Colombia - Power Development Finance Project

Docrrmmtof TheWorld Bank FOR OFFICIALUSE ONLY Report No. 12116 PROJECT COMPLETION REPORT COLOMBIA POWER DEVELOPMENT FINANCE PROJECT (PEN) (LOAN 2401-CO) JUNE 30, 1993 Trade, Finance, Industry and Energy Division Country Department I11 Latin America and The Caribbean Regional Office This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  ISA Interconexi6n Electrica S.A. EEEB Empresa de Energia Electrica de Bogota EMCALI Empreeae PQblicae de Cali EPM Empresas Pfiblicas de Medellin ICEL Instituto Colombiano de Energia Electrica CVC Corporaci6n- Aut6noma Regional del Cauca CORELCA Corporaci6n Electrica de l a Costa AtlSntica JNT Junta Nacional de Tarifas FODEX Fondo de Moneda Extranjera FEN Financiera Electrica Nacional I D B InterAmerican Development Bank SAR Staff Appraisal Report CURRENCY EQUIVALENTS Currency Unit - Colombian Peso (ColS) Year Year Average Yearend FISCAL YEAR = CALENDAR YEAR  FOR OFFICIAL USE ONLY THE WORLD BANK Washington, D.C. 20433 U.S.A. Office of Director-General Operations Evaluation June 30, 1993 MEMORANDUM TO 0 SUBJECT: Project Completion Report on Colombia Power Development Finance Project (Loan 2401-COl Attached is the "Project Completion Report on Colombia - Power Development Finance Project (Loan 2401-CO)" prepared by the Latin America and the Caribbean Regional Office, with the Borrower providing Part 11. The project was both ambitious and innovative: one of the first Bank operations to use the "B" loan co-financing scheme, it aimed at (i) providing external finance for completing Colombia's 1984-85 hydroelectric development program, and (ii) establishing Pinanciera Energetica Nacional (FEN) as a strong development banking institution to which some appraisal responsibilities might be delegated for future Bank sector lending and through which improvement in sector finances would be promoted. Whereas the first objective was accomplished partially (the project did not result in further "B" loans as expected), the second was not, due primarily to weak Government commitment during project implementation and deficient Bank supervision. In spite of recent changes in Government policy aimed at broadening FEN'S role, persisting financial issues in the power sector at the time of PCR preparation made its sustainability as a development banking institution uncertain. Accordingly, the project is rated unsatisfactory overall, its institutional impact negligible, and its sustainability uncertain. The PCR provides an extremely detailed and thorough analysis of the circumstances and performance of the project. In light of the pilot nature of this project (one of the first major financial intermediation operations in the power sector) and the significance of the issues revealed by the PCR, an audit is planned. Attachment This docuntnt has a restricted distribution wd my be wed by recipients only in the perfornsnco of their officiat duties. Its contents m y not otherwise be disctosd without Uorld Bank authorization.   Preface . . . . . . . . . . . . . . . . . . . . . . . . . . Evaluation Sucmury . . . . . . . . . . . . . . . . . . . . PART I: PROJECT REVIEW FROM W ' S P- . 1 Proiect Xdentitv . . . . . . . . . . . . . . . . . . . 3 proiect Obiectivem D Project Objectivem Project Componentm Project Financing ... .. .. . . . . ... . . . . . . e. .a .m. ................... Procurunent Subloanm and Onlending Temm . . . . . . .. .. .. .. ................ Loan Structure . . . . . . . .. .. .. . .. .......... 4. B o i e c t m i a n Project Formulation and Timing . . .................. Innovative project concept . . . . . . . . . . . PEN*l Role . . . . . . . . . . . . . . . . . . . Changem during Lorn Coaaaittoo Review and Negotiationm . . . . . . . . . . a Bank Policiem f o r Financial 1nterPP.diuiam b .. Exclumion of Project Comt Overrun Financing Dimburmeanent Arrangemmntm . . . . . . . . . . . 5. p- . . . . . m a . . . . . . . . . . Utilization of Project Financing Strengthening of ?EN am Pawor Sactor Developrunt Bank . . . . . . . . . . a. Improvomont of Project Analymim and Monitoring Capability . . . . . . . . b . Enhancmnont of Financial Planning and Advimory Capacity c . Mobilization of Dollurntic Savingm . . . . . . . . . . . ... . . . . . . . . . .. Domomtic Lending Bank Suporvimion Activitiem . . . . . . . . a Suporviaion Wimmionm . . . . . . ... b .. Quarterly Project Progremm Roportm Activities Submquent t o Cornplotion This document has a restricted distributionand may be wed by mipienu only in the performance of their ofllcial duties . ttscontents may not otherwim k d i l d without World Bank authorization.  Table of Contents (Continued) Page No. a. Six Month Extension of Subloan Amortization Period . . . . . . . . . . . . . . . . . . 29 b Temporary Exception from 7:l Debt To Equity Limit 32 c.. Increase in FEN'e Responsibilities . .. . . 35 Financial Performance 37 ProjectRiaks . . . . . . . . . . . . .. .. .. . . . .. . . . . . . . . . . . . .1990.and. .1991. . . . 39 Actions which affected Project Implementation . . . . . . . 40 7. project suatainability . . . . . . . . . . . . . . . . . . . . . 42 8. -~erfonnance . . . . . . . . . . . . . . . . . . . . . . . . 42 Major Strengths and Weaknesses . . . . . . . . . . . . . . . . . 42 Unforeeeen Foreign Exchange Risks . . . . . . . . . . . . . . . . 44 Hedgee against Maxi-Devaluation of the Colombian Peso . . . . . . 45 Hedgee against Major Devaluation of US$ . . . . . . . . . . . . . 46 The Bank's New Currency Management Policies . . . . . . . . . . . 48 LeeeoneLearned . . . . . . . . . . . . . . . . . . . . . . . . . 50 9. Borrower Performance . . . . . . . . . . . . . . . . . . . . . . 52 Major Strength6 and Weaknesses . . . . . . . . . . . . . . . . . 52 LeeeoneLearned . . . . . . . . . . . . . . . . . . . . . . . . . 54 PART 11: PROJECT REVIEW FROM BORROWERS PERSPECTIVE Background . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 Borrower'e Report-Introduction . . . . . . . . . . . . . . . . . 61 Achievement of Overall Loan Objectives . . . . . . . . . . . . . 61 Conclusione . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Borrower'sComments . . . . . . . . . . . . . . . . . . . . . . . 63  Table of Content6 (Continued) Page NO. PART 111: SUPPLEMENTAL INFORMATION RELATED BANK LOANS . . . . . . . . . . . . . . . . . . 68 PROJECT TIMETABLE . . . . . . . . . . . . . . . . . . . 70 LOAN DISBURSEMENTS (INTO SPECIAL ACCOUNTS) . . . . . . 71 TABLE 3.2: SUBLOAN DISBURSEMENTS . . . . . . . . . . . . . . . . 72 LE 4.a: FEN'e LOCAL CURRENCY BORROWINGS . . . . . . . . . . . 73 W L E 4.2: FEN'S LOANS TO POWER COMPANIES FROM DOMESTIC FUNDS . 74 LE 4.3: PROJECT COSTS (No data on actual coete available - - eee para. 5.39) TABLE 5.A: FEN'e RETURN ON EQUITY COMPARED TO INFLATION, AS ESTIMATED AT APPRAISAL AND ACTUALLY ACHIEVED . . . 75 5.2: FEN - COMPARISON OF PROJECTED VS. ACTUAL SUMMARY BALANCE SHEETS AND SELECTED PERFORMANCE INDICATORS . . 76 LE 6: STATUS OF LOAN COVENANTS . . . . . . . . . . . . . . 77 ma.ud: USE OF BANK RESOURCES . . . . . . . . . . . . . . 83  T a b l a of C o n t a n t o (Continued) Page No. ANNEXES I. U n i l a t e r a l Six Month E x t a n m i o n of Subloan Amortization P e r i o d. . 84 -A: REPAYMENTS BY PEN ON PROJECT LOANS AS AGREED COMPARBD TO REPAYMENTS TO FEN ON SUBLOANS AS ORIGINALLY SCHEDULED AND AS MODIFIED FOR S I X MONTH EXTENSION O F SUBLOAN REPAYMENT . . . . . . . . . . . . . 93 TABLE B: REPAYMENTS BY ?EM ON PROJECT LOANS, ASSUMING MODIFIED SCHBDULB lKlR BANK W A N , COMPARED TO REPAYMENTS TO R N ON SUBLOANS AS ORIGINALLY SCHEDULED AND ASSUNING MODIFIED SCHEDULE . . . . . . . . 94 a: REPAYUENTS BY PEN ON PROJECT LOANS AS AGREED COMPARED TO REPAYMENTS TO FEN ON SUBLOANS ASSUMING MODIFIED SCHEDULE AND MODIFIED SCHEDULE EXTENDED S I X m H T H S . . . . . . . . . . . . . . . . . . 95 . . . . . . . . . . . . . . . . . . 11. Extanded T a m T r a n o f o n ~ t i o n 96 111. FEN's -- comment. on tha praliarinuy draft of thia report a e n t b y l e t t o r o f J a n u u y 3 , 1992. . . . . . . . . . . . . . . 114  COLOXBIA POWER DEVELOPMENT FINANCE PROJBCT (LOAN 2401-COl Thie La the Project Completion Report for the Power Developnent Finance Project which had two main objectiveer (i) to provide the external financing required to euetain during 1984-85 high priority electricity projecte that were already under conmtruction; and (ii) to mupport the developnent of the then recently created Financiers Electrica Nacional (FEN) ae a financial intermediary which would raiee domeetic and foreign funde to be relent to the Colombian power companiee and a8 a development bank which would etrengthen the financial management of the eector and act ae a power financing etrategy advieer to the Government and the eector. Thie wae an innovative project which broke new ground in the World Bank and which wae prepared and appraieed under a very tight timetable. Although the financing objectivee of the project were largely accompliehed with eome delaye, little progreee wae made in achieving the inetitutional building objectives. In order to determine and explain the reamone for this dieappointing inetitutional performance and to reflect eignificant evente which occurred after completion of dieburmemente, this report ie eignificantly longer than the normal PCR. Thie project wae one of the firet to uee a cofinancing echeme adopted by the World Bank in December 1982 under which it would participate in loane provided by cwnmercial banke ("Bn loane) for projecte financed by etandard World Bank loane ("A" loane). Thie echome wae adopted to encourage commercial banke to continue to provide financing for developing countriee at a time when they were curtailing euch eupport. The project financing coneieted of an "A" loan of USS170 million and "B" loane of USS200 million of which the Bank'e ehare wae USS28.2 million. The cloeing date of the "A" loan wae extended by one year to December 31, 1987. The final dimbureeunente from epecial accounte eetabliehed in Colombia'e central bank to eubloane to the power companiee wae made on the "A" loan on December 1, 1987, and on the "B" loane in July 1988. Preparation of thie report began in January 1990. A draft was eent to the borrower, FEN, and ite comments thereon, which are eet forth in Part I1 of the report, have been reflected ae appropriate in Part I. Drafte of the report were aleo eent to the Government and the cofinanciera, but no comnente were received. The Trade, Finance, Induetry and Energy Divieion, Department I11 of the Latin American and Caribbean Regional Office (LA3TFIE) prepared parte I and I11 of the report based, interu,on the Staff Appraieal Report, the loan documente, records of the World Bank and the Final Evaluation Report eubmitted by FEN.   PROJECT COMPLETION REPORT COLOMBIA POWER DEVELOPMENT FINANCE PROJECT - 1LOAN 2401 COL Proiect Obiectives and Deecription 1. Oblectives. The project had two main objectivee: firet, to provide the external financing required to sustain during 1984-85 the scheduled pace of eeveral high priority electricity development projects that were under construction, and in thie connection to re-enliet the eupport of foreign commercial lenders to the power eector and to Colombia through a cofinancing package; second, to eupport the development of Financiers Electrica Nacional (FEN). FEN had been created in 1982 in conformance with an undertaking made by the Government in 1981 under the Guavio Hydro Power Project (Loan 2008-CO) when it was recognized that the local currency cost component of the power sector inveetment program could not be met entirely through eelf generated funds. FEN'e development wae to be eupported in several respects: to enable it to eerve as an effective channel for power sector lending projects in Colombia; to strengthen the financial management of the eector by enhancing FEN'e ability to monitor the financial performance and proepecte of Colombian power companiee and to act ae power financing strategy adviser to the Government and the eector; and to enhance FEN'e image in the domeetic capital market and thue help facilitate ite acceee to future eavings (parae 3.1 and 4.5). 2. Deecri~tion. The project coneisted of eubprojecte representing a two year time-elice (1984-85) of priority ongoing electricity developnent projecte of ISA, EEEB, EPM, CVC and COFtELCA, several of which were already partly financed by Bank loans. Subprojects coneieted of the portion of the civil worke, goode and eervices required for the electricity development projects during the period covered (para 3.2). 3. Financinq. Thie project wae one of the firet to use a scheme adopted by the Bank in December 1982 to encourage commercial banks to continue to provide financing for developing countriee at a time when they were curtailing euch eupport. In addition to a normal Bank loan, referred to ae an "A" loan under thie echeme, the Bank would aleo participate in the comercial bank loans, referred to as "B" loane. The project financing in thie came coneieted of an "Aw loan of USS170 million from the Bank and "BW loan. of USS2OO million, of which USS28.2 million wae from the Bank and the balance from commercial banks. The "A" and "B" loane would finance the mubloane made by FEN to finance the foreign exchange coete of the eubprojecte not financed from other sourcee (parae 3.3-3.5).  Proiect Deeian 4. The project wae deeigned primarily as an emergency financial aeeietance project with an institutional component geared to strengthen the newly created FEN a8 a power eector developnent bank. The project concept wae innovative and broke new ground in the Bank. It was prepared and appraieed under preeeure and a very tight timetable. Thie reeulted in delays in obtaining Loan Committee authorization to negotiate and in completing the negotiatione (parae 4.1-4.7). The time taken during the unusually extended Loan Committee review wae needed to respond to question8 raieed about departure8 from Bank policies for financial intermediariee and to clarify whether the propoeale were coneietent with the Bank'e policy that coet overrune are normally not eligible for Bank eupplementary financing. The latter ieeue aleo contributed to the delay in completing the negotiatione. Becauee theee ieauee were reeolved in a shorter time than wae needed to complete the arrangement6 for commercial bank financing, there were no adverse coneequencee ae a result of theee delaye (parae 4.12-4.20). 5. Both the Staff Appraieal Report (SAR) No.4771-CO and the Preeident'e Report N0.P-3750-CO clearly epecify the role envieaged for FEN ae a power eector development bank, and the loan document8 contain pertinent covenants coneietent with that vieion. In retroepect, however, it is clear that the Government and FEN were not ae etrongly committed ae appeared during appraisal and negotiation to enhancing FEN'e ekille so that, in addition to raieing funds for the power eector from domeetic and foreign sources, it would be able to carry our broad reeponeibilitiee to improve power eector financial management (parae 4.8-4.10). Thue, there ie no reference to this broad role in FEN'e description of the project'e objectivee in the Final Evaluation Report which it prepared (Part 11, para 2.03), and there ie a dieclaimer that thie wae one of the project'e main objectivee in FEN'e letter commenting on a draft of thie report (reproduced ae tranelated in Part 11, para 3.00). 6. It is aleo clear in hindeight that both the Colombian authorities and the Bank undereetimated the difficultiee which FEN would encounter in establishing iteelf to fulfill its broad monitoring and advieory role with reepect to power eector finance8 and to eerve ae an intermediary to which eome appraisal reeponeibilitiee might be delegated for future Bank eector lending operatione. Aa a result, time target8 set forth in varioue undertaking8 agreed during negotiatione to achieve progreee in theee reepecte proved to be unrealistically ehort. In all theee reepecte, the Bank underestimated the rieke related to FEN'e ability to fulfill ite role ae a developnent banking inetitution and diemieeed them too blandly (paras 4.11). 7. The disbursement proceduree, in combination with the reporting requirements and the failure to audit etatemente of expendituree, were not adequate to confirm that project financing wae limited to foreign coats and that the limits on the extent of retroactive, and to a lesser extent eupplemental and overrun, financing were adhered to (parae 4.21-4.25, 5.34 and 5.36). Aleo, the projections of the conetruction program8 and financing requirement8 of the power companiee for the 198411985 project time slice made at appraieal by the Colombian authoritiee and the Bank were overeetimated. The elower pace of conetruction than forecaet ia the primary reaeon for the lag in eubloan diebureemente, which in amount and extent beyond December 31, 1985, were eignificantly greater than expected when the loan wae approved and which required a one year extension of the December 31, 1986, cloeing date.  The final diebursanent to clear the advances made from the commercial bank loane occurred in July 1988 (paras 5.2-5.7). 8. Initially, the Bank cloeely monitored project execution and FEN operatione. Ae time paeeed and it became increaeingly plain that FEN'S role in reeolving the power sector'e probleme would be of minor importance, the Bank gave priority to effort8 to prepare what became the Power Sector Adjustment Loan approved in December 1987, to that loan'e eubeequent eupervieion, and eince 1990 to preparation of a poeeible eector loan. In thie context, the attention given to eupervieion of thie project diminished eubetantially well before the loan wae fully diebureed. Overall, the eupervision effort was ineffective (para 5.22). In addition, not only wae there delay in implementation of most of the component8 of the project deeigned to etrengthen the newly created FEN'e capabilitiee, but for the moet part the impact of theee efforts ham been ineubetantial (para 5.8). The principal ehortcomings were: (a) There wae a one year delay in obtaining the coneulting aeeietance to etrengthen FEN'e technical capacity to evaluate loan propoeale and monitor project implementation, relatively little uee of the loan funde provided for thie purpoee, and virtually no independent appraisal activity by FEN through the end of 1990 (para 5.9). (b) FEN encountered difficultiee in designing computer eyeteme which would serve ite neede, ae well ae thoee of the power companiee individually and the eector ae a whole, for monitoring hietotical financial performance and for financial planning. The study of eector finances and the formulation of a proposed financing etrategy, which FEN undertook to complete by September 1984, was eubmitted only in draft form in January 1986. The final report wae never eubmitted, nor did the echeme for an annual exchange of views on power eector financing strategy ever take place ae agreed in the loan documente. Inetead exchanges of viewe took place in the context of the preparation and eupervieion of the Sector Adjuetment Loan and eubeequent activities (paras 5.10-5.12 and 9.1). (c) Both the mobilization of domeetic eavinge and extent of domeetic lending by FEN fell well ehort of the appraisal projections and the covenanted targets. Thie wae due to the deterioration in the financial situation of many of the power companiee which made it difficult or impoeeible to obtain the intermediation of comnercial bank8 which wae required ae a condition of FEN'a domeetic loane to power companiee. That deterioration wae not reaeonably foreeeeable at the time the project wae appraised, eo that neither FEN nor the Bank ehould be faulted for the failure to reach projected borrowing and lending levele. The reeponsibility for the emergence and pereistence of deteriorated financial conditions in the power eector lie8 with the Government (paras 5.13-5.20 and 5.71).  (d) The ineffectiveness of Bank supervision is indicated by such matters as failures to: (i) notice that the consultant who was retained, inter alia, to propose key financial indicators to be monitored had failed to do so and to follow up on this omission (paras 5.24-5.25); (ii) arrange for the audit of statements of expenditures (para 5.34); (iii) be aware of and follow up on the decline in real terms of FEN'S profitability, both during supervision missions and when dealing with the issues raised by FEN'S unilateral six month extension of the repayment period for project subloans (paras 5.28-5.29, 5.46 and 5.65-5.67); (iv) continue monitoring compliance with the covenants in the Power Financing and Guaranty Agreements (paras 5.31 and 5.35); and (v) assess critically the expectations that subloan disbursements would be completed by mid or year end 1986 (paras 5.4-5.6 and 5.35). 9. 9. When informed by FEN in July 1988 that it had unilaterally extended the subloan amortization period by six months without requesting the Bank's prior agreement, the Bank responded in a formally correct manner. It admonished FEN to comply strictly with the provisions of the loan documents in the future. Some of the supporting information for the Bank's conclusions that the extension would have no material adverse effect on FEN'S finances and that FEN would continue to have adequate financial ratios is missing from the project files. It appears that the Bank's financial analysis of this issue was superficial. Based on a retrospective review of this matter, it also appears that if this extension had been analyzed adequately, the Bank would have become aware of two important issues: the decline in FEN'S rate of return on equity in comparison to the rate of inflation; and the mismatch between the repayment terms of the external loans obtained under this project and of the subloans mode from theee eources. Had the Bank become aware of these issues, it could have suggested corrective measures (paras 5.41-5.48). 10. Temporaw Exception from 7:l Debt to Equitv Limit. Since 1988, FEN hoe not complied with its undertakings to maintain its debtlequity ratio within the limit of 7 to 1. The ratio, as of the year end, was 7.4 in 1988, 10.0 in 1989 and 8.8 in 1990. The shortfall in required equity to comply with these covenants as of the year end, using year end exchange rates, was the equivalent of about US$6 million for 1988, US$42 million for 1989 and US$33 million for 1990. By letter dated June 20, 1989, PEN requested a temporary exception from the debtlequity limitation. FEN referred to several major foreign credits to be made available to it shortly which would cause it to  exceed the limit. It also referred to pending legislation pursuant to which FEN would receive additional capital contributions of about Co1$30 billion in 1989. It stated that this would bring it back into compliance with the debtlequity limit. It concluded with a statement of its intentions to submit information by the end of 1989 in support of a request for a permanent increase in the debt limit. The Bank agreed with this request and apparently also to a further request made in June to extend the temporary exception to December 31, 1990, The first increment of equity capital to be provided to FEN under the new law was paid in by the Government in December 1990 in the amount of ColS19.76 billion, equivalent at the year end exchange rate to about US$35 million. As indicated above, this was not sufficient to bring FEN back into compliance with the covenanted debt limit. In September 1991, FEN requested comments from the Bank and the Interamerican Development Bank (IDB) on a proposed increase in its permissible debtlequity ratio to conform to much higher limits allowable under Colombian law applicable to financial institutions under which FEN now operates. As proposed, the limit would be at least 12:l and possibly as high as about 20:l. Both the Bank and IDB are evaluating this request in the context of the role and financial risks under which FEN will be expected to operate in the future (paras 5.49-5.55). 11. Increases in FEN'S Responsibilities - 1990 and 1991. As a result of new laws and other regulatory changes, there has been a substantial increase in FEN'S responsibilities in 1990-1991. FEN is now authorized to serve the entire energy sector, although its activities continue to be primarily focussed on the power sector. It may now also lend, not only for investment programs and projects, but to refinance debt service obligations in order to rationalize the functioning of the energy sector in accordance with the Government's policies. In addition, FEN is no longer subject to the requirement that credits made from domestic savings must have a bank guaranty or be rediscounted. As a result there has been a very large expansion, by about six times and predominantly in 1991, in the amount of its outstanding domestic borrowings. These, plus the substantial addition to its capital provided by the Government in December 1990, were used principally to finance debt service payments of the power companies due on loans guaranteed by the Government. The power companies need this assistance because the repayment terms of their loan financing impose financial burdens on them which are not reasonably related to their earning power, even assuming that they were efficiently operated and managed and the level and structure of their tariffs were appropriate. These developments highlight the need for FEN to provide more ambitious term transformation on a systematic and transparent basis, in order to apply effective and realistically achievable financial discipline to the performance of Colombia's power companies (paras 5.56-5.64). 12. The actions taken in 1990 to increase FEN'S responsibilities are indicative of a change in the attitude of the Government, which now expects FEN to act as an effective development bank in support of government policies to improve the functioning of the power sector. FEN is to assist in the formulation of performance or management contracts which the power companies would enter into with the Government and which would detail measures to be taken by them to improve their efficiency. Fulfillment of these contracts will be supervised by the government, and continued access to FEN credits will be conditioned on compliance with these contracts. Companies owned by the national government which fail to comply with the contracts are to be subject to restructuring or liquidation (para 9.1 (iv)).  Decisions Adversely Affectina Project Imvlementatio~ 13. A major weakness in the Bank's supervision was the failure to aeeign to this activity a staff member who was an expert on financial intermediariee and capital markets. The effectiveness of the Bank's efforts to improve FEN'e institutional capability would have been greatly improved had such a staff member been made available and participated in most, if not all, of the supervision miesione. It would also have been appropriate to have designated such staff member as the projects officer. This staff member would have provided an element which was missing during the implementation of the project: an expert representing the Bank who would gain the confidence and respect of FEN officials and who would have provided continuity to the Bank'e efforts to encourage and when necessary prod FEN to fulfill the broad development banking role envisaged for it (para 5.70). The project'e handling was also adversely affected at times during appraisal and supervision by the absence of a qualified power utility financial analyst (para 8.2). 14. On the Government's side, a major factor affecting implementation was the decision to let the real tariff levels fall after the 1985 devaluations, since it had a direct bearing on the internal financing of the utilities. This prevented FEN from taking a more active part as a financial inetitution, lending local currency funds to the utilities, and as a development institution capable of imposing conditions that would improve the financial and managerial poeition of its borrowers. These two capabilitiee go hand-in-hand since the conditionality cannot be successful unlese a permanent relationship is established with the borrower that would make it poseible for FEN to demand actions, even if politically difficult to implement for the borrower, that would make it worthwhile for the beneficiaries of FEN loane to comply. This relationship, and the role of FEN ae a development institution as well, were elements lacking in the organization of the power sector in Colombia. It remains to be seen whether the efforts of the current Government to reinvigorate FEN will be successful (para 5.71). Performance 15. - Bank. The major strength demonstrated by the Bank in its handling of this project is its capacity and that of its staff to be innovative and respond promptly to the power sector's need for additional financing for ongoing projects. This was a very significant accomplishment. Despite the limited progress made in achieving the project's institution building component, and despite the shortcomings in the Bank's performance during design, appraisal and implementation of the project cited herein (para 8.2), the project was a worthwhile undertaking. There was a genuine need for an institution such as FEN with the capabilities and responsibilities which the Bank envisioned when the project was proposed, and there still ie such a need. The Bank did not pursue the appraisal report vision of FEN effectively, but the opportunity to do so remains. The lessons learned in executing thie project should be valuable in pointing the way to improved performance in the future (para 8.1). 16. FEN and other Colombian authorities have called attention to the very large increases in the project's debt service costs in both Colombian peso and US dollar terms over what was anticipated at appraisal. The substantial increases in debt service costs resulted from two sources: (i) the unexpected acceleration in the mid-1980s of the crawling peg adjustment of the value of  -viii - the peso in relation to the dollar including a 51% devaluation in 1985 and (ii) the unanticipated substantial devaluation of the dollar vis-a-vis the Japanese yen (Y) and other foreign currencies such as the Deutsche mark (DM) beginning in 1985, which resulted in much higher debt service costs for the "A" loan under the Bank's currency pooling system and for the yen denominated "Bn loan. The Bank has been criticized because of its failure to provide solutions which would have protected the utilities and FEN against these unforeseen exchange risks. Because the possibility of a maxi-devaluation of the peso in relation to the US$, or of the US$ against other currencies, was not "likely" when the project was designed and negotiated, this criticism is inappropriate. Moreover, in the context of then prevailing circumstances, the practicability and merits of the Central Bank providing a hedge against a -5-devaluation of the peso are questionable. It may be fair, however, to criticize the Bank for taking too long to formulate its institutional response to the problems resulting from the devaluation of the US$ subsequent to mid- 1985. Consideration should be given to undertaking retrospective studies of what might have been achieved and at what cost had hedging against the devaluation of the US$ in relation to other foreign currencies been implemented for the project loans (paras 8.3-8.18). 17. - FEN. The major strength shown by FEN as a result of this loan was its proved ability to mobilize financial savings in the domestic market through innovative and suitable short- and medium-range instruments (para. 5.24). Its major shortcomings were reflected in the very limited progress made in improving its technical and managerial skills to serve as a power sector development bank in such respects as the ability to evaluate loan proposals and monitor the implementation of projects for which it provides financing, and to serve ae a puwer sector financing strategy advisor to the Government and the utilities. These shortcomings, in turn, reflect a lack of commitment to these objectives by the Government and FEN and to consequent inadequate follaw through by them (paras 4.10 and 9.1). Lessons to be Learned 18. The lessons of experience under this project that may be relevant to future Bank-financed projects include (see paras 8.19, 9.2 and 12.2-12.4): (a) Emergency Financial Packages; New Lending Models. (i) A special procedure could be established by the Bank for dealing with emergency financial packages under specific conditions, so that ad hoc arrangements would not have to be devised as was the case for this project. (ii) When new lending models are being implemented, particularly as exemplified by this project when they involve an early application of a new cofinancing scheme in the success of which both the Bank and its borrowers are highly interested and when they are part of an emergency package for a client sector: more rather than less financial analysis should be applied;  and the financial and institutional aspects should be more closely supervised, particularly if the technical component is weak. Such supervision should be entrusted to Bank staff and not delegated to consultants. (b) Institutional Development. When a principal objective of a project is to assist a newly created financial intermediary to serve as a development bank for a capital intensive infrastructure sector and strengthen its financial management, major responsibility for the Bank's activities during the design, appraisal and supervision stages should be assigned to a staff member who is an expert on financial intermediaries and capital markets. Key financial indicators to be monitored should be agreed during appraisal. When there are shortfalls in performance, appropriate follow up action should be taken to see that corrective measures are adopted. (c) proiects Involvinn more than One Sector. When projects involve more than one sector and the staff expertise required for the different sectors work in separate divisions, Bank management must ensure that there is adequate cooperation and coordination between or among the divisions so that staff experts who should be involved in appraisal and supervision of such projects are made available when needed. The Bank should be more agile in forming multi-disciplinary teams across division lines when needed and not attempt to compensate for organizational rigidities by using consultants. (d) Realistic Scheduling. When establishing dates to be specified in loan covenants for carrying out studies and accomplishing institutional improvements, be realistic in estimating the time needed and avoid setting overambitious targets. (e) Disbursement and Reporting Requirements. When disbursements are to be made on the basis of advances into a special account, and the subsequent actual use of the loan funds for project purposes is to be accounted for by disbursement from the special account, it is essential that the appraisal report estimate the rate of expected disbursements on both bases. Disbursement arrangements from the special account and reporting requirements should be designed so that they provide a basis for confirming that the financing was used for the intended purposes such as foreign costs, and that other limits or requirements were observed such as the amount of and time limits for retroactive financing. Staff should make sure that the Borrower arranges for the required audit of statements of expenditures (paras 4.21-4.25, 5.34 and 5.36).  (f) Leniency. There are no advantages to be derived from showing leniency in the event of non-compliance with loan covenants. In the case of this loan this is particularly critical, especially with regard to FEN'S failure to carry out various commitments designed to enhance its financial planning and advisory capacity (paras 5.10-5.12). Less leniency would have stimulated more the institutional strengthening of FEN as a development banking institution. (8)Hedninn Foreian Exchanne Risk. The Bank should reconsider its role with respect to the devastating effect that foreign exchange movements can have on the financial position of borrowers, particularly on producers of non-tradeable8 such as the utilities, in the absence of hedging mechanisms against this risk. The question which should be addressed is whether the Bank is doing all that it should to assist borrowers to address this issue appropriately, including provision of adequate technical assistance and advice to do so. In the case of Colombia, the role of the central bank should be reviewed in this context, since it should be able to provide advice about the costs and potential benefits of hedging and to coordinate and consolidate hedging activities on an economical and efficient basis for borrowers who desire to hedge. Until recently, the Central Bank monopolized all foreign exchange transactions as a result of the exchange control regime (Decree 444 of 1967), but this is no longer the case (see para 5.61). (h) Extended Term Transformation. The feasibility of extending FEN'S term transformation role should be studied in collaboration with Colombian authorities so that the burden of debt service requirements on Colombian power companies is realistically, reasonably and transparently related to their earning power. This subject is discussed further in Annex 11.   COLOMBIA POWER DEVBLOPHBNT FINANCE PROJECT [LOAN 2401-COL PART I. PROJECT REVIEW FROM BANK'S PERSPECTIW PROJECT ID- Project Name : Power Developnent Finance Project Loan Numbere: 2401-CO; B-6-CO; B-7-CO RVP Unit: Latin A m e r i c a & Caribbean Region LA3TFIE Country: Colombia Sector : Energy Subsector: E l e c t r i c Power Sector Development Obiectivecl ( A t timu of ~ r o i e c ta D s a i e a l L 2.1 I n support of t h e Government'e goal t o ineure t h e provision of t h e e l e c t r i c i t y supply required t o support growth and employment, t h e main development objective8 of t h e power eector w e r e to eupply a t l e a e t cost t h e electric energy needed to f a c i l i t a t e growth of productive and commercial a c t i v i t i e s , and t o help eupply e l e c t r i c i t y t o a larger eector of t h e population, only 54% of which had acceee to it. The achievement of t h i s objective was constrained by a ehrinking eupply of funde from t h e international c a p i t a l markets, and by t h e emall e i z e of Colombian financial markets and institutione. Policy Context 2.2 In May 1983, t h e Government reviewed i n d e t a i l t h e j u s t i f i c a t i o n for, and t h e f e a s i b i l i t y of, a l l new inveetmente i n t h e 1983-90 power investment program, taking account, f o r t h e f i r s t time, of competing demande f o r resourcee from other p r i o r i t y sectore, with a coneequent reduction, by almost one t h i r d , of t h e planned c a p i t a l outlaye f o r e l e c t r i c i t y . The authorities recognized t h e need t o manage and monitor more d i r e c t l y t h e impact of e l e c t r i c i t y development on t h e reet of t h e economy, i t 8 compatibility with overall public investment plane and t h e technical and financial f e a s i b i l i t y of programs a t t h e national and eector levele. Bank Particination f A t time of oroiect am- 2.3 Between 1950 and 1984, t h e Bank had made 27 loan8 t o Colombia's power sector, t o t a l l i n g US$1,374 million. A t t h e mama t i m e t h a t thim project wae being appraieed, t h e Bank wae aemimting with t h e preparation of t h r e e propoeed  generation and transmission projects. Past loans had assisted t h e expansion of generating capacity, transmission and distribution f a c i l i t i e s i n t h e systems serving Bogota, Medellin, Cali, Cartagena, Bucaramanga and Manizales, including expansion of e l e c t r i c i t y distribution t o low-income areas. Rural e l e c t r i f i c a t i o n had been aupported under several projects. The Bank had a l s o encouraged t h e creation of Interconexion Electrica, S.A. (ISA) t o serve a s an independent, national generation and interconnection company whose shareholders were t h e large municipal power u t i l i t i e s and t h e Government-owned power companies. A number of Bank loans had financed ISA's construction of large hydropower projects too big t o supply any one company a s w e l l a s t h e creation of a national e l e c t r i c i t y grid. The Bank had a l s o been t h e coordinator of a technical assistance program t o strengthen system planning, and t h e executing agency f o r another UMIP-financed program which would enhance t h e sector's management capability f o r t h e construction of large hydroelectric f a c i l i t i e s . Additionally, i n connection with Bank lending, marginal cost t a r i f f studies had been carried out f o r t h e major markets. Overall, t h e Bank had been involved, f o r a period spanning more than t h r e e decades, i n a l l t h e stages of power development i n Colombia, from t h e planning stage through financing and construction of generation and transmission f a c i l i t i e s , t o delivery of service t o t h e f i n a l consumer. This involvement and t h e previous Bank lending t o Colombia had been found largely successful, and it was generally accepted i n t h e Bank and by t h e Government of Colombia t h a t it had contributed t o t h e evolution of a stronger and more e f f i c i e n t power sector organization. 3.1 The project had two main objectives. The f i r s t was t o provide t h e external financing required t o sustain during 1984-85 t h e scheduled construction pace of several high p r i o r i t y e l e c t r i c i t y development projects, and i n t h i s connection t o re-enlist t h e support of foreign commercial lenders t o t h e power sector and t o Colombia through a co-financing package. The second was t o support t h e development of t h e Financiera Electrica Nacional (FEN), t h e then recently established financial arm of t h e power sector, i n several respects: (i)t o enable it t o serve a s an e f f e c t i v e channel f o r sector lending projects f o r power i n Colombia; (ii)t o strengthen t h e financial management of t h e sector by enhancing its capacity t o monitor t h e financial performance and prospects of Colombian power companies and t o a c t a s power sector financial advisor t o t h e Government and t h e sector; and (iii)and t o enhance its image i n t h e domestic c a p i t a l market and thus f a c i l i t a t e its acceea t o future savings. Project Comwnents. 3.2 Theproject,whichhadanestimatedcostofUS$1.6billionatthetFme of t h e appraisal (USS898 million i n foreign c o s t s ) , consisted of subprojects repreeenting a t w o year time-slice (1984-85) of p r i o r i t y ongoing e l e c t r i c i t y development investments of ISA, EEEB, EPM and CVC, several of which w e r e already p a r t l y financed by Bank loans. Contingent inclusion of ICEL and CORELCA was a l s o provided, subject t o t h e i r preparing by November 30, 1984,  Government-endorsed investment and financing programs covering the years 1984- 87, eatiefactory to the Bank. The plans were prepared but not approved by the Government or the Bank. The Bank decided not to declare ICEL eligible but to allow CORELCA to participate in the expectation that eubeequent to project completion its plans would be reviewed again. The project aleo included a program to assist FEN in strengthening its technical capability. All eubprojects to be included under the project were to be part of the revised National Power Expansion Program, in the case of generation and transmission; leaet-cost solutions for completing the network path to the final consumer, in the case of subtransmission and distribution; or additions to plant needed to complement one or both of them. It was agreed that capital outlays for new electricity development projects would be authorized by the Government only if the project in question would be economically justified and there would be available adequate financing that would not affect negatively the existing financial obligations of the company(-ies) concerned (Guarantee Agreement, Section 3.02). Subprojects consisted of a portion of the civil works, goods and eervices associated with the investments, corresponding to the portion of electricity development projects executed during the period covered. Pro1ect financinq 3.3 This project was one of the first to use a scheme adopted by the Bank in December 1982 under which it would participate in cofinancing provided by commercial banks (so-called "B" loans) for projects financed by standard Bank loans (so-called "A" loans). This scheme was proposed at a time when commercial banks had become less willing to provide financing for developing countries. It was expected that even though the Bank's status as a preferred creditor would not be extended to the commercial banke, they would be more willing to provide cofinancing and do so at somewhat more attractive terms if the Bank were a participant in the cofinancing loans. There were several waye in which the Bank might participate, and in this instance it was by taking the maturities last payable. 3.4 The project financing consisted of an "A" loan of USS170 million from the Bank and "B" loans of USS200 million primarily from foreign commercial banke. One of these was a US$ denominated loan for US$175 million, for which the Midland Bank is the agent bank. The other was Yen denominated in the amount of Y5.5 billion, equivalent to USS25 million at the exchange rate then prevailing of Y220 per US$, for which the Industrial Bank of Japan is the agent bank. The Bank's share of the "B" loans, originally expected to be US$30 million, amounted to USS28.2 million as finally agreed. The "Aw loan had etandard country terms. It was repayable after a four year grace period in semi-annual installment over 13 years and was subject to standard variable intereet rates and foreign exchange risks based on the currency pooling system. The "B" loans also had a four year grace period after which the commercial bank shares were repayable in semi-annual installments over four years. No repayments of the Bank's share of the "B" loans were due until the commercial bank shares were fully repaid. Then, the Bank's share of the US$ denominated "B" loan, including US$5 million provided by the Nordic Investment Bank, warn repayable over two years, and its share of the Yen denominated "Bw loan warn repayable over 1 112 years. Interest rateta on the "B" loans were also variable, either 1 518% over LIBOR or 1 112% over the US prime rate for the Us$ denominated loan and 0.4% over the long term prime lending rate in Japan for the yen denominated loan.  3.5 Subloans of FEN t o t h e u t i l i t i e o would finance foreign exchange costs not covered from other oourceo. Bank financing, including its share of t h e cofinancing loans, could be ueed t o finance project costs o r i g i n a l l y expected t o be financed by commercial lendere but, except f o r t h e Mesitas Hydro Power Project, could not be ueed t o finance project coat overruns. It could also be used t o finance intereet during conetruction on existing Bank loans, i n t h e cases of ISA, EEEB and EPM, where t h e Bank'e norma'l c r i t e r i a f o r such financing were m e t . The loan agreements were eigned on June 27, 1984. Retroactive financing from January 1, 1983, was not t o exceed US$29 million from t h e "A" loan and t h e Bank's ehare of t h e "Bn loans. The "B" loans provided t h a t they could finance only expenditures incurred o r paid a f t e r January 1, 1983, subject t o a limit of US$5O million f o r expenditures incurred and paid p r i o r t o January 1, 1984, i n t h e case of t h e US$175 million loan. Procurement 3.6 Goods and works financed under t h e project by t h e Bank Loans, including t h e Bank's share of t h e "Bn loane, were t o be procured under international competitive bidding, i n accordance with Bank guidelines. The commercial lenders' share would follow procurement norme of t h e Bank, IDB, KfW, o r other i n s t i t u t i o n a l lenders i n t h e case of subprojects p a r t l y financed by them, o r local procedures of t h e Government satisfactory t o t h e Bank. Subloans and Onlendina Terma 3.7 FEN could approve eubloans meeting t h e e l i g i b i l i t y c r i t e r i a of t h e project. Subloans financed under t h e "Aw loan required p r i o r approval by t h e Bank. A free-limit, however, wae eetablished f o r eubloans which did not exceed US$4 million f o r any given subproject; except f o r t h e f i r s t three such loans, Bank approval of f r e e l i m i t eubloane was required only before disbursement of Bank fund8 wae authorized. The 'Bw loans could be u t i l i z e d only t o finance expenditure8 for eubprojects e l i g i b l e f o r financing pursuant t o t h e provisions of t h e "A" loan agreement; essentially t h i s meant t h a t t h e subprojects had t o meet t h e t e e t o of e l i g i b i l i t y described i n para 3.2. To be e l i g i b l e f o r subloans under t h e project o r f o r any other new lending from FEN, t h e power companies would have t o be current i n paying t h e i r obligations under all outstanding loans from FEN. Onlending term8 w e r e t o be established upon t h e b a s i s of full-cost recovery by FEN. The foreign exchange r i s k was borne by t h e u t i l i t i e s ; t h i e wae accomplished by denominating t h e subloane i n d o l l a r s equivalent t o t h e various currencies i n which t h e principal of t h e loans t o FEN from t h e Bank and t h e commercial banks would be payable, including t h e basket of currenciee applicable under t h e "A" loan. Interest rates, fees and terms of t h e eubloans w e r e a blend of t h e t e r m s of t h e Bank and co-financing loans, plue a 0.59 epread over t h e i n t e r e s t rate; maximum subloan term was 15 years, including a 4-year grace period. Loan s t r u c t u r e 3.8 I n addition t o t h e ueual Loan and Guarantee Agreements f o r t h e "A" loan, t h e r e is a Power Financing Agreement between FEN and t h e power companies which d e t a i l s t h e i r reopective obligation8 under t h e project (see President's Report, para 63 f o r a f u l l eurmmary thereof). There a r e a l s o separate Loan and Guarantee agreements for each of t h e "BW loans.  4. PROJECT DESIGN AND ORGANIZATION project Formulation and Timinq 4.1 The project waa conceived during a very critical period of Colombia's external financing evolution at a time when the Latin American debt criaia had recently emerged. Commercial aourcea of credit which were part of the financing plana of aeveral large projecta under way in the power aector had completely dried up, and the country waa experiencing aerioua difficulties securing "new moneyn for theae and other purpoaea. Due to thia and to the need to continue implementation of ongoing projects in the power aector, the loan waa deaigned primarily aa an emergency financial aaaiatance project, with an inatitutional component geared to atrengthen the newly created FEN aa a power aector development bank. Although there had not been a formal appraisal miasion, there had been Bank adviaory miasiona to FEN in June and November 1982. The findinga of theae miaaiona, together with a poaitive reaponae from Government during preliminary diacuaaiona, were conaidered an adequate baaia for aeeking management agreement to the propoaed project as aet forth in a combined iasuea/deciaion paper prepared in January 1983. 4.2 A poat appraisal miaaion took place in March 1983. Aa work progreaaed, the main outline of the project remained eaaentially the aame but significant detaila were changed. Thie included aubatantial increaaea in the loan amounta. The original propoaal waa for an "A" loan of US$25 million and a "Bn loan of US$80 million including a poeaible World Bank participation of US$20 million. These were increaaed in two ateps to the amounta finally agreed. Becauae of the need to clarify and refine varioua of the project concepta and detaile, an unusually long eight week period waa required before Loan Committee authorization to negotiate the "A" loan wae obtained in December 1983. The negotiationa were ale0 protracted, and after an initial aeaaion in Washington in January 1984 they were concluded in a aecond seasion in Colombia in February. During negotiations, the Bank reconaidered a position it had previoualy taken, coneistent with a practice followed to promote financial diacipline on the part of Bank borrowera, that the funde provided by the Bank could not be ueed to finance coat overruna on projecta which it had financed under previous loana. Infrequent exceptiona to thia practice, however, were allowed when the circumstancee cauaing the overrun were beyond the control of the borrower, the borrower and the Government had made substantial efforts to meet the coeta of the overrun from their own reaourcea, the project was still economically justified, ita acope could not be reduced and no other sources of financing were available. In reaponae to a Colombian request, the Bank concluded that theae teat8 were met for the Meaitaa project, and it agreed that FEN could utilize US$43.2 million of Bank funda and $19.7 million of commercial bank funds to finance the foreign coat overruns on the Mesitaa project. This waa neceesary becauae it would otherwise not have been possible to use the financing being provided preponderantly for projects already under way, as waa intended when the project was propoaed. 4.3 The loan was presented to the Board for approval in March 1984, nine months later than it had been originally acheduled. The original achedule had foreaeen that the poet appraiaal mission would have visited Colombia in February 1983, that negotiationa would have taken place in April 1983, and that the loan would have received Board approval in June 1983.  Innovative project concevt 4.4 The tight original schedule, the absence of a formal appraisal mission, and the long time it took for authorization to negotiate and then for negotiations to be completed are symptomatic of thie loan's central characteristic: in many respects, it was breaking new ground in the Bank. At the time it was being designed, no institutional framework or tradition was available for a loan that would provide emergency financial assistance to a sector and would be filling the financial gap left by the commercial banks that were withdrawing from lending to developing countries, including Colombia, after more than a decade of liberal lending. The Bank was actively assisting in the establishment of FEN through the June and November 1982 advisory missions when the serious effect of the debt crisis on Colombia's ability to obtain external financing became apparent. Coincidentally, the Bank was considering an expansion of its cofinancing arrangements with commercial banks through a program of "A" and "B" loans. This led the Bank to propose and the Government to agree to an accelerated plan to assist FEN's establishment through participation in this program. 4.5 FEN had been created in January 1982. The need to create such an entity had been acknowledged in a commitment made by the Government in 1981 in conjunction with the Guavio Hydro Power Project (Loan No. 2008-CO) when it was recognized that the local currency cost component of the power sector investment program could not be met entirely through self generated funds. The original concept was expanded significantly by empowering the new entity to raise funds from external as well as local sources. It was intended that FEN would initially focus its efforts on mobilizing voluntary local savings to meet the power sector's requirements for medium and long-term local currency financing, while also establishing itself as a specialized institution to oversee and coordinate the sector's finances and financing strategy. Subsequently, it was expected to expand its operations into the international capital market. 4.6 It was in this context that the appraisal of this loan was undertaken under pressure and with a timetable reflecting the urgent financial needs of the ongoing projects. The nature of the project was basically considered to be a mobilization of resources effort that would be mainly associated with investments that had been already appraised by the Bank. Thus, in the combined issues/decision paper of January 1983, the Region proposed that the project would not require an appraisal report, but only an extended President's Report. This approach was not accepted by Bank management. The fact that it was proposed, however, reflects the urgency of the situation which appears to have affected other decisions; for example, the initial decision (April 1983) not to postpone the loan until a revised 1985-92 expansion program of the power sector wae discuesed with the government and approved by them, despite recognition that the existing 1982-87 program that had been adopted in 1981 overstated investment requirements vis-a-vis the slow-down of demand. This revision was formalized in May 1983, nevertheleee, and the financial projections of the appraisal report are based on the revised scenario. 4.7 The haste and the sense of urgency reflected a real problem, since the Colombian external finance strategy was falling apart without an indication that the authorities were coming to grips with the new developments in the  capital markets, and there was a need t o f i n d complementary sources of funds f o r t h e completion of non-reversible projecta, many of which had been financed by t h e Bank. The s i t u a t i o n of t h e c a p i t a l markets was viewed a t t h e t h e a s a temporary event. It was thought that the "B" loan component would assist the power eector and Colombia i n reetoring normal relations with t h e commercial banke, and t h a t more co-financed projects would take place i n t h e future a s t h e normal way t o channel commercial funds t o development projecte. That t h i s never materialized i e something t h a t could not have been foreseen a t t h e time of appraisal. A year l a t e r , Colombia and t h e commercial banks changed t h e i r respective e t r a t e g i e s i n such a way t h a t Bank co-financing was no longer possible o r necessary i n a "B-loan" context. The new commercial bank strategy entailed relending t o t h e country only all o r most of Colombia's amortizations t o commercial banks with policy endorsements by t h e Bank and t h e IMF, and a p a r a l l e l financing package by t h e Bank and t h e IDB. Accordingly, there is no room i n t h i s scheme f o r "B loan" co-financing. In these subsequent operations, however, FEN has served a8 t h e intermediary f o r t h e substantial amounts of t h e funds borrowed from t h e commercial banke destined f o r t h e e l e c t r i c u t i l i t i e s . FEN'e Role 4.8 A central aspect of t h e loan was t h e r o l e t o be played by t h e newly created FEN a s a development bank f o r t h e power sector. A t t h e t h e of t h e appraieal it w a s thought correctly t h a t FEN w a s not equipped yet f o r t h e larger r o l e it would f u l f i l l i n t h i s respect, i n addition t o tapping domestic and external c a p i t a l markets t o provide financing f o r t h e power companies. It wae anticipated t h a t t h e experience gained i n implementing t h e project, combined with t h e i n s t i t u t i o n a l measures, technical eupport and aesietance incorporated i n it, would contribute t o t h e development of FEN's banking, financial, analytical and policy e k i l l s 80 t h a t it would be more f i t t o carry out i t e broad responsibilities t o improve power sector financial management. The Bank a l s o anticipated t h a t FEN, with these improved e k i l l s , would be qualified t o serve a s borrower and intermediary f o r future power sector lending projects. This was a clever etrategy but i n f a c t , a s pointed out below i n t h e discussion of project implementation and r e s u l t s , very little was accompliehed. 4.9 There are several explanation8 f o r t h i a disappointing outcome. In reporting on t h e poet appraieal mieaion, t h e Bank etated t h a t , predictably, FEN'e role i n sector financee has caueed some appreheneion i n ISA, and t h a t ISA was reluctant t o relinquish i t 8r o l e as coordinator of sector finances, even though ISA o f f i c i a l s recognize t h a t ISA had f a l l e n short i n f u l f i l l i n g t h i a reeponsibility. (Region's letter dated April 29, 1983 t o t h e Minieter of Finance and Public Credit.) The other power companies no doubt also resented and reeisted eroeion of t h e i r existing responsibilities and authority. This etatement is not intended t o imply t h a t t h e power companies did not honor t h e i r obligations under t h e Power Financing Agreement t o furnish information t o FEN with respect t o sub-projects and t h e i r financee. M o s t likely, however, FEN and t h e Government w e r e aware t h a t t h e power companies did not welcome t h e need t o have FEN review and check t h e judgment8 of t h e i r technical e t a f f , and FEN and t h e Government w e r e reluctant t o aeaert FEN'e authority t o do s o when they recognized t h a t t h e qualifications and c a p a b i l i t i e s of FEN's etaff w e r e still largely untested and unproven.  4.10 It is evident in retroepect that the Government and FEN were not as strongly committed as appeared during appraisal and loan negotiation to the enhancement of FEN's status and capability "to monitor the utilities' financial performance and prospects, and act as power financing strategy advisor to the Government and the sector." (President's Report, para 57.) Evidence of this lack of commitment appeared at the very outset of project implementation. The report of the first supervision mission in September 1984 included the following comment about FEN's organization and management: "The organization ...seems capable of carrying out the role ae transfer agent without difficulty. Management does not seem to desire a more effective development role." (Report dated October 10, 1984, Annex 9, para 3). Further evidence of this lack of commitment is the lack of any reference to this broad role in FEN'S description of the project objective in the Final Evaluation Report which it prepared (Part 11, para. 2.03); and the statement in FENWs letter of January 3, 1992, commenting on a draft of this report, that "it is not valid to assert that developing FEN's role as a development bank was one of the major components of the project" (Part 11, Section 3.00). The latter claim ignores the stated objectives and the many significant covenants in the loan documents clearly designed to enhance FEN's capabilities and responsibilities as a development bank for the Colombian power sector. In addition, the Government's decision to let real tariff levels fall after the 1985 devaluations substantially adversely affected the finances of the utilities. This, in turn, impaired FEN's ability to condition its lending to the utilities on their achieving agreed targets for satisfactory financial performance (paras. 5.68 and 5.71). On balance, it is reasonable to conclude that the very limited progress made in achieving the broad objectives for the role of FEN, as defined in agreement with it and the government of Colombia during project preparation, is due principally to lack of effective follow- through by them. The recent (1990) restructuring by the new Government of FEN'S legal organization and domain, however, holds out the prospect that this objective will now be adequately supported and implemented. 4.11 In view of the previous experience with ISA, the likelihood that there would be resistance to imposition of new centralized authority, and that for political reasons the Government would respond weakly to such resistance, could have been anticipated. (See brief discussion of ISA's history in President's Report No. P-4676-CO, for the Power Sector Adjustment Loan, paras 45-47.) Finally, it is clear in hindeight that both the Colombian authorities and the Bank underestimated the difficulties which FEN would encounter in establishing itself to fulfill its broad monitoring and advisory role with respect to power sector finances and to serve as an intermediary to which some appraisal responsibilities might be delegated for future Bank sector lending operations. As a result, time targets set in various undertakings agreed during negotiations to achieve progress in these respects proved to be unrealistically short. In all these respects, the Bank underestimated the risks related to FEN'S ability to fulfill its role as a development banking institution and diemissed them too blandly. (See Staff Appraisal Report, para 4.14; President's Report, para 71.) ahan s rin 4.12 Part of the delay in obtaining Loan Committee authorization to proceed with negotiations was due to the time needed to clarify whether the Government's intentions on national electric tariff policy as expressed in  public etatemente which had recently been made were compatible with exieting and propoeed agreement8 with the Bank. (See memorandum dated November 28, 1983, to Chair of Loan Committee.) The balance of the time wae needed to clarify varioue project concepts and detaile. 4.13 a. Bank Policiee for Financial Intemediariee. Some of the queetione raieed related to departure8 from Bank policiee for financial intermediaries, including entitiee epecializing in financing particular eectore, euch as the urban and water eupply and eanitation eectore in Brazil. The experience of theee similar projects ehould have been utilized eyeternatically to structure thie project. The ieeuee which aroee involved the Region'e original propoeal that only eub-loane in exceee of USS10 million would require prior Bank approval, and that FEN'e decision8 for thoee under thie limit would be eubject to review during project eupervieion. To conform to Bank policy, thie was changed eo that all "free limit" sub-loans would be reviewed prior to authorizing diebureement from the loan; and coneietent with Bank practice for newly eetabliehed intermediaries, the firet three "free limit" eub-loane would require prior Bank approval. There wae aleo a difference of opinion between the Region and eenior management ae to the extent to which it would be appropriate to delegate primary reeponeibility for appraieal of sub-projects to the untried FEN. Thie wae reeolved by reducing the "free limit" to Us$4 million. 4.14 The Region'e revieed propoeale for proceeding with negotiations, as set forth in ite memorandum of December 14, 1983, to the acting Chair of the Loan Committee and approved by him, detailed the baeie on which FEN would appraise eubprojecte in the 1983-87 power inveetment program which had an estimated coet of over USS4 million and which had not been appraieed by the Bank or IDB. For euch eubprojecte which had an eetimated cost of over USS40 million, FEN would undertake their appraieal with technical aeeietance provided, as the Bank deemed warranted, either by coneultante acceptable to the Bank and financed under the "Aw loan, or through Bank etaff participation in the FEN appraieal. For euch eubprojecte which coet between USS4-40 million, the Bank, aa it deemed warranted, would review FEN appraieale in the field. These propoeals did not eurvive the negotiatione. Although there is no specific record on thie matter, preeumably they were dropped becauee the Colombian authoritiee were intereeted in ueing the funde provided by the project predominantly, if not eolely, for financing the completion of ongoing works already appraieed, and in that context it wae not coneidered necessary to define how FEN would appraise new worke. 4.15 The mattere described in the preceding two paragraphs have been mentioned because they illuetrate two ehortcominge of the Bank's work during project preparation and deeign. The firet relatee to the problems referred to in para 4.13, which might have been avoided if the Bank team working on the project had included a etaff member knowledgeable about financial intermediariee and the Bank'e practice8 and policies for euch entities. such a etaff member, who ideally would also have been familiar with financial markets, would have participated in the technical aeeietance missions in 1982 which led to the formulation of thie project and would have taken the lead on all the important aspects of the project relevant to FEN'e role as a financial intermediary and its tapping of local capital markete. Inetead, the Region had to rely on the employment of coneultante for these tasks with only very limited aeeietance from the technical division responsible for industrial  finance in the form of occaeional comment8 on draft reporte eent to it. Had such a staff member been aeeigned to the project team, reliance on coneultante could have been reduced and the effectiveness of the Bank'e work during design, appraieal and mupervision would have been improved. The eecond ehortcoming relatee to the divergence in underetanding between the Bank and the Colombian authorities, which surprieingly had developed at this late stage of project design, ae to the use of the project loane for purpoeee other than financing high priority ongoing worke. Thie point is further illustrated in the diecussion of the ieeuee which aroee during Loan Committee review related to cost overrun financing. 4.16 b. Exclusion of Proiect Coet Overrun Financinq. Questions were raieed whether the Region'e propoeale were consietent with the Bank'e policy that coet overruns on projecte previouely financed by Bank loane are normally ineligible for supplementary Bank financing. In its reeponee of December 14, 1983, referred to in para 4.14, the Region etated that project coet overrune would not be eligible for Bank financing under the project. The authorization to negotiate was conditioned on the Region being satiefied that the total Bank financing, under the "A" loan and the Bank'e share of the "B" loan, for projecte previously financed by the Bank (other than for intereet during conetruction where the normal criteria for euch financing were met) would not exceed the eetimated shortfall in commercial bank financing. The Region wae aleo requeeted to reaesure itself that with the application of the agreed criteria for eubproject eligibility eet forth in the December 14 memorandum, there would be room for the USS300 million of "expenditure8 potentially financeable with Bank reeourceew referred to in that memo, or at least for the USS2OO million of propoeed Bank loane. Thie queetion was raised because of concern that thie might not be poseible if the etatement in that memorandum, that only four contract8 under exieting Bank financed projecte would be eligible for sub-loans, was correct. 4.17 In a memorandum dated December 16, 1983, to the acting chair of the Loan Committee, the Region provided an analysis of USS323 million of foreign coets which would be eligible for Bank financing under the agreed criteria. Thie included USS66.5 million of unfinanced foreign coete of civil worke for the San Carlos and Playae projecte, limited to an amount equivalent to previously anticipated external financing that did not materialize. It also included USS76.5 million of intereet during conetruction on exieting Bank loane for eix projecte still being implemented. The balance of USS180 million was for new subtranemiseion and distribution worke that had not yet been initiated or appraieed. Thie analyeie covered the yeare 1883-87. All of the civil works coete were eetimated to be incurred in 1983-85, including USS37 million for 1983, some of which may have carried over to the following year. Of the intereet during construction, USS2.6 million was estimated for 1983, US$43 million for 1984-85 and USS30.9 million for 1986-87. Thus, for these two categoriee related to ongoing worke, only USS112 million was for 1983-85 and the balance of USS31 million was for the next two years. Of the amounts eetimated for new eubtranemission and distribution projecte, USSS million wae for 1984, USS70 million for 1985, and USS105 million wae for 1986-87. 4.18 Ae indicated in para 4.2, when the Colombian authorities were informed during negotiation8 of the Bank'e proposals for eligibility criteria for Bank financing and the implicatione this had for the use of a large amount of this financing for new projecte, they requested the Bank to reconeider its  position. Apparently focussing on the 1983-85 time-slice proposed as the project period in the green cover SARI they were concerned that so much of the Bank funds, estimated at approximately US580 million, would have to be assigned to new projects to be initiated during 1984-87. They pointed out that this departed substantially from the original concept of the project oriented towards structuring an emergency financing program for electricity projects in progress which gave priority to continuing work on them and assuring their timely completion. To achieve the original objective, they requested the Bank to allow some use of its funds for overrun financing and to authorize a larger amount of retroactive financing, and the Bank agreed. (See Region's memorandum of January 25, 1984, to chair of Loan Committee, and memorandum to files of February 1, 1984.) 4.19 The project record indicates that when the Region made its green cover submission to the Loan Committee, it did not adequately appreciate the need to explain the exceptional circumetancee which would justify supplemental Bank financing for projects which had been previously financed by the Bank and had experienced cost overrune. (See memorandum dated November 8, 1983, responding to comments that such justification was needed which were made by the Energy Department in a memorandum dated October 27, 1983, baaed on a review of a draft green cover appraisal report.) When this iesue was raieed during Loan Committee review, the Region apparently decided that it would be unneceeeary to involve the Bank in any overrun financing in the expectation that the commercial banks would be willing to finance project coat overruns, including any associated retroactive financing, without limitation. Although this in fact did not prove to be the case, the expectation that the commercial banks might have been willing to do so was not unreasonable. In proposing to extend the project investment program period by two years and include a very substantial amount of new subtransmission and distribution work8 for Bank financing, however, the Region clearly misjudged how the Colombian authoritiem would react to this. In retroepect, it would have been much more appropriate for the Region to have presented the juetification for supplemental Bank financing for the overruns on Mesitas project costs (see para 4.2) when questions about the utilization of Bank funds to finance cost overruns arose during Loan Committee review. In retrospect, aleo, it is eurprising that senior Bank management and project advisors did not call thie issue to the Region's attention at a much earlier stage in the project cycle. 4.20 Most if not all of the time taken to resolve thie ieeue and other matters which arose during Loan Committee review and negotiations was needed in any case to complete the arrangements for the commercial bank financing. Accordingly, there were no serious adverse consequences as a result of these delays. Disbursement Arranaements 4.21 A major share of the responsibility for administering the disbursement procedures to ensure that the loan proceeds were utilized for the intended purposes was assigned to FEN. Under these arrangements, the proceeds of the "A" and "Bn loans were initially deposited into US dollar denominated revolving funds established by FEN in the Central Bank. In turn, disbursements from the special accounts were made for expenditures for goode and services and for interest and other charges eligible for financing under each approved eubloan. Replenishment of the special accounts was made upon  receipt of withdrawal applications by FEN based on certified statements of expenditures. Documentation for these expenditures was retained by FEN and available for Bank review. These procedures applied to the major disbursement category for subloans under which 100% of the amounts disbursed by FEN was financed. The only other categories were for a small amount to be utilized by FEN directly to pay for consulting services needed to carry out a program to strengthen its technical capacity, for which 100% of foreign expenditures and 50% of local expenditures was financed; and for the amount to be paid to the Bank for the front-end fee. The Central Bank effectively paid the interest charges on the loans applicable to the amounts deposited in the special accounts until they were withdrawn to pay for FEN'S disbursements on the sub- loans. 4.22 The Region stated, in its memorandum of December 14, 1983, that disbursements from the apecial accounts "will be based upon disbursement percentages applicable to the items financed under the subloan as well as receipt by FEN from the relevant subborrower of a disbursement application and full documentation of the expenditures." In the context that the financing to be provided under this project would finance foreign costs, thia statement implied that the disbursement percentages would vary and that for some items, particularly for civil works, they would be less than 100%. Presumably, it was intended that, in determining eligible expenditures for financing under this project, FEN would apply the disbursement percentages specified to determine eligible foreign cost financing under previous loan agreements for projects which would receive additional financing under thia loan - e.g. for the Playas Project, 56% of civil works expendituree or 100% of foreign expenditures and 100% of the ex-factory cost of locally manufactured goods for equipment. Such a procedure was not adopted. Inatead, FEN was allowed to disburse subloans for 100% of all costs including civil worka. 4.23 There are other respects in which it would have been advisable to reach specific understandings with FEN concerning itm disbursement procedures, namely steps it would take to ensure that eligible expenditurea were limited to those incurred under contracts for which the required procurement procedures were followed and that the limits mpecified for retroactive financing and for supplemental and overrun financing were observed. There were no such understandings. 4.24 The project record does not explain why FEN'm disbursement of subloans for 100% of civil works costs was accepted as consistent with the intention that the project loans would finance foreign costs, nor why it was considered unnecessary for FEN to adopt disbursement procedures for the purposes suggested in para. 4.23. It appears, however, that the need for prior Bank approval for proposed subloans was deemed to provide an adequate opportunity to check that all theae requirements were met. Thim approach placed undue reliance on the reliability of the estimates made as the basis for proposed subloans. In fact, the actual civil works coets for Playas were less than the eatimates. Only the first disbursement application for withdrawal of funds from the special account for the subloans provided details of the expenditures to be financed, including for goods and services of the particular contract, the number of the bill and the dates and amounts of payments thereunder, and for interest paymenta on Bank loans the dates and the amounts of the payments. This application also specifically identified and provided the total amount of retroactive financing. Thereafter all disbursement applications simply  provided the total amounts to be withdrawn for each subloan, without any supporting details. The quarterly progress reports contained an annex listing the items for which subloan disbursements were made during the quarter (see para 5.33), but this information was not adequate to determine the extent of retroactive financing because for the most part dates of payment were not provided. This annex, however, did identify the contracts involved and thus provided a basis to check that project financing was limited to contracts which met the procurement requirements. Although reliance on prior Bank approval for subloans was acceptable to ensure that the limits on supplemental and overrun financing were observed, a report as to how this actually worked out for the individual subprojects would have been desirable as part of FEN'S Final Report. In addition, it would have been preferable to design disbursement procedures for application by FEN to double check compliance with the other requirements discussed in this para. when actual subloan disbursements were made. 4.25 As elaborated further in discussing project implementation, the design of the reporting requirements for the project, combined with that for the disbursement procedures, was not adequate to confirm that project financing was limited to foreign costs and that the limits on the extent of retroactive, and to a lesser extent supplemental and overrun, financing were adhered to (see para 5.36). 5. PROJECT IMPLEMENTATION Utilization of Proiect Financinq 5.1 The implementation of this loan should have been relatively straightforward with respect to its primary objective, which was oriented to assist the financing of ongoing projects. The funds provided by the Bank and the co-financiers were committed within schedule. Using a comprehensive format approved by the Bank and prepared by FEN during the period between loan signing at end June 1984 and effectiveness at end September 1984, initial subloan requests were promptly prepared and submitted. The proposals were limited to the use of the subloans to meet requirements during 1984 and 1985. The amounts and uses requested corresponded closely to plans made in March 1984 to allocate the project financing among the power companies. Over a period of about three months ending mid-January 1985, five subloans were approved which fully committed all of the Bank financing and all but the USS50 million of commercial bank financing which had tentatively been allocated to ICEL (which as it turned out failed to qualify). Supplementary loans to reallocate these funds to CORELCA, CVC, EPM, and ISA were arranged in November and December 1985, within the specified deadline for submitting subloans for Bank approval. There was some delay in making these arrangements when EEEB was unable to obtain authorization from the Bogota Council to accept an additional subloan proposed by FEN. 5.2 Diecussion of diebursements must distinguish between funds advanced by the Bank and the co-financiers for deposit into the special accounts at the Central Bank and payments from the special accounts for actual expenditures by the power companies on the subprojects financed by the FEN subloans. As anticipated when the loans were made, the full amount of the US$ denominated  commercial bank loane totalling USS175 million, including the Bank's share of USS25 million, was diebureed into the epecial account in the eecond half of 1984. There wae a brief delay in the diebureement of the Japanese yen denominated commercial bank loane totalling Y5.5 billion, including the Bank ehare of Y700 million, which was paid into the apecial account on March 29, 1985. Dieburaement of the Bank loan into the special account wae geared to an initial depoeit of USS50 million, made on October 1, 1984, and eubeequent replenishment equal to payments made out of the epecial account for eligible expendituree. The subsequent replenishment8 were limited to the extent that the balance in the apecial account would not exceed the amount of the initial depoeit (Loan Agreement, Schedule 4, para 3). In addition, as stated in para 4.11 of the appraieal report, after USS100 million of the Bank loan had been diebureed, recuperation of the initial depoeit would begin. On thie basis, the rate of diebureement was eomewhat elower than anticipated at appraieal. Comparsd to the appraieal eatimate that the USS170 million Bank loan would be fully disbursed by year end 1985, the amount then diebureed was USS159 million, or 93% of the total loan. Of the total amount then diebureed, however, USS48 million waa held in the epecial account a8 advancee. No additional diebureemente were made in 1986. The remaining USS11 million was diebureed in two inetallmente in February and April 1987. The final payment of USS5.3 million repreaented a final advance to the epecial account requeeted by FEN to facilitate it8 utilization for diebureemente on eubloane. (Part 111, Table 3.1 and page 27, Evaluation Final de la Ejecucion del Contrato del Empreetito FEN-BIRF (2401) - Banca Comercial, prepared by FEN, hereafter referred to as FEN'e Final Report). 5.3 Disbureement of the eubloans to the utilitiee to finance actual expendituree on eubprojecta occurred at a much elower pace than the rate at which diebureemente were made into the epecial account8 (Part 111, Table 3.2 and FEN'e Final Report, Cuadro No. 2). Meaaured on thie baaie, by year end 1985, only USS111 million, or 65%, of the Bank'e USS170 million "A" loan, and only USS270 million, or 73% of the total USS370 million of loan funde provided by the Bank and the commercial banke, had been eo utilized. By year end 1986, USS154 million, or 90%, of the "A" loan, and USS334 million, or 90%, of the total loane were utilized. By year end 1987, all of the "A" loan and all but US$800,000 of the commercial bank loan8 had been utilized. The laet diebureement of USS5.0 million to EPM, on eubloana financed by the "A" loan, was made on December 1, 1987 (FEN'e Quarterly Report for the fourth quarter of 1987). The Bank's USS28.2 million ehare of the commercial bank loane was fully diebursed on aubloana by September 30, 1985. CVC was the only one of the eubborrowere to uee it8 portion of the allocated funde by the end of 1985. Each of the other four power companiee shared in the USS64 million of eubloan diebursements made in 1986 and the USS36 million made in 1987. The final diebursemente to clear the advancee made from the US$ denominated commercial bank loane, in the amounts of US$300,000 to EPM and USS500,OOO to COReLCA, were made in April and July 1988, reepectively (FEN'e Final Report, page 30). 5.4 Eetimatee of the rate at which eubloan diebureemente would be made are not presented in the appraieal report, and apparently none were ever made. Thie wae an omieeion of an eeeential part of the appraieal. Implicitly, because the final date for eubmitting eubloan applications was December 31, 1985, and the cloeing date for the Bank loan wae eet at December 31, 1986, eome delay was anticipated between the advance of funde to the epecial accounts and their utilization for subloans. Under the procedure for  recuperation of the advances to the special account after US$100 million of the "Am loan had been diebursed, subsequent disbursements were to be made only after and to the extent that the Bank was satisfied that all amounts remaining on deposit in the special account had been or would be utilized in making payments for eligible expenditures (Loan Agreement, Schedule 4, para 5). Accordingly, to accomplish full disbursement of the "A" loan by the end of 1985, it is implicit that it was expected that the amount of subloan disbursements then outstanding would be minor. Another relevant point is the provision in the subloan agreements, approved by the Bank, which allowed 20 months from the date thereof as the period for completing disbursements thereunder. For the first five subloan agreements covering all but US$50 million of the project loan funds, this required complete disbursements of the subloans between June and September 1986; for the remaining four subloans, which did not involve the "An loan, this required complete disbursement by March 1987. In fact, only CVC and EEEB met these deadlines. On balance, when the project loans were made, it is unlikely that it was expected that: (i) at year end 1985 there would be as much as USS59 million, or 35% of the USS170 million "A" loan, plus US$31 million, or 16% of the US$200 million of "B" loans remaining to be disbursed on subloans; and (ii) it would take another 12 months after 1986, or even more in the caee of the "B" loans, to complete.the disbursement of the final 10% of the loans. 5.5 The Bank did not become aware of the need for an extension of the closing date until late in 1986 when, by letter dated October 29, 1986, FEN requested an extension to December 31, 1987. Prior to that, the October 1985 and August 1986 supervision missions reported that disbursements would be completed by mid or end 1986, respectively. The October 1985 report attributed the disbursement delays to delays in implementing some of the subprojects and neither report reflected any serious problems in this connection. Considering the amounte remaining to be disbursed on subloans and the pace at which subloan disbursementa had been occurring as of the time of the October 1985 mission, the target date of mid 1986 for completing disbursements reported by it should have been subject to more critical assessment. This goal then required average subloan disbursements of US$50 million for three subsequent quarters, compared to the USS22 million rate for the two previous quarters. In addition, EPM'S relatively low utilization rate of subloans should have been noticed (see para 5.35) and resulted in questions which also would have raised doubts about the likelihood of completing disbursements by mid-1986. 5.6 As of the August 1986 mission, average disbursements of US$25 million for two subsequent quarters were required to complete disbursements by year end, a rate that might well have been deemed reasonable in comparison to the rates attained since the first quarter of 1985. Beginning with the quarter ending September 30 1986, however, there was a reduction in the level and a much more erratic pattern of subloan disbursement than previously prevailed. This should not necessarily have been a surprise, since it appears that the August 1986 supervision mission also overlooked an opportunity to suggest that further analysis was needed on this subject. That mission's report included a statement of subloan utilization as of June 30, 1986, which showed that EPM had made least use of its subloans: only USS24 million, or 40%, of the USS61 million allocated to it had been disbursed, compared to at least 90% of the subloans to the other four power companies. (Mission report dated September 16, 1986, Section 1, para 4.) This disparity should have led the mission to  inquire into the reasons why EPM was making such slow u t i l i z a t i o n of i t e subloans, and t o suggest t h a t FEN should ascertain from EPM when the remaining balance of USS37 million would be claimed and make similar inquiries of t h e other power companies with unclaimed balances a s of t h a t t i m e . Had t h i s been done, the mission might not have concluded so confidently that: "Loan disbursements a r e proceeding satisfactorily. FEN is making arrangements so t h a t by year end there w i l l not be any balance pending disbursement." (Ibid., para 5) 5.7 The reasons cited by FEN t o explain t h e delays i n subloan disbursements were documentation problems experienced by t h e u t i l i t i e s i n t h e i r payments t o contractors and insufficient local currency funds t o execute subprojects a t t h e pace originally planned. It is also relevant t o note that material amounts of subloan disbursements t o finance interest payments during constrl~ctionextended i n t o 1986 and 1987. This was accomplished a f t e r t h e reallocation t o t h i s category of amounts designated i n the original subloan applications t o finance c i v i l works and equipment expenditures and of amounts included therein t o be specified later. This need t o complete the disbursement of t h e subloans through extended financing of interest during construction, is evidence, i n addition t o t h e disbursements made for expenditures incurred i n 1986 and 1987 for c i v i l works and equipment, t h a t t h e projections of the construction programs and financing requirements of t h e power companies for t h e 1984/1985 project t i m e s l i c e made a t appraisal by t h e Colombian authorities and the Bank w e r e overestimated. The slower pace of construction than forecast is the primary reason for t h e lag i n subloan disbursements, which i n amount and extent beyond December 31, 1985, were significantly greater than expected when t h e loan was approved. I f , however, judgment is limited t o t h e lag of about 12 months beyond t h e original closing date, December 31, 1986, which involved only about 10%of project financing, t h e lag may be regarded a s of minor importance. 5.8 Implementation of most of the components of t h e project designed t o strengthen t h e newly created FEN'S capabilities was delayed, and for the most part the impact of these e f f o r t s has been insubstantial. 5.9 3. FEN was about one year l a t e i n obtaining the consulting assistance provided under t h e project t o strengthen its technical capacity, including a study of its technical staffing requirements t o evaluate loan proposals and t o monitor project implementation. The consultant was retained i n July 1985 rather than by June 30, 1984, and t h e study was completed i n September 1985 instead of by December 31, 1984, a s agreed (Loan Agreement, Section 3.01). The Bank reviewed the study report promptly and endoreed its recommendations. The additional technical s t a f f recommended t o meet FEN'S immediate requirements, two well qualified power engineers, was not hired u n t i l 1986, rather than 1985 ae contemplated when t h e loan was made. Other organizational recommendatione contained i n t h e report w e r e implemented. Only about US$12,000 of the USS80,OOO provided under t h e Bank loan t o finance t h i s activity was utilized. In t h i s connection, it is relevant t o note t h a t a l l subloans under the project w e r e f o r ongoing projects, and FEN never used t h e authority it had under t h e project t o appraise independently subprojects for which its loan would not exceed US$4 million (free-limit subloans; see para 3.7). In addition, during  t h e project implementation period (through 1987) and subsequently, it appears t h a t FEN has done very little i f any independent appraisal work with respect t o loans by FEN t o t h e power companies from local o r other external sources. Instead, FEN'S judgments concerning these loans have been baeed primarily, i f not exclueively, on evaluation reports prepared by its borrowers. The one exception has been a USS80 million loan from IDB f o r e l e c t r i c i t y l o s s reductions made i n 1987, under which FEN is t o evaluate eubprojects submitted by t h e power companiee applying c r i t e r i a which meet t h e appraisal standards of IDB. Through year end 1990, however, only USS3.3 million of diebureemente had been made under t h i s loan. 5.10 b. Enhancement of Financial Plannina and Advieorv Ca~acitv. A eecond set of measures t o improve FEN'S technical capacity included an informal undertaking to complete an e f f o r t already begun, using FEN'S own resources, t o deeign FEN8e operating procedures, determine its computer requirements and i n s t a i l computer f a c i l i t i e s compatible with those of t h e power companies as soon a s possible. The objective was t o meet FEN'e requirements, a s w e l l a s those of t h e power companies individually and t h e eector a s a whole, f o r monitoring h i s t o r i c a l financial performance and preparing projected financial etatements needed f o r financial planning (SARI para 2.05). The work on t h i s eecond undertaking was carried out with t h e aesistance of coneultante who w e r e a l s o retained by FEN, ueing i t m own resources, eimultaneously t o carry out a study on sector finances and formulate a proposed financing strategy coneidering alternative scenarios. FEN wae formally committed under t h e project t o complete t h i s study by September 15, 1984, and promptly thereafter diecues it with t h e Government, ISA and t h e Bank. Subsequent t o t h e f i r s t exchange of viewe baeed on t h i s etudy, there were t o be eimilar annual discussions of t h e adequacy of t h e existing financing etrategy f o r t h e power eector and t h e necessary revision8 thereto, not l a t e r than Auguet 31 i n each year beginning i n 1985 (Loan Agreement, Section 4.13 and Guarantee Agreement, Section 3,05) . ISA agreed to cooperate with FEN f u l l y and actively so t h a t FEN could f u l f i l l its obligation8 t o carry out t h e etudy, and t o p a r t i c i p a t e i n t h e annual exchanges of views (Power Financing Agreement, Section 5.01). I n r e l a t e d undertakings, ISA agreed t o prepare by April 30 of each year, and t h e Government agreed t o review and furnish t o t h e Bank f o r its comments by September 30 of each year, a study on t h e updated e l e c t r i c i t y demand growth projectione and t h e related power investment program (Power Financing Agreement, Section 5.03, and Guarantee Agreement, Section 3.13). These were t o serve ae inpute t o t h e annual review of eector financing strategy. 5.11 D i f f i c u l t i e s which should have been foreeeen w e r e encountered i n designing and i n e t a l l i n g computer f a c i l i t i e e which would meet FEN'e requirements and be compatible with those of t h e power companiee. Since a uniform eyetem of accounte had not been prescribed, t h e power u t i l i t i e s did not keep t h e i r accounts and report t h e i r reeulte on a comparable basis. The eubeidiariee of ICEL and CORELCA had poor accounting eyetems, and there w e r e long delaye i n obtaining r e l i a b l e data from them. The Bank was w e l l aware of t h e d i f f i c u l t i e s ISA experienced i n providing consolidated financial statements f o r t h e sector. Moreover, t h e expectation t h a t t h e power financing study would be completed by September 15, 1984, was u n r e a l i s t i c a l l y short, ae indicated i n para 4.11. The coneultante delivered a d r a f t of t h e report t o FEN on Auguet 15, 1985. After extensive review within Colombia, t h e d r a f t was forwarded t o t h e Bank on January 15, 1986. The Bank responded on February 5, 1986, i n a letter which contained extensive and detailed comments, and  extended the deadline for submitting the revised and f i n a l report t o September 15, 1986. The Bank followed up i n April and September 1986. In a telex dated September 24, 1986, FEN informed t h e Bank t h a t the revision of the study was being prepared by a special working group established by the Minister of Mines and Energy, and it propoeed t h a t t h i s revision be presented t o a forthcoming preparatory mission from t h e Bank for a proposed sector loan ae the study referred t o i n Section 4.13 of the Loan Agreement. In response, the Bank agreed t h a t for 1986 t h e review of power sector finances would be accomplished during appraisal of the sector loan. 5.12 The scheme for an annual exchange of views among the Government, FEN, ISA and the Bank, described i n para 5.10, has not been carried out a s such. Instead, the Bank's supervision of the FEN loan merged with e f f o r t s t o prepare the Power Sector Adjustment Loan (Loan No. 2889 CO) approved i n December 1987. Exchanges of views between the Bank and Colombian authorities have been carried out i n the context of the preparation and supervision of the Sector loan and subsequent power sector and project preparation activities. FEN has not played t h e leading role envisaged for it under the project a s power sector financing strategy advisor t o the Government and the u t i l i t i e s . 5.13 3 s . The project included two sets of measures concerning FEN's mobilization of local savings t o be relent t o t h e power companies. First, i n order t o have a better match between the maturities of the securities issued t o r a i s e funds from the domestic capital market and those of the local currency loans made t o the power companies, it was agreed t h a t , by December 31, 1984, FEN would prepare a plan of action acceptable t o the Government and the Bank t o encourage private investment i n medium and long term obligations issued by FEN, and t h a t the Government would take a l l actions neceesary t o enable FEN promptly t o put t h a t plan into effect (Loan Agreement, Section 4.11, and Guarantee Agreement, Section 3.03). Second, i n order t o ensure t h a t FEN raised an appropriate share of the incremental resources expected t o be available i n the domestic capital market, annual targets of the minimum amount of funds FEN would r a i s e from local investors w e r e agreed for each of the years 1984 through 1987; these targets were t o be reviewed and updated, a s necessary, by agreement among FEN, the Government and the Bank a s part of an annual review of FEN's borrowing and lending operations which would take into account the power sector's financing requirements (Loan Agreement, Section 4.12, and Guarantee Agreement, Section 3.04). 5.14 The appraisal report recognized that, given the strong orientation of Colombia investors toward short term paper, it would take t i m e t o a t t r a c t medium and long t e r m funds. It also recognized the practical difficulties involved i n extending maturities, such a s the lack (then) of a free market index of short term interest r a t e s which would be necessary i f t e r m bonde with interest r a t e s adjustable quarterly based on an index w e r e t o be acceptable. The f a c t t h a t t a x regulations w e r e unfavorable t o t h e issuance of medium and long term paper was another problem which the report mentioned would have t o be resolved (SAR, paras 2.11-2.16). It is surprising, therefore, that a target date a s early a s December 31, 1984, was set for submitting and promptly implementing a plan of action t o extend the maturities of FEN's local borrowinge. This proved t o be unrealistic.  5.15 To comply with its obligation t o prepare t h e plan of action, FEN retained consultants who submitted a report, dated February 28, 1985, which analyzes and recommends several alternatives which FEN should consider a s sources f o r r a i s i n g medium and long term financing. Subsequently, as explained i n para 5.24, t h e Bank retained an expert t o review FEN's r o l e a s a development banking institution. H i s report, dated October 14, 1985, i n t e r a l i a , reviews FEN's capacity t o mobilize funds from t h e domestic c a p i t a l market and a l s o analyzes options available t o FEN t o issue longer term securities. Copies of these reports w e r e exchanged between t h e Bank and FEN, and t h e i r contents w e r e apparently discussed i n t h e f i e l d during t h e Bank's preparatory work on t h e Power Sector Adjustment Loan. 5.16 No attempt was made t o implement any of t h e alternatives recommended t o r a i s e medium and long term funds principally because of t h e unwillingnese of domestic banks t o assume primary l i a b i l i t y f o r loans t o most of t h e power companies i n Colombia whose financial condition was deemed by them t o be unsatisfactory. The intermediation of t h e commercial banks through rediscounting with, o r d i r e c t guarantee t o , FEN was required as a condition f o r FEN's loans t o t h e power companies. A s a r e s u l t of t h i s impediment t o its local currency lending operations, FEN was unable t o expand t h e volume of its loans t o t h e power companies a t t h e r a t e anticipated during appraisal, and correspondingly it was unnecessary f o r FEN t o expand its domestic borrowings t o m e e t t h e agreed targets. 5.17 Actual l o c a l currency borrowings by FEN w e r e substantially less than t h e estimates made a t appraisal (Part 111, Table 4.1). A s of year end 1987, t h e outstanding amount of such borrowings was only 25% of t h e appraisal estimate. Both t h e October 1985 and August 1986 supervision missions pointed out t h e reasons why it had not been possible for FEN t o meet t h e specified borrowing t a r g e t s (see para 5.16). The former recommended t h a t t h e Bank accept t h e reduced amount raised for 1984 a s compliance and t h i s was confirmed i n a follow up telex. The l a t t e r reported t h a t t h e amount borrowed locally i n 1985 w a s much l e s s than had been agreed, and t h a t FEN had requested a modification of Section 4.12 ( a ) of t h e Loan Agreement which would substantially reduce t h e amounts t o be raised locally. It recommended that coneideration of t h i s request be delayed u n t i l negotiation of t h e power sector loan then under preparation, s o t h a t t h e decision would take i n t o account t h e neede of t h e power sector f o r such financing a s evaluated during t h e appraieal of t h a t loan. (Supervision Report dated September 16, 1986, Sections I1 and 111). The Banks's follow up telex contained this proposal, which FEN accepted. (Bank t e l e x t o FEN dated September 19, 1986; reply t e l e x dated September 24, 1986). Since then, a s i n t h e case of t h e annual review of power sector financing strategy, t h e annual review of t h e adequacy of FEN'S lending and borrowing operations has been carried out a s part of t h e work related t o t h e Power Sector Adjustment Loan and subsequent a c t i v i t i e s (see para 5.12), instead of i n accordance with t h e express terms of t h e loan documents, i n t h i s case Section 4.12 of t h e Loan Agreement. 5.18 The August 1986 supervision mission was carried out by t h e banking expert referred t o i n para 5.15. Although he noted t h a t it made no sense t o borrow funds t h a t cannot be relent, he was a l s o concerned t h a t FEN had f a i l e d t o achieve its objective t o a s s i s t i n meeting t h e financing needs of t h e power sector by r a i s i n g funds from domeetic financial markets. I n h i s opinion, FEN could increase t h e amount of its local borrowings substantially without  crowding out other participant6 i n theee markete. H e recommended t h a t t h e reluctance of t h e local financing inetitutione t o lend t o t h e financially weak power companiee could be overcome i f legielation w e r e paeeed which would allow them t o charge penaltiee f o r arreare i n payment6 by t h e power companiee. H e eetimated t h a t t h i e , plue other etepe needed t o improve t h e financial condition of t h e power companiee and eetablieh t h e i r earning6 at reaeonable levels, would make it poeeible f o r FEN t o increaee local borrowing6 eo t h a t by year end 1988 they would reach t h e nominal amount eetimated at appraieal t o be attained by year end 1987. Thie wae about two yeare eooner than wae forecaet i n t h e financial projections prepared by FEN and reviewed by t h e Auguet 1986 mieeion. I n actuality, and without enactment of t h e propoeed legielation and with a continuation of financial probleme f o r many of t h e power companiee, t h e nominal amount of Co1$63,000 million of domeetic borrowing6 forecaet a t appraieal f o r 1987 wae not reached u n t i l 1990. I n real terme, of couree, t h e amount of funde raieed by FEN from t h e domeetic financial market through t h e end of 1990 wae e t i l l w e l l ehort of t h e appraieal eetimate f o r 1987. pomeetic Lendinq 5.19 FEN'S lending t o t h e power companiee from domeetic funde hae aleo followed a pattern eimilar t o t h a t of i t e domeetic currency borrowings (Part 111, Table 4.2). The actual amount of euch loane ae of t h e end of 1987 wae 48% of t h e nominal Co1$82,000 million forecaet f o r t h a t date a t appraieal, and t h a t nominal level of lending wae not attained u n t i l t h r e e yeare l a t e r i n 1990. Again, i n r e a l terme t h e amount of euch lending continue6 t o be w e l l under t h e appraieal eetimatee. 5.20 The i n a b i l i t y of FEN t o m e e t t h e domeetic borrowing and lending targete agreed o r projected a t appraieal wae due primarily t o t h e deterioration i n t h e financial eituation of many of t h e power companiee which wae not then reaeonably foreeeeable. Under t h e prevailing circumetancee, FEN accompliehed ae much ae could be expected i n relation t o t h e magnitude of i t e domeetic borrowing and lending programe. Similarly, t h e Bank ehould not be faulted f o r having f a i l e d t o preeeure FEN t o do more i n theee reepecte. The Bank appropriately turned i t e attention t o an attempt t o correct t h e underlying probleme which caueed t h e poor financial performance of t h e power eector through t h e Power Sector Adjuetment Loan. It i e the Government which bears ultimate responsibility f o r t h e emergence and pereietence of deteriorated financial condition6 i n t h e power eector. 5.21 FEN has made eignificant progreee i n extending t h e maturity of its domeetic borrowings which ehould be acknowledged. I t e i n i t i a l borrowings w e r e through eale of 90 day and 180 day E l e c t r i c i t y Certificates (CEV) eold a t a diecount which reeulted i n an effective i n t e r e e t r a t e which was eubetantially positive i n r e l a t i o n t o t h e prevailing i n f l a t i o n rate. These had a good reception and FEN w a s able t o drop t h e 90 day paper and o f f e r t h e 180 day c e r t i f i c a t e s with t h e option of one renewal f o r a further 180 days. Beginning i n 1985, it hae aleo ieeued two year Energy C e r t i f i c a t e s (TER). One fourth of t h e face valuee of these c e r t i f i c a t e s i e redeemable a t half yearly intervals. The intereet payable on each coupon payment increaeee with t h e maturity. The inveetor hae t h e option of postponing redemption t o subsequent half year pointe i n order t o take advantage of t h e higher effective i n t e r e s t rate. The ecale of i n t e r e s t r a t e e i e predetermined on a baeis expected t o be eignificantly poeitive i n relation t o inflation. By 1986, 44% of FEN'S t o t a l  domestic borrowings were raised through the TERs. The relative importance of the TERs increased in each subsequent year, rising to 93% in 1989. In that year, FEN also began issuing a four year version of the TER. Reflecting recent developments in the domestic financial markets, the interest rate payable on each yearly coupon is related to the average effective interest rate, known as DTF, paid in Colombia on certificates of deposit. The interest rate on the first coupon is equal to DTF; for the second DTF plus 3%; for the third DTF plus 4%; and for the fourth, DTF plus 5%. As of year end 1990, the relative importance of TERs had declined to 53% of FEN'S total domestic borrowings. Additional information about FEN'S domestic borrowing and lending activities in 1990 and 1991 is presented in paras 5.56 -5.61. Bank Supervision Activities 5.22 Initially, the Bank closely monitored project execution and FEN operations. As time passed and it became increasingly plain that FEN'S role in resolving the power sector's problems would be of minor importance, the Bank gave priority to efforts to prepare what became the Power Sector Adjustment Loan approved in December 1987, to that loan's subsequent supervision, and since 1990 to preparation of a possible sector loan. In this context, the attention given to supervision of this project diminished substantially well before the loan was fully disbursed. As previously indicated in this section 5 and elaborated further below, overall, the supervision effort was ineffective. The discussion under this heading will be related to: (a) supervision missions; and (b) quarterly reports on project progress. 5.23 a. Supervision Missions. There were four supervision missions: September 1984, March 1985, October 1985 and August 1986. The first two missions were carried out by the project officer, an experienced senior power engineer who was able to handle the entire mission by himself very competently even though it would have been preferable at least for the second mission had he been accompanied by a financial analyst and/or an expert on financial intermediaries. After the retirement of that engineer, he was succeeded as projects officer by another senior power engineer who with a senior financial analyst, also new to the project, carried out the third mission. The fourth mission was performed by the banking expert mentioned in para 5.15. He should have been accompanied by a staff member. All of the reports rated the project status as satisfactory and without serious problems. 5.24 One of the purposes of the second supervision mission was to arrive at a model of the type of quarterly progress report to be submitted. This was accomplished during that mission, with the exception of the financial information for FEN, including the key financial indicators to be monitored. (See paras 5.32 -5.39 for a further discussion of the quarterly reports). It was agreed that this gap would be resolved with the assistance of a banking expert who would participate in the next mission. The mission report does not explain why agreement on periodic financial reporting requirements was deferred. This subject had been covered in the letter which the Bank had previously sent to detail reporting requirements (see para 5.32) including, as part of the annual financial information expected from FEN, key financial ratios to be monitored which were identical to those contained in Annex 2.13 of the Staff Appraisal Report. Indeed, the supervision report included a comparison of the forecast and actual ratios for 1984 using these indicators.  This proved to be the last time such monitoring occurred, however; as explained in the next para, no recommendations on this subject resulted from the visit of the banking expert. The mission report emphasized that the participation of a banking or development finance company expert was also necessary to provide advice on a request from FEN for modification of the formula which determines the minimum liquidity requirement it must maintain (specified in Section 4.06 of the Loan Agreement); for the exchange of views from time to time between the Bank and FEN concerning FEN'S administration, operations and financial condition (Loan Agreement, Section 4.09); for the annual exchange of views among the Government, FEN and the Bank on FEN'S borrowing and lending operations (Loan Agreement, Section 4.12); and to provide advice on questions the mission raised about the basis for FEN'S financial projections reviewed during the mission. Prior to the second mission, in a telex dated March 5, 1985, responding to the request for amendment to the liquidity requirements, the Bank had informed FEN of its intention to send someone with a background in development finance, banking and power utility financing to review FEN's role as development banking institution who, inter alia, would provide advice on that request. 5.25 The consultant retained by the Bank for the purposes indicated in the preceding paragraph visited Colombia in September 1985 and issued a report dated October 14, 1985. The item in the terms of reference, dated Sept. 9, 1985, concerning reporting requirements was related to the liquidity issue: "(e) Discuss with FEN's management its proposal to improve the liquidity approach ... as well as other monitoring indicators for reporting requirement purposes." In this context, after reporting that FEN had decided to withdraw its request to modify the liquidity formula, except for a minor clarification, the consultant concluded that "There appear to be no other problems with regard to monitoring clauses on covenants." (Document # 299.342, English translation of consultant's report, page 25). Bank staff failed to notice that the terms of reference had not adequately focussed the consultant's attention on the Bank's desire to have his advice on the information, including key monitoring indicators, concerning FEN's financial performance which should be included in periodic progress reports. This matter was never again raised. Paras 5.28- 5.29 contains additional comments on the Bank's limited monitoring of FEN'S financial performance. 5.26 The second supervision mission recommended that FEN should analyze the possibility of extending the maturity of its loans (from local sources) to the power companies from five years to seven years, which would be more in line with the sector's needs. This suggestion, adopted by FEN as of June 13, 1985, represents an accomplishment of modest significance which should be credited to that mission. 5.27 The third su~ervisionmission, like the preceding two, reviewed in detail the status of compliance with all of the covenants in the loan documents for this project. In the aide memoire summarizing the findings of the third mission which was discussed and left with FEN, it reported some deficiencies in complying with covenants, including: CVC's delinquency in interest payments to FEN (which was subsequently corrected); Corelca's failure to earn the specified rate of return in 1984; and Government's failure to  submit by the agreed time a price index for the sector to be used for revaluation of assets. The mission informed FEN that the Bank would bring these matters to Government's attention. It also reported the substantial shortfall in local borrowings by FEN in 1984, and stated that it would recommend that this be accepted because as explained by FEN it was unavoidable. The Bank aent a telex to FEN confirming the mission's findings a8 set forth in the aide memoire. A draft telex to the Government, however, which the supervision report recommended to call attention to the shortcomings of CVC and CORELCA and the delay in proposing a sector price index, and request that remedial action be taken, was not sent. There is no explanation in the project files why this telex was not aent; presumably it was just an oversight. 5.28 The fourth su~ervisionmission reviewed FEN'S financial performance for the period 1983-85 in comparison to the forecasts made at appraisal. It concluded that FEN'S financial results for 1985 and the two previous yeara were satisfactory. The rate of return on equity of 21% in 1985 and 1984 and 22% in 1983 was compared with yearly inflation ratea of 16% for 1985, 18% for 1984 and 17% for 1983 and thus indicated to be about four percentage points positive. Although a comparison in this respect with what had been assumed at appraisal was not presented, it would have shown that the appraisal report understated the 1983 performance and had estimated rates of return on equity of 20% and 21% for 1984 and 1985, compared to estimated inflation rates of 22% and 20%, respectively (Part 111, Table 5.1). The mission was evidently misinformed on the average yearly inflation ratea for these yeara which, as officially reported baaed on the CPI, were 20% in 1983, 16% in 1984 and 24% in 1985. On this basis, the performance for 1985 fell somewhat short of satisfactory. 5.29 This mission also reviewed the financial projections for 1986-95 which FEN had prepared (as required by Section 4.02 (b) (ii) of the Loan Agreement.) It characterized the underlying assumptions as "fairly conservativen and concluded that "The financial development of FEN is expected to proceed smoothlyn. It did not present an analysis of projected annual rates of return on equity. Had it done so, this would have shown that they were forecast to be about 20-22% in 1986-89 and 18% thereafter, and that these returns were about equal to the eetimated yearly inflation rates. Although it was appropriate at the time to accept these forecasts as reasonable, FEN'S actual performance in this respect has proved to be unfavorable (see para 5.46). 5.30 The mission'.^ terms of reference called for it to report on FEN's organization and management, including "the changes that have recently taken place through the incorporation of additional ataff...an assessment of the technical and financial evaluation of projects being carried-out by FEN and of the suitability of their staffing, training and methods...(and) an assessment of FEN'S ability to monitor and evaluate the finances of FEN's borrowers..." (Memo., July 18, 1986). The mission's response to this (Supervision Report dated September 16, 1986, Section IV and companion report, dated September 18, 1986, aeeeaaing FEN's potential role under the then proposed power sector loan, section on "Organization and Administration") was perfunctory: a description of FEN'S overall organization; a description of the reorganization of the Technical Vice Presidency and a reference to the recruitment of additional staff for this unit as recommended by the study described in para 5.9; a list of the thirteen professionals and their qualifications assigned to  thie Vice Preeidency; a judgment that FEN ia well prepared to undertake the activities for which it wae created, qualified by the etatement that it ie too early to aeeeee the efficiency of the technical etaff because they have not yet had much opportunity to demonetrate their capabilities; and a recommendation that future mieeione ahould aeaeee the performance of thie etaff and what additional measures may be needed, such ae training or recruiting additional pereonnel. Thie inadequate treatment demonetrates that it was a eerioua mietake on the part of the Bank to aeeign exclueive reeponeibility for thie eupervieion mieeion to a coneultant banking expert. While thie coneultant wae well qualified to deal with development banking and financial market ieeuee, he was lacking in expertiee to aeeeee FEN'a capabilitiee to evaluate and monitor electricity projecte, and perhaps even to aeaeee it8 ability to monitor the finances of the power companiee who were FEN'a borrowere, a eubject which he failed to addreee. Had a member of the Bank'e etaff experienced with power project8 aleo participated in thie mieeion, thie omieeion would probably not have occurred and it ie likely that a more eearching analyeie of the organization and management ieeuee would have been made. Although the reorganization of, and modeet expaneion of etaff aeeigned to, the Technical Vice Preeidency had only recently occurred, it ehould have been poeeible to arrive at preliminary judgments ae to how theee recommended changes were working out and to decide whether it wae appropriate then to recommend further use be made of the eervicea of the coneultant reeponeible for the recommendatione to aeeiet in their implementation. 5.31 The coneultant'e limited fulfillment of one other task specified in hie term8 of reference lead8 to a eimilar conclusion. Thie relate8 to the requirement that hie "report on compliance with loan condition8 ehould update the one preeented in the November 11, 1985, eupervieion report and ehould include your recommendations, if you identify a eituation of non-compliance." Hie report, however, wae limited to a review of the conditione contained in the Loan Agreement and omitted any reference to thoae in the Power Financing and Guaranty Agreement8 which had aleo been covered in each of the previoue eupervieion mieaione. Thue he failed to review with FEN the etatue of compliance by the power companiee with the revenue covenante and what action FEN and the Government were taking where there wae non-compliance. Nor did he follow up on the non-fulfillment by the Government of its obligations to prepare and publieh a price index for the power eector. Again, it is unlikely that theee omieeiona would have occurred if a member of the Bank'e etaff had aleo been assigned to the eupervieion mieeion. 5.32 .O ua te r . The Bank's requirements for periodic report8 were originally detailed in a letter to FEN dated July 13, 1984, which aleo eet forth the Bank'e viewe on eubproject information to be included in eubloan applicatione. As propoeed in thie letter, FEN would have included in each report information on the etatue of execution of each eubproject it8 eubloane were financing, including a diecueeion of problems encountered and eolutione proposed for dealing with them. Thie euggeetion waa reiterated in the Bank'e comment8 on the firat quarterly report submitted by FEN covering the period ending Dec. 31, 1984 (Telex dated Feb. 22, 1985). In reply, FEN stated it had not contemplated that it would be neceeeary for FEN, in ite capacity a8 a financial intermediary, to follow directly and report quarterly the progreee in implementing projecte partially financed by the "A" and "B" loans (Telex dated Feb. 28, 1985). Thie position was accepted by the Bank. The project record contain8 no explanation for thie change of view.  Presumably the Bank recognized that it would be premature to place this burden on the newly created FEN and that the status of implementation of project8 directly financed by Bank loans would be reported to it in the quarterly reports provided by the power companies involved. 5.33 As agreed with the Bank, the quarterly report8 contained a main text which provided information on the status of: (i) diebursement of the "A" and "Bn loans into the special accounts at the Central Bank; (ii) contracting of subloans including sources of funds allocated thereunder; (iii) subloan disbursements; and (iv) balances in the special accounts. Also included was a detailed schedule of ~roiectcosts which implicitly defined costs as the amounte to be provided by FEN ae eubloans and distinguished between foreign and local costs in terms of the currency of the expenditures so financed. It contained a break down for each subloan of the amounts of financing allocated from the "A" and "B" loans to particular projects (8.g. San Carlos or Playas) and works (e.g. a named transmission line or substation), and these amounts were also broken down to show foreign and local expenditures separately. This data was presented separately by source of financing: Bank financing from the "An loan, the Bank's share of each of the "B" loans, and each of the "B" loans excluding the Bank's ehare. One eet of columns showed the project cost as originally established when the subloans were contracted, and a second set as last revised to reflect the changes in allocations proposed by FEN and agreed by the Bank during project implementation. A final set of columns showed the total amount8 diebureed to date for each line item by source of financing but without distinguishing whether the expenditures actually so financed were foreign or local. The reports aleo included annexes listina the items for fandwhichwere during the quarter financed from the 'An loan and the Bank's share of the "B" loans; this listing identified, for civil works, equipment and materials, the pertinent contract, the bills and their amounts, and, for intereet during construction, the pertinent loan, the amounts and dates of the paymente. Finally, the quarterly reports contained an annex showing the status of com~liancewith the Joan conditions. 5.34 This comprehensive report was well designed, on the whole, to keep both the borrower who prepared it and the Bank who received it informed on a timely basis of the status of project implementation. There was one significant omission, however, in the list used to report on compliance with loan conditions, which contributed to the ffz ex~enditures. The list did not refer to that part of the covenant requiring an annual audit of FEN'e accounts and financial statements which specifies that the report of the independent auditor shall include a separate opinion as to whether the proceeds of the Loan disbursed on the basis of statements of expenditures (see para 4.21) were used for the purpose8 for which they were provided (Section 4.02(a), Loan Agreement). None of the audit reports contain such a eeparate opinion, and in thoee audit reports which have a separate section detailing compliance with financial covenants, this requirement is aleo omitted. It is implicit that the terms of reference for the audits did not include thie task as part of the audit. It is also significant that the requirement to audit the statements of expenditures was not included in the list of covenants for which an annual report of compliance was requested in the Bank letter of July 13, 1984, which detailed reporting requirements. This omission, and the failure ever to be aware during project supervision that the statements of expenditures had not been audited, may be attributed to lack of  familiarity of t h e concerned projects s t a f f with the Bank's requirements for statement8 of expenditurea because disbureement on t h i s basis was not normally used for power projects. 5.35 The g c may be quertioned i n several respects. a. There were only two occasions when Bank s t a f f who reviewed t h e reports made substantive comments on t h e Progress Reporting S l i p attached t o the reports f o r circulation purposes. one was i n August 1985, when it was noted t h a t t h e report was incomplete. A follow up letter was written (dated August 21, 1985) t o request inclusion i n t h e future of (i)sections reporting actions taken and progress achieved concerning improvement of FEN'S capacity t o analyze and evaluate the technical and financial merits of projects and t o mobilize domestic savings t o finance t h e power sector; and (ii)a section detailing the status of compliance with a l l loan conditions. Although FEN replied t h a t it would comply with these requests (letter dated August 29, 1985), subsequent reports did not i n fact include t h e additional sections mentioned i n subdivision (i)of the preceding sentence. The Bank, however, was otherwise able t o keep track of these matters (see paras 5.9 and 5.13-5.21). The second was i n February 1987 when a comment was made t h a t extension of the closing date had been approved t o permit completion of disbursements. b. Beginning with the report for t h e eecond quarter of 1986, the annex on status of compliance with loan conditione wae limited t o covenants i n the Loan Agreement. This change passed by unnoticed without a request by the Bank t h a t t h i s annex should continue t o include t h e Power Financing and Guaranty Agreements. c. The much lower utilization r a t e of subloane by EPM i n comparison t o t h a t of the other power companies was never noticed and a s a r e s u l t no inquiry was made t o find out why it was occurring. EPM's relatively low r a t e was consistently reported i n a tabular analyeis of t h e utilization of subloans included i n the reports beginning with t h a t for the f i r s t quarter of 1985. That i n i t i a l report showed no utilization by EPM, whereas t h e u t i l i z a t i o n r a t e for the other four companies ranged from 43%t o 92%. In t h e report for the third quarter of 1985, which was reviewed i n t h e f i e l d by the October 1985 eupervieion mission, EPM's u t i l i z a t i o n r a t e is l i s t e d a t 20%, whereas t h a t of the other companies ranged from 60%t o 100%. The situation i n t h i s respect which prevailed a t t h e t i m e of the Auguat 1986 supervision mission is described i n para 5.6. By implication, it appears that the allocation of loan funds t o EPM exceeded its needs and t h a t consideration should have been given t o the feasibility of reallocating some funds t o other companies i n order t o accelerate disbursements. The question also arise. whether FEN, l i k e the Bank, is culpable for f a i l i n g t o notice and consider EPM's low u t i l i z a t i o n r a t e a problem concerning which remedial action should be considered. I f , on t h e contrary, it did take o r consider taking action i n t h i s respect, t h i s should have been highlighted i n its reports t o t h e Bank. The  Bank requested FEN to report problems encountered and eolutione proposed or effected to resolve them in its first letter to FEN describing reporting requirements and in the telex meeeage commenting on the first quarterly report submitted by FEN (see para 5.32). 5.36 There are some respects in which BK would have been desirable. One concerns the categories of expenditures expected to be financed. As set forth in each subloan agreement, the amounte allocated to each project (e.g. Guavio or Cerrejon 11) or work component (e,g. a specified transmission line or substation) are subdivided among typee of expenditure, such as civil works, equipment and materials, engineering and administration, interest during construction and contingencies ("imprevietoen or "por distribuir"). It would have been much more significant to compare the actual expendituree with the original allocations broken down to ehow the type of expendituree, rather than simply comparing total expenditures. Aleo, it would have been useful if there had been understandings on how various limits on financing agreed with the Bank and the commercial banke would be tracked to eneure that the limits were observed. Specifically, this applies tor a. Su~plementaland overrun financinq. For projects already partly financed by the Bank, project cost Overruns were not eligible for Bank financing except for the Mesitas Project (Supplemental Letter /2, June 27, 1984); as a consequence, in all other caeee involving supplemental Bank financing under this project for civil works, goods and services, the amount of such financing wae to be limited to the amount of project costs originally expected to be financed by commercial lenders, to the extent such financing had not been forthcoming. As stated in para 4.24, it wae intended that these limits would be taken into account in approving subloans; while his was a reasonable approach a review of how this worked out would have been desirable as part of FEN's final report. b. -q. In the case of the Bank, financing of expenditures prior to the date of the loan agreement8 (June 27, 1984) were to be limited to US$29 million from the "A" loan and the Bank's portion of the "B" loans, for payments after January 1, 1983 (Loan Agreement, Section 2.02 (d) and Supplemental Letter $2, June 27, 1984; and in the case of the US$175 million "B" loan, exclueive of the Bank's share, financing of expendituree incurred or paid after January 1, 1983 was limited to USS50 million for expenditures incurred and paid before January 1, 1984 (Loan Agreement, Clause 5 (B) (c)). c. Foreisn Costs. With respect to civil work costs, as earlier discussed in paras. 4.22 and 4.24, the disbursement proceduree followed by FEN and accepted by the Bank, did not ensure that financing under the project was limited to foreign coete, ae was expected. (See the statement that subloans under the project would finance foreign costs and the analysis of the financing of the project'e total estimated foreign coste in the Staff Appraisal Report No. 4771-CO, paras 4.07-4.08 and in the Preeident'e Report No. P-3750-CO, para 58). This was a specified requirement for  Bank financing (Loan Agreement, Section 2.02 (d) (iii)and Supplemental L e t t e r #2, June 27, 1984). It wae simply an expectation, and not a formal requirement, f o r t h e commercial bank financing. With respect t o t h e limits on retroactive and foreign coet financing, pertinent data ehould have been provided routinely, i f not ae p a r t of t h e quarterly reporte, then lee8 frequently, perhaps annually. I n any case, FEN ehould have been requeeted t o comment on a l l t h r e e of theee mattere a s p a r t of FEN'e Final Report. The eignificance of theee comments should not be overstated, eince, on t h e baeie of t h e overall review of t h e project record, it doee not appear that deviatione, i f any, from theee financing limite were material. The f a c t remaine, however, t h a t ae mattere now etand, one cannot be certain t h a t they w e r e obeerved. 5.37 The differences between t h e a s applied a t appraisal and eubeequently during project implementation deserve comment. Ae presented i n t h e appraieal report, t h e eetimated coet of t h e project was t h e eum of t h e coste eetimated t o be incurred by ISA, EEEB, EPM and CVC during 1984 and 1985 f o r ongoing works, i n t e r e s t during construction related thereto and etudiee. (See Staff Appraisal Report No. 4771-CO, para 4.08 and Annex 5.39.) On t h i e basie, t h e project cost i n current money terms wae eetimated t o be US$1,602 million, of which t h e eetimated foreign coete w e r e USS898 million and local coete w e r e USS704 million. This wae exclusive of eetimated inveetmente of USS377 million i n projects of other u t i l i t i e e , USS70 million i n working c a p i t a l increases and USS645 million f o r future construction, i.e. projects juet a t a r t i n g o r recently begun. The costa of t h e construction programs of CORELCA and ICEL w e r e also excluded eince t h e i r participation i n t h e project was uncertain. A8 proposed by t h e Bank i n its letter of July 13, 1984, explaining reporting requirements, t h i a concept would have been continued. What was propoeed wae a table which would have comparative columns ehowing local, foreign and t o t a l coete as eetimated at appraisal and a8 l a e t revised, followed by columns which would show disbureemente t o date and funde remaining t o be diebureed. The "coste" t o be financed by t h e "A" and "B" loans would be l i s t e d i n d e t a i l with separate l i n e items ehowing t h e source and amounte of financing t o be provided f o r each subloan, i n a format l i k e t h a t described i n para 5.33; followed by a section consisting of one l i n e i t e m labelled "Ongoing works committed under other financing1'; and a f i n a l l i n e f o r "Total Coet." The t o t a l s i n t h e column showing t h e "Original Project Coet Estimate" would add up t o US$1,602 million, eubdivided between local and foreign coets a s ehown i n t h e appraieal report; t h e subtotal f o r t h e project coete financed by t h e "A" and "B" loane would be USS370 million; and t h e eubtotale f o r t h e "Ongoing Worke" would be t h e difference8 between t h e Total Costs and t h e amounts financed by t h e "A" and "B" loane. A s propoeed i n t h e July 13, 1984, letter, a l l of t h e financing provided by t h e "A" and "B" loane would have been ehown a s f o r foreign coete. This was consistent with t h e analyeie presented i n t h e appraisal report (para 4.08), which applied t h e Bank'e broad definition of foreign costa a s including indirect foreign coste paid f o r i n local currency. I n t h e report of t h e f i r e t supervision mieeion (dated October 10,1984, Annex 7 , para 5 ) , t h i e wae changed t o show t h e eetimated amounte of financing expected t o be provided ae subloane t o each of the power companies, divided between foreign and local costa. The amounts ueed are identical with those i n a table i n t h e project f i l e s which apparently record8 an underetanding reached during negotiation8 a s t o t h e allocation of  t h e project financing and which dietinguiehee foreign and local coete on the baeie of the currency of payment (Distribution Credito FEN-BIRF, March 1984, Doc. No. 287.062 -9). 5.38 The f i r s t three supervision missions included i n t h e i r reporte a presentation of project coete on t h e baeie just described, including f o r t h e l a e t two a comparieon of the appraisal eetimate with a new eetimate. The l a t t e r , however, reflected only eome minor changes eince appraisal i n the US$ valuee of t h e Japaneae yen financing component of the "B" loans; no changes from t h e appraisal eetimatee w e r e made i n the figures presented for the "Balance of Ongoing Worke", even ae l a t e ae the t h i r d mieeion i n October 1985. Nor wae there any recognition that i f t h i e concept were t o be followed, it would be neceeeary t o increaee the amounte ehown ae "Total Project Cost" and "Balance of Ongoing Worke" t o include the balance of t h e 1984 and 1985 construction programs of CORELCA, once it was determined t h a t some of the "B" loan financing would be allocated t o it. 5.39 For purpoeee of the quarterly progreee reporte, FEN was not required t o report project coete on the baeie proposed i n the July 13, 1984 letter; instead, as deecribed i n para 5.33, it followed a definition of project costs limited t o t h e amount of t h e financing provided under t h e project'e eubloane. The project record does not contain an explanation why the Bank choee not t o ineiet on i t e original propoeal. Preeumably, however, ae indicated i n para 5.32 f o r a eimilar ieeue, the Bank recognized that it would not be feasible t o expect the newly created FEN t o implement euch a proposal. The Bank made no e f f o r t during project implementation t o arrange with FEN t o obtain the data needed t o compare t h e actual coete which the f i v e companiee involved incurred f o r t h e i r construction programe i n 1984 and 1985 with t h e appraisal eetimatee; nor did it arrange t o obtain the complementary data needed t o compare t h e i r actual and estimated financing plane for theee yeare which would be an eeeential part of any euch comparieon. Thus, t h i e epecific comparieon i e unavailable f o r purpoeee of t h i e report, and the eector context, which it would have provided, f o r evaluating the performance under t h i s project of FEN, t h e Government and the Bank, i e lacking. Thie i e not a eerioue shortcoming eince t h e eector context i e w e l l known i n broad terms through the Power Sector Adjustment Loan (President's Report NO. P-4676-CO, November 10, 1987) and the evaluation, "Colombia, The Power Sector and the World Bank, 1970-1987", ieeued by t h e Operatione Evaluation Department (Report No. 8893, June 28, 1990). 5.40 Three significant developmente which occurred a f t e r completion of diebureemente ehould be mentioned: f i r s t , FEN'e unilateral eix month extenaion, effective Auguat 1, 1988, of the period of amortization of subloane; second, a temporary exception from the requirement that FEN'S debt t o equity r a t i o ehould not be greater than 7:l granted November 15, 1989; and third, the expaneion of FEN'e reeponeibilitiee i n 1990 and 1991, particularly concerning i t e domeetic borrowing and lending activities. 5.41 1. a. i x Mont The ieeuee related t o FEN'e unilateral extension of the grace period f o r repayment of the project eubloane are diecueeed i n d e t a i l i n Annex I. The main pointe are eummarized i n the following paragraphs.  5.42 I n July 1988, FEN informed t h e Bank t h a t i t e Board of Directore had adopted a resolution which extended t h e eubloan amortization period by eix months. Thie action was taken without prior coneultation with o r coneent by t h e Bank. Section 3.02 (b) of t h e Loan Agreement, however, providee t h a t , except ae t h e Bank ehall otherwise agree, FEN s h a l l not amend, o r f a i l t o enforce any provision of, t h e eubsidiary loan agreemente. I n order t o determine whether t o approve t h e action taken, t h e Bank i n Auguet 1988 requeeted FEN t o provide it with financial projectione f o r t h e next two yeare eo t h a t it could eee how FEN'S financee were affected. The Bank aleo aeked FEN, when eimilar caeee a r i e e i n t h e future, t o provide it with an opportunity t o comment before decieione a r e taken, ae required by t h e loan documente. 5.43 The requeeted financial projectione w e r e eent by FEN i n September. An internal s t a f f memorandum dated November 28, 1988, r e f e r e to, but doee not include, actual and forecast financial etatemente f o r t h e period 1984-90 prepared from t h e information received from FEN and available i n t h e f i l e s . The memorandum etatee t h a t theee ehow t h a t t h e exteneion of t h e eubloan amortization period w i l l not have any material e f f e c t on FEN'e financee and t h a t FEN would continue to have adequate financial ratioe. Ae recommended i n t h e memorandum, a letter wae eent on December 2, 1988, expreeeing no objection t o t h e modification of t h e subsidiary loan agreements. The letter emphasizee t h a t t h e procedure u t i l i z e d by FEN t o modify t h e eubsidiary loan agreemente without t h e Bank's prior agreement wae not i n accordance with Section 3.02 (b) of t h e Loan Agreement, and it requests t h e pereonal intervention of FEN8e Preeident t o ensure e t r i c t compliance i n t h e future with t h e provieions of t h e loan documents. 5.44 The aforementioned memorandum wae prepared by an aeeietant level analyet. The f i l e copy of t h e letter notee t h a t it wae cleared i n substance by a eenior financial analyet; t h i e matter, however, may have received only cureory attention by him. Coneidering t h e record a s it now exiete with eome of t h e supporting material f o r t h e Bank's conclueione mieeing from t h e f i l e e , one i e l e f t with t h e impreseion t h a t t h e Bank'e financial analyeie wae euperficial, and t h a t t h e Bank, having been presented with a f a i t accompli, decided it had no choice but t o accede t o t h e exteneion with t h e admonition t h a t FEN eeek t h e Bank's p r i o r approval f o r any future modificatione of t h e eubeidiary loan agreemente. 5.45 On t h e baeie of a retroepective review, it appears t h a t had t h i e matter been analyzed adequately, t h e Bank would have become aware of two important iesuee: (i)t h e negative r e a l rate of return on equity earned by FEN since 1985; and (ii)t h e mismatch between t h e repayment terms of t h e external loane f o r t h i e project and of t h e subloane made from these eourcee. Had t h i e occurred, t h e Bank could have euggeeted corrective meaeuree. 5.46 g.expreesed The conclueion i n t h e! aforementioned memorandum t h a t FEN would continue t o have adequate f inanci a l ratioe wae inappropriate ineofar ae it applied t o FEN'S return on equity. A s indicated i n Part 111, Table 5.1, FEN'S r e a l return on average equity wae eignificantly positive i n 1983 and 1984. It became marginally negative in 1985 and eubetantially eo i n 1986 and 1987, when t h e nominal return on equity f o r each of theee years was l6%, i n comparison t o rates of i n f l a t i o n of 19%and 23%, reepectively. The projectione supplied by FEN f o r 1988-90 e x p l i c i t l y called attention t o t h e expectation t h a t t h e r a t e of return  on equity f o r 1988 would be negative by about 10 percentage points. A s shown i n Table 5.1, t h e actual nominal return on equity f o r 1988 was 16.7% compared t o an i n f l a t i o n r a t e of 28.1%. FEN'S projections t h a t its r e t u r n on equity f o r 1989 and 1990 would match t h e i n f l a t i o n r a t e did not, however, prove t o be accurate. The actual return on equity i n 1989 w a s l6.6%, compared t o an i n f l a t i o n r a t e of 25.8%, and i n 1990 t h e actual return was 23.9% compared t o an i n f l a t i o n r a t e of 29.1%. I n evaluating t h e j u s t i f i c a t i o n presented f o r t h e extension, t h e Bank should have noted t h a t FEN'S rate of return on equity f o r several years was substantially l e a s than t h e r a t e of i n f l a t i o n and inquired i n t o t h e reasons f o r t h i s unsatisfactory relationship and t h e remedies t h a t might be proposed f o r correcting it. 5.47 : The s i x month extension of t h e grace period on t h e subloans postponed FEN'S receipt of t h e i n i t i a l repayments otherwise due thereunder beginning i n October 1988. This action, taken a f t e r FEN'S repayments of t h e "A" and "B" loans had begun i n June 1988, increased t h e amount of borrowings FEN had t o make from t h e l o c a l c a p i t a l market t o m e e t its repayment obligation6 on t h e external loans f o r t h e project. The Bank made no specific analysis of FEN'S s i t u a t i o n i n meeting its debt service requirements a s o r i g i n a l l y scheduled and as modified because of the extension. Had t h i s been done, it would have supported t h e Bank's concluoion t h a t t h e extension would not have a material e f f e c t on FEN'S finances, because t h e significant borrowings which w e r e needed would not have an adverse e f f e c t except b r i e f l y i n i t i a l l y . Far more important, however, t h e Bank would have realized t h a t t h e o r i g i n a l relending arrangements, baaed on a 13 year term f o r t h e subloans, had r e s u l t e d i n a significant mismatching of maturities of t h e subloans and of t h e foreign loans, which w a s not anticipated when t h e project was appraised. The analysis a l s o should have led t h e Bank t o consider t h e s i t u a t i o n which would have been applicable i f t h e maximum 15 year term f o r subloans permissible under t h e project agreements had been utilized. This a l t e r n a t i v e would have increased substantially t h e amount and extent of borrowings by FEN needed t o m e e t debt repayment obligations on t h e foreign loans which w e r e larger than t h e repayments it received on t h e subloans. On t h e other hand, it appears i n retrospect t h a t i f t h e issues had been appropriately analyzed i n 1988, t h e Bank would probably have concluded t h a t t h e r e was a compelling case t o recommend aubatitution of an extended 15 year t e r m f o r t h e extended 13 year term f o r t h e subloans. This would have been based on a judgment t h a t t h e higher amount of domestic borrowings needed could be raised by FEN without adversely affecting its p r o f i t a b i l i t y , and t h a t FEN'S unprotected exposure t o foreign exchange r i s k under t h e 13 year subloan term would be reduced. The longer repayment period would a l s o have lessened t h e burden t h e power companies w e r e experiencing i n meeting t h e i r debt service requirements, t h e unstated reason f o r t h e extension. This retrospective analysis a l s o demonstrates t h a t t h e Bank's analysis of repayment terms during appraisal was inadequate, and t h a t t h i s was t h e case t o o during supervision, when t h e d r a f t subsidiary loan agreements submitted by FEN were reviewed and approved without participation of a financial analyst. 5.48 The Bank's f a i l u r e t o recognize, and adopt measures t o minimize, t h e mimmatch between FEN'S repayment terms on t h e "A" and "B" loans and t h e repayment terms t o FEN on t h e subloans made from those sources had adverse consequences. Of much greater significance, however, is t h e mismatch of t h e short repayment terms of t h e loan c a p i t a l made available t o t h e power  companies from both foreign and domestic sources i n comparison t o the long conetruction period8 and ueeful earning livee of the aeeets financed by these loana. Thie ieeue and etepa FEN might take t o provide euitable term traneformation t o t h e power companies t o reaolve t h i e problem are diecueaed i n paraa 5.62-5.64 and Annex 11. Such meaeurea would aleo eliminate the miematching problem8 diecueeed herein for the period eubeequent t o t h e i r adoption. 5.49 9. Sincel988, FEN haa not complied with it8 undertaking8 t o maintain its debtiequity r a t i o within the l i m i t of 7 t o 1 and not t o incur any debt which would cauee it t o exceed t h a t ratio, except ae the Bank ehall otherwise agree (Loan Agreement, Sectiona 4.04 and 4.05). The ratio, a8 of the year end, was 7.4 i n 1988, 10.0 i n 1989 and 8.8 i n 1990. The ehortfall i n required equity t o comply with theee covenants a s of the year end wae about 6% f o r 1988, 44% for 1989 and 26% for 1930. In money terms, using year end exchange ratea, t h e ehortfall i n required equity waa approximately ColS2.1 billion or USS6 million for 1988, ColS18.2 billion or USS42 million for 1989 and ColS19 b i l l i o n o r USS33 million f o r 1990. 5.50 For theee calculatione, the t o t a l atockholdere' equity hae been compared t o t o t a l liabilities ae reported on FEN'e audited balance aheets. I n the independent auditore' report on FEN'e compliance with t h i e covenant, the equity baee i a limited t o t h e eum of paid-in-capital, legal reeerve and retained earninge, and a l l other reserve8 and the amount credited t o equity ae exchange adjuetmente are excluded. A s defined i n Section 4.05 of the Loan Agreement, "equity" meane "the eurn of the t o t a l unimpaired paid-up capital, retained earning8 and reeervea of the Borrower not allocated t o cover epecific l i a b i l i t i e e . " Under t h a t definition, the credit f o r exchange adjuetmente ehould be coneidered t o be part of retained earninge, eince it ehould be counted a s part of net profit8 ae etated i n Annex I, para 9. The treatment of t h e other reeervee i e queationable. Two of them, t h e reeerve deecribed aa "At t h e Board'e d i e p ~ e a land~ t h e reeerve for "Donatione", appear t o be ~ allocatione of retained earninge made a s general provieione without any epecific baaie or commitments, and i f t h a t i e ao they should be included as part of equity. The other two reeervee, for "Loan ProviaionsV1and "Taxation", may aleo f a l l into t h i e category. I f any of these allocations w e r e deemed t o be necessary aimply ae general provieione for a particular expense category, t h e more appropriate accounting treatment would be t o include them aa chargee againat earnings on t h e income atatement and r e f l e c t t h e correeponding credite under a heading other than "equity" on the balance eheet. For purpose8 of t h i e report, a l l of the reeervee have been counted ae equity. I f it i e eubeequently determined a f t e r further consideration t h a t some o r a l l of t h e reeervee should be excluded, the ehortfall i n required equity w i l l be increased. For example, a t year end 1990, i f the reeerves f o r Loan Provieione and Taxation were excluded, the debt equity r a t i o would be 9.0 and the ehortfall i n equity would be 29% and amount t o ColS20.7 b i l l i o n o r USS36 million. 5.51 I n calculating the debtiequity ratio, t o t a l l i a b i l i t i e a are used becauee t h e term "debt" i a defined i n Section 4.05 of t h e Loan Agreement t o mean "any indebtedneea of t h e borrower". The l i a b i l i t y baae uaed by t h e independent auditore f o r calculating t h e r a t i o also include8 contingent l i a b i l i t i e e for credite approved and not diabureed. Thie approach has not  been followed for the ratios presented in the two preceding parae becauee under Section 4.05 the amount of debt is limited to the extent to which it "hae become outetanding". Thus, the ratioe preeented herein are lower than thoee cited in the independent auditors' reports because they are baeed on a higher meaeure of stockholders' equity and a lower meaeure of total liabilitiee than used by the auditors. 5.52 In epecifying debtiequity ratio limit6 in a capital etructure covenant, the Bank's normal practice ie to define debt ae "any indebtedneee of the borrower maturing by ite terme more than one year after the date on which it ie originally incurred." (See Operational Manual Statement No. 2.22, Annex 7, Issued: February 1984.) This definition was not applied for thie loan becauee it wae expected that FEN would be raising eubetantial amounts of its borrowed capital from the domeetic markets on a ehort term baeie in the expectation that such short term borrowings would be refinanced or rolled over at maturity. In retroepect, it appeare that the definition adopted, "any indebtednees", went further than wae neceseary, and that it would be poesible to define "debt" to include such short term borrowing8 ae part of the prudent capital structure limitation while excluding normal short term liabilitiee not ueually counted for thie purpose. If thie approach were applied as of the end of 1990 and total liabilities were reduced by subtracting accounts payable and certain other liabilitiee and accrued expeneee and provieions, the debtlequity ratio would be 8.2 and the ehort fall in equity would be 17% and amount to ColS12.2 billion or USS21.5 million. 5.53 The poesibility of increaeing the permiesible debt/equity ratio was raieed inconclueively during project implementation. The banking expert who carried out the fourth eupervieion mission in Auguet 1986 wae aleo asked to assess FEN'S potential role under the power sector adjuetment loan then under preparation. Among the matters he was aeked to review was the reaeonableness of the 7 to 1 debtiequity limit, including if neceeeary recommendation of an alternative ratio (Terms of Reference, dated July 18, 1986). He proposed that the limit be raised to 12 to 1 in order to accommodate proposed borrowing levels and provide room for additional borrowing. The higher ratio, in his opinion, "would remain at a level that ie acceptable by usual banking standarde." His analysis wae baeed on the financial forecaets prepared by FEN (mentioned in para 5.29) which made no allowance for its participation in the power sector adjustment loan. Thie analysis showed that from 1986 to 1989, FEN'S outetanding debt was otherwise expected to increase from the equivalent of USS411 million to USS527 million, or by 28%. Under the 7 to 1 debtiequity limit, additional debt would be permissible equivalent in amount to USS364 million in 1986 rising to USS696 million in 1989. Under a 12 to 1 limit, the additional permiseible debt would amount to USS918 million in 1986 rieing to USS1.57 billion in 1989. Although this analyeis doee not present a prima facie case for any immediate increase in the debt/equity ratio limit, nor for one so large as recommended, the routing memorandum (dated September 18, 1986) which circulated thie report noted that the acting divieion chief concurred in the recommendation. No action was taken on thie recommendation, probably becauee it wae subsequently decided not to utilize FEN as an intermediary for the power sector adjuetment loan. 5.54 By letter dated June 20, 1989, FEN requested a temporary exception from the 7 to 1 debtiequity limitation. FEN referred to eeveral major foreign credite to be made available to it shortly which would cause it to exceed thie  limit. It also referred t o pending legislation which would expand its reeponeibility t o t h e energy sector and pursuant t o which FEN would receive additional c a p i t a l contributions of about Co1$30 b i l l i o n i n 1989. It stated t h a t t h i s would bring it back i n t o compliance with t h e debt/equity l i m i t . It concluded with a statement of its intentions t o submit information by t h e end of 1989 i n support of a request f o r a permanent increase i n t h e debt l i m i t . This request w a s endoreed by a letter from t h e Minister of Finance dated July 10, 1989. Subsequently, a t t h e request of t h e Bank, t h e Minister of Finance i n a letter dated August 31, 1989, informed t h e Bank t h a t t h e Government had requested t h e Colombian Congress t o authorize contributions of c a p i t a l t o FEN i n an amount s u f f i c i e n t t o bring it back i n t o compliance with t h e agreed debt limitation by June 30, 1989. This letter may have been l o s t i n t h e mail o r misplaced, but a f t e r receipt of a copy submitted by FEN, t h e Bank agreed, i n a t e l e x dated November 15, 1989, t o a temporary exception from t h e covenanted debt l i m i t u n t i l June 30, 1989. 5.55 The new law expanding FEN'S responsibilities and authorizing large increases i n its c a p i t a l was not enacted u n t i l February 8, 1990. On June 28, 1990, FEN requested an extension of t h e temporary exception u n t i l t h e end of 1990 ( t h i s letter is missing from t h e f i l e s ) . This request was endorsed by t h e Minister of Finance i n a letter dated July 3, 1990. There is nothing i n t h e f i l e s t o indicate what action i f any t h e Bank took on t h i s request, but presumably it w a s granted. In December 1990, t h e f i r s t increment of equity capital t o be provided t o FEN under t h i s law was paid i n by t h e Government i n t h e amount of ColS19.76 billion, equivalent a t t h e year end exchange r a t e t o about US$35 million. A s indicated above i n paras 5.49 and 5.50, t h i s was not sufficient t o bring FEN back into compliance with t h e covenanted debt l i m i t . I n September 1991, FEN requested comments from t h e Bank and t h e Interamerican Development Bank (IDB) on a proposed increase of FEN'S debt/equity r a t i o ; and i n December 1991, FEN provided supporting information t o its request f o r a waiver of t h e negative pledge clause covenanted with both institutions. FEN pointe out t h a t under new legislation (Law 1731 of July 4, 1991) FEN now operates under t h e r u l e s and regulations applicable t o financial corporations i n Colombia. FEN proposes t h a t t h e covenanted debt/equity r a t i o of 7:l be amended t o conform t o t h e l i m i t s allowed by Colombian law which allows a maximum r a t i o of risk-weighted-assets t o equity of 12:l. Under t h i s approach, t h e r i s k adjusted basis of assets may range from zero r i s k t o 100%r i s k , including intermediate points. To t h e extent t h a t assets have less than 100% risk, a risk-weighted-assets t o equity r a t i o t r a n s l a t e s i n t o a higher debt/equity r a t i o because a corresponding amount of l i a b i l i t i e s which financed these assets is not counted i n t h e risk-weighted-assets t o equity ratio. This has its biggest impact i n FEN'S case f o r t h e large amount of loans t o power companies financed from foreign loans which w e r e guaranteed by t h e Government and a r e therefore claesified a s having zero risk. Thus, a s of August 31, 1991, FEN'e risk-weighted-assets t o equity r a t i o w a s only 0.9:l when its eetimated debt/equity r a t i o was 11:l. I f FEN w e r e t o increase its risk- weighted-aesets t o equity r a t i o t o t h e l i m i t of 12:1, it would mean an increase i n its debt/equity r a t i o t o about 20:l o r more. Both t h e Bank and IDB are evaluating FEN'S requests i n t h e context of t h e r o l e and financial r i s k under which FEN w i l l be expected t o operate i n t h e future. This w i l l depend on t h e f i n a l structure of t h e power sector, currently being debated i n Colosnbia ae p a r t of t h e proposed new e l e c t r i c i t y law drafted i n December 1991, which may allow greater private participation i n t h e power sector. consideration should a l s o be given t o t h e increased r i s k FEN is now taking f o r  its domestic borrowings (see reference t o Law 51 of 1990, next para) and the further r i s k s it would take i f it expands its term transformation r o l e (see Annex 2) . 5.56 5. Two laws enacted i n 1990 substantially increased FEN's scope of operations. One of these, Law 25 of 1990, is mentioned i n the preceding para. This law authorizes FEN not only t o finance investment programs and projects, but also t o refinance debt service obligations i n order t o rationalize t h e functioning of t h e energy sector i n accordance with t h e Government's policies. specifically, FEN is authorized to: (i)make loans t o sector entities t o finance debt service payments due on foreign loans o r the domestic obligations derived from such loans; o r (ii)t o assume obligations of sector e n t i t i e s under loan contracts i n exchange for new credits under which t h e obligations assumed by FEN would be repaid t o it under new terms and conditions and the Government's guaranty of the original loans would continue t o apply t o t h e new credits owed t o FEN. A s stated previously, Law 25 also authorized large additions t o FEN's equity capital. Reflecting the expansion of FEN's responsibilities t o include the e n t i r e energy sector, t h i s law changed its name t o Financiera Energetica Nacional. Most of FEN's a c t i v i t i e s , however, continue t o be primarily focussed on t h e power eector. The second statute, Law 51 of December 28, 1990, modified Law 25 t o delete t h e requirement t h a t FEN c r e d i t s made from domestic savings must have a bank guaranty o r be rediscounted. It also authorizes FEN t o determine the terms and conditions t o be m e t by borrowers t o be e l i g i b l e for loans from FEN and t o undertake fiduciary operations when deemed necessary t o reorganize the finances of the sector. 5.57 FEN'S annual report for 1990 (English version, page 15) reports an additional development which would permit FEN t o participate more actively i n the national savings market and funnel a greater volume of resources t o the power sector. This was the action of the Monetary Board i n November 1990 authorizing FEN t o a t t r a c t the domestic savings it considere necessary and eliminating previously existing l i m i t s on amounts raised from t h e local capital markets. 5.58 J i ) Lendinq t o Refinance Foreian Debt Service. The eteps taken i n 1990, explained i n the two previous paras, w e r e designed t o give FEN a new function, which is described i n t h e 1990 Annual Report (page 22) a s "External Debt Service Support Loans - CADEX". This replaces a previous program, FODEX (Foreign Exchange Fund), utilized by the Government t o make peso loans t o Government e n t i t i e s , and most importantly the power companies, t h a t otherwise would have had insufficient peso funds available t o purchase t h e foreign exchange needed t o meet t h e i r debt service obligations on Government guaranteed foreign loans. FODEX had two sources of funds for these loans: surplusee accumulated by other Government entities, such a s t h e Coffee Fund; and, t o t h e extent t h a t t h i s source was insufficient, funds made available by the Central Bank on behalf of the Government. The surpluses of t h e other Government e n t i t i e s w e r e loaned t o FODEX on short terms and onlent by it on similar terms. FODEX loane funded by the Central Bank on behalf of t h e Government were repayable over somewhat longer terms. Repayment of t h i e second category of FODEX loans is one of t h e sources authorized under Law 25 t o be u t i l i z e d by the Government t o increase FEN's equity capital.  5.59 Under the CADEX program, payments of US$169 million owed by power companies to foreign lenders and euppliers were budgeted to be financed for the second semester of 1990. FEN'e eources for providing this financing are listed ae US$72.5 million from the 1989-90 Integrated Loan Facility (i.e. the "Challenger" foreign commercial bank loan); US$37 million from capital increases; and USS59.5 million from "others, Short Term" (presumably FEN's borrowings from domestic eavings). Actual remittances abroad under the CADEX program, which had been operating since September, amounted to USS127 million by year end 1990, and the balance of the budget remained to be drawn down in 1991 (1990 Annual Report, English version, pages 9, 22 and 23). 5.60 FEN'S domestic borrowings increaeed by Col$16,394 million in 1990, mainly during the last two months of the year (ibid, page 14). This plus the capital increase of ColS19,760 million pesos in December were the apparent sources for Co1$32,029 million of direct short term loans to power companies made in 1990 and outstanding at year end (Audited Financial Statements, Annex No. 1, page 2). It is implicit that these direct loans were made under the CADEX program and that all loans from domestic sources made by FEN in 1990 under the CADEX program were short term. 5.61 During 1991, the outstanding amount of FEN'e domestic borrowings increased 5.4 times from Co1$65,729 million at the end of 1990 to Co1$356,894 million at the end of 1991 (FEN's letter, dated 3 January 1992, commenting on a draft of this report). As a result of eteps taken in 1991, including freeing the peso so that eince October 1991 most foreign exchange transactions, including thoee of FEN and the power companies, take place in the market, there has been a eubetantial real appreciation of the peso in relation to the US$. Thus, converting the peso values of domestic borrowings into US$ at the exchange rate that prevailed at the end of 1990 (ColS568.73) and the parallel market rate at the end of 1991 (Co1$633.68), domestic borrowings increased from the equivalent of USS115.6 million to USS563.2 million, or by almost five times. The dollar equivalent values are significant because the domeetic borrowings were used principally to finance debt service payments of the power companies due on loans guaranteed by the Government, i.e. foreign loane. 5.62 Financial assistance to the power companies to meet their debt service obligations is needed becauee the repayment terme of their loan financing impose financial burdens on them which are not reasonably related to their earning power, even assuming that they were efficiently operated and managed and the level and structure of their tariffs were appropriate. This mismatch between lending terms and earning power was not deemed to be a problem when the arrangements for this project were agreed in 1984. At that time it wae anticipated that for 1984-87 the individual companies, except for the CORELCA and ICEL groupe, would have adequate debt service coverage and make substantial contributions to their capital requirements from net internal cash generation (SAR Report No. 4771-CO, Section 5). By the time the Power Sector Adjuetment Loan was agreed in 1987, the mismatch wae prominently mentioned as one of the factors to which the financial difficulties of the power sector can be attributed. The Preeident'e Report for this loan mentions that the average term of eector debt is short (about 10 years) compared with the average life of power investment8 which normally exceeds 25 years, and that grace periode are much shorter than the conetruction time required for projects (Report No. P-4676-CO, para 61). As then analyzed, the resultant problems were considered  limited to the period 1987-90, years when the sector's debt service coverage8 would be marginally negative or barely positive (minus or plus about 9%). Under the financing plan designed for this period, it was expected that the sector's capital and debt service requirements would be met by supplementing internally generated funds with a large infueion of equity capital, mainly funded by the Bank's loan, and a borrowing program which was very large in groas terms but modest in net terms after allowing for debt amortization requirements. Implicitly, some of the foreign commercial bank financing included in the borrowing program would be used to refinance some outstanding debt. No such assistance was deemed needed for the years 1991-92, for which the projections indicated the debt service coverage would be 1.24 times (ibid, paras 99-101). 5.63 Jii) Need for More Ambitious and Systematic Term Transformation. In actuality, because most if not all the power companies have found it impossible to service their outstanding loans as well as meet all their other funds requirements, resort has been made to a series of ad hoc solutions. FEN'S role in this context has included onlending the share, designated for the power utilities, of new foreign commercial bank loans obtained by Colombia to refinance current amortization payments falling due on them. Although this has effectively extended some principal repayments on such loans, it has only partially mitigated the problems confronting the utilities because repayment terms on their loan financing are too short. Peso advances to the power utilities to meet their foreign debt service requirements have also been provided from FODEX, but since the obligations to repay these advances were short or medium term, the relief they provided has been temporary. This conclusion apparently also is applicable to the CADEX program. Moreover, even the repayment terms of the domestic loans provided by FEN for investment purpoaes are too short: seven years including a grace period of one year. 5.64 The problems resulting from unsuitable repayment terms on borrowings are accentuated because, under the Bank's standard approach for measuring financial performance for purposes both of analysis and compliance with loan covenants, all debt service requirements, except for capitalized interest during construction, are counted to determine debt service coverage and contribution to expansion from net internally generated funds. In the case of the power sector in Colombia, as well as many other countries, this has resulted in a gross exaggeration of debt service requirements in relation to realistic measures of the reasonable earning power of the sector and individual utility companies. To apply effective and realistically achievable financial discipline to the performance of Colombia's power utilities, much more ambitious term transformation provided on a eystematic and transparent basis would be needed. See para 8.19 (h) and Annex I1 for recommendations as to how this might be accomplished through FEN. Financial Performance 5.65 Table 5.2 of Part I11 compares FEN'S actual performance for the yeare 1983-1990 with the projections made at appraisal for the years 1983-1987 in terms of comparative balance sheets and the key financial indicators set forth in Annex 2.13 of the Staff Appraisal Report (No. 4771-CO). Previous sections of this report have dealt with FEN'S financial performance in relation to its domestic borrowing and lending programs (paras 5.17-5.21), return on equity (para 5.46) and debt/equity ratios (para 5.49-5.55). The discussion which  followa related t o Table 5.2 aupplementa the previoue comment8 with particular focus on two of the ratios, grose epread and annual increment i n net worth. 5.66 Groes apread (net income aa a percentage of year end t o t a l aseete) i e a meaeure of profitability. Actual performance i n t h i s respect compared favorably with the appraisal estimate8 f o r 1983-1985 and unfavorably for 1986 and 1987. The drop i n the actual epread from 4.4% i n 1985 t o 2.7% i n 1986 (compared t o appraisal estimate6 of 4.6% and 3.50, respectively) continued i n aubaequent yeare t o 1.4% i n 1989 and 1.6% i n 1990. The annual increment i n net worth i e another meaeure of profitability Since virtually a l l of FEN'e net income i e retained ae part of its net worth. Like the groaa apread, the actual increment6 i n net worth compare favorably t o the appraieal eetimatea for the years 1983-1985 and unfavorably f o r the next two yeara. The subatantial decline i n FEN8s p r o f i t a b i l i t y which occurred i n t h e yeare aubaequent t o 1985 i a indicated by the drop i n actual annual increments of net worth from a range of 26%t o 34% f o r 1983-1985 t o a range of 17%t o 18%f o r 1986-1989 and the recovery t o 27% i n 1990. Thie meaeure l i k e the r a t e of return on equity (i.e. net worth) i n Table 5.1, when compared t o t h e r a t e of inflation, is indicative of the adequacy of FEN'e p r o f i t a b i l i t y t o preaerve i t e capital i n r e a l terma. The inflation rate8 for each of the yeare 1986- 1990, ranging from 19%t o 299, are ehown i n Table 5.1. Aa a reeult of t h e negative r e a l rates of return earned on equity eince 1985 (aee para 5.46), there hae been an eroeion of about 11%i n the r e a l value of FEN'e net worth from t h e end of 1983 t o the end of 1990, excluding f o r t h i a purpose the additional equity of ColS19,760 million paid i n by t h e Government i n December 1990. 5.67 The decline i n FEN'S profitability eince 1985 i a due t o several factors: F i r s t and foremost, FEN8e exemption from income taxea terminated a t the end of 1985. Second, the lending ratee for FEN'e domeatic loana were fixed for the f u l l period of theae loane (increaeed from five t o seven yeare i n June 1985 - aee para 5.26) whereae i t a domestic borrowing coete w e r e increasing from average annual ratee of about 31% i n 1986 t o 37% i n 1989 and 1990. The adverse impact thia hae had on FEN'S profitability haa t o eome extent been mitigated by the decline of t h e relative importance of FEN'e domestic loans t o its t o t a l income producing aeeeta from 26% i n 1986 t o 9% i n 1990. I n 1989, when domeatic loana w e r e 13% of t o t a l income producing aaeeta, FEN had a negative margin (i.e it had loeses) on i t e dorneatic aaeeta, and it changed i t a policy t o link lending ratea on dorneatic loana t o i t e coat of money raieed i n t h e domestic capital market. (FEN report dated July 12, 1991, "Analyeis Hiatorico de loe Eatadoa Financieroa ... Periodo 1986-1990" obtained by t h e Bank i n April 1992.) Third, FEN wae required t o include a s part of i t a expenses i n 1989 a proviaion of ColS752 million f o r the accumulated d e f i c i t i n a fiduciary account it had been managing since November 1984 involving the adminietration of Social Security Reaervea. A charge of ColS120 million wae aleo made i n 1990 f o r t h e loss on t h i s account incurred i n t h a t year. The 1989 charge wae equivalent t o about 12%of i t e net  income for t h a t year net of the charge. For further d e t a i l s , including t h e implications for FEN'S future management of these resources, see Annex 11, para 35 ( v i ) . It is clear that FEN'S lending margins have been inadequate t o maintain its earnings a f t e r taxes a t a satisfactory level, namely a t l e a s t sufficient so t h a t when retained and capitalized the r e a l value of its net worth is preserved. During project supervision, the Bank failed t o suggest t h a t an increase i n margins would be required a f t e r FEN'S exemption from income taxes was withdrawn; and it was not aware u n t i l recently of t h e inappropriate arrangements f o r FEN'S domestic loans under which FEN was a t r i s k f o r increases i n t h e costs of domestic borrowings needed t o maintain such loans u n t i l repayment. This is another i l l u s t r a t i o n of the inadequacies which occurred i n supervising t h i s loan without participation of a s t a f f member who was an expert on financial intermediaries and capital markets. pro1e c t Risks 5.68 The two main sources of r i s k s of the project w e r e identified a t appraisal, namely t h a t (i)t h a t t h e local currency share of investments would not be available i n a timely manner and (ii)FEN would not be able t o f u l f i l l its role as a development banking institution. Both sources proved t o have significant consequences. The inadequate local currency funding was due primarily t o insufficient net internal cash generation, w e l l below t h e appraisal estimates. The major factors accounting for t h i s shortfall a r e (i) lower revenues resulting from unanticipated decreases i n t h e r e a l t a r i f f level and from actual sales growth inferior t o t h a t foreseen a t the time of the appraisal; and (ii)much higher debt service on foreign borrowings caused by very significant unforeseen changes i n exchange rates, namely a substantial r e a l devaluation of the Colombian peso i n relation t o t h e US dollar and of the US Dollar i n relation t o the basket of currencies (including t h e Japanese yen) determining the foreign exchange r i s k of World Bank loans. FEN was unable t o f i l l t h e gap of local financing through mobilization of funds because it could not lend t o the financially weak u t i l i t i e s through t h e financial system. This inhibited t h e a b i l i t y of FEN t o perform one c r i t i c a l function a s a development banking institution: t h a t of supplier of domestic financing through mobilization of funds i n the local market. The capacity t o i n s t i l l financial discipline among its borrowers, another crucial function of a development bank, could not be developed by FEN either, because a key variable for achieving financial discipline - t a r i f f adjustment - was not within its province and was not carried out by t h e responsible authorities i n l i n e with the loan covenants. Furthermore, there is no evidence t h a t FEN has attempted t o perform its mandated role t o promote greater cost effectiveness among the borrowers through its conditionality, o r that it has been instrumental i n upgrading t h e i r accounting and control procedures. FEN has been more a passive outlet of funde t o a deficit-ridden sector than a t r u e development banking institution. 5.69 I n both respecte, the r i s k s were substantially underestimated. A s pointed out previously, the r i s k s related t o FEN'S a b i l i t y t o f u l f i l l its development banking role w e r e dismissed too blandly i n t h e s t a f f appraisal and President's reports; i n t h i s respect, some of the d i f f i c u l t i e s FEN would encounter should have been foreseen, and many of t h e t i m e t a r g e t s established t o achieve specific r e s u l t s were unrealietically short (paras 4.11, 5.11 and  5.14). The factore which resulted i n the deterioration of t h e financial situation of many of t h e power companies, however, w e r e not reasonably foreeeeable when the loan was made. 5.70 A major weakneee i n the Bank's eupervieion wae the f a i l u r e t o aseign t o t h i e activity a etaff member who wae an expert on financial intermediariee and capital markete. The effectiveneee of t h e Bank'e efforte t o improve FEN'S institutional capability would have been greatly improved had euch a etaff member been made available and participated i n most, i f not a l l , of t h e eupervieion mieeione. It would aleo have been appropriate t o have designated euch staff member ae t h e projecte officer. Thie s t a f f member would have provided an element which wae mieeing during t h e implementation of t h e project: an expert repreeenting the Bank who would gain the confidence and reepect of FEN official6 and who would have provided continuity t o t h e Bank'e efforte t o encourage and when neceeeary prod FEN t o f u l f i l l the broad development banking role envieaged f o r it. Aleo, had such a etaff member been involved when the eubloan arrangements w e r e formulated, it is unlikely t h a t the Bank would have failed t o be aware of and deal with t h e significant mismatch between the repayment terme of FEN'S borrowinge f o r the project and of the eubloane made from thoee borrowinge. The e f f o r t of t h e responsible project divieion t o obtain euch aesistance w a s uneucceeeful. Regional management ehould have been more reeponeive; indeed, the need f o r t h i e additional etaff support f o r t h e project should have been recognized, and action taken t o provide it, ehortly a f t e r t h e decieion t o proceed with t h e project i n January 1983. 5.71 On t h e Governrnent'e eide, a major factor affecting implementation was the decieion t o let the r e a l t a r i f f levels f a l l a f t e r the 1985 devaluatione (para. 8.3), eince it had a direct bearing on t h e internal financing of t h e u t i l i t i e e . Thie reeulted i n the delay i n implementation of t h e echeme which would have allowed FEN t o take a more active p a r t a s a financial inetitution, lending local currency funde t o t h e u t i l i t i e s , and ae a development inetitution capable of impoeing conditions t h a t would improve the financial and managerial poeition of i t e borrowere. Theee two capabilities go hand-in- hand eince the conditionality cannot be eucceeeful unleee a permanent relationehip i e eetabliehed with the borrower t h a t would make it poeeible f o r FEN t o demand actione t h a t would be p o l i t i c a l l y d i f f i c u l t t o implement for the borrower, and t h a t would make it worthwhile f o r t h e beneficiariee of FEN loane t o comply. Thie relationehip, and the role of FEN a s a development inetitution ae w e l l , w e r e element8 lacking i n t h e organization of t h e power eector i n Colombia. It remaine t o be eeen whether the efforts of the current Government t o reinvigorate FEN w i l l be eucceeeful. 6. MAJOR RESULTS OF THE PROJECT Prolect Oblectivea 6.1 Of t h e two main project objective6 (para. 3.1), the f i r e t wae largely achieved, although with delaye: t o provide the external financing required t o euetain during 1984-85 the echeduled conetruction pace of eeveral high  priority e l e c t r i c i t y development projects. The support of foreign commercial lenders t o t h e power eector and t o Colombia, through a co-financing package was obtained only f o r t h i s project. The expectation t h a t there would be subsequent co-financed projects involving t h e "A" and "B" loan format did not materialize because the commercial banks withdrew from additional voluntary lending t o Colombia and have limited new loans t o Colombia t o amounts equivalent t o relending a l l o r most of the amortizations due on t h e i r previoua loans. Under t h i s new strategy, however, FEN has served as the intermediary f o r t h e subetantial amounte destined for the e l e c t r i c i t y companies (para. 4.7). 6.2 L i t t l e progress was made i n achieving t h e second main objective t o support the developnent of FEN i n several respects: t o serve a s an effective channel for eector lending projects for power i n Colombia; t o monitor the financial performance of t h e u t i l i t i e s and t o a s s i s t t h e strengthening of the power sector financial management; and t o enhance its capacity t o mobilize domestic savings. 6.3 FEN has not yet become an effective channel for aector lending projects f o r power i n Colombia because it lacks the p o l i t i c a l and institutional clout t o be an effective development bank and has not fully developed its technical and managerial s k i l l s t o monitor and supervise t h e technically stronger u t i l i t i e s . A s previously noted, the impact of the various project components deeigned t o strengthen FEN'S capabilities as a power sector development bank has been insubstantial (paras 5.8-5.12). 6.4 Because of t h e financial weakness of most of the e l e c t r i c i t y companies, there were severe constraints on FEN'S a b i l i t y t o channel funds mobilized from t h e domestic capital market t o the u t i l i t i e s through the financial system, without endangering its own financial a t a b i l i t y (paras 5.16- 5.20). This has greatly limited its proved capacity for mobilizing domestic savings t o complement t h e internal generation efforts of u t i l i t i e s . U s e of t h e government'a guaranty a s a way t o circumvent t h i s obstacle (see para 7.1) has been counterproductive since it becomes a way for t h e discreet encroachment of the u t i l i t i e s a s dependents of the central government budget, and a permanent source of sector financial informality. 6.5 A t appraisal, the benefits of the project were presented i n qualitative terms without quantifying a return on investment for t h e project (SAR No. 4771-C0, paras 4.13-4.14). The qualitative benefits identified were mainly thoee related t o t h e project objectives, the achievements with respect t o which are discussed i n paras 6.1-6.4. In lieu of a quantified return on investment, the appraisal report stated that it has been established t h a t the projects which a r e prospective subloan candidates would have higher internal rates of return than the 130-150 estimated a t the time of original Bank appraieal. This claim and approach have been criticized i n a report issued by t h e Operations Evaluation Department dated June 28, 1990, Report No. 8893, Colombia -The Power Sector and the World Bank, 1970-1977, Volume 11, Chapter 5, paras 56-62.  7.1 To eneure continued financing for t h e power sector through FEN, it is neceseary t h a t t h e Government define more clearly its role a s a development banking institution and gives its f u l l p o l i t i c a l backing t o t h i s role, s o t h a t FEN can become an effective instrument for policy implementation and control, and a poeitive element contributing t o t h e planning and forecasting of t h e financial needs of t h e power (and energy) sector. The current Government hae taken a number of actions t o accomplish t h i s , but a s of t h i s writing (June 1992) t h e efficacy of t h e steps taken is still uncertain. Prior t o these recent changes, FEN had not functioned as a development bank eince it lacked a base of credit-worthy potential borrowers and t h e capacity effectively t o impoee financial and policy conditionality on its borrowers t o increase t h e i r credit-worthinese. It functioned merely as a window for quasi-fiscal spending of t h e government, which choee t o support the power sector through capitalization of t h e financial losees arising from arrears and defaulte of the power eector u t i l i t i e s . In part, t h i s has been accomplished through relending t o t h e u t i l i t i e s by FEN of its assigned ehare of external credite t o t h e Government and of t h e additional equity capital investments i n FEN made by t h e Government i n 1990. This was done rather than supporting t h e financially weak enterprisee more transparently through direct budget support. This policy choice negatively affects t h e sustainability of Bank sector financing through FEN and, more seriously, it undermines t h e accountability of FEN a s a financial institution and its prospects for survival a s a viable development banking inetitution. See para 9.1 for additional comments on the actions taken by t h e Government i n 1990 t o increase FEN'S responsibilities and authority and how t h i s may affect FEN'S future prospects. m i o r Strenuthe and Weaknesses. 8.1 The major strength demonstrated by t h e Bank i n its handling of t h i s project is its capacity and t h a t of its s t a f f t o be innovative and reepond promptly to t h e power sector's need for additional financing for ongoing projects. Thie was a very significant accomplishment and wae achieved by (i) deeigning very quickly a project which served a s a vehicle t o provide supplementary Bank financing for previously financed Bank projects and also a t t r a c t commercial bank cofinancing for these and other ongoing projects, and (ii)overcoming the institutional barriers for the flexible treatment proposed (para8 4.4-4.7). The Bank should also be credited for t h e very subetantial e f f o r t involving over 100 s t a f f weeke t o a e s i s t i n mobilizing the commercial bank cofinancing for t h e project (memorandum t o f i l e s dated April 13,1984). Deepite the limited progress made i n achieving t h e project's inetitution building objective and t h e shortcomings i n t h e Bank'e performance noted i n t h e next para, t h e project wae a worthwhile undertaking. There was a genuine need for an institution much a s FEN with t h e capabilities and responsibilities which the Bank envisioned when t h e project was proposed, and there is still such a need. Of course, it would have been preferable i f t h e Bank had pursued the appraieal report vision of FEN more effectively, but t h e opportunity t o do  so remains. The leeeone learned in executing thie project should be valuable in pointing the way to improved performance in the future. 8.2 The deficienciee in the Bank'e performance are eummarized below: (a) During the deeign and appraisal etage, the urgency of proceeding quickly resulted in eome propoeale not in conformity with Bank policieo for financial intermediaries (paras 4.13-4.15), Bank etaff was elow in addreeeing the Bank'e policy not to finance coet overrune except under special circumstances (parae 4.16-4.20), the design of the diebureement arrangements was inadequate (paras 4.21-4.24), and the financing requiremente for the 1984-85 project period were overeetimated (parae 5.1-5.7). (b) The supervision of the project was not effective, indicating a lack of commitment of the Bank'e technical etaff with the project. Thie ie reflected in the relative complaieance of the eupervieion reports and the Bank'e review of quarterly progreee report8 with the delays of the diebursemente of the subloane (parae 5.5, 5.6 and 5.35); the Bank's failure to maintain under thie project continuoue overeight of the performance of the utilities and the Government to meet the covenanted ratee of return and other obligations set forth in the Power Financing and Guarantee Agreements (paras 5.31 and 5.35); the inadequate reeponee of the Bank to FEN'e decieion to extend the grace period of the subloane financed with funde obtained under loan 2401-CO and the co-financing agreements, without prior coneultation with or coneent by the Bank (parae 5.41-5.48); the Bank'e failure to obtain agreement on, and monitor during eupervieion, key financial indicators of FEN'S performance (paras 5.24 and 5.25); and the Bank'e failure to be aware of, and requeet corrective action for, the decline in FEN'S rate of return on equity and the coneequent erosion of ite equity capital in real terms (paras 5.28, 5.29 and 5.46). (c) The Bank miseed the opportunity to help establieh FEN ae a key planning and policy implementation inetrument because it did not pursue the appraisal report vieion of FEN ae a full-fledged development bank for the power eector (parae 5.30 and 5.67). (d) The Bank failed to recognize the need for much more ambitioue term traneformation than hae heretofore been attempted by FEN in order to (i) have a better match between loan repayment terme and the earning8 generated by the inveetmente financed by the loane, and (ii) avoid the dietortione that occur in meaeuring financial performance baeed on exceeeive meaeuree of debt service requirements that result when it ie  necessary to refinance debte incurred with repayment terme which were inappropriately ehort (parae 5.56-5.64). Many of theee deficienciee are attributable to the Bank'e failure to aeeign a staff member who wae a epecialiet in financial intermediaries and capital markets to be part of the team responeible for thie project and relying instead on coneultant eervicee; the project'e handling wae also adversely affected at t h e e during appraisal and eupervieion by the absence of a qualified power utility financial analyst (e.g. paras. 5.23, 5.24 and 5.47). 8.3 FEN and other Colombian authorities have called attention to the very large increaeee in the project's debt service costs in both Colombian peeo and US dollar terms over what wae anticipated at appraieal. The eubetantial increases in debt service coete resulted from two eourcee: (i) the unexpected acceleration in the mid-1980s of the crawling peg adjustment of the value of the peso in relation to the dollar including a 51% devaluation in 1985 and (ii) the unanticipated eubetantial devaluation of the dollar vie-a-vie the Japanese yen (Y) and other foreign currencies euch as the Deuteche mark (DM) beginning in 1985, which resulted in much higher debt eervice costs for the "A" loan under the Bank'e currency pooling syetem and for the yen denominated "B" loan. Queetione have been raised about the adequacy of the Bank'e work because of ite failure to provide eolutions which would have protected the utilities and FEN againet theee unforeseen exchange rieke. The diecueeion which followe, parae. 8.4-8.18, wae written in, and reflects information on exchange ratee available ae of, mid-november 1991. 8.4 Ae of Dec. 31, 1987, before repayment6 of the project loane began in 1988, there wae an increaee of 61% in the US dollar amount of the "A" loan under the currency pooling eyetem (from US5170 million to US5274 million) and an increaee of 80% in the yen portion of the "8" loan (from US525 million to US545 million). The US5 portion of the "B" loan (US5175 million), of course, wae not affected by the devaluation of the US$, but the Colombian peso coat of servicing it increased as a result of the unanticipated real devaluation of the peeo in 1985. Since year end 1987, there has been a modeet and uneven strengthening of the US$ againet the other currencies in the currency pool; ae of June 30, 1991, a date representative of the peak of that upward movement, the balance due on the "A" loan in US$ terme wae 51% more than the unadjusted original principal amount (US5198 million compared to US5131 million). More recently, there has been eome weakening of the US$ eo that as of September 30, 1991, the balance due on the "A" loan in US$ terme wae 57% more than the unadjusted original principal amount (US5195 million compared to USS124 million). Compared to the exchange rate of Y123 per US$ as of the end of 1987, the rate improved eomewhat in favor of the US$ during 1988-90. At ite best during 1990, however, when the exchange rate averaged 145 to 1, the additional US$ equivalent coste to eervice the yen portion of the "B" loan were about 50% more than had the 220 to 1 rate in effect when the loan wae arranged been applicable. The yen hae strengthened during 1991, and at the rate of 130 to 1 applicable in mid-November 1991 the US$ value of the remaining balance due on thie loan ie about 70% higher. It ie clear that the interest costa and debt repayment obligatione on the project loans, measured in Colombian peeoe and US$, have increased markedly over what wae expected when the loane were made and have impoeed a much higher debt eervice burden on  t h e p e r companies than then anticipated. This e f f e c t is compounded by t h e impact of other foreign currency loans on t h e u t i l i t i e s ' financee. It hae been eetimated, f o r example, t h a t t h e mid-1980s accelerated devaluation increased t h e financial expenses of EEEB by threefold and of ISA by 60%. 8.5 The Bank's o f f i c i a l i n f l a t i o n and devaluation projections f o r 1983-87, which were adopted i n consultation with Colombian authoritiee and u t i l i z e d f o r t h e project appraieal, did not foresee an adjustment of t h e r e a l exchange rate, although its index showed at the time (1983) that it had revalued by about 30% against its level of 1974-75, which wae considered adequate i n t e r m s of t h e t r a d e policy objectives. This w a s public knowledge, and it did not require an elaborate forecasting methodology t o predict t h a t a major nominal devaluation would take place i n t h e following years. The Colombian government, however, had committed i t s e l f t o a policy t h a t precluded a maxi- devaluation. Given t h i s commitment, most analysts agreed a t t h a t t i m e t h a t t h e r e a l devaluation t h a t w a s needed could not be achieved i n t h e short run, and t h e oolution wae assumed t o take place, more o r less gradually, over t h e medium-run. It came as a surprise, therefore, when a new Minister of Finance achieved t h e exchange rate adjuetment within a year, i n 1985, by means of accelerating t h e "crawling pegw to a level of about 50%p.a. t h a t had not seemed poesible. Thus, although t h e Bank and other analysts w e r e aware of t h e need f o r adjuetment, t h e f a i l u r e t o forecast t h i s maxi-devaluation should not be deemed a weakneee i n t h e forecasting, which was internally consistent and took i n t o consideration t h e information available about future policy. The maxi-devaluation was an event which was not reasonably foreeeeable i n 1983 and early 1984, t h e period when t h e project appraisal was completed and t h e loan w a s negotiated and approved. This is aleo t h e came f o r t h e very substantial devaluation of t h e US$ against other currenciee included i n t h e Bank's currency pool which has occurred subsequent t o mid-1985. 8.6 The Bank has been criticized, not f o r mistakes i n it8 forecasts of i n f l a t i o n and devaluation, but f o r it8 f a i l u r e t o provide contingent fall-back mechanieme t o protect againet a major adjuetment of t h e r e a l exchange r a t e , an event t h a t , it is aseerted, was l i k e l y to occur and t h a t would upset t h e e n t i r e financial scenario f o r t h e power eector, a producer of non-tradeable6 par excellence. It i e contended that the proper path would have been t o eecure from t h e central bank adequate market-priced hedging mechanisms against a maxi-devaluation, and against t h e devaluation of t h e d o l l a r r e l a t i v e to other currencies, s o t h a t t h e borrowing u t i l i t i e e would have only had t o bear t h e r i s k of t h e normal "crawling-peg" devaluation of t h e peeo against t h e dollar. It is a l s o aeserted t h a t i f these mechanisms had been i n place, t h e financial c o s t s of t h e foreign indebtedness of t h e u t i l i t i e s would have been kept within bounde, instead of t h e much greater burdens indicated i n para 8.4. 8.7 Ineofar a s t h i e criticism Fmpliee t h a t t h e suggested hedging mechanisms ehould have been arranged during t h e design and appraieal etages of t h e project and agreed during negotiations, it i e untenable. A s noted i n para 8.5, t h e r e was little o r no basie then t o expect e i t h e r t h e maxi-devaluation of t h e Colombian peso i n r e l a t i o n t o t h e US$ o r t h e major devaluation of t h e US$ i n r e l a t i o n to other currencies. Accordingly, ae of 1983 and e a r l y 1984, it is inappropriate t o characterize the poesibility of either of these events  occurring as "likely" and t o expect t h a t attention should have been given t o t h e availability of hedging mechanisms i f they did. 8.8 Moreover, with respect t o a hedge against a maxi-devaluation, t h e p r a c t i c a b i l i t y and m e r i t s of t h i s suggestion are questionable. The Government undoubtedly would have rejected and refused t o consider such a proposal at a time when it was committed t o a policy of gradual r a t h e r than abrupt adjustments i n t h e real exchange rate. I n any case, it is unclear how t h e equivalent of an "adequate market price" f o r such a hedge would have been determined. I f t h e principles f o r pricing hedges i n competitive foreign currency markets w e r e followed, t h e forward price f o r purchasing dollars with pesos would be determined by increasing t h e spot market p r i c e by t h e r a t i o of one plus t h e i n t e r e s t rate payable t o borrow pesos (actually o r implicitly used t o acquire d o l l a r s a t t h e spot rate) t o one plus t h e i n t e r e s t r a t e which would be earned (actually o r implicitly) on t h e d o l l a r s t o be delivered i n t h e future. This basis makes no allowance f o r any changes i n exchange rates other than those implicit i n t h e d i f f e r e n t i a l i n t e r e s t rates. Any allowance f o r a possible change i n t h e real exchange r a t e would have t o be a r b i t r a r i l y determined, and it is doubtful t h a t t h i s a r b i t r a r y determination would be adequate t o compensate f u l l y f o r t h e devaluation. The Government and t h e Central Bank would a l s o be under preseure t o provide eimilar protection against a real devaluation from a l l other creditors exposed t o t h i s r i s k and not l i m i t it t o producers of non-tradeable6 such as t h e power companies. Thus, a policy under which t h e Central Bank would have provided special protection against a maxi-devaluation is probably one under which t h e r e would have been a significant s h i f t of t h e costs of servicing foreign debts from t h e borrowers who incurred t h e debt8 t o t h e economy a t large. The preferable policy is t o allow these c o s t s t o be borne f u l l y and d i r e c t l y by t h e borrowers and indirectly, t o t h e extent t h a t t h e borrowers a r e able t o pass them on, by t h e i r customers. To t h e extent t h a t t h e Government determines exchange rates, its fundamental policy should be t o avoid so far as possible the occurrence of over o r under valuations of t h e rates and t h e corresponding windfalls o r shocks which those who earn o r who borrow o r use foreign exchange experience when t h e exchange r a t e s a r e distorted and subeequent s t e p s a r e taken t o correct t h e distortions. Also, Government policy should not prevent regulated u t i l i t i e s from passing on increases i n debt service cost8 resulting from devaluations (para. 5.71). 8.9 Because of t h e a v a i l a b i l i t y of very large and highly competitive currency markets involving future exchanges of t h e US$ and other major country currencies, hedging s t r a t e g i e s t o l i m i t t h e currency r i s k s on t h e project loans i n relation t o t h e US$ were feasible. It appears that these could have been implemented by t h e Central Bank a t r e l a t i v e l y modest transaction and other costs, assuming t h a t Colombia would have been deemed t o be a c r e d i t worthy party and therefore t h a t high c r e d i t r i s k premiums would not have been charged. In retrospect, it i e c l e a r t h a t such hedging would have been much preferable than simply accepting t h e uncertain exchange r i s k s implicit i n foreign loan8 denominated i n whole o r i n part i n currencies other than t h e US$. Ae s t a t e d i n para 8.5, however, t h e need f o r such hedging was not apparent i n 1983 and e a r l y 1984 when t h e project appraisal was completed and t h e project loans w e r e negotiated.  8.10 The remaining issue is whether subsequent t o approval of t h e project loans i n e a r l y 1984, and i f so when, t h e r i s k of major and long-sustained devaluation of t h e US$ became s o evident t h a t t h e Bank should have recommended t h a t Colombia adopt, o r coneider adopting, a policy t o o f f e r market priced hedging mechanisms against t h e devaluation of t h e US$ through t h e central Bank. That issue, of course, pertains not simply t o t h e project loans but a l s o t o a l l non-US$ denominated foreign loans t o Colombian borrowers, and not simply t o Colombia but t o a l l Bank borrowers. A r e l a t e d question is whether and when t h e need f o r such hedging mechanisms should have become evident t o t h e Government and t h e central bank independent of any advice from t h e Bank. 8.11 There a r e no simple c l e a r answers t o t h e questions posed i n t h e preceding para. The turning point i n a long sustained period i n which t h e US$ strengthened against other currencies occurred about mid-1985. It is very unlikely t h a t anyone would have been aware of t h i s turning point precisely when it happened. In September 1985, however, a formal announcement w a s made a t a meeting of t h e US and s i x other industrialized countries t h a t they intended t o devalue t h e US$ i n relation t o other currencies including t h e Y and t h e DM i n order t o s t e m t h e widening US trade d e f i c i t . The currency markets are s o huge t h a t what happened subsequent t o t h i s announcement is mainly a t t r i b u t a b l e t o market forcee and only p a r t i a l l y t o t h e intervention of government agencies implementing government policies. Thus, t h e September 1985 announcement should be considered an event which put t h e Bank and its borrowers on notice t h a t a devaluation of t h e US$ was likely. 8.12 From t h e beginning t o t h e end of 1985, t h e exchange r a t e of t h e Y t o t h e US$ moved from 251 t o 201, o r an increase of 25% i n its US$ value, but t h e average exchange rate f o r 1985 compared t o 1984 was v i r t u a l l y unchanged a t about 238. Similarly, t h e DM r a t e t o t h e US$ moved from 3.15 t o 2.46 from t h e beginning t o t h e end of 1985, repreeenting an increase of 28% i n its US$ value, but t h e average rate f o r 1985 compared t o 1984 was v i r t u a l l y unchanged a t about 2.9. There w e r e continued annual increases i n t h e US$ value of these currencies ranging from 23% t o 28% i n 1986 and 1987, which by then w e r e a l s o reflected i n declining average exchange r a t e s and increasing average US$ values f o r these currencies. Since 1987, t h e r e hae been an uneven, modeet strengthening of t h e US$, with movements up and down i n t h e exchange ratee within a r e l a t i v e l y narrow band. It ia noteworthy that the range i n variation of t h e exchange r a t e s during t h i s period is applicable back t o January 1987. Using monthly averages f o r t h e period from January 1987 through September, 1991, t h e r a t e per US$ f o r t h e Y has varied from about 123 (November 1988) t o 158.5 (April 1990), o r a r a t i o i n terms of US$ values of 1.29 t o 1; and t h e rate for the DM has varied from about 1.48 (February 1991) t o 1.99 (June 1989), or a r a t i o of 1.35 to 1. 8.13 I n retrospect, it would seem plausible t o have expected t h a t (i)t h e Bank and its borrowers would have begun an assessment of t h e consequences of t h e devaluation of t h e US$ announced i n September 1985 shortly thereafter; (ii)the problem8 caused by t h e declining values of t h e US$ i n r e l a t i o n t o other major foreign currencies would have been recognized by 1986 o r 1987 a t t h e latest; and (iii)consideration would have been given during t h i s period to adoption of hedging mechanieme t o limit the effects of t h e devaluation. In f a c t , t h e problems w e r e recognized by Colombian (and other) borrowers who w e r e very c r i t i c a l of t h e resultant high cost of Bank loans and other non-US$ denominated foreign loans. There w e r e no suggestions, however, e i t h e r from  Bank o r Colombian sources t h a t offsetting hedging mechanisms might be advisable u n t i l t h e announcement of t h e adoption by t h e Bank of new currency management policies i n January 1989. The Colombian criticism, mentioned i n para 8.6, concerning t h e Bank's f a i l u r e t o secure from t h e central bank market-priced hedging mechanism8 against t h e devaluation of t h e dollar, was made i n 1990, but it apparently w a s not related t o o r stimulated by t h e new currency management policiee i n s t i t u t e d by t h e Bank i n 1989. 8.14 The new policies, which were adopted a f t e r more than a year of study, w e r e deeigned t o a l l e v i a t e d i f f i c u l t i e s which Bank borrowers (and not simply Colombia) had experienced. They were intended t o make t h e currency composition of Bank loane more transparent and manageable both f o r loans made since July 1, 1980, which were eubject t o the currency pool, and the fixed currency loane made p r i o r thereto. The composition of t h e currency pool was t o be targeted s o t h a t f o r every dollar i n it t h e r e would be 125 yen and 2 Deutsche marks ( o r t h e equivalent of 2 Deuteche marks i n a compoeite of Deutsche marke, Swise francs, and Netherlands guilders, three currencies whose exchange rates had a tendency t o move closely together). These t a r g e t s were expected t o create approximate balance among t h e t h r e e currency groups which would account f o r 90-95% of t h e currency pool. Previously, t h e currency composition of t h e pool had varied greatly since its inception. The fixed t a r g e t s would enable borrowers t o manage pooled loans a s families of subloans, one i n each targeted currency. Budget planning would be more effective, and hedging would be f a c i l i t a t e d f o r borrowers with ready market accees who wished t o traneform t h e nature of t h e i r foreign exchange exposure should t h a t be desirable i n t h e context of overall l i a b i l i t y management. The fixed t a r g e t r a t i o e f o r t h e currency pool were t o be achieved by July 1, 1991, but progress achieved by July 31, 1989, was s o considerable t h a t changes i n t h e t a r g e t basket matched changee i n t h e currency pool very closely. Under t h e new policy f o r fixed currency loans, currency r e c a l l s f o r amortization payments on each such loan would be made on a pro r a t a basis. This would eliminate t h e uncertainty a s t o what currencies would be called which had previously caueed serioue budget and foreign exchange planning probleme f o r borrowers. A s i n t h e caee of t h e pooled loans, borrowere would be able to t r e a t t h e fixed currency loane a s families of eubloans i n t h e conetituent currenciee, and borrowers with ready market access could transform currency r i s k using hedging techniques. (Papers e n t i t l e d "Queetions and Anewers on t h e World Bank's New Currency Management Policies" and "Transparency Preeentation on t h e World Bank's New Currency Management Policiesn iesued by t h e o f f i c e of t h e controller, September 1989.) 8.15 A paper issued by t h e Bank shortly a f t e r t h e adoption of t h e new p o l i c i e s contains a detailed discuseion of t h e instruments and technique8 available t o hedge currency r i s k , including forwards, futures, options and swape ("Anticipating and Hedging Debt Service Payments on World Bank Loanew, Office of t h e Controller, September 1989). Among other points made, t h i s paper pointed out t h a t aggregating hedging on eeveral loane would be advantageous becauee better pricing is available f o r l a r g e amounts and because t h i s would be less burdensome administratively. The paper suggests t h a t one way t o accomplish t h i s is f o r t h e central bank t o hedge on behalf of many borrowers. I n addition t o t h e guidance and advice provided by t h i s paper, it and t h e "Question and Answerw paper referred t o i n t h e preceding para s t a t e d t h a t , while t h e Bank i t s e l f w i l l not be able to provide hedging services t o  borrowers, technical aesietance on a limited baeis i e available from t h e Bank and t h a t advice is also available from private sector firms with expertise i n t h i s area. 8.16 There was no direct follow up, either by the Bank o r Colombian authorities, t o the suggestions i n t h e papers noted above t h a t it may be advantageous t o manage proactively the foreign exchange r i s k inherent i n World Bank loans using market hedging techniquee. The lack of interest i n conaidering hedging i n the period since 1989 was not limited t o Colombia; the Bank received no requeete f o r technical aesietance on t h i e eubject. This lack of interest may have reflected a view that the devaluation of the US$ had been exceseive and that the US$ warn likely t o etrengthen i n t h e future, so that it was unnecessary t o incur the coete and the administrative burden of hedging. Such a view would be supported by t h e general opinion t h a t t h e US$ had become undervalued i n relation t o other major foreign currencies on a purchasing power parity basis. In addition, t h e lack of intereet may have been due t o an inadequate dissemination or underetanding of the optione available f o r r and t h e potential benefits and coete of, hedging currency risks. More recently, s t a f f of t h e office of the Treasurer have been involved i n providing technical aesietance on hedging t o Bank member countries. The scope of t h i s activity and its current status have not been reviewed f o r t h i s report. 8.17 To sum upr it is not appropriate t o c r i t i c i z e t h e Bank for f a i l i n g t o include, as part of the project design, hedging arrangements which would have protected project borrowere againet, o r softened t h e impact on them of, the devaluation of the US$ i n relation t o other foreign currencies. It may be f a i r t o c r i t i c i z e the Bank f o r taking too long t o formulate its institutional response t o t h e problems resulting from the devaluation of t h e US$ subeequent t o mid-1985. Even i f t h e Bank had formulated that response sooner, say two yeare e a r l i e r , by January 1987, one cannot be certain t h a t a decision would have been taken t o begin hedging shortly thereafter or whether the decision would have been t o do nothing a s was the case a f t e r t h e January 1989 announcement of the Bank's new currency management policies. It is inappropriate t o c r i t i c i z e the Bank's regional staff responsible for supervising t h i s project for f a i l i n g t o react to the devaluation of the US$ which began i n 1985 and failing t o suggest that hedging against t h e devaluation should be considered. Very few Bank staff had any expertise i n t h i s specialized and complex eubject, and it would have been purely coincidental i f such a staff member w e r e part of t h e project team or of t h e regional management concerned with t h i s project. 8.18 Coneideration should be given t o undertaking retrospective studies of what might have been achieved and a t what cost had hedging against t h e devaluation of the US$ i n relation t o other foreign currencies been implemented for the project loans treated i n the aggregate. Although these would serve only as one i l l u s t r a t i o n of t h e costa and benefits of such hedging, such an i l l u s t r a t i o n should be a useful complement t o the theoretical baeie f o r determining whether, when and on what baeis it may be advisable t o undertake hedging and how t o evaluate the rioke involved. The studies could test a variety of assumptions: different etarting dates, e.g. from the beginning of project implementation ehortly a f t e r loan effectiveness, mid- 1985, January 1, 1987, and mid-1989; different techniques, forwards, futures, options, swaps and, perhape, some combinatione thereof; difference8 i n t h e extent of coverage, e.g. i n addition t o f u l l hedging, various approaches t o  p a r t i a l hedging such a s are suggested i n t h e Bank's "Anticipating and Hedging Debt Service Payments" paper; and t h e combined approach using options s t r a t e g i e s t o hedge t h e Y and DM together which t h a t paper s t a t e s i e leee costly. Instead of attempting t o mirror t h e f i n e tuning t h a t might have been undertaken t o take account of t h e changing currency compoeition of t h e currency pool, a simplifying assumption could be used, l i k e t h e fixed "reflecting pooln t h a t closely tracked t h e f u l l pool and which is described i n t h e "The World Bank'e Currency Pool" (October 1988). The eimplifying aeeumption could include a switch i n t h e l a t t e r p a r t of 1989 to t h e t a r g e t s adopted under t h e Bank's new currency management policies as t h e baeie f o r hedging t h e "A" loan. Both t h e Bank and Colombia would benefit from t h e retrospective studies. A t least as an i n i t i a l check, t h e information obtained would be useful to t h e Bank t o determine whether changes w e r e needed i n i t e policies and procedures f o r interacting with its borrowers on t h e subject of hedging t h e currency r i s k s on foreign loans, and t o Colombia i n determining t h e country's policies i n t h i s respect. beesone Learned. 8.19 The principal leseone learned, with implications f o r other projects of t h e same nature, as a r e s u l t of t h e analyeis of t h e preparation and implementation of t h i s project and of subsequent events relevant thereto, a r e t h e following: (a) E t . (i)A special procedure could be established by the Bank for dealing with emergency financial packages under specific conditione, s o t h a t ad hoc arrangements would not have t o be devised as was t h e case f o r t h i s project. (ii)When new lending models are being implemented, particularly a s exemplified by t h i s project when they involve an e a r l y application of a new cofinancing scheme i n t h e euccess.of which both t h e Bank and its borrowers are highly interested and when they are part of an emergency package f o r a c l i e n t sector: more rather than less financial analysis should be applied; and t h e financial and i n s t i t u t i o n a l aspects should be more closely supervised, particularly i f t h e technical component is weak. Such supervision ehould be entrusted t o Bank s t a f f and not delegated t o consultants. When a principal objective of a project is t o aeeiet a newly created financial intermediary t o serve ae a development bank f o r a c a p i t a l intensive infrastructure sector and strengthen its financial management, major responsibility f o r t h e Bank's a c t i v i t i e s during t h e design, appraisal and supervision stages ehould be assigned t o a e t a f f member who is an expert on financial intermediaries and c a p i t a l markets. Key financial indicators t o be monitored should be agreed during appraisal. When t h e r e are  shortfalls in performance, appropriate follow up action should be taken to see that corrective measures are adopted. (C) project6 Involvina more than One Sector. When projects involve more than one sector and the staff expertise required for the different sectors work in separate divisions, Bank management must ensure that there is adequate cooperation and coordination between or among the divisions so that staff experts who should be involved in appraisal and supervision of such projects are made available when needed. The Bank should be more agile in forming multi-disciplinary teams across division lines when needed and not attempt to compensate for organizational rigidities by using conaultanta. (d) pealistic Schedulinq. When establishing dates to be specified in loan covenants for carrying out studies and accomplishing institutional improvements, be realistic in estimating the time needed and avoid setting overambitious targets. (e) pisbursement and Revortins Recruirements. When disbursements are to be made on the basis of advances into a special account, and the subsequent actual use of the loan funds for project purposes is to be accounted for by disbursement from the special account, it is essential that the appraisal report estimate the rate of expected disbursements on both bases. Disbursement arrangements from the special account and reporting requirements should be designed so that they provide a basis for confirming that the financing was used for the intended purposes such as foreign coats, and that other limits or requirements were observed such as the amount of and t h e limits for retroactive financing. Staff should make sure that the Borrower arranges for the required audit of statements of expenditures (paras 4.21-4.25, 5.34 and 5.36). There are no advantages to be derived from showing leniency in the event of non-compliance with loan covenants. In the case of this loan this is particularly critical, especially with regard to FEN'S failure to carry out various commitments designed to enhance its financial planning and advisory capacity (paras 5.10-5.12). Less leniency would have stimulated more the institutional strengthening of FEN as a development banking institution. The Bank should reconsider its role with respect to the devastating effect that foreign exchange movements can have on the financial position of borrowers, particularly on producers of non- tradeables such as the utilities, in the absence of hedging  mechanieme againet t h i e riek. The queetion which ehould be addreesed i e whether the Bank i e doing a l l t h a t it ehould t o aeeiet borrowere t o addreee t h i e ieeue appropriately, including provieion of adequate technical aseistance and advice t o do so. In the caee of Colombia, the r o l e of the central bank ehould be reviewed i n t h i e context, eince it ehould be able t o provide advice about t h e coete and potential benefit8 of hedging and t o coordinate and consolidate hedging a c t i v i t i e e on an economical and e f f i c i e n t baeie for borrowere who desire t o hedge. Until recently, t h e Central Bank monopolized a l l foreign exchange traneactione a s a reeult of the exchange control regime (Decree 444 of 1967), but t h i s is no longer t h e caee (eee para 5.61). (h) Extended Term Traneformation. The f e a s i b i l i t y of extending FEN'e term transformation role ehould be etudied i n collaboration with Colombian authorities eo t h a t t h e burden of debt service requirement8 on Colombian power companiee is realietically, reaeonably and traneparently related t o their earning power. Thie eubject i e diecueeed further i n Annex 11. 9. BORROWER PERFORMANCE Ma-ior Strenaths and Weakneeeee 9.1 The major etrength ehown by FEN ae a reeult of t h i e loan was its proved a b i l i t y t o mobilize financial eavinge i n the domestic market through innovative and euitable ehort- and medium-range instruments. It also developed the a b i l i t y t o collect financial information from i t e client u t i l i t i e s and t o aggregate it i n t o financial forecaeting modele for the power eector. According t o FEN management, these modele a r e working and enable the institution t o obtain t h e information it neede t o be a prudent lender i n a financially weak eector. I n t h e opinion, however, of the etaff of t h e Bank and t h e IDB, a s w e l l a s other independent obeervere, the modele and the information a r e very deficient and muet be improved. There are even questions about the justification of having t h i e information proceesed by FEN and not by ISA o r a technical eecretariat of the newly created Commiseion Nacional de Energia. Undoubtedly, FEN neede t h i e information i f it is ever going t o become a w e l l functioning development bank f o r t h e power sector. Since it has developed the channel8 f o r t h e collection and proceeeing of financial data from t h e u t i l i t i e e , a practical and perhape the beet solution would be t o leave t h i e responsibility with FEN but require it t o improve the proceee and t h e financial forecaeting modele. The major weakneeeee of FEN a r e or have been ae summarized below: (i)It did not adequately improve ite technical and managerial skille t o eerve a s a power sector development bank i n euch reepecte ae t h e a b i l i t y t o evaluate loan propoeale and monitor the implementation of projects f o r which it provides financing, and t o eerve a s a power eector financing strategy advieor t o t h e Government and t h e u t i l i t i e s .  (ii)It has not had enough political or financial clout to impose a conditionality among its borrowers t h a t would i n s t i l l financial d i s c i p l i n e i n t h e system. Laws 25 and 51 of 1990 not only expanded FEN'B lending functions t o include refinancing debt service obligations i n order t o rationalize t h e functioning of t h e energy sector, but a l s o authorized FEN t o determine t h e terms and conditions t h a t borrowers must meet t o be e l i g i b l e f o r FEN's loans (para 5.56). This gives FEN both t h e responsibility and t h e necessary financial clout t o ensure t h a t its borrowers make significant and timely progress t o remedy deficiencies as a condition of continued financial support from FEN. Assuming t h a t t h e Government does not i n t e r f e r e with FEN's conduct f o r p o l i t i c a l reasons, FEN's future success depend6 on how good its judgment is i n applying its authority t o impose conditionality. (iii)It has been limited in its capability to supply the sector's needs of local funds, not by its savings mobilization potential, but by t h e financial weakness of t h e potential borrowers t h a t a r e not credit- worthy. This is no longer s o since t h e adoption of Law 51 i n December 1990, which authorized FEN t o provide Col$ c r e d i t s without a bank guaranty o r rediscounting. Under t h i s law, FEN must be prepared t o take t h e f u l l r i s k of bad loans and, a s s t a t e d above, i f it is t o be successful, it must see t o it t h a t its borrowers correct t h e i r weaknesses a s a condition of its lending. ( i v ) The Government has not viewed FEN a s a full-fledged development bank f o r t h e power ( o r energy) sector but rather a s an instrument f o r quasi-fiscal spending by means of capitalization of t h e i n s t i t u t i o n t o compensate f o r losses derived from lending t o bad c r e d i t subjects. This could contain t h e seed f o r FEN's future disarray, since it obfuscates t h e financial picture of t h e i n s t i t u t i o n and diminishes t h e accountability of its management. The actions taken i n 1990 t o increase FEN's responsibilities are indicative of a change i n t h e a t t i t u d e of t h e Government. The Government now expects FEN t o a c t a s an e f f e c t i v e development bank i n support of government p o l i c i e s t o improve t h e functioning of t h e power sector. FEN is t o a s s i s t i n t h e formulation of performance o r management contracts which t h e power companies would enter i n t o with t h e Government and which would d e t a i l measures t o be taken by them t o improve t h e i r efficiency. Fulfillment of theae contracts w i l l be supervised by t h e government, and continued access t o FEN c r e d i t s w i l l be conditioned on compliance with these contracts. Companies owned by t h e national government which f a i l t o comply with t h e contracts a r e t o be subject t o restructuring o r liquidation. (FEN's 1990 Annual Report, English version, pagee 8 - 9.) (v) NO major restructuring exercises of ICEL, CORELCA and t h e i r subsidiaries have been carried out, although they a r e t h e l e a e t credit- worthy borrowers of FEN. Under FEN's supervision, studies have been made t o determine what should be done t o improve t h e operations, financial performance and management of theae companies. Following amendment of t h e Colombian constitution i n June 1991, l e g i s l a t i o n is being prepared pureuant t o which theee companies would be reorganized. Although t h e r e now is a reasonable prospect t h a t t h i s w i l l be accomplished i n t h e near future, t h e overall time taken t o effectuate  improvements in the performance of these companies has been unduly long. Lessons Learneq 9.2 The major lesson learned is that the financial problems of the power sector cannot be solved merely by the creation and maintenance of a specialized financial intermediary such as FEN. Until December 1990, FEN'S managers, acting responsibly as prudent bankers, relied on bank guaranties or rediscounts as security for Col$ lending to the power companies, and this limited the availability of such assistance to the companies that were credit worthy. Also, all of FEN'S foreign lending to the power sector has required a government guaranty, and this is still the case. This guaranty may or may not be honored. If it is honored, it contributes to the financial disarray of the sector and to the widespread undiscipline. If it is not honored, it taxes the financial structure of FEN and will lead to its eventual surrendering of independence -e.g. to giving up its role as a bank. The potential of FEN as an agency capable of mobilizing funds to the power sector in the domestic market and internationally will be largely wasted unless the utilities are restructured so that they may become credit-worthy borrowers on their own right, and this cannot be accomplished without a thorough reorganization of the sector. 10. PROJECT RELATIONSHIP 10.1 Although the relationship between the Bank and FEN was harmonious at all times during implementation of the project, the involvement of the Bank's staff with the borrower was lacking in the sense that no interest permeated regarding the institutional performance of FEN or the role to be played by this institution in the development of the power sector. The attitude of the Bank towards FEN has been of benign neglect, which has not been conductive to institutional development or strengthening. This is attributable to the Bank's failure to assign a staff member who was an expert on financial intermediaries and capital markets to participate in the design and supervision of this project (para 5.70) 10.2 FEN'S relationship with the government has not been successful in the sense that the institution has not been able to define a strong role for itself in the organization of the power sector. 10.3 FEN could improve substantially its information gathering process and its ability to deal with the financial problems of the utilities if it could induce them to accept and apply uniform and standardized accounting procedures and to adopt a standardized management information system. This is particularly critical in the case of ICEL. 11. CONSULTING SERVICES 11.1 Consultants played a relatively small but potentially significant role for several key aspects of the project. FEN hired the local consulting firms  of Mejia, Millan y Perry and FEDESARROLLO t o develop a financial forecaeting and analyeie model -ENE and t o carry out a etudy on sector finances and formulate a propoeed financing etrategy (para 5.10). In eubmitting t h e i r d r a f t report, the coneultanta recommended additional work and improvement8 t o complete development of t h e model. The need for t h i e further work wae endoraed by t h e Bank i n i t e commente on t h e d r a f t report which concluded t h a t t h e model, when perfected, would provide a auitable baeie f o r t h e periodic review of t h e financial prospecte and etrategiee of t h e power sector (letter dated February 5, 1986). It i e uncertain what action FEN took t o carry out these reconunendatione (para 5.11). Ae pointed out i n para 9.1, there a r e conflicting views on how effective t h e model hae been. 11.2 An individual consultant, Renato Salazar, was retained t o aeeist i n improving FEN's project analysis and monitoring capacity. H i s work wae competently performed and hie report wae w e l l received by t h e Bank which endorsed hie recommendatione. Although FEN implemented t h e recommendations, it has to date not developed independent appraisal capacity which wae one of t h e objectives of t h e project (para 5.9). This, plus t h e delay i n retaining t h i s consultant and the limited uee made of t h e funds provided under t h e loan f o r t h i e aesistance, are another indication of a lack of genuine commitment by FEN t o t h e objective of etrengthening its capabilities a s a development bank for t h e power sector (see para 4.9). 11.3 A study of t h e capital markets i n Colombia with reconunendatione for FEN about alternative inetrumente t o mobilize eavings i n t h e domestic market on a medium and long term baeie wae conducted by t h e local coneulting firm Servicios de Informacion Ltda. An individual consultant retained by t h e Bank, Celestino Carbajal, also submitted a report which analyzed options t o iseue longer term securitiee which FEN might use t o raiee reeources from t h e domeetic market (para 5.15). Theee studiee and t h e i r conclusions w e r e not u t i l i z e d since FEN wae unable t o develop resource mobilization t o i t e f u l l potential (paras 5.16-5.20). The legal and regulatory changee made i n 1990 have greatly expanded FEN's responsibility and authority t o r a i s e funds from t h e domeetic market, and t h e amount eo raieed increaeed dramatically i n 1991 (paras 5.56-5.61). Ae a reeult, these studies should be of significant u t i l i t y t o FEN for its future domestic borrowing a c t i v i t i e s . 12.1 The data relevant for t h e preparation of t h e PCR for t h e moat part were readily available. There a r e a few eerious gape i n t h e documentation contained i n t h e filee. Copiee of some incoming letters and information eubmitted by FEN o r t h e Government w e r e missing, apparently because etaff had not eent them t o t h e o f f i c i a l f i l e e . Minutes of t h e negotiations w e r e not available. This report had t o r e l y on information contained i n FEN'e Final Report f o r some of i t e e t a t i e t i c a l information and on t h e f i l e s of t h e power eector projects for data on t h e statue of compliance with covenante i n t h e Power Financing Agreement. 12.2 There was no agreement during negotiations on t h e key financial indicators t o be monitored during project implementation. Through an overeight, t h e intention t o reach agreement on these indicatore during t h e  third eupervieion mieeion wae not implemented. In future project6 and coneietent w i t h eetabliehed Bank practice, agreement on a l l key indicators for monitoring project performance ehould be reached during appraieal. 12.3 It proved t o be impractical t o obtain reporting which would compare actual t o eetimated project coete on the baeie of the definition of project coate ueed f o r t h e appraisal (parae 5.37-5.39). In the future, arrangement6 for the collection of euch data ehould be made prior t o preeentation of a project for Board approval. In addition, the design of disbursement and reporting requirement6 ehould be improved eo that actual uees of loan funde, including eubloan funds, could be compared w i t h estimated uees more substantively than proved t o be poeeible in thie instance (para 5.36). 12.4 In dealing w i t h a borrower much ae FEN which reliee on ehort term borrowinge for a eubetantial part of its capital requirements, the definition of debt t o be ueed i n a debt/equity limitation covenant ehould exclude normal ahort term liabilitiee (para 5.52).  - - 57 PART 11: PROJECT REVIEW FROM THE BORROWER'S PERSPECTIVE 1.01 In compliance with Section 3.03 (b) of the Loan Agreement, the Borrower prepared and eent t o the Bank a completion report of t h e project. Based i n part on t h i e f i r s t report, the Bank prepared a preliminary draft PCR and eent it for cotnmente to the Borrower. By letter of January 3, 1992, the Borrower eent i t e comments which were received at the Bank on January 8, 1992. Due coneideration has been given t o these comments i n the preparation of the final t e x t of Part I of t h i e report. 1.02 The completion report prepared by t h e Borrower t i t l e d " EVALUACION FINAL DE LA EJECUCION DEL CONTRATO DE EMPRESTITO FEN-BIRF (2401) - BANCA COMERCIAL," gives an overview of the evolution of the eector economic indicators and a descriptive account of t h e preparation and implementation etages of t h e project. For future reference, the Borrower's report has been included i n t h e project f i l e . 1.03 To r e f l e c t t h e Borrower'e views on the main achievements and lessons learnt, t h i e section contains English tranelatione of: a) the Table of Contents; Introduction, Achievement of Overall Objectives of t h e Loan; and Conclueione of the Borrower's report; and b) t h e f u l l t e x t of the Borrower'e letter of January 3, 1992; Annex 3 contains a copy of t h e original text of the l e t t e r .  2.0 w o n of selected Sectione of the Borrower'e Reoort 2.01 -erg o Reaort - ble of Co- TABLE OF CONTENT8 INTRODUCTION I. RECENT FINANCIAL DEVELOPMENT OF THE POWER SECTOR 1. Overall Framework 2. World Bank Financing for the Sector 11. FINANCIERA ELECTRICA NACIONAL S.A. 1. The Nature, Purpose and Role of Financiera E16ctrica Nacional S.A. 1.1 Nature 1.2 Purpoee 1.3 Role 2. Internal Organization of Financiera Elgctrica Nacional S.A. 3. Enterprieee Holding Stock in FEN 4. FEN'e Financial Resources 4.1 CEV and TER 4.2 "Constant Valuew Social Security Bonde 5. Credit Operatione by Financiera Electrica Nacional S.A. 5.1 Credit Operatione Using Internal Reeources 5.2 Credit Operatione Ueing Reeourcee from External Loane 6. Restructuring and Capitalization of FEN 6.1 Reetructuring 6.2 Government Capitalization of FEN 111. EVALUATION OF PROJECT 2401-CO, COFINANCED BY COMMERCI?LL BANKS 1. Background to the FEN/World Bank Agreement 2. Overall Implementation of the FEN/World Bank/Comercial Banke Loan Agreement 2.1 World Bank (2401-CO) 2.2 Midland Bank 2.3 Induetrial Bank of Japan (IBJ) 3. Dietribution of Financing among Enterprieee 4. Application of Financing, by Local or Imported Investment Component  5. Financial Condition6 Applied to Subloans 6. Achievement of Overall Objectives 7. Impact of the FEN/World Bank/Commercial Banks Loan on FEN'S Financial Position 8. Revaluation of the Debt Reeulting from the Compoeition of the Basket of Currencies 9. Achievement of Objectives in the Financing of Individual Projecte 9.1 Interconexi6n Electrica S.A. (ISA) San Carloe I San Carlos I1 Desviacionee Chivor Jaguae Termozipa V Calderas 9.2 Empreaa de Energia Electrica de Bogota (EEEB) Mesitae Guavio Urban Distribution 9.3 Empreaas Piiblicaa de Medellin (EPM) Guadalupe IV Playae 9.4 Corporaci6n Aut6noma Regional del Cauca (CVC) Salvajina Hydropower Generation 9.5 Corporaci6n Electrica de la Coeta Atl6ntica (CORELCA) Tennoguajira I Termoguajira I1 Cuestecita Substation Electrification: C6rdoba, Sucre and Sur de Bolivar Valledupar Substation Control Center Power Project for Noninterconnected Areas (PEZNI) Expansion of San Andrea Generation Alto Sinii Hydropower Project CONCLUSIONS  - 60- TABLES Ordinary Loans t o t h e Power Sector Loane and Diebureemente ueing External Financing, by Enterpriee and Project: General Table f o r Loan 2401-CO (Annex No. 1) Credit Dietribution, by Enterpriee Application of Credit, by Local and Imparted Invoetment Component and 6. Statue of FEN/World Bank Loan Adjueted f o r t h e Effecte of t h e Eaeket of Currencies Breakdown of FEN Aeeete and L i a b i l i t i e e Coneolidated Balance Sheet Exchange Ratee CHARTS Inetitutional Structure of t h e Power Sector FEN: Organization Charts f o r 1984, 1987 and 1989 Geographical Location of Power Projecte Supported by t h e FEN/World Bank/Commercial Bank Loan External Financing: Annual Diebureemente by Enterprime (Chart No. 1) External Financing: Total Diebureemente by Enterprime (Chart No. 2) External Borrowing: Operating Proceee Breakdown of FEN Aseets and L i a b i l i t i e s Statue of t h e FEN/World Bank Loan Adjuoted f o r t h e Effects of t h e Basket of Currenciee (Chart No. 3) Diebureemente of External Resources, by Source and Application (Chart No. 4 ) Application and U s e of FEN/World Bank/Coanaercial Bank Loan Proceede (Chart No. 5) Balance of FEN/World Bank Subloan, Adjueted f o r t h e Effecte of t h e Basket of Currencies (Chart No. 6)  - porrower's ReDOrt Introduction JNTRODUCTION In accordance with Agreement 2401-CO, signed on June 27, 1984 to govern the FEN/World Bank/Comercial Bank cofinanced project, the purpose of thia report is to furniah the World Bank with an evaluation of loan execution. To this end, the report first preaenta an overall analysia of recent financial developments in the Colombian power sector. Second, it preaente the organizational structure of Financiera Electrica Nacional S.A. (FEN), with a breakdown of ita ahareholdere, financial reaourceo, overall lending programs, and its future prospects after the reetructuring and capitalization process through which it will become Financiera Energetics Nacional S.A. (FEN), with a broader range of activities and management. Finally, the implementation of the cofinanced project ia evaluated, with an examination of the background to the Agreement, ite overall implementation, the dietribution of the financing by enterpriee, financial conditions applied to the mubloanm, the achievement of overall objectives, the impact on FEN'S financial position, the revaluation of the loan amount because of the effects of the basket of currencies, and the achievement of the objective8 of the epecific subprojects financed. porrower'a RePort - Achievement of Overall Ob* Whievement of Overall Loan Oblectivea Basically, the loan had the following two major objectives: first, to provide financial aupport for energy enterprise8 to enable them to complete the conatruction of power generation, distribution and tranemiseion projects, in a cofinanced operation eupported by commercial banks. Second, it was to aupport the development of FEN (Financiera Electrica Nacional S.A.), enabling it to become an effective meane of channeling finance to the eector and to increaae ita capacity for mobilizing domoatic aavinga and providing acceae to future external borrowing. An examination of the implementation process ehowe that these objective8 were entirely achieved, mince the financing contributed toward the completion of a number of generation, tranemisaion, eubtranemiasion and diatribution projects, aa will be ahown later. Aa regarda FEN, it increased its lending activities and became the moat important financing agency in the sector, as ie reflected in ita financial position. A brief analysis of this follows.  CONCLUSIONS 1. FEN made eatisfactory progress a8 a financial intermediary. 2. The reeourcee were fully diebureed and made it poeeible to implement a large number of projects that are now generating revenue for the enterprises concerned. 3. Nevertheleee, the power eector as a whole was prevented from achieving overall poeitive results by the atructural problems of inadequate net internal generation and the high level of external indebtedneee. 4. The change in the value of the loan as denominated according to the World Bank/IDB baeket of currenciee had a coneiderable impact, coupled with the high rate of domeetic devaluation over recent yeare. 5. Internally, FEN was coneolidated a8 a financial agency. In the private inveetment market it firmly eetabliehed varioue type8 of eecurity with different maturitiee, gaining public acceptance and confidence. 6. In light of it8 eucceee and the need to increaee its ephere of activities, it ie ehortly to become Financiera Energetics Nacional S.A.  FEN: FINANCIERA ENERGETICA NACIONAL S.A. Office of the President Santaf6 de Bogota D.C. January 3, 1992 Mr. Graham Smith Division Chief, Infrastructure and Energy Operations Country Department 111 Latin America and the Caribbean Regional Office World Bank Washington, D.C. Dear Mr. Smith: I wish to make the following comments on the Project Completion Report regarding Loan 2401-CO to FEN: 1. Paras. 3.01, 4.01 and 4.05 [now 3.1, 4.1 and 4.51: As regards the objectives of the loan, the Report emphasizes that, from its establishment, FEN was to act as a development bank. Although at appraisal and during the negotiations on the draft Agreement this was mentioned as one of its several purposes, far more importance was attached to the objective of supporting Colombia in its efforts to increase energy supply to the necessary level for supporting development and increasing employment. Consequently, the basic purpose discuesed at that time was to ensure that the necessary projects for increasing power supply could be executed, so that rationing could be avoided. We do not believe that the advisory missions to FEN in June and November 1982 made any headway in establishing FEN as a development bank, because the agency only began to operate at the end of 1982, and its main concern then was its capitalization (in the form of resources from the Electric Power Development Fund managed by Banco de la Repiiblica). According to our records and institutional memory (i.e. our "oral tradition"), FEN's beginnings were very different from the account presented in the Report. At the same time, other activities undertaken initially were the mobilizing of domestic savings and the transfer of resources through rediscounting, rather than the formation of an entity to be responsible for the control and financial coordination of the sector. 2. Para. 4.08 [now 4-61 (FEN'S Role): In view of the above, it is not valid to assert that developing FEN's role as a development bank was one of the major components of the project. In fact, the main effort was devoted to achieving the objective identified in (1) above, and FEN was regarded more as a mechanism for channeling resources to the  u t i l i t i e s , i n i t i a l l y i n local currency and subsequently i n foreign exchange. It was only three or four years later that steps were taken t o direct FEN toward t h e a c t i v i t i e s referred t o in t h e Report, i.e. controlling and managing the sector's finances. 3. Para. 4.09 [now 4.91: One of the factors leading t o t h i s situation was I S A ' s role i n the power sector a t t h a t time. Certainly, t h e National Government, and particularly the Ministry of Mines and Energy (together with the Bank i t s e l f ) , showed considerable hesitation over whether FEN wae t o take over ISA's controlling and coordinating role from the outset. In the event, ISA continued t o be responsible for the technical and financial coordination of t h e sector from 1983 t o 1986, while FEN devoted i t s e l f t o obtaining the resources required by the enterprises for carrying out t h e i r expansion plans. A s a result, conditions i n FEN, the Government and t h e Bank i t s e l f did not favor granting the new agency a leading role i n controlling the sector's finances. 4. Changes i n the management of t h e sector were very gradual. A s the Report points out, the Government took a decisive role i n t h e planning process for expanding the power supply system for the f i r s t t i m e i n May 1983, when CONPES discussed t h e matter. It was then considered t h a t the increase i n power demand would be considerably below the sector's projections and t h a t consequently the construction of new generating projects could be postponed for some time. It was also the f i r s t t i m e that the necessary links between power development and macroeconomic planning had been mentioned, and t h a t the financial constraints arising from the high level of indebtedneee had become evident. The result is t h a t only one new project has been begun since t h a t year, i.e. the Riogrande I1 hydropower plant, mainly because it was planned a s a multipurpose project, and it was its importance i n meeting t h e Medellln area's drinking water needs that was decisive. The remaining projects i n the sector's expansion plans (since a s early a s 1977) are still awaiting decisions on the starting dates for t h e construction of t h e i r main works. A s regards financial issues, between 1986 and 1987 the National Government's lack of confidence i n ISA's r o l e i n t h i s aspect of its activities reached a crisis point, and consequently the Administration a t t h a t t i m e (1986-90) made FEN responsible for the preparation of financial projection6 and t h e i r monitoring under the adjustment plan, and it was i n l i g h t of these a c t i v i t i e s t h a t Agreement 2889-CO was entered into by the Bank and the Republic of Colombia. 5. Para. 5.07 [now 5.91: The Report also raise6 the issue of FEN'S r o l e i n previous years with regard t o loan appraisal. Until t h e end of 1990, FEN provided only capital investment loans. I n general, the purpose of these loans -- including those made under 2401-CO -- was t o supplement the sources of financing for projects already under  construction, many of which had already been studied by the World Bank or IDB. Consequently, it wae considered that there was no need for further appraieal, because one or other of the Banke had already carried out prefeasibility, feasibility and deeign studies on these projects and issued appraieal reports. Nevertheless, in 1987 FEN began to establish appraisal groups to deal with new projects, using the methodology developed for IDB Loan 237/IC-CO. Internally, the loan allocation proceee involver the technical, economic, financial and inetitutional analyeie both of the projects themselves and of the enterprises submitting applications. The appraiaal document is examined by the Internal Credit Committee. This presents recommendations to the Office of the President of FEN, which in turn submits its proposals to the Board. 6. Paras. 5.08 and 5.09 [now para 5.10 and 5.111 The projection models used by FEN and the power sector are baaed on those used by IDB (SPMOD), as revised and upgraded by FEN with IDB approval. These models replaced the FAST used by ISA and the World Bank, with the latter's approval. Consequently, any criticisms in this regard should be discussed with the multilateral banks in question. As regards management of information, this ia of course particularly difficult when so many agencies are involved in providing the basic parameters necessary for preparing projections (e.g. the macroeconomic scenario, pricing, the energy balance, investment, etc.), particularly in light of the coneiderable institutional complexity characterizing the power eector. As a result, models have to be used that can properly represent theme parameters and that have a sufficient degree of detail to be able to produce projections that closely reflect likely developmente in the sector. Naturally, it is aleo useful to have simpler models (and these have already been developed), but they should only be used for eeneitivity analyses, once orders of magnitude for the projections have been calculated using the more detailed models. 7. Para. 5.18 [now para 5.201 As regards mobilization of funds in local currency, the Report states that, in ite initial years, FEN was unable to carry out a massive transfer of reeourcee. It explains (quite accurately) that the main cause was both the amall eize of the Colombian financial eector and also the lack of confidence shown by banks and corporations toward the power enterprises. Nevertheless, no reference is made to the fact that the level of funds that could be mobilized and the financial conditions applied to them were governed by the severe restrictions imposed by the monetary authorities, particularly as regards volume. The result was that the original targets were not met. The position changed dramatically in the final year, since, in November 1990, the Monetary Board (Junta Monetaria) removed the restrictions on the volume and interest rates for FEN'S borrowings. At that time, FEN eecuritiee circulating on the local market totaled ColS54.5 billion. By December 1990, the figure had risen to Co1$66 billion, and over the past year has progressed as follows:  FEN SECURITIES IN CIRCULATION: 1991 BY END OF MONTH MILLIONS OF COLS March June September December Over a single year the funds mobilized have increased more than sixfold, as a result of the measures for reorganizing FEN and, naturally, two other points of which the Report makes mention: the agency's excellent image among investors, and the mobilization mechanisms developed since it was established. These resourcee have provided loans for the enterprises, mainly to be used for servicing government-guaranteed debts. 8. Paras. 5.39 and 5.40 [now 5.56 and 5.571: As has also been mentioned, the reorganization of FEN has been a gradual process, based on the legal reforms of 1990 and 1991. The first of theee wae Law 25 of 1990, together with the corresponding Regulatory Decree (1806/90), by which the agency became Financiers EnergQtica Nacional S.A., with a broader sphere of activities both as regards the enterprises that could apply for loans and the categoriee of projects eligible for financing. In addition, the system of guarantees wae changed. Another reform (also approved in 1990) allowed FEN to make direct loans in local currency, with rediscounting one (but not the only) option for the placement of local credit. The 1991 financial reform increased the potential scope of FEN'S activities by classifying it as a "financial corporation," thus allowing it greater flexibility in its activities vis-a-vis the power sector. (A more detailed account of these points can be found in the documentation concerning the negative pledge clause, prepared by FEN and forwarded to the World Bank. ) All these changes, introduced over several years, have made it possible for FEN to function properly as a development bank, controlling the enterprises' financial management. As such, its role has been confirmed by the present Administration, which drew up a reform plan for the sector in a document approved by CONPES on May 21 this year (Estrategia para la Reestructuracidn del Sector El6ctrico). This defines the functions of the various institutions and, in particular, gives FEN responsibility for entering into performance contracts that set management targets for the main enterprises in the sector. 9. The Report also states that FEN has become a window for quasi-fiscal eupport for the power sector, through the capitalization of the institution. This is not correct, because FEN'S capitalizatione were originally the result of transfers of resources from Fondo de  Deearrollo E16~tri~0,and later came almoet exclueively from FEN'm own earnings. The agency ha8 managed ite reeourcee and ieeued loanm am a financial institution under the eupervieion of the Superintendency of Banke, while seeking to maintain the molvency and mtability of the enterprise in question ae part of the adjustment proceee in the power eector and maintaining proper etewardehip of the reeourcee invested in FEN by local eavere, the international financial inetitutione and the Government itself. The Bank'e comment appears to refer to the mechanism operated by FODEX (the Foreign Currencies Fund), financed through accountm eetablimhed by both the National Government and Banco de la Repdblica. Thie ie ueed to meet the external obligatione of a number of organizatione, mainly in the power eector. Recently (about one year ago), FEN established a line of credit to provide loane for enterprieee to enable them to service their external debte in a timely manner. However, theee loan8 are also granted through FEN'e credit facilitiee, in compliance with the principle8 laid down in ite Credit Regulations. Consequently, failure on the part of the enterprieee to meet their obligation8 toward FEN would reeult in the muepeneion of diebureemente and a poeeible epeeding up of the loan proceee. I hope that you will find theee comment8 useful in preparing the final vermion of the Report. We would be very pleased to discuss them with you if you coneider this necessary. Sincerely, /e/ Cgear Gonzdlez MuAoz President, Financiers Energetics Nacional S.A  PROJECT COWLETION REPORT COLOWBIA POWER DEVELOPHENT FINANCE PROJECT (LOAN 2401-CO) PART 111. SUPPLEHENTAL I~OWATION ---------------------------------- TABLE 1. RELATED BAAR LOANS .................................... h u n t Year Die- (in US$ of bursement m i l l - Apro- ae of Number and Title ion) .................................... ------- --------- va1 Borrwer ----------- --------- 4130192 Purpoee ............................................. 1. 38-CO: Achicaya Hydro-electric 3.53 1950 CVClCHIDRAL 100% Anchicaya unite 1 and 2 (2 x 12 HU hydro) 2. 39-CO: La Ineula Hydro-electric 2.6 1950 CBEC 1OOX The Ineula unite 1 and 2 (2 x 10 IW hydro) 3. 54-CO: Labrija Hydro-electric 2.4 1951 LMRIJ A 1OOX P a h e unite 1 and 2 (2 x 4.4 Hd hydro) 4. 113-COXAnchicaya Yumbo Power 4.5 1955 CBIDBAL 100% Anchicaya unit 3 (20 W hydro) and Yumbo unit 1 (10 W thermal) 5. 215-CO: Yumbo Kxteneion 2.8 1958 CEIDRAL 100% Yumbo unit 2 (10 W thermal) 6. 217-COILa E-lalda 4.6 1959 CBEC lOOX La K-ralda unite 1 and 2 (2 x 13.3 Hd hydro) 7. 225-CO: Gundalupe 12 1959 EW lOOX Guadalupa unite 1 and 2 (2 x 45 IW hydro) and Tronerae unit 1 (18 Hd hydro) 8. 246-CO: Bogota Powr 17.6 1960 EEEB 100% Laguneta unit 4 (18 IW hydro) and Zipaquira unit 1 (33 w thermal) 9. 255-CO: Yumbo I11 Calima I Pwer 25 1960 CVCICEIDRAL 100% Yumbo unit 3 (33 W thermal) and Calim unite 1 and 2 (2 x 30 W hydro) 10. 282-CO: Second Gudalupe 22 1961 EW lOOX Tronerae unit 2 (18 HU hydro) and Guadalupe unite 3, 4 and 5 (3 x 45 HU hydro) 11. 313-CO: Second Expaneion 50 1962 EEEB 100% Zipaquira unit 2 (37.5 Hd hydro) and Colegio unite 1, 2 and 3 (3 x 50 W hydro) P, P, 12. 339-C0: Pwer Expansion 8.8 1963 CVClCEIDBAL 100% Calimo unite 3 and 4 (2 x 30 HU hydro) 00cr m P m 13. 347-CO: Cospique Pwer 5 1963 ELECTBIBOL 100% Coepique unite 2 and 3 (2 x 12.5 IW thermal) P P O 14. 369-CO: Hare 45 1964 EPU 100% Guatape unite 1, 2, 3 and 4 (4 x 70 W hydro) h h) 15. 537-CO: Third Expansion 18 1968 EEEB 100% El Colegio unite 4, 5 and 6 (3 x 5) W hydro and Canoe. (1 x 50 W hydro) IIIIIIIIIIIIIIIIIIIIIIIIIIIIIII=IIII~IIIIIIIIIIII~IIIIIIII~IIIIIIIIIIIIIIIIIIII~IIIIIIIII.IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII~IIIIIIIII~  TABLE 1: Related Bank Loans (Continuation) .......................................... Amount Year Dis- (in US$ of bursement mill- Apro- as of Number and Title ion) val Borrwer 4130192 Purpose .................................... ------- - ----------- ............................................. 16. 57540: Pwer Interconnection 18 ISA Central System Interconnection (230 kV trans- mission line and Substation) 17. 681-CO: Chivor Hydroelectric 52.3 ISA Chivor 1 (4 x 125 hydro) 18. 87440: Guatape I1 Hydroelectric 56 EPtl Guatape I1 units 1, 2, 3 and 4 * (4 x 70 LIW hydro) 19. 1582-C0: San Carlos I Hydro P w e r 126 ISA San Carlos I (4 x 155 HW hydro) 20. 1583-CO: 500kV Interconnection 50 GOVERNMENT 500 kV Interconnection Central System/ Atlantic System 21. 1628-C0: Hesitas Hydroelectric P w e 84 EEEB El Paraiso 3 x 90 W ;La Guaca 3 x 100 W pumping 3 x 10 W P ; Sesquile dam strengthening 1725-C0: San Carlos I1 Hydro P w e r ISA San Carlos 11 (4 x 155 LIW hydro) 1807-CO: Bogota P w e r Distribution EEEB Bogota distribution 1868~CO:Guadalupe IV Hydro P w e r E m Guadalupe IV (3 x 71 HW hydro) 1953-CO: Playas Hydro P w e r E m Playas (3 x 67 LIW hydro) 1999-C0: Village Electrification CORELCA Atlantic coast village electrification 2008-CO: Guavio Hydro P w e r EEEB Guavio (5 x 200 LIW hydro) 240140: Pwer Development Finance FEN P w e r development finance FEN P w e r development finance (Cofinancing) 244940: Rio Grande tiultipurpose E m Rio Grande Hydro (3 x 100 LIW hydro) 263440: Bogota Distribution I1 EEEB Bogota Distribution I1 2889-C0: P w e r Sector GOVERNHENT P w e r Sector Adjustment  Table 2: PROTECT TlMETABLfi WCredit Signature - June 27, 1984 LoanlCredit Effectiveness May 1984 - September 27, 1984 Loan/Credit Closing December 31, 1986 December 31, 1987 December 31, 1987 Loan/Credit Completion - April 30, 1987 i Time taken to prepare project was 11 months. b\ Date of combined Issues/Decisionspaper. - c\ There was no formal Appraisal Mission (para. 4.1).  TABLE 3.1 LOAN DISBURSEMENTS (INTO SPECIAL ACCOUNTS) I Disbursemts ('US$ Millions) Semester Bank Fiscal Ending Actual as Year and Calendar Estimated Actual 4% of Semester MonthiYear Cumulative Cumulative Estimated I1 1. Loan 240140: - -- IIa2. Midland Bank (USS) CofinancingGmup (kludii Bank loan B-6) - (US$ Millions)- II 3. Industrial Bank of Japan (Yen) CofinancingGroup ancludingBank loan B-7) -Qapancsc Yen Million)-  TABLE 3.2. SUB- DISBURSEMENTS (US$ Millions)  Table 4.1 FEN's LOCAL CURRENCY BORROWINGS AS ESTIMATED AT APPRAISAL COMPARED WITH ACTUAL AMOUNTS (Co1.S Millions) Source: Appraisal estimates are from SAR No. 4771-C0, Annex 2.12. Actual amounts for 1983-1989 are from FEN'S 1989 Annual Report (English version), page 54, and for 1990 from audited financial statements. -11 Percentages shown for 1988-1990 are related to estimated year-end amount for 1987.  TABLE 4.2 FEN'S LOANS TO POWER COMPANIES FROM DOMESTIC FUNDS AS ESTIMATED AT APPRAISAL COMPARED WITH ACTUAL AMOUNTS (Col.$ Millions) Appraisal estimates are from SAR No. 477140. Annex 2.12, using total loans to power companies less external obligations and less estimates of official funds which did not actually materialize. Actuals are from audited financial statements. I/ Percentages shown for 1988-1990 are related to estimated year end for 1987.  TABLE 5.1. PEN'S RETURN ON EQUITY COMPARED TO INFLATION, AS ESTIMATED AT APPRAISAL AND ACTUALLY ACHlEVED (ColSmillions and 96) -11 Appraisal estimates are from SAR No. 4771-CO. Average equity is an average of beginning aud end of year net worth amounts as estimated in Annex 2.12 . Profits are the yearly net incomeamounts as estimated in Annex 2.14. The rate of return is a calculated percentage of profita to average equity. The inflationpercentages are the expected local inflationrates as shown in h e x 5.40 of SAR. -21 Actual average equity amounts are an average of beginningand end of year shareholders' equity as reported: for end 1982in Annex 2.12 of SAR Report No. 4771-CO; for 1983-1989at page 52 of FEN'S 1989h u n l Report (English version); and for 1990in the independentauditors*report for that year. The additional ColS 19,760 million of equity capital paid in by the Government in December 1990h been omitted from the year end 1990amount. Had it been included, the average equity for 1990would have been ColS 72,953 million. Actual profits for 1983-1989 are the amounts reported at page 53 of the aforementioned 1989 annual report, except that foreign exchange gains credited directly to shareholders*equity have also beon included (see Annex I, para. 9). The profit for 1990is the amount shown in the independent auditors' report for that year similarlyadjusted. The resultingadditiom to profita are ColS 992 million for 1988, ColS4.233 million for 1989, and ColS 3,379 million for 1990. The rate of return is a calculated percentage of profits to average equity. Excludingthe foreign exchange gains credited directly to stockholdera' equity, the rato of return would be 13.6% in 1988, 5.6% in 1989and 13.9% in 1990. The inflationpercentages are calculated from consumer price indices (1985sloo), period averages, as reported in InternationalFinancial Statistics, 1991yearbook, published by International Monetary Fund.  TABLE 5.2: - ----------.----------------------------------------------------------.---------------- FEN COllPARISOW OF PROJECTED VS ACTUAL SUMMARY BALANCE SHEETS ( i n current ColS milliars) AND SELE'TED PERFORMANCE INDICATORS Liquidity Loans t o Pauer Canpenles Fixed Other assets Total Assets I - Local Obligations 8500 18500 30300 45000 62900 6036 7540 11878 12272 lm 28637 48885 65729 External obligations 25230 35167 60709 91854 0 29078 64243 95149 131134 216456 337138 528669 Officlal Funds ------- ------- -----.-.------ ------- ------- -.-.--- ------- ------- ------- ------- .------ 4369 6139 ------- 8429 11391 826 1194 ---.--- 2537 6769 7959 17131 34335 49104 Total Liabilities 0 8500 58099 71606 114138 164145 1 6862 37812 80658 114210 156801 262424 420358 643502 1 Net Uorth 10148 12439 15569 19765 2-2 305951 13637 17373 21819 25590 29989 35441 41843 72953 Net l n c o r I/ 17811 3150 4176 4837 5W3 3003 3690 44% 3)R 4434 5454 6402 11351 ------- MTIOS Dcbtfiq~it y 2/ 0.7 3.1 3.6 4.6 5.4 0.5 2.2 3.7 4.5 5.2 7.4 10.0 6.8 Gross Spreod 3/ 8.5% 4.m 4.6% 3.5% 3.0% 14.6% 6.7% 4.4% 2.7% 2.4% 1.8% 1.4% 1.6% Incremnt i n Net Yorth 22.6% 25.3% 26.8% 24.5% 24.4% 34.4% 27.4% 25.6% 17.3% 17.2% 18.2% 18.1% 27.1%5/ lncrcrnt I n Loans Outstdin0 4/ 2.2 3.1 1.4 1.5 1.4 1.7 2.5 2.1 1.6 1.4 1.6 1.5 1.7 ------------------------------------------------- 1/ A c t u l net incola for 1986 through 1990 has been r a t a t d m errplaind i n the second parwrqh of footnote U of Table 5.1 2/ Total inrkbtcdrms over net w r t h 3/ Net I n c a r as a p r c c n t w of total w e t s 4/ Net of r e p a m t o 51 B d m pr-adnet-worth of ColS53.193 m ercluslve of ColS19,760 of equity c8pital paid i n by the Goverment i n Deccnkr 1990 Source8: -*.----- a/ kllbr 2.13 of Staff Appraisal Report NO. 4771-W. b/ FEN - l n f o r r & evalkmclon final del c r p r a t i t o FEN-BIRF 2401-W; 8nd FEN- 1990 Annual Report Flle: TABLE-52.W  68 -OF co- - p Forapdbdpvporr&hnLrpprwdub-b.ry&tamtul..Lud ch.M e a nPllla d I..; I a i d ~ l n o o ~ a a p c l d . e s o l l l l t o t t b . ~ k a L . d t h u w i t r 0 ~ ~ . . . p o d i b u r - ~ ~ For hnlMt m h h w 432.433.536 F o r o l i n d t d . m o u n t o f r r t r o u U v m ~ d OWIbrdp- F o r ~ d u l . . L u p v o n v d u d u ~ d u m ~ r r b k & h a l ~ I o I o f P-PtuPdn(Alrrmrrt; &dldapdlld~ r r m t l o ~ r h . l l b . p r u o t d 0 t b . ~ o a o r b d o r n D r ~ 3 1 . 1 ~ ; d r h . l l b . ~ i n ~ 1 p u t o f ~ 0 b . n u d o 0 I C E L d C O R E L . C A d y U h n L h r k a ~ m ~ t ~ c h . n N m r m b r 3 0 . 1 W I d t b ~ 6 n ~ d d ~ ~ . d o r y O c h . h n l r f o r O r ~ p m r ~ ~ m ~ v . ~ p ~ d p r l ~ ~ ( i . ~ m l # U t h r o y h F Z N t o . m p b y ~ b y J ~ 1 4 . l W t o ~ & k . r k n l r . l ~ ~ m-gU=k prgmd (0) 01) ( S u p p l I*)~ u s k hnlr wd&hm 0a l b u mQULidaryb.ny;-b hru.532439 F E N t o n u i n t . i a p m c d u n r d r ~ ~ ~ O d ~ r d r c o n l t b . ~ o f ch.Projut md Yhmbpmj.C(:  to br a d t d d jby Indopadmtauditon acc.mbl. to th.B.nlr; tobrubaiadbtbBralrbyMurh31ofachyu; To Wud. m audit of tlu Nwt.d ~ Q t u r a ~ t o ~ r r t o r t b B . n L a r * p o r ( o t w h w o p . . n d d . S . U u t h . B . n l r Th.Bank dldnot- rrroprbly npurtr Ibodng: Uul luch inlomadaa k induddhtbeuutrfi 0) no kbr thn 60 day. aftu th. nprtd,Itr pwfo-r lrpoltlslhdbdlmda durbgth. 1 2 m o p t h p u l o d p n c ~ t h . a d o f t b ~ r t . r u l d p m J c t d l o r 3.03 (a) t h . f ~ l 2 m o ~ t h p ~ 01) k r a ~ m k ~ ~ ( O & y m r h r r h . d o f ~ ~ y u , u p B - d . w PanS29 p m j r h of ZEN'# b&w. h m u , .c.t.mda of b c c u mdmtnauatr of ~ rrkmi.drmwu wurcr mdapdkadoa of b d a for tb following wm-yupubd. muk f o r b yur 1986-19% 4.040) YENto m.in,.indobUqulty ndo mpclRd in & c h 4.05: Notnut RN a) FBN to takamuombl. utlon to bAq muh ndo to within tbL lLnlLh dac* 1988 5 1 M U pmmptly U tb nth t acrdd. 4-Mb) FEN'm dobUequity nth not to prcad7:l drh any n m bornwing; Not nut P u u 5.49-SSS 4 . m ) PEN, P.b-ry of rchyean a) ~ t.e~or~ly . to aulbodolgl~ t l ~ rtouBt~ ~ :~ L ( A ~ P T p Q d l ~ 1 0 k t O c u m d b y F l L N ~ f d o ~ l t ~ O ) l o r r J d & ~ ~ t l o n w r r l c . ~ o f P E N ~ r t h . u a u p r i o d l a) Mtnuina rumororth.12-monthphl,SO%eonrutlbl. to cub rrlthLcr90 Md &jm, bdmm comrdbl. mithin 30 &jm. m t l u rh.nth.mumofi (A) 116 &td apmQtwa for th.phod, plru; @) 142 FEN'# ariuubdI d drbt rmorrlvllon wrrlc. nqdnmmh forth. YM p.rlod. 4.m ~ ~ r o ~ u ~ ~ . r t . p ~ ~ ~ ~ r b u r r u l ~ i r p m MII ~ h n t r ~ t ~ . u c l u a g m d k 1- opontio~. 4.09 BuakmdFEN, atL.raquutof dthapy, t o a r d m u f m m U m e t o Mat Um withwardtoadmbhtndaa, opn- dllnsdalcolldikuofPEN a o d m to f u d A d h ~ f o ~ p u d o l l k a-kb~lj y w c a v e a-QI1. o p n w uld fio.actldh. 4.10 to ~ b hnL n p r r c ~ ~ dtow h bpRt ~ r d darad b h&&OE r 4.01 MOl (.).nd~*Mtdocumacr; 4.11 PElYbW~tor.ndB.nL~Ikcarb.r31.1Wur.LupLlbl.pl.nof NotM* P a w . S.W.16 &Ionto aubl~EZNtomh-inth. C o h m b h u p i t d m u l u r o n a loq mdd u m mmhub; 4.12b) ~ b ~ ~ ~ m h ~ ~ C d o ~ ~ ~ ~ t h n C d 0NOtMU0 , ~ m Pi ~l N5.16-6.18 n ~ ~ l b 1% 11.800 milhaduhg 1985.14.700 nJllbnd d q 19sLend 17.W mWon -19s7; 4.126) F E N . C I v l r a t D r ~ B m k r h . l l ~ r d m u ~ k ~ ~ A p r i l 1 o f r c b Md Pan 5.17 y r r on PEIU'm p r h r n u l u m of it. born* a d lcd& o w cb v.aku 12-montb priedandit.pmjactd b o r n e a d 1- opontlom d d q t h . ~ a K c * l 2 ~ p r i o d . .I  the powr rector -tor oa thm Q P a m 5.10 5.12 - m t to Wre action. r wlutlon for the dlvolution or Uquidntionof FEN frvonble t h givm to d d r r ~ Elnndd irrlon~tionfor pow- cornpanic.: E.chof the Powu Cornpanic.IIuU: &re itm vcormtl utd fhuaddat.- r w b d by indepodsnt auditon Met .cc.pt.ble to the (b) M d e FPN Ltarthan 4 montb &or md of each y n r c d d topic. of the Met 6aud.l for m u b year: (c) hvideFENrnthtarrhn6month1 r h m d ofeach ynrup-to-dat. Not Met projectiou of the nDuwid &teaad# for the f o U o h mwmyear pried; (a) provide FZN with ocher 6n.nci.l i n f o n ~ t b nu FEN rhllreuoorbly mqud to Not Met ~ b l m F E N t o m l i n t l i n . n u p d . t d ~ i . I ~ o f t h e p o w u u w p . n ~ . 4.01 CPdW apeadhum lhit.tbn. A power eornpmy b not to commit Ltwlf to n m upltal l n p ~ d I M other thuiongobg4rlu until dz-projeeh a n complete unlar: (a) propod nn* projrt b commiullyj d l i d ; and (b) a m 4 ulpendi-, Including new project, .h.Unot a c e d 1%of currant n a Met vdw of itm h r d MI& for the w r r a p o o ~ year, or U it &a, it hu obtained 6naming undw cnaditiom whth will not adrandy affectitm M d waditiool dp r f o ~ofeItmobti&ahutoPEN. (Suppl.manU1lattergivatirtofom going pmjectm .~[dudedfrom thu condition)  -. fOWEfl PART I mNANa@IG STARlSOP , RgpggWCeO&. AGREBMENT I)lsaPmM COMPLIANCE REkLARKS 4.04(a) EPM, CVC W u d i n g CHIDRAL). EEEB. CORELCA. and W to mlintllin Not Met tariffs mauing nta of rr(um M foUou.: In 1985md thaufbr - , 7% 10% CVC 9% 9% EEEB 14% 14% CORELCA 11% 11% m 9% 9% (b) th.pomr wmp.nles of (I) to &.ra in the B n t hro month of eachquarter of Not M u MetoalpbyEPMaadISA th.W yar, the adequacy of the& tariffs and hunisbto th.Bank and FEN the M L t a of w h re4.v: (c) h d d a porrvcompanymtarrpet tocamthe y & n b o f n ( u r n f o r t h e n a t Not M u 124mntb period ineludiq the quarter of the mprt, it b u l d adjust ib tariffs wrdingly to be in effet m later t .the md of the fohwhg qurtar. 4.05 EERB to takm Np (including adj* it. t a m s and ebyu for the supply rnd Not Met d e of dsctrkity) u rsprdrsd to h Rm& h9m inrunrl murca that gmentd by EEEB hpll be equivolcotto mt leu than foUowhg prcentym of EEBB'r invahnpnb hk e d nu& aad ISA for esrhyear - - - - 1984 1985 1986 1987 35% 1% 55% 55% 4.06 Except u FJW otherhe rgrsa, each Pomr Campmy a h d l not 'hew 4long U r ~ e & i n EPM complied. ISA term debt d e v a runonable foncut dom that for syh gear dwing the term of nquatadaad lscdvd the debt, the debt servicewvorye tat indiestor Cm&d cd genendoddsbt arcmpdoru from the aewice rsputemat) would ba not l a a rbPn: rqui~wnmt (I) CORELCA 1J @) 19A.EPM, EEEB, CVC IS - aii) ICEL 13 (from 1/88) 4.07 Euh pomr company, M r lhPrrbol&n of FEN.to tpka 9uaon n e c v to Met PEN &a not p y 4cnrh dividend or m p u c h u Ita ahamof ltoclr for cd. 5.01(a) ISA to coopamtowith FEN to auble FEN to fulfill ita obligatioru under SSbcon Met 4.13 of UuLoan Agr~~mmt. (a) ISA to -bulge v i m with the Gunnutor, the BPnL md FEN on M annual b S Not Met PIN 5.105.12 punrpnt to pmvtionr of p n g n p h (b) of secdon 4.13 of the Loan -rot. 5.02 ISA to cooperntowith G-tor to prepre price indar for the ElectricitySabr Not Met On29 hf8y 1985,LPA unt pvnvnt to Sacdon 3.06of the GumnuteaAgnanmt. the Govemrruotr propoul for wmpudnt a h M tadex, but the GOVMDUP~ bok no .ctloaoathL pmpoul. AIM s- pan. 527 5.03 ISA to pmpm m t lrbr than April 30 in each ymar a study on updntad Met P a m 5.105.12 dahicity d- g m d projcdow and m r h v ~ mpmg- t for pvparof the &ow of Section 3.13 of the Guamtm -mt 5.04 ISA and 1CR.L d ISA and CORELCA. jointly with Gumator, b prepare Met P a n 33 w h e m f e d to in sub-paragraph(c) (ti) (B)of Sadoa 2.03 of Loan Agrsemcnt md Wte dl utioru nacaury to auble Guarantor to fulfill ib obUgatiowunder paragraph (a) uetion 3.11 of Grunnta Agmuncot 6.01 PEN may a v p d dhbuncmcnt of suMouu if (I) I power company defaultr in Not rpplld P a n 537. pymenb to FTiN under W o a m or other loan aysanmt helwem~FEN rnd the Although them w m power company; or (b)a power company defaulta in the prfo-e of lqy htmca\mdv (a) md(b) other obligationby a power company under the Power Agmund or whm FXN auld have actd #ul*idiar~Lorn Agroemmt or nqy other Loan Agroarmnt -em FXN md the it cbou m t to. Th. rspsctive power compmy; or (c) PEFPs ri&t to mnke withdrawah m&r th. wnditlon In (c) nner a m . project loam u suspended.  aithoutthis condition b& impleme&athn of the project did not meedin fall. Reportedly,more recntaMlysiciadicates that the project now meets nnthereqlhduitorin. weptfor aeom- fhccing~whiehh expechdtobedosedhthe near term. 3.03(a) G ~ ~ ~ t o r t o ~ d a C t i o m t o d l e F E N t o p ~ p u e p l n n o f & t o NotMeI FENdidwtpreparestrb d mresourceh theColombii capital marketreferredto in Sdon 4.11 a PIpO. See parar 5.U-5.16 of the Loam AgreemmG and (b) prompuy put PIpO illto effect 3.04 Gnar~ntortoLPLeor-etobetaltmPndru=qtoerPbleFEN Met FEN did not meet targets to meet itr borrowkg targctr hthe Colombian capital marketre(fonh in but thi, wns not due to lack pamppb (a)of Section 4.12 of the Loan Aprrammt. of Governmentsupport relacadtoaccerrtothe cawmarket 3.OS(a) n o t L s t o r ~ A p r i l l o f e e c h y c a r , G ~ u r ~ n t o r t o ~ e r i e r n a i t h t h e Met Para5.11 Baok and FEN for the purpose of paragraph (b) of k t h 4.12 ofthe LOM 4Pe-ePt; (b) Not Inter than Ang& 31 of BDfh year,Guaraniar to achange riear with the Not a~ei Pam 5.1C5.12 Bank, FEN and ISA for the pupme of k t b m 4.13 (b)of the Lorn Agreglmt. 3.06(a) Gunmutorto famish to B m k not Inter than D~rembv31,1984 for review Notluet Para 531 andcommenb, propwal for the comptation of price index of variation of prices of demicsectorgoo& and &IS; (b) star@ Jnnnny 1,1985, commtcand pblisb the v- hthbprice Not M ~ I Para537 ides. -- 3.07 Gluru~tortotnLcdactionatodthepowwcomprnierto~d Not luet Paras5.16,5.20,537 ma;nta;ldectridtytarifhdehargeratlerelr~tofal6ll&of thpir obligationsp m m ~to Section 4.04 of the Power h a d q Agreunmt. ~ t 3.08 GuarantorondortPLesto takaall vtioruto cnavc the timely procarrm~~ of Met the goods and services mqwhd for each projeEt. 3.09 C n a r P n t o r t o t n L c P n a c ~ m o n b ~ ~ m h p h p r ~ ~ o f ~Not Met . to d b Powv Companies to perform their raspeetiveobligationsC i d i q fhadall under thePower Aaamiq Agreement. 3.10 No taxer to be levied on Power Fhnchg Aprranent. Met 3.11(a) Prepueaodhmkhto Bmkmlaterthau N o v c m b e r 3 0 , 1 9 8 4 , t h e ~ d Met Pam3 3 schemesr8ferredto in sub-pnmgrpph (c) (ii)Seetion2.03 of the Loan of A V - d . (b) take,plrd cawICEL and CORELCAto take,PO diem-, Not applicPble Pam 3 3 hc4ubiog h d y provision of budgotuy r r s o ~to mbb mrh power ~ ~ , compcllliesto ury out the plans Pnd schemereferredto. (Cf U e P n w t i o P r n e e ~ , ~ h ~ p r i d o n o f M , t o d l e Met Applicable ody to ICEL and CORELCA to make paymmtr Feqclirodto &e the sub-baas CORELCA and FEN l o w made to them by FEN under the pmjsc  - t 0 u r r y w t u l ~ v e m ~ p l p l l b 1 ( 8 d ~ m ~ m m ~ - - aipBlrforbr,dartL of sbuly urr4d ad by ICEL'r Dirkion& Aworir Conhd M m h k a i ~ o &ah-- .ot thru 31,1984; --a- 31,1984, r p d h i n qeztbata of ICEL'r thecdm, to mrlw the  - 83 - Table 7: USE OF BANK RESOURCES A. STAFF INPUT 12401-C01 (Staff Weeks) Bank FY 1983 1984 1985 1986 1987 1988 1989 Total Through Appraisal - 20.9 16.2 37.1 Negotiations 27.6 27.6 Supervision 2.4 6.3 11.9 6.7 4.6 1.5 33.4 Total 20.9 46.2 6.3 11.9 6.7 4.6 1.5 98.1 B. &4ISSION DATA fLOAEl2401C01 Post-Appniul I 3/83 3 16.6 80-0.2PE 2FA,2CO,U) Po&-Appniul I1 4/83 1 1 1 FA Port-AppniulIII 10183 2 2 1 FA SupervisionI 9/84 1.5 1 1 PE 1 SuporvisionI1 3/85 1.5 0.5 1 PE 1 Suporvirionm 10185 1.5 3 2 (PE. FA) 1 L SupervisionN 8/86 3 3 1 CO 1 M TOTAL 16.5 33.1 i\ 1 -- Probkm-freoor minor problem; 2 = Moderate problem; and 3 = Major Problerm - F F-ial; T Technical; M = Management e\ PE = Power Engineer; PA = Financial Analyst; E = Economist; LO = Loan Oficer. CO = Coonrlta~,LA = Lawyor -d\ No infomution available in the Project File. Mission to promote participationof CoGncein in BLoan.  JwmU Page 1 of 12 PROJECT COMPLmION REPORT COLOMBIA POWER DEVELOPMENT FINANCE PROJECT LOAN (2401-COI UNILATERAL SIX MONTH EXTENSION OF SUBLOAN AMORTIZATION PERIOD 1. By letter dated July 19, 1988, FEN informed t h e Bank t h a t a t a meeting on July 14, 1988, i t e Board of Directore had adopted a reeolution which extended t h e eubloan amortization period by eix monthe. Under t h e reeolution, effective August 1, 1988, t h e eix month exteneion wae applicable t o amortization paymente due between t h a t date and January 31, 1989, and t o a l l eucceeeive paymente previouely scheduled on eubloane made from t h e proceeds of t h e "A" and "B" loane, ae w e l l a s certain eubloane from local reeourcee. Thie action wae taken without prior coneultation with o r coneent by t h e Bank. Section 3.02 (b) of t h e Loan Agreement, however, provides that, except ae t h e Bank ehall otherwiee agree, FEN s h a l l not amend, o r f a i l t o enforce any provieion of, t h e eubeidiary loan agreemente entered i n t o f o r relending t h e "A" and "B" loane. In order t o determine whether t o approve t h e action taken, t h e Bank requested FEN t o provide it with financial projectione for the next two yeare eo that it could eee how FEN'S financee w e r e affected. The Bank aleo aeked FEN, when similar caeee a r i e e i n t h e future, t o provide it with an opportunity t o comment before decieione a r e taken, a s required by t h e loan documente (letter dated Auguet 16, 1988). 2. The information provided by FEN i n reeponee t o t h e Bank's request i e available i n t h e project f i l e e i n t h e form of detailed financial projectione f o r t h e period 1988-1990, dated September 12, 1988. An internal etaff memorandum dated November 28, 1988, refere t o , but doee not include, actual and forecaet financial statements f o r t h e period 1984-90 prepared from t h e information received from FEN and available i n t h e f i l e e . The memorandum etatee t h a t theee show t h a t t h e extension of t h e eubloan amortization period w i l l not have any material e f f e c t on FENte financee and t h a t FEN would continue t o have adequate financial ratios. On t h i e baeie, it recommende t h a t a letter be eent expreeeing no objection t o t h e modification of t h e eubeidiary loan agreements. The letter was sent on December 2, 1988. It refere t o FEN's letter of September 13 and telefax of November 11 eupplying information requested by t h e Bank (both of which a r e mieeing from t h e f i l e e ) ; it emphaeizee t h a t t h e procedure u t i l i z e d by FEN t o modify t h e eubeidiary loan agreemente without prior agreement of t h e Bank wae not i n accordance with Section 3.02 (b) of t h e Loan Agreement; it etatee t h a t ae an exception i n t h i e inetance t h e Bank hae decided t o approve t h e s i x month extension of t h e amortization period of t h e eubloane t o t h e power companiee; and it concludee by requesting the pereonal intervention of FEN's President, t o whom the l e t t e r wae addreeeed, t o eneure e t r i c t compliance i n t h e future with t h e provieione of t h e loan documente. 3. The only specific information i n t h e aforementioned memorandum i e a l i s t i n g by company of t h e amounte of t h e amortization paymente poetponed f o r eix monthe,  ANNEX I Page 2 of 12 totalling USS18.9 million for the subloans made using the proceeds of the "A" and "B" loans and USS11.9 million for subloans made from funds raised locally. These are significant amounts and one would have expected some discussion, beyond the conclusory judgments stated, of the effect of these postponements on FEN'S finances. The memorandum was prepared by an assistant level analyst. The file copy of the letter of December 2, 1988, notes that it was cleared in substance by a senior financial analyst, but this matter may have received only cursory attention by him. Considering the record as it now exists with some of the supporting material for the Bank's conclusions missing from the files, one is left with the impression that the Bank's financial analysis was superficial, and that the Bank, having been presented with a fait accompli, decided it had no choice but to accede to the extension with the admonition that FEN seek the Bank's prior approval for any future modifications of the subsidiary loan agreements. Nevertheless, as demonstrated in the next section relevant to FEN'S profitability, the conclusion that the extension had no material impact on FEN'S finances was appropriate even though the further claim that FEN'S financial ratios would continue to be adequate was not. Im~actof Extension on FEN'S Profitability. 4. It is uncertain what financial ratios were calculated in the analysis which led to the conclusion in the aforementioned memorandum that they would continue to be adequate. It is implicit, however, that a very significant ratio was omitted, namely the rate of return earned on FEN'S equity. As shown in Part 111, Table 5.1, FEN'S real return on average equity was significantly positive in 1983 and 1984. It became marginally negative in 1985 and substantially so in 1986 and 1987, when the nominal return on equity for each of these years was 16% in comparison to rates of inflation of 19% and 23%. respectively. These historic negative returns should not have been characterized as adequate. 5. The projections supplied by FEN explicitly called attention to the expectation that the rate of return on equity for 1988 would be negative by about 10 percentage points. As shown in Table 5.1, the actual nominal return on equity for 1988 was 16.7% compared to an inflation rate of 28.1%. FEN'S projections that its return on equity for 1989 and 1990 would match the rate of inflation did not prove to be accurate. The actual return on equity in 1989 was 16.6% compared to an inflation rate of 25.8%, and in 1990 the actual return was 23.9% compared to an inflation rate of 29.1%. As analyzed below, although FEN subsequently concluded that the extension contributed to a decline in its profitability in 1988 and 1989, this is based on an incomplete statement of its net profits (see para 9). In any case, as a result of its evaluation of the justification presented for the extension, the Bank should have become aware that FEN'S rate of return on equity in recent yeare had been substantially lees than the rate of inflation and inquired into the reasons for this unsatisfactory relationship and the remedies that might be proposed for correcting it. 6. With respect to the subloans made from the proceeds of the "An and nB" loans, the postponed amortization payments were the initial repayments due thereunder at the end of the four year grace period originally provided. Of the total amount of repayments to FEN which otherwise would have been payable,  Am?zu Page 3 of 12 USS17.1 million was due between October 31 and December 18, 1988, and USS1.8 million on January 15, 1989. Since the four year grace period under the "A" and "B" loans also ended in 1988, FEN had already begun, on June 27, 1988, to repay these loans, To make up for the funds foregone as a result of its decision to extend the subloan amortization period, FEN had to increase its borrowings from the domestic capital market. According to FEN'S Annual Report for 1989, the costs of this increase in peso borrowings were not totally recovered and this had an adverse impact on its finances in 1988 which extended into 1989 (English Version, Section headed "Financial Managementw, pages 11 and 14). In the analysis which follows, it is assumed that the extension of amortization of subloans from locally raised funds had no adverse impact because the interest rates charged on such subloans include an appropriate margin over the cost of raising local funds. 7. There are two sources of income to FEN resulting from the extension of the amortization period of the subloans made from the proceeds of the "A" and "B1' loans: interest income and increases in the peso values of these foreign currency denominated subloans due to exchange rate adjustments. The latter ie the more significant of the two sources. FEN's interest income on the amount of the subloans from external sources remaining outstanding due to the postponed amortization is based on foreign determined interest rates (of about 9-10% in 1988-90) applied to the US$ equivalent value of the subloans and translated into pesos at the exchange rate applicable at the time of payment. That income initially would be substantially less than the interest costs incurred at much higher rates (of about 30-40% in 1988-90) on the peso amount of borrowings from the domestic capital market equivalent in value to the postponed amortization payments based on the exchange rates applicable when they were originally due. The peso amount of these borrowings would remain fixed and rolled aver as long as necessary throughout the nine year period for repaying the subloans. The subloans, however, are repayable in pesos equivalent to the foreign currencies in which the source of the subloans, namely the "An and "8" loans, are or were repayable. As a result of the increase in the peso value of the foreign subloans reflecting the devaluation of the peso, the interest income on the extended portion of the subloans will also increase and in time equal and then surpass the interest costs on the fixed emount,of peso borrowings incurred because of the extension of the amortization period. As it would have appeared in 1988, a period of about four to five years would be needed before the break even point is reached. 8. More important, the increase in the peso value of the foreign currency denominated subloans is also available to offset the initially higher interest costs on the domestic peso borrowings incurred because of the extension of the amortization period. The average annual increase in the US$ exchange rate in 1989 and 1990 was about 30%. Thus, taking both of these sources of income into account, the extension of the foreign subloans should not have caused a decline in FEN'S profitability for more than a brief period; the impact of the extension on FEN'S profits would become neutral within at most a year and thereafter would be positive. Parenthetically, it should be noted that the impact of the action taken by Colombian authorities effective in October 1991 to allow the exchange rate to be determined in a free market was not foreseeable in 1988. Earlier in  MmLZ Page 4 of 12 1991, other actions had been taken to revdue the peso. As a result, there has been a substantial real appreciation of the peso. In 1991, as measured by the official Col$/US$ rate, the nominal devaluation of the peso was 24.3%, about the same as the difference in inflation between Colombia and the USA: as measured by the parallel or free market rate, the nominal devaluation was 13%, some 10% below the Colombian-US inflation differential (TheEconomist Intelligence Unit, Country Report No. 1 1992, page 17). Since October 1991, FEN and the power companies have been using the parallel rate to effect transactions in foreign currency and to denominate their foreign currency obligations. 9. As reported by FEN in its annual financial statements for 1988-90, net profits include only that part of its foreign exchange gains which are credited to *direct operating incomen and exclude substantial amounts of foreign exchange adjustments which are credited directly to nstockholders' equity". This distinction is not explained but it may be related to income tax consequences. As so reported, however, FEN'S rate of return on equity shows declines to levels of about 13% in 1988 and 5% in 1989. This is the basis onwhich FEN concluded in its 1989 Annual Report, as mentioned in para 6, that the costs of the peso borrwings to make up for the funds foregone because of the extension of the subloan amortization period were not fully recwered and that this adversely affected its finances. In calculating the higher rates of return stated in para 5, all exchange gains including'those credited directly to stockholdero' equity are counted as part of net profits, as the more appropriate way to measure FEN'S perfomance. On this basis, as stated in para 8, the costs of the increased peso borrowings were fully recwered and the extension of the subloan amortization period had no material adverse impact on FEN'S profitability. Broader Context for Evaluating Post~onementof Subloan Amortization. 10. It is important to note that even without the extension it was necessary for FEN to raise considerable amounts through domestic borrowings in order to meet its repayment obligations under the nA" and "B" loans in 1988. The total of about US$44 million due and repaid on these loans by FEN in 1988 was aver US$27 million more than the postponed repayments due to FEN on the subloans to the power companies made from the proceeds of those loans . Loan repayments by FEN continued to be higher than subloan repayments to it for the next three years. The cumulative negative balance reaches a peak at the end of 1991 of US$57 million as originally scheduled and of US$75 million as modified. This is shown in Table A hereto, which presents a comparison of FEN's repayments on the "An and "B" loans with the repayments to FEN on the eubloans as originally scheduled and as modified for the six month extension of the subloan amortization period. The US$ equivalent amounts in this table, which are used in this analysis, are based on the equivalent values of the loans and subloans as of the time they were made or drawn down and do not reflect the impact of subsequent changes in exchange rates applicable to the nAn loan because of the Bank's currency pooling system and to the Japanese yen denominated "Bn loan. After a grace period of four years, the repayment period is 13 yeare for the "An loan, and four years for the commercial bank share of the "B" loans plus an additional one and one half to two years for the Bank and Nordic Investment Bank shares of the loans. Thus, repayments by FEN are exclusively for the "A" loan from  Amxd Page 5 of 12 1995 to 2001. The repayment period on the subloans is nine years after a grace period extended from four to four and one half years. 11. From 1992 through 1997, the position reverses and FEN would receive more in repayments from the power companies than it must repay on the "A" and "Bn 1 0 ~ s . On a cumulative basis the net position becomes positive by 1995 as originally scheduled and by 1996 as modified. The maximum positive position is reached in 1997 in the amount of USS43.4 million as originally scheduled and of USS41.8 million as modified. In 1998, when the final payments to FEN on the subloans are made, the amounts FEN receives from these repayments by the power companies are once again smaller than FEN'S repayment obligations. Thereafter, the remaining balance of USS32.66 million on the "A" loan is to be repaid in 5 semiannual installmenta, ending March 1, 2001, using the cumulative positive balance of funds received from repayment of the subloans. 12. Thus, FEN'S finances have had to reflect the need to borrow local funds to compensate for its net negative position with respect to repayment obligations related to the "A"and "B" loans during 1988-1991. In subsequent years, FEN'S finances will reflect the maintenance of the borrowings as long as necessary. As explained in para 8, the extra carrying costs of these borrowings should be fully offset by the interest income earned on the corresponding arnount of subloans that remain outstanding plus the foreign exchange adjustments in their pesovalues. During 1992-1997, FEN'S financeswill also reflect the consequences of annual net positive positions. One consequence will be a decrease in the amount of foreign currency denominated subloans, on which FEN earns interest and benefits from increases in their peso values as a result of currency devaluations in relation to the peso. As an offset, FEN would be able to utilize the annual surpluses to repay local borrowings or for additional local loans to the power companies or other peso investments. Except as stated in the next paragraph, these offsets should be adequate to protect FEN from any material adverse effects on its profitability because it had to borrow local funds during 1988-1991 to meet higher repayment obligations on the "An and "B" loans than the amounts repaid to it on the subloans made from those sources. 13. The situation changes when there is a cumulative surplus of funde repaid from subloans over the amounts repaid by FEN on the "A" and "B" loans. When this occurs, in 1996 under the revised amortization schedule as compared to 1995 under the original schedule, the outstanding amount of subloans will be less than the outstanding amount borrowed by FEN to finance them. From that time, FEN will bear the foreign exchange riek on the borrowings to the extent that their outstanding amount exceeds that of the subloans, and after the subloans are completely repaid in 1998, FEN will bear the entire foreign exchange risk. There can be no assurance that FEN'S earnings on the peso surpluses accumulated to repay the borrowings will be adequate to offset fully the foreign exchange risk which can no longer be passed on to the power companies. 14. In this context, the six month extension of the subloan amortization period effective August 1, 1988, has a positive effect because it reduces to a small extent in time and amount FEN'S unprotected exposure to foreign exchange risk. Its incremental impact, however, is of relatively minor importance. The  ANNEX I Page 6 of 12 significant problem which should have been recognized is the much more substantial impact of the mismatch between the repayment terms of the external loans obtained under this project and of the subloans made from these sources. One question which arises is why this was not addressed during the project design stage or during the early stage of implementation when the subsidiary loan agreements were proposed for approval. Another question is whether remedial action might have been proposed for consideration had the mismatching problem been recognized when the request for approval of the six month extension of the subloan amortization period was made in 1988. Alternatives to Reduce Mismatching which Deserved Consideration: ti) at A~praiSal. 15. One step, permissible under Bank policy, which could have been considered during the project design stage to minimize the mismatching problem was an extension of the grace period on the "An loan from four to five years in exchange for a reduction of the repayment period from 13 to 11 years. In attached Table B, the schedule for repayment by FEN of the "A" and "B" loans which would have been applicable if this change had been made is compared with the repayments on the subloans made to FEN by the power companiee shown on two bases: as originally scheduled based on a 13 year term, including a four year grace period (Heading A), and assuming that the term had been set at 15 years, including a four year grace period (Beading B). This second basis shows the subloan repayment requirements which would have been applicable under the maximum subloan terms permitted under the Bank Loan Agreement, Section 3.02 (a). These alternatives should be compared to the repayment schedules actually agreed and adopted (Table A, Beading A) to illustrate the specific comparisons that might have been made if these choices had been analyzed during appraisal. 16. Under the first basis of comparison (Heading A of each table), the subloan repayment requirements are identical and are those actually adopted and originally scheduled. This comparison indicates that the one year extension of the grace period on the "A" loan in exchange for a two year reduction in its repayment period, when blended with the shorter repayment terms of the "B" loans which were as liberal as could be arranged, would have provided a significantly better match for the subloan repayment terms actually adopted. Both the negative and positive variances would have been substantially less. The maximum cumulative negative position would have been USSS1.6 million in 1992 compared to USS57.1 million in 1991. The maximum cumulative positive position would have been USS36.3 million compared to USS43.4 million in 1997, and the period during which FEN'S debt service requirements on the "A" loan would continue after complete repayment of the subloans would have been reduced by one year. As a result, FEN'S need to borrow locally to make up for the negative differences, and the amount and extent of its unprotected exposure to foreign exchange risk after the cumulative position becomes positive would have been reduced. 17. The results are mixed for the second basis of comparison, which assumes both that the "A" loan would be repaid w e r a shorter period, a 15 year term including a five year grace period, and that the subloans would be repaid over a longer period, a 15 year term, including a four year grace period (Heading B  LBELz Page 7 of 12 of Table B), instead of the terms actually agreed for both the "A" loan and the subloans (Heading A of Table A). On this basis, the negative variances would have been much more substantial. The cumulative negative peak of USS83.9 million in 1992 would have been USS26.8 million more than under the repayment terms actually adopted and the negative position would have extended over ten rather than seven yeare. On the other hand, the annual repayment requirements for the power companies would have been very substantially reduced and better adapted to their debt service capacity. Also, PEN'S unprotected exposure to foreign exchange risk would have been minimal. 18. There is no indication in the project record that any consideration was ever given to the alternative that the Bank loan might have a 16 year term, including a five year grace period, rather than a 17 year term including a four year grace period. Apparently, this basic step in the analysis which should have been carried out during appraisal was omitted. This oversight may have occurred because the Bank personnel concerned with this project, +includingmanagement, mistakenly assumed there would be no significant mismatch between PEN'S repayment requirements on its borrowings for the project and the repayments it would receive from the power companies on the subloans made in relending those borrowings. This conclusion is reinforced by the categorical statements made, both in the Staff Appraisal Report (para 4.09) and the President's Report (para 60), that the onlending terms would be such that the full foreign exchange risk would be borne by the power companies. Had this matter been analyzed as it readily could and should have during appraisal, the problem would have been identified. In that event, it is likely that the Bank would have propored a one year extension of the grace period for the "A" loan in exchange for an equivalent reduction in its term in order to mitigate the problem. Whether the Colombian authorities, in particular the Government, would have accepted such a proposal, is less certain, but given the resultant advantages noted in para 15 for reducing both the negative variances which would have to be financed and the extent of PEN'S exposure to foreign exchange risk, it seems likely that they would have. 19. There is more uncertainty about the conclusions that would have been reached had the choice between a 13 and 15 year term for the subloane been analyzed during appraisal (para 17). Since it was agreed that the maximum subloan term would be 15 years, this was a possible, and perhaps the likely, choice. To recommend this, however, the Bank would have had to be confident that a much higher and prolonged negative funds flow from the project's financing and refinancing arrangements could be managed by FEN. At the time, PEN was a newly formed institution whose ability to raise funds from the domestic capital market had only been briefly tested under very favorable conditions (Staff Appraisal Report, para 2.13). Nevertheless, projections were made that the total resources which PEN could mobilize fromlocal investors would increase fromthe equivalent of US$220 million in 1984 to USS528 million in 1987, without crowding-out effects (Staff Appraisal Report, paras 2.17 and 3.24-3.27 and Annex 5.40). In this context, and considering the advantages which the power companies and FEN would derive from a 15 year subloan term, it seeme likely that the Bank would have recommended it. On the other hand, FEN apparently preferred the 13 year term. Presumably this was because after analyzing the choice FEN recognized that it would have a substantially smaller negative funds flow to finance and it attached  4HwU Page 8 of 12 more importance to this than to the consequent higher exposure to foreign exchange risk. Under the circwnstances, the only conclusion one can confidently reach on the choice between a 13 and a 15 year subloan term is that the Bank missed an opportunity to influence a significant decision concerning project design by failing to analyze this choice during appraisal. m~ortunitiesafter ADDraisal to Consider Alternatives to Reduce Mismatching. 20. There were two subsequent occasions when the mismatching problem should have been recognized and addressed: first, when the draft subsidiary loan agreements were proposed for Bank approval in 1984 following loan signing on June 27, 1984; and second when the Bank was considering whether to apprcwe the rix month extension of the eubloan amortization period after receipt in late July 1988 of the notification from FEN that this had been done. To illustrate choicer that might have been analyzed on these occaeione, Table C compares the repaymentr by FEN on the "A" and "B" loans as actually agreed with the repapmentu FEN would have received on the subloans if their term were 15 years (Heading A) and if the 15 year tern were also adjusted for the six month extension of the amortization period (Heading B). 21. The comparison between the 15 year subloan term (Heading A of Table C) and 13 year eubloan term (Heading A of Table A) is pertinent to the analysis which should have been made on the first occasion. The results would have been similar to those discussed in para 17 but not as good: the cumulative negative peak would have been higher (USS88.7 million, or USS31.6 million more than under the arrangements actually adopted); there would have been comparable advantage6 to the power companies because of lower annual debt repayment obligationel and FEN9# exposure to foreign exchange risk would have been substantially reduced but not virtually eliminated. As an additional part of its analysir on the firet occasion, when the focus of its review was on the subsidiary loan agreements, the Bank should also have noted and become concerned about the significant mirmatch between the repayment terms of the "An and "B" loans and of the subloans. If it had, it would probably also have considered the possibility that it would still be timely to propose to amend the repayment terms of the "A' loan to extend the grace period in exchange for a reduction of the repayment period as discuused in paras 15-17. Thus, the conclusions stated in paras 18 and 19 are also applicable as of this early stage of project implementation (para 5.1). The omission of an analysis of these choices as part of the Bank's review of the subsidiary loan agreements may be attributed to the failure to assign a financial analyst to supervision of the project at that time. 22. The second occasion following appraisal which ehould have called the Bank9r attention to the mismatching problem was FEN's decision to extend the subloan amortization period by six months. FEN'S letter informing the Bank of thir extension was received on July 27, 1988, five weeks prior to the date, September 1, 1988, vhen the first semiannual amortization repayment of the "An loan wau due. It is unrealistic to expect that, within that five week period, the Bank9e review of FEN'S decision would have been carried out so well and so expeditiouuly that the potential benefits from amending the Bank loan agreement to extend the grace period by one year in exchange for a two year reduction in the amortization  ANNEX I Page 9 of 12 period would have been identified and that the agreement of all concerned needed to effectuate that amendment would have been obtained. 23. On the second occasion, however, it would have been both timely for the Bank to consider a longer extension of the subloan amortization period and reasonable to expect that it would examine the merits of such a step on its own initiative. As previously indicated in paras 10 and 11, the Bank's analysis of the effects of the six month extension of the subloan amortization period should have included a comparison such as in Table A between FEN'S schedule for repaying the "A" and "B" loans and the subloan repayment schedule as originally specified and as modified for the six month extension. Moreover, the decision to extend the subloan amortization period was taken about six weeks after the effectiveness of the Power Sector Adjustment Loan (Loan No. 2889-CO). The Bank was no doubt aware of the financial difficulties then being experienced by the power companies in meeting their debt service requirements, and it must have assumed that this was the unstated reason for the extension. In those circumstances, one could reasonably expect that the Bank, as part of its review, would have considered whether a lengthened repayment schedule for the subloans would be advisable. Thus, the analysis should also have included a comparison of subloan repayment schedules based on 15 and 13 year subloan terms, each extended sixmonths, with FEN'S repayment obligations on the "A" and "B" loans, like that presented.in Heading B of Table C and Heading B of Table A, respectively. 24. As could be expected, a comparison of extended 15 and 13 year subloan terms would have shown that the negative funds flow associated with the longer subloan term would be substantially higher and more prolonged than with the shorter term. As illustrated in Tables C and A, the cumulative negative peak would be USS103.2 million in 1992, or US$28,4 million more than USS74.8 million in 1991, and the cumulative negative differences would be sustained for 11 rather than eight years. By 1988, however, FEN'S ability to raise large sums from the domestic capital market was well established. Also, as discussed in paras 7 and 12, such borrowings to finance the negative variances would not adversely affect FEN's profits. On the other hand, the comparison would have shown that an extended 15 year term would lessen the debt service burden on the power companies and FEN'S exposure to foreign exchange risk. It appears, on balance, that there was a compelling case to recommend substitution of an extended 15 year term for the extended 13 year term which would have become apparent if the issues raised by FENss decision to extend the subloan amortization period had been appropriately analyzed. 25. As described above, there have been significant adverse consequences as a result of the failure to recognize, and adopt measures to minimize, the mismatch between FENss repayment terms on the "A" and "Bn loans and the repayment terms to FEN on the subloans made from those sources. Of much greater significance, however, is the mismatch of the short repayment terms of the loan capital made available to the power companies from both foreign and domestic sources in comparison to the long construction periods and useful earning lives of the assets financed by these loans. This issue and steps FEN might take to provide suitable term transformation to the power companies to resolve thie problem are discussed in paras (5.77-5.79) and Annex XI. Such measures would also eliminate the mismatching problems discussed in this Annex for the period subsequent to their adoption.  ANNEX I Page 10 of 12 TABU A. REPAYMENTSBY PENON PRoJEcX'LOANS ('A' AND 'B' LOANS) AS AGREED COMPARED TO REPAYKENISTO FEN ON SUBLOANSAS ORIGINAUY SCHEDULED AND AS MODIFIEDFOR SIXMONTH EXTENSIONOF SUBLOANREPAYMENT (us3 Miuioor Scc note) - B. AJ Modified for 6 M o d Externion of Subloan AU US$ amountsam bucd on rtrtcdUSS equivdodvalueru o f b timothe louu and arblornr were -do. n e y do mot mflcct the impactof aubaequenichangesin oxclungcrater applicable totbo Wodd Bank 'A' loan of USSl7Omillion a d tho Japanow Yon deaominatod 'B' b.Such oxctuago nto clunga, of cauno, would not bo applicable to tht UsS175 million demminacod'B' lorn. Specifically, for tho World Bank 'Ag lorn, tho amountslhown do pql rrfloct chrngeain USS equiv.lenl valuer due to tho Wodd B.nlr*acwr~ocypooling ayatcm affecwFEN'# repaymentsto tho World Bank and tho rcpaymcntaby Lo povcr canprniea of the poaion of r u b h applicrbh ta tho relendii of tbora Iluda. P a tho Japuica Yon dcaomimted cofinrachg.tho repaplentr by PENare b a d onthe oxchangonto of Y220per US$ (or a total equivalentvalueof USSZSmillion) applicablearhontholoanw u aged; md therepaymentsby thopowwcomp.nieauo bued ontho US$ cquivrlontvduc therwf when therubloamworodnwnd m . AJ a r e d of thia laltor ictor, tho mdUS$ cquivllontv.luoof tho ~ b l o u ir r h mar USS370.52 u million,&thanUSS369.5 millionwhichothvrvire would h v eboon applicrbkdowing Tor the totalof USSOJ million u ~ fw d tho ~ - 0fee payablcto tho World Bank and for consulting u m c a to d FEN.  ANNEX I Page 11 of 12 TABLEB. REPA- BY FEN ON P R V LOANS (.A9 AND 'B' LOANS) ASSUMINOMODIFIEDSCHEDULEFOR 'Ag LOAN (15 M W OF 17YEARTERM, INCLUDINaFIVEINSI7WDOFFOUR YEAR OBACB PERIOD) COMPARED TO REPAYWWIS TO FEN ON SUBLOANS AS ORIQINALLYSCHEDULEDAND ASSUMINO MODIFIEDSCHRDUL.8(15 IN- OF 13 YEAR TERM, INCLUDINO FOUR YEAR ORACE PERIOD) (USMilliom Seenote) - wn3: Ibe lloto for Table A rlm rpplierto thu Uble.  ANNEX I Page 12 of 12 TABLEC. REPAYMENTSBY FEN ON PRO= MANS ('A' AND 'B' LOANS) AS AGREED COMPAREDTO WAYMENIS TO FEN ON SUBLOANS ASSUMING MODIFLEDSCHEDULE(15 INSTBADOF 13 YEAR TERM, INCLUDINGFOURYEAIL GRACEPERIOD) AND MODIFIED SCHEDULE-ED SIX MONTHS (Uss WLiom See note) - m: Tbe W forTable A dm applierto thir table.  Page 1 of 18 COLOHBIA POWER DEVELOPMENT FINANCE PROJECT (LOAN 2401-C01 Overview 1. Colombian power companies have a genuine need for extended debt repayment terme eo that their debt eervice requiremente would be realietically related to their earning power (parae 5.62-5.64). To accomplieh thie, it ie prima facie reaeonable to believe that eignificantly longer and more eyetematic term tranefonnation than now available ie feaeible and could be provided by FEN without undue riek, both with respect to domeetic funds raieed through Colombia's capital markets and foreign borrowinge. Thie would also be coneietent with FEN'e expanded role, under Lawe 25 and 51 of 1990, to refinance debt eervice obligation8 in order to rationalize the functioning of the energy sector (parae 5.56-5.61). The conclueions stated herein, however, ehould be regarded ae tentative and eubject to confirmation or modification after a thorough etudy, more complete and detailed than the limited analyeie made for the purpoee of thie report. Moreover, theee conclusione ehould not be regarded ae a panacea but ae a complement to the many other eeeential  ANNEX I1 Page 2 of 18 remedies needed to correct deficiencies in the management, operations and pricing policies of Colombia's power utilities and power sector. Indeed, unless these deficiencies are corrected, the proposals for extended term transformation discussed herein would not be feasible. ColS Denominated Loans 2. Domestic Capital Markets. The domestic markets from which FEN raises Col$ resources have strengthened and matured substantially since 1982 when the initial studies for this project were made. One illustration of this is that, since 1989, there has been a free market based measure of average short term inter~?trates (DTF), paid by banks and other financial institutions on certificates of deposit, which FEN has used as a basis for setting variable interest rates on its four year borrowings (para 5.21). The amount of funds raised from the domestic markets by all issuers participating in them has also grown significantly. Although the orientation of these markets remains primarily short-term, apparently much of the savings attracted by the ample real interest rates payable is rolled over on maturity. Over the seven year period from the end of 1983 to the end of 1990, FEN'S outstanding borrowings from the domestic markets have increased at an average annual rate of about 40% in nominal terms and 14% in real terms. Thus, it appears that FEN can fairly confidently expect that the short and medium term funds raised by it in the domestic capital markets will remain available to it indefinitely as long as it pays competitive interest rates for them, and that FEN can reasonably consider these funds as a solid foundation for prwiding long term loans to the power companies. 3. Interest Rate Risk. FEN would be protected from interest rate risk by specifying that the interest rate charged by it on long term Col$ denominated loans would be variable and would be tied to its average interest cost for funds borrowed from the domestic capital markets plus a specified margin (commonly called a "spread") deemed adequate to cover its operating expenses, taxes, provisions for losses, and a reasonable profit. Unfortunately, this practice was not adopted by FEN and, as pointed out in para 5.46, its profits have not always been adequate. In the future, a reasonable profit should be understood to be an amount which, together with appropriate profits from other activities, would result in total annual net earnings after taxes at least sufficient, when capitalized as part of retained earnings, to preserve the real value of FEN'S stockholders' equity. The adequacy of spreads and FEN'S profit objective are subjects discussed in more detail in paras 31-37. 4. Partial Caoitalization of Interest. Under prevailing inflationary conditions in Colombia, a preponderant part of high nominal interest rates represents compensation to maintain the real value of the loan principal. When interest is paid on this basis, most of the payment represents real repayment of principal; this imposes a heavy burden on the borrower and reduces the average real maturity of the credit. The proposed new term transformation arrangements should relieve borrowers from this burden by capitalizing an appropriate part of the nominal interest payment. Of three methods for partially capitalizing interest analyzed during appraisal, the one identified as preferable specifies the rate of interest in advance (e.g. 8%), with the rest to be capitalized as part of the loan amount and amortized as  ANNEX 11 Page 3 of 18 part of the scheduled loan repayments. Other methods which might be used would predetermine on a percentage basis either the amount to be capitalized and amortized, with the rest paid as interest, or vice versa. The first method was deemed preferable because by maintaining the amount of interest constant it facilitated financial planning and reduced administrative costs (SAR No. 4771-C0, Annex 2.1). 5. Extended Repayment Terms. Specific determination of how far to extend the repayment terms of FEN'S loans to power companies needs further study. The following discussion presents criteria and general considerations pertinent in this respect. 6. Normally, grace periods during which no loan amortization is required are commensurate with the period needed to construct and place into operation the assets financed by the loan. FEN should adopt this standard for setting the grace period on its domestic loans for investment projects. To illustrate what this would mean in contrast to its present practice of a limited one year grace period, the following norms are cited. For distribution and transmission investments, a one year grace period would generally be adequate, although for a multi-year program of such investments and for some high voltage transmission projects, a longer period may be justified. Longer construction periods are needed for generation projects. Experience indicates that these can be estimated more reliably for thermal generation than for hydro electric projects. For the former, a grace period of about three years usually would be appropriate; for the latter, a grace period of at least five years generally would be required and longer periods may be needed depending on the size and complexity of the particular project. 7. FEN should have, or as necessary retain, the engineering expertise to review and be satisfied with borrower estimates of construction periods. Borrowers should understand that they will be held accountable for their mistakes in underestimating construction periods. FEN should adopt a restrictive policy on extending grace periods and providing additional finance when projects are delayed or have cost overruns - e.g. limiting relief to situations (i) which were beyond the control of the borrower and not reasonably foreseeable, and (ii) where the borrower was efficiently managed and operated and could not meet the costs from internally generated funds. If these conditions are not met and financial assistance is necessary, other conditions should be imposed, such as changes in management and adoption and implementation of measures to correct deficiencies and improve performance. 8. In addition to the length of the grace period, f durinnaspartoftheloanamountisanotherimportant financial policy issue. As an element of prudent policy, lenders usually require cash payment of interest during the grace period except in cases where it would be unreasonable to expect the borrower to be able to generate the required funds for this payment from operations. Typical exceptions would be a construction loan to a newly established company which has not yet begun operations, or a loan to an operating company to finance a very large addition to capacity which has a long construction period. If FEN has not already adopted such a policy, it would be advisable to do so as part of the expansion of its term transformation role. It should be noted that the policy followed  Page 4 of 18 in determining whether to capitalize interest during the grace period as part of the loan amount is not dependent upon the accounting treatment of interest during construction, and vice versa. In accordance with generally accepted accounting practice in Colombia as well as elsewhere, interest costs on funds borrowed to finance plant additions, incurred during the construction or implementation period, are capitalized as part of fixed assets. Interest costs applicable to works for which the construction or implementation period is one year or less, however, may be charged to operations. This accounting practice is applicable irrespective of the source of funds to pay the interest costs. 9. Standard repayment periods, exclusive of grace periods, should be established by FEN. These should vary depending on the type of investments financed and allow longer repayment periods for loans financing assets with longer useful lives. The standards should be based on studies of funds generation which Colombian power companies would have available for debt service assuming that they were efficiently managed and had prudent capital structures (e.g. debtlequity ratios no higher than 60140 after allowing for revaluation of assets). The studies should assume that FEN would renegotiate domestic currency loans now outstanding so that the proposed standard repayment periods would henceforth apply to them as well as future loans. They should also allow for extensions in the repayment periods of foreign debts which would result from refinancing arrangements discussed in paras 15- 30. The objective should be to have prudent annual debt service coverages, say at least 1.5. 10. Debt repayment periods, exclusive of grace periods, normally would be materially less than the useful life of the asset financed, typically not more than 75-80%, and not less than 50-60%, of the life. To illustrate, repayment periods might vary from 12 to 15 years for loans financing distribution and sub-transmission assets with lives of about 20 years; 15 to 20 years for thermal generation and high voltage transmission assets with lives of 25 to 30 years, and 20 to 30 years for hydroelectric assets with lives of 40 or more years. The feasibility of shorter repayment periods at the low end of the range would generally depend upon the borrower having a relatively high equity component in total capitalization. 11. Although maturity ~remiaare customarily part of the interest rate structure, the reasons for them are greatly diminished in cases where variable lending rates are linked to short term borrowing rates expected to be adequate to offset inflation risks. Nevertheless, there is a risk that unexpected adverse changes in conditions may occur, and this risk increases the longer the repayment term. Thus, FEN may wish to consider including modest maturity premia in its interest rate structure by charging somewhat higher spreads on loans with longer maturities. This might also be associated with offering an incentive to more credit worthy borrowers to accept shorter repayment periods. Under such an arrangement, there might be a standard range of spreads which are successively higher as the term of 'the loan increases, with the lower end of the range applicable to borrowers who choose and qualify for shorter repayment periods - e.g. 12 instead of 15 years for a transmission and distribution loan.  ANNEX I1 Page 5 of 18 12. Eligibility and other Policy Considerations. For a power company to be eligible for FEN'S financial support, whether for Col$ or foreign currency denominated loans, FEN should be satisfied that it is efficiently operated and managed and is credit worthy, or that it has adopted and is satisfactorily implementing a program which will result in its becoming so within a few years. One test of credit worthiness would be the prudence of the capital structure and the adequacy of the debt service coverage ratio (see para 9). 13. In order to limit FEN'S term transformation risk, consideration should be given to adoption of a policy that requires the power companies, to the extent feasible, to give priority to use of net internally generated funds, after payment of debt service, for financing components of their investment programs with the longest useful lives and to limit their borrowings from FEN for their investment program to those elements with the shortest useful lives. Such a policy, however, will only reduce the risk, since it is likely that some FEN financing will be required for those elements of a power copmpany's investment program with the longest useful lives. This qualification is particularly relevant to a program that includes hydroelectric generation. 14. There are other policy implications of an expanded term transformation role for FEN relevant to whether its domestic borrowings would "crowd outw other borrowers and to maintenance of an adequate liquidity buffer and debtlequity ratio. These are discussed, beginning at para 38, after considering FEN'S term transformation role for foreign borrowings. Foreian Currency Denominated Loans 15. The repayment terms of foreign borrowings by the power companies are mismatched in relation to their earning power. This applies both to direct foreign borrowings by the power companies and to foreign currency denominated loans extended through FEN. As a result, many power companies have had insufficient funds available to meet all their obligations, including debt service, and it has been necessary to resort to various ad hoc solutions, including the CADEX program instituted by FEN in 1990, to provide short or medium term financing to them so that repaymnents due on foreign loans could be made. These solutions have provided only partial and temporary relief. A more systematic and transparent way to establish appropriate debt repayment obligations reasonably related to the earning power of the power companies is needed so that they would be subject to effective and realistically achievable financial discipline. (See paras 5.62 -5.64.) To accomplish this, a potential solution which should be explored would be for FEN to establish and manage a pool of foreign resources (hereafter referred to as the "foreign facility") which would be used and made available for this purpose to all interested and qualified power companies, as described below. 16. Fundinn Sources. Establishment of the foreign facility would depend on obtaining the support of foreign lenders. The most likely source would be foreign connnercial banks. The type of financing available for the facility probably would be limited to lines of credit which could be drawn upon for stated periods of time and which could be renewed, and medium term credits which would only be partially amortized with the balance due at maturity subject to refinancing on similar terms if the parties so agreed. The  ANNEX I1 Page 6 of 18 facility should be presented to potential lenders as an important element of an effort to rationalize and improve the performance of Colombian power companies, which are to be held accountable for meeting debt service requirements reasonably related to their earning power. Thus, the presentation should emphasize that (i) the facility's objective is to refinance debt service due on outstanding and future foreign loans so that grace and repayment periods are reasonably related, respectively, to the construction or implementation period, and useful life of the assets financed; (ii) its availability is limited to qualified companies (para 12); and (iii) the interest rate and foreign exchange risks assumed by FEN would be passed on to the power companies benefiting from the facility. On this basis, lending probably would be forthcoming for the facility because it would be viewed as financially sound. Government support for this approach to foreign lenders, of course, would also be necessary; and the foreign lenders no doubt will expect that their loans to FEN for this facility would be guaranteed by the Government. In time, when the power companies have demonstrated that they have improved their efficiency and that their financial performance is sound, it may be possible to obtain funding for the foreign facility from foreign lenders without a government guarantee. 17. Whether short and medium term borrowings from foreign commercial banks will prove to be a solid foundation for FEN to provide long term loans to the power companies will depend on the continuous availability of such borrowings over many years. In turn this will depend on how well the power companies perform in meeting their debt service obligations on the extended terms made available by FEN and whether Colombia continues to be a credit worthy country. It will also depend on how well the world economy performs, since the availability of credit from foreign commercial banks would be adversely affected by poor performance of their home country economies. 18. There are clearly risks to FEN in proceeding with term transformation on the basis suggested. It is reasonable to conclude, however, that taking these risks is worthwhile, considering the advantages to be gained from establishing the foreign facility. To a limited extent, it should be possible to mitigate the risks by maintaining adequate liquidity and equity buffers (paras 42-46). 19. It may also be possible for FEN to raise funds domestically to finance the foreign facility through US$ (or other foreign currency) denominated borrowings. If this were authorized by the Government and the central bank, such borrowings might well prove to be another significant funding source for the facility. Most likely, such borrowings would have to be short term initially, with some scope for extending maturities later, and the interest rate would have to be related to dependable indices such as LIBOR or the US prime rate. As in the case of Col$ borrowings, once such a market were established in Colombia, the prospect of maintaining and gradually increasing outstanding balances from year to year should be good. At present, the foreign exchange market in Colombia has been liberalized and there is a very substantial supply of US$ available (para 5.61). If it were deemed advisable to do so, US$ (or other foreign currency) denominated borrowings could be limited to lenders who have foreign exchange to pay for the notes issued by FEN and to whom payments of interest and principal would be in foreign  A-NNE Page 7 of 18 exchange. Alternatively, they might simply be foreign currency denominated loans paid for in ColS. 20. Implementation Arrannements. FEN would agree with interested and qualified power companies to (i) assume their obligations under foreign loans in exchange for their agreement to repay these obligations to FEN on an extended basis, and (ii) to extend the repayment terms of FEN'S foreign currency denominated loans to them (e.g. the subloans under this project or those using other external credit resources such as the Concorde and Challenger Loans). Extended grace and repayment periods would be established in accordance with standards and policies adopted after further study, as suggested in paras 5-10. The interest and amortization payment dates, with respect to month and day (e.g. June 15 and December 15) and frequency (e.g. semiannually), would remain the same under the new loan owed by a power company to FEN as under the original foreign loan assumed or extended by FEN. 21. The agreements would specify that the new loan would have two distinct components for determining its principal amount and the interest payments due thereon: an "original" and a "foreign facility11component. This distinction and the other arrangements described herein are necessary so that the foreign exchange and interest rate risks that FEN assumes or incurs in order to provide extended term financing to the power companies will be appropriately matched with those it charges to the power companies for this financing. 22. The basis for denominatinn principal amounts in terms of foreign currencies payable that was prescribed for the original loan (i.e. the assumed loan or the previously agreed foreign currency denominated subloan that FEN is extending) would continue to apply with respect to the original component of the new loan; the comparable basis that would be applicable to the foreign facility component would be stipulated in relation to the resources used to finance that facility (see paras 26-28). Similarly, the interest rate arrangements agreed for the original loan, plus a spread which FEN would charge on assumed loans (see para 30). would be the basis for determining the amount of interest payable on the original component of the principal; and those specified for the foreign facility (see para 29) would be the basis for determining interest payable on the foreign facility component. Repavments of principal under the new loan would depend on the grace and repayment periods agreed in each case. Subsequent to the grace period, when there would be no repayments, level repayments would be due on each repayment date. They would be level in the sense that they would represent an equal proportionate amount of the total outstanding principal of the loan (i.e. the sum of the original and foreign facility components) over the remainder of the repayment period. For example, if the new loan were to be repaid in 40 semi-annual installments, the initial repayment would be determined by dividing the outstanding total principal of the loan as valued on the repayment date by 40; the second repayment by dividing by 39; and so on. Repayments would be applied proportionately to the original and foreign facility components of the new loan's principal amount. 23. The foreinn facility component of the new loan would represent two elements. First, to the extent that repayments under the new loan were less than those that were required under the original loan, that difference would  ANNEX IT Page 8 of 18 become part of the principal owed under the new loan represented by the foreign facility component. Second, any interest payments that would be capitalized as part of the loan amount only because of the provisions of the new loan would become part of the foreign facility component. 24. An illustration of the ~roposedarrangements may be useful. Initially, the original component would constitute 100% of the principal amount of the new loan. Thus, upon effectiveness of FEN9s assumption of a foreign loan owed by a power company, FEN'S loan portfolio would include a new loan with a principal amount identical to that of the assumed loan now part of its financial obligations. Similarly, upon effectiveness of an agreement for PEN to extend the repayment terms of a foreign currency denominated loan, a new loan would become part of its loan portfolio and it would have a principal amount identical to that of loan it replaces. In each case, the entire amount of the new loan, upon effectiveness, would be recorded as the original component. 25. As the new loans are repaid by the power companies at a slower pace.than the original loans FEN has assumed or extended, FEN will need other resources to meet the principal repayments on the assumed loans and on the borrowings it made to finance the subloans which had been extended. This is also true of any interest payments which under the original loans would have been paid in cash but are to be capitalized under the new arrangements. As FEN utilizes other resources for these purposes, commensurate changes would be made in the records it maintains for its loan portfolio to record the original and foreign facility components of the new loans. This would be accomplished as the occasion requires either by: transfers from the original to the foreign facility component recorded - in the same amount as, and simultaneously with recording, the repayments made by FEN on an assumed loan, and in the amount that would have been required to be repaid to FEN on the due dates as prescribed under the original terms of a foreign currency denominated subloan by FEN to a power company, had the repayment terms not been extended; or recording interest payments capitalized as part of the loan amounts under the original component if the original loan arrangements provided for such capitalization, and under the foreign facility component if the capitalization occurs pursuant to the new extended loan arrangements. As previously indicated, as loan repayments required under the new loan arrangements are made, the amount thereof will be allocated proportionately to reduce the original and foreign facility components of the pertinent loan. 26. Foreign Facility Currency Pool. It has been assumed in this discussion that the foreign facility would be funded by foreign borrowings. In order to distribute the impact of foreign exchange and interest rate changes on these borrowings fairly among all the beneficiaries of the foreign facility, it is recommended that these borrowings be maintained in a separate foreign facility  ANNEX Ix Page 9 of 18 currency pool, the performance of which would serve as the basis for achieving this objective. Broadly stated, there are two choices for managing the foreign borrowings. The simplest would be for all the borrowings to be denominated in a single currency, say the US$, in which case the only relevant exchange rate would be that of the single currency chosen. The second solution would be for FEN to establish a foreign facility currency pool involving a number of foreign currencies and to allocate the foreign exchange risk in the same way as under the World Bank's currency pool so that the impact of exchange rate changes on all loan balances funded by the pool is identical. In either case, the average interest cost of the foreign borrowings made to fund the foreign facility would also serve as the basis to which a spread would be added to determine the variable lending rate charged on the foreign facility loans. 27. Under the second solution, it would be necessary to decide what limits if any should be followed with respect to the number and composition of foreign currencies to be involved. The World Bank's experience with its currency pool led it to adopt a targeted currency composition because this would simplify management of foreign exchange risks by its borrowers including facilitating hedging (see para 8.14). It is questionable whether FEN would be able to manage a pool of foreign borrowings so that it would have a dependable targeted currency composition. To accomplish this, it would be necessary for FEN to have available adequate sources of several different foreign currencies which could be drawn down and maintained in fixed proportions. Even if this proved to be possible, there would also be a question whether it would be advantageous, or deemed to be so by the power companies, to have a mix of foreign currency financing sources, with different foreign exchange and interest rate risks, rather than rely on a single foreign currency financing source such as borrowings denominated in US$. 28. There are several reasons for deciding to rely solely on US$ denominated financing for the foreign facility currency pool, without prejudice to the right to change that decision in the future if a changed course becomes advisable. Because ex ante determinations of foreign exchange and interest rate risks are very speculative, it is unlikely that there would be a sound basis for concluding that an alternative course would be advantageous. There is a very broad, large and highly competitive international market in which to obtain US$ denominated financing at variable interest rates tied to reliable indices such as LIBOR or the prime rate of major US banks. Use of other currency denominated financing would not be precluded if such financing could be swapped for a US$ denominated loan at a lower interest cost than payable for directly obtained US$ financing. Finally, this is the administratively simplest basis to use. 29. Onlendinn Rates. As indicated in para 26, variable lending rates for the foreign facility component of extended foreign currency denominated loans would be based on the average interest cost of the borrowings in the foreign facility currency pool plus a stipulat'edspread. The spread should be adequate to cover costs and contribute appropriately to meet FEN'S reasonable profit objective, as suggested in para 3 for Col$ denominated loans. Instead of a constant spread for all foreign currency denominated lending, spreads may  ANNEX I1 Page 10 of 18 be structured t o allow for modest maturity premia, as suggested i n para 11 for Col$ loans. 30. In addition, the interest payable on the original component of assumed foreign currency denominated loans would be determined by adding an appropriate spread t o the interest rate payable as agreed for the original loan. Although FEN did not serve as an intermediary i n the arrangements for the original loan, its agreement t o assume responsibility t o repay that loan under the original terms is a commitment of part of its borrowing authority for which it is entitled t o some compensation. The duration of the remaining l i a b i l i t y under the original loan which FEN assumes w i l l be a pertinent factor i n determining an appropriate spread, particularly i f FEN structures spreads t o allow for maturity premia. Thus, i n those instances where the remaining duration is comparatively short, FEN may conclude that a relatively small spread compared t o its average spread on foreign currency denominated loans would be adequate compensation for assuming liability for such loans as part of its extended term transformation efforts. No additional spread would be needed for the original component of foreign currency denominated loans extended by FEN. The spread provided under the arrangements agreed when such loans were originally made should be adequate t o compensate FEN for its role as intermediary. General Considerations 31. Adeauacv of Spreads. A review of the adequacy of FEN'S present practices for establishing spreads and onlending rates is one of the matters dealt with herein requiring further study . The purpose of t h i s review would be t o determine what changes may be needed and would be feasible t o improve FEN's overall level of profitability and the distribution thereof among its various sources for generating profits. Paras 3 and 11 (and para 29 by reference thereto) contain brief statements of suggested principles applicable i n these respects. Additional comments pertinent t o the recommended review follow under three sub-headings. 32. a. Profit Objective. Para 3 suggests that a reasonable profit objective for FEN is an amount of annual net earnings after taxes a t least sufficient, when capitalized as part of retained earnings, t o preserve the real value of FEN'S stockholders' equity. This is equivalent t o stating that FEN'S nominal annual rate of return on its average equity for each year should be a t least equal t o the average rate of inflation for the year and that a l l of its earnings should be retained rather than distributed. In actuality, except for a very small amount of dividends paid in cash, it is FEN'S policy t o retain a l l of its net earnings after taxes. A s pointed out i n the main text (paras 5.46 and 5.66 and Table 5.1) ,however, FEN's after tax rate of return on average equity has been less than the rate of inflation i n each of the years 1985-1990. As a result, the value of FEN's stockholders' equity has depreciated i n real terms by about 11% over the period from the end of 1983 t o the end of 1990. 33. The proposed profit objective is quite modest, identical t o stating that the real rate of return on equity should not be less than zero. This may be justified on the grounds that FEN functions basically as a service company  ANNEX 11 Page 11 of 18 that provides essential services and assistance t o a f f i l i a t e d Colombian power companies by arranging financing f o r them, and t o the Government by monitoring the financial performance of the power companies and strengthening power sector financial management. A s such, it may be concluded t h a t FEN'S financial performance would be satisfactory i f it provided the services a t no p r o f i t and no l o s s i n r e a l terns, which is what would be required i n the USA with respect t o dealings between a service company and a f f i l i a t e d u t i l i t i e s . An alternative view would be t h a t FEN is not simply operating as a service company but a l s o a s a financiera taking r i s k s i n extending c r e d i t t o the power companies, and t h a t it should be compensated f o r t h i s r i s k by earning r a t e s of return on equity approaching but not i n excess of those earned by privately owned financieras i n Colombia. Under the l a t t e r view, the zero r e a l r a t e of return on equity would be a minimum p r o f i t objective, and a maximum r a t e of return on equity t h a t was deemed reasonable would have t o be established a s the maximum p r o f i t objective. The increase i n c a p i t a l resulting from earning and retaining a positive r e a l return on equity would be advantageous because it would support an expanded lending program to the power companies in real terms, without c a l l i n g on the Government t o fund increases i n FEN's equity c a p i t a l base t o t h a t extent. Clarification of FEN's p r o f i t objectives w i l l be necessary as p a r t of the review of the adequacy of spreads. 3 4 . b. Distribution of P r o f i t s among Revenue Sources. FEN's two principal revenue sources a r e peso and foreign currency loans. I f it were feasible, it would be desirable f o r FEN t o earn approximately the same r a t e of p r o f i t on a l l of its loans so t h a t each of its borrowers would contribute t o FEN'S p r o f i t s proportionally t o its use of FEN'S lending f a c i l i t i e s . As a p r a c t i c a l matter, t h i s i d e a l distribution of p r o f i t s cannot be realized. For example, i f maturity premia a r e introduced i n t o the lending r a t e structure a s suggested i n paras 11 and 29, the i d e a l would have t o be modified t o achieve uniform p r o f i t margins on loans with comparable maturities. More fundamentally, it must be recognized t h a t both the amount of n e t p r o f i t produced by any given amount of spread (e.8. 0.5%) and its percentage relationship t o equity w i l l vary over time a s costs other than borrowing costs, the volume of lending operations over which these costs may be distributed, the amount of income earned on investments, the extent t o which equity c a p i t a l is u t i l i z e d f o r lending purposes, and t h e amount of the equity base vary. Accordingly, appropriate changes i n i n t e r e s t r a t e margins may be necessary from time t o time t o achieve desired p r o f i t objectives. This may be the reason f o r the range of margins, from 0.5% t o 0.9% per annum over its i n t e r e s t cost f o r borrowing foreign funds, t h a t FEN has used t o s e t its lending r a t e s f o r foreign currency loans (PEN Annual Report f o r 1989, English version, page 2 0 ) . Moreover, t o the extent t h a t FEN'S p r o f i t objective includes a positive r e a l return on equity, the judgment a s t o t h e specific amount of return, o r range of returns, deemed reasonable may vary from time t o time a s circumstances change. Thus, the objective of a uniform r a t e of p r o f i t on a l l loans can only be applied prospectively and imperfectly when spreads over average borrowing costs a r e periodically established as the basis f o r determining lending r a t e s t o be charged f o r future loans. I f an identical spread, o r a uniform scale of spreads related t o loan maturity, is applicable f o r a l l loans, whether peso o r foreign, t h a t is the most t h a t may be accomplished t o achieve the desired p r o f i t objective while adhering t o the ideal distribution of p r o f i t s as closely a s possible. Similarly, other loan charges such as commitment fees  ANNEX I1 Page 12 of 18 should be applicable uniformly for all loans. There should be an appropriate justification for any departure from this approach; e.g. charging higher spreads and thus earning higher profits for riskier loans. 35. The initial step in carrying out the review of the adequacy of FEN'S interest rate margins would be to determine the annual amount of profits realized in past years from each of the peso and foreign currency loan categories and compare this to the average annual amount of loans outstanding in each category. This would identify the extent of differences in average profit rates for the two categories. A separate component of this analysis would be a comparison of the average gross spreads earned for each category measur?d by the differences between direct operating income from, and expensee for, interest and commissions, as a percentage of the average amount of loans outstanding. If the resulting differences in profit rates or spreads are significant, the analysis should be extended to identify the principal factore accounting for the differences. A number of technical points affecting this analysis are discussed below. (i) The average annual amount of loans outstanding in each category should preferably be calculated on the basis of month or quarter end balances to minimize distortions due to uneven changes in outstanding amounts during the course of the year. (ii) Income on Investments. FEN also earns income on peso and foreign currency investments, which essentially represent holdings acquired om part of liquidity buffers maintained for each of the lending activitiee. To the extent that it is possible to determine that borrowings were the source of funds utilized to acquire these investments, the pertinent interest and commission costs of such borrowings should be subtracted from the income (interest, commissions, amortized discounts and monetary correction) derived from the investments. The resulting net income or loss should be attributed to the pertinent lending category and treated either as a source of income in addition to that derived from the spreads, or an additional item of expense, pertinent to the lending category. (iii) Peso loans mav be financed from equity capital as well as from borrowings. This should be assumed to be the case whenever the amount of peso loans (factor A) is greater than the amount of peso borrowings considered to be the source of funding such loans (factor B). In that event, the total interest and commission income from peso loans should be allocated to the two sources. The portion of this total obtained by applying the ratio of factor B over factor A should be considered the direct income obtained from peso loans financed from peso borrowinge. This should be compared with the interest and commission expenses attributable to the amount of borrowings utilized to finance peso loans to calculate the average gross spreads on peso loans. The balance of the total direct income from peso loans should be treated as a source of income in addition to that derived from spreads. (iv) Foreinn Loans under Power Development Finance Prolect. Foreign loans have been financed exclusively from foreign borrowings, and,  ANNEX 11 Page 13 of 18 except for interest spreads, there has been a precise sparmetry (or nearly so) between the terms of the foreign borrowings and the loans financed by them. The one and only instance in which there has not been such a symmetry occurred under this Power Development Finance Project. In this case, as pointed out in the main text, there was a mismatching between the repayment terms of the foreign borrowings and the subloans financed by them, which was accentuated by FEN's decision in 1988 to extend the subloan amortization period by six months (para 5.47). For the years beginning with 1988, the gross spreads on these subloans should be calculated separately from that for other foreign loans. This separate calculation should include, as an element of direct income in add.'tionto interest, the exchange rate adjustments applicable to these subloans, including the amounts thereof credited both to direct operating income and to stockholders' equity (see Annex I, paras. 8 and 9). It should also include, as an element of direct expense in addition to the interest costs on the foreign borrowings, the interest and commission costs attributable to the peso borrowings incurred to meet repayment obligations on the foreign loans in excess of the amounts repaid under the subloans for as long as necessary until this deficiency is made up (see Annex I, paras. 10-13 and Table A). (v) Allocation of Expenses other than Borrowinn Costs. To calculate the amount of net profits attributable to peso and foreign loans, it will be necessary to allocate the net expenses resulting from all other items of income and expense, both operating and non-operating, other than borrowing and investments. Except for income taxes, it is suggested that this allocation be made in proportion to the average amounts of outstanding loans in each category. Income taxes could be allocated in proportion to the earnings before income taxes of each category, excluding that portion of credits for exchange rate adjustments made directly to stockholders' equity. (vi) s FEN'S expenses in 1989 include a special provision of Co1$752.3 million for a deficit that had accumulated to the end of that year in a fiduciary account which FEN had been managing since November 1984. Of this amount, losses had been incurred in the amount of Co1$286.1 million for 1989, Co1$348.7 for 1988 and Col$117.5 million in prior years. In 1990, FEN'S expenses include a provision of Col$119.5 million for that year's loss in this account. The resources administered through this account were provided by the Social Security Institute (ISS) in exchange for obligations (identical to those for Constant Value Social Security Bonds) to repay them at a value related to inflation plus interest of 5.5% per year on the adjusted value. The contract under which FEN administered these resources required it to account for any lossea resulting from financial and credit decisions which produced earnings less than the cost of the resources. The provision charging the accumulated deficit in this account to FEN'e earnings for 1989 was made at the direction of the Banking Superintendency. FEN'S responsibility to make up the losses sustained in this account was confirmed in Law 48 of December 25, 1990. That law also provided for the termination as of 1991 of the fiduciary administration contract and the simultaneous  AWNEX IT Page 14 of 18 transfer to FEN of ownership of the assets previously administered thereunder, so that these assets and the corresponding liabilities will henceforth be reported in FEN'S financial statements. In calculating annual profits from peso lending, the expense charged in each of the years 1988-1990 for the loss sustained on this account should be limited to the amount attributable to that year's results. For years prior to 1988 if the information is readily available, the yearly loss (or gain if any) in this account should be included in the calculation, or the cumulative loss from inception to the end of 1987 charged to 1987. Beginning with 1991, the ISS funds should simply be considered as an additional source of domestic borrowings available to FEN for its peso linding activities; the uses made of these funds will no longer be separately identifiable and their costs ehould be treated as part of the costs of the pool of domestic borrowings. The lessons learned from this experience should include FEN'S belated recognition, noted in its Annual Report for 1990 (English version, pages 16-17), that inadequate spreads were charged for relending the ISS resources and that better returns could be achieved from investing the surplus funds. Additional information about this special fiduciary account is set forth in the addendum to this Annex. 36. After completing the analysis of historical results, a comparable analysis should be made for 1992 and at least one but preferably several subsequent years based on forecasts of FEN'S financial performance. The object should be to determine a schedule of spreads that is judged to be feasible to implement and reasonably likely to produce the desired rate of return on equity. In preparing the forecasts, it will be necessary, of course, to take into account the spread and lending rate arrangements already contractually agreed for loans made by FEN and those to be committed pursuant to arrangements specified in FEN'S agreements with its lenders. As suggested in para 9, however, if the recommendations made herein for extended term transformation are adopted, FEN will be able to apply new spread and lending rate arrangements not only to future peso loans but also to presently outstanding peso loans which would be renegotiated. Similarly, there will be a substantial potential for FEN to adopt and implement new spread and lending rate arrangements for foreign loans which it would assume, and for the extended term portion of outstanding foreign loans which it would renegotiate (see para 22). FEN'S ability to renegotiate outstanding foreign loans and to apply new terms to future foreign loans to be funded from outstanding credits may be subject to obtaining necessary approvals from its lenders. The financial forecasts should make appropriate allowance for these potentials. 37. F F . To ensure that FEN'S lending rates appropriately reflect its current borrowing costs, FEN ehould standardize its practices in this respect. It is suggested that its policy should be to apply stipulated spreads to average borrowing costs to determine new lending rates at least semi-annually, and perhaps quarterly. It is also suggested that interest should accrue based on the new lending rate from the time that the new rate is effective. 38. Crowdinn Out. The question whether FEN'S borrowings from the domestic capital market would "crowd out'' other borrowers has been raised and answered  ANNEX 11 Page 15 of 18 i n the negative during project appraisal (SAR No. 4771-C0, paras 3.26 and 3.27) and during project implementation when FEN'S potential f o r mobilizing domestic savings was evaluated (see reports of experts referred t o i n paras 5.15 and 5.18). A t the time of appraisal, access t o the domestic c a p i t a l market was regulated by the Government (SAR No. 4771-C0, paras 2.13 and 2.14). Since November 1990, however, FEN is no longer subject t o Government control i n t h i s respect, and it appears t h a t it i s f r e e t o compete f o r resources from the domestic c a p i t a l market t o r a i s e the amounts it deems necessary t o fund its activities (para 5.57). On t h i s basis, and on the assumption t h a t other sources of domestic finance a r e f r e e t o compete with FEN i n offering financial assistance t o the power companies, the question of crowding out may not be pertinsnt. Because the power sector is an important productive sector, it may be argued t h a t it should be able t o compete, directly o r through FEN o r other lenders, f o r resources from the domestic c a p i t a l market, s o long as it does s o without any special advantages o r preferences provided by t h e government. This reasoning would be strengthened i f the power sector were operated and managed efficiently, but despite the shortcomings i n t h i s respect, t h e argument may be valid. Nevertheless, i n view of the magnitude of FEN'S expanded participation i n domestic c a p i t a l markets t o obtain resources t o support the extension of power company debt repayment terms, as pointed out i n para 39, it is l i k e l y that both the Bank and Colombian authorities w i l l wish t o consider t h i s question again as part of the evaluation of FEN'S expanded term transformation role. 39. The one t h i r d increase i n FEN'S Col$ borrowings which occurred i n t h e l a s t two months of 1990, and the phenomenal 440% increase i n 1991, were necessary principally t o support its CADEX program t o provide assistance t o power companies t o meet foreign debt service requirements. I n addition, the ColS19.76 b i l l i o n increase i n equity provided t o FEN by the Government i n December 1990 was used f o r t h i s purpose. (Paras 5.59-5.61). I n US$ terms, using year end exchange rates, the increase i n Col$ borrowings i n t h e l a s t two months of 1990 was equivalent t o about USS28 million, the increase i n equity i n December 1990 t o about US$35 million, and the increase i n Col$ borrowings i n 1991 t o about US$438 million. No information i s available concerning the continuing requirements FEN may have under t h e CADEX program t o maintain and increase its borrowings from the domestic c a p i t a l market. This w i l l depend on how f a s t the power companies can r e a l i s t i c a l l y repay the Col$ loans made t o them under the CADEX program and what additional ColS loans they w i l l need t o meet future debt service payments on outstanding foreign loans. It appears that a l l the Col$ loans made under the CADEX program in 1990 were short term (para 5.60). It is unlikely, however, that the full amount of these loans and the much larger amount of Col$ loans made i n 1991 could be repaid within a year o r less, although t o some extent they may have been bridge loans expected t o be repaid within the short term. I n addition, it is l i k e l y t h a t the power companies w i l l need additional financing assistance t o meet future foreign debt service requirements. Thus, on balance, FEN may have t o r a i s e substantial additional amounts from the domestic c a p i t a l market i n 1992 and subsequent years t o support the CADEX program a s currently being implemented. 40. An alternative and preferable solution t o the problems power companies are experiencing i n meeting t h e i r foreign debt service requirements would be the foreign f a c i l i t y which FEN would establish and manage, as discussed herein  ANNEX 1% Page 16 of 18 (paras 15 - 30). Implementation of this proposal would relieve the domestic capital markets from the very substantial demands for funds which FEN would require for the CADEX program as now being executed. Although comparable resources would have to be obtained from foreign capital markets, those markets are many times greater than the Colombian market. Moreover, in financing the power sector, it would be better fiscal policy, and consistent with long standing Colombian practice and policy, to borrow foreign funds to meet foreign costs, including foreign debt service costs, when internally generated funds are insufficient to pay such costs. 41. If a decision is made to rely on Col$ borrowings to meet part of the funds needed to extend the repayment terms of foreign loans, this assistance should be provided on a more systematic basis than now being used, One possibility would be to modify the procedures discussed herein for the foreign facility to include a domestic facility which could be utilized as a substitute source of funding the extension of the repayment terms of foreign loans (see paras 20 -25). If this were to be done, partial capitalization of interest, as recommended in para 4, should also be applicable. 42. Liquidity Buffer. Because of FEN's reliance on term transformation and the risk of collection delays, it was agreed during appraisal and confirmed during negotiations that it would maintain at all times a liquidity buffer equivalent to not less than two months (116) of estimated annual expenses plus one month (1112) of the principal amount that would become due and payable to local investors during the year (SAR No. 4771-C0, para 2.31). The estimates of expenditures and principal repayments on which these requirements would be based are to be made in February of each year, and one half of the total requirements are to be held in assets which could be converted into cash within 90 days and the other half within 30 days (Loan Agreement, Section 4.06). As an additional protective measure needed because of FEN'S reliance on term transformation, it was also agreed as a condition of effectiveness that FEN would be provided with access to short term borrowings from the Central Bank to cushion it from an unexpected drop in savings mobilization, similar to that available to other financial intermediaries. This would provide coverage for short term liquidity problems resulting from an unexpected drop in savings mobilization. It was expected that FEN'S quota in this respect would be 100% of paid in capital and legal reserves for a maximum of 90 days annually. (SAR No. 4771-C0,para 2.12; Loan Agreement, Section 6.01.) During implementation, in 1985, FEN requested a reduction of the minimum liquidity requirement but subsequently withdrew the request (paras 5.24 -5.25). 43. An analysis of FEN's audited financial statements for 1990 indicates that FEN met the minimum liquidity requirements with a comfortable margin. Applying the definitions of Section 4.06 of the Loan Agreement in retrospect, rather than prospectively based on estimates as literally required, 116 of expenditures in 1990 was about ColS12.5 billion. Assuming that the total outstanding Col$ debt at year end would be repayable within a year, which is an overstatement, 1/12 of the principal due to local investors was about Co1$5.5 billion. This results in a total minimum liquidity requirement of Col$18 billion, or USS31.5 million at the year end exchange rate. The auditors report that at the year end, FEN had assets convertible into cash in  ANNEX 11 Page 17 of 18 less than 30 and 60 days, respectively, of Co1$22.2 billion and Col$19.9 billion, well in excess of the minimum requirement of Co1$9 billion in each category (Auditors' Report for 1990, section on compliance with accounting and financial clauses of Loan No. 2401 CO). Using rough estimates of the impact on interest and principal payment requirements resulting from the very large increase of domestic borrowings which occurred in 1991 ((para 5.61), total minimum liquidity requirements may have increased by Co1$33 billion to Co1$51 billion, or US$79 million at the year end exchange rate. 44. The present minimum liquidity requirements were established at a time when it was expected that (i) FEN'S receipts from repayments of foreign currency denominated loans would match its expenditures to repay its foreign currency borrowings and (ii) FEN's term transformation activities would be limited to Col$ borrowing and lending. With the expansion of FEN'S role to include term transformation of foreign currency denominated borrowing for the power sector, the minimum liquidity requirements should make an allowance for FEN's annual requirements to repay foreign as well as domestic loans. For example, the liquidity requirement could be reformulated to refer to FEN's total, rather than local, principal repayment requirements for the year as the base for computing the 1/12 component of the requirement. The determination of what would be a prudent liquidity buffer under FEN's expanded, and riskier, term transformation role, is an important aspect of the studies recommended hereunder. In this context, of course,'allowanceshould be made for the income earning potential of short term investments as an offset to the costs of maintaining the liquidity buffer. Taking this into account, it may be the case that the net costs of maintaining a large liquidity buffer may not be so significant. Also, of course, the availability of a liquidity quota from the Central Bank would also be considered. Assuming that this is still available in an amount equivalent to 100% of paid in capital plus legal reserves, at year end 1990, it would have amounted to a line of credit of Co1$54.5 billion or US$96 million, that could be utilized for a maximum of 90 days annually. 45. Debt Equity Ratio. FEN undertook to maintain its debt equity ratio within the limit of 7 to 1, under the agreements for this project. As explained in paras. 5.49-5.55, this limit was exceeded in the period 1988-1990 under circumstances such that the Bank granted temporary exceptions through Dec. 31, 1990. It was expected that the additions to FEN's equity, which the Government was authorized to make under Law 25 of 1990, would bring FEN back into compliance with the 7 to 1 limit, but the Bank has no information as to whether this in fact has occurred or is likely to occur in the near future. 46. Maintenance of a prudent debt equity ratio is another measure to protect FEN from the risks involved in its term transformation activities because of inability to obtain funds from expected refinancing of part of its debts or from scheduled repayments of loans by some of its borrowers. Considering that the 7 to 1 limit established at the beginning of its operations was a relatively high limit and that its risks are increasing as a result of its expanded term transformation role, any increase in the 7 to 1 limit at this time would be of dubious merit. Some clarifications of definitions and details of the calculation, however, would be appropriate (paras 5.50-5.52).  ANNEX I1 Page 18 of 18 47. In addition to sources of increased Government contributions to FEN's equity specified in Law 25 of 1990, another significant source may be the prepayment of outstanding foreign loans to power companies which the Government is now considering, as a result of Colombia's recent significant improvement in its balance of payments and foreign exchange positions. The Government has not yet indicated how this would be accomplished in relation to the power companies which are the borrowers of the loans which may be prepaid. To maintain appropriate financial discipline on the part of the power companies, one method that may be advisable would be for the Government to make the prepayment in exchange for a note from the power companies to continue to service the amount prepaid by the Government, and for the Government to transfer the note to FEN as an equity contribution. The terms of the note, under which service of the amount prepaid would continue, might be the same as would have continued to apply if the prepayment were not made. More appropriately, the repayment period should be extended as suggested in para 20, and the foreign exchange risks and interest bases might be simplified as suggested in para 28. ADDENDUM Information about the funds of the Social Security Institute (ISS) which FEN has been administering in a fiduciary capacity is contained in the Audited Financial Statements for 1988-1987, Annex C, pages 30-31: Audited Financial Statements for 1989, Note 20 and Schedules 1,2 and 3; and FEN's Annual Report (English version) for 1989, pages 14-16 and for 1990, pages 12, 13, 16 and 17. In the appraisal report (SAR No. 4771 - CO, para 2.10 and Annexes 2.2 and 2 . 5 ) , these funds were treated on the basis that they would be reported as part of FEN's financial statements, with the resources shown as liabilities under the caption "Official Funds", their uses as assets and the attributable income and expenses under pertinent categories on the Income Statement. The SAR states that the resources made available from this source on a monthly basis represent proceeds of Constant Value Bonds issued by ISS to be managed by FEN and are to be repaid over 25 years at 5.5 percentage points above their inflation adjusted value. The use of the ISS resources was limited to financing the acquisition by power companies of domestically produced goods and services (a restriction which was lifted by Law 25 of 1990) and to make short and medium term temporary investments in high yield liquid securities. (FEN's Annual Report for 1989, English version, pages 14 and 16). As of the end of 1989, loans and investments in this account totalled ColS11 billion, of which loans were 31% and investments 69%. Since inception of the account in November 1984, total income earned on loans and investments has been insufficient to cover total costs, of which about 93% represents the costs of the funds provided by ISS and 7% commissions paid to FEN (Audited Financial Statements for 1989, Schedule 3 and for 1988 and 1987, Annex C). FEN'S Annual Report for 1989 states that because there was little demand for these resources on the terms provided, it was difficult to administer them profitably. No explanation is provided as to why the account's surplus funds could not be invested at yields higher than the cost of the funds, which in 1989 appears to have been about 29%.  A N N E X1 x 1 Page 1 of 6 .: .-:,.- ,.-- Santai5 de Sogotd, D.C., - ; ci;2 ,;:-.:.? Senor SFL4;iX?:I SkIITii Jefe 3ivisidn Infrasstructura y Energia Departamento 111 Oficina iizgional para America Latina y El Caribe Banco :vIundial :,Vashin,~tonD.C. Apreciado senor Smith: En atencibn a1 Inforrne de TerminaciSn de Proyectos corrtspondiente a1 PrCstamo 2401-CO entre el Banco Mundial y la FEN, me perinito hacer a contii~uacibn10s siguientes comentarios: 1. Parsgrafos 3.01, 4.01 y 4.05. En cuanto a 10s objetivos del PrSstamo, el Informe hace Bnfasis en que desde su creacidn la FEN debid ejercor funciones como Banco de Desarrollo. Si bien en la dpoca de evaluacidn del credit0 y de la negocizcibn de las minutas del contrato Cste fue enunciado como uno entre varios propbsitos, se di6 nuchisirna in5s importancia a1 objetivo de apoyar a Colombia en sus esfuerzos para co~npletar una oferta electrica que diera respaldo a1 desarrollo y a1 ern?leo. Por tanto, el propdsito fundamental expresado entonces f u s asegurar la construcciBn de proyectos de expanstdn elCctrica para evitar racionarnientos. No creernos que misiones de asesorfa a FEN en junio y noviembre de 1982 hayan logrado avances en la constitucidn de la FEN como Banco de Desarrollo cuando la entidad apenas empet6 a funcionar a finales de 1982 con una preocupacidn blsica centrada en su capitalizacidn proveniente de 10s recursos del Fondo de Desarrollo EIctrico, administrado por el Banco de la Repdblica. Segfin nuestros registros escritos, Y la "tradicibn oral" el origen de la FEN fue bien dtferente a1 que se quiere seRalar en el Inforrne. A1 tismpo, otras actividadas iniciales fueron la captacidn de ahorro local y la transferencia de recursos mediante el mecanismo del redescuento, antes que la conformacidn de una entidad para el control y la coordinacidil financiera sectorial.  ANNEX I11 Page 2 of 6 Pardgrafo 4.08. El papzl de la FEN. Siendo cierto lo anterior, no es vPlida la afirmaci6n de que un aspect0 central del credit0 fuera el papel de la FEN corno Banco de Desarrollo para el sector elSctrico. Xealinente 10s mayores esfusrzos tendieron a1 curnplimiento del objetivo destacado en el punto anterior y se vi6 a la FEN mas corno un canal de recursos, primer0 en inoneda local y luego en divfsas, hacia las empresas del sector eldctrico. Tan solo trzs o cuatro anos despuds se empezd a proaover la transforrnacibn de la FEN para que se dedicara a 10s proflsftos que se mencionan en el Informe tendientes a1 control y manejo de la situacidn financiera sectorial. Paragrafo 4.09. Uno de 10s factores que condujeron a la situacidn mencionada era el papel que jugaba ISA en el sector electric0 por la Bpoca Ciertamente, en el Gobierno National, particularmznte en el Ministerio de Minas y Energla, y tarnbign en el propio Banco, se rnostrb una gran cautela en cuanto a que FEN asumlera desde un comienzo funciones de control y coordinacibn que venfa desernpehando ISA. En la prsctica, durante el per'iodo 1983-1986 ISA continue mznejando la coordinacidn tecnica y financiera del sector, inientras que la FEN se dedicd a la consecucidn de 10s recursos que demandaran las empresas para cumplir sus planes de expansi6n. Como resultado, ni en la FEN ni en el Gobierno, ni en el prop10 Banco, hub0 las condiciones propiclas para que la nueva entidad asumiera funciones de Ilderazgo en las finanzas seceoriales. 4. Las transformaciones a1 interior del manejo sectorial fueron muy graduales. Como lo anota ei Informe, por primera vez en mayo de 1983 el Gobierno asurn16 un papel decistvo en las definiciones sobre el plan de expansidn el&ctrica, cuando el CONPES debatlb el terna Entonces se consider6 que el crecirniento de la demanda de energla electrica estarla bastante por debajo de 10s prondsticos del sector y que por tanto se podZa aplazar notableinente la construccibn de nuevos proyectos de generacfdn. Fue tainbisn la primera vez en la que se mencionaron la$ necesarias correlaciones entre el desarrollo eldctrico y el plan macroecon6mico y se empezaron a evidenciar las restricciones financieras derlvadas del alto endeudamiento. El resultado ha sido que tan solo un proyecto nuCvo se ha infciado desde ese ano. Es el caso de la Central HidroelBctrica de Riogrande 11, en cuya definici6n prim6 el que se trataba de una obra de mClltlple propbsito, por lo que su importancia para cubrir las, necesidades de agua potable en el lrea de Medellln fue decisiva Los demss proyectos en 10s planes de expansidn preparados por el sector, aJn desde 1977, est5n pendientes de una definicien sobre la fecha en que deba iniciarse la construccibn de sus obras prlncipales.  ANNEX I11 Page 3 of 6 En c'ctanto a1 aspect.:, financisro, entre 1385 y 1987 hizo crisis la deseonfianza rnanifestada por el propio Sobierno Nacional hacia el papel de ISA en estas rnsterias. Es asi colno en la definicidn del plan de ajustt la adrninistracidn del perlodo 1385-1990, sz encornendd a la FEN la preparacidn de las proyecciones financieras y el sdguimienta de las misrnas, labor que sirvid corno referencia para la contrstacidn del Credit0 2 8 8 9 4 9 entre el Banco y la RepGblica de Colozbia. 5. Farsgrafo 5.07. Otro terna dal Inforne, sobre el papel de la FEN en 10s aAos anteriores, se refiere a la evaluacldn de 10s pr8stamos. Hasta finales de 1990 la FEN solaqent2 otorgd crdditos destinados a la inversibn. En general, esos crGditos, incluldos 10s proveniant2s del Prdstamo 2401-CO, tuvieron como propdsito coinplementar las fuentes de financiacidn de proyectos en construccidn, rnuchos de 10s cttalcs hablan sido examinados previamente por el Sanco klundial o el BID. Por lo anterior, se consider6 que no era llecesario un nuevo proceso de evaluacidn dado que esos proyectos tenfan estudios de prefactibilidad, factibilidad y diseno e inforrnes de evaluacidn de uno u otro banco. Sin embargo, la FEN desde 1987 ha constituTdo grupos de evaluacidn para 10s nuevos proyectos, con base en la metodologla desarrollada con ocasidn del PrBstamo BID 237/IC-CO. Internament~, el proceso de asinnacidn de crdditos contempla el andlisis tdcnico, econdmico, financidro e &stitucional tanto de 10s proyectos corno de las empresas solicitantes de 10s recursos. El documento de evaluacidn es sometido a la . consideracijn del Cornit6 Interno de Crddito quien presenta recomendaciones a la Presidencia de la FEN y dsta a la Junta Directiva 6. Paragrafos 5.08 y 5.09. La bass de 10s modelos de proyecciones utilizados en la FEN y por e l sector electrico proviene de 10s modelos utilizados por el BID (SPMOD) 10s cuales han sido revisados y mejorados en Ia FEN; con el reconocirniento del propio BID. Estos modelos reeinplazaron el modelo FAST utilizado por ISA y el Banco Mundial, con la acegtacidn de este Gltixio. Por tanto, las criticas sobrs el particular debleran resolverse conjuntainente con 10s bancos multilaterales. En cuanto a la informacidn es evidente la dificultad en su manejo, sobre todo cuando hay tantas entidades a cuyo cargo se tienen par6rnetros fundamentales en la preparacidn de las proyecciones (escenario macroeconbrnico, tarifas, balance energgtico, inversiones, etc.) y, ademb, teniendo en cuenta la gran complejidad institutional del sector electrico. Lo anterior hace que se requieran rnodelos que representen adecuadamente estas caractzrlsticas. Solo modelos con algtin nivel de detalle pueden producir proyecciones que sean cercanamente representativos de la posible evoluciBn del sector. Desde luego es conveniente tener modelos mas Calle 71A ND6-200 1 ~20 3osota. D.E.C O ~ O T ~Teij; ~157.1)2:7?:00 ~ j : ~ [ 5 7 - 121 19776 3 , S 1 ~ a"20;l FE>,ICO ~ ~  - 117 - ANNEX 111 Page 4 of 6 mencionados en el Informe relatives a 13 excelente imagen con que cuenta la entidad entre 10s ahorradores y a 10s mecanismos de captacibn d~sarrollados durante su existencia. Estos rscursos se han trasladado en crbditos a Iss empresas, principalrnente destinados a1 servicio de la deuda que cuenta con SarantPa de la Nacidn. 8. Par'agrafos 5.39 y 5.40. Como ya se rnenciond tambien ha sido gradual el proceso de transformacldn de la FEN, el cual se concretd con las reformas legales efectuadas entre 1990 y 1991. En primer lugar, con la Ley 25 de 1990 y su Decreto Reglarnentario 1805/90, que la convirtiera en la Financlera Ener~Stica Naclonal S.A. - FEN, ampllando su radio de accidn tanto en lo relativo a las empresas a las que puede otorgarles crddlto como en cuanto a 10s rubros elegi3les para financiacibn. AdemBs, se cambid el rCgimen de garantfas. Tambidn dursnte 1990, una nueva reforms, pernitib a la FEN otorgar crCditos directos en rnoneda local haciendo que el redescuento fuera una opcibn para la colocacibn de 10s crdditos locales, pero ya no la Gnica alternativa. En 1991, con la reforma financiera, se dieron mayores posibilidades a la FEN, inscribiBndola dentro del regimen correspondiente a las Corporaclones Financieras, por lo que la entidad tendr3 mayor flexibilidad en el desarrollo de sus actividades frente a1 sector de energla (sobre estos t e n a s hay explicaciones m5s detalladas en 10s documentos preparados por la FEN y entregados a1 Banco Mundial sobre el tema de la cl5usula de prenda negativa.) Todas las transforrnaciones mencionadas a lo largo de varios aios posibilitan una efecttva funcidn de la FEN como Banco de DesarroIlo, controlador de la gestldn financiera de las Empresas. Este papel ha sido confirmado por el Gobierno Nacional durante la presente -4drninistracidr1, la cual defini6 el plan de reforma del sector en el docurnento aprobado por el CONPES el 21 de mayo de este afio, "Estrategia para la Xeestructuracidn del Sector Eldctrico". All1 se definen las tareas de 12s diferentes entidades y, en particular, a la FEY s e le asigna la suscripci4n de convenios de desernpeno con las principales empresas del sector donde se les definan aetas de gestidn. 9. Se hace mencidn tarnbisn en el Informe a que la FEN se ha convertido en una ventana cuasi-fiscal de recursos hacia el sector eldctrico a travCs de la capitalizaci5n de la entidad. Esta aprzciacldn no corresponde a la realidad, puesto que las capltalizaciones de la FEN han provenido originalrnente del traslado de 10s rscursos existentes en el Fondo de Desarrollo Eldctrico y m8s adelante casi en su totalidad de las proplas utilidades de la FEN. El manejo de 10s recursos y la colocacibn de 10s cr6ditos han sido 10s correspondientes a una entid'ad financiera vigilada por la Superintendencia aancaria, buscando preservar la sanldad y estabiiidad de ia empresa en bencficio del proceso de ajuste del sector  ANNEX 111 Page 5 of 6 . ! . ;'FlNANClERA : ENERGETICA ' NACIONAL S.A sirnplificados, 10s cuales ya se han desarrollado tamblSn, pero 6110s solamente deben utilizarse para cizrtos an3llsls de sensibilldad, luego de haber calculado 10s Brdenes de magnitud de las proyeccfones con 10s rnodelos mas detallados. Paragrafo 5-18. Sobre el tzna de la movilizaciSn de fondos en rnoneda local, el Informe plantea que la FEN en sus grlrneros afios no pudo llevar a la prsctlca un traslado masivo de recursos. Se explica, con razbn, que la principal causa estriba en prlmer lugar en el tamailo reducido del sector financier0 colornbiano, pero taabidn en la desconfianza de 10s bancos y corporaciones hacia las entidades del sector electrico. Sin embargo, no s e hace referencia a que el nivel de las captaciones y las condiciones financieras de las misinas estaban regfdas por las autoridades inonetarias con fuertes restricciones, en particular en cuant:, a su volurnen. Todo ello condujo a que no se cumpliaran las rnetas sobre el particular establecldas originalmenee. Lo anterior cambid dr3sticaixente en el Gltimo a o . En efecto, en noviembre de 1990 la Junta Monetaria levant6 las limitaclones a la FEN para sus captaciones tanto en 10s montos coino en las tasas de interbs. En~oncesel nivel de tltulos de la FEN en circulacidn en el mercado local ascenda a unos 54.500 millones. La cifra pas6 a unos 66.000 millones en diciembre de 1990 y luego ha tenido la siguiente evolucidn durante el transcurso del presente d o : m v L O S FEN EN CIRCULACION, 1991 FIN DE Marzo 128.818.0 Junio 222.472.8 Septiembre 297.928.2 Diciembre 356.893.7 En un afio el nivel de capcaciones sz ha hecho mas de sets veces mayor, lo cual es un resultado. del conjunto de medidas reIativas a la transformacijn de la FEN y2 expuesto y, desde luego, a 10s dos elementos  ANNEX 111 Page 6 of 6 I .\ -'(FINANCIERA I ENERGETICA NACIONAL S.A. electrtco y del buen cuirlado de 10s recursos colocados en la FEN tanto por 10s ahorradorzs locales como por 10s bancos internacronales y por el propto Gobierno. El cornentarto del Banco parece referirse a1 mecanismo del Fondo de Monedas Extranjeras (FODEX), con cuentas tanto del Gobierno Naclonal como del Banco de la Repfibllca, que sirvieron para atender obligaciones de varias entidades, principalmente del sector electrico, con el exterior. Recientemente (desde hace un afio, aproximadamente), la FEN ha establecido una llnea de crddito para otorgar prCstainos a las empresas tambien con el objeto de que sirvan oportunamente su deuda externa. Sin embargo, la asignacidn de estos prestamos se hace tambien a travds del proceso crediticio de la FEN, dando cumplimiento a 10s principles establecidos en el Reglamento de Crddito de la Financiera Por tanto, 10s incumplimientos de las empresas en sus obligaciones con la FEN conducirlan a que se suspendan 10s desembolsos y a la eventual aceleracidn de 10s crbditos. Espero que estos comentarios sean de utilidad para ustedes en la preparacidn del Informe definitlvo. Con-gusto estaremos dlspuestos a discutirlos con ustedes sf lo consideran necesario. Cordial saludo, Presidente I vi7$