Colombia - Third Medellin Power Project

-4.URCULMTINU £ty CONFIDENTIAL Reort No. Jt'n liiLIi4AliULNliL DLiNN rULL LuLULN Af1UUllUN uLJ uLVLLUrIzINL PtKUKMAINCE AUDI) or COLUMBIA: THIRD MEDELLIN VOWER PROJECT (LOAN 369-GO) May 23, 1974 Operations Evaluation Department  Tnnn Tn. Qro on 4-b r Pm,1,hlionc do M.dollin (PPM)- with Guarantee of the Government of Colombia, was closed in November r% - n Mef0lV-wng-erf-c.IlMnc- .adit tl.ea e-axoytent tn Tl.Thiflh the physical and economic objectives and financial targets stated in the documents on tne vasis U which the lan wo approved by the Executive Directors were attained, and examines the role of the Bank in connection with LtIe projctL in tis lightL. The report is based on a proje cometion report, cverin the engineering aspects, submitted to the Bank by the borrower in July 1973 (Integral: Desarrollo Hiudroelectrico del n.o Nare, Central de GuatapA, Primera Etapa - 280 MW, Informe de Terminacion), on review of Bank files and of the financial and engineering quarterly progress reports submitted by EPM during the project's construction, and on two days of discussion with EPM and its engineering consultants in January 1974, inuuuing a visit to the plant-site. Use was also made, in preparation of the report, of work carried out in 1971 by the predecessor Division to the present Operations Evaluation Department (IBRD Report No. Z-17, Operations Evalua- tion Report: Electric Power, March 10, 1972 and Report No. Z-18, Bank Operations in Colombia: An Evaluation, May 25, 1972). No separate verification of the financial data used was undertaken but the reports of the External Auditors who have examined EPM's accounts each year since 1960 were studied and information material to the present enquiry was used. The valuable assistance provided by the Empresas Publicas de Me- dellin, its General Manager and senior staff is most gratefully acknowledged. Note: Currency Equivalent: Annual Average non-preferential Exchange Rates used in this report are as follows (in Colombian Pesos per US dollar): 1963: 9.00 1969: 17.37 1964: 9.00 1970: 18.49 1965: 10.50 1971: 20.41 1966: 13.50 1972: 22.58 1967: 14.73 1973: 24.79 1968: 16.38  PERFUKAINCE AUUIT OF COLOMBIA: THIRD MEDELLIN POWER PROJECT (LOAN 369-GO) Table of Contents Page Summary i Introduction 1 Project Implementation 2 Prolect Costs 4 Economic Justification 6 Financial Performance 9 Institutional Development 12 Conclusions 13 Tables 1 Prnjet Costs 5 2. EPM: Self-Financing of Investment 9 3. Estimntp nf FPM'q Revnne Loss due to Stealing of Electrical Energy 11 Annex Tables I. Approximate Costs of Guatape' I Project Supported by IBRD Loan 369-CO of 1964 Generation and Load III. Empresas Publicas de Medellin: Growth of Electricity Supply Summary Sources and Applications of Funds 1961-72 Map Empresas Publicas de Medellin: General Layout of the Electric System Chart Empresas Publicas de Medellin - Power: Load and Capacity Development, A c-a' aA Foecas 1 Q 71 tL LU L uIL V'J~L . L 1L J. I .1  This report deals with the third of four power loans which the Bank has made to the Empresas Publicas de Medellin (EP), tne second largest regional power company in Colombia with a service area of some 2.9 million people. Loan 369-CO, in the amount of $45 million and with the longest term ever granted by the Bank on a power loan (35 years), was signed in February 1964, amended in December 1967, reduced to $39 million by cancellations in 1969 and 1970, and finally closed in Novem- ber 1972, nearly four years after the originally established closing date (December 31, 1968). The centerpiece of the project, accounting for more than 85% of expected and actual total costs of about $75 million equivalent, was a complex hydroelectric plant diverting water from the Nare River, at some 1850 meters, and discharging it into the Guatape River, at about 1000 meters, and involving tunnelling of some 13 kilometers and construc- tion of a large powerhouse 650 meters below the surface. There were three major delays - nine months due to contractor difficulties on the powerhouse access tunnel, one year on the dangerous penstock tunnel (in which twelve people were killed), and one year because of supplier's failure.to adhere to contracts for control equipment, However the small dam involved was realigned and built 5 meters higher than originally planned, to impound three times as much water, and cost savings enabled all four 70 MW turbine-generator units envisaged for Stage I of the Guatape plant to be procured instead of the two initially provided for under the loan. Given the considerable difficulties of nature involved and poor performance of some of the contractors, especially those for control equipment, the project was very well executed. Load growth, both in the Medellin area and on the system created by the achievement of interconnection among the major cities of central Colombia late in 1971, was such that Guatape' had it been completed on schedule, would have been premature, but, as actually completed in 1971-72, was very timely. It accounted for nearly half of EPM's generation in 1972 and well over half in 1973, when it ran at a capacity factor of more than 66%. and it has been supplying energy to the interconnected network since March 1972. In view of the further development that was possible within original cost estimates and the trend that fuel prices have since assumed, there is no doubt that the extra investment required to build Guatape rather than substitute thermal plants vields more than the 12% rate of return originally estimated, despite the delay in realization of benefits. A two-circuit 220 kv transmission link between the plant and Medellin, six new substations around the city or in neighboring areas and expansions of most existing substations have been satisfactorily completed, but the transmission and distribution program which was sup- pnqed tn nrnnunt for nearly 15% nf total nroipct exnenditures and actuallv received only about 10% was not sufficiently defined in the appraisal - ii - reDort to enable any firm statemnt now aq to how muh of it h hpn accomplished. Expenditures lagged severely, apparently due partly to Rhortagp of Ppgo ri-_qniiri-, and As az nresuilt of theuneTnesaet EFM's registered customers grew considerably more slowly than appears ton have-, bee,n expectedA - at only abu 4;5% p. atruh1968 -A only 6.5% p.a. over the whole period 1963-72, despite major efforts to make k, LL.1- - o LL-.tt...CL 1.SCL LL D JtJU aLLULL L II iL. been growing about 6% p.a., it can be estimated that the proportion of 73% in 1968 and recovered to only a little over 80% by 1972. Relative neglect of distribution and of metering/billing control by EC7 (and by the Bank) ledU to tne rise of very large stealing of energy - by small industry and large industry in established areas and by oour people in areas where ErN faileu to proviae racilities - which, at its peak in 1968, accounted for 16% or more of total generation. It is estimated that EFI lost to stealing (in excess or the inevitable min- imum) some 240 million Pesos, or about $15 million, in potential revenues between 1964 and 1972. Although the Bank supervision mission in October 1964 noticed the emerging problem and raised it with EPM, follow-up letters then and in subsequent years failed to rerer to it, despite the serious fiaancial consequences it was to have, and concentrated rather on the poiLtically difficult need to adjust tariffs in light or inflation; no further supervision reports were filed before 1969. There have been untimely delays in EPM!s adjustment or power tariffs to keep up with in- flation, particularly in 1965-66 and in 1972-73, but over time they have kept up well (better than if they had been indexed at the outset) and the main reason for EPM's failure to reach most of the important financial targets agreed with the Bank in connection with the 1964 loan agreement has clearly been loss to stealing. Internal self-financing of construc- tion expenditures fell repeatedly below the 40% agreed with the Bank for each five-year period starting December 31, 1963 - whereas it would have reached well above that level had stealing been held to more normal levels - and it rose substantially only with the drop-off in construction expen- ditures in recent years as Guatap6 Stage I was drawing to a conclusion. Self-financing has sharply deteriorated for some of the other less costly services for which EPM is responsible which were referred to in the loan agreement but not checked by the Bank during project execution. EPM's electricity department exceeded the 60/40 debt-equity limit agreed with the Bank, beginning in 1967. The deficiency in project supervision and in attention to distri- bution aspects of the project is partly explained by the Bank's preoccupa- tion in the period with the greater problems of some of the other Colombian power companies to which loans had been made and with the much bigger and more important issue of interconnection among them. But the Bank does seem to have given less attention to Medellin than would have been worth- while; more and better project supervision, better response to EPM in connection with its problems with foreign suppliers and also a somewhat less accommodating attitude on financial matters might have helped EPM to offer its community better service. Most of the large contingencies - 111 - wisely built into the original estimates of the project's toreign ex- change costs proved fortunately superfluous but were cancelled quite late, having cost EPM $250,000 in commitment charges, while those that were spent were devoted to interest payments to the Bank, includ- ing about $1 million which the Bank disbursed against 'interest during construction' although it was for interest payments after the plant was completed and earning revenues. The Bank's appraisal mission gave useful help in securing local borrowings, from U.S. A.I.D. and from a local development finance company, but it appears that the availability of these funds caused EPM to slacken its efforts to raise tariffs. The 35-year term to the loan was justified, given the long life of hydro- electric facilities, on the basis of national considerations of debt- service and exchange-earning prospects, which in retrospect are of un- certain validity, but. irrespective of their merit at the national level, they may have been inappropriate for a company enjoying the fundamental streneth but facing the political problems that EPM does. However, the more serious shortcoming on the part of the Bank was, once having helped to gather the necessary financing, not to promote and assist a pattern of expenditure which would not only have brought earlier benefits to consumrs hut also have greatly eased EPM's financial problems.  PERFORMANCE AUDIT OF COLOMBIA: THIRD MEDELLIN POWER PROJECT (LOAN 369-CO) Introduction Medellin is the second largest city of Colombia, following Bogota, and the capital of the Department of Antioquia, and Empresas Publicas de Medellin (EPM) is the second largest power company, supplying areas with a total 1973 population of some 2.9 million out of Antioquia's 3.3 million: the city itself (population of about 1.3 million), neighboring towns in the metropolitan area (about 600,000), other small towns aggregating about 100,000 population and, through the Electrificadora de Antioquia, some 45 rural municipios with a total population of 900,000: in addition, since 1972, it has been selling in bulk to the interconnected network which now links the main centers of central Colombia. EPM's zenerating plants are entirely hydroelectric. EPM also has separate departments resoonsible for water, telenhone and sewerage services to the city and to limited areas around it. EPM has been one of the larger borrowers from the Bank for elec- tric pweo l-, nunt4na to ate for Al3 mi lin rnmmi tmpnt-c in f nir separate loans, the most recent of $56 million in January 1973. This n,r Fn-r nnra( niiq i f- A n nn (1 ir T.7i ti- t-10 f-i fl I -n i r fc frir T.14 1 f-k. RMI,- made Loan 369-CO in the amount of $45 million in February 1964; final disbursements out Of this loan were made in November 1972. Guatape hydroelectric projectil, taking advantage of the difference in elevation1 betee th reRver, at some 185 meters, and thet= GUtapU River, at about 1000 meters, in the central range of the Andes crossing .LC fUtLCC OU Li CI.. C C. AL. JflC LL PLUJCtL_t_ L PLU V.LUr_U J-UL aL two-circuit 220 kv transmission link between the plant and Medellin, soeQ o l 1oees t . IU L_ r-L.LI. LC LU L-L.e west, anIL0U t exp CaanbUi Ut L~L rrl UL 6LL-L- bution system. The Guatape" project had been studied since early 1961 by EPM and in March 1963, the project was appraised in June, and the loan negotiated in Jetcel . ±eL o r CLUo U a LLLCLCUbL LLe an1 J-yea t, longer than any other loan the IBRD has ever made for power, which was Lelt LU Ue consistent with Le average life of the works to be built but was justified by the Bank mainly in terms of restraining the growth of uUuias debt service vbuen anu setuing a good example to the newly formed Consultative Group for Colombia; service on existing foreign det was L6O Ln 1964 declining to only 5Xr by 1969, but the country's "relatively sluggish export outlook" advised some lending on very long terms for the ambitious development program then envisaged. The loan became effective in August 1964, six months after signature and somewhat 1/ The fourth loan, in January 1973, was principally for the second stage of this same project. - 2 - later thLian epcd, due t the t~L~Mime requi. .atL toI fulfil 1-1 ments that the Colombian Government imposed on EFM in return for its guart e ofLO tIL 'l oan±I . . m AO 4 iJ jO.Ls.LLLi. .ji .L.a..c the loan other than a requirement that consultants be engaged for detailed iu vecemuver 1967 t e Executive virecturs approveu an amenUment to the loan agreement splitting the original loan into two tranches, of e%1 miLilou at J'oo l auu $ milliou at 67i, te gi IBT, L-in tst rate at the time, with the second tranche, consisting of then-expected cost savings on the original project, available for purchase of two 70 MW turbine-generators additional to the two included under the 1964 loan. A year later, work on the project had progressed sufficiently far that EPM felt it safe to apply for cancellation of $2 million of the first tranche and $1 million of the second tranche, which was duly approved by the Bank in January 1969. In August 1970 a further $2.5 million was cancelled from the first tranche and $0.5 million from the second tranche, reducing them to $36.5 million and $2.5 million respec- tively. But in October 1972, fifteen months after the Guatape station had finally come into operation when contractor claims were being set- tled, the Bank agreed to transfer $550,000 back to the lower-interest first tranche, to cover small additional costs on the basic project and $381,000 -of interest currently due to the Bank, which was added to the "interest during construction" item in the List of Goods. Final balances under the loan were therefore $37.05 million under the first tranche and $1.94 million under the second tranche. Repayment to the Bank 'egan in 1969, in accordance with the five-year grace period of the original loan. Project Implementation The generation portion of the project, which accounted for about 85% of originally expected costs, has been completed largely as origin- ally envisaged but with a delay of some 2.5-3.0 years. The first two 70 MW units, originally expected to be completed in December 1968 and June 1969 respectively, came on line in July 1971; the second two units, foreseen for completion in December 1969 and December 1970 although not included in the original loan in order to keep its size down and the construction/grace period within about five years, actually came on stream in March 1972. Delays arose in many parts of this complex project, involving considerable tunnelling, but there were three stages at which they became critical to the progress of the scheme as a whole. The main elements of the project are a small earth and rock fill dam (Santa Rita) on the Nare River; a 4.9 km inlet pressure tunnel leading from a vertical-drop inlet shaft in the reservoir so created to the penstock, with a vertical- shaft surge tank 150 meters upstream of the valve chamber; a 900 meter penstock at a 48% slope; a powerhouse built for four 70 MW units in a cavern 650 meters below the surface, with vehicular access by a 2 km tunnel; and a 4.6 km free-flow tailrace tunnel conducting the water from the power plant to the Guatape River. The first major contract ment of the next stage of construction was delayed by an overrun of some ntne monhU (LU April 1966) in comUpleUig te access tunnel, uUe to poor organization of the Canadian contractor in the first fifteen months of work, difficulties in securing import licenses for his equipment and spare parts, and serious water seepage problems encountered about a third of the way in. The second major delay occurred on the penstock tunnel, which was finally completed in July 1969, or about two years after the date foreseen in the original plan and a year behind contract schedule. It formed part of the main civil works contract for the project which was let to an Italian-Peruvian consortium and included also the powerhouse caveryi and inlet tunnel. It was the most difficult item in the project. There were three serious accidents, one in September 1967 when 5 people were killed by electrical curreni's in the rock setting off explosives placed, a second late in 1968 when 3 people were killed and another seriously injured due to breakage of the cable supporting the cable car in the tunnel, and a third in March 1969 when 3 people were killed because the drilling platform came loose and another was killed in the subsequent rescue operations. The first accident required a major change in the construction method. These tragedies and the serious construction delays reflect above all the sheer difficulties and dangers of excavating a 900 meter long tunnel at a slope of 480. The Santa Rita dam, which was the only item greatly changed from what was foreseen in the appraisal report, being built to an alignment such that it could be incorporated in the higher dam required for the second stage of Guatape and with an additional 5 meters height impounding a reservoir of 73 million cubic meters instead of the 20 million foreseen at appraisal, was also delayed, both in construction, due to poor organization of the Italian-Peruvian consortium which also won this contract, and in the subsequent filling of the reservoir, due to difficulties in procurement of land which would be flooded and in settlement with the neighboring town; but these delays in the dam and reservoir were never on the critical path for the project as a whole. The third major delay - and in many ways the most expensive of all, because it postponed effective operation of the plan from mid-1970 when virtually everything was completed until July 1971 - was in the supply of a few small but critical items of control equipment. For some of these items EPM was able to provide temporary substitutes, but for the excita- tion cubicles, contracted with AEG, the generator control boards, contracted with Brown Boveri of Germany, and the turbine control board, subcontracted by Escher Wyss with AEG, all delivered with about one year of delay from contract schedule, EPM could only wait.1/ Numerous visits were made to 1/ It is interesting to note that another recently completed IBRD-assisted power project - Karachi Electric Supply Corporation's 125MW Korangi 3 thermal unit .- was also delayed at the end, in 1969, by late supply of control eauioment. The supplier in this case was Brown Boveri of Switzerland, and the delay much less, about four months. Germanty, JouL tiie mandEuacurersL appar.enlLy hdU inerab))leUL prob'UlemsL o4 overcommitment and lack of labor; EPM feels that the Bank could have obtained a higher priority for themu by suggeting LU L mt! n LIIULakLUIeL that their poor performance might affect the Bank's viewpoint in discus- sions about future contract awards under IBRD loans, but, despite rM s representations, the Bank apparently felt unable to intervene in borrower- contractor relations. The last of the essential items required from AEG and Brown Boveri finally reached the project site in the first quar- ter of 1971 and, after adjustments for differences between design specif- ications and what was actually delivered, test runs of the whole plant started in June. As regards the transmission and distribution component of the project it is impossible to say with any precision how much of the works envisaged at the time of appraisal have been accomplished, since the appraisal report was almost silent on the matter. There seem to have been substantial delays - less than half the foreign exchange envisaged for this category !ad been spent by the end of 1968, the original target date for completion of the whole transmission and distribution component (see Annex Table I) - but a number of major works have been successfully accomplished, including the 220 kv transmission line to Medellin, comple- tion of the 110 kv circuit around Medellin, construction of six new sub- stations and expansion of most of the existing ones. Since completion, Guatape I and its transmission link with Medel- lin have generally operated very satisfactorily. Production increased from 180 million kwh in 1971 to 1270 in 1972, when it accounted for nearly half of EPM's gross generation, and 1,610 million kwh in 1973, the first full year of operation, when it ran at a capacity factor of more than 66%. Peak load taken has been 285 MW. The only serious outage occurred in November 1972 when the plant was out for most of nine days as a result of air being caught in the inlet pressure tunnel due to garbage blocking the trashracks cf the intake tower in the reservoir. Project Costs Actual total costs cannot be specified with any precision, partly because of difficulties in defining exactly what should be included for comparison with original cost estimates and partly because of the frequent changes that have occurred in the parity between the Colombian Peso and the U.S. Dollar; Table 1 shows three different estimates of actual costs, of which the most accurate may be the last one, based on EPM's Quarterly Progress Reports about the project, although it probably somewhat under- states land acquisition costs (see the estimate based on EPM data for this item) and interest during construction.l/ 1/ Adjustment to include interest during construction on all local cur- rency expenditures on the Dower plant (assuming interest rate of 10%, compared to 8.5 - 14.0% rates at which EPM's marginal local borrowings were made at the time and a 7.7% averaae rate of inflation in wholesale (continued on page 6) Table 1 PROJECT COSTS (in US $ millions) Appraisal Report 7ased on Integral Base on EFM Data Based on Progress Reports Foreign Domestic Foreign Domestic Foreign Domestic Foreign Domestic Exchange Currency Total Exchange Currency Total Exchange Currency Total Exchange Currency Total Land, Access Roads, Etc. - 2.24 2.24 - 3.20 3.20 0.09 4.52 4.61 - 3.76 3.76 Civil Works 16.61 1:3.26 29.87 17.10 15.,0 32.70 17.10 17.94 35.04 17.10 20.49 37.59 Powerplant Equipment 8.94 0.67 9.61 7.55 2.05 9.60 7.55 1.89 9.44 7.55 1.49 9.04 Engineering 1.00 2.22 3.22 1.12 4.98 6.10 1.12 4.98 6.10 1.12 5.65 6.77 26.55 18.39 44.94 25.77 25.83 51.60 25.86 29.33 55.19 25.77 31.39 57.16 Transmission & Distribution 4.77 3.91 8.68 4.85 1.15 6.00 4.85 6.77 11.62 4.85 1.00 5.85 31.32 22.30 53.62 3C.62 26.98 57.60 30.71 36.10 66.81 30.62 32.39 63.01 Interest during Construction 5.00 0.46 5.46 8.38 1.02 9.40 8.38 3.25 11.63 8.38 3.99 12.37 36.32 22.76 59.08 39.00 28.00 67.00 39.09 39.35 78.44 39.00 36.38 75.38 Price Contingencies 3.09 3.38 6.47 Physical Contingencies 5.59 4.32 9.91 Subtotal 8.68 7.70 16.38 TOTAL 45.00 30.46 75.46 39.00 28.00 0 7.00 39.09 39.35 78.44 39,00 36.18 75.38 Note: Domestic Currency converted into US dollars at the average exchange rates prevailing in the years when the expenditures were made. Sources: Appraisal Report No. TO-391b of January 28, 1964. Integral: Desarrollo Hidroelectrico del Rio Nare, Central de Guatape, Primera Etapa - 280 MW, Informe de Terminacion, July 1973. EPM: Data provided in January 1974 by EPM (see Annex Table I part A). Progress Reports: Data given in EPM's quarterly financial progress reports to IBRD since inception of project construction (see Annex Table I part B). Measured in US dollars, actual total costs of the expanded project - including the higher dam and larger reservoir and four 70 MW turbine-generators instead of two - have probably been about the same as (or at most a few percentage points above) the appraisal report's allowance, including substantial price and quantity contingencies, for the original project. Direct foreign costs, covered virtually entirely by the Bank's loan, have been substantially less than originally provided; foreign exchange contingency provisions of some $8.7 million, to the ex- tent they have been used, have been devoted virtually entirely to Interest during (and after) Construction. Domestic currency overruns, measured in US dollars at average exchange rates prevailing in the years when the expenditures were made, have compensated the foreign cost saving and have been particularly important on the civil works, due to the delays and difficulties encountered, on interest during construction and on engin- eering; however, with regard to the latter, the engineering consultant reports that the final costs attributed to this head include EPM's own engineering and administration costs for the project, which were not in- cluded in the original budget, while actual payments to the consultants, at $3.2 million equivalent, were as originally forecast. Substantially less than originally envisaged appears to have been spent in local cur- rency on distribution expansion. The principal reasons for the foreign cost savings are, first, that equipment was procured at exceptionally favorable prices due to nrincinal bids being called at a time when order-books were lowy' (four units plus all auxiliary items were obtained for some $7.5 million, com- nared to thp bagic budget figure- without contingencies. of nearly $9 million for only two units) and, second, that the rock encountered in the riuil woarks qnPr11v P-nod no that construction oroblems were much less than they might have been in the large amount of tunnelling involved. Tn.ql rnet MvPrrin- qp to h VPrV nrPdoinantly dni to clayv. Ecnomic Tietifirnt-inn The oTriinl ecnonmic iiusHf-irntin-n o)f (',lit-AnewasQ hnqp( nn n forecast of EPM's peak demand, assuming growth of about 11% p.a. 1963-70 nn,l17 cl iaht-1 -% h,.1nT.7 t-h- I oncr_-t r hi ctr1ri rnl t-rpntind ~ n I Qr.9 r nn cii I t-. n study of alternative development plans for EPM through 1978 suggesting of total costs and implying that the return on the extra capital invest- m-ent requr ed tA buil G-. 1 O - ,, e' _ni-l- f--kn + .-on- _J-j -4: +F -1,n .- r, (continued from preceding page) 11LL D - ,ULUl 1ULII WUULU LILLita L6iI.Lb LLCi UL ' bU le J 111 10J11 equivalent, which would be partially offset by exclusion of about $1 million interest paU LU te onk aLLUL ULeLte11 UL constructLon but charged to the loan and capitalized. 1/ This situation seems to have changed, as noted, by the time that the control boards were actually being manufactured. - 7 - blLZ WUU LU U IL/0. LLL: DiIIl& b CLpiL*bL LLL.bLULL 11ULU LLtLL CVeLge residential consumption on EPM's system was already high, at about 4,500 kwh p.a., and it suggested that some allowance should consequently be made for saturation; it projected a peak load growth averaging 9% p.a. 1963-70 (see Annex Table 1I). Since the Bank had also begun in 1963 to urge its major Colombian power borrowers to consider the advan- tages of interconnection, the appraisal mission also investigated the prospective need for Guatape on the assumption that interconnection would be accomplished in 1967. The Colombian companies' and consultants' studies suggested that, whether for the EPM system alone or for an inter- connected system, load growth was likely to be such that Guatape would be required by about the end of 1968 or early 1969. The appraisal mission's studies indicated lower load growth for both contexts, such that, taken literally, they would have recommended a postponement of Guatape, by one year or eighteen months, to mid-1970. Nonetheless the mission recommended adhering to the timing implied by the higher load forecast. Although the scenario has in fact been very different, it must in retrospect be considered fortunate that the Bank agreed to finance Guatape' when it did, mainly in view of the delays in the plant's completion. Taken in the context of the Medellin system alone, it finally came on line at a time when, despite even slower growth of peak load in the 1960s than the Bank had predicted, reserve capacity was becoming relatively small and when there likely would have been energy shortages had it not been that 1970/71 was a very good water year; a simulation an lysis carried out by the Operations Evaluation Department and Planeacionl' in 1971 showed that already in 1969, a bad water year, EPM had only about 6% surplus energy from its all-hydro system. As regards interconnection, negotiations tqok much longer than the Bank had anticipated, the interconnection company2/ was finally only formed late in 1967 and the initial network completed late in 1971, a few months after the first units at Guatapd{. Here again load growth was slower than expected so that, when the interconnected sys- tem finally came into being, it had reserve generating capacity ostensibly of the order of some 30-35% of peak load, a relatively high level; but again the 1971 simulation analyses strongly indicated that Guatape's energy contribution was needed at that time, the urgency of the need being limited only by the happenstance of the previous fifteen months' heavy rains. While Guatape would therefore clearly have been premature had it been completed on schedule, there is no doubt that, as completed, it was very timely. As regards the 12% return to the incremental hydroelectric investment, as estimated by the appraisal mission in 1964, it has not been possible to repeat the analysis of alternative develooment programs in ilhe course of this audit, but the trend of fuel prices in the recent past and in prospect and the fact that a lareer plant has been built 1/ The Colombian National Planning Department. 2/ Interconnexion Electrica, S.A. (ISA). within basically the original budget are certainly sufficient signif- icantly to outweig' tne negative effects u Lue lunger plant coustruction period, so that a recalculation on the same basis today would show a higher return. As regards the distribution program, of such unclear scope, envi- saged under the loan, no specific justification was given, but it was pointed out that metropolitan Medeillin's population was expected to grow at about 6% p.a. and that EPM would probably be increasing its service area. Population growth has indeed been in the neighborhood of 6%, accord- ing to current estimates, and EPM's service area has been expanding. But until the last four years, following the period that was supposed to be covered by the distribution program, EPM's electricity customers were growing at only about 4.5% p.a. - which means that the proportion even of metropolitan Medellin's population enjoying EPM electricity service must have been falling, perhaps from some 79% in 1964 to about 73% in 1968.1. However,these years saw substantial growth of illegal makeshift connec- tions which, dangerous and inefficient as they are, may have been enough to keep the proportion effectively supplied roughly constant. In 1969, stimulated by realization of the large revenue loss which was resulting from stealing of energy - tests showed that more than 50% of the energy going through certain substations went unbilled - EPMK started a major dis- tribution program in the poorer parts of Medellin. A- average of nearly 15,000 new residential consumers were added in each of the years 1970-72, compared with about 5,000 in most preceding years (see Annex Table III) and 9,000 of these each year were connections un-er the special program.2/ However, the number of new connections has been declining since 1970, the major effort has been heavily concentrated to the northwest of Medel- lin, the northeast is only beginning to be touched, and little work has been done in the marginal areas of t-he other cities in the region, let alone in the smaller towns other than a few located close to the existing installations. It appears that distribution, which has accounted for only a remarkably low 5-10% of total power investment by EPM even in the most recent years, has been an area of underinvestment, particularly severe in 1964-68, but not yet overcome. At a very general level, then, it is clear that the assistance provided under the project in this area, if too little and too late, was nonetheless very worthwhile. 1/ Estimates based on the following approximate asLumptions: metropolitan Medellin's population grew at 6% p.a. from some 950.000 in 1964 to 1,600,000 in 1973, almost the entirety of EPM's direct residential con- nections have been in this area, and each residential connection is assumed to correspond on average to 7 people. The proportion of An- tioquia's inhabitants outside metropolitan Medellin (Romp 1-7 million people in 1973) receiving public utility electricity supplies,mainly through the Electrificadora de Antionnia iq hplipvpd to hp ahont 15% 2/ Calculations on the ame bai' as the d re n ing footnot e ygget tla- the proportion of metropolitan Medellin inhabitants enjoying EPM elec- tricitv Rprvicp hv flip end oif 1q79 mayi havein beean just oer 80%. -9- Financial Performance EPM'q financial nerformance has been disaoointing compared with appraisal report expectations. The formal provisions of the loan agree- man have norhanc hon adhered to hut the financial tarcets established in the appraisal report have not been met. They key target 'was for EPM's elect.Irict depnaonrtmnti to -enerai-o enn-,hl rfrnho i-n nnnr-p i-nt-crrinl11V 40% of construction expenditures during each five-year period subsequent LUI _ IJLC1L-C -. Id . . -L-L.1O _7 t 1964-68, 1965-69 and 1966-70, as Annex Table IV shows, although it was LL Li bUs equCenIL fiveCJemOl F OL.L%JkLao W .L - v-L construction period of Guatape" (1964-72), time-spans including the recent years wLIen Unstuct iU expeniurLueL siave v- eentmorar4.y mors mo.o m c. EPM also undertook to generate from each of its other departments' opera- tions a 'reasonable surplus' to finance thi epc*4-,- expanson bu since the 'reasonable' level was never defined, as it was for power, and Lucre is rIU evLueIIce LHIS unueIL LeALLL wab rve sLu C uJ ve xtc-1y the Bank, it is not possible to say categorically whether objectives were met or not. Actual performance has been as follows. Table 2 EP: Self-Financing of investment (Share of investment covered by net internal cash generation in %) 1964 1965 1966 1967 1968 1969 1970 1971 1972 Water neg.a/ 34.4 10.3 21.9 46.1 62.5 23.2 12.1 10.2 Sewerage 121.0 48.0 58.9 60.5 46.9 22.9 14.7 9.9 12.8 Telephone 14.5 17.0 neg.a/ 50.9 45.7 49.0 46.1 35.5 42.3 Power 31.3 36.2 37.6 33.0 41.3 30.3 48.9 75.4 84.b EPM 25.2 33.7 28.1 34.2 43.0 34.9 41.8 48.1 78.3 Total Investment in mln. Pesos 198 215 258 236 308 460 410 365 189 hi nn n nr% C It i% 1 ic/ n in o or C 1n n 1 i A n n in min US$Z.' L..U /_V . 1 :7.iJ -U . U 10. 0 /LU.J L)/_4. L l. V 0 .+ a! neg.= negative: i.e. debt service in excess of operating income and depreciation. b/ converted at average exchange rates prevailing for each year. Only in the last years (except for 1968) when investment has been rapidly falling has EPM been able to exceed a 40% self-financing rate over all operations, in which power is very dominant, and these years saw a sharp deterioration of self-financing for water and sewerage. Finally, from 1967 on, EPM's electricity department was unable to adhere to the 60/40 debt-equity limit agreed with the Bank. One reason for the Door financial nerformance has been the diffi- culty in keeping up with inflationary pressures. The general consumer price level in Medellin slightly more than doubled between 1964 and 1979 - 10 - EPM's power employees are well paid and in the highest income brackets of the country; average wage costs to EPM (including prestaciones so- ciales) nearly tripled in current prices (and increased nearly 40% in real terms), to reach about $3,800 eauivalent in 1972. But there was no increase in the number of employees of the electricity department between these two years and there were some benefits from scale economies, so that average total production cost per unit sold slightly less than donihled_ Aq id FPM's avrrae qalps nrict in Medellin: in US dollar terms the retail sales price remained extremely low relative to most other nower systems In the orld, at ahon Picht-tnths of a 1T cent ner kwh (and only about six-tenths for the lower-priced first 200 kwh of residen- tn 1 monthly onumntinn1 While novrall rnf- and nrices have therefore more or less kept in step over the long run there have sometimes been delays i n aisngy tari ffsa in ine- w.4i inflatniion, on nrrnQi nn inrrenqrri by difficulties in securing approval from national Government authorities. For- inst--,ce t-ho 1f3nk-'s prnoetins in 1QI.64 ShowedA the neod for a _5% tariff increase in July 1965 and this was urged on EPM in discussions, I/l- 1--, L compLIIj/anly eventuLLatlly di ded thLa iat Januar 196 would.LL be earlyvJL.I.L.... enough and it finally obtained approval from the Government, which had for application of the increase (raised to 38% by that time) only from Tune 19OAA haA a t5 inceace hen annlied From Tnnarr, ac PM cnnah- thi company would have earned some 17 million Pesos additional in 1966, million Pesos to 1965 revenues, both of which contributions to investment expenituresLwould WLIt-Ltave beenL very tilCy. Mu £I.LL h-O I estsA4tt-CU1tL.C- W-- . encountered in introducing the large 1968 tariff increase (also 38%), bw#- +-er wenre-nn c nside-rale Ael- , 4- 4..., .n .r a . nr .-n c ,,10'i Despite the emphasis put on the matter of delays in tariff increases in iBank C/. . t crepondceC with II LI , LIoweveLr, 4 sL s tr L 1,4 ng> d-Lat eve LI LILIaA t LL 4 0P been indexed to average real costs of production on the basis of 1964, W LCI LUCy bee LU iidVC UCha Lelatively adUequaLz- .. Lt=VULiuUb WUU.LU aULUdILy have been less by some $1.5 million equivalent over the whole period 1YoJ-i£ Lan they actually were - uaily due tu large Size vt cue step increase in tariffs in 1968. A more important reason for the bad financial performance has been the large amount of energy lost to svealing. Total. system losses rose steadily from about 13.5% of gross generation in 1960, including an estimated 1.0% stealing, to a peak of about 28% in 1968, including 17, stealing. Investigations of this problem in 1968 and 1969 suggested that the main factors, of roughly equal importance, were stealing by existing consumers who fixed wires to by-pass meters (particularly small repair shops in some of the central parts of the city, but also some large enter- prises) and stealing by people setting up household, and stringing a line 1/ Analysis by the Operations Evaluation Department in 1971 showed that tariffs were sufficient in 1964 to yield a 10% rate of return to aver- age net fixed assets in operation revalued to allow for inflation up to that year. - 11 - to the nearest EPM circuit, in the marginal barrios where EPM failed to provide distribution facilities, in part because it was a one-time policy of the Municipality of Medellin to try to discourage immigration and speculation in marginal lands by refusing to provide public services. As mentioned, EPM began major efforts in 1969 to deal with these problems and to replace private lines in marginal areas with properly installed equipment, but losses to stealing still amounted to as much as 5% of gross generation in 1972 (see Annex Table II). The financial impact of this stealing has been substantial, as illustrated in the table below, based on the assumption that stealing could have been held to the 1% level of 1960 and valuing energy stolen in excess of this amount at EPM's average sales price for each year. Table 3 Estimate of EPM's Revenue Loss due to Stealina of Electrical Ener2 Estimated Losses to Contrahnd Average above 1% EPM Sdles nf Cr C,n- Price Rvenue T.n-q (mln kwh) Colombian centavos mln. Pesos mln. $ equiv.a 1964 85 9.7 8.2 0.9 1965 135 9.9 13.4 1.3 1966 145 11.6 16.8 1.2 19A7 184 13 'Yb A 1 7 1968 262 16.2 42.4 2.6 10D 1 /. AQ 4 1 0 1970 204 18.4 37.5 2.0 1071 140 10 7 9O 7 1 4 /1 LaJ LUsi. L.a. l L. 1972 109 16.7 18.4 0.8 240.3 14.9 al/; ConJtveteda ati. averA.age- excnge rk6ates- JC.CLA prevaaigeac year. Tnspec"t--on of Annex Table TSI-OW tha Jbn EC tese amonLsI, orsoetin approximating them, had been collected EiF's electricity department would have met th1-e 40fl self-fnncng targets ~ as. r,-- W -1 fl'7LL, CVL-..L five-year period subsequent to December 31, 1963.1! According co the .L LvlLii LLIC LLLaVJLL=%L.L CoLOZ)LLtJii.UL LLLCL. CL.it bt.CCt1.IL6 WaZ5 U)Y SqUaLLe _L6 ILI marginal barrios whose relatively low consumption (monthly average of 260 kwh per family according to experiace under the program for con- nection of such barrios) would have enabled them to benefit from the lowest blocks of the residential tariff, this remains easily true; re- ve tue loss would have been somewhat more than 75% of the amounts shown in Table 3, which, if collected, would have raised self-financing to about 45% for the relevant five-year periods. - 12 - report of a Bank supervision mission in October 1964 (the single such report filed on this project prior to 1969) EPM at that time expected to reduce thefts substantially from the approximately 7% level they had then attained, but it appears that in fact nothing was done before late 1968; Bank letters between 1964 and 1969 relating to financial perfor- mance referred only to the need for early action to raise tariffs. To the extent that the gap in financing plans resulting from EPM's poor financial performance was filled (as opposed to being closed by cutting down on investment, particularly in distribution), this was done with credits from local banks and from Government sources. Since EPM felt that a further tariff increase could not be introduced in the near future following the 60% increase applied in April 1963, the Bank's appraisal mission gave considerable attention to helping EPM secure local loans to contribute to the project's Peso requirements. The U.S. A.I.D. agreed to contribute out of counterpart funds Ps. 20 million, which were eventually provided in 1964 and 1965 in the form of twelve-year loans at 10% interest from the Government-owned Caja Agraria and Instituto de Fomento Industrial (IFI). The Corporacion Financiera Nacional (CFN) agreed to provide a line of credit, up to $2 million for two years, on which EM would have to take the exchange risk; about half of these funds appear to have been used, and CN was in tact virtually the only Sider to EPM, other than the Bank, in the years 1966-68 when the Peso shortage seems to have been most acute. In 1969 EPM managed to secure a further Ps. 20 million from IFI out of U.S. A.I.D. counterpart funds on equally favorable terms with the previous loan. Smaller borrowings have been made, on 4-10 year terms, from a greater variety of local in- stitutions in subsequent years. However the total of local borrowings between 1964 and 1972 amount to only about 40% of the revenues estimated to have been lost on stolen energy over the same period. Institutional Development The Bank had given a good deal of attention to the constitutional and organizational structure of EPM in connection with the first two loans, a number of its suggestions had been implemented, and in 1963 it found the company's technical and financial management good. It set no further institution-building objectives in connection with this loan, but it did require that EPM should "(a) consult with the Bank before EPM's Junta Directiva introduces any modification in the statutes of EPM and (b) in- form the Bank of any proposed amendments to national legislation and national and municipal decrees and resolutions affecting the statutes of EPM or its status and operations." EPM appears to have fulfilled these obligations, drawing the Bank's attention early to certain changes in municipal decrees affecting the company that were ,rought up in the Municipal Council in 1970 and keeping the Bank informed of subsequent developments. A number of changes were eventually approved, including transfer of responsibility f6r elacting the 6 voting Board members of EPM other than the Mayor of the city entirely to the Municipal Council (previously the Mayor had chosen 4 out of the 6) and addition of the Chief of the Municipal Planning Department and of a delegate oJ the Governor of Antioquia to the Board as non-voting members. Legal disputes over these changes prevented effective operation of any Board for much of 1971, but the acting General Manager continued to direct affairs, in collaboration with the Mayor who remained ex officio Chairman of the Board. The changes mentioned were eventually fully implemented, and a Board constituted along the new lines has been operating errectively, selecting a new General Manager, with his team of departmental managers, in April 1973. The Board is now more representative of different inter- est groups in the city of Medellin itself, although, since the company is essentially a municipal subsidiary reinvesting all profits but paying a small turnover tax (averaging about 7% of net profits, but a little higher for the electricity department) to the municipality, it has no representation of the growing areas outside the city of Medellin where EPM operates in the electricity field. Conclusions Loan 369-CO was timely and the project for which it was given proves to have been economically worthwhile. Except for parts of the distribution component, the project was well executed by EPM, its con- sultants and contractors given the difficulties of nature with which they had to contend. While its financing was crucial due to the inability of the Colombian financial system to generate funds for a project of this magnitude and delayed return, the Bank gave little additional positive help toward Proiect execution, but it maintained a generous and accom- modating attitude. It may be questioned whether on some occasions the Bank may not have been excessively generous with this essenilally strong company, Zor instance in approving substantial use of loan funds in 1971 and 1972 against interest during construction even though con- struction was essentially completed and revenues being earned, and in givine the comnany the benefit of a 35-year amortization term decided on national economic grounds but, in retrospect, perhaps not fully justi- fiable on that basis in view of the sharp growth in non-traditional exports that Colombia has proved capable of and the consequent decline of the national debt-servicing ratio to some 12-14% in recent years - although it may, as was hoped at the time, have contributed to Colombia's ability desoite the financial crisis of the late 1950s, to raise the sub- stantial amounts that it did borrow in subsequent years from countries that had not nreviously been lending to Colombia. A ,arder attitude on the part of the Bank toward EPM could have helped strengthen EPM's financial discinline- even though the introduction in subsequent loan agreements of a target minimum rate of return to be earned on revalued assets may redune the direct influence of the easy terms of Loan 369-CO on financial performance. As regards the large contingencies that the apnrncnl micQinn hilt into the nrnlet cost estimates- in light of earlier experience in the power sector in Colombia, it appears that they were esSenti.al on t-he co'ltost sidend n-lroblyl wicp on thbP fnrPig<)n cost side, in view of the uncertainties involved in tunnelling; none- the fact that EPM paid over the years some $250,000 in commitment charges LJI. LLIC YU LLL UI WLL L t. £CCJ l.t, ... Lt -FF-JLL-L' - 5-aL t at L - 14 - 1/ amounts should have been cancelled earlier.- While the assistance pro- certainly helped the project to go ahead it is noteworthy that the appraisal L~a1U L4L £~. L.LLL I1'.l .L.L1 L %J ULU L . U-t, .L U L OA.-LLX_UUL_ Q £LIL1 0 postponement of the proposed tariff increase from July 1965 to January 17UU LOLLC availabiUity UL Lhe UINT CLULL LU LI[ L Ual CULLP-nCy -xpen- ditures. But the more serious shortcoming of the Bank was, once having Helped to arrange the necessary financing, not to pursue te matter of adequate investment in the high-return field of distribution. The Bank's preoccupation with the bigger and more important issue of inter-regional interconnection, the strength of EPM s own management and trust in its Colombian engineering consultants (who were not in fact concerned at all with distribution), together with the very excellence of the progress reports they sent to Washington, hardly justify but all may help to explain why the Bank devoted insufficient supervision effort to the project and largely neglected distribution both during appraisal and suosequently; no attention was ever given to the other services for which EPM was responsible despite the reference in the loan agreement; things went seriously astray with electricity distribution expansion and metering/billing control, which has delayed and somewhat limited the impact of what was otherwise a very successful project, and the Bank seems to have become substantially aware of these problems only in retrospect. 1/ For instance, the $4 million put into the second tranche in December 1967 proved more than twice as much as required, for equipment of which most prices were already known with fair precision since most of it was bought under options EPM obtained in bids for the first two turbine-generators. 讓’〕！;!:!”〕〕‘!’朧 !!!!！〕！ 〕．！〕〕〕〕〕〕〕 “〕〕〕：;！〕〕！〔〕！！〕！〕！！「 “〕〕！〕！〕〕！!!!！〕！：〕〕〕 ::：。！：〕〕：! &―〕〕；::!!.::！〕：:！〕〕〕」！ :-！〕！、〕〕〕．!：〕〕：!！〕〕〔 莖〕〕〔〕〔〕〔．〕〔〕〕、〕！‘〕！〔／ &．〕〕！〕〕；.-!::!.!：一〕〕 〕〕！!！〕！．〕〔〕！!！〕！!! Annex Table II E.preses Publicas de Medellin: Forecast and Actual Sales, Generation and Lose, Average Rate of Increase 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1963-70 1963-72 FORECAST Sales: Residential 406 400 425 52o 570 619 675 737 805 869 952 9.0% Industrial 177 185 212 239 265 295 328 364 405 450 500 11.1% Commercial 51 53 57 62 66 71 77 83 89 96 1.04 Street Lighting 33 34 32 33 35 37 39 41 43 45 47 Others 43 53 53 68 75 83 92 101 113 125 138 Total (Gwh) 710 725 779 922 1,011 1, 105 1,211 1,326 1, 55 1,585 1,741 9.5% Station Service & Losses (Gwh) 114 127 157 198 204 210 214 234 255 280 309 as % of gross generation (13.8) (14.9) (16.8) (17.7) (16.8) (16.0) (15.0) (15.0) (14.9) (15.0) (15.8) Gross Generation (Gwh) 824 852 936 1,120 1,215 1,315 1,425 1,560 1,710 1,865 1,950 8.2% Annual L,ad Factor (%) 63 65 54 56 56 56 56 56 56 56 56 System Peak Load (MW) 149.9 148.3 199.6 229 248 268 291 318 350 380 418 9.0% ACTUAL Sales: Residential 425 494 515 553 582 598 608 631 736 845 924 5.9% 7.2% Industrial 212 246 283 300 320 340 358 414 448 518 608 8.9% 10.6% Comercial 57 61 64 70 77 81 87 99 115 127 140 Street Lighting 40 39 40 41 42 43 44 40 35 37 40 Others 45 61 77 84 ill 121 122 147 162 158 520 of which Other towns 4 18 31 35 44 53 61 73 83 92 120 ISA - - - - - - - - - - 303 Total (Gwh) 710 725 779 901 979 1,048 1,132 1,183 1,219 1,331 1,496 1,685 2,232 7.5% 10.6% Station Service & Losses (Gwh) ill 122 149 196 254 322 344 409 473 492 457 442 414 as % of gross generation (13.5) (14.3) (15.9) (17.8) (20.6) (23.5) (23.3) (25.6) (27.9) (26.9) (23.3) (20.6) (15.5) in trans.ission/transformation 3.5 4.5 4.2 3.1 3.4 3.9 3.2 3.3 2.8 3.0 2.7 2.9 1.8 in primary distribution networks 4.1 3.7 3.5 4.1 4.7 4.2 4.7 5.3 4.1 3.8 4.4 4.3 3.9 in distribution transformers 1.8 1.8 1.9 1.8 1.8 1.7 1.7 1.6 1.6 1.5 1.5 1.4 1.2 in secondary distribution networks 1.3 1.1 1.1 1.1 1.1 1.2 1.3 1.3 1.4 1.3 1.7 1.8 2.0 in meters 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.1 by meter reading errors 1.6 1.6 1.6 1.6 1.5 1.5 1.5 1.4 1.4 1.4 1.5 1.5 1.4 by fraud & contraband 1.0 1.6 3.5 6.0 7.9 10.8 10.8 12.5 16.4 15.7 11.4 8.4 5.1 Effect of billing lag (Gwh) 3 5 8 3 3 3 2 5 6 6 12 23 24 Gross Generation (Cwh sent out) 824 852 936 1,100 1,236 1,373 1,478 1,597 1,698 1,829 1,965 2,150 2,670 8.7% 10.4% Annual Load Factor (%) 63 65 54 58 61 59 58 59 59 60 61 58 60 System Peak Load 149.9 148.3 199.6 215.8 231.5 267.0 289.0 309.8 327.0 349.5 370.8 422.4 600.7 8% 12.0% LHypothetical Sales; fraud at 1% 710 730 802 956 1,064 1,183 1,277 1,367 1,480 1,600 1,700 1,844 2,3417/ 8.6% 10.5% Annex Table III mpresas Publicas de Msedellin: Groth of Ele,tricity Supply Capacitv, Distribution System, Customers and Sales 196)-72 % . % . % % % % % Rate of Growth 1960 increase 1961 increase 1962 inerease 1963 increase 1964 Lncrease 1965 inerease 1966 inrease 1967 increase 1968 increase 1969 increase 1970 increase 1971 intrease 1972 1963-68 1968-72 1963-72 Generatng Capailty Installed (MW) 137 137 227 227 248 308 443 443 443 443 443 583 723 Suhstaion Capacity (MVA) 309 300 410 516 616 636 702 839 839 900 Dstribution Transformers (MVA) 269 12.3% 302 11.3% 336 8.3% 364 7.1% 300 4.1% 406 1.07 419 4.9% 430 13.0a/ 626 8.6% 571 4.27 598 5.9% 633 5.1% 8.45// 6.5'a/ N't lncease in Year 33 34 28 26 16 4 20 96ý' 45 27 35 Gress Add,tion (Empresas) 46 42 43 43 31 14 30 39 39 Gress Addition (Developers) 14 22 17 12 18 12 16 42 21 Primary Feeders (ksv) 958 9.4% 1048 9.4% 7146 8.2% 1240 6.0% 1314 5.5% 1386 3.9% 1440 5.2% 1515 6.6% 1615 9.5% 1768 1895 5.8% 5.7% 5.8% Increae i Year 90 98 94 74 72 54 75 100 153 Secondary Lines b,ilt (kes) 70 98 112 97 74 80 170 Numer, el Customers ('1?s) 99.9 5.4% 105.3 5.1% 110.7 3.7% 114.8 6.4% 122.1 4.7% 127.8 4.3% 133.3 3.3% 137.7 4.3% 143.6 6.8% 153.4 13.8% 174.5 8.3% 189.0 6.9% 202.1 4.6% 8.9% 6.5% No. ef Residential Cestmees ('O0Os) 86.8 5.3% 91.4 5.4% 96.3 4.8% 100.9 7.2% 107.2 4.7% 112.2 4.6% 117.4 3.17. 121.0 4.2% 126.L 7.5% 135.6 14.6% 155.4 8.9% 169.2 6.9% 180.8 4.6% 9.4% 6.7% IecreaseinYear 4.6 4.6 4.9 4.6 6,3 5.0 5.2 3.6 5.1 9.5 19.8 13.8 11.6 Electification Program ('000s) .. -- .. .. .. .. 0.9 1.4 2.7 6.', 13.ý .3 6.0 Toel Sales (Gh) 710 2.1% 725 7.5% 779 15.7% 901 8.7% v79 7.1% 1.68 8.0% 1132 4.5% 1183 3.0% 1219 9.2% 1331 12.4% 1496 12.6% 1685 14.5% 19296/ 6.2% 12.2% 6.8% Residential Sales (Gwh) 414 e8. 6U7 6.4% 433 16.2% 503 4.2% 524 7.4% 563 5.2% 592 2.7% 608 1.5% 617 3.2% 637 16.5% 742 14.8% 852 9.3% 931 4.2% 10.8% 7.1% Av. Residenlal Sales (kwh p.a.) 4,771 4,458 4,500 4,987 4,892 5,013 5,044 5,022 4,891 4,694 4,775 5,037 5,151 Total lypothetical Sales (Goh) 710 2.6% 730 9.9% 802 19.2% 956 11.3% 1,064 11.2% 1,183 8.0% 1,277 7.1% 1,367 8.3% 1,480 8.1% 1,600 6.3% 1,700 8.5% 1,844 10.3% 2,038 9.1% 8.3% 8.0% Gros . Genor,tio (Ch) 824 3.4% 852 9.9% 936 17.5% 1,100 12.4% 1,236 11.1% 1,373 7.7% 1,478 8.1% 1,597 6.3% 1,698 7.7% 1,829 7.4% 1,965 9.4% 2,150 10.0% 2,367ý 9.1% 8.7% 8.9% Peak Demand (MW8) 150 - 148 35.1% 200 8.0% 216 7.4% 232 15.1% 267 8.2% 289 7.3% 310 5.5% 327 7.0% 350 6.0% 371 73.8% 422 6.9% 451 8.6% 8.4% 8.5% a/ Adjusted to allowlfr the fact that installed distrIbution transformer capacity nas reassesed in 1969, and at leest 40 MVA of the increse rteorded for ti en rpresente histatistical adjustment rather than installatio-n during the year. b/ Exlolding sales to 15A. Note: Figures refer to production or sales during the cotre of the yenr, r to capacity as of the end of the year, as appropriate. Annex Table IV Empresas Publicas de Medellin Electricity Department Summary Sources and Mplications of Funds 1961-72 (In millions of Pesos) Total Guatape Construction Total Total Total Total Period Total Total Total 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1961-65 1963-65 1966-66 1963-68 fro 1964 1964-69 1965-69 1966-70 F 0 RECAST Internal Cash Generation: Gross Internal Cash 28.90 27.80 57.98 74.84 94.24 117.35 128.40 137.47 283.76 227.06 383.22 610.28 552.30 552.30 Less: Debt Service 6.15 7.35 27.51 13.93 30.43 45.62 30.32 45.18 87.37 73.87 121.22 194.99 167.48 167.48 22.75 20.45 30.47 58.91 63.81 71.73 98.08 92.29 196.39 153.19 262.10 415.29 384,82 384.82 Reductions in Working Capital 8.09 - - - - 10.53 - - 8.09 - 10.53 10.53 10.53 10.53 Borrowings (more than 1 year): Foreign 28.14 96.65 66.30 107.28 120.33 114.66 133.65 84.69 418.70 293.91 333.00 626.91 560.61 560.61 Domestic - - - 20.00 15.00 - - - 35.00 35.00 - 35.00 35.00 35.00 28.14 96.65 66.30 127.28 135.33 114.66 133.65 84.69 453.70 328.91 333.00 661.91 595.61 595.61 Total Sources 59.99 117.10 96.77 196.19 199.14 196.92 231.73 176.99 638.18 482.10 605.63 1908.73 990,96 990.96 Construction Expenditures 58.98 105.05 91.65 179.81 187.72 186.92 199.13 144.79 623.21 459.18 530.84 990.02 898.37 898.37 Investment in Interconnection - - - 1.00 6.00 10.00 7.00 - 7.00 7.00 17.00 25.00 24.00 24.00 58.98 105.05 91.65 180.81 193.72 196.92 206.13 144.79 630.21 466.18 547.84 1,014.02 922.37 922.37 Increases in Working Capital - 12.05 5.12 5.38 5.42 - 25.60 32.19 27.97 15.92 57.79 73.71 68.59 68.59 Total Applications 58.98 117.10 96.77 186.19 199.14 196.92 231.73 176.98 658.18 482.10 605.63 1,08.73 990.96 990.96 Construction Financed Internally 38.6% 19.5% 33.2% 32.6% 32.9% 36.4% 47.6% 63.7% 31.2% 32.9% 47.8% 41.0% 41.7% 41.7% ACTUAL Internal Cash Generation: Gross Internal Cash 28.90 27.80 50.93 66.25 68.96 87.50 105.60 134.11 177.33 200.48 222.38 261.72 242.84 186.14 327.21 513.35 1,324.33 462.42 573.50 705.02 Less: Debt Service 6.15 7.37 14.72 21.01 14.84 22.03 50.77 52m23 7 M3.30 69.99 50. 124.28 64.09 30.57 125.03 173.60 508.95 l6M8 213.17 268.32 22.75 20.43 36.21 45.24 54.12 65.47 54.83 81.88 104.03 130.49 141.88 137.44 178.75 135.57 202.18 337.75 815.38 30154 360.33 -436.70 Reductions in Working Capital 8.09 5.12 - 22.16 1.37 7.83 11.69 - 13.49 - - 59.96 36.74 23.53 19.52 43.05 116.50 43.05 34.38 33.01 Borrowings (more than 1 year): Foreign 28.14 81.37 38.96 72.82 67.86 96.93 331.15 191.40 165.94 182.83 171.32 55.95 289.15 179.64 619.48 799.12 1,336.20 760.16 853.28, 968.25 Domestic - - - 4.34 26.37 3.99 3.95 10.12 28.61 . -5.26 10.79 9.18 30.71 30.71 18.06 48.77 102.61 48.77 73.04 51.93 28.14 81.37 38.96 77.16 94.23 100.91 335.10 201.52 194.55 188.09, 182.11 65.13 319.86 210.35 637.53 847.88 1,438.80 808:92 926.31 Ih020.17 Total Sources 08.98 106.92 75.17 144.36 149.72 174.21 401.62 283.40 320 318.58 323.99 062.53 535.35 369.45 859.23 1, 8 2,70.6 1,155 321.02 1,499.89 Construction Expenditures 58.98 106.92 61.18 144.56 149.72 174.21 166.30 198.19 343.03 267.15 188.30 162.38 521.36 355.46 538.70 894.16 1,793.84 832.98 1,031.45 1 148.88 Exchange Adjustments (For. Debt) - - - - - - 235.32 59.38 30.96 36.61 52.27- 100.15 - - 294.70 294.70 514.69 294.70 325.66 362,27 Increases in Working Capital - - 13.99 - - - - 25.83 - 14,89 83.42 - 13.99 13.99 25.83 39.82 124.07 25.83 :25.83 40.65 lotal Applications 18.98 196.92 75.17 144.56 149.72 174.21 401.62 293.40 312.07 323.99 262.53 531.35 369.45 859.23 _228.6 2 _37002 1.483.88 Construction Financed Internally 38.6% 19.1% 59.2% 31.3% 36.2% 37.6% 33.0% 41.3% 30.3% 48.9% 75.4% 84.6% 34.3% 38.1% 37.5% 37.8% 45.5% 36.2% .34.9% 38.0% Note: All Actual Figures are based on Audited Accounts. COLOMBIA EMFKEAS PU Li.A S t MtUtLLIN TRONERAS General Layout of the Electric System 36eMW I(~TENCHED, loans DIVERSJON25C, 2C) * Hydro plants in operation R. ConcePcn D Hydro plant under construction Transmission lines - oAvERSI TUNNE Rivers 50 MW iZI~~~~~-Dnms~~~r t________ ______ÅnAl I.Oý fT dG Reservoirs MIRAFLORES DAM TRONERA ¡'' ¡W -----Tunnels I 1,AND RESýRVOIR DAM A[B R.D. loons RESERiRailroad RE 225 CO, 282 CO) Service areas of the Medellin Power Company . R.D projects5 o 5 lo I5 MILES o 5 1o I5 20 25 GJ u/ I I I KILOMETERS ~tre!~ros !~GRNDE Dn Motlos Sto Domingo 70 Barbosa Concepción, Alejandria BE-DAM :,,n, SanVicente GUATAPE 1 W 264MWN 7 - 3,.6 RO loon Guatope0 ( 9 On Rinonn11o -Granada 0 El Santuario ,.-~ - -. OEL LOrmen CiTOBER !971 IRRD-3542 R  COLOMBIA EMPRESAS PUBLICAS DE MEDELLIN - POWER LOAD AND CAPACITY DEVELOPMENT ACTUAL AND FORECAST (1955-1971) 600- - -L-T282- C o¯¯ ~----- -- - --- 00 60006 60 7 Fare~s C ~xciy LEGEND. ACTUAL CAPACITY ACTUAL EFFECTLVE CAPACITY (1963 - 69)1/ z ACTUAL LOAD .09 lod F0,o. Cs 500 FORECAST CAPACITY Lo 28? Co. FORECAST LOADS R-______n 1 uoi growth1 LOAN 223 Co. . LOAN 382 Co. . LOAN 369 Co. 44 1/ WHEN DIFFERENT FROM INSTALLED CAPACITY 3 Lad Forecast NOTE. DATA FOR FORECAST LOADS ARE END OF 418 Loo 369 Co. YEAR FIGURES. (9.3%oannua rowth 400 L____-- _____ _____ _____ __________ .ioadFore.cst ______400 398 L- - -- Lon050 7/% ann~1 - Loon 225 wth) 80 T 353 35 343 5 -. 349.3 94 02 32 3 27 317 ;30 -- 336 300 300 263 2 -. 289.0 -1 2 10 9 8 231 245 o369 Co. 20 27 200 200 182 147.5 149. 1 137 137.1 .... .... .. .*. .... .. 125 125.4 137137 Lon 382 Co. 3117.1 . Date of AgR,ement ate oF Agremen ¯ f Agreement (Loan 225 Co.) (Loan 282 Co.) (Loan 369 Co.) 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 (972 IBRD-5783(R)