Document of The World Bank FOR OFFICIAL USE ONLY FILE CO Report No. 2638 PROJECT PERFORMANCE AUDIT REPORT COLOMBIA: CALI WATER SUPPLY AND SEWERAGE PROJECT (LOAN 682-CO) August 22, 1979 Operations Evaluation Department 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. FOR OFFICIAL USE C PROJECT PERFORMANCE AUDIT REPORT COLOMBIA: CALI WATER SUPPLY AND SEWERAGE PROJECT (LOAN 682-CO) Table of Contents Page No Preface (i) Project Performance Audit Basic Data Sheet (ii) Bighlights (iv) Project Performance Audit Memorandum I. Project Summary 1 The Project 1 Project Costs 2 Sales, Revenues and Financial Performance 2 Incremental Financial Rate of Return 5 Institutional Matters 5 Bank's Role 6 II. Main Issues 8 Demand Projections 8 Form of Financial Covenants 9 III. Conclusions 11 Attachment PROJECT COMPLETION REPORT 1. Introduction 12 2. Project Preparation and Appraisal 14 3. Project Implementation, Operation and Cost 16 4. Operating Performance 21 5. Financial Performance 24 6. Institutional Performance 32 7. Project Justification 34 8. Bank Performance 34 9. Conclusions and Important Lessons to be Learned 36 Annexes 1 - Summary of Revisions to Project Scope - Project Execution 38 2 - Comparison of Projected and Actual Work Programs - Water Supply 40 3 - Review of Cost Estimate 42 4 - Appraisal Projections and Actual Statements of Income and Expenses (1970-1977) 46 5 - Appraisal Projects and Actual Statements of Cash Flow 1970-1977 47 6 - EMCALI Appraisal Projects and Actual Balance Sheet - 1977 49 7 - Incremental Rate of Return 50 LaR 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. (i) PROJECT PERFORMANCE AUDIT REPORT COLOMBIA: CALI WATER SUPPLY AND SEWERAGE PROJECT (LOAN 682-CO) PREFACE This report presents the results of a performance audit of the Cali Water Supply and Sewerage Project in Colombia (Loan 682-CO) for which a loan of US$18.5 million equivalent was made to Empresas Municipales de Cali (EMCALI) in 1970. The loan was signed on June 4, 1970, became effective on September 4, 1970, and closed on June 30, 1977, following disbursement in full. The Project Performance Audit Report consists of a Memorandum (PPAM) prepared by the Operations Evaluation Department (OED) and a Pro- ject Completion Report (PCR) prepared by the Water Supply and Sewerage Division, Latin America and Caribbean Projects Department (LCPWS). OED has reviewed the PCR against the Appraisal Report and other documents and discussed the project with Bank staff. On the basis of the above abbreviated review, the PCR was found to be both comprehensive and objective and OED agrees in general with its conclusions. The PPAM summarizes the concept and history of the project and discusses two specific matters in more detail -- the demand forecast (which turned out to be too high) and the rate covenant (which in retrospect might have been formulated differently). Following normal procedures, a draft copy of this Report was sent to the Government and Borrower for comments. However, none were received. PROJECT PERFORMANCE AUDIT BASIC DATA SHEET COLOMBIA: CALI WATER SUPPLY AND SEWERAGE PROJECT (LOAN 682-CO) KEY PROJECT DATA Appraisal Actual or Item Expectation Current Estimate Total Project Cost (US$.million) 33.45 40.80 /1 Overrun (%) -- 22 /2 Loan Amount (US$ million) 18.50 18.50 /3 Disbursed ) -- 18.50 /3 Cancelled ) Repaid to ) February 1979 -- 2.34 Outstanding to ) -- 16.16 /3 Date for Completion of Physical Components 12/73 6/78 Proportion Completed by Appraisal Target Date (%) -- 33 Proportion of Time Overrun (%) -- 126 Incremental Financial Rate of Return (%) 13.5 6.0 Financial Performance Satisfactory Worse Institutional Performance Satisfactory Worse Cumulative Estimated and Actual Disbursements CY (US$.'000) 1970 1971 1972 1973 1974 1975 1976 1977 (i) Appraisal Estimate 900 5,060 12,200 18,500 18,500 18,500 18,500 18,500 (ii) Actual - 800 3,100 4,700 8,500 15,400 17,700 18,500 (ii) as % of (i) 0 16 25 25 46 83 96 100 OTHER PROJECT DATA Original Actual or Item Plan Revisions Current Estimate First Mention in Files or Timetable -- -- 6/21/61 Government's Application -- -- 2/27/69 Negotiations n.a. 3/30/70 Board Approval n.a. 5/26/70 Loan Agreement Date 4/06/70 Effectiveness Date 9/04/70 9/04/70 Closing Date 6/30/74 31/12/75 6/30/77 Borrower ) Empresas Municipales de Cali (EMCALI) Executing Agency ) Fiscal Year of Borrower Calendar Year Follow-on Project Name Second Cali Water Supply & Sewerage Project Loan Number 1523-Co Amount (US$ million) 13.8 Loan Agreement Date 6/20/78 /1 Excluding interest during construction /2 Project scaled down 73 Plus exchange adjustment US$2.66 million - iii - PROJECT PERFORMANCE AUDIT BASIC DATA SHEET COLOMBIA: CALI WATER SUPPLY AND SEWERAGE PROJECT (LOAN 682-CO) MISSION DATA Month No. of No. of Date of Item Year Weeks Persons Manweeks Report Identification 4/66 n.a. 2 n.a. n.a. Preparation 2/69 n.a. 1 n.a. 4/69 Appraisal 8/69 -- 2 4 10/69 Total n.a n.a Supervision I 7/70 0.3 2 0.6 Aug. 1970 Supervision II 11/70 1.5 2 3.0 Dec. 1970 Supervision III 4/71 1.0 2 2.0 May 1971 Supervision IV 8/71 2.0 3 6.0 Sept.1971 Supervision V 4/72 1.0 4 4.0 May 1972 Supervision VI 8/72 0.6 2 1.2 Aug. 1972 Supervision VII 11/72 0.2 3 0.6 Nov. 1972 Supervision VIII 2/73 0.2 2 0.4 Mar. 1973 Supervision IX 6/73 0.3 3 0.9 June 1973 Supervision X 9/73 1.0 1 1.0 Oct. 1973 Supervision XI 3/74 0.5 1 0.5 Apr. 1974 Supervision XII 4/74 1.0 1 1.0 May 1974 Supervision XIII 9/74 1.0 2 2.0 Oct. 1974 Supervision XIV 2/75 0.5 2 1.0 Feb. 1975 Supervision XV 4/75 0.6 1 0.6 Apr. 1975 Supervision XVI 5/75 0.3 1 0.3 May 1975 Supervision XVII 9/75 0.6 3 1.8 Nov. 1975 Supervision XVIII 2/76 0.3 1 0.3 Feb. 1976 Supervision XIX 5/76 0.3 2 0.6 June 1976 Supervision XX 10/76 2.0 3 6.0 Nov. 1976 Supervision XX1 2/77 1.6 2 3.2 Apr. 1977 Completion 3/78 0.6 2 1.2 June 1978 TOTAL 15.8 35.0 COUNTRY EXCHANGE RATES Name of Currency (Abbreviation) Peso (Col$) Year: Appraisal Year Average Exchange Rate: US$1 = 17.33 Completion Year Average US$1 = 38.07 - iv - PROJECT PERFORMANCE AUDIT REPORT COLOMBIA: CALI WATER SUPPLY AND SEWERAGE PROJECT (LOAN 682-CO) HIGHLIGHTS Loan 682-CO was made to Empresas Municipales de Cali (EMCALI) for high priority works to improve water production and expand water, sewerage and drainage networks. The project suffered a time overrun of some four years, mainly because of a shortage of funds at a time of high inflation and institutional problems. Since the demand for water fell short of projections, however, the delay in building the treatment plant did not adversely affect EMCALI's ability to satisfy demand and in fact excess capacity now exists. Some drainage and sewerage components were dropped from the project (partly through lack of funds and partly because design changes were dictated by unforeseen trends in flow patterns) but conversely the program of house connections was expanded; because of unexpected developments in the direction of urban growth, the treatment plant location may no longer represent the least cost solution. There was a cost overrun of 84% in current prices (after adjustment for pro- ject changes) but a small cost underrun in real terms. The incremental financial rate of return is now calculated at about 6% compared with 13.5% at appraisal. Sales of water were seriously overestimated at appraisal 1/ and tariff increases lagged behind soaring operating costs, so that compliance with financial covenants proved increasingly difficult and the Borrower suffered serious cash problems. Institutional improvements were also, by and large, not achieved, due mainly to frequent management changes. The project has, however, met the objectives of overcoming supply deficits and bringing water and/or sewerage service to about 140,000 to 180,000 low income residents. The following points are of special interest: (i) Changes in project scope and their consequences (PPAM para. 2; PCR para. 3.02 and Annex 1); (ii) Reasons for cost and time overruns (PPAM paras. 3- 4; PCR paras. 3.03, 3.06-3.11 and Annex 3); (iii) Over-estimated sales projections (PPAM paras. 5 and 20-26; PCR paras. 4.01-4.03); 1/ Compare the discussion of the over-estimate of sales in the Project Per- formance Audit Report for the Colombia First Bogota Water Supply Project (Report No. 2003, March 29, 1978), which is relevant to the discussion of this same subject in the present report. (iv) EMCALI's financial problems (PPAM paras. 7-8; PCR paras. 5.01 - 5.03); (v) EMCALI's management problems (PPAM paras. 12-16; PCR para. 3.03); (vi) possible variation in the form of rate covenant (PPAM paras. 27-31); and (vii) excellent Bank response to lessons of experience, applied in subsequent project (PPAM para. 34). PROJECT PERFORMANCE AUDIT MEMORANDUM COLOMBIA: CALI WATER SUPPLY AND SEWERAGE PROJECT (LOAN 682-CO) I. PROJECT SUMMARY The Project 1. Loan 682-CO for US$18.5 million was made in May 1970 to Empresas Municipales de Cali (EMCALI), an autonomous municipal agency providing water supply, sewerage, storm drainage, electricity and telephone services in and around the third largest city in Colombia. A second loan (1523-CO) of US$13.8 million was made to EMCALI in March 1978. The project, to be constructed during 1970-1973, comprised those works in the Master Plan for water, sewerage and drainage considered to have highest priority - i.e., to increase water treatment capacity (since existing installations were expected to become inadequate in 1972); to construct main sewerage and drainage works to alleviate flooding and river pollution; and to connect many more premises to the water and sewer networks, mainly in low income areas. The project was estimated to cost Col$670 million (US$33 million) plus Col$82 million of interest during construction; the Bank loan was to finance 49% of the total (including US$7.5 million of local costs), the remainder being financed by government loans (24%), consumer contri- butions (9%) and internal cash generation of EMCALI (18%). In addition, private developers were expected to construct another Col$207 million worth of additions to the service networks. 2. There were a number of modifications to the project as executed. Some major components of the sewerage and drainage program were deleted in 1974, partly due to the shortage of counterpart funds to meet escala- ting costs (water supply had a higher priority), and partly because the observed growth of the sewage load and stormwater runoff had by this time rendered the design of these components inadequate. They were eventually redesigned and included in the second project. On the other hand, the program of house connections in low income areas was substan- tially expanded, to bring water and sewerage to an estimated 140,000 to 180,000 low income residents, and measures were instituted to improve the capacity of the existing treatment plants. These measures were initially prompted by slow progress on the new plant but then deferred the necessity for the new plant so that construction tended to lag still more. 3. The new treatment plant was finally completed in 1978 (about 4-1/2 years behind schedule) and the original loan closing date of June 1974 was twice extended (eventually to June 1977). The reasons for delay were many and inter-connected, one compounding another, but principally stemming from political interference and inadequate funding. In 1970, a new city administration set out to reassert municipal authority over the "empresas"; it replaced the General Manager, changed the constitu- tion of the Board, tried to transfer tariff policy decisions to the -2- Municipal Council, and challenged the legality of the consultants' contract (and indeed the legality of the entire Bank Loan Agreement). In 1972, the position was largely restored but the damage had been done; the consultants' dispute alone held the project up for six months and the staff turnover and deteriorating morale created further delays. Each delay, in a climate of growing inflation, increased costs. Con- versely, inadequate revenues and lagging government financial contri- butions (para. 8) exacerbated EMCALI's cash problems, so that it had to defer contracts, in turn causing still further construction delays and cost increases. There were also other causes of delay, such as material shortages, a minor structural failure at the treatment plant, and the rebidding of a major contract. In general, however, procure- ment went smoothly and there were few problems between EMCALI and its consultants. Project Costs 4. If actual costs had eventuated as forecast, the completed project would have cost Col$642 million compared with Col$670 million for the project as originally planned. By the end of 1973, prices in terms of Colombian pesos had reached 1.6 times their 1969 levels com- pared with only 1.4 times forecast in the Appraisal Report 1/; thus, completion on schedule would have meant a cost-overrun of some 15%. In the event, prices rose to 3.7 times the 1969 level before the pro- ject was completed, and the ultimate cost was Col$1,179 million -- an 84% overrun. Interest during construction also doubled. In US dollars, the cost overrun was less pronounced (27%), reflecting exchange rate differentials, and in real terms there was a small (1%) cost underrun, though some individual components cost substantially more than expected, even in real terms. Unfortunately, however, EMCALI's programmed re- sources in real terms fell far short of expectations. Sales, Revenues and Financial Performance 5. For reasons discussed more fully in paras. 20-24, water sales in the period 1970-1977 (by volume) proved to be well below the level projected at appraisal. By 1977 they were not much more than half the quantity anti- cipated for that year (82 million m3 as against 147 m3) and in fact had still not reached the level forecast for 1972. Total sales for the ears 1970-1977 were thus only 592 million m3, compared with 852 million m proj- ected at appraisal, a shortfall of 31%. 1/ The Appraisal Report forecast 8% annual inflation (local) to 1973, thereafter nil. Actual average annual rates of inflation were 13.1% to 1973 and 22.6% for the period 1973-77 (PCR para. 3.08). - 3 - 6. The adverse effect on EMCALI's finances of this overestimate of water consumption was aggravated by tariffs which gradually fell behind costs. Water supply and sewerage tariffs had been traditionally low in Colombia, though official policy had begun to encourage financial self-sufficiency coupled with some income re-distribution, and in 1968 (during project pre- paration) EMCALI's tariffs had been more than doubled, despite strong public resentment. Covenants in the Loan Agreement provided for a rate of return on revalued net fixed assets of 2.5% in 1970 rising to 8% by 1978; since cash balances for all services were pooled, and to avoid pos- sible cash flow problems in EMCALI's other services which might prejudice water supply finances, minimum rates of return were also specified for the electricity and telephone operations. In the event, these services helped to cushion water supply deficits. 7. With constant urging from the Bank, EMCALI raised tariffs from time to time, including automatic monthly increases during much of the project period, so that average rates in 1977 (PCR Annex 4) were 3.6 times the 1970 level - almost in step with general inflation. However, in the face of lagging sales growth, EMCALI was unable to maintain the unit cost of supply which in 1977 was 31% higher than forecast (Col$1.60/m3 compared with Col$1.22/m3) even using the price levels assumed in the Appraisal Report. Thus, while 1977 total revenues were 4.8 times the 1970 level, the corresponding factor for operating expenses was 6.3 times and for depreciation 8.4 times (PCR para. 5.04). Operating costs rose pre- dominantly in administrative expenses and certain production/distribution costs -- possibly aggravated by a generous labor contract awarded by the outgoing 1972 Administration 1/. Accordingly, EMCALI's return on fixed assets for water supply never approached the covenanted levels after 1971; telephone finances also deteriorated severely, though the return on electricity operations generally surpassed the levels agreed. 8. Aside from revenue shortfalls, EMCALI had other drains on its finances. At the outset, unforeseen investment costs had to be met for ongoing programs which were delayed and insufficiently controlled and for commitments to a public development authority for services in newly- paved streets. Government loans did not materialize and did not in fact reach the programmed level until 1976. Similarly, it appears that the planned collection of capital improvement levies was not implemented to any extent, although the Bank had carefully investigated EMCALI's right to impose these charges during negotiation (Memorandum of Under- standing, April 7, 1970) 2/. As a result, EMCALI has had to finance 1/ There was also a considerable overrun in "Non-operating Expenses," the exact nature of which is not clear. 2/ With regard to sewerage and drainage levies, EMCALI was still reporting as late as January 1976 that it was "near to putting this system into practice." its extra investment costs with the aid of temporary advances from the electricity and telephone undertakings (a practice which obliged those undertakings to seek outside loans and which is more strictly controlled under the second loan) and from suppliers, together with high-cost local bank loans which EMCALI is now having difficulty in funding. The down- turn in EMCALI's finances may clearly be inferred from the table below: EMCALI -- INVESTMENT FINANCING 1970-77 (excluding private developers' activities) Projected in Appraisal Report Actual Col$ million % Col$ million % Internal Cash Generation (less debt service and working capital) 125 12 59 3 Current creditors 9 1 442 24 IBRD Loan 371 36 519 28 Government and other local loans 195 19 644 35 Consumer contributions (excluding developers) 373 37 128 7 Reduction in Cash Balances 51 3 1,073 105 1,843 100 LESS Added to Cash Balances 54 5 TOTAL (Investment Program) 1,019 100 1,843 100 9. In retrospect, the tariff increases imposed by EMCALI were perhaps as much as could have been expected in a period of recession and high in- flation. Public resistance to tariff increases may have been strengthened by the published profits of EMCALI. Revalued depreciation did not appear in the books while certain capital receipts were treated as revenues, so that the accounts showed a much healthier picture than the one developed for rate of return calculations. For example, a net loss of Col$82 million on a "revalued" basis for the year 1977 became a profit of Col$43 million in the published accounts. Nevertheless Cali's tariffs are lower than Bogota's and EMCALI could, as the PCR points out (para. 9.04), have reached the 1977 rate of return target by increasing its tariffs by only a further 5% annually over the project period. 10. EMCALI could also certainly have done more to improve its cash flow by (a) imposing the improvement levies more consistently, - 5 - and (b) taking advantage of rising property values to reclassify resi- dential consumers in higher tariff categories based on those values 1/. Such reclassification would have enhanced revenues markedly without an overt tariff increase; however, the effect on individuals could be severe, and perhaps for this reason EMCALI persistently lagged in implementing reassessments. With hindsight, it would have been preferable to make the improvement levies and the reclassification of consumers the subject of loan covenants (though the appraisal team understandably did not foresee the full scope for reassessment). In the second project, both matters were indeed covered by loan conditions. 11. EMCALI's cash flow problem was one of overspending as well as under-earning. Bank missions, besides pressing for improved revenues, also (during later stages of the project) urged better control of operating costs and postponement of investments. Incremental Financial Rate of Return (IFRR) 12. The IFRR on the project is calculated to be around 6.0% compared with 13.5% forecast at appraisal. The reduction is attributable to the fact that sales were lower and operating costs (in real terms) were higher than anticipated in the appraisal calculation. Institutional Matters 13. EMCALI (with responsibility for several different services) is a complex entity needing good management systems. Its management performance as evaluated in some Bank reports gives a curious impression of stagnation. In the Appraisal Report for the first project (1970), the utility was des- cribed as "a fairly well managed organization" with some weaknesses in training, budgeting, management information, internal auditing and maintenance operations (weaknesses all addressed by formal loan covenants). In the PCR (para,. 6.02) EMCALI is described as "acceptably managed," but needing further improvements in budgeting, management information and coordination. The Staff Appraisal Report for the Second Project (1978) proposed appropriate covenants on these matters and on internal audit -- this time providing consulting assistance. 14. In contrast to the implication in the foregoing of little change during the intervening eight years, EMCALI's management efficiency seems in reality to have fluctuated considerably. There were times when super- vision missions found EMCALI's managerial team to be excellent and times when all the staff familiar with EMCALI's finances seemed to have disappeared. The reason clearly lies in the frequent changes of management suffered by EMCALI. The project was hardly under way when political changes appeared to 1/ There are 13 tariff categories. The lowest-rated consumer paid Col$8.80 a month in 1977 'for a minimum consumption of 26m3, while the highest paid Col$430.40 a month for only 20m3. All paid the same rate for consumption above the allowance. - 6 - threaten all the plans and policies on which the Bank loan was predicated. The General Manager and other key staff either resigned or were replaced and political interference did great damage to staff morale and motivation. Until this point, excellent progress had been made in training programs, budgeting and other administrative improvements. Thereafter, however, it was a slow uphill struggle to recover lost ground, complicated by further management changes -- there were six General Managers during the project period so that there was usually no time for them to learn the job and no sustained momentum in improving administrative practices. Many departmental managers also changed after each election. The fluctuating management performance continues to date -- supervision missions for Loan 1523-CO reported it to be "deteriorating" in July 1978 and "improving" three months later. 15. Nevertheless, there have been achievements. Billing and collec- tion have continued to be efficient. Service to consumers has improved (except for telephones where there is a waiting list) and recent tariff increases have aroused little criticism after skillful publicity emphasized this fact. Staff numbers for water and sewerage have been contained; the numbers per connection and per 1000 population served are less than in 1970 (PCR para. 6.04) despite understaffing at that time in the sewerage department, and the present levels are reasonable for a developing country 1/. EMCALI has benefitted from the experience of the project in learning how to implement and administer major capital works programs. Technical staff turnover appears to have been reduced and fringe benefits are favorable, though salaries are still lower than in the private sector. 16. On the other hand, unaccounted-for water continues to increase for reasons which are not clear 2/. Budgeting, cost control and information systems continue to be imperfect -- as recently as July 1978 there was little or no effective budgeting or expenditure control and no regular system of management or financial reporting. Such reports as existed were said to be seriously behind schedule and containing errors 3/. Though there appear to have been improvements in some fields since appraisal, it must be concluded that the institutional objectives of the project have not been fully achieved. Bank's Role 17. The Bank devoted intensive supervision to this project. Between mid-1970 and mid-1978, there were 21 missions. Excluding a gap of 18 months towards the end of the project (possibly caused by personnel constraints) these missions averaged about one every 16 weeks. In all, supervision 1/ 6.6 staff per 1000 connections and 93 staff per 1000 population served. 2/ Although in 1977 it was twice the level projected at appraisal, EMCALI believes that the appraisal projection (15%) was unrealistic (PCR para. 4.04). 3/ It is, however, clear that here a recent reorganization has caused some disruption. missions took up some 7-1/2 man-months of Bank staff. It is interesting to note that the composition of the "project team" changed ten times dur- ing the project period, so that EMCALI was not alone in having continuity problems. From the beginning, as soon as the serious nature of the over- estimate of sales (para. 5) became apparent,.the main thrust of super- vision was on finance. Though the Bank expressed concern from time to time with various tendencies which it saw as contrary to the spirit of its partnership with EMCALI, it was the failure to achieve the covenanted rate of return which was almost continuously highlighted; the Bank repeated- ly emphasized the need for increased tariffs. Supension of the loan was considered several times, but each time the Bank stopped short of actual suspension -- either because EMCALI's future projections were more favorable or to await more accurate information (the parameters for financial projections on this project were often blurred and capable of varying interpretations). In retrospect, the mere possibility of suspen- sion -- supplemented by the Bank's expressed doubts about the advisability of extending the closing date or of appraising a second project -- probably helped EMCALI's management to keep the utility's finances from deterio- rating even further. Actual suspension, on the other hand, although perhaps formally justified, might have been counter-productive. 18. There were certain specific matters in respect of which Bank missions (assisted by resident staff) played a particularly useful part: (a) at the time of the 1970-71 political upheaval, project staff appeared before the new Board of EMCALI and persuaded that body (i) to accept the Loan Agreement; (ii) to con- firm the disputed 'consultant's contract; (iii) to accept the tariff program; (iv) to implement a telephone rate increase; and (v) to leave future rate-fixing decisions to EMCALI; (b) they drew EMCALI's attention to a looming problem which the utility had apparently not perceived, namely the need to ascertain the actuarial cost of a generous staff pension scheme; and (c) regular representations were made at government level to assist EMCALI, e.g. to hasten the flow of government funds and to speed import licences. 19. In general, the Bank's participation appears to have helped EMCALI to recover from the deleterious decline in its efficiency in the period 1970-72 and to secure such improvements as have been achieved in its institutional performance. On the other hand, the Bank may have been at fault in the matter of the overestimate of sales (see para. 24). - 8 - II. MAIN ISSUES Demand Projections 1/ 20. The project highlights the importance of realistic projections of demand for public utility services, since these govern the scope and timing of the project. As early as December 1970, the Bank found that its appraisal projections of sales during the project construction period were overestimated - by about 10% for 1970, and it was hoped, lesser proportions in subsequent years. By 1974, it was expected that sales would have caught up with projections. In the event, they never did (para. 5). 21. The overestimate stemmed from errors in the base estimates of the city's population, the number of connections and the water consumption per head 2/. Population projections by a local consultant, using historic growth rates, had estimated Cali's population in 1970 at 980,000, rising (at a lower growth rate) to 1.16 million in 1973. It was later ascertained that birth and migration rates had fallen well below the assumed rates, so that the true population in 1970 was probably only 810,000. It did not reach 1.16 million until about 1977; thus, the overestimate was of the order of 20-30%. Additionally, the number of connections was overstated at apprai- sal. EMCALI had reported having some 6,000 standpipes, whereas there were in reality only about 100 serving an estimated 6,000 families. This error affected the engineering consultants' analysis of consumption patterns. Finally, water losses in the distribution system appeared also to have been underestimated. 22. The combination of these errors would distort the picture of population served and per capita consumption and may have implied a sup- pressed demand greater than really existed -- though per capita consump- tion did in fact exceed the targeted level in 1972. It fell dramatically thereafter, however (instead of rising, as foreseen) and of course at all times was subject to a much smaller multiplier (population served) than estimated, so that total sales were well below projections. 23. The decline in per capita consumption can be attributed to three main causes: (i) most new connections were in low income areas, where consumption would understandably be below the average; (ii) tariffs were restructured so as to reduce the "fixed allowance" and consequently increase the incentive to economize (since a greater proportion of water consumed was now charged by quantity); and (iii) there was a downturn in industrial growth. Of these, the first is almost certainly the most significant. 1/ The following discussion of the overestimate of demand and its impli- cations for the project has several interesting features which are analogous'to those identified in the Project Performance Audit Report for the Colombia First Bogota Water Supply Project (Report No. 2003, March 29, 1978). 2/ Future population growth was also overestimated. - 9 - 24. The errors in the basic data could not, perhaps, have been detected though it is curious that the 60-fold error in the number of standpipes was not noticed, e.g., by visual evidence. The fall in per- capita consumption, however, could have been foreseen, especially as the distribution works planned were intended expressly for low income areas and in retrospect the Bank was probably at fault in this respect. The tariff restructuring, too, was foreseen at appraisal but not (apparently) its effect. In the second project, demand projections appear to have been much more conservatively based. 25. Errors in demand projections meant that the major project component -- the treatment plant -- was not required at the time anti- cipated; in fact, on its final completion, it was still not immediately needed even though four years behind schedule. The Staff Appraisal Report for the Second Project commented that "considerable excess production capacity now exists." Moreover, there are indications that, because the physical direction of the city's growth has been differ- ent from that foreseen, the plant may be located at a point no longer offering the least-cost solution (PCR para. 9.07). 26. These developments may give rise to a question whether the Bank should not have pressed at an early date for a reassessment of the need for the treatment plant. The treatment plant was, however, already under way in 1973 and, in the previous year, per capita consumption had reached its peak (269 lcd). Revised projections made in mid-1973 in fact indicated water shortages within six months and accordingly measures were taken to improve the existing supply. As it proved, these measures were sufficient for longer than expected. Form of Financial Covenants 27. The Loan Agreement called for tariffs to ensure a rate of return on revalued net fixed assets of not less than 2.5% in 1970 for water supply and sewerage, rising to not less than 8.0% in 1978 and after. The utility did not comply with this covenant after 1971; from 1972-77 its actual rate of return only once exceeded 1%. EMCALI claimed these goals were unattainable, blaming not only lower-than-estimated sales, but also what it felt had proved to be an inappropriate revaluation index for fixed assets (and depreciation expense), namely the local Cost of Living Index, which it felt was heavily weighted for extraneous items like food. Against this, Bank staff pointed out that an index based on construction costs would not have differed much and may well have provided even higher multipliers in recent years. In any event, the same index was proposed for the Second Project and again accepted by EMCALI. 28. The PCR (para. 9.04) makes the point that the 1977 target could have been reached by a cumulative additional tariff increase of only 5% - 10 - per annum over the project period 1/. Nevertheless, it is noteworthy that, for the second project, the rate of return requirement was dropped from 8% to 3%, and the requirement related to less than half the fixed assets (drainage works and excess treatment plant capacity being excluded). It was considered that the former rates of return were now "no longer reasonable" and would generate excess cash. 29. No financial covenant will achieve its objectives if it is un- realistic, or if the Borrower ceases to be committed to it, feeling it has become 'unattainable. In this case, not only did the Borrower perceive it as unattainable for the reasons mentioned above, but there are also indications that the covenant was regarded in Colombia as rather academic -- the Borrower's own accounts showed a much more favorable position, since they were kept on a different basis (para. 9) and perhaps gave him an unwarranted sense of security. 30. EMCALI's problem has always been one of cash and the revised covenant for the second project (para. 28) seems to have been based on cash requirements rather than on a notionally "correct" rate of return. However, latest projections (November 1978) show serious cash shortfalls in 1981 and 1982, despite achieving the required rate of return. The question, therefore, arises as to whether a simple cash generation covenant would have been more appropriate for both projects. Such a covenant has its drawbacks, but at least the Borrower might have identified with it more thoroughly. Furthermore, it might have avoided difficulties encountered in these projects concerning: (i) the appropriate index for revalued assets and their exclusion from the accounts (as mentioned above); (ii) cash sources (improvement levies) not reflected in the formula but whose absence proved important; (iii) late transfers of completed works to fixed assets; and (iv) uncertainties over the classification of "operating expenses." 31. Admittedly, the question as to whether or not a variation in the form of rate covenant would have been more appropriate is conjectural. Given the public resistance to tariff increases, the frequent changes in management and the delays in completing the project, it could be argued that a simple cash generation covenant would not have produced better results, since tariff increases would in any event have had to be im- posed to meet whatever target might have been agreed upon. However, in 1/ This estimate appears correct, provided that the substantial charges in the accounts for "non-operating expenses" (Col$45 million in 1977) do not in fact include any expenses which should be charged to opera- tions and hence affect the rate of return. The auditors have been recently asked to analyse this item. - 11 - the opinion of the audit, the factor of Borrower identification is impor- tant and should be given due weight when establishing rate covenants. III. CONCLUSIONS 32. The project appears to have been appropriately conceived on the basis of available knowledge at the time. However, the assumptions were not borne out by events - viz., the fall in per capita water demand and the unforeseen growth in waste and storm water loadings, combined with funding shortages which led to the excision of two components. The project was competently installed (though with cost and time overruns) and the equipment functions satisfactorily. Above all, it has brought water and sewerage to 140,000 - 180,000 low income residents whose sanitary environment had been unsatisfactory. Financially and institutionally, however, the results have been disappointing; the lack of progress stems from the political upheaval early in the project which could not have been foreseen. 33. The Bank's input appears to have been prompt, persistent and appropriate. The financial and administrative shortcomings which came to light did not arise for want of warnings, representations and advice from the Bank, and there are indications that the Bank's actions did much to minimize and correct the adverse tendencies which threatened the project. 34. Lessons learned from this project were applied in the second project. Notably, such lessons related to more realistic demand projections, imposing conditions about improvement levies and tariff reassessments, the provision of consulting assistance in administrative reforms, and limits on inter-utility cash transfers. As for the future, a critical factor would seem to be, the improvement of EMCALI's implementing and operating efficiency by obtaining a firm commitment from government and the entity for maintaining strong management. Attachment - 12 - COLOMBIA - CALI I WATER SUPPLY AND SEWERAGE PROJECT (LOAN 682-CO) PROJECT COMPLETION REPORT I. Introduction The Borrower and the Project 1.01 The borrower under Loan 682-CO is Empresas Municipales de Cali (EMCALI) which is an autonomous city government agency responsible for the provision of water supply, sewerage and stormwater drainage, electric power and telephone services to the city of Cali and some surrounding municipali- ties. The loan is for US$18.5 million and was approved on May 26, 1970 and signed on June 4, 1970. It was the first loan to EMCALI, and was to finance 49% of the cost of a new water treatment plant, distribution mains, new service connections, and sewerage and drainage facilities. In March 1978 a US$13.8 million loan for the Second Cali Water Supply and Sewerage Project (Loan 1523-CO) was aproved which will finance water storage, trans- mission and distribution, reforestation, sewerage laterals, collectors and an interceptor, stormwater drainage networks and a canal. Information in this report has been obtained from EMCALI, from supervision reports and from staff associated with the project. Population 1.02 The population of Colombia at the end of 1977 was estimated to be 25.2 million. Its recent growth rate has been 2.8% per annum, which is down significantly from the rate of 3.2% per annum which prevailed in the early 1960's. About 70% of the population lives in what the Government defines on an administrative basis as urban centers, where the population growth rate is 4.6% per annum, down from 6.0% per annum in the early 1960's. Colombia has several large urban centers, including the capital city of Bogota with about 3.5 million population, Medellin 1.5 million, Cali 1.1 million and Barranquilla 0.7 million. There are twelve more cities with 0.1 to 0.3 million inhabitants. Present indications are that migration to the cities and urban growth rates will continue at existing levels. Sector Organization 1.03 Until the end of 1975, Instituto Nacional de Fomento Municipal (INSFOPAL) had direct responsibility for planning, financing, constructing, and operating water and sewer system in municipalities over 2,500 population, excluding 21 large municipal empresas. Under decrees issued in December 1975 and June 1976 designed to decentralize the sector, INSF0PAL assumed responsi- bility for planning, coordination, financing, and general supervision for all communities over 2,500 in population. Regional empresas (EMPOS) and municipal empresas were given executing responsibility for design, construction and - 13 - operation of water supply and sewerage systems under INSFOPAL's guidance. Municipal empresas may become independent from INSFOPAL when they demonstrate technical and financial viability, and at present the empresas in Bogota, Cali, Medellin and Palmira are autonomous. At the present time INSFOPAL is in the process of restructuring its organization and redefining its relationships and responsibilities. Because of inadequate manpower in the EMPOS and municipal empresas, INSFOPAL has found it difficult to delegate project preparation and construction activities, and at the present time the decentralization envisioned in the decrees is not working as well as had been anticipated. Sector development and supervision for communities of less than 2,500 inhabitants is the responsibility of Instituto Nacional de Salud (INAS), which is expanding rural systems rapidly to improve service levels. Sector Development 1.04 The availability of water supply and sewerage facilities in Colombia ranks among the highest in Latin America. At the present time, about 76% of the urban population (served by four independent municipal empresas and those under INSFOPAL) has access to piped water and about 62% is connected to sewerage systems. Service levels in the rural areas (served by INAS) are rela- tively low, with about 18% of the population having access to piped water and about 5% having access to sewerage systems. For the most part, the urban population which is not served with piped water and sewerage systems consti- tutes the urban poor. Sewerage and industrial wastes are largely untreated and the rivers on which the major cities are located are becoming increasingly polluted. Sector Finances 1.05 Investment in the water supply and sewerage sector aggregated US$392 million equivalent during the period 1965-75, of which an estimated 45% was financed from internally-generated funds and domestic borrowings, 35% from the national budget and 20% from external sources. The Government's stated financial objectives for the sector are to have each empresa set tariffs at levels which will recover all operating and financial costs. At the same time, however, tariff structures must subsidize low income consumers and provide for income redistribution. For many empresas, the financial viability objective still must be attained, and tariff increases are often a pre- requisite to assistance from the national treasury. External assistance to the sector commenced in 1961, and through 1975, US$232.4 million equivalent had been committed. Of this amount, 68.2% was from the Bank, 24% from IDB and the remainder from USAID, the U.S. Eximbank and KfW. Bank lending to the sector commenced in 1968 and now includes 7 loans totalling US$172.4 million for water supply and sewerage in Bogota, Cali and 16 medium-size cities. Sector Objectives 1.06 The Government's objectives for the decade of the 1970's were to provide water and sewerage facilities to 80% of the urban population by 1980. The water supply objectives may be met, but the sewerage objectives will not. A major constraint has been lack of financial resources, but -14 - there have also been managerial and technical staffing constraints which have retarded sector development, particularly in the small-size cities. 1.07 Longer range, the Government plans to provide piped water to 90% of the urban population and 60% of the rural non-dispersed population by 1990. Sewerage service level objectives are 80% and 50% respectively. The Government has explicitly recognized the need to expand sector service to the poor, and it is expected that this will be accomplished along with social prices for water and sewerage services. Sector objectives for 1990 are attainable, provided sufficient external and internal financial resources are available. II. Project Preparation and Appraisal Origin 2.01 In 1968 EMCALI determined that the production limit of its existing treatment capacity would soon be reached and enlisted the services of consultants, Gonzales and Llano (two Colombian firms) and Hazen and Sawyer (United States) to prepare a feasibility study for a new treatment plant. Initial discussions were held with the World Bank about financing for the treatment plant, and the Bank convinced EMCALI of the need for a compre- hensive long-range development plan. A master plan for the provision of the water and sewerage services to Cali through the year 2000 was prepared by EMCALI and submitted to the National Government and the Bank in 1969. The project proposed for Bank financing included the water treatment plant, booster pumps, transmission mains and storage. Through the initiative of the Bank, distribution networks for lower-income sectors of the city were added together with sewerage and drainage components. Preparation 2.02 Feasibility studies for the treatment plant carried out by the consultants considered alternative water supply sources and plant locations and determined that the proposed Puerto Mallarino Project was the least- cost solution. Also, the consultants studied the feasibility of various alternatives for the sewerage components before adopting the selected alternatives. The drainage works were studied by Buck, Seifert and Jost (United States). Final design for all components was undertaken by Hernando Gonzales Hurtado (para 2.05) and Hazen and Sawyer. It was determined that EMCALI had the managerial capability to implement the project and that the financing plan was feasible. Appraisal 2.03 Appraisal took place in September 1969 after preparation of the master plan and definition of the project. Negotiations were completed in April 1970 and the loan was declared effective on September 4, 1970. There were no unusual delays between project identification and loan effectiveness. - 15 - Negotiations 2.04 The key issues during loan negotiations were the need for institu- tional improvements, the financing plan for the project, the need for EMCALI to increase water and sewerage tariffs to improve the Water Supply and Sewerage Division's financial position, and the establishment of minimum rates of return for the water, power and telecommunications divisions. Consultants 2.05 Consultants for final design and construction supervision were required by the Bank. EMCALI selected Hernando Gonzales Hurtado (one of the local firms that participated in the feasibility studies) with the proviso that Hazen and Sawyer would provide the required foreign consulting services. The Bank assisted EMCALI in drafting the terms.of reference for the consult- ants. Project's Role in the Long-term Plan 2.06 EMCALI's Master Plan for water supply and sewerage covered general investment requirements through the year 2000, and contained a detailed invest- ment program with cost estimates for the period 1970-82. The Bank-financed project consisted primarily of facilities considered to be of highest priority, such as water treatment capacity and pipelines and distribution networks, sewerage networks, mains and interceptors, and drainage canals which had been scheduled for construction in the 1970-73 period. Project Description 2.07 The water supply component of the project included a new intakS on the Cauca River, a water treatment plant with an initial capacity of 4 m /sec expandable to 12 m /sec, two booster pumping stations, 19.5 km of water trans- mission mains from 30" to 60", two storage tanks, about 13.1 km of distri- bution mains from 12" to 30", distribution networks in four unserved low-income areas, and house connections. The sewerage component included a major interceptor, 1.60 m in diameter in its downstream stretch, a pumping station, a pressure line, trunks and secondary sewers, and house connections. The storm drainage component included several drainage canals, of which Napoles Canal was the largest. Financial Covenants 2.08 The project provided for revaluation of fixed assets and for subse- quent annual adjustments. Water tariffs were to be restructured for simplifi- cation and to improve revenues, and rate increases were agreed upon to improve EMCALI's financial position. Specifically, EMCALI agreed to maintain tariffs for water and sewerage services at levels sufficient to produce gradually increasing rates of return on revalued net fixed assets with the objective of 8% by 1978. - lb - Institutional Covenants 2.09 The project envisioned a number of institutional improvements among which the following were covenanted. EMCALI would increase the number and quality of its technical staff and train 440 skilled workers in 1971-72. The budget system would be improved to assure proper control over expenses, and the accounting system would be modified to support budgetary control and improve capital investment records and financial reports. The internal audit function in EMCALI would be strengthened. On the operations side, EMCALI would improve its meter repair and maintenance, its maintenance records for the distribution system and would initiate a leak detection program. Finally, a Master Plan Department was to be created to program and control the expan- sion program and to act as liaison between EMCALI, the Bank, consultants and contractors. III. Project Implementation, Operation and Cost Project Inception 3.01 There were no special conditions of effectiveness and the loan became effective on September 4, 1970 as scheduled. Much of the design work was completed or underway at the time of loan effectiveness, and at that time the work program was on schedule. Project Revisions 3.02 There were a number of revisions to the original scope of the project as described in Annex 1. Notable additions were the Canaveralejo Reservoir for runoff regulation, expanded works for laterals and house connections and the upgrading of the existing San Antonio and Cauca treatment plants to meet demand prior to the delayed completion of the Puerto Mallarino treatment plant. The San-Antonio Plant upgrading was financed by the Bank, and the Cauca Plant with EMCALI's own resources. In 1974, the Bank agreed to EMCALI's proposed deletion of the Aguablanca sewerage interceptor, which was the largest subproject in the sewerage component, and to a large segment of the Napoles Canal which was the largest subproject in the drainage compo- nent. The principal reasons for the deletions were the rapidly escalating costs of the project coupled with fund shortages, the addition of the Canaveralejo Reservoir, and the fact that completion of the water supply component with available funds was considered to be of higher priority. There was also some concern over the technical feasibility of these compo- nents. By 1974 it became clear that the urbanization pattern and extended water distribution and sewerage networks had increased sewerage wastes and made the design of the Aguablanca Interceptor inadequate. Also, actual rainfall and runoff proved to be greater than the Napoles Canal had been designed for, and construction would have overloaded an already over-congested stormwater drainage system. EMCALI agreed to redesign these components and to start construction in 1976 with its own resources, but because of fund shortages was unable to do so. These components have now been redesigned and have been included in EMCALI's 1978-81 investment program which is being financed under the second Bank loan. - 17 - Implementation Schedule 3.03 The project was scheduled for completion by the end of 1973. In 1971 the EMCALI Board of Directors became politically dominated and EMCALI lost its traditional independence. The result was the deterioration in the quality of management, and lack of budgetary and financial controls. In spite of regular Bank requests to increase property assessment values and tariffs, the Board refused to take any action, and during 1972 EMCALI's financial position deteriorated steadily. In August 1972, EMCALI was informed that the Bank was considering suspension of the loan if tariffs were not increased in order to achieve a more satisfactory financial position. In October 1972, the National Government "intervened" EmCALI on grounds of bad administration and financial management, and replaced the Board of Directors with a more balanced membership. The new Board subsequently took action and recruited a qualified general manager and financial and admin- istrative managers, increased tariffs and prepared budgets and financial projections. While technical management of EMCALI was not seriously affected the overall management situation and financial constraints hampered project implementation, and at the end of 1973, the originally scheduled completion date, only about 33% of the work had been completed. A revised implementation schedule was prepared in mid-1973 which called for a completion date of December 1975. Subsequently, cost overruns coupled with insufficient tariff increases and slow National Government fund releases, technical difficulties (structural failure and erroneous soil mechanics studies), and structural steel and cement shortages contributed to further delays, and construction of the treatment plant was not actually completed until mid-1978. The other subprojects experienced lesser delays and were all completed prior to the end of 1977. See Annex 2 which compares the estimated and actual completion dates for the various components of the project. 3.04 The political interference in EMCALI could not have been predicted at the time of appraisal. In early 1971, the Bank recognized the deteriorating management and financial situation and from May 1971 to November 1972 mounted five supervision missions in an attempt to help correct the difficulties, but only after the new Board was appointed were the necessary corrective actions taken. Apart from these difficulties, the original construction schedule was tight and would have required efficient execution and timely funding to have been completed on schedule. Also, because of capacity additions to existing treatment plants coupled with slower than expected demand growth (see paras. 4.01-4.03), there was no urgency to complete the new treatment plant. Finally, in retrospect, the size of the project was perhaps too large and the time period too short in relation to EMCALI's executing and financial capacity. Procurement 3.05 EMCALI had no difficulties in complying with Bank procurement guidelines. There were some procurement delays which resulted from the Exterior Commerce Institute's system for approval of import licenses. - 18 - EMCALI considers that international bidding resulted in cost savings, provided useful technical information and aided in selection of the most appropriate technology. However, EMCALI did not group procurement contracts into large- size packages which increased administrative requirements for EMCALI and the Bank, and also created complications related to compatibility of equipment purchased from numerous foreign suppliers. Project Costs 3.06 Project cost at the time of appraisal was estimated to be Col$670.4 million (US$33.45 million equivalent) of which the foreign cost was Col$179.8 million (US$8.9 million equivalent). In addition there was estimated to be Col$80.2 million (US$4.00 million equivalent) of interest during construction on the Bank loan and the Government loan. The cost estimate included 25% for technical contingencies and 34.6% for price contingencies, which assumed 8% per annum increases in local costs and 3% per annum for foreign costs. The Bank loan of US$18.5 million was to cover the foreign exchange costs of the project including interest during construction on the Bank loan (US$2.1 million) and Col$150.3 million (US$7.5 million equivalent) of local currency expenditures, equivalent to 30.6% of local costs. For purposes of converting project costs into US dollars, it was assumed that the devaluation rate for the Colombian peso would be 5% per year. 3.07 As shown in Annex 3, page 1, which adjusted the appraisal cost estimates for subsequent changes in scope of the project financed by the Bank, the revised cost estimate for the project was Col$642.4 million, excluding interest during construction. The estimated final cost of the project amounted to Col$1,178.9 million at the end of 1977, representing a cost overrun of 84% as shown on the table below. Including interest during construction, the total cost of the project was Col$1,354.3 million, representing a cost overrun of 87%. Annex 3, page 2 provides a detailed breakdown of project costs by category and a comparison with the original cost estimate. - 19 - Total Cost of the Project (Col$ millions) Percent Increase Appraisal Revised Estimated vs. Description Estimate Estimate 1/ Final Cost 2/ Revised Estimate Water Supply 467.5 454.2 834.0 84% Sewerage/Drainage 202.9 188.2 345.0 83% Subtotal 670.4 642.4 1,178.9 84%. Interest during Construction 80.2 80.2 175.4 119% Total 750.6 722.6 1,354.3 87% 1/ Reflects change in scope of the project. In 1969 Colombian pesos and directly comparable with appraisal estimate (see Annex 3, page 1). 2/ Current Colombian pesos. 3.08 The cost overrun resulted from the combined effects of rapid cost escalation and delays in project execution. Local costs had been expected to increase by 8% per annum through 1973 when the project was originally expected to be completed, while actual cost escalation during this period was 13.1% per annum. Similarly, foreign costs had been forecasted to increase at a rate of 3% per annum, while in fact they increased at a rate of 8.3% per annum. Thus, even if the project had been completed as scheduled, there would have been a cost overrun of an estimated 15%. The more significant factor in the cost overrun was the four year delay in project completion, as during the period 1974-77, local cost inflation and foreign cost inflation were at average rates of 22.6% and 11.7% per annum respectively. When the project costs are deflated to 1969 values on which the cost estimate was originally made, the total actual project costs were about Col$504 million compared with Col$509 million estimated, representing a difference of about 1%. 3.09 In US dollars the adjusted cost of the project was US$32.0 million and the actual cost US$40.8 million, representing a cost overrun of 27.5% (see Annex 3, page 3). The much lower cost overruns in US dollars reflects the devaluation of the Colombian peso which offset a large part of the local inflation. With respect to the foreign exchange costs of the project, the appraisal estimate was US$8.0 million and the actual cost US$13.0 million, (both including indirect foreign exchange) representing a cost overrun of 62%. This overrun was accounted for by more rapid than expected inter- national inflation and physical contingencies (paras. 3.08 and 3.11). - 20 - 3.10 The cost overrun was not spread uniformly among the components of the project, and in fact most of the items experienced cost overruns below the average. The critical component was the treatment plant which was projected to cost Col$113.8 million (including contingencies) and actually cost Col$368.8 million, 224% above the estimate. The principal reasons were the extended period of construction for this component and the large technical contingency (para. 3.11) which produced a cost overrun in real terms of 46%. 3.11 The appraisal cost estimate included 25% of direct project costs for technical contingencies to cover physical contingencies and under estimates of original costs. On an overall basis about 94% of the technical contingency provision was utilized. However, in the case of the treatment plant the technical contingency required was 82%, while for the other water components none was required as the original cost estimate overstated actual costs (principally on pipes) by Col$27 million. The technical contingencies for the sewerage and drainage components were more than adequate (Annex 3, page 4). Disbursements 3.12 A comparison of the estimated and actual disbursement of Bank loan funds for the project is presented below, which highlights the delays in project implementation previously discussed. Cumulative Disbursement of Bank Loan Funds (US$ 000) Appraisal Year Estimate Actual 1970 900 - 1971 5,060 800 1972 12,220 3,100 1973 18,500 4,700 1974 - 8,500 1975 -- 15,400 1976 -- 17,700 1977 -- 18,500 Operations 3.13 The installed water supply and sewerage mains and network exten- sions are technically appropriate and functioning in accordance with design specifications. The major subproject, the treatment plant, has not yet been operating on a commercial basis (para. 4.06). The individual components of the treatment plant have been tested and conform to the design specifica- tions, and the plant is expected to meet all performance criteria. - 21 - Performance of Consultants, Contractors, Suppliers and Borrowers 3.14 The consultants were almost exclusively responsible for project feasibility and design (paras. 2.01 and 2.05) and performed acceptably in most areas. EMCALI observed a tendency on the part of the consultants toward the use of sophisticated technologies (such as proposed telemetering and remote control devices), and noted deficiencies in the feasibility studies for the Napoles Canal and Aguablanca Interceptor (para. 3.02). Because of the lack of.involvement by EMCALI's staff in project design, particularly in the early stage of project implementation, the technology transfer was not as great as had been expected by EMCALI. As part of a study for the Role and Use of Consultants in Bank Group Projects 1/, the major engineering and supervision contracts with consultants were evaluated and it was noted that strained consultant-client relationships had existed. EMCALI was, however, generally satisfied with the performance of Hernando Gonzales Hurtado and Hazen and Sawyer. 3.15 In late 1974 Hernando Gonzales Hurtado and Hazen and Sawyer had a major disagreement over pipe selection and Hazen and Sawyer was terminated by the former. Subsequently, Hernando Gonzales Hurtado terminated its contract with EMCALI. EMCALI then retained Indecom-Planes (Colombia) along with Gilbert and Associates (United States) to assist in completing the project. 3.16 As noted previously (para. 3.03), project implementation was initially very slow. It is difficult to separate the managerial and finan- cial deterrents to progress from the technical aspects, but it is clear that for the first three years of project construction, overall performance of EMCALI was not wholly satisfactory. Subsequent to 1973, project implemen- tation for most subprojects proceeded on schedule and performance was satis- factory. EMCALI has benefitted substantially from construction of the project and now has the capability of designing and supervising construction of much larger programs. 3.17 With the exception of structural steel and cement shortages in Colombia and delayed shipments of ancillary equipment for the treatment plant, the performance of suppliers may be regarded as satisfactory. The works were executed by a number of local contractors and their performance was on the whole satisfactory. IV. Operating Performance Water Production and Sales 4.01 As indicated in the following table, water production and sales .were substantially below the levels projected at the time of appraisal: 1/ Report No. 1824, December 8, 1977. - 22 - EMCALI Projected and Actual Water Sales and Production 1970-1977 (Million M3 ) Water Produced Water Sold Unaccounted for (%) Year Projected Actual Projected Actual Projected Actual 1970 84 82 68 62 20 24 1971 92 84 76 67 18 20 1972 106 98 89 71 16 27 1973 120 99 102 73 15 26 1974 132 100 112 77 15 23 1975 145 107 123 80 15 25 1976 159 112 135 81 15 28 1977 173 118 147 82 15 30 4.02 The reasons for the failure of water sales to reach projected levels were basically overestimation of both population growth and per capita consump- tion. The 1969 report of a local consultant (Juan Marchant), estimated Cali's population in mid-1970 at 980,000. The long term and current growth rate at that time was 7.9% per annum, reflecting heavy migration from the countryside as mechanization reduced the requirement for farm labor. At a reduced growth rate of 5.7% per annum, the population of Cali was projected to be 1.44 million by mid-1977. In fact, Cali's population was only 810,000 in mid-1970, and because of reduced birth rates and migration the city's population growth rate was actually 4.5% per annum. The result was that in mid-1977 Cali's population was 1.1 million, or 76% of the projected figure. 4.03 Per capita consumption had been projected to grow from 240 liters per capita per day in 1970 to 310 1/cd in 1977. In fact, per capita consumption, after peaking at 269 1/cd in 1972, declined to 224 lc/d in 1977. In large part this occurred because the bulk of new consumers connected to the system were from low income groups which used less water per capita than the average. It appears that industrial demand grew in line with overall demand, while it had been projected to grow at a slightly higher rate, and this factor also contributed to the variance from the forecast. Finally, reduced per capita consumption was in part due to the fact that the minimum consumption block for low-income consumers was reduced and the percent of connections metered increased from 96.8% to 98.7%. 4.04 It must be noted that EMCALI's unaccounted-for water increased from 24% in 1970 to 30% in 1977 which was twice the projected level of 15%. EMCALI believes the increase is a result of high consumption through unmetered connections in newly connected areas coupled with illegal connections in these areas. EMCALI pointed out that the original objective of 15% for unaccounted- for water was unrealistic and that the 20% figure which was attained in 1971 is the lowest percentage that can be realized. Under Loan 1523-CO the objec- tive has been revised to 23%. - 23 - Project's Role 4.05 The components of the project concerned with extending water supply and sewerage services have generated quantifiable benefits. For water supply, where transmission mains were completed by the end of 1972 and distribution networks and house connections partially completed (see Annex 2), it may be assumed that incremental sales after 1972 are attribut- 3 able to the project. These amount in aggregate through 1977 to 38 million m versus 174 million m projected. The reasons for this very large shortfall have been covered in the previous section. Sewerage laterals and house connections were also partially completed by an early stage, and incremental revenues from sewerage after 1972 may also be attributed to the project. 4.06 The water transmission and distribution, sewage collectors and networks, and drainage components of the project are all functioning in accordance with design specifications. The major component of the project, however, the Puerto Mallarino Treatment Plant is not yeS in commerc al operation as EMCALI's treatment plant capacity of 5.0 m /sec (4.2 m /sec safe yield) has been sufficient to mset average demand which is currently 3.7 m /sec, and peak demand of 4.2 m /sec in the dry season. EMCALI estimates that there are no operating cost savings associaSed with water production from the Puerto Mallarino facilities versus the 3.0 m /sec Rio Cauca Treatment Plant, and the decision regarding which plant to operate as the base load plant has not been taken. It is expected that the new capacity will be required to meet demand later this year and the Puerto Mallarino Plant will initially be used to allow periodic shutdowns of the existing plants for major maintenance. Over the longer run, this treatment plant, which is one of the most modern in Latin America, is expected to produce good quality water efficiently and, at projected water demand growth rates of 4.0% per annum based on 1977 per capita consumption levels, should meet requirements until 1995. At the present time, however, because of its size in relation to demand growth, it represents substantial advance investment on the part of EMCALI. Had the demand growth forecasts been more in line with realizations, its construction would have been delayed, and perhaps the size and location changed. Project Benefits 4.07 The project has enabled EMCALI to increase water supply service levels from 76.0% in 1972 to 91.9% in 1977, for the most part by extensions of supply mains and networks to low-income sections of the city. The project has also provided flexibility in the system, especially with respect to treatment capacity, and increased storage and new transmission mains have increased the reliability of supply throughout the system. Water waste by consumers has been reduced through metering programs and through tariff restructing. Water supply, and sewage network extensions and drainage have improved health and the quality of life for the urban poor as well as resulting in increased property values in the areas covered. The project generated employment in the construction industry and improved the technical performance of EMCALI and contractor personnel. - 24 - V. Financial Performance General 5.01 Water supply and sewerage account for 66% of EMCALI's assets and 24% of revenues, compared with 20% and 64% respectively for power distribution and 14% and 12% respectively for telephone service. As a general statement, tariff levels since 1972 have not been sufficient to meet capital expenditure requirements not covered by the Bank and Government loans, and the financial position has been less than satisfactory. The loan covenant specified rates of return on revalued net fixed assets for the three operating divisions, and with the exception of power, actual results have been well below the agreed objectives as indicated below: EMCALI: Covenanted Versus Actual Rates of Return 1972 1973 1974 1975 1976 1977 Water and Sewerage Covenanted 4.0 4.5 5.0 5.5 6.0 7.5 Actual 3.4 2.3 0.0 0.6 2.9 0.0 Power Covenanted 7.0 7.0 7.0 7.0 7.0 7.0 Actual 9.8 5.3 8.6 5.2 8.5 8.8 Telephones Covenanted 7.0 7.0 7.0 7.0 7.0 7.0 Actual 8.2 5.1 3.0 (0.9) (4.7) 2.6 5.02 During the period of project implementation, EMCALI objected to the use of the cost of living index for workers for the city of Cali as a basis for the annual revaluation of net fixed assets. This index-escalated much faster than had been expected, although in line with inflation in Colombia overall. EMCALI now attributes this to be a major factor in their failure to meet the covenanted rate of return requirement. 5.03 The original rate of return objective for water and sewerage, which was 8% from 1978 onward, is no longer either realistic or attainable and under the second loan has been revised downward to 5.0% by 1984. The interim 1978 objective of 3.0% is attainable if property reassessment is completed and monthly tariff increases continued, both of which have been covenanted. Because the Water Supply and Sewerage Division's assets include the treatment - 25 - plant which will be only partially utilized, very high tariffs would be required to satisfy the agreed rate of return objectives. Accordingly, in calculating the rate of return under the second loan, the treatment plant assets will be included in the rate base only to the extent its installed capacity was actually used during the relevant year. Tariffs will still be above the average incremental cost of supply and will generate revenues sufficient to cover debt service, working capital and capital expenditure requirements. The rate of return objectives for the power and telephone divisions were raised to 8% under the second loan, with an intermediate objective of 5.5% for the telephone division in 1978. Tariff increases have already been affected to allow these objectives to be attained. Water Supply and Sewerage Division Income Statement 5.04 Annex 4 compares projected and actual revenues and expenses for the Water Supply and Sewerage Division for each year 1970-1977, and the following table indicates the percentage increase between 1970 and 1977, and the percent the 1977 actual figures were of the original estimate, both in current Colombian pesos and in pesos used in the Appraisal Report projections: 1/ 1/ 1969 costs escalated at 8% per annum through 1973 and held constant thereafter. - 26 - EMCALI - Percentage Increase in Water Produced and Sold and Revenue and Expense (1970-1977) and Variations from Appraisal Projections 1977 Actual Percentage Increase/ 1977 Actual Percentage 1977 Actual Percentage (Decrease) vs. Projec- Increase/(Decrease) Increase/(Decrease) tions in Appraisal from 1970 2/ vs. Projections 2/ Report Col$ 1/ Water produced 44% (32%) (32%) Water sold 32% (44%) (44%) Average Revenue/m 262% 143% (10%) Revenues 380% 37% (50%) Operating Expenses 529% 134% (14%) Depreciation 737% 55% (43%) Interest 293% 41% (78%) Net Income (698%) (403%) (111%) 1/ 1969 costs escalated at 8% per annum through 1973 and held constant thereafter. 2/ Current Colombian pesos. 5.05 In spite of the fact that water sold increased by only 32% and was 44% below the projected volume, revenues were up 380% reflecting substantiall increased tariffs which produced an increase of 262% in average revenue per m These higher tariffs were necessary as operating costs increased by 529%, of which the major part was attributable to rapid inflation which averaged 17.8% per annum versus 8% per annum which had been forecast. The result was a loss in 1977 of Col$82.0 million (on revenues of Col$351 million), which was the fourth consecutive year of loss for EMCALI's Water Supply and Sewerage Division. 1/ 1/ This loss reflects the charge for depreciation based on revalued fixed assets and exclusion of valorization levies from revenues, while EMCALI's published financial statements base the charge for depreciation on acquisition cost and include valorization collections as non-operating revenue. On the latter basis the result in 1977 was a reported net income of Col$43.2 million for the Water Supply and Sewerage Division. - 27 - 5.06 The actual 1977 results can not meaningfully be compared with the appraisal estimates for 1977 (column 2 above) because of the substantially greater inflation rates than had been estimated at the time of appraisal. For purposes of comparison therefore, the key revenue and expense figures for 1977 have been converted into pesos used in the Appraisal Report projections (column 3 above). Here it can be's en that in constant terms tariffs in fact provided average revenues per m 10% below forecast. When this lower revenue per m3 is combined with the greatly reduced water sales, total revenues were only half the forecast level. Operating expenses, which had been estimated based on water production and sales did not quite reach the projected levels, but increased more rapidly than3water production. As a 3 result the unit cost of water sold was Col$1.60 m in 1977 versus Col$1.22 m projected (both in pesos used in the Appraisal Report projections). The major factors accounting for this unit cost increase versus the projection were administration expense, and production and distribution costs excluding direct labor, power and chemicals. Depreciation expense in total was well below the appraisal estimate, but in unit costs almost identical to the appraisal projection. The net loss in 1977 of Col$ 82.0 million, was Col$30.1 million in Appraisal Report terms compared with Col$27.1 million net income forecast for that year. 5.07 It is concluded that EMCALI's unfavorable income performance in both current Colombian pesos and in pesos used in the Appraisal Report versus the Appraisal Report projections principally reflects the fact that EMCALI did not achieve projected water sales volumes, was not able to maintain unit costs of production, and did not increase tariffs sufficiently to produce revenues to cover escalating costs. Cash Flow 5.08 EMCALI's Water Supply and Sewerage Division originally projected and actual cash sources and requirements for the years 1970-1977 are presented in Annex 5, and summarized below: - 28 - EMCALI - Water Supply and Sewerage Division Statement of Cash Flow 1970-1977 (Col$ million) Percent Projected/% Actual/% 1/ Variance Sources Internal Cash Generation 852 36% 550 24% (35%) Bank Loan 371 16% 519 22% 40% Government Loans 180 8% 378 16% 109% Other Loans 15 1% 266 11% 1,673% Customers Contributions 903 39% 178 8% (80%) Current Liabilities 9 - 442 19% Total 2,330 100% 2,33 100% Requirements Capital Expenditures 1,019 45% 1,843 77% 81% Networks Constructed by Developers 530 23% 49 2% (91%) Debt Service 536 24% 381 16% (30%) Working Capital 191 8% 110 5% (42%) Total 2,276 100% 2,383 100% 5% Increase/(Decrease) in Cash 54 (51) 1/ Current Colombian pesos. 5.09 Overall sources and requirements for the period 1970-1977 were in line with original projections, although the components of the sources were substantially different. Because the original time period for completion of the project was 1970-1973, all comparisons with the appraisal forecasts are affected by the extended construction period and the concurrent inflation and cost overruns on the project. Internal cash generation was only about two-thirds the projected level as a result of the poor net income situation previously discussed. As a result, it met 24% of total cash requirements as opposed to 36% which had been expected. Because devaluation rates were much greater than projected and the Bank loan was disbursed on a delayed basis, its Colombian pesos equivalent was 40% more than projected. The Bank loan met 22% of EMCALI's cash requirements during the 1970-1977 period versus 16% originally anticipated mainly because the capital expenditure program was curtailed (para. 5.10). Customers' contributions (networks constructed by developers, connection fees, costs and valorization 1/) which 1/ Improvement levies on beneficiaries of public works projects. - 29 - were projected to be the largest source of funds, were only 10% of the projected amounts, which necessitated the sharp increase in local borrowing, including a more than doubling of the Government loan, together with short-term borrowings and increased other current liabilities. 5.10 Regarding cash requirements, capital expenditures in total exceeded the projections by 20% but the works undertaken by EMCALI directly were 81% over projected levels. The major reason for this was cost inflation, as the actual volume of work undertaken on the project was less than had been contemplated in the original scope of the project (para. 3.02) and, more importantly, a large portion of the works scheduled for the 1974-1977 period was postponed because of cash shortages. Requirements represented by networks constructed by developers were only about 10% of those projected, which is the principal reason for the relatively modest overall overrun. Debt service was below projections because of the slow drawdown of the Bank loan (para. 3.11). The debt service ratio in 1977 was 1.4, which is satisfactory, and for the period 1970-1977 it averaged 1.6. 5.11 Sections 4.02 and 4.03 of the General Conditions specify that principal and interest on the loan shall be paid in the currency or currencies in which the loan funds were withdrawn or in currencies used to purchase those withdrawn. EMCALI states that this provision adversely affected its financial position by increasing interest expense, loan amortization and the amount of long-term debt outstanding as a result of the substantial depreciation of the US dollar versus European currencies which were principally used for loan withdrawals. In fact, just over 50% of the loan was denominated in US dollars, and about 36% in hard currencies of European and Middle East countries. This mix of currencies resulting from procurement and Bank options is not unbalanced in favor of hard currencies. In 1978 EMCALI would have lower debt service payments by about US$134,000 (7.5%) if the entire loan had been denominated in US dollars, and this has not been a major negative factor in EMCALI's finan- cial performance.. 5.12 In summary, below expectation water sales and revenues and low customer contributions coupled with substantial project cost overruns, were primarily responsible for the cash shortages faced by EMCALI during project construction. Project Financing 5.13 The original and actual financing plan for the project is summarized below: - 30 - Financing Plan for the Project (Col$ million) Appraisal Report Estimated Projection (%) Actual (%) 1! Variance (%) Requirements Project Works 670 - 1,179 - 76% Interest During Construction 82 - 175 - 113% Working Capital 39 - (17) - (143%) Total Requirements 791 - 1,337 - 69% Sources Bank Loan 371 47% 519 39% 40% Government Loan 180 23% 377 28% 109% Net Internal Cash Generation 175 22% 344 26% 97% Customer Contributions 65 8% 97 7% 49% Total Sources 791 100 1,337 100% 69% 1/ Current Colombian pesos. 5.14 In spite of the fact that total cash requirement for the project exceeded projections by 69%, the financing plan did not change markedly in percentage terms, although the absolute amounts had to be increased substan- tially to cover the cost overrun. The Bank loan, however, met only 39% of the actual cost of the project (including interest during construction and working capital) as opposed to 47% originally expected as a result of the cost overrun and because the rates of local inflation exceeded the rates of devaluation of the Colombian pesos. The government loan had to be increased by more than 100% to meet the cash shortfall, and covered more of the cost than originally expected. Net internal cash generation was about twice that forecast, and met a slightly higher percentage of the total cost. Balance Sheet 5.15 EMCALI's water supply and sewerage division's projected and preliminary actual balance sheets as of December 31, 1977 are presented in Annex 6 and summarized below: - 3A1 - EMCALI Water Supply and Sewerage Division Balance Sheet December 31, 1977 (Current Col$ million) Appraisal Report Preliminary Projection Actual 1/ Variance % Assets Fixed (net) 1,932 2,168 12% Current/other 286 333 16% Total 2,218 2,501 13% Liabilities and Equity Current Liabilities 133 287 116% Long-term Debt 515 1,621 215% Equity 1,570 593 (62%) Total 2,218 2,501 13% 1/ For balance sheet purposes assets are not revalued and are not directly comparable to projections. Net revalued fixed assets for determination of rate of return are calculated separately. 5.16 Even though EMCALI's water and sewerage divisions' capital expendi- ture program for the years 1970-1977 was curtailed, the value of the empresas assets at the end of 1977 exceeded projections by 12% as a result of more rapid than anticipated inflation which increased the cost of the assets installed. Because of the high cash requirements for capital expenditures and the poor net income and cash.generation performance of EMCALI during the period, requiring a high level of debt financing, long-term debt increased by more than 200% and at December 31, 1977 it accounted for a rather high 73% of total capital, compared with 25% originally projected. Tariff Structure 5.17 EMCALI's tariff structure is designed to cover in aggregate operating, maintenance and debt service costs and is not cost related by category of consumer. The tariff structure is socially oriented, with the higher income consumers subsidizing the cost of service to the low income - 32 - consumers. This is done by classifying customers into one of 13 categories based on property values, with those in the categori s representing higher property values paying a substantially higher rate/m for their minimum consumption. 1/ Periodic asset revaluations automatically increase revenues without any nominal increase in the rates, and in the past EMCALI was often slow in adopting the higher valuations and recategorizing consumers. Rate of Return 5.18 The incremental financial rate of return on the project estimated at the time of appraisal was 13.5%. Based on the actual cost of the project, incremental volumes now expected and average tariff levels prevailing in 1977, the expected incremental financial rate of return on the project is 5.9%. 2/ The major reasons for the reduced return is the fact that water sales volumes have been and are expected to be much lower than anticipated, tariffs were only 90% of p5ojected levels and operating expenses were 31% higher (both in terms of m ) than projected levels (in real terms). Project costs were about the same as had been projected at the time of appraisal in real terms. The rate of return calculation is presented in Annex 7. VI. Institutional Performance Management and Organization 6.01 EMCALI is one of the few large city public utilities in the world which has combined responsibility for water and sewerage, electric power distribution, and telephone. As such, it is a complex operation which needs good top management and a proper functional organization. The General Manager of EMCALI is appointed by the Mayor of Cali and there have been six changes since the project was appraised. The institutional and financial difficulties which arose in 1971 and 1972 as a result of political inter- ference have been previously referred to (para. 3.03). 6.02 EMCALI's organization has been along functional lines and a reorga- nization in early 1977 increased the autonomy of the three operating divisions. At the present time, EMCALI is an acceptably managed and organized public utility. However, further improvements are required, particularly in EMCALI's financial management. Improvements in management and organization have resulted from both the Bank's and EMCALI's efforts. Institutional Covenants 6.03 At the time of appraisal, weaknesses were identified in EMCALI's training programs and management information, budgeting and accounting systems and covenants were incurred in the Loan Agreement regarding improve- ments. The concerned areas of operations were regularly followed up during 3 1/ Categories 1-3 pay less than Col$ 1.00/m of mnimum consumption while categories 11-13 pay an average of Col$18.27/m. 2/ All costs and benefits in 1969 Colombian pesos. - 33 - loan supervision and there have been basic improvements. The covenants have been satisfactorily met. However, during appraisal of the Second Cali Project it was determined that further improvements were required in the management information and budgeting systems. Also, provisions were made in the second loan for strengthening of the coordination of the administration and operation of EMCALI's divisions. Employment 6.04 In 1970, EMACALI had 1,751 employees of which 632, or 36%, were in the water and sewerage division. By 1977 there were 3,001 employees in EMACALI and 972 employees (33%) in the water and sewerage division. EMCALI has in fact expanded its water supply and sewerage division staff only in line with increased operations as indicated in the following table: EMACALI - Number of Employees and Relationship to Population Served Increase 1970 1977 (Decrease) Population served (water) 640,000 1,050,000 64% Connections (water) 89,200 147,100 65% Total employees 1/ 1,751 3,001 78% Water and Sew. employees 1/ 632 972 54% Employees per 1,000 connection (water) 7.1 6.6 (7%) Employees per 1,000 population served (water) 99 93 (6%) 1/ Includes contract employees and students. 6.05 Employment in the water supply and sewerage division expanded at a lesser rate than EMCALI overall, the population served and the number of connections. In 1977 the employees per 1,000 connection served was lower than in 1970, and the absolute level indicates efficient operation in comparison with other developing country utilities. Management Consultants 6.06 During the period of project construction, EMCALI used the services of local management consultants to a limited extent, principally to assist in improving electronic data processing and information systems. - 34- Reporting 6.07 Quarterly Reports of covering project implementation, operating results and financial performance have been submitted to the Bank with some delays. All additional requests by the Bank for information have been satisfied. VII. Project Justification Water Distribution and Sewerage 7.01 The basic objectives of the project were achieved. Water supply service levels in Cali through house connections were increased from 76% in 1972 to almost 92% in 1977, one of the highest for any large city in Latin America. The bulk of new areas served were low income neighborhoods. While the initial project called for low income neighborhoods covering about 500 hectares to be served, EMCALI actually extended service to low income neighborhoods covering 800 hectares with an estimated population of 180,000. Similarly, service levels for sewerage were increased from 73% in 1970 to 82% in 1977, again one of the highest levels for any city in Latin America. Sewerage service also was installed principally in low income areas where an estmated population of 140,000 was benefitted. The water supply and sewerage mains and distribution networks were efficiently installed and represent the least cost solution to increasing service levels. Treatment Plant 7.02 While the treatment plant will be operated at a lower capacity than envisioned (see para. 4.06), it is well designed and efficient and at the lower demand growth now forecast will meet EMCALI's requirements into the 1990's. In light of the actual development of the city to the south and availability of water from the Pance River not required for agricultural purposes, the chosen site of the treatment plant may not be the least cost location. The pumping requirements for the rejected Pance River site now appear to be less than for Puerto Mallarino, and the longer transmission main from that location could now be used to supply distribution networks in south Bogota which now have to be supplied from the north. The completion of the Aguablanca Interceptor, now scheduled for the end of 1981, will greatly improve water quality at the Puerto Mallarino Treatment Plant as was planned as part of the original project. VIII. Bank Performance Appraisal Issues 8.01 There were no major issues raised during appraisal. EMCALI was considered to be a fairly well managed utility which needed no more than the typical institutional covenants. The project was considered to be satis- factorily prepared by both the Bank and EMCALI. Based on the population - 35 - projections prepared by a local consultant and on EMCALI's water demand growth projections derived therefrom, the water treatment plant was required. The failure of demand to materialize has been previously noted (paras. 4.02 and 4.03) as well as the changes in project scope (para. 3.02). EMCALI believes that the latter were caused by deficiencies in the feasibility studies pre- pared by local and foreign consultants. The concerned subprojects have been redesigned by EMCALI and approved by the Bank, and construction will commence in 1979. Loan Supervision 8.02 Because of the institutional and financial difficulties which became apparent in 1971, the project was heavily supervised. From the time of loan effectiveness on June 4, 1970 to the end of 1977, there were a total of 21 supervision missions. The visits were followed by mission reports high- lighting deficiencies and by letters and discussions with EMCALI's management and municipal and national government officials urging that corrective actions be initiated where required. It is believed that this pressure on the part of the Bank was instrumental in reducing political interference in EMCALI and in accelerating institutional improvements. 8.03 On the technical side, the Bank induced EMCALI to undertake long- range planning and to improve the quality of its personnel. The Bank assisted EMCALI's staff in project construction and responded positively to EMCALI's proposed changes in project scope, financing of modifications to an existing treatment plant, procurement of stocks, and foreign procurement practices. Bank Performance 8.04 It is EMCALI's opinion that the Bank's performance was altogether appropriate. EMCALI considers that the Bank's staff were well informed on the project and valued their advice on institutional and project implementation matters. EMCALI found that the Bank responded promptly to their requests and maintained satisfactory supervision over the project's development. EMCALI believes that the Bank did over-emphasize some aspects of the project such as the rate of return on revalued fixed assets, the hiring of consultants and payments to contractors and suppliers. This was attributed to the Bank's uncertainty regarding EMACALI's capability to undertake these responsibilities on its own. In all other aspects, EMCALI believes the Bank's attitude was one of flexibility and cooperation. Working Relationships 8.05 The working relationships between the Bank and EMCALI were, and are considered to be good by both parties. Covenants 8.06 Both the Bank and EMCALI considered that the covenants were appropriate and useful in improving EMCALI as an institution. - 36 - IX. Conclusions and Lessons Learned 9.01 The project experienced two broad categories of difficulties, those related to institutional deficiencies, including EMCALI's financial situation, and those related to the technical aspects. The lessons learned are generally applicable to all Bank-financed projects in the public utility sector and the experience was immediately valuable in designing the covenants for the second Cali loan approved in 1978. Institutional 9.02 For the most part the political interference and the resulting institutional difficulties which took place in 1971 and 1972 (para. 3.03) could hardly have been predicted. The appraisal mission noted that all indications were that partisan influence on EMCALI's Board was kept to a minimum. The Bank did act promptly and properly in exerting pressures to correct the deficiencies, including the possibility of loan cancellation. The lesson learned is that the Bank can have some positive influence on Governments and institutions when political interference adversely affects implementation of a Bank-financed project. Financial 9.03 The poor financial results in EMCALI's water and sewerage division resulted primarily from the failure to raise tariffs to compensate for increasing costs. Apart from the monthly increases agreed to by EMCALI, it was understood that revised property assessments would be used in assign- ing consumer categories (para. 5.17). It would have been preferable to incorporate this requirement in the loan agreement, and this was accomplished in the second loan. Also, while EMCALI considered that the Bank paid undue attention to the matter of finances, the attention had only modest tangible results and perhaps more serious actions should have been taken to have EMCALI effect tariff increases. 9.04 EMCALI contended that the periodic revaluation of assets in line with the cost of living index for Cali made the rate of return covenant too harsh (para. 5.02). To meet the 7.5% rate of return objective for 1977, for the Water and Sewerage Division, EMCALI's tariffs would have had to be 40% higher, which amounts to an increase of only 5% per annum (0.4% per month) from 1970 over actual tariff increases and a real increase of only 30% for the seven year period. In view of the rapid inflation which prevailed, the revaluation of assets was essential to make the covenant meaningful. While the required tariff increases may have been difficult politically, if EMCALI had met the covenanted return in 1977 its debt service ratio would have been 3.0 (versus 1.4 actual) and internal cash generation would have met 65% of capital expenditures (versus 12% actual). Technical 9.05 The technical difficulties experienced in project implementation for the most part appear to be a result of deficiencies in the project feasibility studies. The main deficiencies, with respect to the treatment - 37 - plant, were the overestimation of population and population growth and future water demand and insufficient least cost analysis of the location alternatives (para. 7.02). The error in estimating population (para. 4.02) was perhaps unavoidable,. but the population growth projection failed to recognize in full the declining trend in urban migration growth rates. The assumption of increasing per capita water consumption (para. 4.03), when a high percentage of the new connections were to be made to low income households, was the critical determination of the need for the plant and proved to be invalid. It may be noted that the demand forecast was made in conformity with the method normally used at that time. Because of under or overestimations of demand which have resulted from that method, demand forecasts are now made, where possible, by projecting volumes by category of consumer. The water demand projections prepared for the second loan take into account the lessons learned since 1970. 9.06 The delay in completing the treatment plant in part corrected the situation regarding the over-estimate of demand. If the actual demand forecast had been realized, Cali would have faced water shortages from 1973 until the time of completion of the Puerto Mallarino Treatment Plan in mid-1978. 9.07 EMCALI believes that if the decision regarding treatment plant location were to be made today, the rejected site on the Pance River south of town would be preferable (para 7.02). Because of the relatively little time since the feasibility studies were made in comparison with he useful life of the treatment plant (potential expanded capacity of 24 m should meet demand for about 45 years) it is not logical to assume that the trend in city expansion would alter so drastically as to raise doubts about the selected location. Accordingly, the lesson to be learned is that where the siting aspects are unclear, the feasibility studies should thoroughly examine all alternatives under different city development assumptions in order to give as high a probability as .possible to having the selected project be the least cost alternative. The incorrect assumption regarding city development also affected the designs for the Aguablanca Interceptor and the Napoles Drainage Canal. Latin America and the Caribbean Regional Office June 1978 - 38 - ANNEX 1 Page 1 of 2 CALI WATER SUPPLY AND SEWERAGE PROJECT LOAN 682-CO Summary of Revisions to Project Scope - Project Execution 1/ Treatment Plant 1. The treatment plant was basically constructed as designed with the exception of the intake, pumping and preliminary treatment of raw water. The original design called for a gravity flow from the intake to an excavated lagoon that would act as a combination grit chamber and preliminary sedi- mentation. The screens and low pressure pumps were to be installed at the outlet of the lagoon. Actually the low pressure pumps were installed at the intake structure and the lagoon was replaced by a reinforced concrete structure which included mechanical screens and grit chambers. Distribution 2. Booster pumping stations built were those for the low income "Barrios" of Siloe and Terron Colorado, which represent about 74% of the work considered at appraisal time. 3. Instead of constructing two storage tanks of 15,000 m3, only one tank of 15,000 m3 was constructed, and about 30% of a second tank of 9,200 m3 (to be finished with EMCALI's own resources). This in overall represents 60% of the amount of work considered at appraisal time. 4. The comparison of the length and diameter of transmission mains considered at appraisal time and actually installed is as follows: Length Diameter Appraisal Actual 60 6,800 -- 56 -- 8,336 54 4,400 -- 50 -- 5,124 48 4,900 -- 44 -- 3,474 42 1,100 -- 38 -- 1,842 36 700 -- 34 -- 1,604 30 1,600 -- 1/ For effects on project costs see Annex 3, page 1. ANNEX 1 - 39 - Page 2 of 2 The cost of both alternatives is about the same. However, as the length of large supply lines (12-inch to 24-inch) installed was much less than considered at the time of the appraisal, it is estimated that 82% of the combined items for pipelines have been executed. 5. All of the planned distribution network extensions were executed. The material acquired for house connections and the number of house connec- tions installed within the program represents 88% of the cost estimated at appraisal time. 6. Acquisition of vehicles not considered at appraisal time added Col$3.3 million to the cost of the project. 7. The item for tools was exceeded by 34% and the item for land by 84%. 8. The upgrading of the San Antonio and Cauca treatment plants (at a low unit price), to compensate for the delay in the construction of Puerto Mallarino plant, added Col$7.5 million to the cost of the project. 9. The Aguablanca interceptor and the sewerage pumping facilities were deleted altogether. 10. Only 40% of the Napoles Canal was completed. The sewers and canals actually built represent 68% of the amount of work considered at the time of the appraisal. 11. In lateral and sewerage connections, the amount of work executed has been 64%. above the appraisal estimate. 12. The Cafiaveralejo Reservoir which was not considered at the time of the appraisal but constructed as a part of the project for runoff regulation, added Col$12.2 to the cost of the project. COLOMBIA CALI WATER SUPPLY AND SEWERAGE PROJECT COMPARISON OF PROJECTED AND ACTUAL WORK PROGRAMS - WATER SUPPLY Year 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 Quarter 1234 12 3j 123 134 1 2334j1234 Treatment Plant Plant Equipment C Booster Pumping Station Pumping Equipment Storage Tanks Transmission Mains Additional Distribution House Connection Upgrading old treatment plants NOT CONSID RED AT THE TIME OF APP AISAL APPRAISAL 1Z / ACTUAL COLOMBIA CALI WATER SUPPLY AND SEWERAGE PROJECT COMPARISON OF PROJECTED AND ACTUAL WORK PROGRAMS - SEWERAGE Year 1969 1970 1971 1972 1973 1974 1975 1976 1977 Quarter 1f23 4 123 1 2 34 231234 12 33123 1 234 1 Interceptor DELETED Sewers __I__ Lateral and house connections Drainage canals Cafiaveralejo Reservoir NOT CON IDERED AT A PRAISAL TIM APPRAISAL ACTUAL 0 - 42 - ANNEX 3 Page 1 of 4 CALI WATER SUPPLY AND SEWERAGE PROJECT LOAN 682-CO Review of Cost Estimate Modified Project Estimate for Works Actually Executed Differ6nce in amount Modified Appraisal of work Project Item Estimate executed Estimate ------ (in 1969 Col$ million) ---------- Treatment Plant 72.5 0 72.5 Booster pumping stations 8.3 -2.0 6.3 Storage tanks 4.6 -L.8 2.8 Pipelines 98.4 -17.7 80.7 Additional distribution 46.8 0 46.8 House connections 42.2 -5.0 37.2 Tools 3.8 +1.3 5.1 Vehicles* -- +3.3 3.3 Land 4.5 +3.8 8.3 Upgrading old plants* -- +7.5 7.5 Interceptors 17.3 -17.3 0 Pumping facilities 15.0 -15.0 0 Sewers and canals* 42.2 -18.1 24.1 Laterals and house connections 46.1 +29.7 75.8 Caiaveralejo reservoir* -- +12.2 12.2 Construction Cost 401.7 -19.1 382.6 Engineering 31.5 +1.6 33.1 Total Base Cost 433.2 -17.5 415.7 Technical contingencies 98.3 -4.7 93.6 Project cost in 1969 prices 531.5 -22.2 509.3 Escalation 138.9 -5.8 133.1 Project cost in current prices 670.4 -28.0 642.4 * Grouping of works is as modified by EMCALI for purposes of supervision of work progress and work payments. - 43 - ANNEX 3 Page 2 of 4 CALI WATER SUPPLY AND SEWERAGE PROJECT LOAN 682-CO Table of Revised Cost ---------------(in Col$ million Appraisal Estimate 1/ Actual Results 2/ Items Local Foreign Total Local Foreign 3/Total Treatment plant 34.8 37.7 72.5 177.0 191.8 368.8 Booster pumping station 5.8 0.5 6.3 17.3 1.5 18.8 Storage tanks 2.8 -- 2.8 7.1 -- 7.1 Pipelines 52.1 28.6 80.7 108.8 92.4 201.2 Additional distribution 39.8 7.0 46.8 40.2 7.1 47.3 House connections 27.9 9.3 37.2 40.0 25.0 65.0 Tools 1.4 3.7 5.1 1.9 5.9 7.8 Vehicles -- 3.3 3.3 -- 5.2 5.2 Land 8.3 -- 8.3 11.3 -- 11.3 Upgrading old plants 6.5 1.0 7.5 21.8 3.3 25.1 Sewers and canals 23.7 0.4 24.1 87.5 1.5 89.0 Laterals and connections 71.7 4.1 75.8 170.7 9.8 180.5 Cafiaveralejo reservoir 11.9 0.3 12.2 23.5 0.6 24.1 Construction cost 286.7 95.9 382.6 707.1 344.1 1,051.2 Engineering 24.8 8.3 33.1 96.3 31.4 127.7 Technical contingencies 70.2 23.4 .93.6 -- -- - Price escalation 99.8 33.3 133.1 -- -- -- Project cost in current prices 481.5 160.9 642.4 803.4 375.5 1,178.9 1/ In 1969 Colombian pesos. Adjusted in accordance with amount of work actually built. 2/ Current Colombian pesos. 3/ Including foreign indirect component.as in appraisal estimate (14% weighted average). ANNEX 3 -44 - Page 3 of 4 CALI WATER SUPPLY AND SEWERAGE PROJECT LOAN 682-CO Table of Revised Cost -----------------(in US$ million) 1/ Appraisal Estimate Actual Results Items Local Foreign Total Local Foreign 2/Total Treatment plant 2.01 2.17 4.18 5.69 6.14 11.83 Booster pumping station 0.33 0.03 0.36 0.48 0.09 0.57 Storage tanks 0.16 -- 0.16 0.22 -- 0.22 Pipelines 3.01 1.65 4.66 3.46 3.02 6.48 Additional distribution 2.30 '0.40 2.70 1.78 0.31 2.09 House connections 1.61 0.54 2.15 1.54 1.07 2.61 Tools 0.08 0.21 0.29 0.09 0.26 0.35 Vehicles -- 0.19 0.19 -- 0.23 0.23 Land 0.48 -- 0.48 0.52 -- 0.52 Upgrading old plants 0.37 0.06 0.43 0.60 0.10 0.70 Sewers and canals 1.37 0.02 1.39 2.70 0.04 2.74 Laterals and connections 4.14 0.23 4.37 6.30 0.35 6.65 Cafiaveralejo reservoir 0.68 0.02 0.70 0.90 0.03 0.93 Construction cost 16.54 5.52 22.06 24.28 11.64 35.92 Engineering 1.43 0.48 1.91 3.50 1.40 4.90 Technical contingencies 4.05 1.35 5.40 -- -- -- Price escalation 2.00 0.66 2.66 -- -- -- Project cost in current prices 24.02 8.01 32.03 27.78 13.04 40.82 1/ Adjusted in accordance with amount of work actually built. 2/ Including foreign indirect component as in appraisal estimate (14% weighted average). ANiNEX 3 Page 4 of 4 - 45 - CALI WATER SUPPLY AND SEWERAGE PROJECT LOAN 682-CO Comparison of Predicted and Actual Contingencies (in Col$ million) Appraisal Estimate.Y Actual Results 2/ Component Local Foreign Total Local Foreign Total Treatment plant 34.8 37.7 72.5 34.8 37.7 72.5 Technical contingencies 7.8 8.5 16.3 39.7 19.6 59.3 Price escalation 11.1 12.1 23.2 102.5 134.5 237.0 Total for component 53.7 58.3 112.0 177.0 191.8 368.8 Other water components 144.6 53.4 198.0 144.6 53.4 198.0 Technical contingencies 32.6 12.0 44.6 -23.6-3/ -3.1 3/ -26.7 3/ Price escalation 46.3 17.1 63.4 127.4 90.1 217.5 Total for component 223.5 82.5 306.0 248.4 140.4 388.8 Sewerage & drainage component 107.3 4.8 112.1 107.3 4.8 112.1 Technical contingencies 24.2 1.0 25.2 19.6 0.8 20.4 Price escalation 34.5 1.4 35.9 154.8 6.3 161.1 Total for component 166.3 6.9 173.2 281.7 11.9 293.6 Total for construction 443.5 147.7 591.2 707.1 344.1 1,051.2 Engineering 38.3. 12.9 51.2 96.3 31.4 127.7 Total for project 481.5 160.9 642.4 803.4 375.5 1J78.9 Actual Cost Overrun as a % of Construction Base Cost Treatment plant 54 54 54 409 409 409 Other water components 54 54 54 72 162 96 Sewerage and drainage component 54 54 54 163 148 162 Aggregate total 54 54 54 147 259 175 1/ In 1969 Colombian pesos. Adjusted to actual amount of work executed. 2/ Incurrent Colombian pesos. 31 Negative results due to price overestimation. COLONBIA CALI WATER SUPPLY AND SEWERAGE PROJECT EMCALI WATER AND SEWERAGE DIVISIONS Appraisal Projections and Actual Statements of Income and Expenses (1970-1977) (Col$ million) 1970 1971 1972 1973 1974 1975 1976 1977 YEAR ENDING DECEMBER 31 Actual Proj. Actual Proj. Actual Proj. Actual Proj. Actual Pr0j. Actual Proj. Actual Proj. Actual Proj. ater Produced (Million M3) 82 84 84 92 98 106 99 120 100 132 107 145 112 159 118 173 Water Sold (Million M3) 62 68 67 76 71 89 73 102 77 112 80 123 81 135 82 147 Unaccounted-for Water (%) 24 20 20 18 27 16 26 15 23 15 25 15 28 15 30 15 RATES Average price Col$/M3 1.16 1.16 1.35 1.34 1.64 1.54 1.95 1.73 2.15 1.73 2.69 1.73 3.55 1.73 4.21 1.73 (Water - 62.5%) (Sewerage - 37.52) REVENUES Water Billings 43.5 49.3 55.0 63.6 71.2 85.7 87.4 110.3 101.8 121.1 132.5 133.0 177.3 146.0 217.7 158.9 Sewerage Billings 28.5 29.6 35.3 38.2 45.2 51.4 54.8 66.2 63.7 72.6 82.6 79.7 110.2 87.5 132.9 95.3 Other 1. 0 9 1 3 1.1 1.8 1.4 3.1 1. 3 2.0 3.2 2.2 4.1 2.4 5.3 2.6 Total Revenues 73.0 79.3 91.6 102.9 118.2 138.5 145.3 178.3 169.1 185.7 218.3 214.9 2p1.6 235.9 350.9 256.8 EXPENSES Power 2.5 3.5 3.7 4.4 4.7 6.0 5.6 7.5 8.6 8.7 10.5 9.8 15.1 10.9 17.6 12.1 a Chemicals 3.7 3.5 4.3 4.3 4.4 4.9 3.2 6.8 6.5 7.5 11.7 8.2 15.3 8.6 16.4 9.6 Production Labor 1.9 1.9 3.7 2.0 2.6 2.1 3.3 2.9 4.8 3.4 6.9 3.4 8.1 3.4 10.8 3.4 Maintenance Treatment Plant 0.6 0.5 0.5 0.5 1.6 0.5 3.8 0.7 4.3 0.8 6.5 0.8 8.1 0.8 11.3 0.8 Maintenance Distribution System 9.7 10.8 12.3 13.0 16.4 16.5 19.3 20.5 29.5 21.7 37.8 22.9 46.1 24.3 53.7 25.6 Other Prod. and Diatr. Costa 2.0 2.1 4.6 2.4 12.2 3.0 14.6 3.5 19.4 3.8 22.4 4.0 22.4 4.2 33.1 4.4 Administration Water/Sewerage 2.3 7.9 2.8 9.2 2.3 11.3 4.4 13.6 6.6 14.4 8.1 15.2 7.9 16.1 11.0 17.0 Administration Encali 6.7 2.4 9.2 3.0 13.7 3.9 14.7 4.9 21.4 5.1 23.0 5.4 26.8 5.7 45.0 6.1 Other Administrative Costa 3.1 2.6 2.0 3.0 2.6 3.8 3.6 4.5 5.4 4.8 5.9 5.0 7.1 5.3 9.9 5.7 4% Contribution to Municipality 2.9 3.2 3.6 4.1 4.7 5.5 5.6 7.1 6.5 7.8 8.6 8.5 11.4 9.3 13.5 10.2 Municipal Services 1.7 1.7 2.3 3.1 4.1 2.7 5.9 3.4 6.3 3.8 7.3 4.2 9.2 4.6 11.2 4.9 Total Expenses 37.1 40.1 49.0 48.0 69.3 60.2 84.0 75.4 119.3 81.8 148.7 87.4 177.5 92.2 233.5 99.8 Income Before Depreciation 35.9 39.7 42.6 54.9 48.9 78.3 61.3 102.9 49.8 113.9 69.6 127.5 114.1 142.7 117.4 157.0 Depreciation 14.7 27.2 23.9 33.2 27.7 44.3 42.6 57.3 50.7 63.4 63.0 68.5 77.0 74.6 123.0 79.3 Income Before Interest- 21.2 12.5 18.7 21.7 21.2 34.0 18.7 45.6 (0.9) 50.5 6.6 59.0 37.1 88.1 ( 5.6) 77.7 Interest 8.0 12.7 12.5 20.7 10.8 34.7 8.6 51.2 8.7 60.7 13.5 58.3 21.1 55.8 31.5 53.2 Income After Interest 13.2 ( 0.2) 6.2 1.0 10.4 ( 0.7) 10.1 ( 5.6) ( 9.6) (10.2) ( 6.9) 0.7 16.0 12.3 (37.1) 24.5 Non-Operatiig Income/kpense (Net) 0.5 0.9 ( 2.7) 1.1 ( 2.1) 1.4 3.2 1.8 ( 3.5) 2.0 (19.6) 2.2 (34.9) 2.4 (44.9) 2.6 Net Income 13.7 0.7 3.5 2.1 8.3 0.7 13.3 ( 3.8) (13.1) ( 8.2) (26.5) 2.9 (18.9) 14.7 (92 0) 27.1 Retqrpa on Net Plant (Z) 1/ 6.0 2.6 4.2 3.6 3.4 4.4 2.3 4.9 0.0 5.5 0.6 6.0 2.8 6.7 0.0 7.9 Operating Ratio (1) 50.8 50.3 53.5 46.6 58.6 43.5 57.8 42.3 70.6 41.8 68.1 40.7 60.9 39.5 66.5 38.9 I/ Income Before Interest Divided by Net Fixed Assets Excluding Customers and Urban Developers' Contributions. Which are: 351.7 474.8 446.0 609.7 625.0 765.3 805.8 939.2 1006.1 916.0 1064.0 938.8 1307.0 1014.2 1129.0 985.2 COLOMBIA CALI WATER SUPPLY AND SEWERAGE PROJECT EMCALI WATER AND SEWERAGE DIVISONS Appraisal Projections and Actual Statements of Cash Flow 1970-1977 (Col$ Million) 1970 1971 1972 1973 1974 1975 1976 1977 YEAR ENDING DECEMBER 31 Actual Proj. Actual Proj. Actual Proj. Actual Proj. Actual Proj. Actual Proj. Actual Proj. Actual Proj. Sources of Funds Income Before Interest 21.2 12.5 18.7 21.7 21.2 34.0 15.7 45.6 (0.9) 50.5 6.6 59.0 37.1 68.1 (5.6) 77.7 Depreciation 14.7 27.2 23.9 33.2 27.7 44.3 42.6 57.3 50.7 163.4 63.0 68.5 77.0 74.6 123.0 79.3 Increase in Social Sec. Reserve 2.1 1.3 0.6 1.7 1.1 2.2 1.4 2.9 9.1 3.1 12.5 3.3 7.7 3.4 79.5 3.7 Non-Operating Income 0.5 0.9 _2.7) 1.1 (2.1) 1.4 3.2 1.8 1.5) 2.0 (19.6) 2.2 (34.9) 2.4 (44.9) 2.6 Total Internal Cash Generation 38.5 41.9 40.5 57.7 47.9 81.9 65.9 107.6 55.4 119.0 62.5 133.0 86.9 148.5 152.0 163.3 Project Loans - Bank - 18.0 13.5 77.7 53.7 143.2 35.0 131.9 99.7 - 210.5 - 78.3 - 28.8 - Project Loans - National - 8.0 - 45.0 29.9 80.0 27.3 47.0 60.9 - 10.0 - 250.0 - - - Other Loans 83.1 15.0 31.0 - 8.5 - 10.3 - 12.5 - 23.8 - 25.5 - 70.7 Total Loans 33.1 41.0 44.5 122.7 92.1 223.2 72.6 178.9 173.1 - 244.3 - 353.8 - 99.5 - Customers' Constirubion 5.6 19.2 7.9 23.8 14.1 121.5 20.5 86.3 19.5 29.5 16.6 30.1 20.3 29.6 33.4 31.5 Developers' Contribution - 29.0 - 55.2 4.4 62.1 14.2 60.6 2.0 75.1 17.8 79.8 4.9 81.6 6.5 87.0 Increase in Current Liabilities 4.0 0.8 0.1 1.1 8.8 1.5 12.5 2.1 69.5 0.8 156.3 0.8 62.5 0.3 85.9 0.8 TOTAL SOURCES OF FUNDS 131.2 131.9 93.0 260.5 169.3 490.2 185.7 435.5 319.5 224.4 537.5 243.7 528.4 260.5 367.3 282.6 --------------------------------------------------- ------------..-... .... .... .... ... C ao COLOMBIA CALI WATER SUPPLY AND SEWERAGE PROJECT EMCALI WATER AND SEWERAGE DIVISIONS Appraisal Proiections and Actual Statements of Cash Flow 1970 - 1977 (Col$ million) 1.970 .1.971 1.972 1.973 1.974 1.975 1.97 1.8977 i'EAR ENDING DECEMBER 31 Actual Proj. Actual Proj. Actual Proj. Actual Proj. Actual P-*oj. Actual Proj. Actual Proj. Actual Proj. kpplication Of Funds ?roiect Works 26.8 29.9 30.5 143.7 80.2 261.8 103.7 235.0 261.3 - 438.C - 385. 5 - 182.S - .orks done by Urban Developers - 29.0 - 55.2 4.4 62.1 14.2 60.6 2.0 75.1 17.8 79.8 4.9 81.6 6.5 H7. 3 'rw Program - - - - - - - - 26.7 43.7 70.4 143.4 7t.6 99.4 83.5 i ?.2 Pr:or Program 72.5 15.0 - - - - TotOl Investment 99.3 73.9 30.5 198.9 84.6 323.3 117.9 295.6 290.0 118.8 327.0 223.2 470.0 181.0 272.5 13-.2 .ebt Serice ru rest on Pnor Loans 8.2 9.1 12.5 10.6 10.8 9.4 8.6 8.4 8.7 7.3 12.5 6.5 2!.1 5.6 2-.1 nt-r-i on Proj'ct Loans- Bank - 3.0 - 5.8 - 12.3 - - 20.9 - 18.5 - 23.2 - 27.6 - 2(.9 ni-e3t on Project Loans - National - 0.6 - 4.3 13.0 - 21.9 - 24.3 23.8 - 22.6 lo'al Intercst 8.2 12.7 12.5 20.7 10.8 34.7 8.6 51.2 8.7 (0.7 13.5 58.3 21.1 55.8 25.1 53.2 1mort'zation on Prior Loans 18.7 15.9 30.1 18.9 29.6 20.2 21.6 21.5 27.7 11.6 19.5 12.2 26.9 8.1 44.5 8.1 00 trort:zaion on Project Loans - Bank - - - - - - - - 5.2 4.4 1.4 9.3 15.2 9.9 15.6 10.6 Arnor. ation on Project Loans - Nationl - - - - - - - - 7.6 - 3.7 - 9.9 - 11.4 Total ,rnortization 18.7 15.9 30.1 13.9 29.6 20.2 27.s 21.5 329 23.6 31.9 30.2 42.1 27.9 60.1 30.1 TOTA. DEBT SRVICE 26.9 28.6 42.6 39.6 40.4 54.9 36.2 72.7 41.6 :14.3 45.4 88.5 63.2 83.7 85.2 83. 3 ncrease in Inventories 3.5 4.2 (5.0) 4.9 (3.1) 7.5 6.6 7.8 (!.6) 2.1 48.0 2.1 4.7 2.2 (12.8) ?.4 n.wrease in accounts receivable on Customers'. Coatrbution (1.4) 13.8 (0.8) 12.3 2.1 99.5 5.4 51.6 4.7 (11.4) - (12.2) (3.4, (13. 8 9.8 (1. .rtas*n accounsts c :vable on Sale 1.2 2.1 (0.3) 3.8 0.8 5.7 3.1 6.6 1.9 2.8 1.7 3.2 2.9 3.6 1.1 C. ther 4.7 - 1.7 - 2.2 - 6.5 - (7.0) - 2.4 - 11.4 - 15.1 - TOTii, APPLICATIONOF FUNDS 136.2 122.7 68.7 259.5 127.0 4C1.6 176.7 434.3 329.6 196.6 625.4 304.8 548.8 256.7 370.i 209.D s==... n-= ..a m. .- a ..... sma.. a =-.= mn- qI=- ==- -- Zash Surplus for Year (5.0) 9.2 24.3 1.0 42.3 (1.3) 9.0 1.2 (10.1) 27.8 (87.9) (C1.1) (20.4) 3.8 (3.5) 73.) -ash at Beginning of year (1.5) (1.5) (6.5) 7.7 17.8 8.7 60.1 7.4 69.1 8.6 59.0 36.4 (28.9) (2-L4) (4c.') 20. ( ash at End of lear (6.5) 7.7 17.8 8.7 60.1 7.4 69.1 8.6 58.0 t6.4 (28.9) (24.7) (49.3) (23.9) (52.9) 52.: Bebt Service Coverage 1.3 1.5 1.0 1.5 1.2 1.5 1.7 1.5 1.2 1.5 1.5 1.5 1.8 1.8 1.4 2.0 0 Ms~ S.C -49- ANNEX 6 CALI WATER SUPPLY AND SEWERAGE PROJECT LOAN 682-CO EMCALI Appraisal Projection and Actual Balance Sheet - 1977 (Current Col$ million) ASSETS Projected Actual Variance Fixed Assets (net) 1,931.6 2,143.5 11% Current Assets 285.9 357.9 25% Total Assets 2,217.5 2,501.4 13% LIABILITIES AND EQUITY Long-germ Debts 514.6 1,591.0 209% Equity 1,569.8 593.5 62% Total Capital 2,084.4 2,184.5 5% Current/Other Liabilities 133.1 316.9 138% Total Liabilities & Equity 29217.5 2_501.4 13% ANNEX 7 - 50 - LOAN 682-CO CALI WATER SUPPLY AND SEWERAGE PROJECT Incremental Rate of Return d/ (in Col$ million of 1969) Incremental Cost Incremental Benefits Year Investment a/ Operations b/ Total Gross c/ Net 1970 24.0 -- 24.0 -- -24.0 1971 24.3 -- 24.3 -- -24.3 1972 43.3 - 43.3 -- -43.3 1973 41.7 0.4 42.1 2.4 -39.7 1974 81.7 1.0 82.7 6.4 -76.3 1975 108.4 4.5 112.9 9.7 -103.2 1976 75.6 6.4 82.0 11.8 -70.2 1977 45.0 9.4 54.4 12.5 -41.9 1978 34.2 9.4 43.6 16.3 -27.3 1979 32.4 10.1 42.5 20.2 -22.3 1980 15.6 10.7 26.3 24.2 -2.1 1981 5.5 13.4 18.9 30.9 12.0 1982 -- 15.8 15.8 35.5 19.7 1983 -- 18.3 18.3 40.4 22.1 1984 - 20.9 20.9 45.3 24.4 1985 - 23.6 23.6 50.5 26.9 1986 -- 26.5 26.5 55.9 29.4 1987 22.0 29.4 51.4 61.6 10.2 1988 20.8 32.5 53.3 67.3 14.0 1989 10.0 35.6 45.6 73.4 27.8 1990 5.5 38.9 44.4 79.8 35.4 1991 2.3 42.4 44.7 86.4 41.7 1992 2.5 46.0 48.5 93.3 44.8 1993- - 49.7 49.7 100.5 50.8 1994 -- 53.2 53.2 107.9 54.7 1995-2020 -- 57.1 57.1 115.8 58.7 a/ Investments during 1970-77 corresponding to the first Cali Water Supply and Sewerage Project (financed by Loan 682-CO), reduced by customer contributions. Investments during 1978-81 correspond to the components of the Second Cali Water Supply and Sewerage Project (financed by Loan 1523-CO) that are neces- sary to increase sales from water production facilities built under the first project. Investments during 1987-92 correspond to similar investments as in 1978-81. but substituting initial investments corresponding to the Yumbo subproject with the additional investment required for this subproject at that time. b/ Operating costs have been diminished by the 4% contribution to the Municipa- lity and the Municipal Services. It has been assumed that non accounted-for water is going to be reduced from the current 30% to 23% in 1981. c/ Incremental benefits have been considered from 1973 on because in 1972 a substantial part of additional distribution works and house connections were completed that allowed incremental sales. Incremental benefits have been calculated assuming a 4% yearly increment of sales up to the year 1995 in which the current production capacity will be fully utilized. d/ The incremental rate of return (discount rate at which present values of costs and benefits are equal) is 5.9%. MAP I COLOMBIA CAlI WATER SUPPLY AND SEWERAGE PROJECT MAJOR ELEMENTS Built-up area High/ Low level service area boundary \f Unserved barrios Roilways -- - Existing main sewer line ø 60" Proposed primary supply mains Ø l,602 Proposed main sewer lines v Changes in diameter 42 Storage o i 2 3 Kilometers o 0 Miles -r - -Puerto Mallarino -irf Treatment Plant - -1- - R1 ca San Antonio reatment Treatmeni -' -- -I Plant I Unión de Vivienda PoDular Aguablanca II El Rodeo Pumping UI SiloéSiar.on III Siloé " IV Terrón Colorado 54 _ Paci,- Bucaramang Ocen ôBogota . Palmira Cali COLOMBIA A9 o200 FEBRUARY 1970 IBRD 2751R
Группа Всемирного банка · Project Performance Assessment Report
Colombia - Cali Water Supply and Sewerage Project
Открыть оригинал документа
Полный текст размещён на сайте публикующей организации. lawenc.com индексирует метаданные и ведёт на официальный источник.
Полный текст
Основные сведения
Организация
Группа Всемирного банка
Тип документа
Project Performance Assessment Report
Страна
Колумбия
Источник
Всемирный банк