Morocco - Second Water Supply Project

Docennt of The World Bank FOR OFFICIAL SE ONLY Rprt NO. 8725 PROJECT PERFORMANCE AUDIT REPORT - No0Rocco SECOND WATER SUPPLY rROJCT (LOAN 1724-NOR JN 8, 1990 Operations Evaluation Department This d.comen b.u a restr~e disrhut~. vd miy be asd by recpdnt .mdy in9 the Kmýof ]Fame of Curre ( revlation) Dirtam (DR) Exchage Rate 4ra$ual Reýårt (as o November 1Ø78) US$1 - DH 3.82 Interveling lerars (average) US$1 - DH 6.76 Completion Year (1987) US$1 -DH 8.2T ONEP - Office National de aEu Potable FISCAL YEAR OF BORRO ER January - Deceber 31 *~ * To 0ftD BANK wash"gon. D C 203- ORke rs DeCtIr-C~6era June B, 1990 MEmORAUM TO rus I\1R DIRECTORS AD ERESIDENT SUBJECT: Project Performance Audit Report on Morocco Second Water SuRgy Pro1ect (Loan 1724-MOR) Attached, for information, is a copy of a report entitled "Project Performance Audit Report on Morocco - Second Water Supply Project (Loan 1724- iOR" prepared by-the Operations Evaluation Department. Attachment V i docment ukat ditbuto an ay be se by weiins on the prrnce of thei oftial dtes. Jis contt may n othewise be disclosed withou W"rl Bak authowibaion ma ona uaoNLY 9OROCCO SECOND SUPPI PROJE (1~ 1724-MORN Prefac........... .............. ..... ..... ...... Baotc Data Sheet ........................... ........ ...... it Evaluattob Sumnary.. ......................... ........ Iv 1. BACKGRoND ......... . .............. .............. 1 nI. TE PROJECT AND IS OJECVES......................... 2 ii IMPN ZTATION . ........................................ . IV. PROJECT .COSTS AND DISUSE . .................. 7 Performanc of Contractors and ultants .... ... .... .. 7 Institutional Developmet ........................... 7 Operattonal,Aspects ..\..................................... 9 Financial å8c ............. ........................... 9 V. RESULTRAND SUSTAINABILITY ... ........ ... 11 Economic Assessment . 12 Environmental and Social Aspects .......'. 12, VI. THE ROLE OF THE ANK....................................13 YII. U"SIONS ANDLESSOS This document has a restricted distrbution and may be used by mdptts only in the petfoniane of t oiea duties. Its comtenws may nototherwise be disclosed World Bank authorisation PROJECT PERPO VC9 AUDIT REORT MOROCCO SECOND WATER SUPPLY PRO1ECT ( A C 172- Thii report ipresents the results of the Performance Audit of the Second Water Supply Project for which Loan 1724-MOR of US$49.00 million was approved on June 7, 1979. The loan was mtde to the Government of Morocco, ut on-lent to the project executing agencies, Office National de 1 Eau Potable (ONZP) and the wmnicipal water and electricity distribution enter- pries (R gies) of Casablanca,' Kenitra and Rabat-Sal The Loan Agreement and Project Agreement were signed on July 2, 1979 and the loan became effective on February 4, 1980.. The loan was fully disbursed, final disbursement . taking place ilt February 1988; the loan had formally closed on December 31, 1987, three and a half years after the original\ closing date. The Project Performance Audit Report (PPAR) consists of a Preface, an Evaluation Summary, and Project Performance Audit Report prepared by tho Operations Evaluation -Department (OED) . OED has re"ewed the PCR, the \Staff Appraisal and Preident' s Reports, the legal 4ocuients and Bank files nd the transcri\t of the meeting of the Executive Directors at which the loan was approved, and has discussed the project with Bank staff.involved. An Audit mission visited Morocco in January 1990 and discussed the project vith goverment and p*oject officials and visited the site- of the project. The Audit finds that the PCR gives a candid and concise account of the ,project expGrience and draws appropriate conclusions, with iich the Audit concurs. The PPAR provides additional comments and colup ions on various project aspects. Following standard OED procedures, copies of the draft PPAR were sent to the Government and the implementing agencies for comments. No comments were, however, received. 1N Li p a it ii j 11, 7 1 1!e Id 4'U @- -§8 W \påte No. o kays in speclixotions Lelr. en=g Field RDrosented 7hrough apprat 4.5/75 1 10 Enineaer 6/715 n.-Fin. An. 7/ 2 8 En.-fin. An. 3/78 3 14 Eg.-2 Fin. An. 6-7/78 4 15 2'Eng.-2 fin. Ani 10-11/78 4 28 1 Eng,-2 F An-1 Cns AprismL throg kbojrd Approv - 4/79 1 10 Fin. An.. 5/79 1 5 Fin. An. oSaerd Approva thruh Effctvenss 7-8/79 2 10 2 Eng. S10/9 2 16 1 Eng.-1 Fin. AK s ervision UWD 2\ 20 1 Eng.-1 Fin. An.a109V 3/81 2 14 1 Eng.-1 Fit. An. 6/81 1 18 1 Eng. 9/81 1 ,7 1 Eng. 1/8 2 14 1 Eng.-1 Fin. An. 5/82 2 14 1 En.4 Fin.\ An, 10/82 2 14 1 Ena.-1 Fin.lAn. 2/8 2 5 1 Eng.-1 Fin. An. 6/83 2 20 1 Eng.- Fin. An. 10-11/83 1 14 1 Fin. An. 3/84 3 24 \ ng.-Fin. An.-Procur. 9/84 3 18ý En.-Fin. An.-Procur. S-4/05 2 17 1 Eng.-1 Fig. An. 4/86 3 11 Eng.-fin. AM.-Procur. 12/86 15 1Eng.'- Fin. An 10/87 2 20 1 Eng.-1 fin. An. 3/88 1 24 1 Eng. 7/88 PCR 2 117 1 Eno.-1 Fln. An. 10/88 PCR 2 -9 1 Eng-1 Fn. An. 9THER P M Borrower: Govermnt of Norocco . ftecuting Ag~nies: G=EP and.the tgIes of Casaiånca, Ke tra and Rabat-So. Third IMter &Mpy Poject Lomn 2061NR USM .ImfItlon m Iationäl ter ygp1y Rehabilitstion Projectoan 2825-KOR U860.0 mittien MOROCCO SECOND WATER SUPPLY PROJECT (LAN 1724-HOR) 1. orocco*a rainfall is irregular and seasonal, with most falling in the mountains, so that water supply in coastal cities depepds largely on sUrface water from the rivers; groundwater is not sufficient to -supply populat4on concentrations once they exceed a certain density. However, use of SAarfgce water involves costly storage, -treatment and transmission pipe-. lines, sometimes over 14ng distances. In 1977,- roughly 40%No Morocco' s pqpulation lived in urban areas; urban populations had in fact been growing at nerythree times the rate of growth in rural areas, reflecting rural mgratio to the cities, of which one result, was the emergence of shanty- towns, where some 30% of urban dwellers lived in poverty. Water and sewer systems were inadequate in many cit es. Though most urban dwellers had somq kind og access to piped water, more than half had no private connec- tp* but relied on publtc standpipes and the. like; ia the shanty-towns, as many as 2,000 people might depend on one standpipe for all their wat4r needs. Sewer systems serv%d only atb)ut one in three .f the inhabitants of larger towns, with sewage discharged untreated into receiving waters.. Parasitic and infectioUs diseases were common in the shanty-towns because of unsatisfactory sanita conditions (PPAR, paras'. 1-3). 2. -orocco' s water .and severage- sector organl:zao is -dispersed among several entities, under different Ministr1es. --ater resourp6 alloca- tions, as well as tariffs, are controlled by central governmentt with Office National de l'Eau Pitable (ONEP) producing bulk water for sale, mainly to municipal distribution enterprises (R4gies) or communal autori- ties. Needing to upgrade water and sever ystems everywhere, but\with lie4ted funds, the Government awarded first priority to urban areas, where theadaittedly costly -investment would benefit most people--in particular, the mid-Atlantic coaspaI) area between Rabat and Casablanca, containing 40% of urban -population and producing 70% of GNP. This area was supplied largely throygh long pipelines from dams on rivers; the most recent such scheme had cbme into service 11 1976 as a result iof a Bank-aided project (Casablanca-Rabat Bulk Water Supply Project, Loan 850-MOR), supplying water from the Bou Regreg river near Rabat. The dam built for that project could supply a further 5m3/second if additional treatment and transmission facil ities were constructed, and the Government has requested the Sank to finance tl4s scheme as the next priority (PPAR, paras 4-5). The Project and Its Obiectives 3. project components included additional treat6ent plaih, pump- ing static'w and transmission pipelines for the Bou Regreg system; but, because of lending program constrAints, the main pipeline could -not b- included in the Bank loan, and it was in fact financed by the Zedeiral Republic of Germany. The Bank-aided package did, however, inclu4e another priority area, that of Greater Aadir where the tourism developments of the.1970a were straining water resources; here again, though, the original subproject had to, be out back; it now covered only improvement of water productiin facilities, the originally pro osed water, distribution and sew- erage components being deferred. Af innovative feature of the project was the financing of a revolving fund in each of the three municipal RAgies in the Bou Regreg supply area, to enable them to- offer credit terms for cer- tain house-connection costs -to low-income householders; hitherto, those costs had to \be paid in advance, which deterred the demand for connections. The project also included studies to optipize ONEP's operations and to inveptigate future water resources for Agadir; and a sector and tariff study.: Project objectives may therefore be summarized as constructing works to meet growing demand and obviate future water shortages; providing finance to encourage private connections and thus Improve access to safe water; and, studies to permit rational allocation of water resources, to review the organization, investment decision proceso and pricing policies -ti the whole sector, and to improve OWP's operational efficiency (PPAR, paras. 6-12). ImpleMnation 1narience 4. Procurement and construction proceede4 competently and smoothly for the, most part; however, the project became inereasingly subject to slippage, mainly because of administrative delays in letting contracts, and staff shortages in ONEP. ' These delays threatenidp water shortages in the Agadir area, mhere project slippage reached two years by 1980; and, by 1982, increased water demand in 'the coastal region likewise threateped. shortages in that area within three years. Both threatp were averted by technical ejidients, involving additional works in the case of the Bou Regre& aponeut. The sain' Bpu Regreg works were virtually completed by June 1984, -but the Agadir conponent took anwtb4 year and-.a half. The' household connections program also 1agged, but eventually more connections were installed than had been forecast at appraisal Junevenly distributed, however, among the thrdo Rdgies). The irogram was so enthusiastically received that it has bebu repeated and extended to other #reas under a subsequent project. Final completion of this project came in 1987, about 4-1/2 years late. the sector study was, after all, not carried out under this projet but the other studies were duly completed, with grat#fying success in the case o theTlNBP efficiency studies (PPXR, paras. 7-20). 5i Much of the delay under this project appears to have been attribu- table to cumbersome pproval procedures and to transmission delays in the Treasury, through which loan disbursements,passed; the Rdgies in particular had great difficultyA4n securing, timely reimbursement of qspenditure. The delays in payment were at times apparently aggravated by cash constraints in the Treasury, and those in approval of contracts, by conscious policy of the Gprernment to slow down investment for fiscal. and econobic reasons. Extenhles of the closing date were also needed ty permit completion of the house-connections program (PPAR, para. 21). 6. Project costs were about 53% abqve estimates .in terms of Dirhams, of which the PCR estmates about half to be due to exchange ratq factors, and the remainder to the time overrun and" to project modifications. '(No figures are available of costs in U.S. dollars.) The excess wap made good almost entirely by increased government equity,' ONEP*s contribution from internal cash geneatIon being negligible. Disbursementa also lagged cor- respondingly; by the original closing date, only about two-thirds of the loan had been drawn vdown (PPAR, paras. 22-23). 7. Contractors and consultants performed satisfactorily for the most part. Institutional development was likewise good, with ONEP and th Rdgies performing competently in their respective spheres; ONEP managed effectively to keep its operating costs and staff levels down. Project reporting was timely and of good quality, and inter-agency relations appear to have been reasonably harmonious. However, ONEP's accounting was defec- tive and substantial consultant input was needed to \modernize the system and improve efficiency -(PPAR, pares. 24-30). 8. Two financial constraints hampered the project executing agencies continuously--inadequate tariff levels, and non-payment of bills by p4blic authorities. There was a rate of return covenant under the loan; and to meet this, ONEPs rates would have bad to be increased by 35% iokediately. Howe*er, the Bank chose not to insist on a tariff increase, and instead adjusted the accounts to reflect a notional credit for taxes on water bills, assumed to be reftinded to ONEP as part of .u.apital grants from the Government; by this device, the covenant' was for the time being satisfied. Thereafter, 'tariff adju trants proved persis difficult to secure, until 1982 and ONEP never achieved the target rate of return in any year; the requirement was ruvised under a subsequent loan to a cash generation covenanti,but the problem of inadequate tariffi is still unresolved. The Rdgies, too, are losing money on water supply, made good from electricity profits. In the meantime, ONEP's revenue shortfall has been- made good by Government equity contributions in the billions of Dithama, Paradoxically, the other financial problem subsisted in the failure of 9 erment aqd other institutional consumers to pay bills promptly, despite covenants the contrary, so that build-up of mutual indebtedness persists, in spite of continuing eforts to regulate the maatter, including periodic lump-sum payments (PPM, paras. 32-36). Results and Sustainability 9. Th project achieved its primary objective, in that watqr demand in the projott areas was successfully met, and no water shortage's arose. The house connections program was likewise successfully implemented, (though unevenly, and with some financial loss to the Rdgies). ONEP and the Rdgies are considered to be mature bodies, competent to plan ahead and execute necessatry works and to maintain exipting installations satisfacto:- ily. To thisextent, the project benefit* appear to be well sustainable (PPAR, paras. 37-38). I-ooatei and Enviroxniental -Asees 10. \ The economic juatification for this project rested upon the fact that tfie then current water tariff reflected the long-run average incremen- tal cost of ater at an appropriate discount rate, and under a prqgressive tariff structure, the consumers continued to demonstrate their\ willingness to pay this cost. Tariff reviews in Norocco continue to be based on the incremental cost concept; however, if the tariffs ate not imposed, or ate '- vil - delayed, consumers' willingness is not tested. No detailed evaluation of pObject costs and benefits was provided in the AR and\ has not been at- tempted in the PCR or by the Audit, because of\ overlapping costs and bene- fits of non-project components PPAR, 'paras. 39140). 11. Water and sewerage are vital components of environmental concerns; and although this project contributed to Improvement of the environment in avoiding general water shortages and in providing safe water by means of s household supply, to thousands of low-income residents, it did not address the problem of inadequate sewerage. However, a number of follow-up Bank projects in Morocco have contained sewerage components, including a major project for Casablanca. One environmental improvement proposed in the -project tas, however, not comprehensively carried out, namely, the provi- sion of additional standpipes in shanty-town areas, due partly to the dif- ficulty of securing payment for the water so consumed (PPAR, pares. 41-43). The Role of the Bank 12. The Bank played a prominent part in identifying and preparing the project; projections were accurate and much attention was paid to-formulat- ing the hou.e \connections program. Supervision of the project, rombinad with prepatatiobi of subsequent projects, has .resulted in a sustained pres- ence in the , Mdrocco water an# sewerage sector, over many yas. an1 has produced a fruitful dialogue and a harmonious relationship with counter- parts in Morocco. Rapeated representations were made (though not always successfully) over matters which seemed to undermine project benefits, such as inadeq,a te. tariffs,, unpaid bills and slow transmission of loan withdraw- als; in -an effort to alleviate the last-mentioned problem, thy bank intro- duced a system of telexed disbursement notificati6ho, increased disburse- ment percentages under a Special Action Program and recourse to direct paymet, procedures. The project presents a good example of cooperation between the Bank and executing agencies (PPAR, paras. 21 and 44-45). 13. The Audit suggest ,two lessons from this prjaect (apart from t o 'routine and* obvious ones). First, projec, officers should not close their minds to extraneous 'lactori outside the narrow confines of the project, which may undermine the project objectives- -suh asnational economic 0ol- icy and government aduministrative bottlenecks; these must be allowed tok, and preferably addressed through sectoral or structural adjustment, pperA- \tions. Sepond.y, #bere studies affect A number of >aencies, (some outsi& the project, or even the itor), and authoFitative coordinatton panel is essential. 0f V) A 0 V x, ,.N -' - Q R C PERFORMANCE AUDIT REPORT MOROCCO SECOND WATER SUPPLY PROJECT (1OAN 1724-MOP) I. tt R 1.01 The Kingdom of Morocco, situated on the northwestern corner of Africa, has four distinct climatic zones--the coastal areas, the mouttains, the 'interior plains and the desert. Rainfall is irregular and seasonal, most falling in the short winter period, and therefore water storage, eith- er natural or artificial, is necessary to compensate. The rainfall in the mountain areas is two to three times that received on the coast, so that available water is 'unstly concentrated in rivers (21 km3/year, as against only 4 km3/year of exploitable groundwater). Groundwater supplies are not sufficient to supply population concentrations once they exceed a certain density, and recourse must therefore be had to surface water, which may have to be transported long distances from the reservoir to the cities supplied; In addition- to the cost of transmission, surface water also needs expensive \treatment for human consuption. In Morocco's mid-Atlantic coastal region (the main project area, for Loan 1724-HOR), water may be transported as far as 135 kilometers. 1.02 In 1977, Noroccl's population was approximately 18 - million, of whom about 40% (7 million) lived in urban areas. this comparatively high degree of urbanization seemed likely to continue; wrban population had b#en growing at 4.S% annually, compared with only 1.8% for rural populatich, reflecting migration to urban areas. One result of this transfer of popu- lation was the emergence of shanty-towns; 30% ,of urban dellers were esti- mated to live in these slums, with incomes below the povetty level. 1.03 '-At the time of appraisal of this project, water systems in many cities and population centers were operating at maximum capacity. Though almost all urban residents had some kind of access to piped water, only 45% had house connections (in rural areas, only 5%); the rest depended on standpipes or public sources, such as mosques or water vendors. The situa- tion v4s -particularly bad in the shanty-towns, where a standpipe sometimes served as many as 2,000 consumers, from as far away as 500 meters., sever systems were. even more inadequate; in the large towns, only about one in three of the inhabitants were comntedj and sewage treatment was almost non-existent. The sanitary conditions in 'the shanty-towns thus contributed to a high incidence of parasitic and infectious diseases. Water quality, however, was good in urban areas, though of poor bacterial quality in many rural areas (because of contamination from inadequate wast disposal methods). .04 Sector organization was, and is, somewhat fragmented. The Hydraulics Department of the Ministry of Equipment' controls water. re- sources, with final allocations being made by a sppcial commission in the Prime Minister's Office. The main water production 'agincy, Office National. de l'Eau Potable (ONEP), under the same Ministry, sells bulk water to muni-' cipal enterprises (Rdgies) in the lmain cities and towns; the r6gies, whih Ik -2- come under the Interior Ministry, also distribute electricity. In other urban centers and rural areas, distribution is by ONEP, the municipality or (in unchartered rural agglomerations) the commune, under the Ministry of Agriculture. Goverment policy, however, has been to create more Rdgies, to enable ONEP to concentrate on water development and prod ction. Tariffs are controlled by the Government. 1.05 Faced with the need to upgrade water and sewer systems almost everywhere, but with limited funds, the Government had awarded highest priority to the main urban areas, where the admittedly high cost of stor- age, treatment and transmission would benefit most people; and of these urban areas, the xid-Atlantic coastal area, between Casablanca and Kenitra, and first preference, because of its concentration of population and indus- try: the area contained 40% .of urban population, and produced 70% of CNP. Growth of water demand here had required a succession of long transmission pipelines from surface water sources,, of which the most recent entailed supply from a dam on the Bou Regreg river, near Rabat; the dam, pipeline and ancillary works were financed with assistance from a 1972 Bank loan (850-MOR for the Casablanca-Rabat Bulk Water Supply Project), the main component have entered into service in 1976. This project increased avail- able supplies by 4m3/second; however, the storage impounded by the dam could supply a further 53/second (9m3/second if the dam were raised to its intended full height). II. THE PROJECT AND ITS OBJECTIVES 2.01 The Seccl water Supply Project had a long gestacioh period. From as early as 1972, various possible components were being considered and discarded, because they were thought not to be in a sufficient state of readiness, or because alternative financing was available. Eventually, in 1977, the Government asked for 4 second Bou Regreg project to be the main item, because of the need to increase capacity in the mid-Atlantic area by the end of 1981--three years earlier than forecast at the time of Loan 850-MOR, demand having risen more rapidly than expected; together with any other components for which loan finance might be available, The city of Agadir was anotpher priority; rapid tourism development in the 1970s looked like producing water shortages by 1980 in the absence of further invest- ment, and accordingly a component for -water production,. distribution, and sewerage for that city was identified. Fluctuating constraints in the lending program, however, gave rise to a proposal to defer the Bou Regreg pipelines and treatment plant altogether (since the feasfbility studies might not be ready in time), and concentrate Bank assistance on Agadir; hvt subsequently the Government pressed for the whole Bou Regreg project to be appraised, as the studies were now ready. Eventually, in view of the lend- ing program conitraints, the Agadir component was limited to water produc- tion facilities, and the Bou Regreg project reinstated; however, the lat- ter's pipeline componer; would be financed by the Federal Republic of Germny 2.02 Normally there\would be some risk in appraising a major project of this nature on feasibitM-ztudies alone\; however, in this case, the works in the Casablanca-Kenitra area were t6 some extent a duplication and rein- forcement of existing similar installatony . \,The ali et, rorund condi- 3 tions, and li"kly operating costs would therefore be already known fairly precisely, and the works Involved had already been foreseen at the time of apprais4 of Loan 850-ROR. For this reason, only 10% was added to base cost estimates for physical cortingencaes; nevertheleqs, this provision seems on the low stde. 2.03 The physical works included in the project thus comprised: Bulk Water Suply. Casablanca-Kenitra: Facilitiq- to utilize the additional 5m3/sec available from the Bou Reg: 4i, including a treatment plant, pumping stations, and a new 4 pipeline to Casablanca, with additional pipelines to Rabat & -,41ance to Sal6; and ancillary works and buildings. Agadir Bulk Water Suslv: A well-fixed collection system, equipment of 12 boreholes with pumps, 38km of transmission mains, a treatment plant and chlorina- tion station, and a balancing reaervoir; with ancillary facili- ties, including power lines. 2.04 For both schemes, the sizing and location of the facilities were, selected from a number of alternatives as providing the least-cost solution at appropriate discount rates. Total estimated cost (including other com- ponents mentioned in the next paragraph) was DH 697.23 million US$182.52 million equivalent) with the Bank contributing US$49 million, and the Federal GermaAt Government US$33.34 million equivalent; thus jointly cover- ing the total estimated foreign exchang .cost of US$82.34 million. The project did not, include any works on the retail distribution side of the system, since 01EP and the Rdgies\were considered t4 be well-equippe4 and competent to execute distribution system extensions, as and when neided, with their own resources. 2.05 An innovative feature, to which the Bank devoted much attention, was finance for a revolving ifund within each of the three RAgies within the Bou ReSreg supply area (Caablanca, Rabat-Sal4 and Kenitra), to provide credit facilities to poorer houndholders and thus encourage the increase of house connections. Unde* existing arrangements, the cost of a connection could be paid in installments by low-income houspholders, but t*e contribu- tion to the distribution network (typically at last US$380) had to be paid in advance (unless already financed by- a developer); and this requiremet deterred demand for connections. Eligibility criteria for entitlement tb this facility were agieed, but proved too rigid and had to be twice amended (under the Third Water Supply Project (Loan 2006-MOR) slightly- more flexi- ble criteria were agreed). This component was expected to tZadfit 48,000 households by 1983 and increase the connection rate to 80% in areas where individual supply was feasible. It did not, of gourse, solve the problem in shanty-towns, where house connections were not \sually practicable;' but the loav condi ions included an undertaking that the U4gies would be\ requtred to pro*ide more staidpipes to meet the needs of this population group (Loan Agr6ement Section 3.-0llf)).. In all, some 300,000 people below the urban poverty threshold would, for the first ti e, have running water. The finance for Oe revolving funds were advanced by the Bank and the Government in equal shares. 2.06 Also included in the project were a number of important studies: (a) A water resources management study and investigation for Agadir, where the source, bf supply after 1987 was still not identified, to be crried out under the control of the Hydraulics De/partment of the Ministry of Equ4pment. (b) A study. to optimize the water treatment process at the Bou Regreg (i*luding training of local staff) And thus minimize use of chem- tcals. (c) An efficiency review of ONEP's operaiions in minor centers, where losses were incurred on non-remunerative small oupoly systems. A sector and tariff study (not BaM*-financed), to be' carried, out by he Pqme Minister's Secretariat for Economic Affairs. 2.07 'The project and its objectives may therefore be summarized as follows: (a) Physical works (and, in Agadir, research) to meet growing demand and future needs. (b) Facilities to encourage house connections for low-income 'gitoups, and thus improve access to safe water, whilst permitting continued effective self-financing by the R6gies. (c) Studies to permit rational iallocation of water resources, to im- prove ONEP's. operational e ficiency, and to review the organiza- tion, investment decision process and pricing\,policies in the whole sector. Other institutional aspects were addressed by loan conditions. 2.08 These objectives were well conceived and appropriate; and the project was timel (except perhaps insofar as it was appraised on the basis of feasibility studies only, which entailed some minor engineering revi- $ions after completion of ,etailtd design). 2.09 The project was prepared by ONEP and its consultants, with inputs- from a number of Bank missions; appraisal took place in October-November 1978, and on\June 7, 1979, Loan 1724-MOR for US$49 millio4 was approved (an earlier allocation of US$41\ million for the poject having been increased in order to finance, inter'alia, another 13,000 connections). The borrower was the Kingdom of Norocco, and the beneficiaries ONEP and the three R.gies . Loan effectiveness was delayed for some four motths, to February 4, J 90, due to delays in finalisipg, legal documentation, ii Morocco. 3.01 Consruction of the physical works in the project 'appears to have been competent and trouble-free, except that two\contracts had to be re-let through failure of contracting firms (June 1985)', which caused a 9-months' delay in completing the components concerned. The project files disclose no problems or disagreements on tecwcal matters, and pro;urement went smoothly, ONEP being by this time experienced in this field. However, a considerable amount of time was lost in starting up; already by June 1980, both the Bou Regreg works and the Agadir component were reported to be a year behind schedule, attributable to staff shortages; the final design and bid documents need0a more revision than expected, and, becapse ONEP's staff was somewhat overloaded pending engagement of consultants for construction supervision, comments could not be made quickly. Another cause of slippage was the administrative delay in letting contracts (the Tender Board took 6-8 months to award the Agadir contracts) with a consequent risk of higher costs, and possibly eventual water shortage in Agadir. By November 1980, the estimated slippage on the Agadir c9mponent had reached two years, again because of delays in the award of contracts; but the threatened water shor*age had been averted by various technical expedients. It was -not, until mid-1981 that 4 contracts for the Agadir first Phase were let; the second Phase Was held up awaiting the results of a study of a certain aquifer which might obviate the need for surface -water and a treatment plant. (In the event, these were preferred.) 3.02 ' By 1982, increased demand in the coastal area threatened water shortages in 1985; the demand projections in the fpasibility study were being updated for the Third Water Supply Project, and it now appeared that a 'high" variant calculated therein was proving more accurate than the Omediup* variant actually adopted. Again, however, the threatened shortage was averted, thi time by adding a booster pump and an additional length of force nain to the project, and by rehabilitating an existing groundwater field. 3.03 By June 1984, the main Bou Regreg works were virtually complete4, except for some operational buildings and staff housing where the contrac- tors had defaulted, and for groundwater rehabilitation. However, the Agadir component was estimated still to require over a year more to comple- tion,\and the study not until early in 1986. The house connection program was also lagging. Final completion came at the ebn of' 1987, about 4-1/2 years later than forecast at appraisal. The \loan had already formally closed on -June 30, 1987, but disbursements continued until F bruary 1988, the loan\account being then funll disbursed. 3.04 The hou ehold connectiow program carried out by the R4gies like- wise got off to a belated start, with the subsidiary agreements between the' Government and theelves not executed until nine mpnths after loan signa- ture. Thereafter )progress varied; by August 1981, Xenitra and\ Casablanca had completed only about half the number of connections, forecast for that date, but Rabat-Sal6 had exceeded the target. One problem' faced by the two first-ipentioned. R4gies was how, to deal with leased apartments, where the water connection Ight involve the lessees in higher rents; and the PCR notes other constr ints, (para. 5\03). All Rdgies had difficulty in getting funds passed on'by the Treasury (discussed below). By projec. completion, N ' . NV k, -6- Casablanca still had not achieved much more than half the target program which had been overestimated for that city (some districts were equipped outside, th project, by other ppbitc ageties) but the other two had ex- ceeded theirs; in all, 52,000 conections were Installed, (serving 417,000 people) compared with 48,000 forecast at appraisal. The credit facilities were enthusiastically received by prqspective consumers, and, . under the Third Water Supply Project, a similar program was extended to ONEP s own consumers am'# those of eight other Rdgies, as well as providing funds to meet the unsatisfied demand in the three coastal R4gies. 3.05 The provision of additional standpipes called for under the proj- ect appears to have been hampered by the Inability of the R4gies to secure payment from the municipalities for the water consumed at standpipes; and the requirement to extend these facilities was not repeated under the Third Water Supply Project. In two of the Rdgies, there are not actually fewer standpipen,\due to the expansion of prt4ate-connections. 3.06 The studies comprised in the project were (with one Important exception) carried out--in the case of those aimed at the operational effi- ciency of ONEP, with gratifying success. The ieservoir management study resulted iO important cost-cutting measures (e.g., by use of different chemical combinations to combat eutrophication of stored water, which had previousoy require high chemical dosage of treatment, and skillful manage- ment of keservoir dapacity to benefit from off-peak power rates); and the review of operating proceduresi in smaller centers likewise produced Impor- tant savings. The water sectok study was, however, not carried out by the govetament agency responsible, except for the ujdating of a tariff study; the review of sector organization was deferred and now forms part of the National Water Supply Rehabilitation Project (Loan 2825-MOR). N3.07" References has been made to delays in approval of contracts and in \settlement of sums due on account of expenditures, and hence in payments to contractors, consultants and Rdgies. To some extent the delays were due Merely \to cumbersome administrative procedures, since the loan proceeds had to pass through the Goverment Treasury; though, in later stages, it ap- peared that the government-appointed financial controllers of the sector organizations were under instruction to hold up approval of contracts, since the Government wished, for fiscal and economic reasons, to slow down investment expenditures. Delays in passing on loan disbursements plagued thq project authorities throughout; the Bank ascertained that it could take up to nine months to reimburse a Rdgie for expenditure incurred. Admini- straive impediments are difficult to short-circuit ig they are part of what.a government perceives as an essential element in the system of checks and balances; the delays under this project seea, however, to have been exacerbate4 by the financial constraints under which the Moroccan Govern- ment labored during later stages. of the project- -indeed, at one point, it sepme4 to the Bank as if the Treasury regarded Bank disbursements as a kind of cagh float. The Bank did what it could to resolve this 'difficulty, by establishing a time-consuming system fimikediate notification of disburse- ments, direct by telex to phe ultimata recipients, so that ;hey could chase up the proceeds; by persuading the Government to agree to T 4irect payment proce4ures as from 1983; and by increasing disbursement percentages under a Special Action Program in 1984. 7 Iv. mus cosm7- avABT M =COSTS, "N IBuREKimS 4.01 The table of "Project Cost and Financing" in the PCR (p. 21) i indicative only; it shows costs only in dirhams, and project financing only In dollars, at the exchange rate ruling at appraisal. The conclusionO to be drawn must to that extent be inferential. The table shows project costs,, including the additi nal Bou Regreg components, as DR 1063.8 mil- lion, compared with DH 697.2 million estimated at appraisal, an overrun of 53%. About half the cost overrun in ditrhams is attributed in the PCR to the progressive devaluation of that currency against costs in foreign ex- change, with the remainder arising from the time overrun and from project modifications. The excess cost was made good by greatly inreased govern- ment contributions, equivalent \ to 70X of total costs., compared with only about 50% forecast at appraisal. A similar picture. is presented by the table of ONEP's funds flow during 1979-87 (PCR para. 6.06); of total in- vestment expenditure of DH 3175 million (compared with DH 3331 .million forecast, including updated projections used for the Third Water Supply Project), 70Z was financed from goverrment contributions, as against only 46% forecast. ONEP's own contribution from internal cash generation was negligible (6%) for reasons discussed under *Financial Aspects. 4.02 The slippage in the construction peogram caused disbursements to lag correspondingly; appraisal forecasts showed 85% of thi loan to be drawn down by aid-1983, when the bulk of the works were expected to have been. completed, but in fact less than one-third of the loan had been disbursed by that date (PCI, p. 15). Apart from the delays in execution mentioned earlier, ONEP was initially. slow in applying for disbursement; one early component (the booster pumps) was An fact procuted and installe before ONEP submitted the first vihdrawal Application. 6NEP thereafter a-&eed to make maximum use of the Bank's direct payments procedure, in order to avoid the build-up of large balances of undisbursed funds not cArresponding to the progress of the works.. By the original closing .date (June 1984) only about two-thirds of the oan ha# been drawn down; final payment did not take place until February \1988, d6spite the increased disbursement percent- ages applied under the Special Action Program. Performance of Contractors and Consultants 4.03 With the exception of two contractors .ho became bankrupt and had their contracts re-tendered, performance of coitracts and suppliers, both competently, and the PCR (para. 11.05) notes *xamples of flexible a-Ast- ments of consultants' contracts by agreement to impr?ve projet execution. InstitutionalDelomt 4.04 The principal executing agency under the project, 01E, had been created in 1972 from its predecessor RIX (dgie. des Exploitations Industrielles), an entity responsible for 4 number of dispar*te and uncon- nected activities besides water supply. ' The. extraneous activities were eliminated under the first Bank-assisted project, leaving ONEP free to concentrate on building up the managerial, technical and financial re- sources required for a large scale potable water producer. By the time of appraisal, ONEP had acquired a consideradle expertise in technical tters, -8- though financial and accounting services were still inadequate. A new organization structure had been implanted. including separate technical departments responsible for large projects, small-to-medium projects, land operations respectively, and this\has proked effective. A training divi- sion had just been established, and this gradually provided a continuing stream of qualified technicians. 4.05 In the accounting field, the defective ianual system was over- hauled by consulutantsnder the first project, and eventually computerized, though not without mahy teething problems and necessary modifications; and ONEP found ift necessary to hire additional consultant in 1982 to oversee the computerization of cost accounting, and also to complete the inventory ,of assets--a cause of repeated qualification of additors' reports. In addi- tion to the studies under the project, ONE# also established a commission to examine operational efficiency generally,. with good results, including an innovative system whereby important monitoring indices are computerized to produce warning signals when the indices get out of line with predeter- mined limits. $taffing was also controlled, to 4bout 17% below appraisal estimates for 1985, and operating costs per m3 of water produced rose by only just over half in the ten years to 1907, despite greater inflation (the wholesale price index near4y doubled in that time). 4.b6 As part of the Greater Agadir component of the project, the Bank urged from the outset the creation of an independent inter-communal R6gie to be responsible for .ater supply, electricity distribution and also sew- erage, preferably with greater autonomy than the existing R4gies, and which could serve as a model for future R4gies. The Rdgie was eventually estab- lished (though it did not take over sewerage) and the Government has con- tinued to create municipal enterprises, to permit ONEP progressively to concentrate its resources on bulk water supply alone. 4.07 Inter-agency relationships appear to have been generally harmoni- ous; an# c6operation between ONEP and the Hydraulic Department contributed to effective dev lopment of low-cost groundwater supplies. Customer-sup- plier relations between the Rdgies and ONEP must at times have been diffi- cult, in view of the tight financial climate (see "Financial Aspects" ) and the PCR mentions some disagreements over use of water resources (PCR, para. 10.01). However, the Audit was informed that relations arq not improved, especially sinct the appo ntment of a government Director of Rdgies ad Conceded Services. 4.08 ONEP, besides being subject to government control over major pol- icy matters, such as tarif s and budgets, is also, like other public organ- izatigns in Morocco, subject to "a priori" control of expenditures by a finandial controller and an accounting agent. At appraisal it was consid- ered, based on experience under the previous project, that in practice the system does not detract form ONEP's operational autonomy. As already noted, however, there were times when the fir\anciai controller, whether as an act of policy or otherwise, held up the award of contracts. The Audit understands that a relaxation of controls is under consideration. 4.09 Reporting by ONEP on project execution was in general timely and of good quality; the files disc ose few instances of the 'ank havtng to .9- pursue outstanding matters. Comp tance by ONEP with loan covenants was, on the whole, satisfactory (PCR, pp. 22-23). orrtional AsMectS 4.10 The project succeed d, in both the areas concerned, in meeting water demand at all times. A temporary surge in demand in the coastal region in 1980-81 was successfully met; subsequently- demand fell below the appraisal estimates and no water shortages have developed. Water quality is goodi t6ough slightly aggressive; except that in Agadir 102 of samples recently were found unsuitable for drinking, for reasons being investi- gated. Unaccounted for water has not been a problem in ONEP' a system, except briefly in Agadir (due to #qu1ty meters): and it is understood that in the other Rdgies in the. project area reasonably satisfactori network efficiency prevails, with the enterprises managing their systems competent- ly. * The Audit visited the Bou Regreg complex and found the entire opera- tion to be impeccable. No details are available in the project documents of service levels in the project area as a result of the project, or of sales by'consumer category. Such statistics would usually be included in the monitoring in#icators, but in this case, where the main project entity sells 90% of its production to other authorities, !phis would . have been inappropriate. Information obtained by the audit shows that in the three Rdgies in the house connetion program, populations served by house connec-. tions were as follows: No. of Pri.te Customers ( 00) 12A 1211 1M211 Casablanca 163 237 312 Rabat-SalU 48 89 131 Kenitra 14 19 30 I of Ponulation Connected feat.) A J,flg 12g Casablanca 60 - '69 Rabat-SA6d 60.15 67.4 Kenitra 57.1 73.9 Ja Source: Goverrment Report, April 1989 (may be slightly underestimated). Financial Asnects 4.11 Execution of this project would have been unusually uneventful and trouble-free had it not been bedeviled by two serious constraints, both somewhat outside the contro. of either the project agencies or the Bank. One wasa the inordinate degree of delay imposed on wA4 project activities,- from ,ontract approval to release of funds, by cumbersome VVocedures of consultation and approval. This\ problein has already been outlined in the section Implementation". 442 -The other stumbliiR-block co*erned finandal transactions--in patticular, tariff levels and payment o ,jills -by public bodies., Tariff levels in Bank-assisted public utilities projects are usually the subject - 10 - of a rate covenant intended to ensure that, after meeting operating ex- penses (including depreciation), surplus revenue remains to cover'interest payments and finance a portion of capital expenditures, thus 'reducing re- course to 'borrowing or to equity contributions from the often overloaded government budget. In this project, A rate of return covenant was agreed: ONEP was to earn a surplus, of 21 of the value of its net fixed assets up to 1980, ris4ng to 51 in 1981 and 7% from 1982 onwards. This was a slight modificatiion of a similar undertaking under the first Bou Regreg project, to reflect the fact that ONE? had not, as expected, taken over a remunera- tive private water concession. Appraisal projections assumed, perhaps optimistically, annual tariff reviews to compensate for the effects of inflation on operating costs (estimated at 11% in 1979 and h% a year there- after), which would enable ONEP to meet its obligationi; and the Government agreed as a loan condition to take all measures needed to enable ONEP to estaalish and collect charges at the necessary levels (Loan Agreement :Section 4.03). No specific mention appears th have been made, however, of any obligation to allow the Rdgies to revise their retail tariffs in step with increases in the rates they had to pay ONEP (a frequent oversight in B"~n-financed water supply projects). 4.13, ONEP's accounts hA shown a not loss in 1977; despite a substan- tial tariff increase 1i6 that .year; an lamediate further increase 6f 35% was 10dicated to permit ONEP to meet the 2Z *ate of return target. However, t4is requirement was not pressed by t*e Bank; instead, the appraisal report adjusted the accounts to offset losses on ONEP's rural water systems, by inserting a notional credit for taxes on water sales-handed over to the Government, but considered to be refunded to ONEP as part. of annual capital grants. By this expedient--admittedly not an unreasonable concopt--the 2% commitment would be satisfied. In the. light of subsequent experiece, and' indeed of most of the Bank's previous experience in public utility\,proj- ects, it might have \been better to have used the leverage of loan negotia- tions to secure the 35% "riff increase instead. It is always difficult to persuade government to authorize higher utility _charges n their citizens, regardless of commitments 6der loan covenants;and ONEP's average revenue per %3* of bulk water sold remained 'virtually static through 1982; there- after, annual increases took effect, sometimes substantial\ (though in one instance the R6gies were ordered' to hold back on the corresponding retail increase). Nevertheless, ONEP never achieved the 2% rate of return in the years up to 1987 (even with the aid of the notional tax credits); indeed, in five of those yea s, the rate of return was negative. ONE 8s accounts for ,the years to 1987k are summarized in the PCR (pages 16-18). 4.14 As a result, ONEP's balance sheet at the end of 1985 (the horizon used at appraisal) showed an accumulated lost of Dh 483 million, compared with the appraisal forecast of DH 127 million surplus; the shortfall was more than uid good by Government equity contributions of DH 2.8 -billion (as against Dh 1.3 billion forecast). Such over-reliance -on the Treasury is not conducive to autonomy of enterprises, nor can it be appropriae for a goverikent Xith limited invtent funds; there is no reason why the taxpayer should subsidize the fortunate posessor of a piped water supply. Under Loan 2006-MOR, the rate of 'return covenant was replIaced by-a- require- sentat ONEMP--a operating expenses =Should not exceed 662 of operating revenues; but since this 'working rat4ow formula taks no account of hap- tal needs. depreciation of interest crges, its main effect in a climate of tariff restraint is no more than to keep pperating costs down. In Loan 2825-NOR a fresh requiremt was introduced, or ONEP to generate internal- ly a specified proportion of its capital requirements. This seems a much more appropriate mechanism. However, the Audit examined ONEP' s accounts for 1988 and these st.ll showed a loss for the year and an accumulated deficit of DH 580 million. The R4gies also show annual deficits, subsi- dized by profits from electricity sales. An up-to-date tariff study has, however, just been completed. 4.15 The other financial aspect which plagued the Noroccan water supply sector (and others) during the project period was that of unpaid bjlls owed by public agencies and municipalities- -the result of cumbersome budgetary and disbursement procedures, compounded by shortAges of funds. Despite the usual loan covenant to pay government bills promptly, this issue preoccu' pied every- supervision mission, and bulks large in the project files., From time to time, lump-sum payments are effected together with other measures, such as issue of Treasury Bonds in satisfaction of claims. and the cancella- tion of mutual accounts outstanding; but the arrears continue to exercise ONEP and the R6gies. Some administrative improvements were implemented (such as authorizing cut-of of certain departments. for unpaid 'bills, and, requiring public sevants to pay for domestic consumption). By the end of 1988, ONEP's accounts receivable for water sales totalled DR 451 million- -about nine months' revenue from water sales-largely due by the Rdgies (themselves owed by aunicipalities and government departments). This does,. however, represent an improvement over the situation in some previous years. and efforts to clear the 'reciprocal debts -are- continuing. Arrears from ONEP's private customers appear not to be a problm. V. RESULTS ANDSUST&AINBLIT 5.01 The project achieved its primary objedtive in that water demand in the project\areas was successfully met, the investment program being modi- fied to meet unforeseen short-yerm exigencies, and no water shortages arose. ONEP and the Rdgies have shown themselves able to plan ahead and execute necessary works in a timely manner to -meet the needs of -the popula- tions they ,serve, subject to availability of finance, and to maintait the existing installations satisfactorily. To this -7xtent, the project -bene- fits appe t- be wAl sustainable. Institutional capability is li-ature, and, in the case of 01?, has grown steadily since its inception. 5.T2 The house, connections component of the project was likewise exe- cuted to -an extent -beyond that forecast (though unevenly distributed among the the Rdgies) and is being repeated under a subsequent project. Though the expansion of the population served by house. connections benefitted the R4gies, in having more paying customers able to command a steady supply, some financial -loss was incurred on this pXogram for reasons outlined in the PCR (para. 6.09), and, if not remedied, ihis could be a distacentive to continued improvement of service levels. There is also a. continuing loss, in that ionsumers who would have previously used standpipe=ater, paid for--- by the municipality at more than he bulk supply price, now have private connections and their. ,onsumptiop falls almost entirely within the_ Xowest lif-eUne- tariff bracket at lesi than half the cost to the R4gies of, the i . -12- water consumed. The revolving funds are also faced with loan repayments in foreign exchange. 5.Q3 The project as appraised represented the least-cost solution for exphading \vater production facilities in the project areas, at discount rates Of 89-12%. No detailed evaluation of economic costs and benefits, appears in the appraisal report; but projections had been made in tariff studies, ",hich, showed that the then current, retail. tariff equalled the long-rk aerage incremental cost of water at a disciunt rate of 8%, and ainde, qWder\ a progressiv tariff structure, consumers continued to demon- strate their willingness to pay that price for additional water, the in- vestmen*s were consfided to be economically justified. As, is usual in water supply projects, no account was taken of benefits difficult to quant- ify, such as improyed public health and the social benefits to the urban poor. Calculation for tarike reviews in Norocco -continue to be based on long-range incremental costs, though if the tariffs are not imposed in a timely manner, consumers' willingness to pay is not properly test6d. 5.04 'n economic evaluation of the completed project -as not attempted, since its benefits cannot be disaggregated from t*oa_of parallel ,and over- lapping projects (by, for example, the R4gies). Environmental and Socil Aspects 5.05 Any increase in the volume of -ater supplidd to a. community gives rise to a corresponding increase in the volume of, waste water to be dis- posed of; and, in the project area, sewerage And sewage disposal already. -lagged behind water supply. Indeed, in some rural areas, the quality of the water -supply itself was impaird. by pollution from unsanitary domestic waste disposal. The problem was equally acute in the wuban shan -town Axeas, ith vongested habitation, inadequate 4rainage nd few tublic toiletre Even- where sewer systems existed, effluent was' discharged un- treated into receiving waters. 5.06 Vhile this project did not directly adress the sanitation aspect, the Bank was well aware of the situation, and in collaboration with the Moroccan authorities has mounted a number of initiatives to 1amedy it. A iiajor US$116 mill1on roject for Greater Casablanca Severae (Loan $826-MOR), preceded by an Engineering Loan, was prepared in parallel ith the National Water SpPly Reha ilitation Project (Loan 2825-MOR) ,both approved in 1987. Previous Oroj ects has also contained sewerage coMponents, including the Bay of Agadir Tourism Projeet-(rnnc202A(RY the Rabat Urban Project (Loan 1528-MOR), the Second Urban Project (Loan 1944-OR1, the First Housing Project (Loan 2245-MOR) the ?Ilot Project for a nomtMnal rnfiastructur* Fund (Loan 2272-MOR). ver Oent policr is now to useVerag4 progressively to Afgies, begitning Casablanca. 5.07 The haie connections qomponent of the project -also. contributed to the improvement of environmentl conditions, in that inhabitints of the poorer areas were enabled .to acquire a household supply (either individual- ly or in conceri- with neighbors), Insted of having to add to the conges t&n=and water, spillage- around public standpiphs, and to;carry thw ater K 13- long distances with thv risk of pollution en route. Alternatively, they would no longer have to pay infloted prices to v,#ter vendors, and thus had more to spend on food, for example; so that this omponent had social \over- tones. VI. THE ROLE OF THE BANK 6.01 the Bank's role in the water supply sector in Morocco ho.. been a sustained one, beginning with the preparation of the Casablanca-R4at Bulk Watek Supply Project in the early 1970s, and continuing through successive projects to the present day. For much of this period, the counterpart in Morocco has beet the same (ONEP), so that a continuing harmonious relation- ship and an effec,tive dialogue have been built up, helped by some degree of continuity in the Bank personnel assigned to the sector. Review of the project files and discussions with Bank staff suggest that at no time were there any serious -disagreements concerning this project. Tile Bank repeat- edly identified itself with the interests of ONEP, as refXecting the best approach to ensure successful achievement of project objectives, ,and- did vbat it could to press ONEP's case with higher authority, in such matters as tariffs,. settlement of outstanding bills and (more successfully) 'the expedition of disbursements and payments; the proj eci files are ft41 of, repreqentations and entreaties on these matters. Bank staff also appear to have had a hanios persional relationaip with government officials, even though this was not always .fruitful it 0roducing he results the Bank sought. 6.02 Bank staff played a major role in helping the project benefi- ciaries to identify and prepare th investment program up to a praisal, and being concerned with the overall project rather than individual components, probably helped to foster inter-deartmental - cooperati, in or cco. Demand projections developed at appraisal were reasonably accurate, and ,the Bank obviously- devoted a great deal, of detailed attention to the formula'- tion Of the scheme- for assisted house connections, though the Audit re- ceived, complaints from R4gies, that they were not consulted at negotia- tions. Supervision of the project by Bank staff was also regular and sus- tained, understandably so snce it was coupled with preparation or ac6ini- stration of other sector projects; on average, missions visited Morocco every five months over a period of 68-1/2 (-)years. Altogether, the project pres"nts an example of good cooperation between the Bank - and executing agehd'ies, though both parties were hampered by extraneous con4t aints. VII. CONCIWSIONS N IESSONS 7.01 Apart from the continuing problems of tariffs and financial rela- tionships, this project can be charactefzed as successful. The works were executed competently, they achieved their objectiAres, and they have raised, or 'at least maintained, the conditions of servic4 for, the water consumers in the proje t area. Admittedly they took longer, atid\ cost more, than had been foreseen; but this is more the rule than the exception in major eagi- neering projects of this lnd, and much of the delay was due to extraneous - 14 - (: factors, which (lik the financial problems) neither of the project author- ities nor the Ba kvap ip a posi$on to al ate, and to Ainforeseen prob- lems arising from the house connections program. 7.02 The PCR offers a uvuer aggestions as to the lessons offered by experience under the project (PCR, paras. 8.03 and 9.04); some of these are fairly obvious a4 are (or should be' well known, while others, apply spe- cifically to Morocco, and the Audit sees no reason to dissent from them. One conclusion which may have wider significance is as follows.: tVhere a project comprises physical works by a specific agency In a specific local- ity and a specific subsector, there may perhaps be a tendency that the project preparation and appraisal will focus excessively on that agency, and may lose sight of the external constraints (not always clearly percep- tible to projects 'staff) to which it may be subject--for example, the na- tional economic situation, or the administrative practices of the central governmnt--which may prejudice the project agency's deoication and compe- tence to carry out the project in all Its aspects. such constraints, par-', ticularly insofar as they affect cost recovery, must be recognized and realistically allowed for in project formulation, and preferably addressed by the Bank through timely sectoral or adjustmet operations. 7.03 Or a related subject, where project attdies may affect a anber of institutions, both inside and outside the project sector or subsector, and possibly with conflicting interests and priorities--as f6r example, the water resource studies updar this project--it seems essential to establish an authoitative coordination panel at the outset, otherwise decisions may be delayed or eveb abortive. L, :41