Rapport d'achèvement de la mise en œuvre: Projet d'approvisionnement en eau de Port-au-Prince en Haïti (Crédit 2052-HA)

Documentof THEWORLDBANK FOR OFFICIALUSE ONLY ReportNo. 19981 IMPLEMENTATION COMPLETION REPORT HAITI PORT-AU-PRINCE WATERSUPPLYPROJECT (CREDIT2052-HA) December21, 1999 Finance,PrivateSectorand Infrastructure Department CaribbeanCountryManagementUnit Latin Americaand CaribbeanRegion This document has restricted distribution and may be used by recipients only in the performanceof their official duties. Its contents may not otherwisebe disclosedwithout World Bank Authorization. Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized CURRENCY UNIT: Gourde EXCHANGE RATE FOR US$1.00 (Yearly Average) 1989 : 5.00 1995 : 15.1 1990 : 5.00 1996 : 15.7 1991 : 6.00 1997 : 16.7 1992 : 9.80 1998 : 16.8 1993 : 12.8 1999 : 16.9 1994 : 15.0 WEIGHTS AND MEASURES Metric System FISCAL YEAR October 1 to September 30 ABBREVIATIONS AND ACRONYMS AFD Agence Francaise de Developpement CAMEP : Centrale Autonome Metropolitaine d'Eau Potable IDA : International Development Association IDB : Inter-American Development Bank POCHEP : Projet de Poste Communautaire d'Hygiene et d'Eau Potable PPF : Project Preparation Facility SNEP : Service National d'Eau Potable UFW : Unaccounted For Water Vice President: David de Ferranti Country Director: Orsalia Kalantzopoulos Sector Director: Danny Leipziger Task Manager: Emmanuel Njomo FOROFFICMIL USE ONLY Table of Contents Page No. Preface ....................................... i Evaluation Summary ....................................... ii PART I: Project Implementation Assessment..................................... 1 A. Background .I B. Project Objectives.2 C. Achievement of Project Objectives.3 D. Major Factors Affecting the Project .9 E. Project Sustainability.1 F. IDA Performance.12 G. Borrower Performance .13 H. Assessment of Outcome.14 t. Future Operations.15 J. Summary of Findings and Key Lessons Learned .15 PART HI:Statistical Annexes.................... . 18 Table 1: Summaryof Assessments.19 Table 2: Related Bank Credits.20 Table 3: Project Timetable .20 Table 4: Credit Disbursements: Cumulative Estimated and Actual .21 Table 5: Key Operational Performance Indicators.22 Table 6: Key Financial and Institutional Performance Indicators.23 Table 7: Number of Water Connections.24 Table 8: Project Scope and Costs .25 Table 9: Project Financing.28 Table 10: Allocation of Credit Proceeds ................................. 29 Table 11: Economic Rate of Return................................. 30 Table 12: Financial & Economic Evaluation................................ 31 Table 13a: Etat des Resultats................................. 32 Table 13b: Key Elements of Income Statements ............................. .33 Table 14: Bank Resources: Missions................................. 34 Table 15: Bank Resources: Staff Inputs ................................. 35 Table 16: Status of Legal Covenants ................................. 36 Appendix: Borrower's Contribution................................. 38 Map: IBRD No. 21564 This documenthas restricted distributionand may be used by recipients only in the performance of their officialduties. Its contentsmay not otherwisebe disclosedwithout World Bank Authorization. IMPLEMENTATIONCOMPLETIONREPORT HAITI PORT-AU-PRINCEWATER SUPPLY PROJECT (CREDIT 2052-HA) PREFACE 1. This is the Implementation Completion Report (ICR) on the Port-au-Prince Water Supply Project in Haiti. It was supported by Credit 2052-HA in the amount of SDR 15.2 million (US$20 million), which was approved on June 27, 1989 and became effective on February 14, 1990. The last disbursement took place on November 29, 1999 at which time an undisbursed balance of SDR 0.38 million was cancelled. 2. The Credit was suspended in October 1991, following a coup d'etat. The suspension was lifted in December 1994 following which the project was amended on December 7, 1995 to adapt it to the post-embargo environment. The original Closing Date was extended to June 30, 1999. 3. The project was restructured in December, 1998 with one of the main aims of fostering a deeper private sector involvement in the water sector in Haiti. Had this effort been successful, the revised project Closing Date might have been further extended. However, the water sector reform law ( prepared with IDB's assistance ) that would have helped bring about significant changes has not yet been approved by Parliament and attempts to use an alternative law have not been successful. The project closed on June 30, 1999. Co- financing for the project was provided from the Agence Francaise de Developpement (AFD) and the OPEC Fund. 4. This ICR was prepared by Bernard Decaux (consultant) and Emmanuel Njomo (Task Manager), who visited Port-au-Prince, Haiti, from October 23 to 30, 1999. It is based on information obtained during this mission and contained in the project files. The borrower contributed to the preparation of the ICR by providing facts, figures and views. The ICR was reviewed by Oscar Alvarado, Abel Mejia and Max Pulgar-Vidal. 5. A translated summary of the borrower's report on the project is attached as an Appendix. The full report is available in the project files. IMPLEMENTATIONCOMPLETIONREPORT HAITI PORT-AU-PRINCEWATER SUPPLYPROJECT (Credit 2052-HA) EVALUATION SUMMARY Introduction 1. Port-au-Prince is a city of almost 2 million inhabitants or about 25% of Haiti's population. During the last two decades, the number of inhabitants has more than doubled, mainly because of immigration from rural areas. This has led to a rapid development of low income neighborhoods, with little access to potable water and deficient sanitary conditions. Project Objectives 2. The original Government request was for IDA assistance to improve water services in the Port-au-Prince Metropolitan area and strengthen the institutional capabilities of Centrale Autonome Metropolitaine d'Eau Potable (CAMEP). The specific project objectives were to: (a) develop CAMEP into an operationally and administratively efficient and financially viable water utility; (b) improve and expand water services in the project area; (c) execute a first stage investment program to rehabilitate existing water supply infrastructure; (d) formulate a concrete plan for improving the sanitary conditions in Port-au-Prince; and (e) attain the Government agreement on a strategy and a plan of action for achieving more dynamic and efficient water sector. These objectives were addressed by three project components: expansion of the water supply system through infrastructure investments, institutional development, and technical assistance and studies. Project Implementation 3. Credit 2052-HA was approved on June 27, 1989, and became effective on February 14, 1990. The original closing date was June 30, 1994; it was extended to June 30, 1999, to make up for the delays resulting from the three year suspension that followed the coup d'etat and the slow start of the project which was partly due to difficult procurement problems. The project was restructured in December, 1998 primarily with the aim of fostering a deeper private sector involvement in the water sector in Haiti. Had this effort been successful, the closing date might have been extended further. However, the water sector reform law that would have helped bring change about is still held up in the Parliament and attempts to use an alternative law have not been successful. The preparation of this law was financed by the IDB. The project closed on June 30, 1999. Total project cost is calculated at US$35.4 million of which, US$20.5 million were drawn from the IDA credit. In addition, cofinancing in a form of two loans, totaling about US$15 million was obtained from the French Agency of Development and the OPEC Fund. Only US$0.3 million came from the water supply company and the Haitian Government. iii 4. Project implementation did not really start until 1995, when the project was reactived after the three-year suspension. After disbursements were resumed in 1996, progress in expanding and rehabilitating the water supply works was relatively satisfactory. However, progress in the implementation of the commercial, financial and administrative management components was deficient and plagued by delays. Technical assistance provided by well- known foreign consulting firms between 1996-1998 did not fully meet its objectives, particularly in the commercial and human resources area. Project Results 5. The project's physical objectiveshave been partially achieved. A well field was constructed with seven productive wells, with a daily capacity of 19,760 cubic meters as planned. Pumping stations and new reservoirs have been installed. Transmission pipes were put in place and water mains were installed or rehabilitated. House connections increased by 4,600 since the project was reactived in 1995, which is below the 10,500 initial target. A number of house connections have been disconnected due to non payment of bills by water users or technical failures. Only 1, 000 water meters have been installed compared to 6,500 originally expected. A low percentage of existing meters is effectively working and/or read. Population served has increased but water users served through legal connections represent only about 18% of the total population against 25% ten years ago, reflecting among other things the rapid increase in population of the Port-au-Prince Metropolitan area ( about 78%) during the project period. 6. The unaccounted for water is estimated no less than 55%, which exceeds the already high 50% average of the countries in Latin America. Other key performance indicators such as water billings and connections remain unsatisfactory. Improvements in commercial and financial management have been limited. There are still problems with water quality and reliability. There is a serious leak problem, and the program of control of connections and meters remains inadequate. The poor results achieved in the institutional performance of the company (which was conceived as the project's most important objective, as confirmed by the 1998 Amendment to the Credit Agreement) outweigh the progress made on physical works, leading to an overall "unsatisfactory" rating of the project. The financial rate of return is estimated at -8.26% due mainly to the high rate of illegal connections and a collection rate of approximately 80%. The economic rate of return is estimated at 7.08% compared to the 10.1% estimated during appraisal. The result of the economic rate of return reflects the considerable differential between the prices charged by private vendors and that charged by CAMEPI which are deemed closer to the real economic benefit. IDA Performance 7. IDA's performance was unsatisfactory in identification, preparation and appraisal phases of the project. The difficulties inherent in attempting to strengthen the water utility over the five-year project period, given Haiti's subsequent history of instability and weak institutional setting, were vastly underestimated. After the project was reactivated in 1995, following a three year suspension of the credit, IDA staff enthusiasm proved short lived given the deep seated problems in the water company and the slow progress achieved. A more I CAMEP'spricesof about US$ I/m 3 are estimated closeto marginalcost,according to experience in other regionalutilities. Vendors chargeaboutUS$4/m 3 on average. iv fundamental restructuring of the project should have been undertaken following the lifting of the suspension. Subsequent supervision efforts as well as technical assistance by reputable international firms were not sufficient to solve the difficult issues encountered during the implementation of this poorly designed problem project. Borrower Performance 8. The overall performance of the Borrower was uneven. At the start of the project, the entire management and cadres of CAMEP were fired and replaced by inexperienced staff. CAMEP's performance was further impaired during the three-year suspension which prevented the inflow of external resources that would have financed investments needed to increase access to potable water and improve service. The relatively numerous changes in personnel in the commercial and financial departments were detrimental at times. Nevertheless, the working relationship between CAMEP and IDA staff was often good, and the dialogue contributed to the timely resolution of a number of specific problems relating to procurement, disbursements,institutional development and audits. Assessment of Outcome 9. The project outcome is rated as unsatisfactory on the basis of the original project objectives, even though physical objectives have been reasonably achieved and the rate of project implementation attained represented a commendable outcome under the political and social conditions existing in Haiti. The target relating to production capacity was met fully. About 50% of the planned transmission mains were constructed. Two pumping stations were constructed, as opposed to a single station expected during appraisal and 83 km of distribution pipes were laid, instead of the 60 km planned. However, the number of house connections amounted to only 2,580 or 50% of the new connections envisaged under the project. CAMEP was in the process of planning a major push to expand house connections during the visit of the ICR mission. Because of rationing, the estimated water consumption is still 46% below the expected water demand, although illegal connections and other methods of indirectly obtaining CAMEP's supply of potable water abound. 10. Unaccounted for water is high (55%). Only two-thirds of the bills are collected. The collection period is 355 days. 2 Only about two-thirds of the bills are collected. The collection period is 355 days partly as a result of CAMEP's reluctance to write-off its bad debts because of its impact on operating profits. The ratio of collections to billings increased from 51% in 1994 to an average of 77.5% between 1995-1999. CAMEP showed an operating profit annually through out the entire 1995-1999 period, averaging about US$0.8 million. CAMEP also made a modest profit during this period except in 1998, when it incurred a loss of about US$1 million. Nonetheless, CAMEP's profits are overstated because it did not service the debts owed to the state. The average tariff level has been raised, but a number of de facto structural distortions remains, due to the quasi absence of meters which prevents the application of the tariff structure. 11. A positive outcome of the project has been the sharp increase in the number of people served through stand pipes in low income neighborhoods. Only about 350,000 people, or 2 Thisis mainlycausedby the needto providesufficient quantities of water intothe network,in orderto avoid negative pressuresandhence contamination. Besides,thereare many illegalconnections and leaks. v 18% of the population, are served through legal connections. Another 734,000 are estimated to obtain potable water services either through neighbors or illegal connections. About 514,000 residents (up from nil in 1994) of low income areas are served through water kiosks managed by water committees selected by the residents themselves, with the assistance of NGO's. Despite the good progress in expanding the access to potable water during the ten- year project implementation period, the overall outcome is still considered unsatisfactory. This negative assessment is based on the levels of key performance indicators, linked to the project's original objectives, which remain deficient. In addition, important institutional objectives were not achieved in spite of the combined efforts of the project co-financiers, international consultants and IDB's support for sector reform. Despite the achievements in the physical aspects of the project, CAMEP can not be considered financially viable. The bacteriological quality of the water being supplied by CAMEP continue to be mixed, though considerably improved, and water rationing remains a fact of life in Port-au-Prince. Furthermore, the hoped for reform of the sector, although within reach, has not materialized. Summary of Findings, Future Operations and Key Lessons Learned 12. The failure to achieve institutional strengthening outweighs the reasonable physical accomplishments, as it introduces great uncertainty in project sustainability. The present administration and donor agencies support a major reorganization of the entire water supply sector, in the hope that institutional weaknesses will be eliminated with a deeper involvement of the private sector, by putting CAMEP under a management contract, which would be subsequentlyfollowed by a concession. 13. Even with a management contract for a period of about five years, CAMEP will need significant resources to finance the investments proposed in the new Port-au-Prince master plan for potable water, which was prepared by an IDA-financed consultant. The needs of the water sector in Port-au-Prince are enormous. Tentative projections for the period 2000-2005 indicate that investments required may amount to US$200 million. The bulk of this amount would have to come from external donors including IDA. Beyond 2005, when a private concessionaire would eventually take over, some of the investments could be financed by such concessionaire, the remainder (substantially less than in the period up to 2005), would be made available to support works that are needed, but which are not attractive to the private sector (e.g. water supply in marginal areas, technical assistance, etc.). An alternative approach would be to continue to support the above-mentioned community-basedprogram of water supply to low income sections in the Port-au-Prince area, where a fast growing population lack water distribution networks, but is nonetheless increasingly being served through water kiosks, managed by community groups with NGO support. This program has already made a successful start and deserves to be continued with donor support. Its decentralized management and focus on poverty alleviation would be in line with IDA's overall development strategy for Haiti. Key Lessons Learned 14. The project provided several important lessons for water supply operations. * Early on in project preparation, consideration should have been given to private sector management or ownership as an alternative to relying on the public sector. vi Institutional strengthening is a very difficult and time-consuming task. The role of the private sector and the modalities for its involvement should have been properly evaluated during project preparation. The option of introducing a management contract, concession or affermage should have been studied as a viable alternative before the project was approved. Even though this is now evident and a law has been drafted to reform the sector and enhance the role of the private sector, the law is presently stymied in Parliament. Project design should be flexible to take advantage of unexpected developments during project implementation. The project had originally envisaged the construction and improvement of about 150 stand pipes to serve residents of low income areas. During project implementation, CAMEP expanded this approach to include the provision of bulk water to water kiosks that were managed by the community with the assistance of NGOs. This approach turned out to be quite successful, as it led to considerable savings in potable water expenditures and fostered community self-help and participation in their own development . At present over half million low-income residents are benefiting from this program. They are among CAMEP's best paying customers, since they pay when they receive service. This approach should be studied to determine its replicability in other low income areas in Haiti and in similar situations in other countries. * A careful assessment of the feasibility of institutional strengthening programs and the establishment of realistic objectives in this regard should have been undertaken from the onset. This is especially true in unstable political environments where frequent government changes result in high staff turnover in public sector institutions as well as frequent policy reversals. In situations where a project supports a phased investment program, project performance indicators should also be phased. * Clear and realistic objectives and goals were not established prior the commencement of project implementation. The two main objectives of the project - extension of service coverage through physical works and institutional development - were initially presented on the same footing. As project implementation proceeded, the institutional development aspects of the project began to lag behind the physical aspects. The IDA began to stress institutional strengthening (which was reflected in amendments made to the project) and AFD supported additional technical assistance. This re-orientation of emphasis was not fully appreciated by the management and employees of CAMEP. This led to misunderstandings and failures during implementation. Poor project design is often very hard to correct during implementation. * Close cooperation is necessary among co-financiers in order to achieve shared objectives. Working in tandem would lead to more information sharing, and a consensus on project performance targets. Given the gaps in potable water services, all donors were concerned about water facilities network expansion; however, there were divergent views on institutional matters, in particular with regard to the role and timing of private sector involvement as well as on the urgency for attaining institutional targets. I....I . IMPLEMENTATIONCOMPLETION REPORT HAITI PORT-AU-PRINCEWATER SUPPLY PROJECT (Credit 2052-HA) PART 1: PROJECT IMPLEMENTATIONASSESSMENT A. Background 1. At the time of appraisal in 1988, water service levels for Port-au-Prince's 1 million residents were very low and among the lowest in the Western Hemisphere. Only about 15% of the population had a house connection; another 65% obtained water from a neighbor with a house connection; about 5% relied on public stand pipes; and about 25% resorted to other means such as cisterns, water vendors and natural surface-water sources. Water was severely rationed and supply limited to less than six hours per day; many people received water no more than two hours every other day. The water provided was generally unsafe and often insufficiently disinfected. In addition, Port-au-Prince did not have a water-borne sewerage system. 2. Given the lack of safe water and the poor sanitary conditions prevailing in the area, diseases generally associated with unsafe water and poor sanitation ranked second among all causes of mortality. Infant mortality, with a reported 77 deaths per 1,000 live births, remain among the highest in the Western Hemisphere (compared to between 25 to 55 recorded in other Latin American countries). 3. Two agencies are responsible for operating water supply facilities in Haiti: Centrale Autonome Metropolitaine d'Eau Potable (CAMEP) for the Port-au-Prince area; and Service National d'Eau Potable (SNEP) for the rest of the country. In 1978, SNEP received a US$6.6 million IDA Credit (747 - Ha) in support of an investment program including the expansion and rehabilitation of water systems in seven provincial towns and institutional strengthening of SNEP. The project was completed in 1984 with a 2.5 year delay. Objectives of the project were achieved only partially. A Project Performance Audit Report (PPAR) issued in 1987 found that physical components were executed, but financial and institution-building goals were not met. SNEP remained a very weak institution with chronic funding problems and poor management, largely unable to operate and maintain the water supply facilities under its responsibility. The PPAR concluded that in spite of the project's shortcomings and limited achievements, there appeared to be no cause for doubt that it was justified, if only for social and health reasons. 4. One very important and perhaps overriding factor for the sector's limited potential for revenue generation and CAMEP/SNEP water performance was without doubt the high proportion of the Haitian population below the poverty line, currently estimated at over 65%. IDA had a long-standing strategy of providing assistance to Haiti through lending operations in the social sector with a strong impact on improving the living conditions- of the poor. 2 Support to the water sector matched closely the IDA strategy, as the improvement of water services to the inhabitants of Port-au-Prince would meet the social aspects of this strategy. It was also hoped that assistance to CAMEP would promote more efficiency in the operation of an important public enterprise, and the introduction of more appropriate cost recovery policies would ease the government's need to provide operating subsidies. On these grounds, an IDA credit supporting a Port-au-Prince water supply project was approved in June 1989. B. Project Objectives 5. The project, to be executed during 1989-1994, was to support the execution of the first phase of a 1988-1998 investment program proposed by consultants in the Port-au-Prince Water Supply master plan. 6. The original Government request was for IDA assistance to improve water services in the Port-au-Prince Metropolitan area and strengthen the institutional capabilities of Centrale Autonome Metropolitaine d'Eau Potable (CAMEP). The specific project objectives were to: (a) develop CAMEP into an operationally and administratively efficient and financially viable water utility; (b) improve and expand water services in the project area; (c) execute a first stage investment program to rehabilitate existing water supply infrastructure; (d) formulate a concrete plan for improving the sanitary conditions in Port-au-Prince; and (e) attain the Government agreement on a strategy and a plan of action for achieving more dynamic and efficient water sector. These objectives were addressed by three project components: expansion of the water supply system through infrastructure investments, institutional development, and technical assistance and studies. 7. The detailed Project components were as follows: * InfrastructureInvestments a. Water Supply Infrastructure consisting of: (i) construction of a well field with an average daily capacity of about 20,000 cubic meters/day and connection of the well field to the water transmission system at the reservoir known as R120 including: construction of six wells, installation of pumps, electrical power lines, and pressure pipes (10 kilometers); (ii) replacement/construction of transmission mains conveying water from existing sources to the transmission system. b. Expansion and Rehabilitation of water transmission, pumping and storage infrastructureconsisting of: (i) construction of about 50 kilometers of transmission mains (200 to 800 millimeters diameter). (ii) construction of three storage reservoirs with an aggregate capacity of about 5,000 cubic meters (R120, Gentil and Madame Baptiste). 3 (iii) construction of a pumping station (200 cubic meters/hour). (iv) installation of monitoring and regulating equipment of CAMEP water systems. c. Distribution Network and Connections (i) construction of about 60 kilometers of water distribution pipes (less than 100 millimeters). (ii) installation of about 5,000 new and the rehabilitation of about 5,500 existing connections. (iii) installation of about 2,500 water meters. (iv) installation of about 2,600 flow limiting devices. (v) construction of 100 new and improvement of 50 public stand pipes. (vi) installation of about 50 fire hydrants. d. Construction of an Operations Building. * Institutional Development of CAMEP (i) acquisition of equipment, including vehicles and tools, required for maintenance and operation of CAMEP water supply facilities. (ii) carrying out a program of technical assistance and training to promote the development of management capabilities within CAMEP. * Technical Assistance and Studies Provision of technical assistance for: (i) supervising and engineering of the works included in the Project. (ii) studies for purposes of developing and designing a follow up project including final designs and specifications. (iii) studies for the purpose of preparing a plan to improve sanitation and waste disposal facilities and services in the Project Area. C. Achievement of Project Objectives 8. The objective of expanding the water supply infrastructure has been relatively well achieved despite continued shortcomings at the distribution level. While physical achievements were relatively satisfactory, the project objectives related to institutional strengthening were unsatisfactory. Key performance indicators on the basis of the original project design (unaccounted for water, water billings and collections) were not attained. Total project cost is calculated at US$35.6 million of which US$20.5 million were drawn from IDA funds, US$12.9 million from Agence Francaise de Developpement (AFD) and the Government loan and US$2 million from the OPEC Fund. CAMEP contribution was minimal, at US$0.3 million. This amount is nearly identical to the appraisal estimate of US$ 35.5 million. However, funds allocated to CAMEP institutional development and technical 4 assistance were 50% higher than anticipated (US$6.6 million instead of US$4.4 million) due to the need to address deficiencies in CAMEP performance which were exacerbated by the 3- year suspension due to the coup d'etat. 9. An economic ex-post evaluation was carried out, using incremental revenues generated by the investments and the price paid for water delivered by water vendors as a proxy for CAMEP's customers' willingness-to-pay. However, to take into account the increase in quantity demanded due to CAMEP's lower price, it was assumed that the quantity purchased by CAMEP's customers prior to the switch in supply, represents only about one- tenth of their current purchases from CAMEP. On the cost side, it includes investments carried out for expansion and rehabilitation of the water supply system (IDA, French AFD, OPEC) as well as incremental operating costs. Expenditures related to operations improvement and technical assistance were also included to the extent they represented actions taken to improve water supply investment productivity. On the benefit side, incremental revenues generated from tariffs together with connection fees, as well as consumer surplus brought about by savings due to the differential between the average vendor price (Gds. 132/m3) and CAMEP's tariffs or final price to Kiosk customers (Gds. 15.8/m3). The internal economic rate of return (IERR) of the project was estimated at 16%, taking the revenues and savings by kiosk customers and CAMEP's regular customers. If only the savings of CAMEP's regular customers are taken into account, the IERR drops to 14%. If only the savings of the kiosk customers are taken into account the IERR drops further to 4%. The financial internal rate of return (IRR) was estimated at -1.4%. 10. Project completion was delayed because of political problems that resulted in the 3- year suspension of the project between October 1991 and 1994. The suspension was lifted in December 1994 and the project was amended on December 7, 1995 to adapt it to the post- embargo environmentl. Investments carried out in 1991-93 by the French AFD were modest, totaling US$0.8 million and consisting of studies and some drilling activities. IDA did not disburse any amount before 19962. No operating costs nor benefits were obtained by the project before and during the suspension period and the financial and economic evaluation was thus carried out for the period starting in 1995. 11. The result of the economic evaluation appears reasonable. However, the project is clearly not financially viable. The negative financial return is explained by much lower demand than anticipated 3 , and to the delay in upgrading the commercial system, a process that is still underway. In addition, the high proportion of illegal connection and "customers" obtaining service from their neighbors (about 46% of total) has meant that CAMEP could only charge for less than half the water produced. Unaccounted for water 4 was about 55% in 1998, the level of accounts receivable measured in number of days of billing rose from 216 at appraisal in 1989 to 355 in 1999 and the ratio of collections to billings ranged from 51% in 1 The Amendment to the CreditAgreement reflected the Bank'sefforts to help the Government of Haiti and CAMEP in gaining flexibility for project execution. 2 Except for a PPF (US$837,000) disbursed in 1990/91. 3 Estimated annual water consumption was 24.5 million cubic meters in 1995 compared to 39.7 million anticipated at appraisal. Consumption remained at that level until 1998. 4 Technical and administrative losses. Technical losses are caused by poor pipe connections, faulty water meters and broken pipes. 5 1994 to 77% in 1999, with a high of 84% in 1996. Despite the high proportion of illegal customers, CAMEP was able to cover its operating costs fully, following the lifting of the suspension. Nevertheless, although CAMEP did show positive net income during the project period except for 1998, actual net income was negative because CAMEP was not able to service its debts to the state. Considerable wastage occurs as a result of a lack of meters and the flat rate billing system for consumption. The working ratio 5 has in fact increased from 71% in 1995 to 76.7% in 1999, whereas it was assumed to decrease by half between appraisal and project completion. These project outcomes support the overall unsatisfactory rating given to the project. 12. It became evident in 1998 that the project, as originally designed would not be able to meet its original development objectives. The key problematical objective called for transforming CAMEP into an operationally and administratively efficient and financially viable utility. CAMEP was not financially viable nor was it expected to become so before the project closing date of June 30, 1999. Other original development objectives that were not expected to be attained were: (i) improving the bacteriological quality of the water being supplied by CAMEP; (ii) eliminating severe rationing; and (iii) attaining agreement on a strategy and a plan for achieving a more dynamic water sector. The principal cause for the failure to attain these project development objectives was the three year suspension of the projects in Haiti. As a result of the hiatus, project implementation fell behind as CAMEP's network deteriorated due to lack of maintenance. Given the institutional development constraints, the objectives of the project were simplified 6 in order to bring them in line with the skills available in Haiti. Care was taken to reorient the project emphasis toward the institutional development of CAMEP as opposed to stressing only physical implementation. Physical Implementation. 13. Implementation of the infrastructure components was generally satisfactory, despite intermittent delays. After substantial initial problems, procurement went relatively smoothly. The supply and installation of some transmission and distribution pipes had to be re-tendered because all the offers received exceeded the project budget. The construction of the operation building was delayed because the firm that originally won the contract went bankrupt as a result of the embargo that was imposed following the overthrow of President Aristide. Improvement in operations has taken place and include an increase in water production and an expansion of transmission and distribution pipes network. A. Water Supply Infrastructure (i) Construction of a well field at Tapage Galette: production capacity reached 19,760 cubic meters as planned. Eight wells (as opposed to 6 planned) with a depth of about 120 meters have been drilled of which seven are productive. Pumps and related hydraulic and electrical equipment have been installed. An electric power line of 4 kilometers (against 11 km planned because the circuit was modified) has been constructed. About 20 kilometers of pressure pipes (against 10 S Ratio of operating costs to operating revenue. 6 The DevelopmentCredit and Project Agreement were formally amended on December 21, 1998. 6 kilometers planned) with diameters ranging from 200 millimeters to 800 millimeters have been installed. Accelerated urbanization in the Port-au-Prince area led to the expansion of the pressure pipes network and to changes in the original plan for transferring water to reservoirs. (ii) Transfer of water from existing sources to the transmission system. Water was to be transferred from seven existing springs and one new source (Gentil Spring). The Gentil Spring was not developed due to excessive price claimed by the landowner. Work was undertaken on only four sources leading to the replacement of 5.6 kilometers of transmission mains instead of the 11.2 kilometers planned. It appears that the reallocation of funds to other priority tasks, led to the decision not to replace some existing pipes considered as still in working condition. B. Expansion and Rehabilitation of Water Transmission, Pumps and Storage Distribution Construction/rehabilitation of 19.3 kilometers of transmission mains was undertaken compared to 40 kilometers originally planned. The number of storage reservoirs to be constructed/expandedwas reduced from three to two but with an identical combined capacity of 5,000 cubic meters. Due to the cancellation of the Gentil Reservoir component (since the Gentil Spring could not be developed), and of the Mrs. Baptiste Reservoir in the same area, the works plan was modified: an additional water tank was put in place at the Christ-Roi Reservoir; a pumping station was installed at the Nazon Reservoir site; an additional water mains was installed between the Bourdon and Nazon existing reservoirs; and the distribution network around the new Reservoir R120 was extended, with a new pumping station installed at the Freres 1 Reservoir. The result was to facilitate the "sectorisation" of water distribution, allocating specific springs and reservoirs to particular urban sectors, thus ensuring better control of distribution and consumption. C. Distribution Network and Connections. About 80 kilometers (as compared to 60 km planned) of distribution pipelines with a diameter of less than 200 mm were constructed. The number of new and rehabilitated water connections was less than originally planned, due to difficulties in reaching some overbuilt areas but mainly due to the reluctance of potential customers to pay 3,400 gourdes (US$200) connection charge, which is a sizable sum for low income people. About 3,600 new meters were purchased in May 1999 by CAMEP but only 1,000 have been installed (compared to 2,500 water meters planned). Present tariffs discourage the use of meters, particularly since water is still severely rationed in a number of areas. Flow limiting devices have been abandoned (2,600 were planned) due to improper use of such devices by users. Only 14 fire hydrants have been installed in new sections instead of 50 planned. 14. While the project's original objectives were only partially achieved in the distribution field, highly satisfactory results were obtained regarding the installation of public fountains in Port-au-Prince poor districts and slums. CAMEP support of a potable water supply 7 program initiated by a French NGO 7 in 1994 led to a significant improvement in the service to low income areas that were being served by private vendors using water trucks. The program initially encompassed 14 districts with about 210,000 inhabitants. Currently about 514,000 customers are served under the program. This program consists of the construction of water kiosks inside the shanty towns. CAMEP provides the bulk water but the management, operations and maintenance of the kiosks are undertaken by water committees, elected by beneficiaries. The water is sold by attendants employed by the water committees. Even though water is sold by containers, the price to users (US$1 per cubic meter) is not negligible for the poorest families, but affordable for the majority of families, who previously had to purchase water from water trucks at 2 to 3 times the price. The revenues generated are sufficient to cover all expenses of the local organizations, including the payment of water to CAMEP, the maintenance and some revenue left for investment in small collective infrastructures (especially in sanitation). 15. About 20 km of pipes have been installed, providing water to 60 fountains. In 1998, it was estimated that 50 more stand pipes and 500 homes were connected to the system. This program was financed in 1994-1998 by the European Union and the French Agency for Development (AFD). In 1998-1999, AFD also financed installation of stand pipes supplied by CAMEP in 13 other districts. At these stand pipes water is also sold by the container (at a price roughly equivalent to about US$1/m3). Total population in these 27 districts totaled 500,000 people, i.e. about 25% of the almost 2 million population living in the Port-au-Prince area. Although it entails an extension of the original scope of the project, the positive results obtained by the public stand pipes program have compensated to a significant extent (from a social view point mainly) for the relatively modest results achieved by CAMEP in the distribution field. Besides, these customers have evolved to be among CAMEP's best paying customers, since they pay at the time that service is provided. Institutional Development of CAMEP (a) Acquisition and utilization of equipment 16. Equipment, including vehicles and tools, was purchased for maintenance and operations of CAMEP facilities, including the works included in the project. The amount spent on trucks and pickups was US$604,000 while equipment purchased was valued at US$2,797,000 8 . The largest item was equipment for the rehabilitation and construction of 10,000 connections supplied in July 1998 for US$1,029,000. Other big items included equipment for repairing leaks (US$418,000), macro and micro meters (US$121,000), water pipes 12" to 1/2" (US$411,000) and electronic and computers equipment (US$391,000). Actual payment of US$3.4 million was thus 3.5 times higher than the US$1 million estimated at appraisal. This substantial deviation is an indication of the extent of the work required for maintenance and repair of a water network in extremely poor condition. 7 GRET (Groupe de Recherche et d'Echanges Technologiques). 8 Excluding US$622,000 for regulating equipment to maintain water pressure included under water transmission infrastructure investments. 8 (b) Development of CAMEP's management capabilities 17. At the time of appraisal, CAMEP was an inefficient water utility. It had made some progress towards improving its operational efficiency with the assistance of French consultants under a French Agency for Development (AFD) Emergency Program and an IDA PPF (US$764,000) but had a long way to go for achieving levels comparable to the Latin American average 9 . IDA consequently accorded the utmost priority to assisting CAMEP in its efforts to become a viable entity. Monitoring indicators were set at appraisal (and revised downwards in 1998 when it became clear that the original objectives were not achievable). 18. Technical assistance was provided by AFD. A detailed and useful organizational and management study was performed in 1995/1996by a French water company (Lyonnaise) at a cost of US$535,000. Some training was provided for CAMEP staff in France. It was immediately followed in 1997/98 by the arrival of a team seconded by another French company (Compagnie Generale des Eaux, CGE). They worked inside CAMEP on such issues as reporting, accounting, budgeting, financial planning, electronic data processing. Total cost of this assistance was US$1.42 million. Unfortunately it appears that poor working relationship between CAMEP staff and CGE personnel substantially reduced the transfer of expertise and the efficiency of the technical assistance. The local staff had the impression that the management of CAMEP would be on its way out, with the possibility of a deeper private sector involvement in CAMEP. This view led to the resentment of the CGE experts. The outcome was thus relatively unsatisfactory. At present the principal problems remain: a lack of a viable customer billing and accounting system and adequate information on the customer data basel; poor management of accounts receivable; treatment of debt service payments due to the state; ad hoc change to journal entries; and lack of an updated fixed asset register. Technical Assistance and Studies (a) Supervising and engineering of the work included in the Project 19. Another French consulting firm (SAFEGE) was selected in 1991 to assist CAMEP in preparing a detailed work program and supervise submission of bids and signing of contracts. In 1995, SAFEGE was again selected to help CAMEP to reactivate the project, after the embargo was suspended. It revised project costs, analyzed bids and generally supervised the execution of work. Total SAFEGE assistance amounted to US$3.26 million over the years 1991 and 1995-1998 and was entirely financed by AFD. SAFEGE's performance has generally been considered satisfactory under difficult conditions. (b) Follow-up project 20. A Belgian consulting firm (TRACTEBEL) was selected to carry out a Potable Water Investment Program covering the period 2000-2015 with detailed project costs estimates and 9 For instance, 20 staff per 1,000 water connections against 6-10 in Latin America. Also 4% of university graduates at CAMEP compared to 8%-15% in Latin America. 10 Although some progress has been recently made following a census of water customers conducted by a consultant financed by IDA. 9 plan of action up to 2005 for the Port-au-Prince Metropolitan area. The consultant's Master Plan submitted in November 1998 recommended accelerated rehabilitation efforts in order to reduce technical losses; installation of meters and improved management to reduce unaccounted for water ( estimated at 55%); and an increase in the water rates in order to moderate over consumption. The cost of the study (financed by IDA instead of AFD as originally planned) was US$826,000. (c) Sanitation Master Plan 21. IDA supported the preparation of a SanitationMaster Plan for the Port-au-Prince area, undertaken by consultants (a French/Canadian/Haitianconsortium) at a cost of $398,000 in 1998/99. The Plan provided an excellent diagnosis and made detailed recommendations although a financially viable strategy remains to be developed to address the enormous needs, via a demand responsive approach. (d) Census of CAMEP customers 22. At appraisal a census of customers was not formally included in the Project but was definitely considered useful since the lack of data on effective users was one major source for under-billings and non payment by customers. A French consultant (CGE) was selected to carry such a census based on 50,000 "fiches" and on that basis water users (regular or potential) were located through an aerial photo system. CGE was supposed to install software to manage the "fiches" and to train CAMEP staff in its use, in order to enable CAMEP to survey other areas and to undertake future updates of the customer data base. The census was completed in July 1998 and is now being continued by CAMEP trained staff. The census was conducted in 1997/98 at a cost of about US$1 million, financed by IDA. D. Major Factors Affecting the Project 23. Credit 2052-HA was approved on June 27, 1989, and became effective on February 1990, seven months after Board approval. Only a few month later, CAMEP faced significant problems in its organization, finances and operations. The Government replaced CAMEP's Director General and all department heads. The previous team had been able to initiate a number of positive changes in CAMEP's organization. Their departure was motivated by a strike of CAMEP staff opposed to the company's improved efficiency. Under the new management, CAMEP's administrative operations were not coherent and communications at all levels were inadequate. The new Director General fired the company's external auditors aggravating the situation and delaying the preparation of the audit report. In addition to the time lost due to these changes, the project fell behind schedule due to delays in the procurement of major civil works contract. The l