 38.9 58.5 77.8 33.3 84.0 (iii) Actual/Estimate % - 29.0 51.4 69.6 86.7 98.8 100.0 OTHER PROJECT DATA Original Actual or Item Plan Estimate First Mention in Files or Timetable - 05/20/76 Negotiations - 09/25/78 Board Approval - 11/14/78 Loan Agreement Date - 04/09/79 Effectiveness Date 07/10/79 08/21/79 Closing Date 12/31/82 12/31/84 Borrower Empresa de Energia Electrica do Bogota (EEEB) Executing Agency Empresa do Energia electrica do Bogota (EEEB) Fiscal Year of Borrower January 1 - December 31 Follow-on Project: Project Name Gavio Hydroelectric Power Project Loan/Credit Number 2008-CO Amount (US$ million) 359.0 loan/Credit Agreement Date May 1981 /a This was changed to 142 under Loan 1807-CO. - iv - MISSION DATA Mo/Day/ No. If No. of Staff Date of Type of Mission Year Dajs Persons Weeks Report Appraisal 12/18/77 10 3 6 -- Appraisal Follow-up 01/27/78 8 2 2 11/02/78 Total 18 8 Supervision I 07/28/79 5 1 1.0 09/06/79 Supervision II 07/17/80 11 1 1.5 08/05/80 Supervision III 10/13/81 7 2 2.0 10/03/81 Supervision IV 11/23/81 5 1 1.0 01/08/82 Supervision V 04/30/82 7 2 2.0 05/28/82 Supervision VI 06/24/82 3 1 0.5 07/30/82 Supervision VII 06/27/83 5 2 2.0 07/12/83 Supervision VIII 02/17/84 5 1 1.0 03/07/84 Total 48 11 11.0 STAFF INPUT (WEEKS) FY: 1976 1977 1978 1979 1980 1981 Preappraisal .1 1.2 4.7 - - - Appraisal - - 66.8 9.1 - Negotiation - - - 8.1 - - Supervision - - - 4.9 10.0 3.9 Other - - - - - - .1 1.2 71.5 22.1 10.0 3.9 FY: 1982 1983 1984 1985 1985 Total Preappraisal - - - - - 6.0 Appraisal - - - - - 75.9 Negotiation - - - - - 8.1 Supervision 7.9 4.7 3.7 .9 7.2 43.2 other .1 .7 01 - .9 8.0 5.4 3.8 .9 7.2 134.1 - v - PROJECT PERFORMANCE AUDIT REPORT COLOMBIA: LAS MESITAS HYDROELECTRIC PROJECT (LOAN 1628-CO) EVALUATION SUMMARY 1. The principal components of the project are two in-line bydroelec- tric plants (El Paraiso and La Guaca) near the Rio Bogota, linked by several tunnels and penstocks, using sequentially the same water, taking advantage of an unusually large head (1,900 a),1/ and making use of additional flow available in the river from another ank-financed project to bring domestic water supply to Bogota through diversion of sources in another watershed (the Chingaza diversion).2/ The layout of the project is shown on the last page of the PCR. Other Tiportant elements were the strengthening of an existing dam (Sesquile), and the construction of appropriate transmission facilities to deliver the project's output to Empresa's system, and to integrate it with the developing national network. The project remains uncompleted, because certain elements of the pressure tunnel system serving El Paraiso failed in test in May 1985. The original contractor remains on-site, and is carrying out remedial work under the direction of Empresa and its consulting engineers. The current cost estimate contains a notional US$20 million to defray the expense of this work. The tunnel is expected to be completed in December 1986, four ana one-half years late. ii. A concatenation of events and factors (discussed in the PCR in Sections III and V) has prevented the timely completion of the works, and provoked a deterioration of Empresa's financial position vis-a-vis that forecast in the appraisal. Although the PCR votes (Key Project Data, p. iii) "insitutional performance" is now "improving", and indeed is reasonable, that improvement is against the base of the early 1980s, when Empresa was shaky. The cardinal reasons for both physical and financial problems are: a. unforeseen, and probably unforeseeable, sub-surface conditions; a large number of investigations, that can be characterized as reasonable under the circumstances, were carried out, and the results did not zaise questions in the minds of the designer, Empresa, the international experts who reviewed them, nor the Bank's appraisal team; 1/ This head has already been exploited by a series of existing plants in cascade, essentially in parallel to Las Mesitas, using the then existing flow in the river. (See paragraphs 6 and 7.) 2/ See Staff Appraisal aport 2078a-CO, November 2, 1978, para. 4.03 et seq for a detailed description of this complex. - vi - b. failure of government to implement, in timely fashion, increases in Empresa's tariffs (particularly in 1980-1981) in cnntemplation of its undertakings (specifically Section 3.03 of the Guarantee Agree- ment, requiring government to permit Empress to comply with Section 5.05 of its Loan Agreement, to, inter alia, adjust its tariffs); C. a loss in value of the peso vis-a-vis foreign exchange at an average compound rate of about 20%, vs. about 9% foreseen at the time of appraisal; d. unforeseeable droughts which curtailed sales and revenues; e. general slackening of economic activity in Colombia (indeed, world- wide) during project implementation, which also affected revenues; f. the initiation of construction of the very large Guavio Project by Empresa at a time when Las Mesitas was still absorbing major management efforts, and requiring substantial expenditures; and g. the increased costs, principally interest and engineering, attribu- table to the long delays in completion. iii. Of the reasons under (b) - (g) inclusive, the more serious were (c), (f) and kg). They are discussed in Section II of this PPAM. The under- ground problems encountered (item a) are treated in somewhat more detail in 3ection IIIo iv. Some might question the design of the most upstream tunnel (Granada), which failed in test, particularly the extent to which it was unlined. It would be facile to raise such a question with the benefit of hindsight, and a clear-cut view may never emerge. The cost overruns in the penstock supply contract can be explained in a more straight-forward manner. In any event, these two elements contributed substantially to both delays, and increased costs. v. The PPAM examines the above factors. However, they are best viewed against the background of the electric power sector in Colombia, the role of the Bank in carrying out its appraisal and supervision functions, and finally, an examination of the lending environment which prevailed in the Bank and Colombia at the time of the appraisal. This background is elabo- rated on in the PCR, Sections I and II, as well as in an internal Bank paper, "Colombia, Power Sector Loan, Project Brief", July 30, 1986. It is also discussed in Section T of the PPAM. vi. The audit recognizes the great benefits which have accrued to Colombia and its electric power sector over the long association the Bank has enjoyed with the country and its electric power jupply agencies. Measured by world-wide standards, Colombia has done a most creditable job in the develop- ment of electric power. For example, it successfully accepted the challenge of interconnecting, initially, four major cities--Bogota, Cali, Manizales, and Medellin--in some six years, a major achievement. Comparable advances in - vii - providing dependable, low-cost electricity over wide regions in the United States, e.g., required decades. vii. Finally, Empresa, together with its consultants and contractors, has confronted the problems associated with the execution of Las Mesitas in a straightforward, business-like and conservative manner. PROJECT PERFORMANCE AUDIT MEMORANDUM COLOMBIA: LAS MESITAS HYDROELECTRIC PROJECT (LOAN 1628-CO) I. BACKGROUND The Structure of the Sector 1. Empresa is but one power supply agency in Colombia's large and complex electric power sector. It is municipally-owned, although it wau originally formed as a private corporation with Colombian capital. Its production is mainly hydroelectric, with one steam plant burning local coal. Its market is Bogota and its immediate environs. Other major cities are served by municipal systems, viz., Medellin (Empresas Publicas de Mdellin - EPM), Cali (Empresas Municipales de Cali - EmCali). Operating subsidiaries of the central government agen:y, Instituto Colombiano de Energia Electrica (ICEL), provide electricity in secondary centers in the central and southern departments. (ICEL is the successor to the national Instituto de Fomento Electrico y Aprovechamiento de Aguas--ELECTRAGUAS--itself operating many systems country-wide, some acquired after the expropriation and purchase of American & Foreign Power Company properties.) Another national agency, Corporacion Electrica de la Costa Atlantica (CORELCA) operates large gas- fired steam-electric stations in the northern departments, as well as a large transmission system, to supply its operating subsidiaries in that region with bulk power. Finally, the Cauca Valley agencies (Corporacion Autonoma Regional del Valle de Cauca--CVC-and its subsidiary jointly-owned with EmCali, Central Hidroelectrica del Rio Anchicaya Ltda--CHIDRAL) supply bulk power to EmCali, and serve a large number of secondary cities in Valle and Cauca. 2. Until the late 1960s, these systems (or their predecessors) were operated independently, as a consequence of their relative isolation from each other, the size of their markets, and to some extent, lack of national oversight of the sector as a whole. The Bank and the Colombian systems of Bogota, Cali, Medellin, and Manizales (Central Hidroelectrica de Caldas - CHEC, then a subsidiary of ELECTRAGUAS, now ICEL) cooperated to carry out studies which clearly demonstrated the desirability-both from the economic/ financial perspective, as well as Lhat of operations--of interconnecting these major markets. This endeavor was actively supported by the 'entral Government, and led to the creation of Interconexion Electrica Sociedad Anomina (ISA) which has since not only integrated the markets in Ientral Colombia just mentioned, but has also tied in with the later-created CORELCA network in the north so that a widespread national grid is now in place. Transmission is generally carried out at 230-ky, and a major north-south ISA-CORELCA 500-ky tie has recently been placed in operation, assisted by Bank financing. Both ISA and CORELCA have authority and responsibility to construct major plants (generally hydro and steam, respectively), as do ICEL, - 2 - CVC, EPM, and Empresa. ISA, as initially conceived,1/ has developed into the focus of planning and dispatch, its original owners (Empresa, EPM, CVC and CHEC) having been augmented by CORELCA and ICEL, which now represents CHEC, as well as others of its subsidiaries. The Bank has participated in financing a large portion of the expansions of all these components of the national system. 3. Finally, the sector includes the Ministry of Mines and Energy, which exercises general oversight; Departamento Nacional de Planeacion, which reviews central planning; the Junta Nacional de Tarifas; and Financiera Electrica Nacional, which provides financing, and which has also been assisted by the Bank. 4. Generally speaking, the sector organization is appropriate, and is working well, with the understanding that central system planning and opera- tions control needs strengthening (ISA), and t'e vestiges of inter-regional differences still prevail, as for example, reservation by certain large muni- cipal systems of major plants which would be more logically financed, con- structed, owned, and operated by ISA. Such a plant is Las Mesitas. The Atmosphere during Appraisal 5. At the time the Bank was preparing to appraise Las Mesitas--having accepted the proposition the project would be executed by Empresa rather than ISA--its views of the prospects for growth and increased levels of economic activity in Colombia were high. Resolution of the question of the order in which another plant and associated transmission should be built delayed appraisal of Las Mesitas,2/ so that by December 1977, when the appraisal was carried ouL, major civil engineering contracts had already been placed. While acceptance by the Bank of these circumstances was unusual, the appraisal team3/ noted that, on the positive side, having in hand bids from prequalified contractors lent credibility to the cost estimates. Neverthe- less, the Bank saw fit for its own account to hire a consulting geologist, who agreed, by and large, with the views of Empresa's consultants as regards sub-surface and surface conditions, and the design of structures to accommo- date them. The Bank's cost estimate upon which the Loan and Empresa's financing plan were based included contingency allowances of 24% for civil 1/ The original advocates (including the Bank) of creating a new entity to interconnect markets foresaw, as a major advantage, the ability of ISA's owners to pool resources to construct new plants, the size and cost of which were beginning to be large, in contemplation of the needs and resources of any one agency. However, traditional regionalism prevailed, and the owners reserved to themselves the construction of certain plants. 2/ op. cit., para. 4.18. 3/ Including a former Bank staff member with unusually extensive experience in dealing with Empresa, its system, and past construction activities. - 3 - engineering works, 15% for equipment, and 9% p.a. for price escalation. At the time, these allowances appeared reasonable. 6. In 1973, the then-estimated cost of Las Mesitas was examined at Empresa's initiative by a U.S. expert in such matters, skilled in reviewing such proposals in behalf of contractors preparing bids. The Bank ard the Colombian perties did n3t see fit to repeat this exercise in 1977, but the appraisal notes that there were substantial risks attributable to known geologic conditions, and problems which had arisen in connection with the parallel system cone.sructed by Empresa in the 1960s and 1970s at essentially the same site.4/ There had also been tunnel problems at still another plant (Chivor) const7ucted in the region by ISA and placed in service September 1977.5/ Further tunnel problems would appear in the diversion of domestic water7through the Chingaza Project carried out by Empresa de Acueducto y Alcantarillado de Bogota. Both these projects, and all Empresa's earlier projects, enjoyed Bank financing. 7. The records show the Bank was not always time.y in carrying out the appraisal which seems to have been under substantial pressures of time. For example, at a late stage, the Bank's (then) Central Projects Staff (CPS) insisted on an ecological study of the effects of the project, which was then carried out. Another recommendation, which appears not, to have received any attention at the Loan Committee level of review, was made by CPS that, in view of the substantial uncertainties revolving around the civil engineering works underground, and in cognizance of the problems experienced in the tunnel for the earlier plant at essentially the same site (Empresa's Dario Valencia, originally El Colegio II) mentioned in para. 6 above, the Colombians retain an expert Board of Review to oversee not only the design of Las Mesitas as it evolved, but also construction activity. No action was taken either at the Regional level, or by the Loan Committee. However, in the case of the Guavio hydroelectric plant now under construction by Empresa, financed by Loan 2008 of May 1981, Empresa and the Bank agreed upon institut- ing such a Board. 8. Finally, in contemplation of the Codigo Fiscal of Bogota (1985), Empresa was required to retain for construction supervision, consultants other than those who carried out the feasibility studies and design. Law 150 of 1976, and Decree 222 of 1983, require the same consideration, presumably to enable infant firms to enter the market. However, unlike the Law and Decree, which allow exceptions for unusual and/or complex circumstances, Article 349 of the Codigo does not. Empresa is absolutely bound by the Codigo, because it is a creature of the municipality. It is now moot whether the presence of several consultants contributed to delays, permitted lacunae 4/ op. cit., para. 4.24. 5/ Project Performance &udit Report 2720, October 29, 1979. in overall contractor surveillance, and generally added to the cost.6/ In any event, such arr.gements increased the risk of such problems arising. Engineering requirements for the 4-1/2-year construction period programmed were estimated at 4,760 man-months, or the equivalent of 88 engineers full- time, a very large effort indeed. The construction work itself was broken up into a number of contracts, more than six in total; although the engineering coordination might have been easier with fewer contractors, the work appears to have been well-executed. The financial failure of one Colombian contrac- tor, whose resources were limited, may well have been provoked by Empresa's own then-straitened finances, which resulted in delayed progress payments. In any event, it seems unwarranted for the Bank to criticize the contractor (PCR, para. 3.3.2): he had been selected by Empresa with the advice of its consultants and the consent of the Bank, and, following his bankruptcy, sought relief in arbitration, and indeed was found to have a case with some merit. The arbiter found for both Empresa and the contractor on different issues. In the course of its visit to Bogota in carrying out this audit, OED was advised both by Empresa's consultants and its other contractors, that prior to his collapse, the contractor in question had workedeatisfactorily. This contrasts with the Region's judgment. 9. At appraisal, Empress and the Government undertook to assure ade- quate financing by providing, inter alia, timely tariff adjustments where appropriate. That Empresa's finances were considered less than perfectly managed is reflected in the requirement (Schedule 2, Part G, Loan Agreement) that expert advice be made available in this field, and a training program instituted. The Bank's analysis of Empresa's financing plan yielded the unusual forecast that Empresa would be able to finance more than 70% of its capital requirements during the carrying out of Las Mesitas, and this target was agreed to and inc- orated in the Loan Agreement (Section 5.05). It was not achieved, more because of inflation's effects on operating expenses and devaluation's effect on debt service, than failure of revenues to develop. In any event, both Empresa and the Bank accepted a target which was unreal- istically high.7/ 10. Thus, as the curtain rose on this drama, expectations on both sides, Colombian and the Bank, were high for the economy, for Empresa's finances, and for the orderly and uneventful execution of the project. Unfortunately, events conspired to prevent their materialization. 6/ Engineering costs are coming in high, about twice the original estimate, attributable principally to the long extension of project execution time. 7/ Colombian power agencies had, in the past, been contributing at most 40%, about equal to the local currency fraction of construction costs, See para. 15, below. - 5 - II. THE PRINCIPAL DIFFICULTIES 1!. OED believes the problems experienced by Empresa during the execu- tion of Las Mesitas may be thought of as falling into three broad categories: 1. Cost Overruns; 2. Net Revenue Shortfalls; and 3. Overambitious Construction Planning. 12. As to "Cost Overruns", they, in turn, arise from four principal elements of the project: (i) The increased cost of the penstock supply contract. This terminol- ogy is misleading, insofar as there were two contracts placed for penstocks, one with a foreign firm, and one with a Colombian firm. The Colombian contractor was unable to deliver the material and equipment required, and the vdlue of the foreign contract was increased by this extent. Moreover, not only penstocks were involved: the foreign contractor also supplied the steel for tunnel lining (in amounts greater than originally anticipated, as work progressed and the need arose), and was called upon to furnish a valve not contemplated in the original design. Finally, steel prices escalated sharply during the period between soliciting offers aud awarding the contracts. All these effects combined to prodrt-e what, at first glance, appears to be an unreasonable jump in costs. It probably was not, given the circumstances. All in all, these factors added some US$25 million to the overall cost of (PCR, para. 3.2.1a); (ii) The additional work required to establish the integrity of Granada tunnel, now underway at a cost not yet fully known, but estimated to be about US$20 million (PCR, para. 3.2.1b); (iii) The additional engineering required by virtue of both the above difficulties, but attributable largely to the much longer than anticipated construction period. An additional cost of some US$19 million will have been incurred (PCR, para. 3.2.1c), so that engi- neering costs will likely be twice the appraisal estimate; (iv) Finally, owing to delays in completion, interest charges mounted up, and are likely to be an additional US$71 million, or a total nearly five times the original estimate. 13. Thus, in toto, these four factors alone account for an increase in project cost of se US$135 million equivalent, 49% of the original estimated total cost of US$276 million equivalent (including interest during construc- tion). The total cost overrun is likely to be about 81%, including interest (PCR, Annex 5). -6 - 14. As suggested in the Evaluation Summary, para. iv, with the benefit of hindsight, one might question the design of Granada tunnel and its related structures. The design was accepted by the Bank in the course of the ap- praisal, and rightly so. Indeed, it had been carried out by an engineering firm of outstanding repute and vast experience in not only such designs, but also particularly the specific requirements of Las Mesitas' site, and certain critical design parameters had been reviewed by foreign experts of high international standing. The Granada tunnel system difficulties are cardinal to the failure to complete Las Mesitas on time, and are examined in more detail in Section III. The Bank was well aware of the problems which were emerging, as revealed in the first Project Supervision Report (see p. iii, for the chronology of the supervision activities carried out at intervals of roughly six months on the average), which reported the project's estimated cost had already risen by US$67 million equivalent (26%) by July 1979, some seven months after the Loan had been approved. Although subsequent supervi- sion missions reported virtually continuously increasing costs, the Bank appears not to have believed there was any action it might have taken to ameliorate circumstances, and indeed, there probably was not much the Bank could have done. Nevertheless, the Bank proceeded to process the Guavio loan in the face of mounting difficulties (para. 18, below). 15. "Net Revenue Shortfalls" were occurring simultaneously, as economic activity was taking place at a pace less than anticipated; inflation was more than twice that expected, driving up operating expenses; the peso was being devalued, increasing pari passu Empresa's fnreign debt service burden; and a prolonged drought was experienced during 1980-81 which so curtailed hydro- electric production that power supply in Empress's service area was at one time limited to five hours daily. As shown clearly in the PCR (Annexes 8-12), none of the financial performance expectations materialized. Indeed, by 1980, it had been recognized that the self-contribution target was unreal- istic, and in connection with Loan 1807-CO, it was reduced to 35% from 70%. Finally, revenue loss due to shortfalls in sales was exacerbated by failure to place into effect tariff increases as agreed, particularly during 1980- 1981 in the drought period, when it might well have been imprudent to attempt to raise tariffs while rationing supply. In spite of these delays, actual average tariffs in effect were about as projected in the appraisal, until 1982 when dramatic increases much greater than forecast were becoming the rule. However, the rub was that prices were also escalating, and devaluation was increasing Empresa's debt service burden very substantially. All told, during the project execution period of 1978-1982, these effects combined to reduce revenues by only a minor amount, from the Col$39 billion forecast in the appraisal to Col$38 billion. Hovaver, total contributions from internal cash generation for the same period fell short of projections by Col$10 billion, 42%, as operating expenses and debt service increased dramatically. 16. Finally, Empresa undertook the construction of the Guavio project, a very large hydroelectric installation foreseen neither in connection with the appraisal of Las Mesitas, nor the then generally accepted national plan for new generation developments agreed upon by the major power supply agencies, ISA and ICEL, and reviewed by the National Planning Office. Guavio - 7 - was financed, in part, by Loan 2008-CO of May 1981, at a time when Empresa was already experiencing severe strains, not only financially, but also in managerial capabilities. 17. The decision to construct the US$1 billion, 1,000 mw Guavio Project was obviously sanctioned by the Bank. This undertaking imposed a large addi- tional burden on Empresa,o/ although during 1978-1980, construction expendi- tures had been incurred at a rate far below that anticipated, viz., Col$10 billion vs. Col$17 billion (PCR, Annex 9). However, during 1981-1984, expen- ditures accelerated to Col$89 billion vs. Col$46 billion projected when Las Mesitas was financed. It seems evident, again with the benefit of hindsight, initiation of Guavio exacerbated the deteriorating circumstances in which Empresa found itself, as a consequence of the several factors discussed above, 18. Guavio was appraised in December 1980 by a Bank team none of whose members had participated in Las Mesitas' appraisal three years earlier. The Staff Appraisal Report (3408b-CO), May 6, 1981, notes in para. 2.02, discus- sing Empresa's construction activities then underway, "The disbursements of the fourth loan9/ began in September 1979 and the construction is progress- ing generally sitisfactorily although with a delay of three to six months and some US$62 million of cost overruns due to changes in base prices (actual prices were higher than original cost estimates)." In contemplation of these facts, the term "generally satisfactorily" seems unwarranted. Recall that the appraisal report for Las Mesitas (2078a-CO, November 2, 1978) in para. 4.18 tells us "Approximately 90% of the Mesitas civil works have been under contract since September 1977. The largest part of these, the tunnel and shaft works, for which the propoied loan would cover US$28.1 million in foreign exchange costs, was contracted under ICB consistent with the Bank's guidelines...." These works were described in para. 4.11 as "the Mesitas items most prone to estimating errors". Therefore, there was every reason to believe the cost estimate for Las Mesitas was reasonable. Nevertheless (as noted in para. 14 if this memorandum), by July 1979, i.e., 18 months before the Guavio appraisal, a Bank supervision team was reporting, that Las Mesitas was already some six months late, and its cost was likely to be exceeded by US$67 million. Finally, by December 1980, it was evident that Empresa's sales and revenues were not materializing as forecast during Las Mesitas' appraisal: 8/ Empresa's plant and works in progress then totalled US$1.4 billion equivalent. ISA undertook to finance 40% of the local currency cost, or some 15% of the total cost, i.e., about US$150 million equivalent. 9/ Las Mesitas. -8- Net Cash Sales Revenues Generation Gwh Col$ million Col$ million 1977-79 1980-83 1977-79 1980-83 1978-79 1980-83 Las Mesitas Appraisal (December 1977) 11,079 20,788 22,417 44,963 4,257 24,035 Guavio Appraisal (December 1980) 10,896 18,687 10,414 56,539 -- 26,076 Actual 10,896 16,639 10,414 50,436 2,898 15,733 All these circumstances were known at the time Guavio was appraised, and Empresa's finances were deteriorating, at least so far as its ability to contribute substantially to the construction of Guavio was concerned. They were apparently not considered serious by the Bank at the time the loan for uavio was being processed. 19. In summary, it would appear that, despite all reports about Empresa's straitened financial circumstances being attributable to failure of sales to materialize, and late tariff adjustments, the major factors produc- ing the current problems are: (i) design and construction difficulties at Las Mesitas: these increased costs directly, and caused consequential increases in engineering and interest expenses, as time was lost. This seems to be a good example of the so-called "Ripple Effect"; (ii) financial burdens worsened by external forces, viz., devaluations; and (iii) construction and financing of the ambitious Guavio Project by Empresa alone. Of these contributing factors, Emvresa had control over only the last; but then, so did the Bank. III. THE GRANADA TUNNEL PROBLEM 20. By referring to the layout attached to the PCR, it can be seen that Las Mesitas is quite complex: there are three tunnels, two penstocks, a siphon, a surge chamber, and two power plants. It is an altogether ambitious and challenging project, designed cleverly to take advantage of the topogra- phy and to develop an extremely large head in two manageable "bites". The principal underground difficulties arose in the Granada tunnel, and its appurtenances. -9- 21. In its original conception and alignment, Granada appears to have been prudently designed. There seems to be a general consensus on this point among Empresa, its design engineers, and the general contractor. (The super- vising engineer has assiduously avoided commenting upon the design.) The designer had foreseen about 3-1/2 km of concrete lining (out of a length of 13 km) for structural reasons, the balance to have been supported by rock- bolts, ribs, and shoterete.10/ Before construction began, the tunnel align- ment was rearranged to shorten the length of the adit near km 7 ("Ventana"). This brought the tunnel under a lower overburden profile, which increased the support requirement because the hydraulic head in the tunnel when operating would be greater than the in situ pressure in the surrounding rock. In this section, some steel lining was introduced to assure the tunnel's integrity. As work progressed, locations requiring greater than design support were encountered, and concrete lining was increased to 6 km from the 3-1/2 km originally contemplated. An appropriate grouting program was carried out. 22. A potential problem was foreseen at the surge chamber location, just upstreim of the first power plant, El Paraiso. Here, the topography was such that the tunnel was encased in a formation known as a "nose", i.e., a narrow ridge, with its downstream end falling off sharply. The possibility of causing hydraulic fracturesll/ in this zone was recognized, and tests were carried out to better quantify the risks. The results of these tests were reviewed not only by the concerned Colombian agencies, but also by several geologists, experts in rock mechanics, and construction specialists of international repute. The consensus was that it would be prudent to increase the steel lining contemplated for the tunnel at this zone, but that the surge-chamber would not require any special support or lining. 23. Subsequently, the tunnel was tested to full operational pressure in May 1985, and it failed in two zones, attributable to hydraulic fractures: the surge-chamber, and near the adit.12/ Leakage of some 400 litres per second was observed, substantially above an acceptable level. The tunnel was dewatered, designs reviewed, and a program of remedial works, now nearing completion, was undertaken. Although the contractor argues that the entire tunnel should be concrete lined for structural reasons, the designer poInts out that the failures (except the surge-chamber) were in zones where concrete lining was in place. The remedial works undertaken will see the tunnel lined with reinforced concrete some 65% of its length; steel lined sections 10/ The contractor calls this the "New Austrian Tunnel Method", to which the designer takes exception. 11/ Fracturing of the rock surrounding the tunnel by virtue of (1) the high pressure in the tunnel, and (2) the low restraining forces in the supporting rock. 12/ One consequence was the contamination of the water supply system for the town of Las Mesitas, because the waters used by Empresa in all its plants are Bogota's untreated sewage. Empresa quickly intervened to correct the problem, and make restitution. - 10 - increased in length; the surge-chamber steel-lined; drainage galleries driven near the surge-chamber; and extensive consolidation grouting effected throughout. 24. Empresa, its designer, and its contractor (who may disagree with the final design) are all working harmoniously and cooperatively to complete the job successfully. As this was being written, the remedial works were virtually complete, and pressure testing began for September 1986.13/ Pend- ing the outcome of this final step, the success of the remedial wFrks will not be fully established. It seems clear that the original design was appro- priate, and the remedial works underway to meet the fallures are also appro- priate to the circumstances, and conservatively based. IV. THE BANK'S ROLE IN ADMINISTRATION 25. A review of the record reveals what appears to be an unusually large volume of correspondence between the Bank and Empresa regarding pro- curement, the tone of some of which reflects pique. Document preparation in particular was questioned by the Bank, but so was the award process in some instances. (Recall that most construction contracts had already been placed by the time Loan 1628-CO was approved.) The Bank sent14/ to Bogota its Pro- curement Engineer, presumably to deal with Empresa's ;Eaff "on the ground", which was prudent, because Empresa's management and staff were not familiar with Bank procedures. Unfortunately, the results seem to have been equivocal. 26. The intensity of supervision seems to have been reasonable. The PCR (p. iii) indicates there were eight supervision missions. (The files indicate there were 9 or 10: the difference in counting is probably attribut- able to nomenclature. Not all supervision missions dealt with the construc- tion problems.) While supervision by the Bank was revealing the deteriorat- ing situations, both financial and in project execution, little seems to have been done by the Bank to attempt to ameliorate things. Perhaps, under the circumstances, the Bank had little to contribute. 27. It is tempting in the light of events to conclude that the Bank and the Colombian agencies moved forward with its processing too quickly: divided engineering responsibilities, fragmented construction activities, and failure to institute a Board of Review all may have contributed to making a .difficult job more difficult than it need have been. 13/ Empresa began to fill the tunnel Septemer 17, 1986, and expects to reach full head October 6, 1986. As of this writing, testing has been uneventful. 14/ There is evidence in the Bank's archives that such a visit was made, but the record is silent with respect to its outcome. * - 11 - 28. It is odd that this project was never selected for discussion at the SVPOP level in the course of the Bank's global semi-annual project imple- mentation reviews. Indeed, there is no evidence it was discussed at the Regional Vice-President level. The Region notes that such reviews are called for only when Senior Management guidance is sought in connection with problem solving. 29. OED is compelled to take note, having reviewed the Bank's files, interviewed m&ny key Bank staff and held discussions with Empresa in Bogota, that the various stages in the carrying out of Las Mesitas--conception, design, procurement, construction, testing--took place over a period when Bank staff were frequently shifting assignments, and personalities coming into contact with Empresa and the project were changing. Thus today, none of the staff directly involved in loan administration has a clear view of more than a small part of the project and its development. The Bank is, in this Project at least, losing its "institutional memory", an apparent consequence of staffing expansion, and the promulgation of the rotation policy. While these factors may be inevitable in the operation of today's institution, nevertheless, there is a substantial cost associated with this phenomenon. V. REALIZATION VS. EXPECTATION 30. The principal purposes of the Project (as defined in Schedule 2 of the Loan Agreement)15/ were to: (i) provide additional generating capacity to Empresa's system by December 31, 1981. This was not achieved: completion is taking five years longer; (ii) carry out related works, including strengthening of an existing dam at a site remote from the Project, and construct transmission facilities appropriate to the Project, and Empresa's and ISA's requirements. The dam was improved, but works were delayed several years, albeit within the date prescribed by the Loan Agreement, Transmission is in place and operating; (iii) study the valuation of Empresa's assets; this was not done under the project, and the work was transferred to Loan 1807-CO. It is complete; 15/ The Latin America and Caribbean Regional office of the Bank (the Region) would prefer the statement of purpose, as it appears in the President's Report. Thus, item (i) would be "to satisfy part of forecast power demand at least cost". This appears to have been achieved [PCR, para. 4.1.2 (ii) 1. However, Empresa was not a "party" to the President's Report. - 12 - (iv) strengthen Empresa's financial management. This appears to have been achieved, if financial performance is used as the criterion for rendering judgement. When one makes such a judgement, one must recognize that original financial targets were substantially lowered. Indeed, original targets seem in retrospect to have been wildly optimistic. Both Government and Empresa should receive credit for the very substantial tariff increases which were event- ually instituted (PCR, Annex 8). VI. CONCLUSIONS 31. Given the problems experienced at El Colegio, Chivor, Las Mesitas, Chingaza, and now Guavio, it seems obvious that future plant designs in the area should lean towards conservatism, Rnd be subject to review before time becomes of the essence. OED agrees with the PCR recommendation (para. 7.2.1) that a "Board of Consultants would be advisable". OED sees the possibility that the timings of both Las Mesitas and Guavio were accelerated by perceived urgency which was not, in the event, warranted. It seems that the financial aspects of the appraisal and the subsequent document review were somewhat superficial under the pressures of time (para. 7, above). Explanations notwithstanding, the project is seriously late and coming in greatly over cost, Here, OED must disassociate itself from the PCr finding that the project is "successful". It is "successful" to the extent that if costs are cast in cc- tant-value currencies, the overrun is much smaller (as para. 3.2.1 of the PCR notes), and the demand for power did not materialize as foreseen, so that the delay has not been as serious as it might have been, but this view is tantamount to making a virtue of a vice. The Region argues that expansion plans must not be static, but reviewed and revised in the light of changing conditions. Whereas the audit agrees with this, in princi- ple, it did not find indications that. in the ease of Las Mesitas, a substan- tial part of the delays was, at one point in time, planned. It rather seems that postponements of commissioning were integrated into the planning, once the delays had occurred. This appiies, in particular, to those attributable to the unanticipated difficult geological conditions in several places (PCR, para. 3.1.5) and to the lacking water tightness of the Granada tunnel. 32. The Bank, in dealing with what appears to have been a relatively new management team at Empresa, failed to recognize, in timely fashion, the problems that team was having with its (the Bank's) procurement policies. 33. Las Mesitas seems a clear case where "complex and unusual" circum- stances prevailed, and the Bank should have attempted to call into question the application of Article 349 of Bogota's Codigo Fiscal, which fragmented the engineering (para. 8, above). Again, had not judgements been blurred by a sense of urgency, this matter might have been prosecuted, and eventually fever civil engineering contractors and consultants, with the attendant need to coordinate all their activities, employed. - 13 - 34. The possibility for the Bank to have intervened more aggressively in connection with its supervision activities (PCR, para. 7.5,1) may well be exaggerated. Frequency does not seem to have been unreasonable. One wonders what Bank staff, visiting from Washington, might have seen that the people in the field did not. By the time supervision began (reasonably soon), the die was east: the cost of the penstock contract would prove to be underesti- mated, and the underground troubles were lying in wait. 35. OED must lament the loss of the Bank's "institutional memory" as personnel are reassigned. Twelve Bank staff at Headquarters (or retired), who were involved at one stage or another were interviewed: none had the opportunity to follow Las Mesitas from conception to the present day. 〞／咨f妒吵紛Jw學子 一名／一 - 15 ATTACHMENT 3. I N G ETE C S. A. Conments from the Borrover's; DesAgn Agineers INGENIEROS CONSULTORES CAWWRA 0 W 30 A- 30 0000T* COLOM&A ~?A00 AMO 9099 V94AX wto 044618 CAI"S: INIXTEC TUara"o Rev 4100 MT. 460. 004. 9" December 11,, 1986 Mr. Brian Shields Chief Energy, Infrastructurg and Urban Devel opment Operations Evaluation Department The World Bank 1818 H Street N. W. Washington D. C. 20433 Ref.: Mesitas Project (Loan 1628-CO). Dear Mr. Shields: We are glad to comment on the two reports transmitted with your letter of November 17, 1986. 1. TUNNEL LINING: To correct a few discrepancies in the report, please refer to Annex 1. 2. OTHER COLOMBIAN PROJECTS: To clarify reference made to other Colombian projects$ please refer to Annex 2. 3. COST OF THE PROJECT: In reference to the overrun in cost and in time, comments appear in Annex 3. - 16 - ING ETEC 4. SESQUILE DAM: Regarding Sesquile , where the protection of the dam is not finished, the situation is as follows: Tne investment so far has been US$3.5 million (1977 prices) in the reinforcement of the foundation in the downstream berm. The upstream berm, not yet built, would cost US$6.4 million (1977 prices), about US$ 8.0 million at present prices. Empresa apparently plans to start this work soon. 5. PRESSURE TUNNELS: HYDRAULIC FRACTURE: Regarding the hydraulic fracture phenomena, that have resulted in problems in so many projects: Chivor II and Mesitas in Colombia, Helms and Bath County in the U. S., Kotmale in Sri Lanka, Hattelberg in Austria and Herlandsfoss, Bjerka, Askor., Birte and Skar in Norway, please refer to Annex 4. 6. CONSULTANTS AND CONSULTING BOARD: Regarding the use of consultants and consulting boards a few comments are pertinent: Ingetec has used individual consultants with great success. The system is very flexible and experts are invited as required, after very care- ful screening. They are essential for the transfer of technology. Their use depends on the type of project and type of problem. They may be replaced at will and new consultants may be added as needed. INGETEC - 17 The use of Consulting Boards is more conventional. For large complex projects, like Guavio, a Board is a very satisfactory solution. For small projects a Board may be a hindrance. The selection of a Board is a problem and it is not easy to replace members. A Board of Consultants could have helped in Mesitas, as it has in simi- lar projects like Salvajina, El Cajon and Guavio. There was no shortage of consultants; a total of 12 international per- sonalities were consulted by Empresa, Impregilo and Ingetec. No single consultant was entirely correct in his advice (See Annex 5). Itq,etec still feels that the use of consultants by contractors rarely is conducive to good results. In Chingaza, the consultant brought by ICA, recommended lining 18 km of tunnel with reinforced concrete. This was incredible and unnecessary. In Guavio, a group of experts brought by the contractor (SB-CB) made proposals better oriented toward claim negotiations than toward the benefit of the project. In Mesitas, Impregilo brought two consultants who, in all their wisdom, recommended conventional unreinforced concrete lining from Rodeo to Ventana as a solution for all problems. As already indicated, the problems had nothing to do with rock support but with water pressure, hydraulic fracture and leakage of contaminated water. Unreinforced concrete is well known as a useless element to resist internal pressure and avoid leakage. INGETEC - 18 - 7. LESSONS: a. Geological Investigations: The Bank Operational Staff questions the adequacy of the stu- dies and suggests that "a more extens*ve geological prospecting using more drilling, geoseismic and geoelectric methods and other modern technologies would always be well worth and should be requested for future hydroelectric projects." With the experience of Colegio, Chivor, Chingaza, Salvajina, Tunjita and Rio Negro Tunnels, Mesitas, El Caj6n and Guavio behind us, we feel very strongly that there is a clear limit to the usefulness of geolo- gical investigations and that no amount of additional prospecting would yield information that would prevent surprises in tunnelling. In the case of pressure tunnels, systematic measurement of rock stresses, after excavation, is needed to determine whether the lining or the rock is to resist the water pressure. Empirical rules are useful only for design layouts. In the case of dams, the eseewMe concentration of geological investigations is justified and, specially for high dams, very detailed geological studies are necessary to avoid surprises. b. Tunnel Alignment: We believe now that the shorter adit may not have been a wise decision. It was difficult, at the time, to predict the behavior of the INGETEC - 19 - sector of tunnel with low rock cover. Empirical rules for adequate cover had proven successful in several projects, so that there was no objection to the reduced cover. The short adit brought savings in tunnel construc- tion and during inspection and repairs. c. Rock Stresses: Tne "nose" theory of rock stresses became internationally known afer 1982. Thus, the cracking of the surge shaft was a surprise for many of the international consultants. d. Cost Estimates: Penstocks: The original estimate was low in the case of the penstocks. This was caused by a great increase in tne unit price of steel between the time of the order of the steel liner for Chivor II and the Mesitas penstock. (This may have been due to lack of competition for very high pressure large diameter pipe). Another problem was a low estimate of the metal appurtenances to the surface penstocks (stiffening rings, supports, expansion joints, manholes, etc). Again lack of competition could have been a factor. e. Cost Estimates: Tunnels: Although we consider ourselves very experienced tunnel engi- neers, especially after Chingaza, the construction program for the Granada tunnel was short and the cost estimate should nave had greater contingen- cies. Tnere are frequent complications in tunnelling and even a very - 20 - ING ETEC capable contractor like IGL could be short of the type of personnel required to overcome, with little delay, the short sections of very dif- ficult ground . f. Colombian Contractors: Local contractors have a role in projects like Mesitas, but the fragility of the financial structure of all such contractors has to be recognized. The contract administration has to consider such conditions, it has to be very flexible and timely financial assistance has to be provided to assure completion of the work. There are few, if any, local contractors with adequate financial resources. Colombia has not had a policy of clear discrimination in favor of local builders. This would be a first requirement to develop a group of prosperous firms. Mexico and Brazil are two countries where such discrimi- nation has been in effect for several decades with telling results. g. Red Tape: There is a bureaucratic straight jacket in Colombia that delays every project considerably. Without this problem the repairs of the tunnel could have taken half the time, with considerable saving in cost. This problem also affected the handling of all contracts. There is such a delay in approving and legalizing a contract that all agi- lity in hiring and firing contractors, or, for that matter, engineers or suppliers, is lost. This is a colossal handicap in contract administration. - 21 - ING ETEC To make things worse, the tendency is toward more red tape. FINAL COMMENT : The Mesitas Project has been under testing since early October '86. It is now generating 200 Mw and should be in normal operation by the end of January '87. The tunnel tests were completed in early November '86 and no further work to reduce leakage is anticipated. The total cost of US$500 million for 600 Mw, with conveyance (intake, tun- nels, penstocks, and tailrace) dimensioned for 1200 Mw, makes Mesitas one of the most economical projects in Colombia. Sincerely yours, INGETEC S. A. C. S. Ospin -22- ANNEX 1 COMMENTS BY INGETEC MESITAS PROJECT (Loan 1628-CO) LINING OF THE GRANADA TUNNEL There are discrepancies in the length of original and final lining. The situation was as follows: Granada II Tunnel: Original % Final % Length of lining: Conventional concrete lining 6 316m 57 5 703m 51 Shotcrete lining 3 378m 31 2 906m 26 Unlined 791m 7 510n 5 Circular steel lining 531m 5 623m 6 Conventional reinforced concrete lining 50m 0 599m 5 Reinforced shotcrete lining Om 0 725m 7 SURGE CHAMBER SHAFT LINING The surge shaft was originally lined with shotcrete (224 m) and conventional reinforced concrete ( 8 m). The final lining was: Steel lining 223 m, reinforced shotcrete lining 9 m. Another comment that is pertinent is that, if everything had been anticipated in the original design, some of the reinforced conventional enncrete lining and most of the reinforced shotcrete lining would have been replaced by steel - 23 - lining. It was difficult and expensive to place additional steel lining after testing the tunnels and ,therefore,it was limited to places where it was essential, like the surge chamber and a short section near the adit. - 24 - ANNEX 2 COMMENTS BY INGETEC MESITAS PROJECT (LOAN 1628-CO) TUNNEL PROBLEMS IN COLOMBIA Reference is made to a number of Colombian projects and it would be worthwhile to review what happened. Colegio: Project built in 1964 to 1967 near Mesitas. Delay in tunnel due to heavy ground, soft rock. Tnere were no leaks as the rock proved impervious. There was no hydraulic fracture, although the tunnel was under pressure, with a 75 m head. There were several major slides along the penstock and near the powerhouse. The Colegio experience, to a great extent, helped to avoid slides in the Mesitas surface penstock, where surface slides have been few and unimportant. Chivor II: The project is located 150 km to the northeast of Bogota. The problems developed in 1982 had to do. with leakage through karstic tubes at Muros and as a result of hydraulic fracture, near the main pressure shaft. As soon as this accident took place, Ingetec added steel lining to the Mesitas tunnel, near Petras Blancas, and probably avoided another problem. - 25 - Unfortunately, this lining was not extended to include the surge shaft. Chingaza: The project is located 20 kms to the east of Bogota. The tunnel is 40 km long with 10 km lined with conventional concrete and 30 km with very thin shotcrete lining or without lining. The problem was caused, apparently, by a very odd condition where a reac- tion developed between water, calcite and pyrite, expanFW and breaking the very thin shotcrete lining andvrockfalls. There were about 40 rock- falls and the tunnel was closed in a number of places. Two kilometers of tunnel were lined with conventional unreinforced concrete. Therefore, there are now 12 km lined with conventional concrete lining,out of the total of 40 km. The project is now in service. The accident in Chingaza had no lessons for Mesitas. - 26 - ANNEX 3 COMMENTS BY INGETEC MESITAS PROJECT (LOAN 1628- CO) OVERRUM IN ZOSTS AND ADDITIONAL TIME REQUIRED FOR CONSTRUCTION Construction Time: The construction of the project was programmed to start in 1976 and be completed by the end of 1981. The tunnel was started by the end of 1977 and was completed by September 1986, with a five year delay. The delay had to do with the trouble in crossing an unstable area, with additional steel lining at Pefas Blancas, with the first test and repairs of the door at Ventana, with the second test and repairs at Ventana, Rfo Seco and Pefas Blancas and with the third test and repairs at Ventana, Rio Seco and Pefias Blancas. The Mufla III pumping station was completed with a delay of nearly two years. The two powerhouses were completed by mid-1983, with a delay caused by the failure of the contractor. There were delays in connection with the supply of all the equipment, but the critical factor was the delay in the erection of the equipment that was completed in late 1984,and with a considerable delay in the completion of the penstocks ( completed in mid-1985-.); it,took more than a year to. repair cracks that developed near shopwelds along the pipe. Summarizing the delays, the critical path is controlled by the tunnel, but it is good to remember that the penstocks were completed only in mid-1985, so the actual delay caused by the tunnel alone was about 1 year*. - 27 - Cost Overrun: The cost overrun is clearly described under Annex 5 in the report prepared by the Bank. It would be necessary to add 6.4 million dollars corresponding to Sesquil6 upstream embankment and 0.5 million dollars corresponding to the supervision of this work, for a total of 6.9 million dollars. Pending: Purchase and installation of equipment for pressurizing the three powerhouses (Muia III, Paraiso and Guaca),in order to reduce the corrosive effect produced by hydrogen sulfide coming from the water in the turbine pits. Civil morks for these installations are complete. (A reserve of US$0.3 million would be ample). Engineering costs: (million US dollars) Actual Original Difference estimate Project cost 279.3 186.0 93.3 Engineering cost 36.1 17.5 18.6 Engineering percentage 12.9% 9.4% .3.5% The longer period of supervision is not the-k* factor for the high cost of engineering. It is a factor, but in addition there are a number of new engineering efforts required by fundamental changes in the design: a) Complete design of the conveyance to duplicate the capacity, including intake, tunnels,siphons, surge chamber, valves, penstocks, tailrace, etc. b) New design for the substation at Paraiso and Muha III with the SF6 system, - 28 - c) Complete change of the refrigeration system at each powerhouse. d) A complete design of the tailrace system at Paraiso. e) Preparation of bid documents for the completion of the powerhouses of Paraiso and Guaca when the contractor was changed. New documents for the fabrication and erection for the Paraiso spillway. f) Engineering assistance. In connection Wa emergency work Vs6644nv ivflooding in the area of the project,that took place in 1982, including work for Tinta warehouse and other engineering services. - 29 - ANNEX 4 COMMENTS BY INGETEC MESITAS PROJECT (LOAN 1628-CO) PROBLEMS IN THE CONSTRUCTION AND TESTING OF THE GRANADA II TUNNEL IN THE MESITAS PROJECT. There were several problems in the Granada II tunnel: 1) During excavation there was a bad spot with water, fractured rock, flowing sand and crushed rock. This is a normal situation in a tunnel and it was solved with conventional methods. This sector of tunnel was concrete lined and it has behaved well during testing. 2) The change in design, to shorten the adit, was made just before the proposals were received for the construction of the tunnel. This de- cision reduced the tunnel cover in a long sector and may be related to the hydraulic fracturing near the adit door (Ventana) and to the leakage at Rio Seco. 3) The first problem with the tunnel at the adit (Ventana),whethe first pressure testAag tv g on April 18, 19841-rhere was a small, structural failure in the frame of the entrance door. This was repaired quickly, but delayed the full itesting of the tunnel. 4) The first full pressure testing caused four failures: a) Hydraulic fracturing at Ventana caused by a low strength rock stratum crossing the tunnel. This was a geological defect impossible to anticipate,that changed the distribution of the stresses in the rock - 30 - and created a condition of failure in a sector where the tunnel cover should have been adequate. . . b) Cracks in several areas near Ventana and under Rio Seco. c) Leakage at Rio Seco caused by a pervious zone. This area was protected by a short steel liner. d) Hydraulic fracture in the shaft of the surge chamber. A sec.tor of the vertical shaft developed cracks. After these failures it was decided to add steel lining to a sector of tunnel near Ventana, to do some additional grouting at Rio Seco, to build a drainage gallery and drainholes near Pefas Blancas and to seal cracks in the surge shaft. The new testing of the tunnel did not prove satisfactory. The quality of the water leaking in Ventana and at Peflas Blancas made leakage unacceptable. Although the tunnel could be operated,if it had been handling clean water, the effect of the contaminated water was such that it was decided to stop the test and to do additional work in the tunnel and in the surge chamber shaft, to reduce the leakage to manageable flows at Peflas Blancas and Ventana. A short comment on the contaminated water probably is pertinent. The Rio Bogot& system (Canoas, Salto, Laguneta and Colegio) with an installed capacity of 550 Mw and 25 km of pressure tunnels, some lined in plain concrete and some unlined, is located along the Bogotf river canyon. These plants have been in service between 20 and 50 years. The steel lined sectors are very short and mostly concentrated at the head of the penstocks. There has *not- - 31 - been in these 50years,a single complaint caused by leakage from the tunnels. The surface area in this canyon is heavily populated. Apparently, the leakage from the tunnels, if any, is filtered by the rock and mixes with ground water of good quality. The effect of the hydraulic fracture at Mesitas was to have larger cracks where the filtering effect was absent ; there was no dilution and the quality of the leakage was identical to the quality of the watea in the tunnel. Thus it was essential to control the leakage. The final program was as follows: a) To line with reinforced concrete some sectors near Ventana and Rio Seco. b) Additional grouting in several sectors of the tunnel to reduce permeability. c) To steel line the surge shaft. d) To line with reinforced concrete 500 m of tunnel upstream of the surge shaft,as an additional precaution to avoid leakage. -32 - ANNEX 5 COMMENTS BY INGETEC MESITAS PROJECT (LOAN 1628-CO) LIST OF TUNNEL CONSULTANTS WHO PARTICIPATED IN THE MESITAS PROJECT EMPRESA'S CONSULTANTS - Dr. Gabriel Fern&ndez, U.S.A. and Colombia - Dr. R. S. Schmid, Motor Columbus, Switzerland INGETEC'S CONSULTANTS - Dr. Raymond Benson, CanadA - Dr. Tor Brekke, Norway and U. S. A. - Dr. Einar Broch, Norway - Mr. J. Barry Cooke, U. S. A. - Dr. Johann Golser, Austria - Mr. James Libby, U. S. A. - Mr. William F..Swiger, U. S. A. - Mr. John Trantina, U. S. A. IMPREGILOIS CONSULTANTS - Dr. Thomas Locher, Switzerland - Dr. Sergio Marchini, Italy  ..33 ATTACHMENT 2 Comments from the Borrower IliggggggM Bogoti, diciembre 26 de 1986 Page_1 of 2 E-556/87 January 13, 1987 Spanish/Colombia OED/30 DBB:bas EMPRESA DE ENERGIA ELECTRICA DE BOGOTA Bogota, December 26, 1986 Mr. Brian Shields Energy, Infrastructure and Urban Development Operations Evaluation Department The World Bank 1818 H Street N.W. Washington D.C. 20433 REF: Mesitas Project, Loan 1628-CO Dear Sirs: We refer to your letter of November 17, 1986, concerning the OED report on the Mesitas Hydroelectric Project. Please find attached our comments on this report. Yours sincerely, Juan Manuel Latorr6 Deputy Technical Manager AVENIDA "EL DORADO" No. $$*$1 APARTADO AEREO 4453 CONMUTADOR: 22148 11 - CABLES: ENERGIA - TELEX: 441242 BOGOTA, D. E. CA*2469003 - 34 - ATTACHMENT 2 Page 2 of 2 Comments on the World Bank Letter of November 17, 1986 Page vii (f) A number of unforeseen events that made additional works necessary delayed the coming into operation of the Mesitas project. This resulted in the completion of the works at the El Colegio Mesitas coinciding with the start of works at Guavio. Page 2 Regional considerations may have influenced the selection by Note 1 the enterprise of plants to be constructed; but in this case the decision was based on such important aspects as geographical location, bearing in mind the difference in levels over short distances and proximity to other existing generating stations. Page 3 Since the Mesitas project was to be a system running parallel Item 4 to the old set of generating stations set up by this enterprise on the Bogota River, regulated by the water of the Chingaza Project of the Bogota water supply system, it was preferable to have it constructed by the same enterprise. The operation of two systems which would have been necessary had some other group been responsible for construction would clearly not have been practical. Page 8 The contract for the supply and assembly of the pressure Item 12 (1) conduits was awarded to the Italian firm of Accialeria e Tubificio di Brescia S.p.A., which subcontracted the manufacture of some of the conduits to the local firm of Tissot. Responsibility for supplying the totality of the materials involved remained the responsibility of Acciaieria e Tubificio Di Brescia S.p.A. Page 17 Since the designer was in a position to provide his own Item 25 consultants to review the design, a committee of consultants was not established for the project. Page 22 The technical staff used at the various stages of construction Item 32 for the project were highly experienced, having been more than 15 years with the enterprise. Bogota, December 29, 1986 - 35 - COLOMBIA PROJECT COMPLETION REPORT EMPRESA ELECTRICA DE BOGOTA (EEEB) MESITAS HYDRO POWER PROJECT I- INTRODUCTION 1.1 Bank Lending to the Sector 1.1.1 Since 1950, the Bank has made 30 loans to the Colombian Power Sector totalling US$1.9 billion (this includes a loan to Empress de Energia Electrica de Bogota, EEEB, for US$171 millions approved by the Bank in November 1985 but not yet signed). 1.1.2 At the start of Bank lending, in the 50's, the Bank's lending to the sector was to individual utilities (CHIDRAL, CHEC, LEBRIJA, EPM and EEEB). The objective was to expand the coverage of the electric service at minimum cost in the area of each individual utility. However, the most economical and large hydro potential sites were beyond the needs and investment capacity of isolated systems; hence, interconnection and pooled investments became a major objective in the second half of the 1960's. To this end, the Bank strongly supported the creation of ISA (Interconexion Electrica S.A.). ISA was created with the purpose of interconnecting the regional power systems, defining the expansion program for the Interconnected System and developing large hydro sites. In the early 1980's, the Bank supported the creation of the Financiera Electrica Nacional, FEN, to assist the sector in tapping the domestic and external capital markets to fund its heavy financing needs. At present, the Bank is focussing its efforts on improving the efficiency of the weaker utilities in the power sector, strengthening sector finances and providing reliable electric service at minimum cost for the country. To this end, the Bank is currently considering a Power Sector Loan. 1.2 Sector Organization 1.2.1 The power sector is highly decentralized, so there is no central authority in the sector to directly enforce Government policies, planning and tariff decisions. Electricity service is provided by many types of companies, including some very weak local utilities in the poorer regions. Formulation of policies, definition of expansion programs and tariff approval for the utilities are in several hands, rendering the decision-making process slow. - 36 - 1.2.2 Sector planning has Improved markedly since the creation of ISA (1967) which was established to provide a rational framework for sector expansion. It has done this by planning and interconnecting its shareholders' systems, and undertaking much larger and more economical projects than would have been feasible under previous arrangements. 1.2.3 The National Planning Department (DNP) is responsible for planning for the power sector, relying on the technical planning which is carried out by ISA, and has the responsibility for overall coordination. Formulation of particular policies and regulations for the energy and power sectors is the rtsponsibility of the Ministry of Mines and Energy. ISA is responsible for preparing the generation and transmission expansion program for the Interconnected System. This program is further analyzed by DNP and after approval of DNP and the Council of Ministers, it becomes the official national expansion plan for generation and transmission. Total sector investments are consolidated by DNP and JNT (National Tariff Board) sets maximum tariff levels and defines tariff structures. The weakest part of the process is that which is assigned to the Ministry of Mines and Energy. In fact, this Ministry is understaffed and has not defined clear policies for power sector development nor for coordinating planning for the optimal utilization of diverse type of available energies. The Bank is currently supporting this Ministry in developing a program to analyze options for energy policies and define strategies for the energy sector. 1.2.4 Public electricity services are provided by about 50 utilities: (a) municipally-owned companies, of which the largest are EBEB (in Bogota), EPM (in Medellin) and EMCALI (in Cali); (b) three national enterprises whose principal purpose is the distribution of power within particular regions: Corporacion Autonoma Regional del Cauca (CVC); Instituto Colombiana de Energia Electrica (ICEL), and the Corporacion Electrica de la Costa Atlantica (CORELCA). CVC, ICEL and CORELCA have a total of 48 subsidiaries, which mainly distribute power at the local level; and (c) ISA, whose shareholders are REEB, EPM, ICRL, CVC, ODRELCA and CREC. II. PROJECT PREPARATION AND APPRAISAL 2.1 Project origin 2.1.1 The Project had its origin in the least cost expansion program agreed with the Government; it was followed by discussions between REEB and Bank staff (May 1976) concerning the possibility of Bank lending to SEEB to partially finance a five-year slice (1977-1981) of its expansion program. The total cost of the program was estimated at US$417 million, of which, US$248 million would be foreign. EREB expected to finance about half of the US$248 million with supplier credits and cofinancing, and the remaining - 37 - US$124 million with funds from the international lending agencies. 2.2 Project Preparation 2.2.1 EEEB prepared the Project with the assistance of ISA and a local consulting firm, with minor help from the Bank in the final stage of project preparation. EEEB prepared and justified its five-year investment program on the basis of the Government's national least cost program for generation expansion, which showed Mesitas as the next project to be built. The local consulting firm prepared the feasibility study and final design of the project. The Bank engaged an individual consultant to review the geological aspects of the project, mainly tunnels and caverns, which were the main technical concern of the Bank. The Bank also assisted EEEB in preparing a financing plan and advised it on financial measures to improve its financial situation. 2.3 Project Objectives 2.3.1 The objectives of the project were: (a) to meet forecast power demand at least cost; (b) to reduce the risk of earthquake damage to an existing dam (Sesquile); and (c) to strengthen EEEB's financial performance and management. 2.4 Project Description 2.4.1 The Project constituted a major part of EEEB's 1978-1982 investment program. It comprised: (a) construction of the Mesitas hydro power development; (b) remedial work on the Sesquile dam; (c) a valuation study of EEEB's assets; and (d) engagement of a financial advisor and establishment of a training program for EEEB's key financial staff. EEEB's total investment program consisted of the project described above and the 1978-1982 distribution program. The total investment program amounted to US$557 million, of which US$261 million corresponded to the project. 2.4.2. The Mesitas hydro power development was the main component of the project. It comprised the following parts: (a) two power plants, set one downstream of the other, baving a total capacity of 600 MW, namely: - 38 - (i) the El Paraiso Power Plant, having an installed capacity of 270 MW, and (ii) the La Guaca Power Plant having an installed capacity of 330 MW; (b) an additional pumping station for the existing Muna head pond; and (c) 100 km of 230 kV double-circuit transmission lines. 2.5 Project Appraisal 2.5.1 The Bank appraised the Project in December 1977 and January 1978. The key issues raised in the appraisal were related to: (a) availability of financing for EZEB's needs in addition to the proposed Bank loan; (b) implementation of the schedule for electricity rate increases approved by JNT in 1977 through 1980 and maintenance of the end-1980 level in real terms thereafter; and (c) strengthening of financial management. Commitments on these matters were sought, and obtained, during negotiations. 2.6 Negotiations, Presentation and Effectiveness 2.6.1 Negotiations started on September 25, 1978. Representatives of the Bank, EEEB and the Government reached agreements on the following: (a) Government's commitment to take actions to enable EEEE to obtain the additional financing needed for the Project; (b) EEEB's commitment to increase electricity rates to obtain a rate of return of 12% per year and the Government's commitment to enable BEEB to do so, and EEEB's commitment to achieve a self-financing ratio of not less than 73%; (c) BEEB's commitment to contract two consultants to assist its financial staff, and to carry out a training program to strengthen management capacity. Board approval occurred on November 14, 1978. Loan and Guarantee Agreements were signed on April 9, 1979 and the Loan Agreement was effective on August 21, 1979. 2.6.2 Loan conditions were the following: loan amount was US$84 million; the term was 16 years including a four-year grace period; the interest rate was 7.35% (fixed) and the commitment fee was 3/4 of 1%. - 39 - 2.7 Major Covenants of the Loan and Guarantee Agreements 2.7.1 EEEB and the Government complied with all the major covenants, except the following conditions which were not fully met: (a) Section 5.05, requiring EEEB to generate fun4s from internal sources not less than 73% of its -apital expenditures during the period 1981 to 1984. This ratio was excessively high and difficult to achieve by any power utility. The Bank agreed to eliminate this covenant under Loan 1807-CO (Bogota Distribution), and to rely on a rate of return covenant to ensure sufficient internal cash generation. However, as the work in progress in an expanding company like EEEB is large (40% higher than the net fixed assets in service at the end of 1984), the rate of return is not a fully satisfactory measure of financial performance, and therefore it was agreed to establish again a self-financing ratio covenant of a level of 35% (which is in line with the level normally agreed in power loans) under Loans 2008/2634--CO (Guavio and Bogota Distribution II, respectively). EEEB's self-financing ratios for the period 1978-84 were generally below the covenanted level. The actual ratios were, respectively: 77%, 50%, 25%, 19%, 35%, 29% and 26%. (b) Section 5.06, requiring EEEB to generate an annual rate of return of 12% (this was changed to 14% under Loan 1807-CO)o EEEB's actual rates of return were short of the requirement: 8.5% in 1979, 8.4% in 1980, 10.4% in 1981 and 11.1% in 1982. However, after that, the rates of return were satisfactory: 12.8% in 1983, and 13.2% in 1984. The covenant requires that the company calculate the rate of return each quarter in respect of the twelve months period beginning with such quarter and the twelve months period immediately preceding it. Any shortfall or overrun would be carried forward and subtracted or added, to the operating income used for the forthcoming twelve-month period. This concept seems to have been helpful to REEE in planning its tariff adjustments. (c) Section 5.09, requiring EEEB to prepare a program satisfactory to the Bank to revise the structure of its tariffs correcting the imbalance among major categories of consumers. EERB was unable to prepare the program during the project implementation. Under Loan 2634-CO, which was approved in November 1985, EEEB agreed to take action to correct the structure of its tariffs as a condition of loan effectiveness. Initial actions to fulfill this commitment were taken in February 1986 (para. 5.2). (d) Section 5.10 requiring REEB to reduce its accounts receivable to 17% of its total sales in 1981. Actual percentages were: 25 in 1979, 29 in 1980, 34 in 1981, 37 in 1982, 33 in 1983 and 35 in 1984. The ratio is expected to drop to 25% in 1986 through a compensation of accounts with BAAB (para. 5.4.1). These accounts receivable arrears are mostly from public sector entities. - 40 - (e) Section 3.04 of the Guarantee Agreement requiring the Government to grant import licenses, to ensure the timely procurement of the goods and services required for the project. The Government on occasion delayed the approval of the import licenses, causing delays in the execution of the project. III. PROJECT IMPLEMENTATION AND COSTS 3.1 Implementation of the Project 3.1.1 Project implementation took much .more time than anticipated during the appraisal. Current expectation is end-1986 for the Project to be operating fully, which compares witb June 1982, the completion date estimated at appraisal. Annex 3 shows actual and forecast project components and completion dates for components. 3.1.2 Although Parts A, B, C and D of the Project (Mesitas hydro plants and transmission lines) were completed by end-1984, the Mesitas plants will not start operating before end-1986 due to extensive repairs to the main tunnel which were needed after harmful leaks appeared during initial testing. This new operation date means eight years of implementation instead of the three and a half years estimated at appraisal. 3.1.3 Difference between actual and estimated implementation time is explained by three types of causes, which are discussed in the following paragraphs: (a) underestimation of the implementation time at appraisal; (b) geological problems; (c) financial difficulties; (d) problems with contractors; and (e) management and institutional problems. 3.1.4 The construction schedule prepared for the appraisal was too optimistic. Although at the time of appraisal the contract for the project's main civil works had been awarded, and the construction schedule was based on the contractor's proposal, experience in similar projects would have suggested the use of a more conservative implementation estimate, to take into account difficulties and occurrence of contingencies inherent to this kind of project. 3.1.5 Unanticipated geological conditions were the principal cause of the delays. First, the construction of the main tunnel was interrupted because a soft rock spot, aggravated by a high flow of ground water, was hit. The tunnel was plugged at the weak front and a bypass had to be excavated. - 41 - Second, bad geological conditions were found at the end of the power tunnel making it necessary to extend the tunnel steel lining and to excavate a drainage gallery. Finally, when the Project was completed and hydraulic tests were made on the tunnel, abundant leaks occurred from the tunnel due to hydraulic fractures in the lining. These leaks were considered a potential for causing landslides near to the downstream villages. EEEB had to study alternative solutions to this problem and contract the required tunnel repairs. Repairs have not yet been completed and are now expected to conclude by end-1986. 3.1.6 EEEB's financial difficulties and to a lesser extent those of one of the contractors, were also partially responsible for the delay. EEBB's financial diffibulties arose because the Government delayed approving tariff increases during 1980-1981; also, EEEB was not able to obtain timely additional financing for the project due to the deterioration in country economic conditions and the onset of the Latin American debt crisis. Due to this, REEB delayed payments to its contractors, which caused delays in construction and contributed to the bankruptcy of the power house contractor. After this bankruptcy, EEEB had to initiate a bidding process again to contract the completion of the power house. 3.1.7 In our view, weaknesses were also evident in EEEB's difficulties to manage and coordinate the project. Board interference in company management, and delay in taking decisions were factors that affected project Implementation very negatively . In addition, in the early years of the project, EEEB had been unable to attract suitably qualified staff due to low salaries, an issue that had not been detected at appraisal. On the other hand, procurement was split into too many contracts for EEEB's coordination capacity (though in this the Bank may have been also partly at fault for not focussing on the procurement packaging early on). Part of the reasons for the project delay was also because the Government delayed the approval of import licenses, mainly for cables and steel lining. 3.1.8 Construction contractors as well as consultants were also partially responsible for the delay. The bankruptcy of the power house contractor was a contributing factor. The contractor, a local financially-weak company, was not a good choice. It was overextended financially from the start and was particularly vulnerable to delays in payments by EEEB; high interest rates on loans it obtained to keep the works going ahead further deteriorated its financial situation. The contractor slowed down its works and finally stopped them all together. Finally, the Project Engineers were also responsible for some delays, because their estimates of the penstock cost were too low and the penstock design had to be modified by increasing steel thickness. These two factors meant last minute additional financing needs and caused an increase in construction time. 3.1.9 Implementation of Part D of the Project (strengthening of the Sesquile Dam) was also substantially delayed. Works started three years after the loan effectiveness date (due to EEEB's financial difficulties) and were completed by mid-1985. - 42 - 3.1.10 The study on valuation of assets (Part F of the Project) was carried out under Loan 1807-CO, by agreement with the Bank. It was completed on November 1984 and EEEB has started to implement its recommendations. The study was sent to the Bank for comments at the end of 1984 and the Benk agreed with its recommendations. The main conclusion is that the value of fixed assets was understated by about 28% at the end of 1982, because the index used for asset revaluation understates the real construction costs of power projects in Colombia. This situation is now being followed-up under Loan 2401-CO (FEN), which requires the preparation of a suitable asset revaluation index. 3.1.11 Strengthening of EEEB's financial management was achieved through the engagement of consultants to advise them on specific matters i.e.: insurance, tariffs, and financing. 3.1.12 The project was carried out without any negative ecological or environmental impact. 3.2 Project Costs 3.2.1 The actual cost of the project was some 60% above the appraisal estimate, in current prices. The actual cost is US$414 million, including the costs for completing tunnel repairs, which compares with US$261 million estimated at appraisal. These figures do not include financing costs (actual financing costs are US$86 million). One of the causes for the cost increase has been the incidence of escalation, due to the delays and longer construction period. Expressed in 1977 prices, the actual project cost is US$312 million which compared with US$204 million estimated at the appraisal yields a 53% increase. The main causes for this cost increase are: (a) penstock costs and steel quantities were undervalued in the initial estimates; these items account for US$25.0 million cost increase; (b) tunnel construction costs increased by US$22.0 million due to lining reintorcement and repairs, construction of the bypass in the slipped area and premium EEEB paid to the contractor to accelerate the tunnel lining; and (c) engineering costs doubled because of the longer supervision period required for the project. It accounts for a US$19.0 million cost increase. Annex 5 compares the actual and the forecast cost by project component and also shows actual and forecast sources of financing for the Project. 3.3 Procurement; Consultants' and Contractors' Performance 3.3.1 Procurement for the Project was carried out under the Bank's guidelines without major problems. Procurement for equipment and supply of materials and for contracting the main civil works was through ICB. Procurement for minor civil works financed by EEEB's own funds was through - 43 - LCB. Local consultants for project engineering and supervision were contracted following procedures agreed on by the Bank. 3.3.2 With the exception of the powerhouse Contractor, the performance of EEEB's constructors was considered satisfactory by EEEB and by the Bank. The bad choice of the powerhouse Contractor showed the inconvenience of dividing the project execution in small contracts. A separate contract for the powerhouse civil works was not attractive enough for a sufficient number of large and experienced contractors. 3.3.3. The performance of EEEB's consultants was considered satisfactory by EEEB. However, in general, the difficult geological conditions prevailing in the country and the geological problems faced in some of the projects designed by local Consultants (Chivor, Chingaza, Mesitas), suggest that in future projects that the Bank considers for financing, it would be advisable to require the Borrower to contract a Board of Consultants to review the geological studies, design, execution timetable and cost estimate and follow-up the project execution. 3.3.4 EEEB also engaged consultants to prepare the study on asset valuation (para. 3.1.10) and for strengthening EEEB's financial management (para. 3.1.11). The work performed by the consultants was considered satisfactory by EEEB and the Bank. IV. PROJECT JUSTIFICATION 4.1 Justification of the Project 4.1.1 The appraisal report justified the undertaking of the Project because it would meet the generation needs of the interconnected system in early 1982 at the least cost among the available alternatives at the time of the appraisal. At that time, it was estimated that the project would have an Internal Rate of Return (IRR) of 10.4% on the basis of: planned tariff increases and using shadow prices for energy sales, operation costs and domestic components on civil works, machinery and equipment. 4.1.2 Based on actual data and updated forecasts, the Project will: (i) meet the needs of the interconnected system by the end of 1986 (because demand growth was,lower than expected, as was overall growth of the economy, no other generating project was needed to compensate Mesitas' four-year delay); and (ii) have Rate of Return of 10.0% on comparable basis using: actual project cost and actual EEEB's system operating costs, actual tariffs and planned tariff increase, planned energy loss reductions, and the current estimate of 1.0 for shadow prices in Colombia for energy sales, operation costs and domestic components on civil works, machinery and equipment. At today's tariffs, without taking into account the planned tariff increase, the Rate of Return would be 9.3%. - 44 - 4.2 Demand 4.2.1 Due to the recession in the country, and the effect of the rationing of electricity derived from a severe drought (1980-1981), electricity demand growth was much lower than forecasted at appraisal. Actual demand growth for the national interconnected system in the period 1977-1983 was 8.0% per year in comparison with 10.8% forecasted. 4.2.2 Demand for electricity in EEEB's system also grew significantly less than the appraisal forecast, as shown in Annexes 6 and 7. The forecast sales growth was 11.4% per year, but actual sales growth was only 5.3% per year. This lower growth was due to the effect of the above-mentioned drought and to the impact of the country's recession particularly in the industrial and commercial sector and of tariff increases. Due to this *slowdown in overall load growth throughout the country, the delay in operation of Mesitas did not affect power service. 4.3 Losses 4.3.1 Although the control of energy losses was not a specific objective of the project, such losses affect directly the economy of the project and the overall performance of the company. Energy losses were much higher than forecasted in the appraisal. As shown in Annex 6, energy losses increased from 14.5% in 1977 to 23.5% in 1983, while the appraisal was based on losses maintaining their 1977 level. This increase in losses has been due mainly to increased theft of electricity and also to insufficient investments, because of financial constraints, for rehabilitation of deteriorated and/or overloaded circuits. These two factors are being addressed under the Bogota Distribution II Project (2634-CO), approved by the Board on November 1985. EEEB has already set up a Unit to program and coordinate efforts to drastically reduce losses. The goal is to limit losses to 13.2% of gross generation by 1990. 4.4 Operating Costs 4.4.1 Unit operating costs doubled from 1977 to 1984 (USel.21 to USc2.40 per kWh). This increase was due mainly to increased energy purchases from ISA and also to increases in administration costs. The reason for larger energy purchases from ISA was, again, drought in EEEB's system. The increase in administration cost was due to higher than expected staff number and salaries. While salary increases are considered justified, staff number increase has deteriorated performance indicators (see Annex 12) as shown belov: No of Customers per Employee Mesitas Guavio Year SAR SAR Actual 1979 234 - 216 1980 244 - 211 1981 248 224 207 1982 259 231 208 1983 302 238 209 1984 281 245 202 - 45 - Under Loan 2634-CO, EEEB has undertaken to strengthen its management and operating performance and will seek to reduce its operating costs through a training program for improvement of managerial skills and a loss reduction program. 4.5 Rate of Return 4.5.1 On the basis of actual data, current and planned tariff increases and planned loss reductions, the rate of return of the project is very close to the appraisal estimate: 10.0%, against 10,4% estimated at appraisal (See Annex 11). If tariffs remain at present levels (in constant prices) and other assumptions remain the same, the rate of return would be 9.3%. A critical factor affecting EEEB's revenues is the level of energy losses, about 24.5% of the gross generation in 1985. If these losses remain the same instead of dropping to the planned figure of 13.2% in 1990, the rate of return would be only 8.8%. V. FINANCIAL PERFORMANCE 5.1 Comparison of Actual and Forecast Data 5.1.1 Annexes 7 to 9 compare the actual and appraisal forecast income statements, sources and applications of funds, balance sheets and performance indicators for the period 1978-84. EEEB's financial performance during that period fell short of the expectations held at appraisal. The company had been committed to a large expansion program to meet the system growth, however, there was a slow down in the economy and sales were subs -ntially below appraisal expectations. Despite these unfavorable factors, key financial ratios were maintained at acceptable levels (see Annex 12). The rate of return was lower than the covenanted 1/ but increased steadily, from 6.5% in 1978 to 13.2% in 1984. The self-fina7cing ratios for the period 1978-84 were, respectively: 77%, 50%, 25%, 19%, 35%, 29% and 26%, which were lower than the 73% covenanted in Loan 1628-CO except in 1978 (this ratio was later reduced to 35% under Loans 2008/2634-CO). Debt-service coverage ratios were generally above the covenanted level of 1.50 Although the debt-equity ratio, deteriorated from 34/66 in 1978 to 47/53 in 1984, EEEB's capital structure is still satisfactory. 5.1.2 Electricity rates increased substantially during the period (see para. 5.2.1). However, EEEB's internal cash generation was lower than expected and contributed to EEEB's difficulties in paying to ISA and to some contractors. The reasons were: a reduced sales growth (11.4% forecasted in SAR compared to an actual rate of 5.3%), higher than forecasted energy losses (para. 4.3.1) and increases in operating costs (para. 4.4.1). Also, there was a substantial increase in the investment program (construction of Guavio hydroelectric project). Because of the reduced sales growth and the increases in operating costs, EEEB was unable to comply with the revenue covenants (Sections 5.05 and 5.06 of the Loan Agreement). The Bank took an understanding view on these developments, recognizing that the 1/ The rate of return was increased from 12% to 14% for 1981 and thereafter under Loan 1807-CO. - 46 - situation of the country changed significantly compared to the one envisaged at appraisal and that EEEB and the Government had made extraordinary efforts to overcome these difficulties, particularly in the area of tariffs. 5.2 Tariffs 5.2.1 Annual average tariffs were somewhat lower than the expected level for the period 1978-1980, but improved steadily in the period 1981-84 reaching Col$6.28/kWh in 1984, at which time they were one of the highest in Colombia and exceeded appraisal estimates by 173%. In mid-1978 prices, the average tariff increased at an annual average rate of 14% in the period 1978-84. The present average tariff is close to the national average medium term marginal cost; however, the tariff structure shows large cross-subsidies among various major consumer groups, as reflected in the following table: Average Tariff during the First Semester of 1985 Col$/kWh Residential 3.85 2/ Industrial 10.86 Commercial 15.59 5.2.2 The average residential tariffs are too low while the industrial and commercial tariffs are high. This is mainly due to the low tariff to consumers with small load. To address this problem, EEEB agreed to prepare a program of rate structure revisions and to put the program into effect by April 1980. Bank missions frequently raised this concern during supervision, but due to its financial difficulties, EEEB focussed its attention on the overall increases in tariffs. The Bank agreed to this approach (see para. 5.2.3) as it reflected the priorities at the time. EEEB's freedom of action was also limited in that, following consumers protests in 1981, the Government did not allow the company to increase the electricity rates to the lower income sector of the population. 5.2.3 During project implementation, the Bank agreed that, in view of the financial issues facing EEEB, the goal of increasing the overall tariff level should take precedence over tariff structure revisions. In October 1984, the JNT issued Decree No. 2545 establishirg a new tariff structure, which the Bank considered adequate. EEEB agreed, in connection with Loan 2634-CO, to revise the structure of its tariffs within the framework of that decree. In February 1986, EEEB made some changes in its tariff policies, decreasing the monthly rate of increase to non-residential customers from 2.75% to 2.2%, and is planning to increase in the near future the monthly rate of increase of residential customers from 1.5% to 2.2%. Though admittedly modest, this would be the first step to arrest the increasing cross-subsidization that was being caused by the policy of tariff increases that was being followed. Much, however, still remains to be done in this area. EEEB has also made changes to the structure of the residential tariffs, which will facilitate simplifying the present twenty four blocks tariff to six blocks in the future. 2/ The tariffs for residential consumers located in the upper blocks are similar to the industrial and commercial tariffs whereas the tariff of consumers located in the lower blocks (below 200 kWh/month) are extremely low. At the end of 1985, consumers with consumption up to 200 kWh/month paid a Col$1.76/kWh; while those with consumption up to 2000 kWh/month, paid Col$13.81/kWh. - 47 - 5.3 Investment Program 5.3.1 While the Mesitas hydroelectric project, which is the subject of this report, was being constructed in 1980, EEEB decided, with the approval of the Government and the support of the Bank, to undertake Guavio, a large (1000 MW) hydroelectric project to meet the estimated demand requirement. At the time of the appraisal of Mesitas, Guavio was not in the system's least cost expansion program, but became part of the program in updates made in 1980. Soon after the Guavio loan was made, in 1981, Colombia entered into a receseion, the Latin American debt crisis started, access to external financing became difficult and a heavy strain was placed on the company finances. To meet the additional financing requirements, EEEB obtained additional financing from the Bank (through FEN) and from commercial banks. Additional funds are still needed to complete Guavio, but obtaining such financing is not expected to be a major problem, because of Colombia's improved external finances. IDB has indicated that it is willing to consider a supplementary loan and the Bank is considering a sector loan which could, inter alia, serve to meet EEEB's funding needs for Guavio, by funding interest during construction on Loan 2008-CO until the project is commissa.oned. 5.4 Accounts Receivable 5.4.1 As a percentage of sales, accounts receivable increased from 21% in 1978 to 35% in 1984. This high ratio is influenced by the arrears from the Empresa de Acueducto y Alcantarillado de Bogota (EAAB). These arrears are expected to be settled in the near future by offsetting their amount against arrears that EEEB has with EAAB for its capital contribution to the Chingaza Project (a water supply scheme that would increase the flow of water for generation at the Mesitas hydro project; EEEB has part ownership in Chingaza). After making allowance for this adjustment, accounts receivable would still represent 25% of annual sales in 1984, which is high. During proj,ect supervision, the Bank pressed EEEB to comply with the accounts receivable covenant (Section 5.10 of the Loan Agreement). The company responded by establishing a special unit (Seccion de Cobranzas), by making special arrangements with customers with overdue accounts to reduce their indebtedness and by intensifying its cut-off program for delinquent cohsumers. It is expected that EEEB will reduce the level of overdue receivables in the near future and the company has agreed, in connection wi:h Loan 2634-CO, to reduce the percentage of accounts receivable to 17% by Lhe end of 1986. VI. INSTITUTIONAL PERFORMANCE 6.1 The Borrower 6.1.1 During the period 1977-1985, EEEB made valuable progress in financial management, system operation, project management and planning. One important step for strengthening its financial management was to hire - 48 - management consultants to carry out a study of the company's managerial, budgetary and accounting systems and financial planning. The study was completed and the recommendations were gradually implemented during 1983 and 1984. To improve its system operation, EEEB completed the design of its control Center. At present, it is being implemented under Loan 2008-CO. To improve the execution of its hydro projects, REEB created two project units (Mesitas and Guavio) supported by local consultants. To improve its planning, the company contracted Italian coneultants to develop planning tools and train its staff. Finally, to expand its distribution system, EEEB carried out the Bogota Distribution Project, financed under Bank Loan 1807-CO. 6.1.2 Despite all these efforts, EEEB experienced a deterioration of the distribution system, evidenced in energy losses, which increased from 14.5% in 1977 to 24.5% in 1985. This may reflect management weaknessess which appear to have now been corrected. However, the deterioration in country economic conditions probably accounts - at least partially - for the increase in theft, so any assessment of management's performance must be placed in the appropriate context. It may be noted too that the execution of two large projects, Mesitas and Guavio, absorbed most of EEEB's financial and human capabilities, to the detriment of the other activities. In this regard, it would seem that the Bank underestimated the burden posed on EEEB's management by these two projects. 6.2 The Government 6.2.1 In general, the Government supported EEEB well in carrying out the project. Although during 1980-81 the Government delayed its authorization for timely tariff increases, after that it allowed EEEB to increase its tariffs to the levels requested by the Company, and pressed it to adapt its tariff structure in accordance with the Bank's recommendations. The Government's support in facilitating EEEB's access to external financing was particularly important when country conditions deteriorated and BEEB's finances became strained. In this regard, the Government's support for the creation of FEN is particularly noteworthy. Government's delays in approving import licenses for cables and steel lining caused delays in the execution of the project. 6.3 The Bank 6.3.1 The Bank made important contributions to the Project and EEEB's development programs in general. Its financial covenants and its constant pressure on EEEB and on the Government were decisive factors for the tariff increases that EEEB applied in 1982 and thereafter. 6.3.2 The Bank's technical support was also important. Although field supervision appears to be low (only 11 staff-weeks were spent on field supervision; this is less than 2 staff-weeks per year), supervision work in the Bank's headquarters was intensive, including frequent discussions with the project consulting firm. However, given the complexity of the project, a - 49 - v 're extensive field supervision may have improved project management and h Iped EEEB to find solutions to some of the technical problems faced during project execution. 6.3.3 The Bank's financial support to the project and to HEEB in general was important. In 1984, the Bank approved a loan of US$170 million, through FEN (Loan 2401-CO), of which US$47.9 million went to EEEB to provide additional financing for the project. During the execution of the Project, the Bank approved three additional loans to BEEB to finance its expansion program: US$84.0 million for Bogota Distribution Project (Loan 1807-GO); US$359 million for Guavio Project (Loan 2008-CO); and US$171 million for Bogota Distribution II Project (Loan 2634-CO). VII. LESSONS TO BE LEARNED 7.1 The Project 7.1.1 On balance and based on the project's objectives, the Mesitas hydro power project should be judged as successful. Although geological conditions which, it would seem, were unforeseable, are substantially delaying project operation, they are being appropriately addressed and completion by end-1986 seems very probable. The actual project cost, although some 60% higher than estimated at appraisal, still represents a very low cost per kW (690 current US$ per kW). 7.2 Adequacy of the Studies 7.2.1 The geological condition that 'affected project execution and similar difficulties that other projects faced in Colombia (Chivor, Chingaza), suggest that a more extensive geological prospecting using more drillings, geoseismic and geoelectric methods and other modern technologies would always be well worth and should be requested for future hydroelectric projects. A review of the studies and the follow-up of the project execution by a Board of Consultants would also be advisable. 7.3 Management 7.3.1 The decision to go ahead with Guavio placed a major strain on EEEB's management, as EZEB concentrated most of its financing and human resources in the execution of Mesitas and Guavio. This suggests that when large projects are to be carried out, it is advisable to pay special attention to the capability of management and of the organization to efficiently handle all its operations, in addition to the project, looking into alternative project execution arrangements, if it is concluded that there is a risk of burdening management beyond its capabilities. It may be noted also that changes in top management that took place in 1982 weakened the company's management, thereby compounding its problems. The situation seems to have improved considerably now. - 50 - 7.4 Procurement 7.4.1 Procurement was split into too many contracts in comparison with EEEB's capabilities, which delayed project implementation. In addition to this, a small civil work contract resulting from this procurement arrangement (powerhouse contract) was one of the reasons for the project delay (see paras. 3.1.7 and 3.3.2). In connection with future projects, the Bank should discuss procurement programming with the borrower during project preparation taking into account their constraints and seek an agreement on packaging at the time of negotiations. 7.5 Supervision 7.5.1 During the period of project execution, there were frequent Bank missions to EEEB, to prepare and appraise Loans 1807-00, 2008-00, and 2634-CO. Through these missions, the Bank followed up closely on developments in EEEB's finances and on institutional issues. On the other hand, the Bank's field technical supervision was not frequent. Because of this, it would seem that the Bank sometimes became aware of problems which arose during project construction when it was too late to make a valuable contribution to solving them. This suggests that when faced with projects that are likely to present project-spec:fic issues, it is not advisable to relax technical supervision, under the assumption that the frequency of general contacts provides for adequate supervision. -51- gAM= I Peae 1 o 2 EMPRESA D OEERO ELICIRICA DE 8000TA (LEB) sITAS HYDRO POWER PROJECT (128-C0) Suery of Bank Londing to the Sector Borrowers and Year of Loan Account Loan No. Agreement Project Descripton (MUS8) Government 1588-Ce 1978 500-kV Intorconnection Central System/Atlantic Syetem) s.00 60.00 FEN 2401-CO 1984 Power Development Finance Project 170.0 Power Development Finance Project (Cofinancing) 80.00 200.00 ISA 875-CO 198 Central System Interconnection (280 kV transmission lines and substation) 16.00 601-CO 1970 Chivor I project (4 x 125 MW hydro) 62.30 1682-CO 1978 San Carlos I (4 x 1656 MW hydro) 128.00 1725-CO 1979 San Carlos I and II (4 x 158 MW hydro) 72.06 268.86 EEEB 246-CO 1960 Laguneta unit 4 (1x18.0 MW hydro) 17.60 Salt II units 1 and 2 (2x88.0 MW hydro) Zipaquira unit 1 (1 x 88.0 MW thermal) 818-CO 1962 Zipaquira unit 2 (1 x 87.5 MW thermal) 50.0 El Cologio units 1, 2, and 8 (8 x 50.0 MW hydro) 587-CO 1988 El Cologlo unite 4, 5, and 6 (8 x 50.0 MW hydro) 18.00 Canoes Project (1 x 50.0 MW hydro) 1828-CO 1978 Mositas Hydro (El Paralso 3 x 90 MW; La Ouace 3 x 100 MW; pumping 8 x 10 MHP; Sesquile dam strengthening) 84.00 1807-CO 1980 Bogota Distribution 87.00 208-CO 1982 Guavio Hydro (5 x 200 MW) 359.00 2884-CO 196 Bogota Distribution II 171.00 786.6 EPM 225-CO 1959 Tronoras unit 1 (1 x 18.0 MW hydro) 12.00 Guadalupe III units 1 and 2 (2 x 46.0 MW hydro) 282-CO 1961 Troneras unit 2 (1 x 18.0 MW hydro) 22.00 Guadalupe III unite 8, 4, and 5 (3 x 46.0 MW hydro) -52 - A~ i plREA P ENEROKA EL.ECTRICA 0E 8000TA (EEB uESITA HYOR POVER PRJECT (162-C0i Suary of Bank Lendism to lh StIm or sorromre and Year of Lan Account Lten No. Aa e Pr.J Dscr al n mU) 169-CO 1964 Quaøpø I unIte 1 and 2 (2 x M.@ MW hydro) g/ 43.m0 874-CO 1978 Gustpa II unI6e 1, 2, 8, and 4 (4 x 76 W hydro) 86."0 168-CO 1986 uadalup* IV Hydro (8 x 71 M) 126.N0 1958-CO 1981 Playo Hyd-o (8 x 07 M) 86M 2449-CO 1984 Rio Grande Hydro (8 x 10 M) and water supply 164.66 69.6 CVC/CH^IRAL 88-CO 196 Anchicaya unIte i and 2 (2 X 12.6 MW hydro) 8.68 118-CO 1955 Anchicaya units 8 (1 x 26.6 MW hydro) 4. 6 Yumbo unit 1 (1 x 10.0 W herml) 218-CO 1958 Yumbo unit 2 (1 x 10.0W thermal) 2.88 265-CO 196 Yumbo unit 8 (1 x 38.6 MW thrmal) 26.N0 Callm unit* 1 and 2 (2 x 80.6 MW hydro) 889-Co 1968 Cellma unIts 8 and 4 (2 x 30.6 MW hydro) 8.8e 44.88 CHIEC 39-C0 196 La Insula unIs 1 and 2 (2 x 16.6 MW hydro) 2.6M 217-CO 199 La E~eralda un1te I and 2 (2 x 18.8 MW hydro) 4.60 7.26 LEBRI.A 64-CØ 1951 Palmas unit* 1 and 2 (2 x 4.4 MW hydro) 24_ 2.40 ELECTRIBOL 847-CO 1968 Cosplque unit* 2 and 8 (2 x 12.6 MW thermai) 5JO 5.N6 CORELCA 1990-CO 1982 Atlantic Coast VIIlag* Eloetrtfleation 38.0 88.6N Total Loen Amounts 1,9=9.88 pf Subsequently Incresød to 4 x 76 MW ANNEX 2 Page 1 of 2 COLOMIBIA EMPRESA DR ENERGIA ELECTRICA DE BOGOTA (BEEB) MESITAS HYDRO POWER PROJECT (1628-CO) MaLjor Covenants of the Loan and Guarantee Agreements 1. The Loan Agreement for Mesitas Hydropower Project required EEEB to: (1) carry out the project with due diligence and efficiency and in conformity with appropriate administrative, financial, engineering and public utility practices (Section 3.01); (ii) employ consultants whose qualifications, experience and terms and conditions of employment were satisfactory to the Bank (Section 3.02(a)); (iii) furnish to the Bank for comments the results of the study on valuation of the Borrower's fixed assets (Section 3.02(b)); (iv) take all necessary measures to ensure that the project were carried out with due regard to ecological and environmental factors (Section 3.03); (v) furnish to the Bank the plans, specifications, reports, contract documents, and construction work and procurement schedule, for the project (Section 3.06(a)); (vi) maintain records and procedures adequate to record and monitor the progress of the project (Section 3.06(b)); (vii) manage its affairs, maintain its financial position, plan its future expansion and carry on its operations in accordance with appropriate business, financial and public utility practices and under the supervision of experienced and competent management (Section 4.01); (viii) furnish to the Bank an audited financial statement four months after the end of the preceding year (Section 5.02); (ix) take all the steps required to ensure that the cumulative funds from internal sources at the end of each year during the period 1981 through 1984 were not less than 73% of its cumulative investments (Section 5.05); (x) establish and maintain tariffs to generate a return in the year 1978 and each year thereafter at a rate of at least 12% (Section 5.06); - 54 - ANNEX 2 Page 2 of 2 (xi) put in effect a program by April 30, 1980 satisfactory to the Bank to revise the structure of its tariffs (Section 5.09); and (xii) reduce its accounts receivable to 20% of its total sales in 1978, 19% in 1979; 18% in 1980; and 17% in 1981 and thereafter (Section 5.10). 2. The Guarantee Agreement required the Government to: (i) make Dr cause to be made arrangements to enable EEEB to be provided promptly with funds as were needed to meet the expenditures required for the carrying out of the project (Section 2.02); (ii) take all action necessary or advisable on its part to enable EEEB to obtain loans, in addition to the loan (Section 3.02);and (iii) take all action, including the granting of all necessary authorizations, import licenses, foreign exchange permits and all other approvals required, to ensure the timely procurement of the goods and services required for the project (Section 3.04). 3. Compliance on the part of EEEB and the Government with the above covenants is discussed in detail in Chapters 2 and 3 of the report. - 55 - ANNEX 8 EWPRESA DE WERIA ELECICOA DE 0 affiEE MESITAS YO POVER PROJECT (1428-C) ProUGff Componente and Foream and Actual Coml"eon Datae quantilty Completion Dstos -AppraitalI-- - -------Approag ----- Unit E9tite Actusl E9timatos Actual Commentary A. Pumpini Station for Muno Head Pond m3/5 52.6 62.5 7/81 9/84 8. El Paraiso Power Plant - Tunnel km 12.8 12.6 4/79 7/84 Repaire needed in the tunnel would bo comploted by 9/86 - Penstock km 4.0 3.9 12/Sl 3/84 - Electrom~chanical equipsent Mw 279.0 278 2/81 11/83 - Expørimentol Operation 6/82 12/8 Current estimate for ful] operation le 10/88 C. Le Guace Power Plant - Tunnel km 1.3 1.3 4/79 1/84 - Penstock km 5.3 6.3 12/81 3/84 - Electromchanical Equipment MW 330 324 12/81 11/83 - Equipment Operation 8/82 12/88 Current estimate for full operation is 10/88 0. Transmiesion Lines - 230 kV El Paraiso- La Guaca line km e 8 10/Sl 10/88 - 280 kV La Guaca- ISA system line km 8 j 10/Sl 10/83 - 230 kV Le Guaca- Torce line km 86 86 10/Sl 10/88 E. Sesqulle Dam Roinforcoment 1/82 $/85 F. Asset Valuation Study 3/80 11/84 -56 - ANNEX 4 COLOMBIA EMPRESA DE ENERGIA ELECTRICA DE BOGOTA (EEE) MESITAS HYDRO POWER PROJECT (1628-CO) Schedule of Disbursements Ending Actual as % Fiscal Year Month and Actual Appraisal of Appraisal and Semester Calendar Year Total Estimate Estimate 1979-2 Jun 1980 - 25.0 - 1980-1 Dec 1980 10.5 39.9 26.3 1980-2 Jun 1981 16.4 56.5 29.0 1981-1 Dec 1981 23.6 73.3 32.2 1981-2 Jun 1982 38.9 75.7 51.4 1982-1 Dec 1982 51.5 78.0 66.0 1982-2 Jun 1983 58.5 84.0 69.6 1983-1 Dec 1983 64.6 76.9 1983-2 Jun 1984 72.8 86.7 1984-1 Dec 1984 78.8 93.8 1984-2 Jun 1985 83.3 98.8 1985-1 Dec 1985 83.9 99.9 1985-2 Jun 1986 84.0 100.0 Closing Date 12/31/84 12/31/82 5/21/869 - 57- Altm 5 ~IPSA DE E~IA LECIRICA DE MGrA (EEB) =ESrrS RD1 PCER PIMCr (1628-<) Pject Gost 2n=mry and Souroe of FinancIng (In øillin of US$) Actual Forecast Dfference fran Forecast local Foreiga Total local Forei Total 1ercent ost cost ut cst Cost Cost (%) ~estas Civil Works (incl.land) 92.4 53.9 146.3 49.4 32.1 81.5 64.8 79.5 E pn 23.1 85.5 108.6 11.6 80.5 92.1 16.5 17.9 Trransiuid~ 2.5 7.0 9.5 0.3 5.0 5.3 4.2 79.2 Egneerig 33.3 1.6 34.9 16.5 1.0 17.5 17.4 99.4 Subotal 151.3 148.0 299.3 77.8 118.6 196.4 102.9 68.0 SesquileStenhein Civil Works 3.5 - 3.5 1.7 2.6 4.3 (0.8) (18.6) EqupDent 0.1 4.9 5.0 0.9 1.4 2.3 2.7 117.4 0.8 0.8 0.8 - 0.8 - - Subotal 4.4 4.9 9.3 3.4 4.0 7.4 1.9 15.9 Finnial 0nsta ad- - - - 0.5 0.5 (0.5) (100.0) Total Bese Cost 155.7 152.9 308.6 81.2 123.1 204.3 152.9 74.8 55.0 50.3 105.3 31.2 25.3 56.5 48.8 86.4 Total Project Cest 210.7 203.2 413.9 112.4 148.4 260.8 153.1 58.7 Interest during Constminnm (IDC) _ 85.6 85.6 15.0 15.0 70.6 470.7 Total Project Cost after IDC 210.7 288.8 499.5 112.4 163.4 275.8 223.7 81.1 .OU 00 C F FDMANCG Bank (1628-00) - 84.0 84.0 84.0 84.0 - - Bank-Ff (2401-D) - 47.9 47.9 - - - 47.9 57.0 Supplers and umrLal Banks - 88.9 88.9 - 53.8 53.8 35.1 65.2 oal inn (m,FE) - 19.7 19.7 - 10.6 10.6 9.1 85.8 B'S on reoure 210.7 48.3 259.0 112.4 15.0 127.4 131.6 103.3 Total 210.7 58.8 499.5 112.4 163.4 275.8 223.7 81.1 - 58 - AIE 6 GHLR8IA DER!S lE UahIA RfEIO& lE 110W (iE) EMO M! PUE (162&0) Elqt~ety Supl and 2ECR‡gmo 1977 1978 1979 1980 1981 1982 1983 sal Acua SAR 1~Um Sm Mtuel SAR C!b=I SäR ~ SAR AM-w1 Mctua Residen~a 1,237 1,416 1,398 1,603 1,550 1,816 1,661 2,ýååö 1,594 2,339 1,942- 1,98 Comer 589 616 622 676 617 742 638 815 569 895 595 576 1ndstia 995 1,138 1,052 1,248 1,135 1,369 1,184 1,502 1,135 1,648 1,130 1,120 oen en 353 408 340 442 363 48D 419 520 470 564 439 443 Satotal 3 3,578 3,412 3,9 3;WW5 Z,507 3,902 4,897 3,768 5446 4,1%6 4,127 St~ame Use 77 75 75 75 77 75 74 90 65 105 79 71 Losses 6 tna= e for 603 687 666 756 _89 818 985 883 863 949 1121 Mff8 1bmal Consupimn 3,854 4,340 4,153 4,800 4,634 5,300 4,961 5,870 4,696 6,500 5,306 5,485 EMB 9do Plant 2,843 2,028 2,22 2,608 2,877 3,107 2,764 3,492 2,767 4,468 2,896 3,337 E flhral Plait 737 800 710 800 655 800 639 800 510 800 672 597 1å 274 1,512 1,241 1,392 1,101 1,393 1,558 1,578 1,418 1,232 1,738 1,550 14xmu DEnd 805 840 819 929 891 1,025 917 1,136 875 1,258 983 1,006 od letor 59.0 59.0 58.0 59.0 59.0 59.0 62.0 59.0 61.0 59.0 61.0 62.0 LUses & Tbeft 14.5 15.8 16.0 15.8 19.2 15.4 19.9 15.0 18.4 14.6 21.1 23.5 - 59 - ANNEX 7 COLOMBIA EMPRESA DE ENERGIA ELECTRICA. DE BOGOTA (EEEB) MESITAS HYDRO POWER PROJECT Actual and Forecast Sales Growth Rates (%) and Percent (%) of Sales by Customers Category, 1977-1982 1977-1982 Percent of Percent of Growth Rates % Sales in 1977 Sales in 1982 Actual Forecast Actual Actual Forecast Residential 9.4 13.6 39.0 47.3 43.0 Commercial 0.2 8.7 18.6 14.5 16.4 Industrial 2.6 10.6 31.3 27.5 30.3 Government 4.5 9.8 11.1 10.8 10.3 Total 5.3 11.4 100.0 100.0 100.0 4/17/86s - 60 - 3 i* rn4ft..#..*-wa I an ila PIffilf I i g dill fl§ i R gaIgeIi Tf 4 1 å= iI å R Roll u11g g 5 g g Z: i ja g iai ggi g 1 agsi - ni<43p-: :. : r..;fni a g a gs seeassi g ag a ii - 61 - lj i a i 1 fill a 19II MI F ie 0,1115 1 silli 11 9i s eill *~ is M M m i ^ o. ~1 4 l fl~~e fl fl0 el l Nill g Mill EFISi 2 2 til ~ gg° l " & ° " m g *. fi l - ;FLc fallì OM§l å R1|9IPIP -1" Iflål Tl mit ii iii a 191 aliffil P ä 1 Railt m- i i i 1m Ee lpflå 1 111 T MI§ f81x11 11 °i "I-njol g "%Iii Fäl l 5 smilig 1 11g1 39° 951 9 M915 sai-llul E sasi el¡gl°ali l a l2 1 lil0 gill T jfe lRe iojg I E ¡"91|1 m°' g ¡ 2:1 °-z RER IA nle| lå 1 ~i ä*al | t ~» all MI-ri- -ff * iuj isfi i '--9 10,4 CaG.0BIA UESITAMSNYOELECTRIC POER PRO3EC ERPRESA DE EMERIA ELECTRICA DE OMUTA Coapartion of Actual and forecast Batao hefts (1977-19840 (ellifon 04 crrent Cl) ---1978- ---1979--- --1980- ---195-- -1982-- ---1933-- -984-- Varia- Varia- Varig- varla- Varla- arlä- Variä- SAR Actual tIon t2 5AR kcual ti 12> MA$ ktal tios 25 SAR Actual tio ian S(R Stoal tin on) SM O Actual tio IZI M Attad i. (Nt ASSET3 gross FiRed Assets in Service 23684 20003 -15.5 29224 2386 -9.7 36056 35690 -1.6 44873 45650 1.7 71565 89872 25.6 8 0 187008 3.0 9 1317 43.4 Actuautated dsoreciation 7707 7047 -8.6 9763 0066 3.1 12053 13995 16.1 1i8 19335 31.6 17960 29979 67.9 21 4 38759 77.3 287.0 ffet Fixed Afsets U Srio 15977 129 -li .9 19461 16320 -16.8 24003 21495 -10.4 3085 2631-0511.5 530 5993 11.5 59292 69329 -5.3 6529 8141 5.4 ir in progrss 2509 1892 -24.6 9351 4687 -49.9 15734 18331 -28.0 16953 24956 47.5 849 43000 4964.8 940 71625 7519.7 1013 146M39 13216.8 Total Fitad Aslets 18496 14853 -19.7 28812 21007 -27.1 39737 32826 -17.4 47038 51171 8.8 54554 802993 88.6 60222 139954 532.4 66300 89490 M%.3 Investantt Ifvestent in 894 2333 2035 -12.8 3725 2953 -20.7 5919 4293 -27.5 9390 6355 -32.3 85254 732 -48.7 22311 %O -7. 322 i1270 -.4 Other lovtestents 280 418 49.3 290 426 52.1 290 451 61.1 20 461 64.6 290 4380 1464.3 28 524 1730.1 380 580 1971.4 total lnvstom ts 2613 2453 -6.1 400 3379 -15.6 6899 4744 -23.5 %70 6816 -29.5 15534 12212 -21.4 22657 14721 -35.0 32908 11074 -4.1 CY Current Atsets Cash 282 341 20.9 323 284 -12.1 465 256 -0.5 625 352 -43.7 741 469 -36.7 911 59 -35.3 197 263 119.5 Accounts Receuvahle 626 1474 135.5 905 1629 00.0 1280 2602 103.3 1694 4323 155.2 2058 6198 201.2 2477 8311 35.5 3031 120 314.1 Ifventory 1074 1069 -0.5 1331 1224 -8.0 1648 1581 -4.1 193 3097 56.2 2529 6041 138.9 2910 7589 160.8 1358 5650 68.6 other 40 0 -100.0 41 81 97.6 42 206 30.5 43 41 932.6 44 576 1209.8 4 80702226.1 4 392 48.3 Total Current feets 2022 2894 42.6 2600 3218 23.8 3431 445 35.4 4345 8216 89.1 5372 13284 147.3 6344 57559 176.9 7621 23219 2M,4 ltt ASSETS 23121 20190 -12.7 35417 27604 -22.1 49367 42215 -54.5 61055 66203 8.4 75460 129399 70.1 89223 172241 93.0 106935 236713 121.6 EIlI 4ND LIAOILtif Capital Stock and Sarplmö 4079 3854 -5.5 6089 5044 -17.2 9855 M492 -34.1 15009 O234 -45.1 22374 t0653 -52.4 30224 14972 -50.5 39868 18473 -U3.1 Revatestion Surpis 11183 9127 -18.4 16563 13133 -20.7 20206 19302 -4.5 23628 29880 22.2 26730 70967 165.5 30106 66241 16.5 33942 10607 212.5 Total EqGity 55262 12985 -14.9 22652 18177 -19.8 30061 25794 -14.2 39637 37114 -3.9 49104 81520 66.0 60330 108213 67.9 73910 124551 68.7 Long-Tere Deht 6207 5158 -16.9 10355 6040 -41.7 16278 9010 -44.6 18903 20041 6.0 22048 31539 52.8 31826 52762 121.4 26634 71909 2M0.0 Loss. Current Pmrtton 443 489 10.2 79 701 -11.2 1190 1238 4.9 2144 2464 14.9 2976 5369 80.4 3632 6130 6. 4140 5804 48.2 not LMg-fero DOt 5764 4 170 -9.0 9566 5339 -44.2 85098 7772 -48.5 M659 17577 4.9 172 2170 47.7 20194 4663 130.9 22494 74105 229.4 Other Litaltin 994 830 -6.1 1074 1108 3.2* 1302 1612 23.0 1571 2194 39.7 1884 370 105.4 2243 853 280.5 2655 12162 3M.8 Cerent, tiatlIttlus Acco t Payetabe 423 571 35.0 648 985 52.0 883 2667 202.0 1330 4034 203.3 1774 4750 167.8 2197 5134 133.7 3054 1144 274.7 Accomnts Pallo to ISA 345 152 -55.9 688 533 -22.5 843 1677 98.9 612 164 169.0 650 3055 363.0 627 229 252.2 692 525 674.9 Cerrent Portson of Mebt 443 488 10.2 789 701 -11.2 1180 1238 4.9 2101 2464 14.9 2976 5369 80.4 3632 6130 68.8 4140 40.2 Other* 0 498 0 761 0 1455 0 1174 0 1695 0 2309 0 lutot Current Lscblsitius 211 8709 4.1 2125 2910 40.2 29 7037 142.2 4M6 938 129.0 5400 14929 174.6 456 15861 45.1 7876 2595 229.5 TOTAL LIA8LITIES AND E[ITT 23121 20190 -12.7 35417 27604 -22.1 49367 42215 -14.5 65053 66203 8.4 7540 129389 70.1 89223 572245 93.0 106815 236773 121.6 fl~.. av#~.t ...l.. *..... SS *f.fl tflfl 8ame fin ne es..., san. a~.. *an . ••••s »... .å~ *~hl s - ~~ -~ COLONDIA HYD0 POVER PROJECT EAPEM BE EMERBIA ELECTRICA DE BOOTA Rate of Return based an actual project cost Killion of 4id-t977 USDis A. 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1980 1989 1990 1991 1992 Hesitas lvestmeots bi Civil dorts FC 0.8 5.1 7.3 7.5 5.9 8.4 3.9 0.1 6.3 2.1 LC 0.8 7.0 13.6 16.1 10.3 9.1 14.6 6.8 4.3 2.0 Equipaet FC 0.0 0.0 5.0 29.8 33.7 12.9 7.6 5.1 2.9 0.2 LC 0.0 0.0 0.0 3.1 7.7 4.7 4.1 2.6 0.4 .0 Eng I Ada FC 0.0 0.0 0.0 0.1 0.1 0.1 0.0 0.3 0.0 0.6 LC 0.5 2.8 3.7 4.6 3.9 3.3 4.3 2.3 1.1 0.6 Total FC 0.8 .1 12.3 37.4 39.6 21.4 11.5 5.6 9.2 2.9 LC 1.3 10.5 17.4 23.9 21.9 17.1 25.0 11.7 5.8 2.5 Total 2.1 15.6 29.7 61.2 61.5 38.4 36.5 17.2 15.0 5.5 Nositas Operation costs (LC 7.5 7.5 7.5 7.5 7.5 7.5 7.5 Distrihtion Investoest FC 0.4 18.6 22.7 24.1 17.7 1.9 LC 8.4 18.6 22.7 24.1 17.7 1.9 Total 16.7 37.1 45.4 48.2 35.3 5.7 bistrib.Adolo.Costs ILCI 2.0 0.1 6.6 9.3 9.7 9.7 9.7 TOTAL COST FC 0.8 5.1 12.3 37.4 39.6 21.4 11.5 5.6 17.5 21.5 22.7 24.1 17.7 1.9 0.0 0.0 LC 1.3 10.5 17.4 23.9 21.9 17.1 25.0 1.7 14.2 30.6 30.4 38.2 34.5 19.1 17.2 17.2 Total 2.1 15.6 29.7 61.2 61.5 38.4 36.5 17.2 31.7 52.1 53.1 62.3 $2.2 20.9 17.2 17.2 BENEFITS at Energy Sale c/d/ LC 6.2 29.0 50.0 73.0 78.1 78.1 78.1 *nection Charges LC 9.9 21.1 23.4 25.7 3.4 Fuel savings FC 4.7 6.7 3.7 0.4 TOTAL DENEFITS 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 20.6 56.8 77.1 91 81.5 78.1 78.1 NET BENEFITS -2.1 -15.6 -29.7 -61.2 -61.5 -38.4 -36.5 -17.2 -31.7 -31.3 3.7 14.8 46.9 60.6 60.9 60.9 INTERNAL RATE OF BETURN 1 0.100295 al Current esticate for shadow prices iss 1.0 bi Actual Project Costs Espressed in Bid-1977 Constant Prices IN c/ Energy Losses decreasin free 24.5 % in 1986 to 13.2 1 in 1990, accordin vlth current Losses Reuction Progrft di Current tariffs for I and 10 1 increase in 1%7 and thereafter, according ith current tariff increase plan. 一64- 劃‘”輯，“華“&,~ 森馴•“，登“義“’壯 T。―&&&,&&&&:& 一，!}；〔〕〕！！〕！！〕〕 盒蘿！&&&，華，,&&&& 矗遞―華召賽邑“”露召””森。 ―。―叢姿疲藝亳寧為召：’遞盡 童篡！&&&,,,,&&&, 籐土廈―實姿糁盔‘,”姿，’召乏 名擊．&_ 】’&！〕蝨‘&&’召盔‘:& ,:i盒i-’薇斗華華丰華華’；拳 暴泵基土萬―要巫禺墾論“二：”森雲 婁旦之擊一‘__ 藹I奮’〕！！〔！〕‘!!!!！〕〕 “垂萋喜鑪―f認‘f甲甲妒莘甲“必 蘿粹乏逕引挺；邊選”&:&‘姿。 一】誌―&&&,&&&,&&&& 織麻 付