Colombia - Medium and Small Size Cities Water Supply and Sewerage Project

Document of The World Bank FOR OFFICLAL USE ONLY ReportNo. 10818 PROJECT COMPLETION REPORT COLOMBIA THIRD MEDIUM AND SMALL-SIZE CITIES WATER SUPPLY AND SEWERAGE PROJECT (LOAN 1726-CO) ,TTT' Infrastructure and Energy Operations Department III Latin America and the Caribbean Regional Office This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  FISCAL YEAR January 1 - December 31 CURRENCY EOUIVALENTS Currency = Col $ Exchange Rate Effective January 31, 1991 US$1 = Col$568.7 ABBREVIATIONS BCH - Central Mortgage Bank EMPOS - Municipal Sanitary Works Companies EPMs - Municipal Public Companies FINDETER - Financiera de Desarrollo Territorial INSFOPAL - Instituto Nacional de Fomento Municipal MOPT - Ministry of Public Works and Transport MOH - Ministry of Health OED - Operations Evaluations Department PCR - Project Completion Report  FOR OMCIAL USE ONLY THE WORLD BANK Washington, D.C. 20433 U.S.A. iffice of Directo,wGeneril Opetatmns EvaluMtitln June 30, 1992 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Completion Report on Colombia Third Medium and Small-Site Cities Water SuRR1lv and Sewerage Project (Loan 1726-CO) Attached, for information is a copy of a report entitled "Project Completion Report on Colombia Third Medium and Small-Site Cities Water Supply and Sewerage Project (Loan 1726-CO)" prepared by the Latin America and the Caribbean Regional Office with Part II of the report contributed by the Borrower. No audit of this project has been made by the Operations Evaluation Department at this time. Attachment W-7' 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 ONLY PROJECT COMPLETION REPORT COLOMBIA THIRD MEDIUM AND SMALL-SIZE CITIES WATER SUPPLY AND SEWERAGE PROJECT (LOAN 1726-CO) TABLE OF CONTET Page No. Preface i Evaluation Summary ii PART I. PROJECT REVIEW FROM BANK'S PERSPECTIVE I A. Project Identity 1 B. Background 1 C. Project Objectives and Description 2 D. Project Design and Organization 3 E. Project Implementation 4 F. Project Results 7 G. Project Sustainability 9 H. Bank Performance 9 I. Lessons Learned 10 J. Borrower and Government Performance 11 K. Project Relationship 12 L. Consulting Services 12 M. Project Documentation and Data 12 PART II. PROJECT REVIEW FROM BORROWER'S PERSPECTIVE 13 (Conclusions of the Final Report Preparei by INSFOPAL) PART III. !TATISTICAL INFORMATION 14 1. Related Bank Loans 14 2. Project Timetable 14 3. Loan Disbursements 15 4. Allocation of Loan Proceeds 16 5. Planned and Actual Completion Dates 17 of Components 6. Project Cost and Financing 18 7. Project Results 19 8. Status of Loan Covenants 21 9. Use of Bank Resources 26 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.  TABLE OF CONTENTS (continued) Page No. ANNEXES Annex I - Chapter of Conclusions of the Final Report Prepared by INSFOPAL 27 Annex II - Comparison of Forecast and Actual Income Statements (1990-1987) 31 Annex III - Comparison of Forecast and Actual Balance Sheets (1981-1987) 32 Annex IV - Appraisal and Actual Monitoring Indicators 33 Annex V - Actual Income Statements (1982-1987) 37 Annex VI - Actual Balance Sheets 40  PROECT COMLETioN REPORT COLOMBIA THIRD MEDIUM AND SMALL-SIZE CITIES WATER SUPPLY AND SEWERAGE PROJECT (LOAN 1726-CO) PREAGE 1. This Project Completion Report (PCR) deals with the Third Medium and Small- Size Water Supply and Sewerage Project, for which Loan 1726-Ca in the amount of US$31 million was approved on June 12, 1979. The Loan was closed on June 30, 1986, three years behind schedule. The loan became fully disbursed on June 18, 1987. About US$0.5 million was cancelled in November 19s.J. About US$29.3 million of the Loan funds were on-lent by Lnstituto Nacional de Fomento Municipal (INSFOPAL) to twenty one beneficiariee. The remaining US,'.2 million financed a technical assistance program. 2. Parts I and III of the PCR were prepared by the Infrastructure and Energy Division, Country Department III, of Latin America and the Caribbean Regional Office and are based, inter alia, on the Staff Appraisal Report; the Loan and Guarantee Agreements; supervision reports; final report prepared by INSFOPAL; correspondence between the Bank and the Borrower; and internal Bank memorajida. Part II of the report reproduces the chapter of conclusion,: c;f the final report prepared by INSFOPAL. This PCR is overtiiie by fcur years. As the PCR is overdue by four years, Parts I and II were not sent to the cofinanciers and the Borrower for comments.  - ii - PROJECT CMPLET'IO REOT COLOMBIA ThIRD MEDIUM AND SHALL-SIZE CITIES WATER SUPDLY AND SEWERAGE PROJECT (LOMA 1726-CO) Proiect ObJectives 1. The main objectives of the Project were to continue strengthening INSFOPAL's operational capabilities; to contribute to the development of financially viable companies in the cities to be benefitted by the Project; to expand the water and sewerage facilities in 23 medium and small-size cities in order to provide potable water and sewerage service for approximately 472,000 and 654,000 additional inhabitants, respectively; and to reduce unaccounted-for water to not more than 25% by 1986. Implementation Experience 2. The Loan was approved by the Board on June 12, 1979, signed on November 30, 1979 and de.clared effective on February 28, 1980. Thereafter, a naior delay occurred in declaring the beneficiaries eligible for disbursement, which on average took place more than three years after Loan signature. This was mainly due the slowness in the implementation of tariff increases and in the preparation of adequate financial projections needed to determine the tariff adjustments and the financial monitoring indicators to be 'ncluded in the subsidiary loan agreements. 3. The Project was scheduled for completion by December i983, but most of the works were completed by June 1987 or three and a half years later than scheduled. The technical assistance component of the Project was only partially completed. The reasons for not completing/delaying the completion of the Project, in addition to the ones mentioned above, were: INSFOPAL and sub-borrowers' management weakness; slow procurement processing; minor changes made in the scope of the Project; and Government delay in providing the counterpart funds for the Project and in approving tariff increases for the sub-borrowers. Project Results 4. In gener&l, the Project fell far short of meeting its objectives because, even though the beneficiaries completed most of the investments originally planned, the kroject was implemented with considerable delay, which contributed to the service levels that were somewhat lower than the target established for 1986. INSFOPAL did not improve its managerial and operational capabilities and was liquidated by the Government in January 1987. The beneficiaries did not improve their financial, operational and planning capability to the level expected at appraisal. Although data are unreliable, it is apparent that the levels of water losses of the majority of beneficiaries were substantially higher than the 25% targeted for 1986. Tariff studies were not based on long-run marginal costs as envisaged at appraisal; and the training program was not implemented (paras. 15-19). 5. It has to be pointed out that the data furnished by INSFOPAL and the beneficiariea regarding the Project (cost, financial statement, and monitoring  - iii - indica'o-s) showed errors and inconsisterncies. Therefore the information prese: in Part III shows the best estimate obtained by the staff who prepared this rport. Nonetheless, the main conclusions of the report will not change. Proiect Sustainability 6. It is doubtful that the facilities built by the Project can be maintained properly in the medium term, given the poor operational and maintenance practices of the .majority of sub-borrowers, and the financial constraints that they are facing at the present time. In the long run, this situation is expected to improve once the decentralization of the sector, which has been underway since the liquidation of INSFOPAL, begins to have its intended effect (para. 23). Lesson Learned 7. The following lessons can be drawn from the implementation of the Project: (a) The Bank should not have aareed to finance the Project given the poor performance records of the borrower in the projects under execution at the time of the appraisal, and the extremely large number of weak beneficiaries. At t;.e time of appraisal, INSFOPAL clearly did not have the commitment to develop itself into a technically competent intermediary. Conditionality under the Project could not have compensated for the lack of commitment. (b) Once the Bank had decided to finance this risky project, it should have assigned enough staff time for intensive supervision. A full- time task manager, an engineer, a lawyer and two financial analysts would have been necessary at least during the first year of Project implementation. However, the allocation of such an extraordinary amount of resources would not have been compatible with Bank's practices in assigning manpower for supervision. Likewise, when significant implementation problems were discovered, the Bank's administration should have taken more decisive actions to compel the borrower to comply with conditionality. The initiation of the technical assistance component and the fulfilling of the conditions for the first disbursement were handled with extreme laxity by INSFOPAL and the sub-borrowers. (c) Financial supervision was weak despite the enormous time spent by supervision missions in reviewing financial information. This was due to a large extent to the inability of INSFOPAL to establish a system to keep track of the evolution of its finances and those of the sub-borrowers. The historical data on financial statements and their corresponding analysie should have been collected and included systematically in the supervision reports but were not. rhe supervision reports on file contain little data on finances. (d) The economic systems of INSFOPAL and most of the sub-borrowers' were weak and unreliable. The audit reports of the independent auditors were generally furnished to the Bank late and with qualified opinions. The Bank could not get INSFOPAL and the sub-borrowers to solve this problem, nor did the Project include adequate provision to deal with this problem even though the SAR recognized it.  EROJECT COMPLETION REPORT COLOMBIA THIRD MEDIUM AND SMALL-SIZE CITIES WATER SUPPLY AND SEWERAGE PROJECT (LOAN 1726-CO) PART I: PROJECT REVIEW FROM BANK'S PERSPECTIVE A. Proiect Identity Project Name: Third Medium and Small-Size Cities Water Supply and Sewerage Project Loan No: 1726-CO RVP Unit: LA3IE Country: Colombia Sector: Water Supply and Sewerage B. Backoround 1. Sector Development Objectives. At the time of Project appraisal, the Government's plans were to provide piped water to 90% of the urban population and 80% of the rural non-dispersed population (communities with over 200 inhabitants) by 1990. Its goals for sewerage service levels were 80% and 55%, respectively. In addition, the Government had explicitly recognized the need to expand services to the low-income population and expected to accomplish this by using cross- subsidization among different income consumer groups. The major constraint in ach..eving these sector goals appeared to be the absorptive capacity of the executing entities, especially in the case of small cities and rural areas. Institution building and training to help remove these constraints were to be daveloped with the assistance of the international institutions. The proposed Project included a technical assistance program for INSFOPAL and the sub- borrowers. 2. Policy Context. The Government's development strategy was aimed at expanding the water supply and sewerage services to a large number of urban and rural population. To this end the Government prepared a national plan (PLANAL) containing a program with priorities for the allocation of investments within the sector. Investments were planned to be financed from external sources, Government and INSFOPAL loans and internally generated funds of the participating utilities. In addition, the expression of interest on the part of a community in contributing to finance a percentage of the investment costs and a commitment to operate and maintain the services were taken into account in determining investment priorities.  -2- C. Proiect ObieEtives and Descrintion 3. Project Objectives. The specific objectives of the Project were: (a) to continue strengthening INSFOPAL's operational capabilitiee; lb) to contribute to the development of the Regional or Municipal Sanitary Works Companies (EMPOS), Municipal Public Companies (EPMI), and water utilities depending on those EMPOS and responsible for services in the cities to be benefitted by the Project, and to create and organize financially viable companies; (c) to provide potable water to approximately 472,000 additional inhabitants of 19 cities by 1983; (d) to provide sewerage service to approximately 654,000 additional inhabitants of 14 cities by 1983; (e) to improve and increase water production and transmission facilities in order to cover the demands of the 1990 population of the above- mentioned cities; (f) to provide urgently needed works to prevent undesirable sewage discharges into open storm drains and to _,roteCt water supply sources from further contaminationt .. (g) to reduce unaccounted-for water in these cities from ar average of 40% at the time of appraisal to not more than 25% by 1986. 4. Project Description. The Project provided f'-- (a) the construction of new facilities or optimization of existing water treatment facilities for the cities of Cartagena, Neiva, Monteria, Riohacha, Girardot, Copacabana, Rio Negro, Calarca, Aguachica, Magangue, Codazzi, Pamplona, Chiquinquira, and Espinal; (b) the construction of pumping and transmission facilities connecting the water production facilities of the city of Barranquilla to the distribution system of the city of Soledad; (c) the improvement of water intakes, pumping stations, and raw water transmission lines in 14 cities; (d) the construction of additional water storage in 16 cities with an aggregate capacity of about 60,000 m3. (e) the extension of water distribution networks in 19 cities, with an aggregate length of about 420 km; (f) the construction of about 80,000 metered house water connections in 16 cities and the purchase and installation of about 62,000 residential water meters; (g) the purchase and installation of water production meters; (h) the construction of sewers in 14 cities, with an aggregate length of about 400 km; (i) the construction of interceptors and outfalls in 11 cities with an aggregate length of about 80 km;  -3 (j) the construction of about 110,000 household sewerage connections; (k) the construction of water pumping stations in four cities and sewage pumping stations in three cities; (1) construction of stabilization ponds for sewage treatment in the city of Sabanalarga; and (m) a technical assistance program for: (i) improvements of the organizational procedures of INSFOPAL, under the responsibility of its Institutional Development Unit, using consultants as needed in the areas of planning, financing, organization, and engineering; and (ii) improvement of the operational capability of the EPMs and EMPOS (including the acquisition of office and laboratory equipment) through the use of consultant services, under the supervision of INSFOPAL's Department of Assistance to Executing Agencies. D. Project Design and Organization 5. Project preparation and design were carried out by INSFOPAL with the Bank's assistance. Individual feasibility studies for each subproject were prepared by INSFOPAL with the aid of local consultants. The feasibility studies were found satisfactory by the Bank. These studies recommended the least-cost solution in each case; however, in two cities, Dos Quebradas and Soledad, additional studies were deemed necessary to select the least-cost alternative for some Project components. 6. The Project was fairly well prepared and its timing was in general appropriate. However, the appraisal team did not conduct an on-site evaluation of the proposed beneficiaries and the works prepared by them; instead it relied on the assessment made by each beneficiary and INSFOPAL.1 Some of the Bank staff who reviewed the SAR expressed doubts about the validity of some of the information collected, and recommended a post-appraisal mission, which was never undertaken. Later, during Project implementation, the inconsistencies in the information provided by the beneficiaries became clear and it became necessary to review the subproject description and prepare new financial projections for the majority of beneficiaries. Another drawback of the Project was that the technical component aimed at improving INSFOPAL in the areas of administration, organization, financing and planning was only generally defined, with no precise procedures established for carrying out this task. For the beneficiaries, this component originally included studies to improve their accounting system and to create an information system on their services and performance. The implementation schedule for this component was also $X optimistic. It called 1 The number of beneficiaries ultimately was reduced from 23 to 21; three of them, Copacabana, Calarca and Dos Quebradas, decided not to participate in the Project and were replaced by Ipiales and Popayan, and Tumaco was cancelled because it did not comply with the conditions for the first disbursement.  -4- for introducing improvements in the beneficiaries' accounting system to permit full disclosure of the statement of sources and applications of funds, asseto revaluation and tariff studies for each beneficiary by October 31, 1979. This was highly unrealistic in view of the fact that the beneficiaries lacked a proper accrual accounting system. Later, in December 1981, the scope of the technical component was increased to include studies on billing and collection, operation and maintenance, budg3t preparation and control, internal control, and planning. 7. The first Project Brief states among the main issues the reorganization of INSFOPAL, its limited absorptive capacity to carry out the Project, slow tariff adjustment procedures, and delays in the Government's counterpart fund contributions. The Project included a technical assistance component and some covenants to deal with thene issues, but actions either were not carried out or were carried out with substantial delays (para. 9). 8. The Project's cost was estimated at US$71.9 million, with US$31.0 million of foreign costs. The Bank loan was to finance the foreign costs. The Government was to finance US$31.7 million of local costs. The balance in the amount of US$9.2 million was to be financed by the beneficiaries. E. Proiect ImDlementation 9. Loan Signature, Effectiveness and Project Startup. Loan 1726-CO was approved by the Board on June 12, 1979 and was signed on November 30, 1979, with some delays. There was also a minor delay in declaring the Loan effective, which took place on February 28, 1980. However, a major delay occurred because the sub-borrowers delayed compliance with the conditions for the first disbursement for about three years. These conditions included presentation of a plan for institutional improvement, signature of the respective subsidiary loan agreement, 2 and approval and implementation of tariff increases. The delay was mainly due to the slowness in the implementation of tariff increases and preparation of adequate financial forecasts required to determine the financial monitoring indicators to be included in the subsidiary loan agreements.3 Even though the preparation of the Project and bidding process continued and it was possible to begin some of the works with the part of the funds provided by the Government, the delay in the fulfillment of the conditions for first disbursement significantly affected the startup of the Project. 2 The subsidiary loan agreements included provisions regarding financial covenants and monitoring indicators, insurance of assets, procurement, etc. 3 As of December 1982 only five sub-borrowers had been declared eligible for disbursement: Sabanalarga, Soledad, Valledupar, Rio Negro, and Ibague. On June 13, 1983 thirteen more sub-borrowers were declared eligible: Neiva, Pasto, Sincelejo, Girardot, Codazzi, Cartago, Espinal, Panplona, Monteria, Riohacha, Magangue, Aguachica, Cartagena y Chiquinquira. Ipiales and Popayan were still working to fulfill the conditions and would eventually subst: ute Calarca, Dos Quebradas and Copacabana. Tumaco was cancelled from the Project.  -5- 10. Implementation schedule. The Project was originally scheduled to be completed by December 1983, but the majority of the works under the Project were not completed until June 1987 or three and a half years later than scheduled. The remaining facilities were completed or planned to be completed by December 1988, 4 about five years later than the target date (Table 5, Part III). The technical assistance component of the Project was only partially completed (Table 7, Part II'). The main reasons for not completing or delaying the completion of the Project, in addition to the ones already mentioned in paragraph 9, were as follows: (a) The procurement process was slow, mainly because INSFOPAL and the sub-borrowers did not have sufficient experience in lealing with the Bank's procurement procedures--which differ in many respects from Colombian law (para. 11). (b, INSFOPAL and the sub-borrowers lacked good management or sufficient number of qualified staff to handle the project implementation. (c) INSFOPAL lacked commitment to the Project and the Bank was unwilling to press the Institute to fulfill its commitments. Part A of the Project included studies for the institutional development of ISSFOPAL and the sub-borrowers. These studies- -which were important to improve the organization, the accounting systems, operation and planning of INSFOPAL and the sub-borrowers--were handled poorly by INSFOPAL, which did not staff the department in charge of this task adequately, did not train this staff and gave low priority to the studies. The Bank, for its part, did not supervise this component closely and did not take strong remedial actions to encourage its acceleration. (d) The Government was slow in providing the counterpart funds for the Project and in approving tariff increases, which were necessary for the sub-borrowers to comply with the financial viability criteria stipulated in the Loan Agreement as conditions for disbursement. (e) Changes were made in the scope of the Project to include other cities (Ipiales and Popayan) and to modify the description and scope of the components of the majority of subprojects. 4 A mission to INSFOPAL in April 1988, to provide advice about the preparation of the PCR revealed that four beneficiaries (Monteria, Soledad, Sincelejo and Girardot) had not completed their works. Furthermore, a visit to Soledad indicated that the works had been halted due to financial constraints, with much of the equipment and materials exposed to the elements and in danger of deterioration. The Bank then asked the Government and INSFOPAL to complete the works of these beneficiaries expeditiously, which they agreed to finish by December 1988.  -6- (f) The Bank was slow in advising INSFOPAL on the preparation of the subsidiary lending agreements. The Bank sent the lawyer in charge of the Project in June 1981 to advise INSFOPAL after voluminous correspondence had been exchanged on this matter. It also seems that there was not full agreement between Bank staff on the content of the subsidiary loan agreements. For instance, a Bank lawyer accepted as satisfactory some of these agreements, which did not satisfy totally the Loan Officer. This required further analysis and time to be settled. 11. Procurement. The procurement process was generally slow, mainly because: (a) INSFOPAL staff responsible for the procurement process were not experienced. Although INSFOPAL had borrowed from the Bank before, the project staff were relatively new and not familiar with the Bank's guidelines. The principal deficiencies were: (i) bid announcements, which did not receive international dissemination; (ii) inadequate time allowed for preparating tenders; (iii) contracts not awarded to the lowest evaluated bidder; and (iv) contracts signed without prior Bank approval; and (v) INSFOPAL's repeated requests to disburse from the loan account in foreign currency to make payments to local suppliers and contractors, which is contrary to the Bank's guidelines. This required a long time to be settled. (b) The Instituto de Comercio Exterior (INCOMEX) was slow in approving import licenses. 12. Proiect Costs. The estimated cost of the Project at appraisal was about US$71.9 million. The final cost was about US$65 million, or a cost underrun of about 9% (Table 6, Part III). This is explained basically by the lower prices obtained in the international and local competitive bidding. 13. Disbursements. The Loan's estimated and actual disbursements (Table 3, Part III) reflect the Project's slow implementation. The original closing date was June 30, 1984, but the Loan was not closed until June 30, 1986, after a two year extension. However, the Bank honored disbursement applications received later than this date, and the final disbursement was made on June 18, 1987. The disbursement percentage for civil works was raised from 37% to 51% in the course of project implementation to help resolve the counterpart funding problem. 14. Loan allocations. The original and actual disbursements by categories are shown in Table 4 of Part III. The original allocations were revised in February and November of 1984 to reflect changes in the Project scope, the creation of a special and a Project account and actual implementation needs.  -7- F. Proiect Results 15. Meeting Project Objectives. In general, the Project fell far short of meeting its objectives. Even though the beneficiaries completed most of the facilities originally planned, the Project was implemented with considerable delays and the service levels of water and sewerage were somewhat lower than the target established for 1986 (Annex 4). 5 INSFOPAL did not improve its managerial and operational capability and was liquidated by the Government by Decree No.77 of January 1987.6 The beneficiaries did not improve their financial, operational and planning capability to the level expected at appraisal. The level of water losses was generally higher than the 25% targeted for 1986. 7 The tariff studies were not done with the scope envisaged at appraisal and the tariff increases were prepared based on gross cash-flow projections that did not take into account the long-term economic cost of meeting the increasing demand. Lastly the training was not undertaken. In retrospect, the lack of commitment to the Project of INSFOPAL and the beneficiaries was the main cause of failure to meet the objectives fully. This lack of commitment was derived, to a large extent, from excessive political interference. 16. Physical Results. The physical targets of the Project in terms of water supply and sewerage facilities were to a large extent achieved, but with substantial delays (Table 5, Part III). It seems that the delay affected somewhat the service level in the majority of the beneficiary cities. 17. Financial Performance. Annexes 1 and 2 showed actual and appraisal income statements and actual balance sheets for INSFOPAL for 1979-1986. INSFOPAL's financial performance during that period was unsatisfactory. Net income was negative throughout the period because its revenues, generated from interest and charges for technical and administrative services provided to the operating companies, did not cover its increasing operating costs and the financial charges on its borrowings. To cover the financing gap and to obtain funds for investment, the Institute was forced to depend on increasing Government contributions, which normally arrived with extreme tardiness. INSFOPAL also had a poorly performing portfolio because it was unable to impose financial discipline on its borrowers. As of 1986, for instance, 37% of the accounts receivable on loans were overdue more than 90 days and the balance of accrued 5 As stated in the evaluation summary, the validity of the information provided by INSFOPAL and the beneficiaries is questionable; therefore, the conclusion regarding the service levels has to be taken with caution. No sufficient information was available to compare the results as of 1983, the year scheduled for Project completion. 6 When first announced, the process of liquidation was to last until December 1989, but was later advanced to September 1989. 7 Water losses include leakages, errors in measurement due to insufficient production and consumption meters, and theft. There was no information available on the weight assigned to these categories in assessing results.  - 8 - income from loans represented 88% of the annual income from loans, which means that the operating companies were paying interest to INSFOPAL with a delay of about 320 days. 18. The financial performance of the sub-borrowers, except for a few of them, was also unsatisfactory. Their operating ratios were generally above the target established in the SAR, as tariff increases did not keep pace with the increases in operating expenses; with few exceptions, the net income was generally negative for all sub-borrowers and the current ratio was less than one, reflecting liquidity problems (Annexes 4 and 6). The sub-borrowers accumulated high levels of accounts receivable and were in default paying their debt service to INSFOPAL. 19. During appraisal the Bank had detected deficiencies in INSFOPAL's and the sub-borrowers'accounting systems but underestimated the magnitude of the problem. Accordingly the Project included only a general component to tackle this situation. Little progress was made on this, which is demonstrated by the fact that during the implementation period (1979-87), all of INSFOPAL's audit reports were furnished to the Bank late and all, except the one for 1985, were issued with adverse opinions, disclaimers or qualified opinions by the independent auditors. The same tardiness and qualified opinions generally characterized the reports on the majority of sub-borrowers. This situation makes the financial data unreliable and therefore they have to be interpreted with caution. In any case, this will not change the basic conclusion that the financial situation of INSFOPAL and of the majority of sub-borrowers was unsatisfactory. 20. The Loan included rate-of-return and self-financing ratio covenants and other monitoring indicators that were never enforced. To a great extent this was due to the inability of INSFOPAL and the sub-borrowers to prepare adequate data on financial covenants, and also the inability of the Bank to supervise the Project adequately. Even though the Bank devoted considerable time to supervision, this was not enough given the large number of sub-borrowers (21), which generally had weak accounting and administrative systems INSFOPAL, which was supposed to undertake the bulk of this work, was not prepared to fulfill this function, and did not staff the corresponding departments and did not train them properly. It also has to be mentioned that the Loan included too many covenants and monitoring indicators, underscoring the concern of the appraisal mission with the poor finances of INSFOPAL and the sub-borrowers. In a situation like this, it would have been better to rely on a few selected easy-to-monitor covenants. The rate of return and the self-financing ratios are generally difficult to monitor, because of the assets revaluation complication and the difficulty in preparing the statement of sources and applications of funds when dealing with weak entities. It seems that it would have been better to use the working ratio a as the main revenue covenant, which is easier to monitor and was suitable in the circumstances. 8 Ratio of the cash operating expenses to operating revenues.  -9- 21. The technical assistance component of the Project, which was the provision included in the Loan Agreement to improve INSFOPAL and the sub- borrowers institutionally, was scheduled to be provided at the beginning of Project implementation. INSFOPAL did some studies on its organization, code of functions and budget preparation in the early years of the Project, but very little was implemented. A comprehensive study of the beneficiaries' administration, finances and operation was started as a pilot study with three of them in April 1984, about four years later than planned, because of cumbersome consultant selection procedures, but mainly due to the lack of INSFOPAL's commitment to the studies. The Bank urged the Institute to accelerate the preparation of the studies, but it responded slowly. The Bank could have been tougher regarding compliance with conditionality. It might have exercised its right under the Loan Agreement to adopt remedial actions like cancellation of many more subprojects, after it became clear that this vital component of the Project was not progressing satisfactorily and that INSFOPAL did not show its real willingness to fulfill its commitment.9 22. Impact of the Project. The Project did not have a favorable impact on INSFOPAL's development. It was liquidated by the Government in 1987 due to its institutional deterioration. The Project did not fully meet the appraisal expectation of the sub-borrowers either. Even though the majority of sub- borrowers completed the physical components and were able to increase somewhat their tariffs as a result of the Project, the financial and institutional development fell short of what was expected at appraisal. Few sub-borrowers improved their operating and financial conditions, and some are now worse off than they were at the beginning of Project implementation (Annexes 4 to 6). The water and sewerage service levels at the end of 1986 were generally lower than those expected at appraisal due to the implementation delay. G. Proiect Sustainability 23. It is doubtful that the facilities built by the Project can be maintained properly over the medium term, given the poor operation and maintenance practices of the majority of the sub-borrowers, and the financial constraints that they are facing due to inflation, devaluation, rising operating costs, low tariff increases and high water losses. In the long run this situation is expected to improve, once the decentralization of the sector (underway since the liquidation of INSFOPAL) begins to take its intended effect. H. Bank Performance 8.01 The Bank's participation in the Project was inadequate. It decided to take the risk of financing the Project, even though at the time of the appraisal INSFOPAL was executing two other Loans with poor performance records, weak organization, a shortage of technically qualified personnel and personnel instability due to frequent changes in top management. The problem was compounded by the large number of sub-borrowers, most of whom with weak 9. The Bank did cancel two subprojects.  - 10 - management. The Bank also failed in the four aspects: (a) It under estimated the staff input requirement to supervise a complex project like this, which would probably have required a full-time task manager, an engineer, a lawyer, and at least two financial analysts in the first year of Project implementation. (b) It did not press INSFOPAL sufficiently to carry out the studies for the institutional development expeditiously, once it noticed the slowness with which the selection process of the consultant was going ahead. In hindsight, the appointment of this consultant should have been a condition of negotiation. (c) It was not sufficiently tough with in respect to compliance with conditionality. And (d) it delayed the declaration of eligibility for the first disbursement to those sub-borrowers that fulfilled these conditions early, because at the beginning the Bank had the intention to declare the eligibility of all the sub-borrowers simultaneously. In this sense, it lost the opportunity to use the good implementation experience of some sub-borrowers as an example to be followed by the rest. I. Lessons Learned 24. The principal lessons learned are summarized below: (a) The Bank should not have agreed to finance the Project, given the poor performance records of the borrower in the projects under execution and the extremely large number of weak beneficiaries. (b) once the Bank had decided to finance this risky project, it should have assigned more staff time for intensive supervision. A full-time task manager, an engineer, a lawyer and two financial analysts would have been necessary at least during the first year of Project implementation. This may have allowed the Bank to monitor the development of the Project in a timely way. However, the allocation of such an extraordinary amount of resources would not have been compatible with Bank practices. Likewise, when significant implementation problems were discovered, the Bank's management should have taken decisive actions to seek remedies. In the case of the Project, the initiation of the technical assistance component and the fulfilling of the conditions for the first disbursement were handled with extreme laxity by INSFOPAL and the sub-borrowers. (c) Financial supervision was weak notwithstanding the enormous amount of time spent by supervision missions in reviewing financial information. This was due to a large extent to the inability of INSFOPAL to establish a system to keep track of the evaluation of its finances and those of the sub-borrowers. The historical data on income statements, balance sheets and sources and applications of funds should have been collected and included systematically in the supervision reports but it  - 11 - was not easy to do so. The supervision reports available in the files contain only few data on finances. (d) INSFOPAL and most of the sub-borrowers' accounting systems were weak and unreliable. The audit reports were generally furnished to the Bank late and with qualified opinions of the independent auditors. The Bank did not press INSFOPAL to solve this problem, nor did the Project include adequate provision to deal in depth with this problem even though the SAR recognized it. J. Borrower and Government performance 25. INSFOPAL's performance was generally unsatisfactory throughout Project implementation even though the beneficiaries completed most of the physical components of the Project, with about three and a half years delay. INSFOPAL performance was poor in the following aspects: the technical assistance components were not completed, revenue covenants were not met, net income was negative in all the period, and portfolio performance was deficient. INSFOPAL's management was weak and faced frequent changes of its top manager. There were five general managers (director general) between 1979 and 1986. INSFOPAL's inability to retain sufficiently experienced and qualified managers and personnel was due to: (a) low salaries compared to other Government entities, as well as to the private sector; and (b) appointments to management, and to a large extent to professional and even support staff positions, on the basis of political expediency rather than qualifications. 26. The Government for its part wanted to have in INSFOPAL an adequately run institution responsible for coordinating, supervising, and financing sector activities. However, it delayed the release of counterpart funds and did little to create an atmosphere of autonomy in which to operate INSFOPAL as a business- like institution. Frequent changes in top management led to changes even to the lower level of management. As a result, INSFOPAL remained a largely ineffective bureaucracy with generally poor staff resources. Under these circumstances the Government opted for liquidating the Institute in 1987 and implementing the process of sector decentralization that had begun in 1986. 27. The decentralization policy was launched at the beginning of the 1980s with the popular election of mayors. Under Law 12 of January 1986, the central Government transferred the responsibility for decision making and administration to municipalities, which were expected to progressivelv assume responsibilities for financing and management of municipal infrastructure, including water supply and sewerage. The adjustments in the sector's national organizations, mandated by Decree 77 of January 1987, established that the National Planning Department (DNP) continue to have responsibility for planning and coordinating sector and overall development policy. The sector's overall management was transferred from the Ministry of Health (MOH) to the Ministry of Public Works and Transport (MOPT). The Central Mortgage Bank (BCH), through its Urban Development Fund, is responsible for mobilizing domestic and external resources for sector investments and evaluating, approving, and supervising sector projects. Law 57 of 1989 expanded the Urban Development Fund into a self-  - 12 - standing apex agency, the Financiera de Desarrollo Territorial (FINDETER), which became operational in May 1990. FINDETER channels funds by rediscounting subloans originated by commercial banks to local governments or municipal enterprises for urban infrastructure investments, including water and sewerage. The Bank was satisfied with the sector's organization following INSFOPAL liquidation. However, it remains to be seen if political interference with sector institutions can be minimized and efficiency increased. K. Proiect RelationshiD 28. The Bank's relationship with the Government, INSFOPAL and the sub- borrowers was somewhat strained. INSFOPAL staff who participated closely with the Bank's supervision missions resented the Bank and believed that it was instrumental in the liquidation of the Institute. L. Consulting Services 29. Fifteen consultants participated in carrying out the technical assistance component of the Project, and at least one consultant per each sub- borrower, participated in the engineering design of the water and sewerage facilities and in the supervision of construction contracts. INSFOPAL, in general, was satisfied with the work of the consultants. However, it considered that some of the firms that worked on the technical assistance components did not have enough experience with water supply utilities and that this delayed the execution of the works. M. Proiect Documentation and Data 30. The Loan Agreement and the Staff Appraisal Report were adequate for Project implementation. The Loan Agreement provided flexibility for introducing changes in Project description, which required only regional approval. However, it included too many covenants, key monitoring indicators and reports to be met/submitted by INSFOPAL and the sub-borrowers. This was difficult to supervise and a heavy load for such weak institutions. INSFOPAL prepared a project completion report, but the information was not fully consistent.  - 13 - PROJECT COMPLETION REPORT COLOMBIA THIRD MEDIUM AND SMALL-SIZE CITIES WATER SUPPLY AND SEWERAGE PROJECT (LOAN 1726-CO) PART II: PROJECT REVIEW FROM BORRWER'S PEEECTIVE I 1. Annex I of the report reproduces the "Conclusions" chapter of the final report prepared by IVSFOPAL (an unofficial translation into English is attached). Its main points are the following: (a) Seventeen beneficiaries completed satisfactorily the works included in the Project, but with a time overrun of two years due to the delays on the part of the Government to provide sufficient counterpart funds and to a slow procurement process. (b) Four boneficiaries (Monteria, Soledad, Sincelejo and Girardot) had not completed their works at the time of the issuance of the final report prepared by the Borrower (July 1988), and INSFOPAL asked the Government to provide the funds needed to complete these works. (c) The Project benefitted a large number of people. (d) The institutional components of the Project were only partially completed. (e) The training undertaken was relatively small. Lastly (f) The tariffs of the beneficiaries were increasing in mid-1988 in line with inflation, due to some extent to the participation of INSFOPAL. 1 This PCR is overdue by four years, and according to the new PCR Guideline. the Operations Evaluation Department (OED) will decide whether to send this report to Government officials for comments.  - 14 - PE-T CT PO EPT COLOMBIA THIRD MEDIUM AND SMALL-SIZE CITIES WATER SUPPLY AND SEWERAGE PROJECT (LOAN 1726-CCO) PART III: STATISTIIAL INFORM&TIO *.Rlt.dank beane --s- (in us$ toar of Disburse- Number and Tttle _Sllon) Approval Borrouer ment Purpose Status 1. 860-CO 9.1 1972 INSFOPAL 902 11 Medium-Sixe water Supply Completed 2. 1072-CO 27 1975 INSFOPAL 91S 1/ 2nd Multi-City Water Com led _________________________ ___________ ~~~supply & ev r83i 3. 1726-CO 31 1979 INSFOPAL 991 it 3rd Water Supply & Completed I I I I Sewerage ~~__________ in 6/87 AL The undiobursed asount was cancelled. 2. Project Ttmetable _ Item Date Planned Date Actual identification __ 6/75 Preparation __ 7/77 Appraisal Hission __ 7/78 Loan Negotiations __ 5/7-10/79 Board Approval _ 6/12/79 Loan Signature _ 11/30/79 Loan Effectiveness __ 2/28/80 Loan Closing 6/30/84 6/30/86 Project Closing 12/83 6/30/87 1/ 1/ The final disbursement took place nearly a year after the Loan was formally closed, going beyond the six-month grace period that was Bank practice until 1988. Since 1989 the Bank has tightened its practices and allows only four months grace to complete disbursements.  - 15 - PROJcT COMPLETION REPORT COLOMBIA THIRD MEDIUI AND SMALL-SIZE CITIES WATER SUPPLY AND SEWEACE PROJECT .OANl 1726-Cm) Cumulative Estimated and Actual Disbursements (in million of USS) Bank Fiscal Year Estimated Actual Actual as Z of and Semester Cumulative Cumulative Estimated 1979-1980 December 31, 1979 0.2 0.0 0 June 30, 1980 2.1 0.0 0 1980-1981 December 31, 1980 4.1 0.0 0 June 30, 1981 10.4 0.0 0 1981-1982 December 31, 1981 17.0 0.0 0 June 30, 1982 22.2 0.0 0 1982-1983 December 31, 1982 27.4 1.3 5 June 30, 1983 29.3 2.9 10 1983-1984 December 31, 1983 31.0 5.9 19 June 30, 1984 31.0 10.7 35 1984-1985 December 31, 1984 31.0 14.3 46 June 30, 1985 31.0 19.3 62 1985-1986 December'31, 1985 31.0 2S.6 83 June 30, 1986 31.0 28.5 92 1986-1987 December 31, 1986 31.0 30.1 97 June 30, 1987 31.0 30.51 98 Date of Final Disbursement: 6/18187 1/ US$460,000 was cancelled because one of the beneficiaries, Tumaco, did not comply with the conditLons for the first disbursement.  - 16 o RJECT c E E COLOMBIA THIRD MEDIUM AND SmALL-SIZE CITIES WATER SUPPLY AND SEWERAGE PROJECT 4. Allocation of Losn Procoeds (in USS) Origtnol Actual Category Allocation Disbursements (1) Pert A of tho Project 860,000 1,249,211.30 (2) Part 9 of the Project: (a) Cartagens 4,600,000 4,963,811.78 (b) Ibagu- 1,380,000 1,716,622.76 (t) Valledupor 1,670,000 1,252,246.12 'd) Posto 1,810,000 2,077,567.47 (a) "*ove 1,320,000 1,865,360.07 (f) Montaria 4,970,000 4,687,192.10 (f) Sincalejo 2,880,000 3,280,938.26 (h) Soledad 1,730,000 1,852,840.88 (1) Riohacho 330,000 291,304.06 (j) Qirardot 590,000 643,461.32 (k) Cartago 630,000 476,136.05 (1) Copecabans 1/ 810,000 Popoyan 2/ 1269650.71 (m) Rio Negro 1,160,000 616,891.01 (n) Do* Quebradas 1/ 910,000 0.00 (o) Calarce 1/ 310,000 Ipiafes 2/ 340420.29 (p) Aguachica 830,000 482,638.56 (q) Tumaco 1/ 460,000 0.00 (r) Magangue 790,000 835,498.14 (a) Codosit 420,000 219,693.37 (t) Pamplona 890,000 792,566.61 (u) Chiquinquira 680,000 423,252.93 (v) Sabanolatga 480,000 500,997.46 (w) Esp;nol 920,000 317,107.48 Adjustments -2177.79 31,000,000 30,140,000 3/ 1/ Component eliminated from the Project 2/ Component added to the Project. a/ A reminder of US$ 460,000 was cancelled from the Poject account, beou Tumaco did not comply with the conditions for the first dlsbgreement.  - 17 - PROJECT COMPLETION REPORT COLOMBIA THIRD MEDIUM AND SMALL-SIZE CITIES WATER SUPPLY AND SEWERAGE PROJECT (LOAN 1726-CO) 5. PLANNED AND ACTUAL COMPLETION DATES OF COMPONENTS Appraisal Actual Months of Project Components Completion Completion Delay Originsl Cartagena Dec-83 May-87 42 Ibague Dec-83 Dec-88 37 Valledupor Dec-83 May-86 17 Posto Dec-83 Jun-87 43 Ne1vn Dec-83 Dec-88 37 Monteria Dec-83 DOec-88 1/ 61 Sincelejo Dec-83 Dec-88 1/ el Soleded Dec-83 Dec-88 1/ 61 Rlohacha Dec-83 Dec-86 24 Girardot Dec-83 Dec-88 1/ 61 Cartago Dec-82 Dec-8S 37 Rio Negro Dec-83 Dec-87 49 Aguachicn Dec-83 Dec-84 12 Magengue Dec-82 Dec-86 49 Codazzi Dec-83 Dec-84 12 Pamplons Dec-83 Dec-86 37 Chiquinquira Dec-83 Dec-88 37 Sabenalorga Dec-82 Dec-84 24 Espinal Dec-83 Dec-86 37 Technical Assistance D.c-S1 Dec-86 2/ 61 Works added to the Project __________________________ Iplales Dec-86 Popeysn Jun-87 ------------------------------_---------- 1/ Expected completion date. 2/ Partially completed.  PROJECT CO0PLETIQN REPORT COLOMBIA THIRD MEDIUM AND SMALL-SIZE CITIES (Filo:Cost1?26) WATEiR SUPPLY AND SEWERAGE PROJECT 6. Project Cocet and Pmnanoing (maillion of USS) Appesisal and Aetual Project Coats Approlsel--- ---Estimteod-- X ODfference Local For.lgn Total Local Foreign Total from Appraiasl I - Project Components Plannod and Executed Cartagena 6.73 4.60 10.33 4.62 '.61 9.03 -12.83 lbegue 1.97 1.37 3.34 2.46 1.87 4.13 23.52 VIagedupe 2.80 1.66 4.26 2.72 1.S7 4.29 0.71 P dst* 2.23 1.81 4.04 2.41 1.96 4.37 8.17 Naive 1.76 1.30 3.08 1.99 2.14 4.13 34.08 Montcnle 6.86 4.97 11.86 4.99 4.16 9.14 -22.90 Sincelejo 2.97 2.86 5.83 3.69 3.49 7.18 23.21 Sol*ded 2.64 1.73 4.37 0.76 1.3U 2.11 -51.77 Riohache 0.37 0.33 0.70 0.93 0.22 1.16 03.92 ic;ardot 0.77 0.69 1.36 0.81 0.58 1.19 -12.69 Cartago 1.01 0.03 1.64 0.98 0.31 1.29 -21.46 Rio Negro 1.09 1.15 2.24 0.99 0.82 1.91 -19.00 Aguaehics 0.93 0.83 1.76 0.89 0.90 1.79 1.69 Magangue 1.02 0.79 1.61 0.80 0.61 1.61 -11.01 Codazzi 0.43 0.42 0.85 0.36 0.32 0.68 -20.31 Pamplono 0.96 0.69 1.85 1.09 0.63 1.72 4.47 Chiquinqulra 0.77 0.66 1.45 0.56 0.64 1.09 -25.07 Sabanaloga 0.94 0.46 1.42 0.69 0.31 0.89 -37.07 Espinnl 1.37 0.92 2.29 1.16 0.96 2.10 -8.49 Technical. Assistanee Program 0.54 0.86 1.40 0.36 1.4a 1.76 27.26 TOTAL COSTS OF COMPONENTS 37.00 26.67 e6.6 7 32.81 26.06 01.47 -6.40 PLANNED AND EXECUTED Works deleted freo Project Copaecabaes 0.60 0.61 1.41 Do0 Qwjebradaes 1.57 0.91 2.46 Calares 0.43 0.31 0.74 Tuawco 1.14 0.46 1.60 Works added to the Project Ipialss 0.53 0.41 0.96 Popayon' 1.22 1.48 2.70 TOTAL PIROJECT COSTS (netuding works added to the Projeot) 40.94 30.96 71.90 84.57 a0.54 65.11 -9.44 Project Financing IsBiD . .3 0.00o 0.96 80.6 0.00 30.54 80.54 Government and MSFOPAL 81.70 0.00 31.70 27.57 0.00 27.5T Beneficiaerle 9.24 0.00 9.24 7.00 0.00 7.00 40.94 80.9 71.90 84.57 30.54 6.11 - = 3== 5=5 1/ Excluding Interect during construction. 2/ Price and physical contingencies hive bien distributed on pro-rate bealsI  - 19 - COLOMBiA THIRD MEDIUM AND SMALL-SIZE CITIES WATER SUPPLY AND SEWERAGE PROJECT 7. Project Results 1/ - Project Studies ---------~~~ - -------- -- - --------------- Impact of Study Purpose as Definod In SAR Status Study Institutional To improve the administ-ration, Development technical capacity and opora- tion of INSFOPAL (a) Revilw INSFOPAL's Met Not lmplemeted code of function* (b) INSFOPAL's Not met manpower needs (c) Improve INSFOPAL Partially Met Thr-e wre minor and beneficiaries Improvements in the accounting accounting systems. system 2/ (d) Prepare a five-year Not met Investment plan (e) Establish a solid wastes unit a, Water supply and to meet revenue requiro(iento Partially met. Finances of beneficiaries sewersae tariffs and reflect es closely as Prepared based on had some Impiovement possible the economic costs revenue requirements. of meting increasing demand Training plan and to ikprove INSFOPAL's Plan was prepared Not Implemented Implementation efficiency 1/ Key Indicators by benefic'ary ore shown In Annex 8. There was no 'nformatlon avallable to estimate the economic rate of return of the Project. 2/ The scorp of the beneficiaries' technical component was incrsesed to includ, studies an billing and collection, operation and olntonance, budget preperation and control, Internal control, and planning. A dianootic wos done by consultants and DNSPOPAL's s*taff for ten cities, and a plan to deal with tho principal problos was prepared, but little or no advance was done In Implementing It 8/ Information was not available.  he W~~~~~p,-6 -fgM9 me.~ ~~~~~~~~~~~~~~~~~~~~~~~ a 41 -$""-t w.fuM eatep. v -h- pq -e*e -o h h 2 2 ' ~~~~~~~~~~~~~~lS t . u I r 6010--orme go-O $ hvA * hx X n e * U S 6CM *u - A u bx Om so Iloilo am pt. hhrnatt p __v'm v