Morocco - Fifth Water Supply Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No: 24070 IMPLEMENTATION COMPLETION REPORT (TF-22631; CPL-36640; SCL-3664A; SCPM-3664S; CPL-36650) ON TWO LOANS IN THE AMOUNT OF US$128 MILLION AND $32 MILLION TO THE KINGDOM OF MOROCCO FOR A FIFTH WATER SUPPLY PROJECT June 28, 2002 FINANCE, PRIVATE SECTOR AND INFRASTRUCTURE MIDDLE EAST AND NORTH AFRICA REGION This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENTS (Exchange Rate Effective as of mid 2002) Currency Unit = Moroccan Dirham (DH) - 100 centimes Dirham = US$ 0.86 US$ 1 = DH11.62 FISCAL YEAR January 1 December 31 ABBREVIATIONS AND ACRONYMS ADB - African Development Bank AFD - Agence Francaise de Developpement BAD - Banque Africaine de Developpement CIRF - Centre Inter-Regies de Formation CLs - Collectivites Locales CSE - Conseil Superieur de l'Eau et du Climat DEPP - Direction des Entreprises Publiques et Participation DGCL - Direction Generale des Collectivites Locales DRSC - Direction des Regies et Services Concedes ESPD - Epidemiology and Sanitation Program Directorate HA - Hydraulics Administration MARA - Ministere de l'Agriculture et de la Reforme Agraire MOF - Ministere des Finances MOI - Ministere des l'Interieur MOPW - Ministry of Public Works and Vocational Training MOPH - Ministry of Public Health NMD - National Meteorology Directorate OECF - Overseas Economic Cooperation Fund (Japan) ONEP - Office National de l'Eau Potable RAD - Regie Autonome Intercommunale de Distribution d'Eau et d'Electricite de Casablanca RADEEF - Regie Autonome Intercommunale de Distribution d'Eau et d'Electricite de Fes RADEEM - Regie Autonome Intercommunale de Distribution d'Eau et d'Electricite de Meknes RADEEMA - Regie Autonome Intercommunale de Distribution d'Eau et d'Electricite de Marrakech RAK - Regie Autonome Intercommunale de Distribution d'Eau et d'Electricite de Kenitra RAMSA - Regie Autonome Multiservies dAgadir RED - Regie Autonome Intercommunale de Distribution d'Eau et d'Electricite de Rabat-Sale TA - Technical Assistance WSD - Water and Sanitation Directorate Vice President: Jean-Louis Sarbib Country Manager/Director: Christian Delvoie Sector Manager/Director: Emmanuel Forestier Task Team Leader/Task Manager: Alexander McPhail  KINGDOM OF MOROCCO FIFTH WATER SUPPLY PROJECT CONTENTS Page No. 1. Project Data 1 2. Principal Performance Ratings 1 3. Assessment of Development Objective and Design, and of Quality at Entry 2 4. Achievement of Objective and Outputs 7 5. Major Factors Affecting Implementation and Outcome 11 6. Sustainability 12 7. Bank and Borrower Performance 12 8. Lessons Learned 14 9. Partner Comments 15 10. Additional Information 15 Annex 1. Key Performance Indicators/Log Frame Matrix 16 Annex 2. Project Costs and Financing 17 Annex 3. Economic Costs and Benefits 19 Annex 4. Bank Inputs 21 Annex 5. Ratings for Achievement of Objectives/Outputs of Components 22 Annex 6. Ratings of Bank and Borrower Performance 23 Annex 7. List of Supporting Docurments 24 Annex 8. Borrower's Comments 25 Map No. IBRIB. 2.5095R   |ICR Type: Core ICR |Report Date: June 28, 2002| 1. Project Data Name: WATER SUPPLY V L/C/TF Number: TF-22631; CPL-36640; SCL-3664A; SCPM-3664S; CPL-36650 Country/Department: MOROCCO Region: Middle East and North Africa Region Sector/subsector: WU' - Water Supply & Sanitation Adjustment KEY DATES Original Revised/Actual PCD: 03/12/1992 Effective: 02/01/1994 09/01/1994 Appraisal: 02/01/1993 MTR: 10/31/1996 10/27/1997 Approval: 11/23/1993 Closing: 12/31/2001 12/31/2001 Borrower/Implementing Agency: KINGDOM OF MOROCCO/ONEP Other Partners: African Development Bank, Overseas Economic Cooperation Fund STAFF Cun-ent At Appraisal Vice President: Jean-Louis Sarbib Caio K. Koch-Weser Country Manager: Christian Delvoie Mahmood Ayub Sector Manager: Francoise Clottes A. Amir Al-Khafaji Team Leader at ICR: Alexander McPhail Sergio Calegari ICR Primary A uthor. Gerard Tenaille 2. Principal Performance Ratings (HS=Highly Satisfactory, S=Satisfactory, U=Unsatisfactory, HL=Highly Likely, L=Likely, UN=Unlikely, HUN=Highly Unlikely, HU=Highly Unsatisfactory, H=High, SU=Substantial, M=Modest, N=Negligible) Outcome: S Sustainability: L Institutional Development Impact: N Bank Performance: S Borrower Performance: S QAG (if available) ICR Quality at Entry: S Project at Risk at Any Time: No  3. Assessment of Development Objective and Design, and of Quality at Entry 3.1 Original Objective: The Bank had actively participated in the sector since 1968 through: four previous loans for water supply projects; one sewerage loan; and one sewerage study. The cornerstone of the Bank's previous involvement had been its efforts to develop ONEP into a strong institution for water production, which represented the majority of the sector's capital investments. In the last two water supply projects, the Bank had been gradually expanding its influence in both the complex area of water distribution (i.e., the Regies) by supporting social house connection programs and encouraging rational water pricing. Whereas investments under earlier loans were geared to the development of water production to cope with the rapidly growing urban water demand, later investments, including this project, were aimed at improving the use of production facilities, particularly by increasing the efficiency of distribution networks. This approach is consistent with the policy to: (a) concentrate on investments that maximize the return on existing facilities and depend less on allocations from the national budget; and (b) delay implementation of costly new investments in large water production and conveyance schemes. In this context, the pmject aimed to ensure that new investments in water production and distribution facilities are the result of coordinated, rational planning between ONEP and the Regies. Within this sector context, the objectives of the project were to: (a) assist the Government in developing a long-term policy on institution building and cost recovery of water services; and (b) to address the needs of the poorest population by improving the delivery of potable water in urban and rural areas, in particular through the rehabilitation of existing facilities. The objectives of the project were clear and realistic. As stated, they were not over complicated and clearly represented an important country development priority. They were consistent with the overall water strategy developed in the four previous Bank projects, and represented a logical extension of previous investments and objectives. The project was also consistent with the Government's objective to rationalize public enterprises, which at the time was being supported by an ongoing Bank-financed Public Enterprise Rationalization Project (PERL I). 3.2 Revised Objective: Although the development objectives were not formally modified, the scope of the project substantially changed, as reflected in the amendment to the Agreemnent of Loan No. 3664-MOR, which was countersigned by the Minister of Finance in February 2000. This amendment: (a) acknowledged the concessioning of urban water services to the private sector by canceling the Bank-financed RAD (Casablanca) and RED (Rabat) components of the project; and (b) canceled two components, financed by other donors, to be executed by the Ministry of Equipment, and for which altemative sources of financing were identified. These were: (i) the rural water supply component (with the Hydraulics Administration); and (ii) the meteorology/hydrology component (with the National Meteorology Directorate). In addition, the Project Agreements with the four remaining Regies were amended to reflect the Govemment's decreased capacity to increase tariffs, recover arrears due by municipalities, and to meet targeted financial objectives. In June 1999, and after taking into account the privatization events and the coming amendments, the Bank proposed to revise the project's Development Objectives as follows: The objectives of the Project are to assist the Borrower in strengthening the its institutional capacities in the water supply sector andjmproving the delivery of potable water in urban and rural - 2 -  areas through increasing autonomy of the Regies, and fostering private sector participation in urban water sector operations in the form of management contracts, leases (affermage) or concessions. However, in November 1999., the Government expressed that it did not wish to amend the Development Objectives, citing that the timing of such action was not desirable because the project would close in just over two years. Faced with this decision, 'the Bank agreed with Government on the amendments described above, but left the Development Objectives untouched. As the Development Objectives were broadly expressed, and as the Bank-financed components were only reduced in amount, the Bank's Executive Directors were not notified of the amendmnents to the Loan Agreement. 3.3 Original Components: The project consisted of the following cornponents: Original Components and Bank Financing Plan - 3-  Component US$ Millions Total Local Foreign Bank Financing Regies 187.5 69.4 118.1 128.0 (Waterworks, TA, Training ONEP 131.3 43.6 87.7 32.0 (Waterworks, TA, Equipment, Training) Hydraulics 14.4 5.5 8.9 0.0 Administration (Works, Equipment, TA, Training) Epidemiology 4.8 0.2 4.6 0.0 & Sanitation Directorate (Equipment, TA, Training) Water & 7.3 0.7 6.6 0.0 Sanitation Directorate (TA, Training) National 8.3 0.1 8.2 0.0 Meteorology Directorate (Works, Equipment, TA, Training_ Total 353.6 119.5 234.1 160.0 The original components were relevant to achievement of the Development Objectives, and well suited to moving the water agenda into the Regies and ONEP's distribution system of small urban centers. With hindsight, the Regies component might have been smaller, but at the time of Effectiveness no one knew what form the Government's privatization program would eventually take. Given the uncertainty, and particularly about whether or not wastewater would be part of the concession for Casablanca and Rabat-Sale, the Bank and Government left in the project adequate funds to cover most investment scenarios. For this component, there was also an increased risk in dealing with the Regies, as the Bank did not have much experience with them beyond the two loans made in 1977 and 1987. The 1987 loan (National Water Supply Rehabilitation Project, Loan 2825-MOR) was still ongoing at the time of appraisal of the Fifth Water Supply Project, and was experiencing slow disbursements in the Regies' component-which should have been an indication for caution. The Government's willingness to adequately raise tariffs was cited as a Project Risk in the SAR. Perhaps for this reason the project included stringent financial covenants for the Regies, recognizing not only slow disbursements but also their short autonomy experience and fragile financial foundation. As it tumed out, and as discussed above, these covenants were not realistic and had to be amended. -4 -  Despite these observations, the Project did take into account an important lesson from previous projects by incorporating into the Regies component a scheme to finance social house connections. The connection charge had been identified in earlier projects as a major constraint to expanding social connections and, therefore, the Fifth Water Supply Project included a 60-month payback procedure. At project end, the Regies component did significantly increase the number of social connections made in FRs, Meknes, Marrakech and Kenitra, and all funding for these schemes was used. The project included many other non-Bank financed components for small amounts, all with different implementing agencies. These components unduly burdened the project, and many of them were subsequently dropped or re-financed by the Government. 3.4 Revised Components: In February 2000, the Project's Loan Agreement was amended to cancel amounts because: (i) water and wastewater services in Casablanca and Rabat had been privatized (US$7 1.0 million); and (ii) less expensive financing was identified by the Government for some of the ONEP components (US$5.0 million). At the Closing Date, another IJS$ 17.0 million was cancelled from the Bank loan, because ONEP was unable to complete the procurement of new information systems and equipment by December 31, 2001. The revised Bank financing plan after the cancellations is shown in the following table. Revised Financing Plan (US$ millions) Component Original Project Original Bank Cancellations Revised Bank Total Total Total Regies 187.5 128.0 71.0 57.0 ONEP 131.3 32.0 22.0 10.0 Hydraulics 14.4 0 0 Administration _ Epidemiology 4.8 0 0 & Sanitation Directorate __ _ Water & 7.3 0 0 Sanitation Directorate _ National 8.3 0 0 Meteorology Directorate __ _ Total 353.6 160.0 93.0 67.0 Therefore, the Project to be financed by the Bank Loan was reduced to four R6gies (Fes, Meknes, Marrakech and Kenitra) and ONEP (with a reduced scope). The Bank's financing was reduced by US$93.0 million from the original US$160.0 million to US$67.0 million. As the components did not change, the Bank did not consider this to be a restructuring, only a reduction in scope, and largely to reflect the Government's eventual privatization plans for Casablanca and Rabat-Sale. The cancellation of the approximately US$17 million at the project's Closing Date, originally intended for -5-  ONEP to upgrade its information system was a result of: (i) a slow start and problems with the principal contractor; and (ii) a new Director at ONEP in the year 2001 who was not in agreement with the overall design and objectives of the proposed IT program. The fact that this component never really got started indicates that there was not sufficient up-front agreement with the Govermment on the overall design and objective. The SAR said little on this topic and only mentions "computer equipment for ONEP and 40 centers" in Annex 2-3. 3.5 Quality at Entry: The project's quality at entry is rated Satisfactory. The project was consistent with the sector strategy and objectives, as well as the government's development priorities. It met the safeguard polices in effect at that time, and it did not result in any safeguard issues being raised during implementation. The standard fiduciary measures (audits, Special Account) were in place at the beginning and throughout the project. Project design incorporated an important lesson from previous projects (the obstacles to increasing social connections) and attempted to deepen the implementation capacity and financial stability of the newly autonomous regies. In particular, the six Regies were selected based on the following eligibility criteria: availability of water supply master plans, management performance, overall capacity to implement the project, and commitments to full cost recovery through the enforcement of "protocoles d'accord" to be made with the Government. There was a Water Master Plan for the sector, which included the cities financed under the project. The Master Plan had proposed a priority phase for the year 2000, in order to meet the needs until the year 2015, which was reflected in the Fifth Water Supply Project's investments. Feasibility studies were completed by the time of appraisal and, as was the government's custom and desire, final engineering design was carried out with financing under the project. All Government staff to implement the project were in place by Effectiveness, and sufficient counterpart funding was always available. The project was prepared without major delay, from 1991 to 1993: its identification began in March 1991, and pre-appraisal, appraisal and post-appraisal missions took place in April 1992, April 1993 and July 1993. All project components were selected and justified on the basis of priorities set forth in master plans studies and other documents submitted by the various beneficiaries. At appraisal, the Bank was aware that a significant private sector contract for a major infrastructure sector was possible in the near future. What form that contract would take (e.g., management contract, lease or concession) was not known, nor was it sure the water sector would be selected, as the power sector was also being discussed. Further, if the water sector ended up with a private sector contract, there was also some doubt that sewage would be included and that perhaps these investments would remain a the responsibility of the public sector. For these reasons, the loan amount was deternined in order to leave the government flexibility over what the would Bank finance. Since the project design predates the existence of the Quality Assurance Group (QAG), there is no official assessment of the project's quality at entry. Two QAG Supervision review reports, issued in August 1997 and August 1998, rated the project as satisfactory. The project took nine months to become effective. However, given there were two separate loans (3664 and 3665), ten implementing agencies (6 Regies, ONEP, and 3 Directorates), six Project Agreements, plus a Loan Agreement and a Guarantee Agreement, it is not surprising it took a long time to get the project approved by the Government. In addition, loans in Morocco have traditionally taken longer than the Bank standard (4 months) to become effective, in part because of the large number of ministries and agencies that - 6 -  need to be consulted, and also because the Govermment wishes to delay the start of the commitment charge as long as possible to ensure the project will start to disburse quickly. Unfortunately, for this project, disbursements never achieved their projected rate. 4. Achievement of Objective and Outputs 4.1 Outcome/achievement of o.bjective: The project is rated Satisfactory for Outcome. The discussion below shows that not all components can be rated satisfactory, but the most important objective, i.e., that dealing with service provision for poor households, was fully achieved. Given the importance of this objective in the sector and the SAR, plus the strong economic rate of return for this cornponent (see section 4.3 below), a satisfactory overall project outcome is justified. The project's first objective, "assist the government in developing a long-term policy on institutional building and cost recovery of water services", was only partially met for the institution building aspect and not met for the cost recovery portion. Institutional development impact is discussed below in Section 4.4 and, therefore, not repeated here. The cost recovery objective was not achieved because the Government did not respect its commitment to increase every year the water tariffs according to the "protocoles d'accord" signed with every Regie. The gap between the Government's commitment and the actual tariffs in effect at the end of 2001 is: 11,3% in FRs, 14,7% in Meknes, 22,3% in Marrakech and 46% in K6nitra. Since November 15, 2000, no tariff increase has been implemented in any of the four Regies. The targets of working ratio were 47% for ONEP and specific for every Regie. The amendments to the Project Agreements, of February 2000, changed this target to a maximum of 80%X' for all Regies, but this objective also was not achieved. At the project's inception, the debt service ratio target was 1.5 for ONEP and all Regies, it was reduced to 1.2 when the project agreement was amended. This target was reached by all the R6gies except by RADEEF. The initial goal of reducing arrears to three months of billing was not achieved. In February 2000, the amendments to the Project Agreements extended this covenant to eight months of billing, starting in 1999. Nevertheless, the four Regies do not comply with the revised target (they ended up between nine and twelve months). For ONEP, the total of arrears, was DH 993 million in 1993, or 12.6 months of billing, and basically unchanged at DH 932 million at project end. The project's second objective was to "address the needs of the poorest population groups by improving water supply services in urban and rural areas, in particular through rehabilitation of existing facilities". The project fully achieved the first part of this objective. As shown in the table below, the program of social house connections carried out by the Regies to serve poorer areas exceeded the SAR expectations. About 61,234 social connections have been constructed, to be compared to the initial target of 47,770 forecast at appraisal. Social House Connections RADEEF RADEEM RADEEMA RAK Appraisal target 15,000 7,000 18,000 7,700 Actually constructed 40,616 7,000 10,268 3,350 The water facilities were rehabilitated and expanded as planned, increasing the coverage of urban water services. However, shown in the table below, accounted for water (efficiency) indicators did not meet expectations in all areas. -7 -  Efficiency Rates ONEP RADEEF RADEEM RADEEMA RAK Actual Rate in 1993: 92% 56% 59% 64% 60% Initial target for 2000: 92% 80% 80% 80% 80% Revised target: 92% 75% 75% 75% 75% Actual Rate in 2001: 92% 52% 67.6% 69% 76.7% The initial target for the year 2000, 80% for all the Regies, was clearly over-optimistic, and was later reduced to 75%. ONEP met its target, but it should be noted that this ratio only concems the production facilities, not the distribution. Nonetheless, for the Regies, the achievement has been very good in Kenitra, good in Meknes and Marrakech, but less positive in Fes, even if 35,000 water meters have been replaced since March 1999 (out of a total of 150,000). According to the RADEEF's management, this low efficiency rate is in large part due to the Medina, where the water network is very old (about 80 years), in bad condition and getting worse. Under the Project about 400 km should have been replaced but, given the difficulty of working in the narrow streets of the Medina, this was postponed so that the replacement of sewerage pipes could be done at the same time. 4.2 Outputs by components: REGEES. This component is rated satisfactory. The performance of RADEEF was very good in addressing the needs of the poorest populations as the number of social house connections reached 40,616, compared to 15,000 forecast at appraisal. Conceming reservoir construction only 17,000 m3, compared to a scheduled 32,000 m3, were constructed. However, reservoirs have a much lower priority in the Development Objectives and SAR than do the social house connections. The financial objectives were not met, but this also can be considered secondary to the objective of increasing the number of social house connections. ONEP. This component is rated unsatisfactory. This rating is supported primarily by the fact that about 53 percent of this component was cancelled (US$ 17 million, out of an original US$32 million). This component consisted of the following four sub-components: 1. Rehabilitation of the Atlantic Coast systems (US$12 million); 2. New MIS (US$17 million); 3. Documentation Center and training (US$0.8 million); and 4. Training (US$2.2 million) The Atlantic Coast sub-component was completed. The new MIS system experienced very long studies, and then difficulties with the firm responsible for the MIS design and equipment. The new director general of ONEP decided to stop this sub-component in 2001, and to review in depth the schedule and scope of MIS. This analysis was still on-going at the close of the project. Concerning the documentation center, construction was never started. On the other hand, the training program was implemented as forecast. As discussed in section 4.5 below, overall institutional achievements for all components were limited. Output indicators (see Annex 1) were well designed for the civil works and financial aspects of the project, but less so concerning the institutional objectives. The civil works output indicators were expressed in number of connections, kms of pipes, network efficiency, and reservoir capacities--all of which were easy - 8 -  to measure. However, there were no indicators or benchmarks in the SAR by which to measure institutional progress, with the result that it was difficult to judge the project's institutional achievements. 4.3 Net Present Value/Econonic rate of retlurn: The methodology used in the SAR to detennine project's economic rate of return relied on an increase in water sales as a result of: (i) increased water availability from investments in production, storage and distribution networks; and (ii) increased system efficiency. However, as shown in the table below, and despite the large number of new connections, water sales increased much less than had been forecasted at appraisal. RADEEF RADEEM RADEEMA RAK TOTAL Water sales (million m3) 1993 39.9 20.9 28.9 12.1 101.8 1997 33.5 21.8 26.9 15.3 97.5 1998 37.3 23.3 29.0 14.8 104.4 1999 36.7 24.1 28.9 15.3 105.0 2000 38.3 23.6 30.9 15.5 108.3 Additional water sold - 1.6 2.4 2.0 3.4 6.2 Forecast at appraisal 16.5 13.0 22.7 7.5 59.7 Gap 18.1 10.6 20.7 4.1 53.5 Water sales did not increase (or actually decreased in the case of RADEEF) for two reasons. First, the initial block in the tariff schedule is set too high at 8 m3/month, while in most countries it is not more than 5 m3/month. This means most low income households can always stay under this level of consumption. In addition, the tariff for the first block has not been raised for some time, and the difference in price per m3 between the first and second block has grown substantially. For example, in FRs, the difference is currently DH4.77 per m3. Therefore, given the liberal per cubic meter limit in the first block and the large price differential between the first and second block, there is a strong monetary incentive for consumers to not consume more than 8 m3/month. In fact, this what is now observed: almost every social house connection consumes less than 8 m3 per month, and many other households, which formally consumed more water, now have reduced their consumption to fall under this limit. Second, as shown in the following table, improvements in network efficiency never reached the 80 percent level that had been predicted at appraisal. RADEEF RADEEM RADEEMA RAK Efficiency rate 1993 56% 59% 64% 60% 2001 52% 67.6% 69% 76.7% Therefore, instead of water sales increasing by 59.7 million m3 per year for all four Regies, they only increased by 6.2 million m3 per year. This means that the revised economic analysis would show negative ERRs and negative NPVs (at a 10 percent discount) for all four Regies. -9-  On the other hand, the project has provided safe water to approximately 735,000 additional people, through 61,234 new social connections (2 families per connection and 6 persons per family). These families used wells, natural water resources or standposts before the project. They were willing to be served by house connections but the main constraint was the cost of the connection. The financial facilities offered by the project (cost of connection paid back in 60 equal monthly installments without interest) overcame this constraint. The NPV and ERR of the social house connection program was estimated for the ICR, and showed the following results: RADEFF RADEEM RADEEMA RAK Vendor Price (DH/m3) 30 50 80 50 NPV(DH) 47,669,016 9,306,688 738,751 6,736,397 ERR 19.2% 13.3% 10.1% 12% The principle assumptions in this analysis are: (i) the prevailing vendor price at the time of appraisal; (ii) a 20 I/c/d increase in water consumption with a house connection; and (iii) the investment made in each Regie. The vendor price is estimated from incomplete data in the SAR, which quotes vendor prices for some of the ONEP small centers as ranging from DH25-DH200 per m3. However, the values shown do reflect the relative known rainfall and hydrological situation in each city. The increase in l/c/d consumption with a house connection is based on information about vendor sales in the SAR, and is considered reasonable based on the Bank's experience in Morocco and other arid countries. The investment amount is net of TA and training for each Regie (US$0.9 million to US$1.6 million). However, given the uncertainty of these assumptions the values used represent conservative estimates. A switching analysis was done on the vendor price which is the most influential assumption. Results are shown below: IVendor Price which I RADEFF RADEEM I RADEEMA I RAK I I gives a NPV=O 1 25.52 44.34 79.68 46.13 | As illustrated, small changes in the vendor price have a big effect on the NPV and, therefore, this analysis should be viewed with caution. However, even if it is difficult to quantify the immediate monetary effect of the social house connections, the improvement in the sanitary situation, the saving of time normally spent queuing and fetching water and potential increase in property and rental values all would add additional economic benefits. Further, the reduction of both infant mortality and of working days lost because of disease were not quantified. Therefore, the true economic benefit of the social house connection program is likely understated. In the SAR, it is stated that social connections were to be targeted to new and existing low income areas. Surveys conducted during project preparation in four Regies showed the areas to be included in the project to be social housing projects and/or bidonvilles. Water service in these areas was either nonexistent or via standpost. The Task Team is confident that these were the areas where investment took place and that few, if any, higher income households were able to take advantage of the connection charge installment plan financed by the project.. 4.4 Financial rate of return: No financial rates of return were calculated for the various components, because the appraisal report had - 10-  not calculated them. There were detailed financial projections made in the SAR for each of the Regies. However, given that out of all the financial targets only the debt service ratio was met by the Regies (except for RADEEF), it is evident that the financial performances predicted during appraisal for each Regie were not met during implementation. 4.5 Institutional development imnpact: Beyond significantly improving customer service and awareness inside the Regies (as a result of the large number o social connections added) and some increase in procurement capacity, it appears the Project had littl- impact on institutional development. The Government's commitment to strengthen the Regies and reinforce their autonomy was problematic for most of the implementation period. Until late 1999, the MOI undermined the performance of the regies by: (i) neglecting to monitor their performance; (ii) delaying and blocking periodic tariff revisions; and (iii) interfering with staffing. In addition, the approval of the Ministry of Finance for award of contracts often delayed project implementation. During this time, the Bank was unable to remedy these concerns through the Project. The ministerial change of December 1999 led to a significant improvement in the relationship between the MOI and the Regies, but this was the result of political factors external to the Project. Despite the lack of institutional impact, the project can still be rated satisfactory. There may well have been substantial institutional improvements and strengthening that took place as a result of the project, but the lack of indicators in this area does not pernit the ICR to make a firm judgement on this issue. Given this uncertainty, there is not enough justification to change the Outcome rating. 5. Major Factors Affecting Implementation and Outcome 5.1 Factors outside the contrcol of government or implementing agency: Not applicable. 5.2 Factors generally subject to government control: As discussed in Section 3.5 above, the major factor affecting implementation and outcome was the Government's decision to cause the main cities, Casablanca and Rabat-Sale, to contract with private firms for operation and maintenance of their water distribution and sewerage systems. This led to cancellation of the RAD and RED components, which meant a reduction of US$71 million in the Bank loan. The other factor affecting implementation and outcome of the project was the Government's reluctance to apply the annual water tariff increases, as stated in the Loan Agreement, as well as the protocoles d'accord signed between the Government and all the Regies. 5.3 Factors generally subject to implementing agency control: Following delays in studies and implementation of two sub-components, MIS and Documentation center, the new director general of ONEP decided to cancel them. This has lead to the cancellation of about US$ 17 million of the Bank Loan 3665-MOR. 5.4 Costs andfinancing: A mentioned above, cancellation of the Casablanca and Rabat-Sal6 investments in the Regies component reduced the Bank loan by US$71 million. Another US$22 million was cancelled from the ONEP component at the request of the Government, but most likely because cheaper financing was identified. For - 1 1 -  example, the MIS contract with the vendor was not cancelled by ONEP and the Bank expects that EU grant funding, which became available in early 2002, will be used to complete this contract. The Regie components that were implemented experienced costs within or below budget estimates and, therefore, it appears that the cost figures used in the SAR were adequate. 6. Sustainability 6.1 Rationale for sustainability rating: The project's sustainability is rated Likely. Thus rating is supported by the following considerations: (i) the civil works constructed under the project were completed to an adequate or better quality; (ii) the Regies and ONEP have the technical skills to operate and maintain the facilities; (iii) the Regies are multi-utilities (water, electricity & sewer) which perrnits them to cross-subsidize operations and provide for a measure of financial stability; and (iv) the Government's renewed commitment to keep the tariff level in line with the needs of the Regies and ONEP (there were tariff increases in all four Regies in 2001). In addition, the Government's commitment to the sector remains high, as does that of donors who remain active and who are funding new investments (especially those donors with low cost financing such as the EU, Japan and France). The policy environment for wastewater remains uncertain in Morocco, and the Bank has experienced concern in this regard in an on-going project (the Second Sewerage and Water Re-use Project, Loan 4010-MOR). However, this issue should not affect the sustainability of the Water Supply V Project, as it mostly dealt with water supply issues and investments. Thus, the lack of a clear and predictable wastewater strategy on the part of the Government should not affect the sustainability of the project. 6.2 Transition arrangement to regular operations: Aside from ONEP's MIS, the project investments have been fully integrated into the regular operations of the Regies and ONEP as every sub-component was completed. Therefore, there has been no transition problem. The project can be monitored in the future through the relevant financial indicators, and such a monitoring system already exists via the extemal audits done for each Regie for the Government. Given the success of the social house connection program supported by the project, another project with this component many be worthwhile identifying. However, the Bank would not consider a water project without a wastewater component and, therefore, we may have to wait on a new water project until the Government's wastewater policies are solidified. 7. Bank and Borrower Performance Bank 7. 1 Lending: The Bank's lending performance is rated Satisfactory. This rating is supported by the overall evaluation of the Bank's performance at identification, preparation and appraisal, as discussed below. Performance at Identification was Satisfactory. The Bank had participated actively in the development of the sector since 1968 through four loans for water supply projects, and two for sewerage projects. The Bank had directed its lending to strengthen the sector institutions while increasing access to safe water by - 12 -  low-income populations in urban areas. The Fifth Water Supply Project was consistent with this strategy and well identified. Performance at Preparation was Unsatisfactory. During project preparation the scope of the project was too ambitious: it covered the six main cities of the country, ONEP and also included four small components for three ministries which are not related to the core project. Obviously, the Bank accepted all requests from local authorities and the project became overburdened with unrelated components. In addition, there were 11 implementing agencies with only a coordination by the Ministry of Interior between the six Regies. It would have been better to concentrate the project on the Regies and, possibly, on one ONEP component. Fortunately, this reduction/concentration in scope occurred naturally during project implementation. Performance at Appraisal was Satisfactory. Cost estimates were done well, the SAR was of a high quality and the Government showed a high commitment to the Project. The design was overly complex, but did include the social house connection program which reflected lessons from previous projects and turned out to be very successful. The SAR recognized the key project risk, i.e., the Government's potential to resist raising tariffs. The project also successfully attracted lower cost parallel financing from other donors. 7.2 Supervision: The Bank's supervision performance is rated Satisfactory. Once it became clear that the government was going to grant concession contracts in Casablanca and Rabat-Sale, the Bank actively advocated that these contracts be competitively bid. However, this advice was not followed and the concessions were awarded by negotiation; on a sole sources basis in Casablanca and from a short list in Rabat-Sale. However, the government did request t:o have a senior Bank project officer based in Rabat to provide on the spot advice and assistance during the transition to the concessions. The Bank official also was on hand to accelerate implementation of the Project in other cities (Morocco had no resident mission at that time). This appears to have had a positive impact on implementation and on Bank performance in supervision which was satisfactory. Overall, there were 13 formal supervision missions, from December 1993 to April 2002, or about twice a year, which is reasonable. The Bank also reacted quickly to the changes in the project components, brought about by the cancellation of the Casablanca and Rabat-Sale components and the availability of lower cost financing for other components (see Section 3.4), when it prepared an amendment to the loan agreement in. February 2000. There were two QAG supervision.assessments, in August 1997 and August 1998. The overall project assessment for both was rated "2" or Satisfactory. 7.3 Overall Bank performance: Based on the discussions above, overall Bank performance is rated as Satisfactory. Borrower 7.4 Preparation: The project was well prepared by the borrower and can be rated Satisfactory. Water resources master plans covered almost the entire country, and a study conducted by MOF and MOI assessed the situation of the Regies and recommended the restoration of their autonomy. A seminar, organized by MOI and the Bank in April 1992 discussed options for enhancing financial and managerial autonomy, and consequently project preparation capacity. The basic recommendation of the seminar was to establish agreements (protocoles d'accord) between the Government and each Regie, which was a condition of loan effectiveness. This was done on time and clarified their respective commitments and obligations. However, - 13 -  the Govenmment's commitments were not achieved. 7.5 Government implementation performance: The perfonnance of the Ministry of Interior was weak until the ministerial change of December 1999 which led to a drastic improvement between the MOI and the Regies. In addition, the Government did not respect its commitments as regards the needed tariff increases. Therefore, its performance can be rated as Unsatisfactory. 7.6 Implementing Agency: The start of the project on the whole was delayed by two years because the final design was financed by the loan itself, as usual in Morocco. Nevertheless, the perfonmance of the Regies was Satisfactory. On the other hand, ONEP's performance was mitigated because, despite years of studies, it was unable to implement a new MIS which represented 50% of the ONEP component. Its performance can be rated Unsatisfactory. 7.7 Overall Borrower performance: Taking into account the recent reform of the Ministry of Interior, as well as the discussion above, the Borrower's overall performance can be rated as Satisfactory. 8. Lessons Learned The main lessons learned are: (a) In middle income and more sophisticated countries, project design should carefully consider the risks between hardware and software. This Project well illustrated that institutional and financial improvements are usually much more difficult for borrowers to achieve than civil works. As a result, and in order to escape from a commitment it now regrets, the Borrower may substitute lower cost financing for Bank financing, or abandon these components altogether. (b) The design of a project should concentrate on well related components, with the same objective. Too many ministries, too many implementing agencies and too many components, are impossible to coordinate and lead to dissipate the effort. (c) An open and competitive bidding for private sector participation is preferred to direct negotiation. During project implementation, the water, wastewater and electricity systems in the cities of Casablanca and Rabat-Sale were privatized. Neither contract was competitively bid; the former was negotiated on a sole source basis, the latter negotiated from a short list. Shortly after the Rabat-Sale concession was concluded, the government started on the process to grant a concession in Tangier. However, this time the process was on an open and competitive basis. Apparently the government has learned the value of competitive tendering through experience-an important experience that should be part of the Bank's PSP discussions with other countries. (d) In middle-income countries, and especially those with a large number of active donors, Bank funding expensive and uncompetitive. In this Project, out of a loan of US$89 million (US$160 million minus the US$71 million for the Casablanca and Rabat-Sale components) approximately US$22 million (or 25 percent) was thought to have been cancelled because the government found lower cost financing. - 14 -  (e) For contracts that are in sectors characterized by fast changing technology (e.g., the ONEP IT contract in this Project), smaller, incremental and sequential contracts are better than one large contract that requires a long time to plan and implement. (f) The Bank must carefully evaluate the proxies put forward to represent government commitment. In this Project, multiple "performance contracts" were made between the Regies, ONEP and the Ministry of Interior, and the Bank executed seven Project Agreements. All were relied upon by the Bank to achieve key policy objectives. However, the contracts were not respected, and the Bank had little recourse during implementation-short of suspension which it was not willing to do after taking into consideration the progress being made in social connections. (g) The Bank has a relatively low understanding of water and sanitation sector financing and needs to do much better in this area. For example, in this Project covenants on water tariffs were not respected and unaccounted for water reduction targets wvere not met, but levels of service and coverage for water supply are good throughout the country's urban areas. So, there must be an adequate flow of funds to make new investment and maintain infrastructure, but the Bank has trouble understanding these flows and designing projects (and their covenants) to adequately portray them. (h) The Bank has to be more thoughtful about the stated objectives for water and sanitation projects. Normally, the Bank insists on a positive ERR and NPV to justify water supply investments. Indeed, the economic analysis for this Project was based on: (i) substantial increases in new water; and (ii) substantial reductions in unaccounted for water. By Project end, none of these variables achieved their predicted targets and, as forecast by the economic analysis, the Project's overall ERR ended up being negative. Yet, the Project was not suspended or ever put in problem status-because the progress on making social connections was above expectations. Thus, the Project's objective was apparently more about providing water to low income families than it was about achieving financial goals. In this case, this objective should have been more emphasized, and a cost effective economic analysis methodology employed. 9. Partner Comments (a) Borrower/implementing agency: Comments prepared by the Borrower are attached in Annex 8. (l) Cofinanciers: Not applicable. (c) Other partners (NGOs/private sector): No comments were received. 10. Additional Information - 15 -  Annex 1. Key Performance Indicators/Log Frame Matrix Outcome / Impact Indicators: IndicatorlMatrix Projected In last PSR ActualLatest Estimate Working Ratio RADEEF 80% 108% RADEEM 80% 86% RADEEMA 80% 97% RAK 80% 84% ONEP 47% 55% Arrears 8 months of billing RADEEF: 249 days RADEEM: 365 days RADEEMA: 365 days RAK: 295 days Debt service ratio ONEP: 1.5 2.0 RADEEF: 1.5 0.8 RADEEM: 1.5 2.8 RADEEMA: 1.5 2.9 RAK: 1.5 1.75 Social house connections 40,616 RADEEF 15,000 7,000 RADEEM 7,000 10,268 RADEEMA 18,000 3.350 RAK 7,700 Network efficiency RADEEF 75% 52% RADEEM 75% 67.6% RADEEMA 75% 69.0% RAK 75% 76.7% ONEP:(on transmission/production) 92% 92% Output Indicators: Indicator/Matrix Projected In last PSR Actual/Latest Estimate RADEEF Pipelines and cvil works 320 km from 60 to 1000 mm Reservoirs 4 reservoirs, 32,000 m3 capacity 2 reservoirs, 17,000 m3 RADEEM Pipelines and civil works 40.75 km from 80 to 600 mm 36.53 km Reservoirs 6 reservoirs, 29,000 m3 capacity 3 reservoirs, 27,860 m3 RADEEMA Pipelines and civil works 480 km from 60 to 600 mm 312.95 km Reservoirs 1 reservoir, 12,500 m3 capacity 1 reservoir, 12,500 m3 RAK Pipelines and dvil works 40.75 km from 60 to 900 mm 36.53 km Reservoirs 5 reservoirs, 23,000 m3 capacity 5 reservoirs, 21,200 m3 End of project -16-  Annex 2. Project Costs and Financing Project Cost by Component (in US$ million equivalent) Appraisal Actual/Latest Percentage of Estimate Estimate Appralsal 'Project.Cost By Component US$ million US$ million REGIES Al. RAD 57.60 0.00 0 A2. RADEEF 14.80 14.80 15 A3. RADEEM 9.90 9.90 13.9 A4. RADEEMA 21.60 21.60 19.6 A5. RAK 10.70 10.70 13.8 A6. RED 44.80 0.00 0 ONEP 111.60 38.70 18.6 MPW: Hydraulics Administration 12.20 0.00 0 MPH 4.10 4.10 0.8 MOI 6.20 0.00 0 MPW: Meteorology Directorate 7.10 0.00 0 Total Baseline Cost 300.60 99.80 Physical Contingencies 15.03 Price Contingencies 37.97 Total Project Costs 353.60 99.80 Total Financing Required 353.60 99.80 Project Costs by Procureme nt Arrangements (Appraisal Estimate) (US$ million equivalent) ProcurementVMethod. Expenditure Category ICB - NCB , the.2 :NB.F Total Cost 1. Works 126.00 39.00 0.00 39.70 204.70 _______ __ (68.30) (17.10) (0.00) (0.00) (85.40) 2. Goods 67.30 6.30 0.60 45.40 119.60 _______ __ (45.30) (4.90) (0.50) (0.00) (50.70) 3. Services 0.00 0.00 23.90 5.40 29.30 ________ __ (0.00) (0.00) (23.90) (0.00) (23.90) 4. Miscellaneous 0.00 0.00 0.00 0.00 0.00 (0.00) (0.00) (0.00) (0.00) (0.00) Total 193.30 45.30 24.50 90.50 353.60 (113.60) (22.00) (24.40) (0.00) (160.00) Project Costs by Procurement Arrangements (Actual/Latest Estimate) (US$ million equival nt) Procurement Method:'::t' ExpendIture Category ICB -.NC,B 2 ... Total:Cost -~~~~~. . . . 1. Works 63.22 0.12 0.00 0.00 63.34 (46.79) (0.08) (0.00) (0.00) (46.87) - 17-  2. Goods 7.63 2.08 0.00 0.00 9.71 (7.70) (1.04) (0.00) (0.00) (8.74) 3. Services 0.00 0.00 11.39 0.00 11.39 (0.00) (0.00) (11.39) (0.00) (11.39) 4. Miscellaneous 0.00 0.00 0.00 0.00 0.00 (0.00) (0.00) (0.00) (0.00) (0.00) Total 70.85 2.20 11.39 0.00 84.44 (54.49) (1.12) (11.39) (0.00) (67.00) "Figures in parenthesis are the amounts to be financed by the Bank Loan. All costs include contingencies. 2' Includes civil works and goods to be procured through national shopping, consulting services, services of contracted staff of the project management office, training, technical assistance services, and incremental operating costs related to (i) managing the project, and (ii) re-lending project funds to local govemment units. Project Financing by Component (in US$ million equivalent) Percentage of Appraisal Component Apprraisal Estimate Actual/Latest Estimate Bank Govt. CoF. Bank Govt. CoF. Bank Govt. CoF. 160.0 160.00 109.60 84.00 67.00 81.00 53.00 41.9 73.9 63.1 - 18 -  Annex 3. Economic Costs and Benefits Backeround. The SAR economic analysis assumed for each Regie significant increase in water sales brought about by project investrnents to: (i) increase water production; and (ii) decrease unaccounted for water. As discussed in Section 4.1, the project did not increase water sales by the forecasted arnount and only achieved an increase of 6.7 m3/ year, versus the forecasted amount of 59.7 m3/year. Given these results, duplicating the SAR economic analysis would result in negative NPVs for each Regie. Despite the lack of achievement in water sales, the project did provide over 61,000 social house connections that will provide significant economic benefit. Therefore, an economic analysis for performed for the social house connection program in the Regies of REDEFF, RADEEM, RADEEMA, and RAK. Assumptions. The principle assumptions in this analysis are: (i) the prevailing vendor price at the time of appraisal; (ii) a 20 UIc/d increase in water consunption with a house connection; and (iii) the investment made in each Regie. The vendor price is estimated from incomplete data in the SAR, which quotes vendor prices for some of the ONEP small centers as ranging from DH25-DH200 per m3. However, the values shown in the economic analysis do reflect the relative known rainfall and hydrological situation in each city. The increase in U/c/d consumption with a house connection is based on information about vendor sales in the SAR, and is considered reasonable: based on the Bank's experience in Morocco and other arid countries. The investment amount is net of TA and training for each Regie (US$0.9 million to US$1.6 million). The period of analysis is 15 years (1993 - 2008) and the NPV uses a 10 percent discount rate. There are assumed to be two households for each connection, and the average household size is six people. Benefits. Benefits are the savings to the household of avoiding vendor sales altogether. Thus, in the with project scenario, each household would avoid 20 U/c/d of vendor costs, having replaced this amount with lower cost water from the Regie system. Households would also buy another 20 UIc/d from the Regie, for a total of 40 I/c/d. The tariff is the actual tariff up to the year 2001, and then held constant until 2008. Increases in the vendor price rise the sarne percentage as the Regie tariff and are also held constant for the years 2002 to 2008. Results. An analysis was done for each Regie, with results as shown in the following table: RADEFF RADEEM RADEEMA RAK Vendor Price (DH/m3) 30 50 80 50 NPV(DH) 47,669,016 9,306,688 738,751 6,736,397 ERR 19.2% 13.3% 10.1% 12% Switching Analysis. The analysis shows a positive NPV for all four cities, but these outcomes are to a large part dependent on the vendor price assumed to prevail in each city. Thus, a switching analysis was done on the vendor price. Results are shown below: |Vendor Price which RADEFF RADEEM RADEEMA RAK I gives a NPV=O 25.52 | 44.34 | 79.68 | 46.13 I As illustrated, small changes in the vendor price have a big effect on the NPV and, therefore, the ERR results above should be viewed with caution. However, even if it is difficult to quantify the immediate monetary effect of the social house connections, the improvement in the sanitary situation, the saving of - 19-  time normally spent queuing and fetching water, and potential increase in property and rental values all would add additional economic benefits. Further, the reduction of both infant mortality and of working days lost because of disease were not quantified. Therefore, the true economic benefit of the social house connection program is likely understated. The excel worksheets are available in the Project File. - 20 -  Annex 4. Bank Inputs (a) Missions: Stage of Project Cycle No. of Persons and Specialty Performance Rating (e.g. 2 Economists, I FMS, etc.) Implementation Development Month/Year Count Specialty Progress Objective Identification/Preparation March 1991 Sr. Sanitary Engineer Appraisal/Negotiation April-May 1993 Sr. Sanitary Engineer October 1993 Financial Analyst Supervision 07/25/1996 2 Sr. Sanitary Engineer (1); Sr. S S Financial Analyst (1) 04/19/1997 2 Sr. Sanitary Engineer (1); Sr. S S Financial Analyst (1) 10/27/97 3 Financial Analyst(l); Economist S S (1); Sanitary Engineer (1) 03/27/1998 4 Engineer (2); Financial Analyst U U (1); Economist (l) 03/09/1999 4 Financial Analyst (2); Economist S U (1); Sanitary Engineer (1) 11/03/1999 4 Team Leader (1); Financial S U Analyst (1); Engineer (1); Economist (1) 05/27/2000 3 Financial Analyst (1); Sanitary S S Engineer (1); Water & Sanitation Sp. (1) 11/23/2000 3 Sr.Fin.Analyst/Ttl (1); Sr. S S Sanitary Eng (1); Environmental Spec. (1) 06/26/2001 4 Sr. Fin. Analyst/Ttl (1); Sanitary S S Engineer (1); Environmental Spec. (1); Economist (1) ICR April 2002 3 Financial Analyst (1); S S Sanitary Engineer (1); Water & Sanitation Sp. (1) (b) Staff: Stage of Project Cycle Actual/Latest Estimate No. Staff weeks US$ ('000) Identification/Preparation 602.5 Appraisal/Negotiation 15.0 Supervision 588.9 ICR 21.0 Total n/a 1,227.4 - 21 -  Annex 5. Ratings for Achievement of Objectives/Outputs of Components (H=High, SU=Substantial, M=Modest, N=Negligible, NA=Not Applicable) Rating 0 Macro policies O H O SU O M O N * NA El Sector Policies O H OSUOM * N O NA OPhysical OH *SUOM ON ONA E Financial O H OSU*M O N O NA E Institutional Development 0 H O SU 0 M * N 0 NA E Environmental O H OSUOM * N O NA Social El Poverty Reduction O H *SUOM O N O NA El Gender O H OSUOM O N * NA El Other (Please specify) O H O SU O M O N * NA E Private sector development 0 H 0 SU * M 0 N 0 NA El Public sector management 0 H O SU O M * N 0 NA D Other (Please specify) O H OSUOM O N * NA - 22 -  Annex 6. Ratings of Bank and Borrower Performance (HS=Highly Satisfactory, S=Satisfactory, U=Unsatisfactory, HU=Highly Unsatisfactory) 6.1 Bank performance Rating O Lending OHS OS OU OHU O Supervision OHS OS OU OHU O Overall OHS OS OU O HU 6.2 Borrowerperformance Rating O Preparation OHS OS O U O HU O Government implementation performance O HS O S 0 U 0 HU O Implementation agency performance OHS OS OU O HU EL Overall OHS OS O U O HU - 23 -  Annex 7. List of Supporting Documents I . Appraisal Report No. 12231 -MOR, October 29, 1993 2. Loan Agreement 3664-MOR with the Kingdom of Morocco, Decem,ber 3, 1993 3. Loan Agreement 3665-MOR with ONEP, December 3, 1993 4. Guarantee Agreement 3665-MOR, December 3, 1993 5. Project Agreements with RAD, RED, RADEEF, RADEEM, RADEEMA and RAK 6. "Contrat-Programme" between the State and ONEP 7. "Protocoles d'Accord" between the State and RAD, RED, RADEEF, RADEEM, RADEEMA and RAK 8. QAG Reports No.1 and 2, August 1997 and August 1998 9. Mid-term Review PSR, BTO and Aide Memoire, October 27, 1997 10. Excel workbook for economic analysis -24 -  Additional Annex 8: Borrower's Comments After perusal of the completion report, the conclusion drawn is that the Fifth Water Supply project was not a success for the RADEEF, since the latter did not meet the stated objectives and did not perform as well as other state-owned companies. However, we would like to underline the following points, based on the comparators used: I . The loan was entirely disbursed (as of January 13, 2000), i.e. 99.98% of the loan amount. 2. RADEEF implemented the planned projects quickly, meeting the prescribed deadlines. 3. RADEEF is the only state-owned company, which respected the original loan closing date without requesting an extension. 4. The rehabilitation of the Medina's water supply system was not completed during implementation of the Fifth. Water Supply Project and had a negative effect on output. There are several reasons behind this, namely: - The loan amount could not finance the Medina's rehabilitation. Funds were therefore used from Loan 4010 (MOR) (Second "Assainissement et reutilization de l'eau"). - The Medina's rehabilitation encountered serious technical problems and many companies hesitated to submit proposals because of these impediments. - The Medina's rehabilitation is a long-term project to be implemented over several years. For this reason, it's impact on output could only occur over the medium term. Since the percentage increase of potable water tariffs was lower compared to other state-owned companies and could not always cover the operating costs, the financial indicators consequently did not increase. The World Bank's second objective, the OBS, was met very successfully. The original number of household connections, originally planned for 15,000, more than doubled to reach a total of 40,616 connections. The objective of providing water connections to the poorest population groups was thus met, eliminating a substantial number of fountains (borne-fontaines?), which discharged water into sewers 24 hours day. In conclusion, the Fifth Water Supply Project was able to correct a substantial number of structural problems, as well as those affecting social and economic development. - 25 -   IBRD 25095R mopocl; Ai editerranean Sea FIFTH WATER SUPPLY-PROJECT - e i.uar - IMPLEMENTATION COMPLETION REPORT - - REGIES LOCATION -* -; CI . / " A * B~P T, *J - )_ ( {' a; J wr _ - D N, 2a *N 9h , A T L 4 N JC :r% " I - .:Id, la . 32' *~~~~ ~ ~ ~ ~ -- - --0 ISOYET . - -rGemm ,JRAO$ RVICEAITL -D2hiI! REGIES (AT APPRAISAL( RDEMREGI ES (AT COMPLETION) 1j2 - . . * ~~~~~~~~~~~~~~~~~~~~~~~~~~~HEADQUARTERS OF REGIES r S g<- Thdbouncbhries,cobro;droo_inesons PROVINCE CENTERS SERVED BY REGIES _t ' ""' MOUNTAINS C -~~~~~~~~~~~~~~~~~~-400 ISOHYRTS K - 3 p -j -'~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~RIVERS AND WADIS