Morocco - Housing Loan and Communal Infrastructure Fund Projects

Document of The World Bank FOR OFFICIAL USE ONLY Report No. PEROMANCE AUDIT REPORT MOROCCO YIsT BOUSlMG LOAN TO CREDIT IMDBIIIER ET HOTELIER - CIM (LOAN 2245-VOR) AND PILOT PROJECT FOR THE COMUAL INFRASTRUCTURE FORD - FEC (LOAN 2272-MDR) JUNE 30, 1993 MICROGRAPHICS Report No: 12168 Type: PPAR Operations Evaluation Department 'lls document has a restricted distrbHmion and may be used by adl only in the performance of their ofieil dulles. Its contents may not otherwise be disclosed WOid Bank uthordzation. Currency Unit: Dirham (DE) Anual Averages 1.2 IM I 1983 19l4 1985 1986 1987 1"S 1989 1990 1221 3.90 4.30 6.00 8.06 9.55 9.62 8.71 7.80 8.21 8.12 8.04 8.15 GLOSSARY OF ACRONYMS BCP Popular Central Bank (Banque Central. Populaire) BDCL Local Commune Development Bank (Banque de D6veloppement des Collectivit6s Locales) BEDS National Economic Development Bank (Banque Natioaale pour le D6veloppement Economique) CDG (GOM's) Savings Management Fund (Caisse de D6p6t et de Gestion) CGI (CDG's) General Real Estate Company (Compagnie G6nrale Imobilibre) CIR Real Estate and Hotel Finance Bank (Cr&dit Imobilier et Hotelier) DGCL General Directorate for Local Authorities (Direction Gn6rale des Collectivit6s Locales) DIC Development Finance Company ERAC Regional Planning and Construction Agency (Etablissement R&gional d'Amfnagement du Territoire) FEC Communal Infrastructure Fund (Fonds d'Equipement Communal) CO Government of Morocco HBM Low-cost housing (Habitation a Ion March) MAT Ministry of Housing and Urban Planning (latterly MO) (Ministbre de l'Kabitat et de l'Am6nagement du Territoire) HOF Ministry of Finance (Minist6re des Finances) NO Ministry of Housing (formerly MEAT) (Minist6re de l'Habitat) MOI Ministry of Interior (Ministre de 1'Int6rieure) OED Operations Evaluation Department ONE National Drinking Water Bureau (Office Nationale de l'Eau Potable) PAR Performance Audit Report PCR Project Completion Report PR President's Report VAT Value-added tax WDR World Development Report FISCAL YFAR OF BORROWER January 1 - December 31 FOR OFFICIAL USE ONLY THE WORLD BANK Washington, D.C. 20433 U.SA Mutc of Drectoi.Gene0a4 Operatione 8valastia June 30, 1993 *lA U TO =IlK EXEMMiE =IE ORS AND THE PRESIDENT SUBJECTs Performance Audit Report on Morocco First Housing Loan to Credit Immobilier et Hotelier - CIA (Loan 2245-OR), and Pilot Project for the Communal Infrastructure Fund - FEC (Loan 2272-Mp). Attached is the *Performance Audit Report on Morocco - First Housing Loan to Credit Iamobilier et Hotelier - CIE (Loan 2245-MOR), and Pilot Project for the Communal Infrastructure Fund - PEC (Loan 2272-MOR)O prepared by the Operations Evaluation Department. The two projects audited in this report provided lines of credit to CIE for a housing program and to FEC to Invest in eligible infrastructure sub- projects in local communes throughout the country; in contrast, several earlier urban projects were specifi physical Investments. Both projects were prepared and approved in 1983, at a time when Morocco's economic prospects were not promising. Thus the funding provided through these two loans was essential to both organizations at the time of project launch. The CIE project provided funding for some 251 of the total housing units financed by CIE during the 1983-90 period. The focus of the Bank funded part of the program was on the low cost housingi its Implementation faced few problems, except land titling uncertainties in a small number of cases. The F&C project, a pilot operation, provided funding for local infrastructure and equipment sub-projectes markets, water supply and sanitation, and urban transport were the most significant. Although the project achieved its physical targets its financial and institutional development was limited. The project, howeverg set in motion a process of further institutional development which allowed the Bank eventually to prepare a follow-on operation with F&C. Overall, the Audit rates the CIN project as satisfactory, its institutional development as partial, and the sustainability of project benefits as likely. The PCR-based rating Is the same for overall performance, substantial for institutional development, and uncertain for sustainability. The FEC projects outcome is rated as marginally satisfactory, its institutional development as partial, and the sustainability of Its benefits as uncertain. The PCR-based ratings weres outcome satisfactory, institutional developmeut substantial, and sustainability likely. This docm*ent has a restricted distribution am my i used by recipients only in the perfore nQ of their offioial daies. Its contents may not othervise be disclosed without ortd eank euthoriztion. l一”哪攔鑼洶編騙啊‘單二鉤騙編，,，碗Jo- L一州”姻觸向謝肉陶”加”一一’”加瞧中魄― 弱，,,,.,&,,,,&,&&,,,&,&&&&,&&,,&,&,&&&&,,,,&&&&”魚磚”啊臘一劉江訪’! ．寥。，&”。。。，,&,,&，。。。”,，。，。。。，&,,,,&,,,,,,&,,&,&,,&”言”叮。。•p繙d&O 9寥”&&&&&”二”二”.．…，二，二”……’”…’&&&&”…•’&.&’爛鑪”言 言露…”&&.”…”二”二●二•••．…’．•．,’二•…’二”.’•…”…””細1。●t 蓄寥，,&,&,,&,,,,,,&&,,,,,,,,,&&&&&,,&&,,&&&&’•館關匯鰓開閑訐下訓勵”磚症”徑 寥寥’.&&.”…”.&&”二•二”...．……’...．……’.”……。…’.0蠶症’蓄 1露．&&．…”&”…”.............................．……””二•．日江0.! ,：。”,，。。。，。，”。”，。。。，,,,,.,,&,,&,,&,&&&,&&,,&,,,”叮妒開d卜讓名，讓，”▼ ！寥．.....．……0 .．……，．…”’二啊團唱蠶遺口神個O劉開勸•閱n必觀名O蹴馴開’111 0言．.......．……，.........................．……讒。•卜開O颼’寥 露t,,&,&&&,,.&,,,&,,&&,,&&&,,,&&,,&,&,,,&,,,,,&,,,，鳥”仰開屆江O&, 。T。。．…。．0．。…。·，&&&,&,,,,，••。，讓肉陣。劇哺”，•••，，觔，”鰍開”寫 LT二”.&.&.&&&&.”二’.&&．…’.．…”二”.”…”&.&.&&.&.’•”&&.0寫症’寥 當，&”二”&．…”．…’.&．••二”.．…’二”二’…”&&”二”&&．……寫10 .1 戲。，&,”。。”,&&&&&,,&,,&&&,,,,,,&&,&,&&&&&&,&,,&”訪們牌名以坤”•鉀開”C 雪，&&&”。．&.&&’二”二””•二。’……’.”二”…’二’.．…•．’二”.0蠶症．忿 觔1’…’....．……’二’．…’…’…””…”二”•二”.&．•”’…’日10&! ”。。。。。。。”。”&,,&,,,&&&&&&,,,,&&,,&，觔，中”•“閱”•『“．邵．-.1”。 tt&,&&&&&&,&&,,&,&,&,,,,,,&&&,&&,,&&&&&&&&&&&&&&&&&O寫盧＇OCO”露 0!&&&&.&.’…’.．…’•．&．…’…’二”……”&”…’&.&&．…寫10.! 0’。·。二。·…”·，…，一”,&&,,&&&,,,,,,,,,&&”鳥悶，讓飽萬物鰍繙纏叫甚’囉 ‘。·。。。。”,,&,&,,,,,,,,,,&,&,&,&&,&&&’彎”贓以柑許P”。于。戲叮，叫甚．▼ 6&&&.&&&&&.&&&&.&&.&&&.&”…”.&&&&.&&&&&.&&’劇必劉戊潤濾蠶唱’11 9,,,&,,,,&&&&&&&&&&&&&&&&&”二”&’…’&&&.&&”馴馴閱”權。越權叩諱煙”寫 ‘··…。．。，,,&,,,,,&”二，，·”·，。•馴鱸每s叩神•，訪，助觀閑”&“•“矯”。 ，。．,，。”&,,&&,,,&,，·，&,,&&&&&&&&&&，鰓。娜卻國開團盧“物計”償’常 ，·················……”壯載磚。，巉娜胛間”。州狗以細•”寥 常二。…。·……。。，,，·，，·，,”··…，，·”·，,&&，陣”。”妒。”“矓’言 C,,&&&,&,,&&&&&&&,&&&,&&&&&&&.&&&’•”.&&&&”。。鳥。•S簡，華雊•寫甚’O 露。。”,&,,,&&,,,,,,,,&,,,&&,&&&”。。”名齋開娜纖輩論。。，•詞•”。；。開開”• 1,,,,&,.&&&,&&&&&&&.&&&&’二”&&.”二’&.&&”••’&&&”•。言幼啊坤：認。X’▼ t&&&&.”。．&&.&.&.&.&&．•二•．&．…”.”二’……”……’&”口顯必衹離訓羶．1 言▲．”…’.&．…’.&．…”……’……”．…”’…•．…””勾鉀論娥寫規O劉馴險奮n口 下下，&&&&&&&&&&&&&&&&&”二’”二”&,.&&.&.&&&&&&&&&”劇濾批編V區曆q釭9甲舊 言．&’…”.,&.&”二’.&.．…’二”二”．……•．”二”二”&.&.’二’…”’二’審劉臚劉絨 觀取囫取「衍〕怔他 一一一一一取勿哥可n才面吋”一― O寫症”口口盧咸嗚開國臨頗個跚U個叨國團咸沁馴認毬鱸名O劉伐期撾州n廈d 煙神 個飾卜鈐館圍閱吃） 日X。•司暱低凶瀘胡馴扛叮觀開颼甚划國幼偽卻網n劍玆劇閱盧名馴n瀘 O劍沁戲閑 勵臨調r勾口叮蔥勿爾師調股 州以O蠶矓甲n調劇勵馴鱸 Zále of C~ Olft (Cont.) Oalte zo. 2. CIE - Tochnical Assistance ........... 33 30 IMC 34 D. Sust~ bility ......... 35 10 CIE 35 2. CDGIFZC ............ 36 IV. MEER POIMS Op SPECIAL IMMST ..................... 37 A. Poverty Impact 37 Ie CIE 00444 404 37 2* IMC 38 3. For Borrovere 9 Loon Am~te vers Inel~ lesat 0 0 0 0 0 0 0 0 4 0 0 0 0 0 39 C. luctitutional Appraleal In U~ of Credit Operatiom .......... 40 D. Role of Subproject Appraisal In FZC Project Dooielon M~ ... 41 9. Inadequate ~ toring In terms of Proj eet Obj ectives 42 10 Zaviror^tal Impact ............... 43 V. FOMMCE op BORROM AM BM 44 A* Experionces of Borrover &ad Zxecuti" Agencies 44 10 CIE 44 2. CDGIIMC ........... 44 ze 1010 Of th* Zaak 44 1. CTE Project ........ 45 2. FEC Project ............. 45 VT. CONCLUSIONS AllD LESSONS 1r2 46 A. CIE Project o ...... 47 3. TBC Project 47 AMM 1 Com~te from tba Borrover 49 MOROCCO FIRST HOUSING LOAN TO CREDIT IMMOBILIZR ET HOTELIER - CIS (LOAN 2245-HOR) AND PILOT PROJECT FOR THE COMMUNAL INFRASTRUCTURE FUND - FEC (LOAN 2272-M0=) 1. This document is a Performance Audit report (PAR) on two lines of credit operations in the urban sector in Morocco. The First Housing Loan to Cr6dit Iamobilier et Hotelier was approved by the Board of Directors on March 15, 1983, and partly financed by Loan 2245-MOR in the amount of US$60 million, which was fully disbursed. The Pilot Project for the Commnal Infrastructure Fund - FEC Vnf, approved by the Board on April 26, 1983, and partly financed by Loan 2272-MOR In the amount of US$16 million, which was also fully disbursed. 2* The PAR consists of an Evaluation Sumary and a Performance Audit Report prepared by the Operations Evaluation Department (OED). Project Completion Reports (PCRs) pre.pared by the Infrastructure Operations Division of former Country Department II tf the former Burope, Middle East and North Africa Region (EMA) were previously submitted to the Board of Executive Directors as Reports No. 10154 dated November 27, 1991 (CIR), and No. 10075 dated November 14, 1991 (FEC). 3. The PCRs provide general accounts of the project experiences and indicate their successes and failures. In order to give a wider and more recent perspective, the Audit provides additional Information about the projects, their history, eventual outcome, the issues raised, and the principal lessons learned. The PAR is based on the PCRs, the Staff Appraisal Reports (SARs), the President'. Reports, project legal documents, project files, legal files, a transcript of the Executive Directors' discusions at Board presentation and other relevant material. It also draws on Interviews with Bank operational staff and with Moroccan Goverrment Officials and management and staff of CIN and FEC in Casablanca and Rabat, as well as field visits to project sites, during the audit mission to Morocco in June 1992. 4. The Audit agrees with the CIE Project's PCR In some respects, but takes a different view of the project's achievements and in particular of the impact on the sector and low-income population. In the case of the FEC Project, the Audit"s analysis of a broader data base has lead to somewhat lower ratings than presented in the respective PCR. S. Following standard ORD procedures, copies of the draft PAR were sent to the Borrower and Executing Agencies for comments on March 26, 1993. Comments and clarifications received from the Ministry of Finance, CIE and CDG, are reflected in, and attached to the report as Annex 1. I zuzMC AMD 11R8T 101US1EG LOAN TO CREDIT nlmDBILIER ET RO'ELIER (LOAN 22454M4R) nASIC DATA SEERT EET PROJECT DATA Actual or Aotuala I AppraLeal Curent of Appasa Total Projeat Coat kUs$ M) 209.5 105.3 50.0 Loen Amount (US$ 9) 60.0 60.0 100.0 Economto Rate of Retu~ 12 15 125 tnstitutional Perfoman*e: Subøtantial Appraisal getløate (Us$ M) 60.0 Actual (Us$ i) 60.0 A*tual sa t of Appraial () 100.0 Date of ial Dbur~ements January 29, 1990 Identification 09181 Prepration 01/82 05/c2 Appratsal 06/82 10/82 Negotiation 12/82 01/83 board Approval 03/15/83 Loean ignture 04/11/83 Loa Ef fectivene8 07/83 08/30/83 Projeot Copff ton 09130190 05/28/91 Loan Clo#1% 12/89 12/31/89 pre~a a 35.9 .1 -- - * --3. s.otat 4.', 0.- lotto" - 5.9 - - - - - - . . . 5.9 lson • 4.3 12.0 8.5 22.1 8.9 7.4 4.3 7.3 4.4 0.4 79.6 6.4 11.5 • - . • • • - - • 17.9 £1 As pr æ Imtituti Data as/RportIng system EMISINATA Date No. of No. of Mnth/Year .Jms Pergona Identlfication 09/81 Preparation 05/82 24 2 Appraisal 10/83 18 5 Negottation 04/83 11 2 Supervision I 02/84 10 2 Supevision II 08/84 10 2 Supervision III 03/85 8 2 Supevision IV 05/85 8 2 Supervision V 10/85 16 1 Supervision VI 12/85 5 1 Supervision VII 02/86 6 1 Supervision VIII 03/86 15 1 Supervision IX 10/86 6 2 Supervision X 03/87 2 2 Supervision XI 10/87 2 1 Supervision XII 04/88 5 2 Supervision XIII 05/89 2 1 Supervision XIV 12/89 4 1 Supervision XV 05189 5 3 Supervision XVI 12/89 7 1 Supervision XVII 02/90 _1 1 Total 253 OTHER PROJECT DATA Borrøvers Credit Umobiller et Hotelier Executing Agenciøs3 Crodtt Iamobiller et Hoteller Iv PILOT PROJECT -OR THE COMMURAL INFRASTR=CTUR FUND - VZ0 (LOAN 2272-MOR) BASIC DATA SOMM RZY PECT -DATA Actual or Actual as I Appraisal Current of Appraisal n Epectation Estimate Etimata Total Project Cost (US$ a) 31.0 11 N/A Loan Amount (US$ a) 16.0 16.0 100.0 Economic Rate of Return n.a. n.a. U.a. Institutional Performance: Substantial 1/ An estimated tine-slice of an ongoing investment program. CUULATIVE ESTIMATED AND ACTUALDISE Appraisal Estimate (US$ a) 16.0 Actual (US$ a) 16.0 Actual as I of Appraisal (2) 100.0 Date of Final Disbursements Januav- 29, 1990 Oriinal Atual Identification 05/80 Preparation 02181 Appraisal 06/82 Negotiation 03/83 Board Approval 04/26/83 Loan Signature 05/23/83 Loan Effectiveness 08/22/83 04102/84 Loan Closing 06/30/89 12/31/89 Preappras t 20.7 30.9 12.4 .2 • • • • • • • 4.2 Apprela • •3.0 16.8 • - • 19.8 Nesotietain • • • .1 • • • • • • • • .1 supervistln • • .1 3.6 6.1 8.7 1.8 2.7 4.8 1.9 16.6 3.6 49.9 Other 3.7 .7 .7 9.4 - • • • • 16.5 j1 As per 0E0 Instituttonmt Date Bas/Reporting System MISSIO DATA Date No. of No. of ~ot 9a 1Dav Peom Preparation 01181 14 4 fteparation 05181 7 1 Preparation 06/81 3 5 Preppra:bal 03182 14 3 Appraisal 07/82 24 3 Appraial 02183 4 2 Supervision 1 11183 3 1 Supervis:Lon II 03184 12 1 Supervision III 10/84 8 1 SupervloLon IV 03/85 6 2 Superviaon V 10/85 5 3 Supervision VI 03186 12 4 Supervision VII 10/87 12 3 SupervLeion VIII 07/88 .2 1 Total 131 OTHER PROJECT DATA Borrowers Goverrment of morocco Executing Agencies: Caise dc Depot et de Gestion (CD) Comwunal Inrastucture und (lEC) v1 MOROCCOs E N CE WITB LINBS OF CREDIT FOR ROUSING AND URBAN DEVELOPMENT FIRST HOUSING LOAN TO CREDIT IWMBILIER ST NOTELIER - CIE (LOAN 2245-MOR) AND PILOT PROJECT FOR THE COmmUNAL INFRASTRUCTURB FUD - FEC (LOAN 222-quR) Introductl n I . The two projects audited in Improve. There were two consequences this report provided lines of credit for the project.: tiLs external to CIR ("Crddit Immoblier et funding they provided become less Hotelier# - Real Estate and Rotel urgent to the borrower., end local Finance Bank) and FEC ("Fonds resource, for CIE and FEC expanded, d'Equipement Commnal - Comiunal reducing the relative Importance of Infrastructure Fund) to Invest In the Bank. hare of the total (par. eligible subprojects throughout the 1.03-1.06). country in housing and local commme infrastructure, respectively. This 3. Morocco'e 4.32 average annual was in contrast to the early urban urban population growth made demand projects in Morocco, which financed for shelter and urban services -'ecific physical investments In Increase rapidly. Currently, 12 selected cities. The project concept million people live in urban area, vae modelled on the development or 48X of the total population. finance company (DFC) approach that UNCHS (N0RTAT) projections indicate the Bank had previously used to this proportion will grow to 711 by provide credit for Ind-trial the year 2010. Challenges of development, including some urbaniaation In Morocco include rapid operations in Morocco itself (paras. eansion of "bidonvilles" or 1.01-1.02)o squatter settlements of precarious shacks built Illegally on unserviced 2. Both operations were prepared land and growth of "informal" and approved while Morocco's economic housing, similarly Illegal, but of a prospects were not promistg: In higher standard than bidanvilles. 1983, the country faced an Sam 20Z of Morocco's urban international payments crisis and the population is estimated to live In need for external finance was scute. these two types of housing. TO Gross domestic product continued to resettle the residents of these fall for the first half of project illegal settlements and meet the implementation until 1985. After growing demands of new urban 1986, however, Morocco's households would require the annual macroeconomic performthae began to provision of at least 151,000 new vii housng units. Morocco's formal agreed between CON and the Bank after housing sector has, at best, been 19810 The Dank's reluctant able to supply 57,000 units In one recognition of 00*' opposition to. year. The Government of Morocco bidonvilles led sector trategy to (GOM) made considerable efforts to focus on sites and services with core increase housing supply, but the housing. The Banka 1982 urban public sector has, at most, been able sector review of Morocco Identified a to deliver only 8,300 units per major role for CIN as a unique annum. Thus, the CIR Project'a focs vehicle for promoting a coherent on the private sector, responsible housing policy. The Bank's strategy for 852 of the formal housing supply, was to take the most successful was appropriate (paras. 1.07-1.11). element of the first two urban projects (i.e.,9 sites and services) 4. Rapid urban growth also and stimulate their efficient increased the pressure for delivery through CIS (paras. 1.16- administrative decentralization in 1.21). Morocco. The Local Commune Organization Law of 1976 augmented Mbectives and V2821121102 local communes' responsibilities and enhanced their access to financial 6. Under this line of credit resources, but central and provincial approachg CIE made loans to public government control over local and private developers for house decision-making remains substantial. construction In the first Instance, Morocco has two tiers of subnational and later as mortgage loans for the government. The first consists of buyers of the houses built. In the provincial governors and prefet. case of the C Project, modelled with considerable responsibilities upon the pioneering Cities and for regional coordination. The Villages Development Bank Project second Is made up of some 859 local (Loan 1826) in Jordan, the final communes, 99 of which are for larger borrowers were local communes and urban areas. The FnC Project sectoral utilities (known as lrftles focussed on the local communes, in Morocco). (paros. 2*01-2.04) providing them with loans to finance infrastructure and equipment. The 7. Project Institutions were key Ministry of the Interior represents players in their respective GCH'. authority for local affairse subsoctors and the financial markets and is responsible for overseeing of Morocco. CIN is the countryos decision-making by the communes principal housing bank with asets of (paras. 1.12-1.15). US$1.7 billion in 1990. Majority ownership of CIEIis held by CDC 5. 00M and the Bank have had (Caisse de Ddpat et de Gootion")s differences about urban sector policy 00M9o very large savings management and strategy in Morocco. Under the fund (para 2.06). Bein wel known first two urban projects, sector to the Bank through four earlier dialogue largely revolved around the touriem projects, CIE was recommended issue of bidonvilles, whether to as the borrower for a housing finance eradicate them (as 0 wished) or to project* Despite its judicial status upgrade them (as the Bank proposed). as an Independent fund with its own Although upgrading was Included in financ statements, 79C was a the first two urban operations on a department of CDGv subject to the trial basis, no further projects including bidonville upgrading ae vill policy and administrative norms of development policy, to assist local that Institution (parse. 2.05-2.10). commues In their efforts to meet their basic infrastructure needs; and 8. At appraisal, FEC was a very (U) to asist FEC In formulating small operation with a staff of only operating principles for project nine. Its job was to lend to local financing, on the basis of rigorous commes and utilities ("r6tes6) for preparation, appraisal and infrastructure projects and supervision standards end criteria, acquisition of equipment. The Bank* and Improving the efficiency of its appraisal report cited a long list of staff in the performance of their weaknesses of F8Cs no track record or functions." (NEC SAR, par&. 4.01) operational procedures and policy; These objectives were Incorporated inadequate staffing; and tabalanced into the legal documents through an financial structure. Most of these agreement to implement the project in problems existed because NEC was not accordance with the Statement of a genuine financial institution, but Policy. (para. 2.37-2.38) simply a department of another agency. These uncertainties led the Inlementation Experience Bank's Legal Department early on to recommend against NEC as a borrower 11. There were practically no for the proposed project (pares. disbursements of the loan to CIA for 2.11-2.17). the first year after loan effectiveness. CINI's explanation is. 9. Objectives of the CIN Project that the Bank underestimated the were tos (i) encourage private and housing costs necessary to meet public developers to produce low-cost official nors, contradicting the ShR housing schemes affordable to income which stated that the ceiling prices groups in the lower half of the urban on which they were based were Income distribution (i.e., with identified by CIOs own market monthly household incomes below studies. After CIE prevailed on the US$330 in aid-1982 prices); and (ii) Bank, the ceiling price, were raised develop CIR's ability to apaise successively by 87.5Z (against an low-cost housing schemes and advise accumulated inflation of 38.42 in DR, private and public developers on all para. 3.03) to a mazimm of DR aspects of low-cost housing design. 150,000 (US$18,720 in 1988) and The funding mechanism involved disbursements began. The PCR converting all (three year) mentions that a special upper limit construction loans into long-term (up of DR 100000 per unit was really to 25 year) mortgage loans. maintained. This limit was not To meet this obligation, it was observed, however, since 13 of the 43 recognised that CIN might have to developer contracts financed by CIE seek additional funds at a later under the projact were for housing date, since the combined maturity of units in the DR 108,955 - 163,333 the loans under this arrangement ois range 0 These inreases ma the up to 28 years, compared with the 17 housing offered under the project year term of the Bank Loan. (pares. unaffordable to Its Intended low- 2.32-2.36) income beneficiaries NIC PrOJect disbursements were also slow at the 10. The goals of the 7EC Project outset, but this later proved typical were: (1) in support of the of municipal development projects covernment's regional and ommunal where noew policies and procedures ix were introduced. As more proposals with the legal agreements. 79C grew were submitted and PEC staff became even faster than CIR with a smaller more adept at appraising them, Bank financial contribution, disbursements began to accelerate corresponding to just 6.12 of its (pares. 3.01-3.04). growth in long-term loans. Over the 1983-90 Period# PEC did not become an 12. CI*s financial performance autonomous financial Institution. weakened between 1983 and 1990 while 001 briefly contemplated creating the project was being Implemented, such an entity, but shelved the plan. This was even true during the second Roweverg Important groundwork was half of this period (1987-90) when laid for this to happen In the future the national economy was picking up. under a follow-on operation that was For nine of the eleven financial recently approved by the Dank (paras. indicators reported in the PCR, 3.16-3.24). actual performance was below the forecast. CIR did, however, maintain a positive interest rate spread and a sound capital structure. CIR grew 14. The project accounted for significantly during 1983-90, but the some 252 of all the housing units Bank contributed only 7.3Z to the financed by CIE during the 1983-90 increase of its long-term loans over period, a significant share of CI's this period. Two decades and six total sector operations. The funding Bank loans may have induced "borrower of each of the 15,449 units completed fatigue" in CIR, whose management under the project involved two came to see the amount of the housing successive financial operations from finance loan as insignificant. CIR's CIE"s point of view. The first was hotel loans worsened during to provide three year construction Implementation of the project (pars. loans to developers to build the 3.05-3.15). house, and the second was to provide long-term mortgage loans to 13. FEC's financial performance purchasers to buy the unit built. during 1983-90 was mixed. This is normal housing finance Profitability declined and losses practice in many countries. In the were reported in 1988 and 1989 after PCR, however, CIN only counts the FEC lost access to the Central Bank's projecs contribution to the second low interest rediscount facility. On Instance, namely the value mortgage the other hand, the maturities of financing. This Ignores construction FEC's liabilities were considerably loans financed under the project that lengthened through the provision of were provided for all units built. substantial long-term loans in 1987. Dy comparing the projects mortgage Despite this improvement, FEC finne only with Cass total remained undercapitalized for most of construction-plus-mortgage financ, the period. CO's two contributions the PCR estimated the projecs of DR 20 million to PEC's equity, contribution to CIE's overall housing when four were expected, did not keep finance operations to be only half FEC from becoming a highly leveraged what it approximately was (pare. operation. On the other hand, FEC 3.25). became more solvent, with a current ratio far surpassing that forecast at 15. Actual physical implementation appraisal. Provisions for losses, of the housing financed presented few which did not exist at all prior to problems, except for land titling the project,1 were made in accordance uncertainties hich made some schemes x infeasible, as had occurred in the housing. The PCR Implied thats first two urban projects. without Ban funds, the low-ost Surprisingly in view of the trend housing program risked becoming towards increasing coiling prices, extinct for lack of CON support. The PCR data reveals a fall in the value increases in the project'. ceiling of subloans per unit during prices, with the Bank'. agreement, implementation compared with that followed the trend of CIS operation. estimated at appraisal. The a. a whole. This accelerated appraisal estimate of the total cost disbursements, but the Audit believes of the houeing component (i.e. the this occurred at the cost of total value of sub-loans to be weakening the project's low-income financed under the project) was DR focus (per". 3.29-3.33). 1,257 million (US$209.5 million). The actual value of sub-loans by 17. CIIIs appraisal and control completion had fallen to DR 891.5 capabilities Impressed the Audit, but million (US$105.3 million), it is difficult to deteroine what increasing the Bank'. share in theae capabilitiea od to the project financing from 29.1% to projectl At appraisal, the Bank had 54.7Z. With 15,449 units built and praiaed the high quality of CIE's funded, the average sub-loan per evaluations. The low-coat housing unit, according to the PCR, fell from apect of these skills aeemed to have an appraisal estimate of D 83,000 been lost, however, and the Audit (US$13,333) to DR 57,706 (US$6,816). miasion was left with the impression The achievement of these dramatic that the CIE/private developer savings Ia not analyzed in the PCR. dialogu on low-coat housing waa not Nor does the PCR explain why, if unit a very active one (para. 3.34-3.36). sub-loan. fell so sharply, was it necessary to successively Increase 18. On the order of 100 local the celling prices of the units? Infrastructure and equipment (pars 3.26-3.28). subprojects were financed wder the FEC Projeact. By subborrower and 16. For CIE operations as a sector, the portfolio included a whole, housing costs in fact rose diverse range of inveatmenta. Rural sharply between 1983 and 1990: 125.4% commuMes, autonomous centers, in the case of public developers, and 'municipaltds', and Or6giesO 423.4Z for private developer., seven (sectoral utilities) took similar times faster than inflation. While shares of the amount lent. the Bank financed units are Commercial infrastructure (mostly affordable by the lower median-income markets) end urban transport were the population, the average C1-financed most popular subprojects. nC kept private sector unit is affordable good control of local investments only by those in the top docile. CII during their construction, but failed mortgage lending also followed a to follow up with adequate trend toward more costly housing. monitoring. Withut adequate staff During 1983-90, the number of resources and equipment for "special regimem (i.e. low-cost) supervision and monitoring, nC is loans awarded per annum halved, while unable to confirm how well the the number of higher cost 'general investments are performing and regime' loans doubled. CIS employed whether they are being properly Bank funds to finance a much reduced maintaine. A cross-sectional low-cost housing program, while using analysia by type of local authority, other resources for higher cost st city etse, region, and poverty Ipact operations In Morocco'e tight of FBC's rich lending experience to financial market. The suetaiusbility date would help policy formulation of the benefits derived from the and clarify sector priorities In housing units financed under the Morocco. project Is likely to be assured by resident-ovners9 care and 19. The Audit mission was Iapressed maintenance In one large scheme by subproject appraisal skills visited in "b.GnCAS however, poor developed in-house through on-the-job maintenance "ublic areas could training by FEC staff. Since PEC threaten su - -sinability of the management felt that the projact's neighborhood as a viable urban rules and procedures were well community (par". 3.42-3.44). understood by all parties, a formal manual was considered unnecessary. 22. As an Integral part of CDC9 This might not have mattered when FC sustainability of the FEC operation operations were small, but an depends on the sustainability of CDC enlarged program requires a clear itself. FC did not become an statement of policy, rules, and Independent financial institution. operational procedures. CDC can, however, come to FEC's aid if the need should arise, as it did 20, Two other key findings should in 1987. The Audit found no reason also be highlighted. First, CIE and to doubt that CDG would be able to CDG/FEC considered the loan amounts fulfil this role, if necessary. of their respective projects Sustainability of the local insignificant in relation to Investments financed under the subsectoral needs, Bank projectwill'depend on the ability of conditionalities, and Bank reporting local communes ad iregies to and information requirements. As a maintain them. Actual performance is result, CIE and FC's growth-minded unknown, because FEC has yet to carry managers were disappointed at the out the necessary monitoring of these Bank's limited financial contribution investments (pars. 3.45-3.46). to their expansion. Second, since the proceeds of a Bank loan to DICs ConclusIoS and Lessons such as CIN and FnC/CDG are fungible within the institution, this kind of 23. The CIR project Is rated operation effectively supports the satisfactory although relative to the work program of the financial optimistic expectations, some results intermediary as a whole, not just the were disappointing. Disbursement targeted components. proved difficult, and concrete impacts on c1H and Mrocco's housing D~.enahA.1Szsector are not easy to see. most inportanta the operation failed to 21. The C18*'s project benefits stimulate the delivery of lw-cost are likely to be sustainable. housing by public and private Although declining profitability and developers, as intended. The short-term solvency are cause for Project's sustainability Is vated a concern, its capital structure likely and Its institutional remains solid. Resource mobilization development as partial (vs. uncertain at levels sufficient to finance and substantial respectively n the increasingly costly lending products basis of the PCR). The main lessons is likely to be CII9a biggest from this exprience include the challenge in sustainig Its following (pars 6.02): Mil (I) Previous Bank-finaned CIN Was relatively smail and (physical) urban development cams to be regarded as projects in Norocco Insignificant by a growth- demonstrated that bidonvilles minded borrower. and low-cost housing were not central to COs housing 24* As a pilot operations the FEC policy, as they were to Dank Project raised modest expectations sector strategy. The lessons with respect to its likely results. of such earlier experiences Its achievements were mixed but should be taken into account, subsequent progress was evidently even if design of the new adequate to Justify the recent project is intended a. a approval of a follow-on operation. substitute for the older The Audit rates the project's overall approaches$ outcome a satisfactory and its Institutional development as partial. (ii) As a corollary, the Bank Becaue of the FEC9s continuing should focus its support on weaknesses In sub-project monitoring, those operations in the DIC the oustainsbility of the project's which are a priority for the benefits is rated as uncertain. The Government and the DPC Itself. PCR-based ratings we overall If not, Bank funding risks performance as satisfactory; merely filling a resource gap Institutional development as left by the local withdrawal substantial; and sustainability a from low priority operations, likely. The principal lessons as appears to have been the Include the following (par&. 6.03): case with CIA's low-cost housing program; (i) To achieve the desired Institutional and policy (1th) Resources provided via a Impact, It is Important for the line of credit to a DYC are Bank to engage the appropriate fugible within the authority in its policy Institution, and in practice dialogue. In this case, CDC support the intermediary as a was a key policy Interlocutor whole and all of its and Its active engagement in operations. Institutional the process should have been appraisal, therefore, should enlisted; cover all aspects of the DIC, and project design may require (U) For a first operation In legally covenanted performance a particular sector, it is criteria for areas of its Important to spell out In operation other than those detail the agreed criteria for directly targeted under the eligible beneficiaries and project (e.g., hotel loans In subprojects. In this case, the case of CIB); aand these details enabled PEC staff to develop a sound appraisal (iv) Bank influence in a DC methodology for evaluating the must be realistically assessed proposals received. In a in terms of the Importance of c=Wlex subsector such as local the Bank loan amount in developmant, it is likewise relation to all new funding important to engage other key likely to be mobilised by the policy ants, such as HOI and intermediary. The Bank loan to the provincial governors, so as to eUSUTe their support for the principles and procedures of the operation. Without euch support, there is a risk that other Investment evaluation criteria may be introduced which conflict with the alas of the Bank-financed projecti and (itt) As in the CIN case, Bank influence in a DC aet be realistically assessed in terms of the importance of the Bank loan in relation to all new funding ltkely to be mobilized by the Intenaediary In question. After the expected follow-on loan did not materialize, Bank influence diminished and CDG/jBC regarded the loan as insignificant in relation to the sector*s needs. MOROCCO FIRST HOUSING LOAN TO CREDIT INNOBILIER IT HOTELIER - CIR (LOAN 2245-HOR) AND PILOT PROJECT FOR THE COIA=NAL INFRASTRUCTURE FAUD - FEC (LOAN 2272-MORl A. Introduction 1.01 Bank involvement in the urban sector of Morocco has been a rich and varied one, spanning nearly two decades. It has Included five lending operations and five sector studies. The two operations assessed here are referred to in this report as the OCII Project" and the "FEC Project" respectively. Together, they reflect an important shift In Bank urban sector strategy, not only for Morocco, but for urban lending as a whole. Thus, in 1983, and for the first time in the region (and probably worldwide also), two Bank loans provided lines of credit to financial intermediaries in subsectors related to urban development. The proceeds were to be on-lent to real estate developers and local commues to finance urban housing and Infrastructure respectively. Two earlier urban projects in Morocco' had used Bank loans to finance specific urban Improvement programs in the cities of Rabat, Renitra, and Wekne. The line of credit approach, by contrast, aimed at using lank loan proceeds to finance Improvements in any city in the country, provided that the proposed subprojects met certain eligibility criteria. 1.02 Technical feasibility and economic and financial performance criteria were agreed to detersine which parts (i.e. subprojects) of CIR's ("Cr6dit Iamobilier at Hotelier" Real Estate and otel Finance Bank) and FEC's ("Fonds d'Equipement Communal" - Communal Infrastructure Fund) overall investment programs would be eligible for Bank financing. This project design was modelled on the development finance company (DIC) approach supported by the Bank to provide credit for industrial development (pars. 2.01-2.04). Its application 1 Following the convention established in the Bank's recent sector paper on decentralisation in Morocco, the French words "communes" and "communal" will be used In place of municipality and anIcipal in this report. This is to avoid confusion, since the word "aunicipalit6" is used in Morocco to refer to a particular kind of "comimne" i larger urban areas. Thus, as used in this report, the term "local communal development" is synonymous with the expression municipal development" widely used in other Bank documents. a Rabat Urban Development Project (Loan 1528-MOR, approved in June 1977 and Second Urban Development Project (Loan 1944-MOR, approved In January 1981). 2 to the urban sector, however, was a novelty. The advantages of this approach, according to the method9s protagonists, included more widespread benefit., more rapid disbursement and permanent institutional gains. The completion of two such operations in Morocco for housing and local development provides a good opportunity to assess whether the results thus far have met expectations in these sectors. B. Macroeconomic Context in Nmcco 1.03 Preparation and appraisal of both projects took place at a time of uncertainty about the prospects of the Moroccan economy. Successive balance of payments deficits and vanishing external funding cuainated in Morocco's international payments' crisis in 1983, coincidentally the year in which both loans were approved.s Thus, the CIN and FEC loans provided welcome relief to the Moroccan Government (CON) in helping to deal with its external payments problems, albeit on a very small scale in macroeconomic terms. 1.04 Notwithstanding macroeconomic stabilization and adjustment programs supported by the IMW (International Monetary aund) and the Bank, pessimism about the future of the national economy persisted at least until 1986. By that year, Morocco's external debt exceeded US$20 billion for the first times, and debt service consumed roughly 40 percent of the country's export earnings, one of the highest ratios in the world at the time. GDP fell continuously and sharply in US dollar terms between 1981 and 1985. In 1981, GDP per capita was US$860. By 1985, it had fallen to US$560, the lowest level since 1978. Thus, identification, preparation, appraisal, and start-up of the CIE and FEC Projects took place at a tine of gloom about Morocco's macroeconomic prospects, and when 00M urgently needed to mobilise resources to balance its external account. 1.05 By contrast, the second half of project implementation (1986-1990) coincided with a significant Improvement in macroeconomic performance. GDP growth resumed, with GDP per capita reaching US$950 in 1990. Unlike the large trade deficits of earlier year., Morocco earned small current account surpluses in 1987 and 1988. This turnaround reduced the need for external financial assistance, and meant that the country's external debt grew anc more slowly than in the early 1980s. Inflation also showed an Inprovementa during 1987-89, the annual rate fell to the 2-3 percent range, in contrast to the 6-12 percent of the 1981-85 period. 1.06 Thus, during the second half of project Implementation, optimism about economic recovery prevailed as Morocco's macro performance Improved. This had two consequences for the projects under review. First, the need for the kind of external funding they provided seemed less urgent from a balance of payments A study of the Moroccan economy during the 1970-1987 period noted thats "In the analysis of stabilization and adjustment in Morocco from 1975 to 1987, 1983 is a natural dividing point. That year the external payments crisis began, and the government was compelled to request emergency financial assistance from external creditors-* Norton, Brenden, Morocoe Analyeis and Reform of acOl Pli=q Econom Development Institute, Washington D.C., 1990. 3 standpoint than it had at the time of appraisal in 1983. econd, more buoyant 0C finances enabled local funding for VIN and FEC to grow, reducing the relative importance of Bank lending in their respective balance sheets. C. The Urban Sector 1. Ravid Urban Growth 1.07 With 4.3 percent average annual growth of the urban population since 1965, fuelled by migration from rural areas, increasingly heavy demands have been being made on infrastructure, services, and shelter in Moroccan citiis. Currently, some 12 million people, or 48 percent of the total population, live in urban areas. According to UNCES (HABITAT) projections, the country's urban population could more than double over the next twenty years, to account for some 71 percent of the total by the year 2010. Fortunately, Morocco's diversified urban system has important settlements located throughout the country, so that many cities can help absorb the pressures generated by expansion of the urban population. The largest urban agglomerations with their estimated 1990 populations ares Casablanca (3.2 million), Rabat-Sal6 (1.2 million), Fe (0.8 million), Marrakech (0.8 million), and Nekn&s (0.7 million).' 1.08 This rapid urban growth has contributed to the proliferation of what are referred to in Morocco as "bidonvillea, or unauthorized low-income squatter settlements of precariously built shacks without basic services. A Ministry of Housing (NO) survey identified 1,064 bidonvilles in Moroccan cities in 1989, housing 890,920 people, or 7.9 percent of the urban population.' Similarly illegal occupation has occurred through the expansion of "habitat clandestine" or informal housing, which now shelters scme 12 percent of the urban population.' Informal housing is also illegal, but is home to higher income residents who build better quality units than those found In bidonvilles. In total, 20 percent of Morocco's urban population lives in housing unrecognized by officialdom. 1.09 0 and local authorities (especially in Casablanca) made clear that these kinds of settlements were unacceptable. Where funds were available, bidonvilles were eradicated and their inhabitants resettled in public housing schemes. The Bank, on the other hand, had encouraged GOM to consider the upgrading of these settlements as a more cost-effective and less disruptive solution. The upgrading approach was tried out on an experimental basis under the first two urban projects supported by the Bank (pars. 1.16-1.17). Much of Estimated data on 1990 urban population and projections taken froms United Nations Center for Human Settlements (HABITAT), Global Renort on Human Settlements 1986, Oxford University Press, New York. Royaume du Maroc, Ministbre de l'Rabitat, Recensement -gnral des Bidonilles, Rabat, June 1989. World Bank, Mrocco: Informal nougnJa Unradin and Prevention Policies and.ognramq EMBS& Urban Projects Division, December 20, 1983. 4 the GCWIBank urban sector dialogue concerned the issue of how to deal with illegal settlements In Moroccan cities. 2. SuLI and Demand for Urban Shelte 1.10 CON's desire to rid Moroccan cities of bidonvilles and informal settlements in the context of rapidly growing urb" population presents an immense challenge. Just to replace bidonvilles and Informal housing and rehouse their 2.4 million inhabitants would require the production of 480,000 new units, assuming one unit per household and five persons per household. A modest ten- year program to replace all these illegal settlements would thus require annual production of 48,000 new units. At least an additional 103,000 units per annum would be needed to meet the demand of new households as city populations continue to grow at 4.3 percent per year. In short, without considering the replacement needs of an aing stock, some 151,000 new units would be required each year. 1.11 Against this, Morocco's formal housing supply in its best year (1987) was only able to supply 54,000 units.' By Implication, informal solutions have to take care of the rest. This means expanding bidonvilles and informal housing as well as overcrowding of the existing stock. OM has made considerable efforts to increase public sector housing supply, both directly through the Ministry of Housing (MO), or indirectly through the seven regional planning and construction agencies (ERACs) and the General Real Estate Company (CGI). Although significant increases in output were achieved, altogether, the public sector accounted for only 8,300 units per annum.* Thus, the call for the CI Project to focus attention on the private sector, responsible for 85 percent of the formal housing supply, was appropriate. 3. Local Government in Cities 1.12 Rapid urban growth also increased the pressure for administrative decentralization in Morocco. Until the Local Commune Organization Law of 1976, the central government in Rabat had been responsible for most investment decisions at the local level. Under this law, however, local comunes were given greater assurance of financial transfers from C0W, although central and provincial goverrment control over decision-making by local communes remains considerable. According to a recent Bank report, moreover, the specific responsibilities of local comunes are not always clear.' MO data for 1984-1989 reproduced In the SAR for the Second Housing Finance Project (Loan 3122-MOR), para. 2.08. Ibid. * In the words of this reports *The decree of September 30, 1976 is ambiguous in its provisions regarding the tasks that the comunes are to carry out and lends itself to various Interpretations." Degentamtion in Noroccal Local Government EZnendites and Manatement, August 31, 1990, World Bank, par&. 9ile 5 1.13 Although only recently reformed, local goverment has formally existed in Morocco since Independence. Administratively, it consists of two tiers. At the first level, the country Is divided Into 40 provinces and 9 *prefectures,# headed by governors and prefects, all of whom are directly appointed by the King and report to the Ministry of the Interior (M01). Today, provincial governors and prefects are responsible fort (1) executing decisions made by elected provincial aseemblies (ii) coordinating GCW sectoral Investments within their jurisdictiona and (iII) submitting local commnesO budgets and spending plans to MO and HO (Ministry of Finance) for approval. Governors and prefects function, therefore, as 00's regional representatives. 1.14 The second tier of local goverment, to which the V80 Project was mainly directed, is made, up of local comunes, Rapid urbanisation has led to an Increase in their numbers, which now total some 859. Of these units, however, only 99 have significant urban development responsibilities. They are the 59 wounicipalitas* and the 40 *autonomous centers* with jurisdiction over the larger cities and smaller urban centers respectively. The remaining 760 are rural communes on the order of what might be called village councils in other countries. Each cowme has an elected local council, which chooses one of its members as president to represent it, fulfilling the role of what Is often a directly elected mayor In many other countries. 1.15 Morocco Is not a federal system of goverment. The autonomy of the local communes is only Incipient and central government still retains considerable control, either directly or through the provincial governors. nown in Morocco as the fministAre de tutelleO of local commnes, NOX is literally entrusted with responsibility for local government affairs. D. Govement Poli and Bank Strate, 1.16 The Issue of how to deal with unauthorized urban settlements In Morocco was first discussed by 00W and the Bank in 1972 during preparation of the Rabat Urban Development Project and the ensuing debate over the merits of bidonville upgrading versus eradication were discussed in the Audit of that project.' Despite considerable efforts Bank missions to persuade the housing ministry to support upgrading, bidonville eradication prevailed as it does today, Re Moraccos Ubat ad BigSAnd Urban =1Develoent _ProesO Loans 1528119§4- Mt Report No. 9729, June 28, 1991, paras. 1.08-1.11 and 4.05-4.13. 6 although NOR did admit upgrading on a trial basis in the Rabat project. Bank recommendations for low-cost solutions affordable by the urban poor to reduce unsustainable subsidies for resettlement progress failed to diinish 0M's hostility to bidonvilles in particular." 1.17 Nonetheless, as if to give the Bank the benefit of the doubt, O agreed to a second urban operation which included bidonville upgrading in the cities of Meknba and Renitra. This became the Second Urban Development Project, approved in January 1981. As the audit of that project pointed out, however, the so-called upgrading operation in Makne was really an urban renewal and resettlement scheme. Lot boundaries on the original site were completely redrawn, forcing the relocation of a large number of families, effectively eradicating the settlement that existed prior to the works." 1.18 No further projects incorporating bidonville upgrading were agreed between OW and the Bank. The last attempt by the Bank came in its 1983 sector paper on policies and programs for informal housing, which did not lead to the hoped for Bank operation. 1.19 The year 1983, when bidonville upgrading was removed from the sector policy agenda, was also when the CIA Project was appraised. This was a key turning point in Bank activity in support of urban development in Morocco. Bank reports and documentation have been reticent in recognizing 0OM's lack of support for bidonville upgrading. The Bank's urban sector review, for example, claimed that O supported upgrading and yet, at the same time, admitted that without external aid, the programs would cease: "Government commitment to the Iplementation of the [slum upgrading) program clearly indicates that the Bank's efforts to promote the establishment (of slum upgrading) through the first (Rabat) and Second Urban Projects have been successful. Continuation of Bank support for slum upgrading is needed to maintain the momentum of these programs which depend heavily on foreign financing by the Bank and USAID." (Morocco: Urban Sector Review and Prolect Identification. 1982, para 6.19.) GCO and local authority actions, especially in Casablanca, made clear their commitment towards bidonville eradication, which even became incorporated in Morocco's national development planning. (Royaume du Maroc, Premier Ministre, Plan d'Orientation your le Dveloosement Sconomigue at Sociale 1988-92, Rabat, Section III.) The Bank'a failure to dissuade CON from pursuing this policy, which is still in effect at the time of writitg, is amply discussed in the audit of the first two urban projects (Report No. 9729, op. cit*, pars* 3.26 and 4.05-4.13). Ibid., pars. 4.14-4.17. 1Moreccos InfoRMal NoMaS: 11nradinaan Preention Policies and Proaram. Report No. 4787-MOR, December 1983. 7 Missions began to focus upon sites and services schemes. These had been Included in the first two urban operations as core house projects and had proved more acceptable to =0M. Thus, the Bank9a 1982 urban sector paper recommendeds Funding for the sites and services projects should be continued, but directed through the housing finance program, which would Include construction of core housingo.. .Because of its central role in housing finance, the CIA constitutes a unique Instrument with which to pursue a coherent and comprehensive housing development policy in Morocco. By funding through this institution, the Bank could achieve far reaching results in promoting the construction of low-cost housing (including core-housntg) for low-income groups.1" 1.20 In pointing a new direction for its urban sector strategy in Morocco, the Bank vanted to encourage the use of an existing Institution (i.e., CIR) to deliver the more successful product (i.e. sites and services) of the earlier projects more efficiently. By concentrating upon core houses, the bank hoped that the beneficiaries would continue to be low-income urban households. In addition to charting the entry of CIs into Bank-financed urban operations, the 1982 sector paper also supported an expanded role for FEC, arguing that *wherever possible, FEC be used as a channel for funds, since this will strengthen he institution and engender more effective and coordinated Bank operations.U Efforts to develop infrastructure and urban services locally through 00m (with the support of local communes) had only been partly successful under the first two urban projects. An alternative approach through a line of credit to FZC with more direct participation by local communes was an attractive alternative for the Bank's urban strategy in the early 1980s. 1.21 These were important years for Bank sector strategy and 00M sector policy in Morocco. A Bank sector paper pointed the way towards urban lending through CIN and FEC, leading to the appraisal of two projects in 1983, precisely when Morocco's international payments crisis broke. The attractiveness of the new approach from the Bank's standpoint was its ability to "wholesale$ lending via financial intermediaries, to which much project management could presumably be delegated. Such delegation held out the hope, which vas in practice to be frustrated, of alleviating what was considered to be the heavy supervision burden associated with Bank urban projects. It was also attractive to GO, CI, and CDC/FEC, all of whom saw the opportunity of ready access to funds when other resources coming from abroad were in short supply. The stage was set. As this report will illustrates however, many of these hopes were not to be fulfilled, especially in relation to the Bank's strategy of targeting project benefits on low-income urban households. 1 Horocco: Urban Sector Review and Prolect Identification, World Bank NA Urban Projects Division, March 19, 1982, paras. xxxift. and 6.22. Ibid., para. 6.17. 8 a. RBank Inolvmn 1.22 Morocco we an experienced Bank borrower, not only for urban development, but for many other sectors as well. CII, although a newcomer to Bank urban operations, had long been responsible for Bank-financed tourism operations. It was borrower and/or executing agency for four projects between 1976 and 1981, specifically* (1) Second Loan to Credit Imobiller at Hotelier (Loan 848, 1972, US$15 million)# (i) Bay of Agadir Tourism Project (Loan 1202, 1976, US$21 million); (ill) Loan to Credit Immobilier et Hotelier (Loan 1279, 1976, US$25 million); and (iv) Fourth Hotel Development Project (Loan 1943, 1981, US$100 million). Among the findings of the audit of these projects were: (1) inadequate marketing forecasts by CII and (i) the poor financial performance of the hotel loan portfolio." An important lesson learned, in the words of the PAR, was: Relying on a DFI alone to pursue a policy dialogue with government, by virtue of the sector knowledge it gains through appraieal and supervision of individual projects, does not appear to be an effective means of developing clear policy recommendations or bringing about needed policy changes." 1.23 The Bank had already had important experience in providing lines of credit for industrial development in Morocco. These were to support small-scale industries, through GON's National Economic Development Bank (BIDE). They included the Ninth BUDE Project (Loan 2037, 1981, US$70 million), and the Small- Scale Industry II Project (Loan 2038, 1981, US$70 milion). To complete this brief reference to projects related to the two operations reviewed in this report, mention should also be made of the follow-on Second Housing Finance Project (Loan 3122, 1990, US$80 million), whose loan io fully committed at the time of writing. This report makes several references to this operation where they help to understand the performance of the first CIR Project. The desired follow-on operation to the FEC Project, however, did not materialise, a. explained later in this report (paras. 2.12 and 2.16). 1.24 Although inovative in their own subsectoro, the CIE and FEC Projects were only a small part of the Bank's portfolio of related operations. This portfolio represented a rich pool of experience from which ideas and approaches could be drawn. Previous Bank experience in the country was very relevant since it covered either the same sector, the same borrower, or the same line of credit project concept as the two operations reviewed here. Report No. 6388, August 1986, pg. aiv. 17 Ibid., para. 42. 9 A. The Line of Credit Anuroach 2.01 In the case of both projects examined io this report, Bank loans were used to finance lines of credit to CIS and FEC as development finance companies (DFCs) or intermediariese. In turn, these DFCs would on-lend the proceeds of the Bank loans to the final sub-borrowers. CII's sub-borrowers (in the first instance) were real estate developers, while FEC's were local communes and sectoral utilities (known as "r4giee" in Morocco). Both projects were innovative in their respective sectors. The CIR Project was the Bank's first housing finance operation in the region, while the FEC Project was only the Bank's second attempt to finance local development through a line of credit, following the pioneering Cities and Villages Development Bank Project in Jordan (Loan 1826-JO), appraised three years earlier in 1980. 2.02 The benefits of DYC lending, especially for industrial development of the kind financed through BIDE, were extolled in a Bank sector policy paper of 1976.' At the time of the appraisal of the CIZ and FEC Projects, Bank enthusiasm for the line of credit approach was high. The satisfactory results obtained from DEC operations became material for technical discussions in standard texts." The Bank saw the potential for larger and faster disbursing loans. The US$60 million loan to CI was more than the combined value of loans for the first two urban projects in Morocco. Even the small loan to FEC (US$16 milion) was seen by both CON and the Bank as merely the first step towards a much larger operation. 2.03 The "wholesale" approach, whereby the Bank could delegate supervision to the DIC was, according to some in the Bank, an attractive feature. In the words of the CIA Project PCR: "the full value of Bank assistance to a sector or subsector is greatly enhanced when Bank loan proceeds are made through financial intermediaries capable of subproject appraisal and supervision. Such capability frees Bank staff from having to scrutinise minute details, allowing them, instead, to focus on broader institutional and sectorial issues" (CIN PCR Evaluation Summary, para. 9). As it turned out, neither the CII nor the FEC Projects led to a reduction in the level of supervision effort required by the is World Bank, DeveloMent_ Finances Comanies - Sector.. Policy Paer, Washington D.C., 1976. See, for example: Baum, Warren C. and Tolbert, Stokes M., Investin in Develoments: Lessons of World Bank Exoerience, Oxford University Press, New York, 1985, pp. 205-206. 10 Bank." From a borrower's point of view nonetheless, a line of credit operation can be attractive, since it holds out the prospect of larger, faster disbursing loans. This aspect appealed to 00 as it headed for an international payments crisis in the early 1980s. 2.04 The design of these operations generally requires that only eligible subprojects meeting criteria satisfactory to the Bank be financed. In practice, however, the fungibility of Bank loan proceeds within the DIC ean. that a line of credit effectively supports the operations of the DIC as a whole, Including those which may not meet the eligibility criteria. For this reason, a thorough institutional appraisal of the agency as a whole has to be a key feature of DIC operations. B. The Project Intitutions 1. MI 2.05 As one of ON's most Important specialized financial agencies, CIS is also Morocco's principal housing bank, with total reported assets of DR 15,1 billion (US$1.7 billion) in 1990. As the name implies, it operates in both the housing and tourism sectors. In the housing sector, CIA lands to public and private developers and makes long-term mortgage loans to house purchasers. For tourism, lending is also made to developers, but for hotels and tourist complexes. Approximately, two-thirds of CII's leading supports housing activities, while the remaining third is destined for tourism. CIS mobilizes resources through long-term borrowing on the local and International markets and through the issue of short-term discount notes. Since 1989, CII has also been able to collect deposits directly from the general public. 2.06 Since its creation In 1920, CIS has been constituted as a private company, although, In practices it Is 0M's most Important instrument for executing national housing policy. CIH's link to 00M Is not direct, since the majority ownership (55 percent) is in the hands of CDG and of the remaining shares in CIN, insurance companies own 20 percent, commercial banks 15 percent, and Norocco's Central Bank 10 percent (CII PCR, para. 20). CIS's administration is centralized in its headquarters in Casablanca. By 1991, however, CIN had acquired some 41 small branch offices throughout the country, most of which were established after 1982. Management operates through six departments (called directorates), each of which deals with both housing and tourism. The financial directorate was responsible for the Bank-supported CIN Project. 2.07 The 1983-90 period, during which the CIS Project was implemented, was one of significant growth for CII. Its staff expanded from some 400 to today's 2o Bank supervision efforts measured by the naber of staff weeks per million dollars of Bank loan for the two operations reviewed here was approximately half that of the first (Rabat) urban project, but twice that of the Second Urban Project. Staff weeks per million dollars of Bank loan for the four projects were: 3.8 - Rabat Urban Developments 0.7 - Second Urban Development; 1.3 - CIE Projects and 1.9 - FEC Project. (Data from respective PCRe.) 11 700, and its total assets nearly quadrupled. CIN9s total Incomoe from interest In 1983 was DR 435.1 million (US$54.0 million), reporting a not profit of DR 72.0 million (US$8.9 million). The corresponding figures for 1990 were DR 1,542.9 million (US$191.8 million), and a net profit of DR 183.8 million (US$22.9 million). This growth was accompanied, however, by a progressive weakening of the Institution's financial structure (detales pares. 3.05-3.15). In terms of day-to-day management, however, CIN remains an efficiently run institution. 2.08 Being well known to the Bank as the result of four tourism projects (para. 1.22), was a key factor recommending CII as borrower and executing agency for the housing finance operation. According to the PCR, however, CIS was chosen because it was the only financial Institution that could grant long-term loans to the housing sector in Morocco (CIN PCR pares. 2.04 and 8.03). At appraisal, nevertheless, the Bank considered the possibility of incorporating the "Banque Centrals Populaire" (BCP) into the project. BCP is a commercial bank with experience in financag low-cost housing programs (I) since 1974. BCP1s eventual participation was ruled out, according to the SAR, due to its tradition of risk aversion (CIE SAR par&. 2.15). The audit was informed that BCP itself did not want to participate in what it considered to be a highly risky operation, although its possible involvement had been sounded out by Bank missions. 2. CDGlPHe 2.09 An analysis of the institutional context of the FEC Project inevitably takes CDG as Its point of departure. From policy, financial, and managerial points of view, CDC to the institution that controls the EC operation. For the Bank, CDC held final responsibility for the execution of the FEC Project. Although FEC was given distinct judicial status through a 1959 law and presents its own financial statements, it is under the direct control of CDG's senior management. At apprasal, FEC was treated as a Odivision# of CDG that had been upgraded to a 'department."2 Another way of understanding FEC is simply as a special fund, whose management is entrusted by CON to CDG and whose policy and administration follow CDC norms. As such, FEC is one of three such funds whose work programs are administered by CDO." 2.10 CDG Is one of CR's most Important financial Intermediaries, with reported assets of DR 12.6 billion (US$1.6 billion) in 1990. Its responsibilities are wide ranging, but are basically to channel savings mobilized 34 The SAR makes a fleeting reference to FEC's 'financial autonomy, since FEC is entitled to borrow in financial markets with CON guarantee (FEC SAR para. 3.01). Decisions to borrow and to seek CONs guarantee are CDG's however. n The other two are RCAR - "R4gIme collectif d'allocation de retraite* (with twice FEC's resources), and CNRA - 'Caisse nationale de retraites et d'essurance" (which Is seven times smaller than FEC). (Data frome Catase do Dep6t et do Gestion, Ramnort Annuel 1990, Rabat). 12 through state Insurance and pension schmes and postal savings accounts into priority investments for 00W. As already mentioned, CDG Is majority shareholder of CIR (para. 2.06). The Investment portfolio financed through CDG Is similarly wide raniing, and Includes housing and real estate, urban land development, touriesm, industry, transportations Insurance, and local communal development. CDG's headquarters are in the capital Rabat, where the majority of ito 522 staff are located. 2.11 At appraisalg FEC was a very small operation, with a staff (on CDG's payroll) of only nine. FZC's function was to channel resources by making loans with maturities of up to eight years to local communes and sectoral utilities ("r6gies") for investment in urban infrastructure and services. FEC's principal source of funds was short-te:m CDG discount notes. Final decisions to approve loans were not FEC's, but were taken by a special Technical Committee, which included representatives of NO and 07, and where FEC was represented by the Executive-Director of CDC. FEC still does not have its own offices, although plans exist to build them. It presently occupies an annex of CDG's headquarters building in Rabat. FEC grew rapidly over the 1983-90 period, while the Bank project was being implemented. Currently, its staff numbers some 44. In 1983, FEC reported total assets of DR 561 million (US$69.6 million) and total interest income of DR 36.7 million (US$4.6 million). The equivalent figures for 1990 were for assets of DR 2,113 million (US$262.7 million) and interest income of DR 207.5 million (US$25.8 million). This growth took place against a background of uneven financial performances profitability declined and the capital structure remained weak, although FEC became more solvent and its risk management improved (details: paras. 3.16-3.24). 2.12 The Banks appraisal report Included a long list of FEC's weaknesses. These included: (i) no proven track recordl (i) lack of formalized policies and operating procedures (iii) lack of rigorous operating principles for project financing; (iv) cursory appraisals of technical and financial feasibility; (v) insufficient personnel in number and experience; (vi) Imbalance in financial structure with a long-term loan portfolio mainly funded by Short-term discount notes; and (vii) undercapitalized for an institution with no proven track record - debtsequity ratio of about 7:1 (FEC SAR, parss. 3.01, 3.06, 3.11, 3.25, 3.30). 2.13 An Important question to ask is whether these problems reflected the weakness of an institution, or the absence of an institution. The Audit is inclined to conclude the latter. The SAR itself hints that it did not consider FEC to be a real institutions "If successful, this pilot operation could lead to an expanded role for FEC, including eventually full legal autonomy and financial independence, so that it can become a aenuine financial institution for the development of the LCs, able to mobilize on their behalf new resources to meet CDG operates in the housing sector through its subsidiary the General Real Estate Company (CGI OCompagnie 06nrale ImmobillareO), which plans to deliver 5,654 middle Income housing units over the next five years. CGI is an important borrower of CIN. The CIA Project financed three CGI housing schemes in the value of DR 120.0 million (US$ 14.2 million) (CII PCR Part III Annex 2). 13 their large investment requirements, and provide another source of finance in addition to the central Government budget."(PEC SAR pars. 1.02 - emphasis added). 2.14 The issue of whether FEC was or was not an institution Is not merely semantic, since policy changes and managerial Improvements as part of an institution building effort under a Bank-financed project must evidently be addressed to a real institution. In the case of the FEC Projects institutional appraisal and institution building were focussed on NEC itself, without due attention to CDG as the agency de facto responsible for the operation. In the view of the Audit, a key factor in limiting the nC Project's progress toward establishing an autonomous institution for local commune financing was that project design did not contemplate the active engagement of CDG in the process. The need for a thorough institutional appraisal (which would include CDC) in this kind of operation is treated as one of the points of special interest below (paras. 4.13-4.14). 2.15 What was missing from FEC that prevented it from being considered a "genuine financial institution" by the Bank's appraisal team? Part of the answer can be found in a memorandum from the Bank's Legal Department at the time of project preparation. Referring to FEC as "a simple account operated by CDG, the memorandum noted that: (1) FEC's autonomy is limited by the CDG's General Director$s authority to grant loans; (ii) FEC's right to borrow, and whether CDG or FEC is responsible for repayment, are unclearl (iii) EC's address is not indicated in its statutes; and (iv) FEC's statutes do not indicate how its staff and directors are appointed. In short, the opinion was that FEC did not have the legal and institutional status to take charge of a Bank- financed project. It raised the basic question "who would be responsiblo for the Bank loan, CDG or FEC?" The SAR itself recognizes FEC as being a part of CDGt FEC was initially managed as a division of CDG and, since 1980, as a department. As such, FEC has no proven track record of its own, and its image and credit-worthiness are strongly linked to CDG's....FEC is managed by CDG, and in practice FEC's chief executive is the General Director of the CDG. He holds substantial control over the policy and operations of the FEC and is involved, through loan approvals, in FEC's daily management. Furthermore, as head of CDG - the main supplier of resources to FEC - he practically retains overall control on EC's financial policy. In addition, the administrative tasks for FEC (recruitment, payroll, staff regulations, benefits, steg) fall within the responsibility of CDG." (FEC SAR paras. 3.01, 3.04) 2.16 At the time of the Audit, FC lacked some of the basic trappings that would normally be found in a financial institution, such as a financial director, a legal advisor and departments responsible for marketing and information, for example. In spite of these misgivings, however, the FEC Project conceived of FEC as an institution in its own right. This may have given rise to expectations of institutional progress, particularly with regard to policy and finance, that FEC a' Legal Department Memorandum of August 17, 1981. 14 as a fund operated by CDG, could not realistically fulfi1. EC could not make itself autonomous from CDG for example. Focussing on 1EC as an institution meant that the project lacked instruments, such as legal covenants agreed to by CDG, which could have encouraged a move towards establishing the genuine financial institution hoped for at appraisal 2.17 During 1985-87, however, 00K did consider creating a local commune development (BDCL - "Banque de DWveloppement des Collectivit6a Locales") out of FEC, along the lines of the Cities and Villages Development Bank of Jordan. The Bank even fielded a pro-appraisal mission in 1987 to consider lending to the institution-to-be. Although draft legislation was prepared, it was not sanctioned and the BDCL vs still-born. It appears that the idea of creating a fully-fledged BDCL aroused much opposition, not the least from commercial banks, who feared it would tighten Morocco's financial markets even further. At the same time, the controversial question of controls whether it be with CDG, MOI, or full autonomy, would have to be answered. With the limited institutional progress, the Bank postponed its further support for local development through FZC. The Region has now informed the Operations Rvaluation Department that a new law has been promulgated under which FEC Io undergoing a full restructuring. A follow-on operation, First Municipal Finance Project, which has recently been approved, includes a specific action plan toward FEC's institutional development. C. Issues In Protet Prenaration 1. g[ 2.18 Spurred by CTI's familiarity with Bank procedures from its earlier experiences under the tourism loans, preparation of the CIN Project was very rapid. Barely twelve monthe elapsed between the identification mission and appraisal. Initially, the Bank had hoped to include a bidonville upgrading component in the CIN Project. Only by March 1982 did the Bank accept GM' as, and hence CI*, strong opposition to such a program (details: paras. 1.16-1.18). From that date, project preparation concentrated on the housing finance component only. This focus was fostered by Bank staff skills in financial analysis which became available at that time. 2.19 With the increasing popularity of the line of credit project concept in the Bank in early 1980., there was strong support by Bank management for taking this approach in Morocco's housing sector through CIN. Not surprisingly, the main issues identified by Bank preparation missions were financialt namely the financial risk of CIE operations involving excessive subsidies and inadequate resource mobilisation by the institution. The income level of potential beneficiaries was not cited as a Issue at this stage. Despite the Banks acceptance of 00's unwillingness to upgrade bidonvilles, the bidonville issue would not go away. The project Issues Paper highlighted concern that Bank funds might eventually be used to finance CIR*o bidonville eradication and resettlement program. U CDG's own views on the autonomy of FEC are contained in Annex 1. 15 2. fg 2.20 Preparation of the nEC Project was much more protracted. Although it wee identified sixteen month. earlier than the CIN Project, the appraisal missione for both operations were fielded at approximately the same time. The degree of Bank discomfort with nC as an Institution matched the degree of its confidence in the familiar CIS. A memorandum early in project preparation set the tone of the Bank management'e miegivinge with FECs To assign to such a weak organization with no track record, the task of on-lending Bank funds In a wide variety of etors (a sort of mini-World Bank) for projects which typically will not have any financial or commercial discipline (e.g. will not Involve the self-interest of an investor) involves real dangers that the Bank funds could be used for quite uneconomic projects." 2.21 Although it was later included in the Bank' e urban portfolio, the NC project was prepared by the region9s industrial development and finance (IDF) division. Perhaps an underlying paradigm of credit worthiness of industrial enterprises Induced an initially critical view of the seemingly high risks involved in lending to local communes that were not driven by commercial priorities. 2.22 To placate misgivings about the operation, its preparation proceeded as a pilot project. Also, Bank missions harnessed Important arguments in its favor, seizing upon support to FEC as an opportunity for a major institution building effort. The project was also seen as a chance to use NC as an instrument for reducing regional disparities in the country. The discussion of these and other issues and internal arguments within the Bank was a fruitful ones but it meant that the nEC Project took approximately twice as long to prepare as the CIS operation. Preparation also required four missions compared with just two for the CIH Project, as well as twice the level of staff effort. D. Project A iratsal 1. 2Z 2.23 Like preparation, appraisal of the CIE Project was rapid, focussing on an analysis and projection of CI's consolidated financial statements and some financial aspects of the agency's housing operations. One-third of the appraisal report is devoted to an analysis of CIR, and more than half of the SAR' s annexes consist of financial tables. Some of the appraisal projections are compared with CIIR's actual financial performance later in this report (paras. 3.05-3.15). Appropriately for a line of credit operation, financial analysis was the focus of project appraisal. Projectione, done for the first time with the use of a computer spreadsheet, contained errors (for Instance, total liabilities for 1982- 86 do not include a portion of equity due to a spreadsheet error), but they did ' Memorandum from IDFDR of July 17, 1980. 16 not affect the computationg monitoring and iplementation of the covenanted financial ratios. 2.24 Scant treatment vas given to CIS policy and priorities. The SAR stated thats *CIZ9s Policy Statement approved by the Board of Directors of CIA in 1974 has been reviewed by the Bank and found satisfactory.* (CIS SAR para. 4.02) * The SAR itself does not review the policy statement, nor is the statement listed among documents available in the project files. In the view of the Audit, the value of a line of credit operation supporting a DIC is considerably enhanced by including the basic policy statement of the agency supported in the staff appraisal report, as was done In the case of the FEC Project (FRC SAR Annez III). In the circumstancess it to surprising that the Bank did not propose improvements to what was a nine-year-old policy statement. This might have been appropriate with regard to policy on arrears, for example, in a financial institution exposed to serious delinquencies on hotel loans which accounted for about one-third of its portfolio. 2.25 On the other hand, the appraisal of the C1 Project gave more attention to the eligibility criteria of housing subprojects and sub-borrowers, which were detailed in Schedule S of the Loan Agreement. Maxiaum ceiling prices of eligible housing were Initially set at DR 32,000 (US$5,333) for the least expensive (core unit) and DR 80,000 (US$13,333) for the osat expensive (valk-up apartment) (CII LA Schedule 5 and SAR para. 3.07).* An appraisal analysis demonstrated that such units would be affordable by the lower half of households on Morocco's income distribution curve. In addition, technical criteria specified maximum standards for physical specifications, such as plot sise and built area, for example. Further eligibility criteria required that all projects had a mininum economic rate of return of 12 percent end, for those costing more than US$500,000, a minimum financial return of 14 percent (CIR SAR pare. 3.07). According to the SAR, the price criteria were derived from CIS market studies reviewed by the Bank (CIS SAR pare. 3.07). In actual practice, the ceiling prices were raised substantially later on during Implementations as discussed extensively later in this report. s' The 80,000 ceiling price corresponded to the bottom end of the range of house values (DR 80,000 - 130,000) under 0099a existing low-cost housing program (RBM -*Habitation a Bon March6m), funded since 1974 by ICP, and later administered jointly with CII. See, World Bank, Morocas-rbna Sector,.ftew and Prolect 9Identification March 19, 1982, para. 4.2. Interestingly, the sector paper highlights affordability problems for low- Income families of the BM houngs * ... given the cost of these (RBM) units, households with monthly income below DR 1,500 (in 1982 prices), or 35 percent of the prospective housing demand, cannot afford them.* 1.7 2.26 The Audit noted Inconsistencies between the ceiling prices set and project cost estimates.a* The appraisal Inconsistency could have been the result of an exaggerated coat estimate or an understatement of the number of units to be financed. In either case, however, the appraisal calculation may have sent a signal to the borrower that higher unit coat parameters had in fact been budgeted than those formally agreed in the projects@ legal documents. If perceived in this way, such a signal would have stimulated CIN to aim for higher cost uniteS9 thereby helping to undemine the project's low-cost housing objectives. 2.27 The reasons for the choice of US$60 million as the loan amount remain a mystery to the Audit. According to the SAR, the value had been "scaled down" from the US$81 million called for (CIN SAR pars. 3.03), although the PCR reports that it had been increased from US$45 million (CIR PCR par&. 4.02). The Audit can only add that, if the average unit cost was equivalent to the median of the ceiling prices, namely US$9,333, a Bank loan of only US$39 million would have been sufficient to finance the 28 percent share of the cost of the 15,000 units foreseen in the SAR. 2. IN 2.28 The appraisal mission focuseed on detailed Institutional and managerial aspects of the nBC operation. However, as already aentioned, this did not include an Institutional appraisal of CDG. A Statement of Policy was agreed which defined nBC's objectives and explicit lending criteria. One of the most important agreements reached was that FBC loans should only be used to finance revenue generating projects (FnC SAR 2.08). Total project costs, excluding technical asistance to C18, were estimated at appraisal to be DR 1,257 million (US$209.5 million), of which the Bank loan would fluance US$59 million, some 28 percent of the total. Construction loans to the value of DR 1,213 million (US$202.1 million) would be provided to build 15,000 new units according to appraisal estimates (CIN SAR par&. 5.02). These estimates Indicate an average cost per unit of DR 80,860 (US$13,473), strangely higher than the maximu agreed ceiling price of DR 80,000 (US$13,333). With four ceiling prices ranging from D 32,000 (US$ 5,333) to DN 80,000 (US$13,333) agreed under the CIS Project, the average imnit cost Ahal have been somewhat below the median value of the ceiling prices, namely DI 56,000 (US$9,333), assuming an equal number of units financed In each of the four price ranges. The Audit noted conflicting technical data between the appraisal report and the Presidents Report concerning the number of housing units to be financed under the CII Project. Contrary to the 15,000 figure given in the SAR, the President's Report (para. 63) and the Bank's Press Release, stated that 13,000 new housing units would be financed. If this figure was correct, the Inconsistency noted in this paragraph would be even more acute, saince average unit costs would be even higher, namely DR 93,300 (US$ 15,546). 18 2.29 The appraisal undertook a le detailed review of F8C's finances than had been the case with C18. The maturity Ibalance of IC's balance sheet, with long-term loans financed by short-term discount notes, was a matter of concern. So too wee FEC's debtsequity ratio of 7s1, a very high rate for an agency without a track record. Recognizing CDG's ability to come to FEC's aid if necessary, the SAR argueds "such an tabalanced financial structure Ishowever, acceptable as long as 1IC's financial liability and credit-worthinese are an integral part of CDG. The financial structure of TIC at this stage to therefore not as isoe relevant to the successful Implementation of the pilot project." (TEC SAR para. 3.30) The SAR wee pointing out, correctly, that PEC was not an autonomous financial institution but did not shift its focus to CDG itself. Such an analysis might have helped Bank missions to develop a deeper understanding of CDG as one of orocco's most important financial intermediaries. With COG's direct involvement, the dialogue about establishing a mechanism for financing local commune development in Morocco might have reached a successful conclusion. 2.30 The longer preparation time of the IC Project permitted a thorough appraisal of the eligibility conditions for TIC lending. A detailed 28-page annex to the SAR spelled out the necessary Information and criteria for 12C subproject appraisals. As a result of a commendable collaborative effort within the Bank, Information was supplied by the relevant operating division, about appraisal criteria for projects in seven sectors, namely: water distribution, electricity distribution, urban transportation, productive facilities, urban development, sewerage, and municipal road projects. The agreed criteria were then set out explicitly in a Side Letter to the Loan Agreement and were very helpful to IC staff in designing in-house appraisal methods. 2.31 In the case of the TIC Project, the Loan Agreement we signed by CON as the borrower and a Project Agreement was signed by both CDG and FTC. In addition, four Side Letters specified specific conditions of the project. In addition to the one just mentioned, which dealt with appraisal criteria, a particularly important one was Side Letter 2, in which the project's financial covenants were detailed (detailst pare. 3.19). This Side Letter was signed only by O, making only the Government formally responsible for the fulfillment of FEC financial performance covenants. a. Prolect Obiectives and DescrJ2tion 1. 0W Projeal 2.32 According to the SAR, the objectives of the CIR Project weres - to encourage private and public developers to produce low-cost housing schemes affordable to Income groups on ,;he lower half of the urban Income distribution curve (i.e. with a monthly household Income below US$330 In mid-1982 prices)l and - to develop C189s ability to appraise low-cost housing schemes and to advise private and public developers on all aspects of low-cost housing design. (CI SAR para. 3.01). 19 2.33 Both objectives emphasized stimulating the delivery of low-coat housing by private and public developers. With respect to the criterion for defining low-cost housing in the present discussion, this report uses the same one as the CIR Project itself at appraisal, namely units costing less than the max1mum ceiling price of DB 80,000 (pars. 2.35). The focue on low-cost housing was appropriate given the enormous shortfall in the supply of this kind of housing in Morocco, as noted earlier in this report (para. 1.10-1.11). The objectives were not stated in the Loan Agreement, which nevertheless contained covenants related to low-cost housing. This could have contributed to a better understanding by all parties of the spirit, as well as the letter, of an operation. Fortunately, the Loan Agreement of th follow-on Second Housing Finance Project (Loan 3122-MOR) with CUS does include such a statement. 2.34 According to the Loan Agreement, the project consisted oft Pa=As A program of subloans for the construction of low-cost housing units and land development for purposes of such construction later on, and of mortgage loans to beneficiaries utilising the proceeds of subloans repaid to the Borrower. PaU.Bs The strengthening of the capabilities of the Borrower9s staff to appraise low-cost housing development and advise the housing construction industry on low-cost designs through: (i) the development and application of a land use model to evaluate the cost-effectiveness of alternate land use patterns suitable for low-cost housing developments (ii) the development and application of a computerised data Information system (including the provision of a computer) to compare and analyse the cost- effectiveness of low-cost housing development, including the housing schemes, financed or to be financed by the Borrower, and the provision of appropriate training for the Borrower's staff In the application of said information systems and (tit) a study to evaluate the status of the housing construction industry within areas served by the Borrower and to recommend suitable measures for expanding said industry's iavolvement in low-cost housing development. 2.35 Schedule 5 of the Loan Agreement detailed the criteria for housing schemes eligible for financing out of the loan proceeds, under Part A above, for which 99 percent of the Bank loan was destined. Four types of housing were considered eligibles (i) core units - DR 32,000 (US$5,333); (ii) one floor partially completed houses - DR 48,000 (US$8,000); (iii) one to two floor completed houses - DR 62,000 (US$10,333); and (1v) three to four room units in 20 multi-floors houses - DR 80,000 (US$13,333) (CIS LA Schedule 5). These values were significantly higher than the low-cost targete set for the Chiles Public Sector Housing Project (Lu 2482-CH), which was appraised at the eae time as the CIR Project. In the Chile Projects core houses were esttmated to cost US$1,870, while the most expensive "economic housing" had a price ceiling of US$9,860 (02D PAR pars. 8). Aside from possible differences in construction costs between the two countries, this suggests thate from the outset, the design standards adopted in the Chilean case were more appropriate In ters of their affordability to low- Income households than those utilized In Morocco. 2.36 The CIN Project provided loans to public and private developers for a period of three years to finance housaing construction. CIS undertook to convert developers' repayments at the end of the three years into mortgage loans of up to 25 years for purchasers. to acquire the units built. Given the Bank loan term of 17 years, such an arrangement can lead to a maturity mismatch, since developer-plus-mortgage loans could have a total maturity of up to 28 years. In the case of Bank disbursements for developer finance made In 1989, final repayments of related mortgage loans may not be due to CIN until the year 2017, seventeen years after the due date of the Bank Loan itself in the year 20001 Underfunded in this sense, the project will oblige CIE to mobilize additional resources at some point in the future In order to sustain the mortgage lending program foreseen under the project. Such a consequence raises concerns about the long-term financial health of CIR, an Institution which, already at appraisal, was recognised as facing a resource gap (CIR SAR para. 4.14). 2. FEC Proleat 2.37 Through Project Agreement covenants, CDG and FEC undertook to execute the project in accordance with the agreed Statement of Policy, in which the objectives were stated (PA sections 2.01(a) and 3.01). The appraisal report summarized them as followes 'In support of the Government's regional and communal development policy, the objective of the project is to assist local comunes In their efforts to meet their basic Infrastructure needs... The project would more specifically assist EC in formasting operating principles for project financing, on the basis of rigorous preparation, appraisal and supervision standards and criteria, and Improving the efficiency of its staff in the performance of their functions.' (FEC SAR pars. 4.01) Appropriately, the FEC Project pursued a twin goal of Improving local infrastructure provision, while at the same tine strengthening the agency that could channel funding for this purpose. 2.38 According to the FC Project Loan Agreement (Section 3.01), the project consisted of the following: (1) the financing by FEC of Investment projects through sabloans to 'Investment entities' In accordance with the provisions of the Statutes, the Statement of Policy, and the Schedule to the Project Agreements and (11) the training of FEC staff in project preparation, appraisal, and supervision through the provision of expert services and fellowships to FEC staff. Through cross references to readily accessible documentation, the project description was au ct and provided a clear guide to the output expected from the contractual agreements between C0N, the executing agencies, and the Bank. OInvestment entities' were understood as local authorities, their associations ("syndicate*), and local utility companies 21 (rftes*). The role of these entities was made clear in the Statement of Policy mentioned earlier, which also provided a detailed list of eligible subprojects. 111. PROJ EMLMBUTIOQN AND REULTS A. Start-up Problems 1. Sl 3.01 Rven though the loan was quickly made effective within four months of signing, practically no disbursement occurred during the first year of effectiveness. This was a disappointing result for a line of credit operation that was expected to disburse rapidly. According to the PCR, the disbursement difficulties were due to the project's technIcal criteria being too rigid and the selling price ceiltg too low (CIS PCR para. 4.04). CIR clearly holds the Bank responsible for thiss "The difficulty [of disbursement] lay in the Bank's undereatimstes of the real cost of low-cost housing as compared to the norms laid down by the Moroccan authorities well before the conclusion of the credit line.* (CII PCR Part 11-1I1.) 3.02 If the ceiling prices agreed at appraisal Inplied the production of housing that was unacceptable by G0's and local standards, the project would of course, have been technically infeasible. The appraisal report offers a different explanation of how the housing types and hence ceiling prices were determineds "Four types of housing units have been identified on the basis of market studies carried out by CIS and reviewed by the mission." (CIE SAR para. 3.07). Irrespective of the question of who was responsible for defining the ceiling prices, there was a hiatus to project start-up through the Inability of CIN to elicit a supply response for housing unite at those agreed prices. 3.03 CIR prevailed upon the Bank to raise these liits which it is worth recallIng, were already considerably higher than those of a siailar housing finance project in Chile appraised at the same time (details pare 2.38). Only after they had been iacreased some 44 percent (by 1985) did the rhytbm of disbursements begin accelerate. By 1988, CI's ceiling price for low-cost housing had been increased by 87.5 percent (against an accumulated Inflation of 38.4%" in dirhams) to D1 150,000 (US$18,270 in 1988). The PCR reported a lower ceiling price liit for the CIN Project itself of DR 100,000 (CI1 PCR para. 5.01). This does not seem to have been observed in practice, however, since thirteen of the 43 developer contracts awarded were for housing unit prices between DR 108,955 and DR 163,333 (CIR PCI Part III Annex 2, page 1). From the point of view of the project's low-cost housing objectives, the Increases made This point, in relation to the rising cost of land and building materials is made once more by CII in Annex 1. Sources 1991 IMf International Financial Statistics. 22 the solution unaffordable by the intended low-income benefioiaries. This aspect is discussed in some detail later in this report (parse. 3.29-3.33 and 4.02- 4.04). 2. U& 3.04 Disbursements against the FEC Project were also low at the outset. This was due to the tardy preparation and processing of subprojects by local communes unfamiliar with the new approach adopted. In this respect, performance of the FEC Project was similar to the experience of municipal development projects in other countries. As more proposals were submitted and PEC staff become more adept at appraising them, disbursements began to accelerate. B. Finangial Pe formMa 1. CIN 3.05 A strengthened CIE was one of the by-products expected from the CIN Project, but CI* financial performance weakened during the 1983-90 period while the project was being Implemented. This was true even during the second half of the period (1987-90), by which time Morocco'a macroeconomic performance had begun to improve. 3.06 Profitability declined, falling below appraisal expectations. CIR's return on investment (measured as net profitastotal assets) fell almost every year from 1.78 percent in 1983 to 1.32 percent in 1990. This was against an appraisal forecast of 2.47 percent by 1986.u Return on equity (measured as net profitesequity) fell from 26.5 percent in 1983 to 19.83 percent in 1990, against an appraisal forecast of an increase of 33.03 percent by 1986. CIH's current ratio (measured as current assetscurrent liabilities) also worsened during the 1983-90 period. From 2.1, in 1983, it had fallen to 0.9, In 1990, affecting CI's ability to meet cash needs to pay short-term debt. With ClI's actual financial performance generally weaker than forecast at appraisal, it is diffi- cult to understand the PCR's conclusion that "CIE financial performance indicators have been in accordance with the projected figures in the Staff Appraisal Report" (CIR PCR para. 6.03)." The appraisal report of the follow-on second housing finance project (Ln 3122-MOR), analysing the same period as the In ite comments on the PAR in Annex 1, CI notes that the decline in the profits was not important, since that decline resulted from substantial loss provisions that had to be made over the years. CIN, therefore, recommends that the analysis be based upon its gross profits which revealed a better performance. Table 4 of Part III of the PCR compares the actual values of eleven performance indicators with the values projected at appraisal. In nine cases out of eleven, actual performance was worse than projected.. Actual profitability Indicators, according to the PCR, were particularly poor, being only one-third the rates projected. The Audit*s analysis of CIT's income statements and balance sheets suggests that profitability ratios reported in the PCR may have been under-reported. 23 PCR, notes thats "the analysis of critical performance Indicators, shows some signs of deterioration (CI I SAR para. 4.14). 3.07 In terms of risk management, CIR had been reluctant to make provisions against possible losses, especially with regard to its high risk hotel portfolio. Provisions for losses as a proportion of total loans were 0.28 percent in 1983. They rose as high as 0.63 percent in 1987, but fell back again to 0.24 percent in 1990. At appraisal provisions were forecast to rise to 1.2 percent of total loans, even not taking iato account the worsening of arrears on the hotel portfolio that actually took place. Action programs for reducing hotel arrears were agreed with the Dank at negotiations. (CIM SAR para. 6.01). 3.08 One modest risk-averting covenant did not seem to have been complied with on two occasions during Iaplementation. To spread risks among sub-borrowers (as well as to avoid the monopoly of the loan proceeds by a few large developers), CIR undertook not to finance any housing project whose value exceeded 20 percent of CI's "aggregate" capital defined as: equity+surplus+free reserves (CIH LA Section 4.10). Nevertheless, a large contract in 1986 for 2,660 units in Kohamadia in the amount of DR 121.8 million (US$14.0 million), corresponded to 34.4 percent of CIR$s reported equity that year. Another, in 1987, for 2,589 units with ANP was valued at DR 174.1 million (US$22.3 million), or 44.6 percent of Cm3's equity. These two contracts by themselves accounted for 31.9 percent of Bank loan disbursements under the project. The PCR did not report non-compliance with this covenant however. 3.09 Other aspects of CIF's finances fared better during the 1983-90 period. For example, debt service coverage (measured as interest income sinterest payments), at 1.5 in 1983 and 1.3 In 1990, remained quite stable. The Loan Agreement required that CIZ maintain a debt service coverage ratio (amortization plus interest payments receivedtamortization plus interest payments made) of at least 1 (CIS LA Section 4.5 a) . The PCR reported compliance with this covenant. Also positive was the fact that CI's administrative overheads remained constant throughout this period. In 1983, administrative expenses were the equivalent of 0.94 percent of total assets and 8.75 percent of total Income. The equivalent values for 1990 were 0.96 percent and 8.69 percent, respectively. These results are consistent with CIR being a well-run institution on a day-to-day basis. 3.10 Concerning CIR's capital structure, performance was mixed. In 1983, CIR's long-term debt:equity ratio vas 11.5, which the appraisal mission considered high, projecting a ratio of 9.8 by 1986. However, CIH continued borrowing heavily and its long-term debt:equity ratio rose to 15.5 by 1988. Recognizing that it was seriously undercapitalized, a major subscription of DR 400 million (US$98.5 million) was made to build up CIt's equity base in 1989. After that, the long-term debtsequity ratio fell back, with CIE reporting 9.7 In 1990.," instead of agreeing an arbitrary debtequity imit, as in some other operations, the CIR Project opted for an Interesting, but complex formula (covering more than two pages of the Loan Agreement), that determined the limit on the basis of differentiated multiples of the values of CIR lending for housing In its comments on the PAR (in Annex 1), CIt adds that the most recent debt equity ratio figures were: 11.32 In 1991 and 10.13 In 1992. 24 and non-housing operations (CI IA Section 4.04). In addition to assuring adequate capitalization, the formula was intended to stimulate housing and discourage other (notably tourism) operations in CI. Although directed to a laudable goal, this formulation risked setting very high ceilings (18.4 in 1990, for example) if CIS managed to shift all its lending into housing through debt rescheduling or reclassification. A better solution for the problem of CIE's overexposure to a poor quality hotel portfolio, in the view of the Audit and with the benefit of hindsight, would have been to include loan provision as legal covenants of the project. 3.11 CIH grew significantly during the 1983-90 period in spite of the weakening of its financial position. In terms of the value of its assets, CIA was nearly three-and-a-half times larger In 1990 than at project appraisal." Accounting for some 12.8 percent of CIR's long-term debt In 1990," the Bank remained an important source of financing for CIN and a stabilizing factor in its balance sheet, but it frovided only 7.3 percent of the funding that led to the growth just mentioned. Perceiving that ClI's growth owed little to the Bank after more than two decades and six Bank loans may have induced "borrower fatigue" in the institution. Growth-oriented management and staff found the loan amount was Insignificant in relation to CI's resource needs. Even so, legal obligations under the project and information and reporting requirements of the Bank and its missions were felt to be demanding. Because of the attention given to this topic by Moroccan interlocutors during the Audit mission, this to treated as a point of special interest later in this report (paras. 4.06-4.09). 3.12 Bven with its record of growth, closing a potential resource gap remained a constant challenge for CIN. A financing plan was Included in the CIX Projectl and efforts to 1aprove resource mobilization through studies of new savings products were included as a component of the follow-on second housing SS Concerning this growth, the PCR concludes thats #...by increasing its income and assets three to four times during project taplementation, CIR has become one of the strongest banks in Morocco* (CI-PCR para. 6.02). The trends of some key CIS financial indicators reviewed here would suggest that growth alone is not a guarantee of the institutional strength of a DFC. In CIR** audited balance sheet for 1990, Bank loans outstanding totalled DR 1,154.6 million (US$143.5 million). This value includes amounts pertaining to the earlier tourism loans and the follow-on second housing loan. s? Over the 1983-90 period, Bank lending in CIROs balance sheet increased by US$89.7 million, or 7.3 percent of the overall increase of CIX's total liabilities of US$1,236.3 million. The main sources of growth during this period weres (i) CON and other loans - US$ 648.2 million; (ii) short-term discount notes - US$248.0 millions (iii) short-term loans - US$145.1 million; and (iv) equity - US$71.5 million. The growth of short-term liabilities was responsible for CI3's solvency problems already mentioned. 25 finance operation. In Morocco's tight local financial marketi" CIR has had to rely upon CDG and BCP as its principal sources of funding. Partly due to this tightness, most CIE bonds Issued on the local market have been subscribed to by CDG. Thus, in 1990 for example, CDG subscribed to 83 percent of CIH obligatory bonds, in the amount of DR 872 million (US$108 million), and BCP took up 64 percent of a CIR uedium-term bond issue to the value of DR 250 million (US$31.1 million). CIE is now entitled to take deposits from the general public, but these cannot be expected to grow at the rates to which CIS has been accustomed. Just to maintain its level of operations, CIR will need to mobilize further resources, especially as key privileges such as the VAT (value-added tax) exemption on its loans expire. 3.13 Through the solid guarantees given, mostly In the form of land and property, arrears on CI8's housing operations, affected only 1.9 percent of its housing loans. In contrast, hotel lending accounted for one-third of CIR's operations, and yet was responsible for two-thirds of the arrears (CIH PCR para. 6.03). The table attached to para. 6.03 of the PCR demonstrates how hotel arrears grew twice as fast as those for housing durig the execution of the CIE Project. As a number of these poor quality loans are held by state-owned enterprises, the chances of a political settlement of outstanding debts, possibly not in the interest of CIN's own finances, is always present. For this reason, a line of credit operation needs to thoroughly appraise other operations of the DFC which might Impinge upon the performance of the sector directly supported by the project, and agree covenants about their financial performance. The need for thorough and broad institutional appraisal for an operation like the CIR Project and agreed Improvements in the performance of poor quality portfolios is discussed as a point of special interest later in this report (paras. 4.10-4.14). 3.14 Although the foreign exchange risk was borne by 00 (Guarantee Agreement Section 3.02), CII as borrower undertook In the Loan Agreement to "...take such steps satisfactory to the Bank as shall be necessary to protect itself against risk of loss resulting from changes in the rates of exchange between the currencies (Including Dirhams) used in Its leading and borrowing operations.* (CIR LA Section 4.08). CII Interlocutors of the audit mission raised the issue of foreign exchange losses arising from Bank loans as CIS did in its PCR (CII PCR Part II III.A.1). By chance though, two trends in international currency rates have so for reduced the cost of repaying Loan 2245- NOR to the Banks (i) the Dirhsm revalued (in terms of the US dollar) as loan amortization began In November 1987; and (ii) strong currencies (notably the Japanese Yen and the Deutsebmark, which made up 52.7 percent of the loan) were already highly valued while disbursements were being made. Although lucky so far, CIE is correct to be concerned about possible foreign exchange risk. As a result, the issue is being more directly addressed under the Second Housing Finance Project. 3$ The Moroccan economy itself continued to suffer from a resource gap, which, In 1990, was equivalent to 6 percent of GDP, given that gross domestic Investment In that year was 26 percent of GDP and gross domestic savings 20 percent (data from the 1992 World Delooment Renort, Table 9). Hence the structural need to seek external funding. 26 3.15 The weakening of CI's financial performance during project Implementation (1983-90) Io a disappointing outcome for a line of credit operation designed to strengthen a DFC's finances." Could the Bank have done more in assisting CI to halt this treadt Given CIB's growth objective and the Bank's limited financial contribution to that expansion, its chances of influencing events was probably limited. Nonetheless, in a case such as thiss detailed and regular monitoring of financial performance indicators Is necessary for the Bank to be well informed. 2. InQ 3.16 Strengthening nC and Its finances was also the hoped for outcome of the line of credit project. Like CI, however, nEC's financial performance deteriorated over the period while the project was being Implemented. In some respects, particularly profitability and capital structure, nC' s performance was weaker than CIPs, while in others (e.g., short-term solvency and risk management), it was stronger. But comparison between nC's and CIROs finances is subject to the proviso that NEC was always an integral part of CDG. This meant that, if willing to, CDC could provide injections of capital, loans, or cash to overcome nEC's financial weaknesses, as necessary." 3.17 PEC' s profitability deteriorated over the 1983-90 period, especially after it lost access to the Central Bank's low interest rediscount facility in 1986. At appraisal, return on equity (measured as net profitsiequity) was projected to grow to 9.20 percent by 1986; the actual rate for that year was only 1.44 percent. After reporting losses In 1988 and 1989, nEC earned a small positive return on equity of 0.66 percent in 1990. Although CON was covenanted to ensure a mInimum 3 percent interest spread for NC, during 1987-90, the actual spread was negative. The increase of nEC resource costs was so sharp that even keeping its lending rates above the 10 percent agreed with the Bank, as it did, was insufficient to ensure a profit. 3.18 As mentioned earlier (para. 2.12), nEC's capital structure was basically unsound if it were to have been an autonomous financial institution. The basic problem identified at appraisal was that NEC's long-term lending was financed for the most part by short-term discount notes (NEC SAR para. 3.30), but its credit worthiness was, in practice, guaranteed by CDG.1 In 1983, for example, 72.0 percent of nC's liabilities were short-term. Limited long-term borrowing meant that the long-term debtsequity ratio in 1983 was only 1.1, while the overall debtsequity ratio was 6.6. After the loss of access to low-cost In its comments on the PAR (Annex 1), CIR states that a more global view of CII finances would reveal as Improvement of the agency's financial situation (up to 1992). In its comments on the PAR in Annex 1, CDG confirmed the "absolute support" CDC always gave to maintain nC's financial soundness. *1 Even so, the weakness of nEC's financial structure was given by the appraisal report as the reason for making the Bank loan to CON and not directly to CDGJnEC (nEC BAR para. 3.30). 27 Central Bank rediscountingg FEC's liabilities were substantially restructured in 1987, through the provision of substantial long-term loans. Long-term liabilities increased from a more 7.4 percent of the total in 1986 to 66.5 percent In 1987. Short-term liabilities, on the other hand, fell from 82 percent of the total in 1986 to 25 percent In 1987." 3.19 Notwithstanding this financial restructurIng the FEC operation remained undercapitalized for most of the 1983-90 period. CON was covenanted under Side Letter No. 2 to the Loan Agreement to ensure that nBC's debtsequity ratio would be less than 7. This target vas met only in 1983 and 1985. By 1990, the ratio had soared to 17.4. 00 made two DR 20 million contributions (in 1985 and 1988) to FEC's equity out of the four annual Install1ants of this amount that were expected. This small Increase in FRC's equity base was Insufficient to bring PEC's capital structure back into balance. A constraint upon the growth of its borrowing would also have been necessary. 3.20 In some other respects, FEC's finances showed improvement during the period when the project was being Implemented. Probably the most important was that the FEC operation became more solvent. Its current ratio (measured as current assetscurrent liabilities) was a very precarious 0.3 in 1983. At that time, sC had Insufficient short-term liquidity to meet its cash needs. The capital restructuring of 1987 brought Important relief to FEC by reducing its current liabilitiesl the current ratio of that year rose to 0.8. Further Improvement was sustained until 1990, when the ratio became a healthy 1.6, due to a further reduction in short-term discount notes among FEC's liabilities. Possimistically, the Bank's appraisal forecast had foreseen FEC's risky 0.3 ratio persisting throughout Implementation of the project. 42 From project documentation, it appears that the Bank did not play a significant role In this tportant decision. The July 1988 supervision mission's Back-to-Office report made only a three line mention of this restructur.,nS, which it learned about from the recently issued audited accounts of 1987. The PCR's reference is almost as brief, but using syntax that might be misunderstood, leaving an Impression that short-term borrowing substituted long-term bonds, when the opposite was in fact the case. The PCR*s sentence should reads *Debt consolidation through the substitution of short-term borrowing by CDG by long-term bonds reduced FEC's maturity mismatch." (FEC PCR para. 5.05) 28 3.21 WeC's risk management also Improved during project Iaplementation. At appraisal, FEC made no provisions for losses, and M05 was covenanted to ensure that reserves equivalent to 1 percent of EC's unsecured loans be made (FEC Side Letter 2). By 1990, FEC's provisione for losses amounted to 0.9 jercent of all long-term loane, probably ensuring compliance with this covenant. VEC lending to local comnunes proved to be low-risk, since repayments were guaranteed through central government transfers. On the other hand, arrears among utilities ("r6gies") worsened, casting them in the same role as CIROa hotel borrowers which spoiled an otherwise good quality portfolio. 3.22 The steady decline In FEC's administrative overhead between 1983 and 1990 indicates an Improvement In the operation's day-to-day management. Thus, in 1983, administrative expenses were equivalent to 5.3 percent of total incomel the proportion in 1990 was 3.9 percent. Although an Important Improvement, this still exceeded the appraisal forecast of 3.2 percent for 1986. 3.23 FBC grew at an even faster rate than CIR during project execution, albeit from a much smaller base. FEC's total assets in 1983 were DR 561 million (US$69.6 million), only 13.9 percent of CIO's in the same year. By 1990, FEC's assets had increased almost fourfold to DR 2,113 million (US$262.7 million), some 15.1 percent of CIN$s. Even though substantial, this growth vas less than the 20 percent per annum projected at appraisal. The Bank's financial contribution to that growth was not significant. During the 1983-90 period, long-term loans reported as liabilities In FEC's finaocal statements Increased by DR 1,633 million (US$203.1 million), while Bank lending on FEC's books increased by only DB 98.9 million (US$12.3 million), 6.1 percent of the total. The Dank loan was intended only to finance a small pilot project, but the audit mission perceived disappointment on the part of CDC and FEC management with the small amount involved. Clearly, CDG and FEC management felt that their hopes for a major Bank contribution to their growth strategy had been frustrated. 3.24 In sumary, FC's financial performance during project Implementation vas mixed. Major growth was accompanied by declining profitability and undercapitalization, but brought some improvements in terms of solvency and risk management. The Bank's hopes of influencing this performance were diminished by the insignificance of the loan amount in relation to FEC's total resource requirements in the eyes of the fund's growth-oriented management. In spite of the improvements, moreover, FBC's finances are still not those of an Independent financial institution and, as noted earlier, OW decided not to proceed with the creation of another autonomous entity to handle FEC type of operations. Important groundwork has been laid, however, for both of these events to occur in the future under the recently approved follow-on operation. 43 The PCRIs review of the Status of Covenants (FEC PCR Part III Table 5) does not refer to those included in the Side Letters to the legal agreements. The Audit itself was unable to confirm compliance, since an assessment requires specific date on FEC loans to "r6ges" which are not published in the financial statements. The Audit recomends that, where possible, legal covenants draw upon published data to make these agreements effective and transparent management Instruments. 29 C. Proiect Reeglts 1. CIX - fousin= 3.25 The project accounted for some 252 of all the housing unite financed by CIN during the 1983-90 period, a significant share of CIR's total sector operations. The funding of each of the 15,449 units completed under the project involved two successive financial operations from CIE's point of view. The first was to provide three year construction loans to developers to build the house, and the second was to provide long-term mortgage loans to purchasers to buy the unit built. This Is normal housing finance practice in many countries. In the PCR, however, CHS only counts the project's contribution to the second instance, namely the value mortgage financing. This ignores construction loans financed under the project that were provided for all units built. By comparing the project's mortgage finance only with CI's total construction-plus-mortgage finance, the PCR estimated the project's contribution to CIN9s overall housing finance operations to be only half what it approximately was. 3.26 The physical Implementation of the CIE Project was relatively straightforward, except for some cases of inadequate land titling affecting some schemes. Unclear ownership rights made some otherwise eligible prograns impossible for CIB to finance, since they did not offer an acceptable guarantee for the loans (CIE PCR para. 5.01). Significantly, land titling uncertainties in some Moroccan cities had also been found to be a major problem in the two earlier urban development projects (PAR Loans 152811944-MOR, para. 6.06). 3.27 As mentioned earlier (para. 2.26), Bank documentation reported contradictory physical targets for the projects the SAR indicated 15,000 units would be financed (CIN SAR para. 5.02), while the President's Report specified only 13,000 units as the target (CIE PR para. 53). The remainder of the discussion In the present report, however, refers to the 15,000 figure. The actual number of units financed exceeded both expectations in any case. As anticipated at appraisal (CIN SAR para. 3.08), the large majority of these units, some 72.9 percent of the total, were located in the Casablanca/Rabat areas. 3.28 The appraisal estimate of the total cost of the housing component (i.e. the total value of sub-loans to be financed under the project) was DR 1,257 million (US$209.5 million). The actual value of sub-loans by completion had fallen to DR 891.5 million (US$105.3 million), Increasing the Bank's share in project financing from 29.1% to 54.71. With 15,449 units built and funded, the average sub-loan per unit, according to the PCR, fell from an appraisal estimate of DR 83,000 (US$13,333) to DR 57,706 (US$6,816). The achievement of these dramatic savings to not analyzed in the PCR. Nor does the PCR explain why, if unit sub-loans fell so sharply, was it necessary to successively increase the ceiling prices of the units? 3.29 An analysis of CIE lending per unit over the 1983-90 period indicates that, for CIS operations as a whole, average lending per unit actually increased significantly, as can be seen from the data in Table 1. For public developers it rose from DR 54,300 (US$6,700) In 1983 to DR 121,300 (US$15,100) in 1990, an increase of 125.4 percent, about double the accumulated inflation of 62.3 percent during the same period. For private developers, lending per unit rose much 30 faster from DR 58,000 (US$7,200) In 1983 to DR 303,600 (US$37,800) in 1990, an increase of 423.4 percent, or seven times faster than price inflation. These are disappointing trends for a project aimed at stimulating the production of low- cost housing. This conclusion assumes, of course, that subloans as a share of 31 total unit costs have been fairly constant over the period. The average private unit financed by CIS is probably affordable only by the top docile on the Income distribution curve." CIR Project landing to developers, which averaged D1 57,706 (US$6,816) per unit, was however, less than half the average of CIB's overall lending to public developers and less than one-third of CIR's typical lending to private developers. 3.30 CI mortgage lending also followed a trend towards more costly housing, as can be seen from the data In Table 2. Between 1983 and 1990, the number of "special regime" sortgage loans (i.e., those for low-cost housing) awarded fell by one half, while the number of higher income "general regime" loans doubled. There was also a significant Increase in the average value of general regime loans. The special regime loans, among which the CIN Project mortgage loans would be classified, maintained a fairly constant value over the period around an average of DR 69,300 (US$8,000), somewhat higher than the likely value of mortgages awarded under the project, namely DR 57,706 (US$6,816). Particularly significant, though, was the sharp decline in the nmer of special regime loans in 1987 and 1988, when only 3,108 and 1,818 were awarded, respectively, in contrast to the 9,779 awarded in 1986. 1987188 would have been the period when Bank funding under the CII Project first became available for mortgage financing, following the repayment of the earliest developer loans. Altogether, CIN committed DR 687.5 million for special program mortgage loans between 1987 and 1990. This was less than the DR 896.5 million that the PCR reports that the project alone made available (CIR PCR Part III Annex I pg. 5). From the data, it would appear that the large majority of CIO low-income mortgage loans since 1987 have been financed through Loan 2245-HOR. A precise estimate If the mortgage loan had the same value as the per unit value of the CII loan to the developer, the buyer'* monthly repayments would be US$ 449 for a US$ 37,800 loan over 15 years at 12 percent Interest per annum. With housing expenses not exceeding 30 percent of income, this would require a monthly household income of at least US$ 1,497. Even deflating for 1982 prices, such an income as Is only enjoyed by the top decile in Morocco would be necessary in order to afford the average housing offered. The distribution is as followss Monthly .ousehold Income .in US 2 of Households I of Total Income - Unoer Limit of Decile 102 4.02 199 202 5.8% 204 302 6.22 225 402 6.82 289 502 -7.4Z 332 602 9.02 387 702 10.12 434 802 11.32 513 902 14.02 799 1002 25.42 - Sources: CIR-SAR para. 3.09 and World Develowment Renort 1992 Table 30. Data refers to the 1982-1985 period In 1982 prices. 32 of the proportion financed io not possible due to the lack of data regarding the conversion of developer loans into mortSa8e operations. It appears, howe"v that CIN relied almost exclustrely upon the Bank loan to fund low-cost mortgage loanag while concentratiag its other resources on expansion of the higher cost general regime progra. 3.31 Because of the funSibility of resources within a DIC9 indlrectlyp the Bank loan allowed CIS to withdraw its regular support from low-cost housiag and concentrate on more popular higher Income operations. The PCR Uapies thato without the Bank loan for the CIN Project* the low-cost progress might have ceased, altogether. In the words of the PCRs *The Project sustatned the pace at which low and moderate-cost houstag constructon loans were granted when, the g=vrnment_nae.OrAmIIt.Xed=n=tgitulanyon.0 (CIB PCR Rvaluation Sumary para. 5 - emphasis added). In the view of the auditq a Bank-financed operation Is bound to be less than fully sucsesetul if the borrower io not committed to Its objectives. The experience of this operation with respect to its low-iacome targeting seems to bear this out. 3.32 Asset iaflation io the bouetag subsectors especially the risiag prite of urban landq was oited by Iaterlooutors of the audit mssion in Casablanca as the reason for the Increasiag cost of heasiag. Unfortunatelyq the available data only perat comperisone for the 1987-90 period. Ovr those four yearog CIN construction lending per unit rose by 51.0 percentg against an Iacrease of 41.7 percent for urban land in Casablancaq 20.0 percent for building materials# and 16.1 percent for consumer goods generally (data from CIN Aaaual Reports). While 33 land prices rose significantly, the tapliat value of the houses for which CIE was lending, rose faster still, suggesting that higher quality units were being delivered. 3.33 By Increasing the cetling prices of eligible housing by 87.5 percent over 1983-90, the CIR Project followed the sae trend as CIA lending generally. As Morocco's principal housing bank, CIR has to uphold the priorities of its shareholders and support 00 housing policy. Financial or political considerations may dictate an approach that does not favor low-cost housing. For the Bank, funding an operation ostensibly aimed at low-Income households and then supporting a strategy that favor. higher Income groups jeopardising an important sector goal. The Bank agreed Increases In the ceiling prices, but there was no explicit revision of thr 5bjectives of the CIR Project, as approved. A Bank supervision mission of harch 1986 objected to the rising trend of house standards, recommending that US$20 million be canceled from the line of credit." Except for that occasion, the Bak was generally supportive of the changes, finally seeing merit in them as the PCR reporteds "the Bank's main strength was its flexibility In adapting the changes in the Loan Agreement which facilitated disbursements of the loan proceeds.. (CIS-PCR para. 8.01). The Bank's flexibility did indeed accelerate disbursements, but at the cost of losing the low-cost beneficiary focus of the project. The project's failure to benefit low-income urban households in Morocco Is treated as a point of special interest later in this report (pars. 4.02-4.04). 2. CIN-* Technical Aseistance 3.34 According to the PCR, the project'e technical assistance component, which aimed at strengtheniag CI's appraisal capacity, Was successfully Implemented. However, CIX interlocutors informed the audit mission that they felt the operation had brought little change to the agency, whose appraisal methods, they said, were well established prior to the project. The Bank appraisal mission's own assessment of CI's capability In this field was quite positives In the course of more than half a century's experience in dealing with housing projects, CIE has developed a sophisticated data set concerning building construction costs and standards and also a good feel for housing market conditions throughout the country. It is against this background that housing projects are evaluated... .CI's appraisals of housing projects are of high quality, but somewhat weak regarding economic aspects. Recently, however, as a result of discussions with Bank assions, CIE appraisers have paid more attention to the economic analysis of projects and have devoted more tine to discussions with developers and to advising them on project concept and design (CIR SAR para. 4.03), a3 Back-to-Office report of April 8, 1986. 34 3.35 The Audit mission was iapressed by CIB's in-house skills and procedures of project appraisal and control. It was more difficult to aseese what CI's present capabilities owe to the project. It was clear, however, that the project stinulated the computerization of project management and accounting tasks that are now carried out with a high degree of professional competence. 3.36 The project *a technical assistance component was to have specifically focussed on low-cost housing, but, consistently with the trend towards financIng higher cost units, CIS does not appear to have treated low-cost housing as a special aspect of technical assistance. CIN was also to have provided technical assistance to developers to help them develop low-cost housing solutions. Ifforts by the audit mission to obtain information from CIR about private developers, however, were unsuccessful. Vrom the CIS staff interviewed, the audit mission had the ipression that CIS is not actively developing its data base on private developers. The mission's contacts with such developers confirmed that their dialogue with CIS on a technical level was not a very active one. 3. P, 3.37 According to the PCR, some ninety subprojects were financed under the loan. The Audit mission was given a list of 109 subprojects by PBC, which included the PCR's ninety, but with some variations on the amounts per subproject. Both figures are considerably higher than the 30 subprojects anticipated at appraisal, but are consistent with a project ained at financing small subprojects, up to a maximm value of US$1.0 million. (ISC-SAR para. 4.09). The PCR provides a detailed breakdown showing the variety of borrowers and the different sectors served (nEC PCR Part III Table 4). The portfolio of investments financed under the loan to very diverse both by type of sub-borrower and sector, with no one type predominating. FEC lending (in terms of value) was shared in siailar proportions by the four kinds of sub-borrowers, namely rural communes (26.6 percent of total), autonomous centers (16.0 percent), "municipalit6sO (33.1 percent), and Or6gies" (24.4 percent). Among sectors, commercial infrastructure (consisting mainly of markets) accounted for the largest share (24.6 percent), with urban transport (which included the acquisition of vehicles) In second place (17.1 percent). 3.38 By including many different types of subprojects in different cities, the investments were consistent with the objective of assisting local communes in their attempts to meet basic infrastructure needs. The level of effort, however, was very small, so that the loan made only an insignificant contribution to meeting the total local infrastructure needs throughout Morocco. nEC$s lending program as a whole could generate a wealth of information about the priorities for local investments in Morocco, from the point of view of local communes and provinces. The Audit recommends that a cross-sectional analysis of this experience be carried out in the future by type of local authority, city size, region, and poverty impact in order to help orient sector policy formulation. Understanding of the priorities of the local communes would be particularly important. 35 3.39 PEC kept very good control of the local Investments it financed during execution, but did not follow them through with systematic monitoring. FEC's follow-up and monitoring unat, with only four professionals and one microcomputer, is seriously understaffed and underequipped. It to responsible for overseeing a portfolio worth some US$300 million spread throughout the country in a multiplicity of small investments. With the unit*a present limited resources, field visits to the subproject sites, for example, are Impossible. In the absence of monitoriag, it Is not possible to be sure about how well these Investments are performing and how effectively they are being maintained. Weak supervision and monitoring was recognised as a problem by CDG and IEC management, who assured the Audit mission that they were addressing this shortcoming. 3.40 Technical assistance and training provided under the project helped strengthen FEC procedures and management. It to Important to remember that, according to the appraisal report, recognised operational procedures hardly existed at all In FEC prior to the Bank-financed operation. A number of FEC staff improved their project appraisal capabilities by on-the-job training during implementation of the project, reaching a high level of technical competence. The Audit mission was Impressed by the in-house skills developed in the use of electronic spreadsheets for project appraisal. Today, some six staff (out of a total of 44) are familiar with the techniques used, a high skill density within the organization. 3.41 An informal on-the-job approach did not work as well, however, for the elaboration and dissemination of the FEC operations manual. A formal manual was not prepared since FEC management felt that the line of credit approach was well enough understood by all staff. While FEC was still very small, this informal approach may have worked, but an enlarged FEC needs to have a clear statement of the policy, rules, and procedures of the operation. The fact that many local commune dossiers of proposed investments had to be returned by FEC because they were incomplete is indicative of an incomplete understanding of the operation's rules. An operations manual would not only clarify such rules, but its use would help highlight where Improvements in procedures could be made. In the view of the Audit, this type of operation benefits considerably by having its rules clearly and formally set out for the information of all project participants. D. Suetainabiill 3.42 In the case of a line of credit to a DYC, there are two distinct dimensions to sustainability, namelys (i) the long term prospects for the DIC itself; and (ii) the physical permanence and continued performance of the specific investments financed through the operation. Each aspect is briefly assessed in turn. 1. SU 3.43 The sustainability of CIN as a financial intermediary in Morocco's housing market is directly itaked to the itstitution's financial performance, analysed in some detail earlier In this section (paras. 3.05-3.15). CIR*s declining profitability, increasing risks of short-term insolvency, and Incautious risk management are of concern to those responsible for its future 36 financial health. On the other hand, CIB8* oapital struature remained solid, even through a period of rapid expansion of its leading operations. Perhape the most Important challenge for CIS In sustainiag its operations In the future I to ensure that it can continue to mobilize adequate resources to avoid a financing gap. This will not be an easy task in Morocco's tight financial market, where interest rates might soar if DICe and banks compete too intensively for scarce funds. Given the Increasing cost of the housing Units it finances, CIS will have to mobilize more and more resources just to maintain the same level of physical output of the recent past. In order to sustain the level of its operations, the audit considers that CIN will inevitably be obliged to develop lower cost financial products, such as loans for core house production, of the kind originally foreseen under the project. 00te* own action would be Important too, in increasing the supply of urban land (principally through deregulation). This would help slow down asset inflation and enable the financial resources of CIE and other banks to reach a greater number of beneficiaries. 3.44 As far as the physical austaluability of the housing financed by CIE is concerned, experience in Morocco and elsewhere has shown that, when they own their homes, residents generally provide the proper saintenance of their housing units. The construction quality of apartment dwellings built by both public and private developers that were visited by the Audit mission In Casablanca was good. Nonetheless, a recent Bank report mentioned poor quality of work as one of the key problems of the Moroccan construction Industry." What did concern the Audit was the poor quality of the public spaces in the very large (30,000 unit) Salmia housing scheme In Casablaca, partly financed under the CIR Project. If these common areas are allowed to continue to deteriorate, falling property prices could undermine the value of mortgages pledged and hence the financial viability of the scheme for CIS. Urgent action by the local authority and residents themselves is necessary to sustain this kind of neighborhood as a viable urban community. Overalls the sustainability of CIS and that of specific investments financed through its operations are rated as likely."' 2. CDGIFC 3.45 As mentioned earlier JA this report and in other Bank documents, 8C was integrally a part of CDG, so its sustainability depended directly on the sustainability of CDG itself. Not being an independent financal agsancy, nRC did not bring together the necessary financial conditions for sustained development, as the analysis of IMC's financial performance above showed (paras. 3.16-3.24). nFC's profitability and capitalization were not adequate for what would be required of an autonomous financial institution. Still, CDG to at the same time nC's guarantor and principal source of funds. As it did through the restructuring of nC's liabilities to 1987, CDG can come to nC's aid when Ann B. 81wan, The Constrotion. Sector in . MNA Countries, World Bank Internal Discussion Paper, Zurope, Middle last and North Africa Region, August 1990, pg. 63. 47 The rating of uncertain sustainability at the time of the PCR review has been upgraded due to more comprehensive Information obtained during the Audit. 37 necessary. Although the Audit did not undertake a detailed review CDG's own fInances, there was no reason to doubt that CDG, as one of Morocco's largest and most important financial institutions, would not be able to sustain PEC's operations through difficult periods, should the need arise. 3.46 The oustainability of the specific Investments made through FEC will depend upon the effective operation and maintenance of them by the beneficiary communes and "r6gies". Since nEC does not effectively monitor these investments (para. 3.39), there is no systematic Information on how well these investments are being operated and maintained. In addition, the majority of them are recent construction and it is still too early to assess the adequacy of maittenance., Nevertheless, the well known limits to local communes' maintenance and operational capabilities is a cause of concern making sustainability of benefits uncertain. IV. OTHER POIMTS8O0 SPECIAL INTEREST A. Poverty sImact 4.01 Both projects assessed in this report aimed to bring benefits to low- Income urban households in Morocco. The CIH Project was the more explicit of the two in targeting low-income beneficiaries by aiming to stimulate the supply of low-cost housing solutions affordable to those on the lower half of the incoZ1 distribution curve (CIR SAR par*. 3.09). The iEC Project's low-income beneficiary targeting was only indirect and Implicit, mainly through the attempt to deliver infrastructure and services to poorer regions of the country. The appraisal report did recognise, however, that the FEC Project was part of the Bank9s sector strategy of extending services to rural and urban low-income groups (nEC SAR para. 2.15). 1. maI 4.02 The appraisal poverty targeting of the CIR Project on households with monthly incomes of up to DR 2,000 (in 1982 prices, or US$333) was modest one in terse of urban poverty in Morocco. A Bank study in 1989 estimated the urban poverty threshold to be a household monthly income of DR 1,030 (in 1985 prices, or US$107) for families in the lowest decile of the income distribution." The more modest appraisal parameters were nevertheless reasonable to help ensure the operation's financial feasibility as well. The 87.5 percent increase in the ceiling prices over the 1983-1988 period (para. 3.03) meant that housing units eligible for CIR finance under the project were no longer affordable by the World Bank, Norocco: Reachina the Disadvantatedt Social Exoenditure Priorities in the 1990s. Vol, 1LIain enort, Oct. 19. 198.. nag. 3 and Anendix 1. ehis study estimat2d tM urban noverty threshold for-Morocco as DR 2.473 wer gerson per year tin 1985 nrces)., eauivalent to a minimum monthly income of DR 1.030 in 1985 urices (U8107) for a five person urban honsehol- 38 original target population. In order to purchase a house at today's top ceiling price of DR 150,000 (in 1991 prices, or US$18,405), a minimum household monthly income of DR 4,456 (in 1991 prices, or US$547) would be necessary. This would be to repay a 15 year mortgage loan at 12 percent interest corresponding to 75 percent of the value of a the unit. 4.03 The risk of failing to target low-inom beneficiaries was recognized at appraisal, but not specifically as a result of the project's delivery of more expensive housing unit. In the words of the appraisal reports "there is a danger that the pressures in demand resulting from a shortage of housing could lead to higher income groups obtaining the shelter facilities targeted for the lower Income population.... As CIR has very limited experience In low-cost housing financing, there is a risk that it will tend to finance subprojects in the highest price range allowed by the eligibility criteria." (CIA BAR pars. 5.04-5.05). In the view of the Audit, the risk that the project would deliver higher standard housing units than planned was exacerbated by the perhaps inadvertent signal the Bank gave in favor of higher cost housing by budgeting average unit costs in excess of the agreed norms (detailss para. 2.26). 4.04 The CIS Project was appraised and presented to the Board as a line of credit that would assist a DPC well known to the Bank stimulate the delivery of low-cost housing in Morocco. The project financed units are affordable by the lower median-income population, but the broad sector Impact was the opposite of that intendeds (i) instead of increasing the supply of low-cost housing, the standard and cost of housing units financed by CIN In general rose significantly (pars. 3.28); (ii) with the sharp Increase of ceiling prices, the CIA Project itself was able to finance more expensive units than foreseen at appraisal (pars. 3.03); and (iii) Bank loan proceeds were nevertheless used to finance what remained of a low priority lower-cost housing program, freeing CIR to use its other resources for priority mortgage lending for higher standard units (para. 3.30). Given that the projectOe objectives were not formally revised, the operation failed to achieve its basic aim as set out in the appraisal report when the loan was approved by the Board. 2. FEQ 4.05 The FEC Project financed local commune subprojects throughout Morocco, some in the poorer regions of the country. Nowever, the location of an investment in a poor region Is not by itself a guarantee that its benefits will be perceived by the poor of that locality. Nonetheless, many of the investments made, notably covered markets and public transport infrastructure and equipment, brought benefits mainly to lower income Inhabitants of the local communes. A definitive conclusion about the poverty impact of the TEC Project as a whole, however, must await the results of an effective project monitoring system that FEC has yet to Implement (detailes pars. 3.39-3.41). 39 B. For Borowers. Loan Amounts. Wre Instanifteant 4.06 Both CIN and CDGIF2C interlocutors made it clear to the audit mission that they considered the Bank loan amounts for their respective projects to be insignificant in relation to the needs of each subsector. In return for these small amounts, they felt that fulfilling conditions agreed in legal documents and meeting the Bank's reporting and information requirements were a heavy price to pay. CIN made the point about the relative unimportance of the Bank loan to it resource mobilization in the PCRt "although Loan 2245-MOR provided the fairly large sum of US$60 million, it represented a small proportion, averaging less than 2 percent of the total funds mobilized by CIE during the loan disbursement period 1983-90" (CIE-PCR Part II II.A)." 4.07 Growth-minded managers of both CIR and FEC were evidently frustrated by the limited contribution Bank funds made to the expansion of their operations. As already noted, CDG vas a much more Important source of funds for both operations as a whole (paraes. 3.12, 3.16, and 3.18). Although the Audit mission did not have access to the details of the conditions of CDG finances, it probably came with fewer technical and policy conditionalities tbpz the Bank's much smaller amounts did. The lesson of these experiences is that, when growth to a DFC's prime objective, the Bank's Influence Is prnportional to its financial contribution to the agency*s expansion. When alternative sources of funding expand rapidly, the Bank's scope for action and influencing events is even more limited. 4.08 For CIE, an institution that has now been responsible for six Bank- financed projects, the risk of "borrower fatigue" (para. 3.11) can be considerable if its management feels that little can be gained for the agency's Srowth objective from a comparatively small loan amount. If this happens, a Bank-funded operation such as the CIA Project and its objectives risk becoming demoralized and its Importance in the eyes of the borrower diminishes (paras. 3.17 and 4.06). In these circumstances, project goals are unlikely to be fully achieved. Future operations are likely to be more successful when the Bank's own strategy for the sector Is in harmony with the objective function of the DFC, namely growth in the case of CII. 4.09 As a new executing agency for a Bank-financed project, the risks of CDG/FEC developing borrower fatigue were lower. In addition, the prospect of a substantially larger follow-on loan -- sincerely contemplated by both 00M and the Bank if the pilot Project were to prove successful -- served as an Important stimulus to PEC management and staff at least during the first half of project implementation. When a second loan was postponed, the Insignificance of the 49 In its comments on the PAR in Annex 1, CDG, confirmed that Bank financing made only a very small contribution to ICs balance sheet. so An analysis of CIR's 1990 balance sheet Indicates a higher, but still very small, contribution of Loan 2245-MOR to CIE's capital structure. By the end of 1990, the US$ 60 million loan was fully disbursed and US$ 16.2 repaid to the Bank, leaving a balance of US$ 43.8 million, or some 3.9 percent of all long-term loans to CIS. 40 pilot project loan for CDG/FEC was brought into perspective. By that time, however, Important improvements to EC's procedures and operations under the project had already been achieved. C. Institutional Anyalsal In Lime of Credit fterations 4.10 The experience of both operations reviewed in this report underscores the importance of the Bank's having a thorough understanding of the responsible DIC in a line of credit operation. In providing a line of credit which is fungible within a DPC, the Bank is effectively supporting that agency'* work program as a whole and not just the eligible components on which the loan proceeds are directly targeted. Medium to long-term development gains will be elusive if a line of credit helps Improve the performance of the target sector only, while the credit worthiness of the institution as a whole is undermined by the deteriorating quality of another portfolio. 4.11 The fungibility of resources within CHI was well demonstrated under the 1housing finance project. Experience under this project demonstrates that, when a Bank loan accounts for only a small share of a DFC's resources, even explicitly targeting the loan proceeds on eligible low-cost housing cannot guarantee that the supply of that type of housing will be stimulated. An alternative project design is thus recommended. Instead of using its funds to finance specific subprojects, the Bank could explicitly agree overall (but precise) targets with the DFC (say, at least two-thirds of all housing units financed each semester to be low-cost). The Bank could then finance a time-slice of the overall program over a given period, disbursing on the condition that the targets had been met. Since such targets would have to be DFC-wide, the Bank's institutional appraisal would have to encompass all the main activities of the institution, especially its appraisal and lending criteria for all sectors. 4.12 The CIE Project, which was ostensibly concerned only with housing, had to face the difficulties raised by CIR arrears on hotel loans. The Bank's appraisal recognized the problems posed by the hotel portfolio and even agreed an action program for reducing arrears, but gave no details of how this was to done, nor how the program was to be followed up (CIN SAR para. 6.01). Despite this program, delinquencies on hotel loans worsened during the implementation of the project. With the benefit of hindsight, it now seems clear that the housing finance project could have helped CIR deal more effectively with this problem if the legal documents had included financial performance covenants aimed at improving the quality of the agency's hotel portfolio. 4.13 The PEC Project experience also demonstrates the Importance of thorough institutional appraisal, but for different reasons. In the PBC case, the key Issue was one of institutional identity and responsibility. During project preparation, the Bank mulled over the first issue, when there was much doubt about the Institutional status of FEC (paras 2.13-2.16). Although FEC was treated as the project executing agency at appraisal, 00 was the borrower and CDG also signed the Project Agreement. Experience has shown that CDG was the "a gag project institution. By focussing its institutional appraisal on FEC, however, the Bank did not develop a fuller understanding of, and relationship with CDG itself. A knowledge of CDG' policies, priorities, management, 41 financial products, and criteria for decision making by the Bank would have been a necessary condition for the Bank to engage it In a sector policy dialogue. 4.14 The principal intention of the 1EC Project was to create a viable and, in due course, autonomous financial Institution. The Bank's postponing of a follow-on operation appears to Indicate that the performance of the pilot project was not considered successful enough to justify Immediate further support to WC. Between 1985 and 1988, 00 did consider creating an autonomous local commune development bank (DCL), but these plans were shelved in favor of continuing with the InC arrangement under CDG. Evidently, the question of creating an autonomous financial agency had Important political repercussions. CDG would itself have been a major player In any such rearrangement. A lesson of this experience Is that, through familiarity with key project agencies, the Bank can review the political pros and cone of such measures to all institutions involved, including CDG. A lot was at stake, since a fully fledged local commune development bank could have become one of the largest financial Institutions in Morocco. D. Role of Subproject ArAigal in _EC _rnJect easion Makina 4.15 The Audit was Impressed by the quality of subproject appraisal methods developed in-house by the FEC team. The techni.al, economic, and financial aspects of all subprojects In all sectors were thoroughly reviewed by FEC staff through the application of evaluation models that used Lotus 123 spreadsheets on microcomputers. The appraisal techniques were developed in-house by FEC staff. The project evaluation and eligibility criteria detailed in the 28 pages of annexes to Side Letter One to the legal agreements proved to be a valuable input for the development of these techniques (para. 3.40). The experience demonstrated that detailed legal agreements can contribute to the technical progress within a client institution. 4.16 What is not clear, however, wae the extent to which these well conceived subproject appraisal techniques were actually used to guide decisions to approve proposals for FEC finance. This was, In part, because the processing of local communesO applications for FEC finance followed a long path, involving Important instances of decision-making outside FBC Itself. A brief review of the process will help set n8C's subproject appraisal work into the broader context. 4.17 Local communes first present their proposals for Investments to be financed by FEC to their awn provincial governors (para. 1.13). The governors themselves evaluate the proposals to determine which are consistent with provincial priorities and regional development plans. Those found to be Inconsistent are not processed further and are not seen by FEC. Precisely how many proposals are turned down at this stage Is nou known to the Audit. Proposals approved by the governors are sent to the Minsstry of Interior's Local Authorities' Department (DOCL - Direction Oft6rale des Collectivit6s Locales) for evaluation. 4.18 The Audit was informed that some 25 percent of the proposals received by DOCLIMDI are turned down at this stage because ofs (i) Incompatibility with local development planes (i) Insufficient debt service capacityl or (iII) 42 infringement of legal or regulatory requirements. The remaining 75 percent finally find their way to M80 via CDC. Only after receiving the proposals from DGCLIMOI, does FMC begin its evaluation. Despite the previous checks, the audit mission was informed that MBC still has to request additional Information for some 60 percent of the proposals, whose dossiers are considered incomplete. FEC informed the audit mission that it generally finds that 90 percent of the proposals are economically and financially justifted, yielding Internal rates of return of at least 12 percent. What the Audit was surprised to learn, though, was that even proposals for subprojects not justified by the appraisal are still sent to the CDG/FEC Technical Committee, which generally approves all the proposals submitted to it. That being the case, the result of the subproject appraisal does not affect the likelihood of the proposal being approved. 4.19 From the process just described, an excellent subproject appraisal technique developed by FEC, is perhaps not being fully used as a management tool to help decision-making. This is for two reasonst (1) the local communes' original proposals are reviewed twice before being appraised by nC, which means that FEC does not evaluate all proposals and some well-justified projects may not even reach the agency if they do not mast other criteria applied by provincial governors and DGCL/MOI and (it) the result of satisfactory subproject appraisal by FEC was insufficient to guarantee financing. 1. Inadeauate_ Monitori -ni Tes of ZnJect gltiea 4.20 Both operations reviewed here aimed at urban sector Improvements, specifically by stimulating the provision of low-cost housing and assisting local authorities to meet their urban infrastructure needs. In the course of implementing these projects, however, executing agencies and Bank missions made only sporadic attempts to determine what their quantitative achievements were in terms of these goals. One of the reasons given for inadequate monitoring was that both local management and staff and Bank missions had tight supervision schedules, overloaded with taske concerning day-to-day project administration. In this respect, successful project monitoring requires that the Bank allocate adequate resources for project supervision. 4.21 Bad emphasis been given at appraisal to defining performance targets in terms of low-cost housing supply and arrangements for their monitoring the CI Project might have met its original low-income beneficiary objectives. With the support of Bank missions, CIE ight have monitored, among other thingsa (i) the annual delivery of low-cost housing as a proportion of the total (CIR efficiently compiles this kind of Information on its own housing loans in its annual reports already) in relation to an agreed targets (1i) the number of private developers engaged in the delivery of low-cost housing and the type of problems they face, with recommendations about the kind of technical assistance they need to help them increase outputs and (Lit) the results of CIE's appraisals, highlighting the kinds of problems identified with low-cost housing projects and suggesting design and other modifications to help overcome them. 4.22 Insufficient supervision and monitoring of the FEC Project was discussed earlier in this report (pars. 3.39 and 3.46) and mentioned by the PCR (FEC PCR para. 9.04). CDG/FEC management are aware of this problem and are 43 planning to strengthen this capability of the TIC operation. The annual monitoring of the following might have been useefuls () Infrastructure deficit, declared need., and actual delivery for each local comunes (it) the actual performance of the infrastructure and services financed by FC loans compared with the projected performance at appraisals (111) local commune' project appraisal capacity and technical assistance needs to strengthen its and (v) Investments financed by nC during the year in terse of region, city size, type of local authority, and beneficiary Income levels. Through its now extensive lending operations, n8C has accumulated a rich body of knowledge about local development in Morocco. Sector dialogue within the country itself and in the Bank, would benefit considerably from a fuller and more widely disseminated understanding of this experience. 4.23 Monitoring Indicators need to be discussed and performance targets agreed at appraisal end, where they refer to the DFC itself, could be incorporated as covenants to the legal agreements. Converting such indicators Into effective management tools can be one way of reducing the risk of such a project losing sight of important objectives. P. RritrnU&m l2oct 4.24 The lack of monitoring makes it difficult to draw conclusions about the environmental impact of the operations under review. Nevertheless, the kind of lavestments made through the CIN and FEC Projects are likely to have had positive impacts on the urban environments in the cities in which they were made. Residents of housing schemes financed by CIH, independently of their income levels, should now be enjoying mtnu environmental health conditions derived from the provision of a safe water supply and through connections to sewerage and stom drainage systems. Although mostly felt b the occupants of the housing schemes as the project's direct beneficiaries, these subprojects should also contribute to an overall Improvement in enviroamental health in the cities in which they are located. It is also reasonable to expect that the majority of the local commune and #regie* investments financed by FEC, such as those for street paving and drainage, water supply and sewerage, and solid waste collection and disposal, will have resulted in environmental Improvements. 44 V. = =QuANDI= nh.M IGWC A. Experiences of Borrwer and Executin Agencies 1. MIf 5.01 Because of CIR's long experience in the sector and with the Implementation of Dank operations, procedures for project identification, appraisal, and administration of subloans were already in place. It was precisely this tn-house ability that made CII an attractive borrower for the Bank. CT's day-to-day management of the project vae indeed efficient, especially with respect to administration of the housing loan portfolio. The excellent quality of CIB's Annual Reports and the housing sector analyses they contain confirms the high level of professional skills available to CI8. CIR/Bank relations were reportedly good, notwithstanding CIN's perception of the insignificance of the loan amount and "borrower fatigue" discussed earlier (details: para. 4.08-4.11). A major disappointment was CIE's unwillingness or inability to pursue the project* low-cost housing objectives. This points to the critical Importance of borrower "ownership* of project goals for their eventual attairment. 2. CDIE 5.02 CDG/FC satisfactorily fulfilled their obligations during project implementation. FEC provided information requested by supervision missions and submitted periodic progress reports to the Bank. Although, like CIR, management and staff in CDG and PSC considered the Bank loan amount to be too small, this did not have as demoralizing an effect on relations with the Bank, since the hope of a large follow-on loan was an Important positive stimulus. FEC was able to perfect its subproject appraisal technique. and apply them to all Investments made, not only those financed through the Bank loan. In spite of occasional setbacks, cooperation between CDG/FEC and the Bank was good. PEC was supportive of the audit mission and provided all the information requested by it. B. Role of the- ank 5.03 The CIZ and the RIC Projects were similar insofar as they each represented the Bank's first Incursion through line of credit operations in their respective sectors. The Bank's role, though, was distinct in each case, especially during preparation. The region's urban projects division responsible for the CIN Project was breaking ne ground in dealing with the financial aspects of housing. In contrast, the IDF division that prepared the FEC Project was familiar with the financial aspects of this type of operation, but had to deal with a different kind of borrower and sub-borrowers than it was used to in industrial finance projects. During supervision, when the urban projects division assumed responsibility for both operations, the Bank's role in both became similar. Now the Bank broke the new ground in each case can provide an object lesson for future operations of this type. 45 1. IR Protect 5.04 A housing finance line of credit was seen by those responeible as a radical departure from earlier physical development operations supported by the Bank in the sector. For the Bank, it was a potentially creative moment when staff with financial analysis skills were deployed for the first time in the urban sector, and staff specialized in physical planning and engineering had to acquire new financial skills. 5.05 Despite its being an innovative operation, and one in which the Bank had no previous experience, the project was prepared and appraised quickly (para. 2.18). Inconsistencies found in appraisal data on project costs (para. 2.26), as well as the start-up problems encountered (para. 3.01), point to lapses in preparation. A better preparation might have allowed the Bank either to design a feasible low-cost housing finance operation in relation to policy and market parameters in Morocco or recognize that the low-cost housing approach it espoused was incompatible with those parameters. Not least of all, the Bank should have taken better note of its previous experience in the urban sector in Morocco, and in particular the previous setback to its strategy of bidonville upgrading which OW opposed (paras. 1.16-1.19). 5.06 According to the CIS PCR, one of the advantages of the wholesaling of Bank lending through a line of credit operation was that Bank staff did not have to "scrutinize minute details," thus permitting them to concentrate on "broad institutional and sectoral Issues* (CIE PCR Evaluation Summary para. 9). This Implied a lighter Bank supervision load, but in actual practice, the CIA Project's supervision effort, measured as Bank staff weeks per million dollars of loan, was approximately twice that of the Second Urban Development Project (Ln 1944-MOR), although considerably less than the Rabat Urban Project (Lu 1528-MOR) (para. 2.02). Bank supervision missions of the CIA Project still had to occupy themselves with many details, although of a different kind from those encountered in the earlier physical urban development projects. 5.07 Much was learned from this experience and applied to later operations elsewhere. In spite of the disappointing results obtained, the project represented a worthwhile effort to set a new direction for Bank urban lending in Morocco which, at the time of its appraisal, risked becoming paralyzed through the divergent approaches of GCM and the Bank to the issue of Morocco's bidonvilles. 2. FEC Polest. 5.08 During preparation and appraisal, the region's ID division brought its experience with line of credit financing to the urban sector. IDP's lack of familiarity with urban sector institutions, however, raised concern about the possible lack of credit worthiness of local communes and dictated a cautious pace to project preparation. Such caution was justified in view of the uncertain institutional status of FEC itself in relation to CDG (paras. 2.13-2.16 and 2.20- 2.21). Project preparation was protracted and the Bank deployed twice the staff resources as for preparation of the much larger CI Project, yet the lack of attention given to CDG In the appraisal process was a major emission by the Bank. More preparation time, however, gave the Bank the opportunity to help introduce 46 some useful management techniques and procedures into EC operations, especially in connection with subproject appraisal. 5.09 Bank supervision began well, but clearly became less intensive as the prospects for a larger follow-on loan diminished. The Bank continued discussions with O about establishing a local comme development Bank, even to the point of fielding a pre-appraisal mission in 1987. The level of staff effort for supervision was greater than that of the CIS Projects and supervision missions had to concern themselves with many details of the operation, such as the conditions of employment of a technical advisor and encouraging nC to make prompt payments to consultants. The effectiveness of staff contribution to the project diminished as hopes for a follow-on loan receded. V1. COWCMUION AND LESSONS LEAR=E 6.01 This section brings together the conclusions and most Important lessons for future efforts from the rich experience afforded by the preparation, appraisal, and Implementation of the CIE and nEC Projects. 1. CIS Proleat 6.02 The CIN Project accounted for a significant share of all housing units financed during the 1983-90 period. The Project is rated as satisfactory, although relative to the optimistic expectations, some results were disappointing. It did not lead to a fast-disburitg loan, nor to the institutional and sectoral Impact that the Bank bad hoped for. Perhaps the expected results of 'wholesalings Bank lending through a DEC In a complex sector such as housing were unduly ambitious. In actual fact, disbursement proved difficult. Concrete project Impacts on CIE as an institution and the housing sector in Morocco are difficult to see. In torm. of its own declared objectives, the CIR Project did finance units that are affordable by the lower median-income population, but it failed to stimulate the delivery of low-cost housing by public and private developers." Cls* long experience in the sector and its established in-house ability to effectively deal with project processing and administration of subloans made it, however, an attractive borrower for the Bank. The Project9s institutional objectives are rated as partial (vs. substantial on the basis of the PCR). CH**s capital structure remains solid, although its profitability has declined. Experience in Morocco reveals that the maintenance of housing units is assured by resident-owners. The Project*s sustainability is rated as likely (vs. uncertain on the basis of the PCR). The key lessons to be drawn from the experience Include the followings 51 According to CIE comments on the PAR in Annex Is this was in spite of CI's disposition to see the project well executed and also its own efforts in a broad campaign to promote this kind of housing among developers. 47 (1) It to important to take Into account the lessons of earlier experience In the sector even if the proposed project design to intended to substitute earlier approaches. Particularly relevant in this case to the protracted 00M/8ank policy housing sector dialogue that occurred during Implementation of the first two urban projects which clearly showed 00*s preference for high standard housing solutions. Unlike Bank strategy for urban development, 00 sector policy did not have bidonvilles and low-cost housing as its centerpiece; (tt) When it focuses its support on those operations in the DEC whtch are not a priority for the Government and the Intermediary itself, Bank funding risks merely filling a resource gap left by the local withdrawal from low priority operations, as clearly appears to have been the case with low-cost housing in the CIR Projectl (tit) The resources provided via a line of credit to a DFC are fungible within the Institution. Consequently, a line of credit project must be understood as one that supports the DEC and all its operations. For this reason, institutional appraisal of a DFC should extend to all its activities. Where the performance of operations other than those targeted (i.eq hotel loans in the case of CIR) can affect the results of the project, agreed performance standards for them should be covenanted In a project's legal documents; (IV) The Bank policy Influence turned out to depend on the importance of the loan amount in relation to new funding likely to be mobilised by CIB. The Bank loan to CIA was relatively small and came to be regarded as insignificant by the borrowerl and (v) As lender, the Bank should evaluate all aspects of DFC. financial performance, Including profitability and risk management in the case of CIE. However, satisfactory financial performance is not a sufficient condition to assure project success. For this, the fulfillment of project objectives, namely stimulating the delivery of low-cost housing in the present case, also has to be achieved. A monitoring system should be in place to measure progress in these areas as well. 2. IEC Project 6.03 The PBC Project was a pilot operation and raised modest expectations with respect to its likely results. Its achievements were mixed. Physical success and partial, but Important institutional development were counterbalanced by poor supervision and manitoring of sub-projects and declining financial performance. TMC remained part of CDC and made negligible progress toward full 48 autonomy which appeared to discourage Iamediate further support by the Bank. Subsequent progress was evidently adequate to justify the recent approval of the First Muicipal Finance project (approved on June 10, 1993). The Audit rates the overall outcome of the project as marginally satisfactory, and its sustainability as uncertain. The PCR-based ratings weres overall performance as satisfactory; institutional development as substantiall and sustainability as likely. The key lessons of the experience of the P8C Project Includes (i) To achieve the desired Institutional and policy Impact, it is important for the Bank to engage the appropriate authority in Its policy dialogue. In this case, CDG was a key policy interlocutor and its active engagement In the process should have been enlisteds (i) for a first operation in a particular sector, it is well worth spelling out in detail the agreed criteria for eligible beneficiaries and subprojects. In the case of the FEC Project, these details enabled RKC staff to develop a sound appraisal method for evaluating the proposals received. It is likewise Important to engage all Important policy agents in a complex subsector such as local development, in this case including VO and the provincial governor., so a. to ensure their support for the principles and procedures of the operations and (ti) As to the case of CIN, the Banke Influence turned out to depend on Importance of the Bank loan in relation to new funding likely to be mobilized by the DIC. After the expected follow-on Bank loan got delayed, Bank influence diminished as CDGIFC came to regard the FEC Project loan amount as Insignificant In relation to the sector's needs. 49 Annam1L Commente from the Borroer and Executina Aaenget (Treolatiom from rnch orig8al) Cateo de Dpot et de Gestion P.O. Box 408 Rabat morocco ApTil 15, 1993 Enersy and Infrastruture DiviDoa Operations avaluation Department World ank Wasington, D.C. For the attention of Mr. Yves Albt Sir With your letter of March 26, 1993 you kindly sent Us a Performance Audit Report on the pilot projet for the Communal Infrastru=ture lund - lEC (Loan 2272-M). Raving namied this report, we have the honor to send you enclosed a memorandum containins a muber of cafente and olarificatione prompted by the conluLonm of this report, whic we would like to see reflected in the final version. We hope you roceive it i order and rmia, ey tmuly your., utidel Lahio Direotor General so Annex 1 Page 2 of 13 Caieae do D6p8t et do Gestion P.O. Box 408 Rabat Morocco April 15, 1993 PEC Project (2272-MOR Performance Audit Revort The use made by 18C of the World Bank loan for US$16 million, signed on March 23, 1983, is the subject of a Performance Audit Report prepared by the Bank staff. The report provides an assessment of the overall performance of the pilot project for 7EC, its institutional Impact, its economic repercussions and the financial performance of EC itself. Examination of this report by the Casse de D6p6t do Gestion (CDC) calls for the following comments and observationsa 1. General context and ietitutional ramework The Performance Audit Report focuses particularly on the degree of the involvement of CDG in project Implementation. On this subject, we need to explain the legal nature of the linkages between CDG and the PED and its Implications. In accordance with its bylaws and those of n1C, prior to its reorganisation which came into effect from the beginning of 1993, CDG was placed in charge of the administrative and financial management of 1EC, as a public establishment with legal status and financial autonomy. The negotiations for the IBRD loan were conducted with the Moroccan Government and of course with CDG through FEC itself. Project Implementation complied fully with the terms of the pertinent loan agreement. It was naturally also carried out under the responsibility of CDG. We atill do not agree, however, with the idea raised in the PAR that PEC's autonomy was limited by CDC. On the contrary, as an establishment with total autonomy, CDG allowed this agency to commit the contracted funds on the best termo. The assessment to the effect that EC's performance would not be that of an autonomous financial agency should also be qualified because, 51 Anne I Pa&e 3 of 13 other than this normative aspect, its involvement and its financial structure are, given its institutional structure, as described earlier, indisting- utabable from those of CDG. In a nutshell, any analysis must be directed in this spirit of organic and operating osmosis between the two organisations. 2. Importance of CDG S =onort As soon as PEC was created, it was to receive technical and financial support from CDC. This aspect should be underlined in the PAR. Until 1986 PEC vas funded primarily in the form of the discounting of bills of exchange by CDC. This type of financing accounted for three-quarters of FEC's total resources in 1986. From November 1, 1986 onwards the discounting of FEC bills of exchange vas no longer eligible for rediscounting by the Bank Al-Maghrib. This change, together with the desire to proceed with a reorganisation of 0EC In order to bring it into tune with the growing financial accountability of local authorities, was to mark the beginning of a consequent consolidation of this agency's resourcest it had refocused the mobilisation of its borrowed funds onto the domestic bond market, one of the leading organisers of which was none other than CDC. Ever since, the bulk of the funds has been contributed by CDC and its services (CNRA and RCAR) via the bond market. In 1989 and 1990 F&C's loan holding status was as followst 1989 1990 CDG subscription as percentage of total issued 191 322 CDG subscriptions and managed services (CHRA-RCAR) 411 72Z In this connection, as vas pointed out In the PAR, IBBD*s assistance is limatd. In terms of outstanding amounts, the IBRD loan accounted for 3.41 of FOC total asets in 1986 and 4.72 In 1990. Most of its funds come from the domestic financial market, In particular the CDG group. Disbursements from the IBRD line of credit contributed only 6.1Z of FEC's total long-term borrowings. CDG's involvement has thus always translated into considerable financial assistance to F8C and has never been an obstacle to its autonomy. Quite to the contrary, CDG constantly encouraged this establishment to search for new sources of financing, while remaining ready to come to its rescue, if necessary. 52 Ane 1 hae et14 3. E2's..2eLtemMnce At the technical level In the early eighttes, RC bad a very small staff. A sustained recruitment drive vas made, however, to order to keep up with the expansion of its activities. Between 1975 and 1960 only 7 people were hired. This =umber vas increased by 19 between 1980 and 1985 and by 14 between 1986 and 1991. It now has a staff of 44. This nmber, which to considered insufficient to the PAR, has nevertheless enabled 180 to develop a proportionately more extensive activity, in terms of both loan, granted and loans disbursed. Compared to other financial agencies, 180 eanages, with a much lower mpersounel coats/volume of business" ratio, to achieve a remarkable performance. The expertise and professional competence of its staff has risen, thanks to this reorganisation. The PAR, in drawntg attention to this aspect, indicates that the audit aission was tipressed by the competence with which projects are appraised by 13C. As for the evaluation and supervision of the project., once they have received financing, this remains rather complex but could be handled with additional necessary technical and human resources. Moreover, though there might be some weakness at the project supervision level, attention needs to be drawn to the fact that the projects were maost often carried out In response to the urgent needs of the c*ome* Interested and that their finsancig, which was backed by the supervisory Ministry and its local authorities, originated more from State assistance to local development than from the sole interest of banks in the profitability of the projects. At the financial level The PAR refers to the choages that have taken place In the financial structure of 13C aisie 1983. These changes have, in several respects, been favorable and beneficial. There are several Indicators of such an effects * A steady IMprovement in total assetes D 561 million in 1983, DR 1,280 milon In 1987 and 2,603 allion In 1991. * A consolidation of resoucess reflected in a reduction In the short-term liabilitleatotal assets ratio from 722 in 1983 to 252 in 1988 and 122 In 1991. In correlation, thanks to the domestic financial market, log-term resources have developed rapidlys the medium-and long-term debtitotal ssets ratio, which was only 1O 53 in 1985 quickly climbed to 661 In 1987 and had reached 832 by the end of 1991. * The project*' return on equity, measured as net profit./equity, did, admittedly, show a significant deterioration, this ratio falling from 6.6% in 1983 to 0.61 in 1987. Losses even had to be reported in the following two years. from 1990 onwards however, this ratio began to recover, showing 0.3% in 1990 and 2.66Z in 1991, even allowing for large provisions for losses. Beyond these developments, this Indicator is still not sufficient by itself to provide an Indication of the financial viability of FIC. Furthermore, any setback or even decline in the results is attributable primarily to an increase In financial costs generated by the use of more costly foreign resources. * The only unfavorable element in the financial activity of PEC undoubtedly still lies in the large volume of arrears payable by the Rfgies de Transport (transportation sector utilities), which totaled about DR 60 million in 1991, and for which reserves were established in accordance with current legislation. Overall, however, FEC's arrears now represent only 3.21 of total loans outstanding. Generally speaking, this problem is now being settledl the arrears should be paid by the local governments to which the pertinent Zwies are linked. What was stated In the PAR should be qualified by the fact that PEC contined, throughout this period, to receive the absolute support of CDG. * Some Important measures were taken in order to Improve the status of FEC, including increases in the rates charged to local governments and rASius. These increases were introduced, In particular following the elimination of FEC's access to the .-entral bank for the rediscounting of its bills of exchange. Thus, In addition to the 101 rate, which came into effect in 1986, five other rates were decided on on April 7, 1989 and were adjusted on July 12, 1990. What is more, it would be wrong not to take account of IBC's special role as a local development fisancing agency, expected, among other things, to provide its assistance to rural communes and autonomous centers, and local governments, most of which have Insufficient resources at their disposal. The rate for rural communes was therefore kept at 102, i.e. less than the cost of external resources, which was 10.601 in 1989 and 10.93Z In 1990. 54 Annez Page 6 of 13 4. Socioeconomic Indast As highlighted in the Project Completion Report and referred to In the PAR, IBRD's US$16 million lne of the credit was used to finance 90 projects. The number of projects submitted to the Bank for approval under the loan was actually 109, 90 of which were approved for a total amount of subloans of DR 134 million, representing 422 of the total cost covered by FOC assistance. The 19 projects not approved by the Bank were eventually financed by 1EC for a total of approsiately DR 226 million. IBD Loan 2272-MOR has been fully disbursed sace February 8, 1990. The sectoral breakdown of the projects financed under this loan should be identified in the report. Two key elements need to be stresseds * 83 projects correspond to infrastructure for local governments. The IBRD loan proceeds allocated to these projects totaled DR 107 million, i.e. 442 of the loan in question. * The sectoral breakdown of the assistance Is extremely diversified. In terms of the projects approved, three sectors account for mst, with 67 projects. These sectors are commercial Infrastructure, sanitation and water supply. In terms of financing, there Is greater dispersion, i.e. 282 for commercial infrastructure, 222 for sanitation, 121 for urban transport and 142 for water supply. In terms of the total financing provided by FEC, these four sectors received 252, 212, 171 and 142, respectively. It goes without saying that the projects financed will definitely have a socioeconomic Impact, since they contribute to the creation and upgrading of infrastructure, the primary benefiaaries of which are ultaimately the low-income and middle-Income segments of the population. 55 Annex 1 Page 7 ot..13 Service du Credit lamobilier et Industriel Division du Cr6dit et des Investisements Direction du TrEsor et des Finances BatArieures Ministry of finance Xingdom of Morocco GAIZL/A167 The World Bank Washington, D.C. For the attention of Mr. Yves Albouva Inrastructure and RnerU Diviston Gentlement Further to the Performance Audit Report you kindly sent me, I have the honor to provide the following coamments a 1* The report notes the slow pace and disbursement difficulties that characterized both the CIN and P8C projects. One of the factors responsible for this problem, which should be mentioned in the report, is the mismatch noted between the criteria established by the Project and the programs adopted by the two institutions involved and considered by them as well suited to the framework of the Project. This is particularly true in the case of the CIO loan, mobilization of which necessitated the relaation of various aspects of the lending conditions to bring them into line with the requirements of CIR's programs. 2. The report also indicates that the CIN Project did not totally benefit the target group. This situation, the causes of which were discussed by CII in the PCR, calls for two commentsa - The project's objective, i.e. to Iaprove the supply of low*income housing, cannot be achieved simply by easing financial constraints. - The report concludes that it is necessary for the Bank's projects to take an overall approach that encompassee all aspects of the activities performed by the financial Institution serving as intermediary for project Implementation. One of the lessons to be learned is that it to also necessary for the Project to have 56 Pane 8.of 13 a broad view of all aspects and all dimensione of what Is being financed. The CII project reveals, In this respect, that there were sm Iportant Issues concerning the supply of low-income housing that had not been euamined and for which this project failed to provide the necessary solutions. What the housing sector needs from the World Bank is financial assistance as well as assistance with Institutional and legal reforms, through a project that to sufficiently broad based to deal with all the constraints of the sector (financing instruments, land tenure Issue, etc.). Faithfully yours, lt Mohamed DAIRM Directeur du Tr6sor at des Finances Ext6rieures 57 CauDIT UmoIlttr UT UorM1Mn (ea) Directorate of Flnanc Canablanca, May 11, 1993 IBXD 1818 a Street, NW Washington, DC 20433 Attentions Infratrutur and 2nergy Divislon, Operationa gvaluation Department Dear Sira, i am pleased to send you herevith the Domments of the CIZ on the Performance Audit Report on the fixst urban development project (oan 2245- (signed) TA Abderrahim PinancUal Director, cM 56 Annex 1 Raa 10 o 13 Comments on the Performance Audit Report on 1BRD Loan 2245-MOR As part of its evaluation of projects financed by ZROD, the Operations Evaluation Department has prepared an Audit Report on the First Urban Development Project in Morocco (Loan 2245-MOR made to the CIN in 1983). The value of this report is evident insofar as it identifies the main lines of the project and highlights its shortcomings, with the aim of adjusting the Bank*s operations and maximizing the induced effects of the loan. We would note, however, that the report sent to the CIR gives rise to two series of comments, the first pertaining to project execution and the second to the financial performance of the CIN. A - PROJECT EXCUTION It should be recalled that the purpose of this line of credit was to encourage private and public developers to produce low-cost housing, affordable to low-income groups, and to develop the CIue ability to appraise housing projects and advise developers on the execution of this kind of housing. To be eligible for financing, the subprojects had to meet the following criterias - The price of land was not to exceed DR 180 per m2l - The average area of the dwellings was not to exceed 100 a2; - Core units were priced at DR 32,000# - One to two floor partially completed houses were priced at DX 48,000; - One to two floor completed houses were priced at DR 62,000; - Three to four room units in multi-story buildings were priced at DR 80,000. These parameters were found to be inoperative, as the core units, even if they could theoretically be produced in line with the above-mentioned criteria, had no appeal for Moroccan households, which led to a total lack of interest by developers in this kind of housing. 59 AnneRA Raa1nita Purthermore, and contrary to what to stated in the report, the Cia, despite its desire to see the project succeed and the extensive sensitisation campaign undertaken vis-A-vis private and public developers, came up against the problem of the non-availability of housing units satisfying these criteria. This situation was due in particular to the quickly rising cost of land and construction materials the report notes the 41 percent rise in urban land in Casablanoa and the 20 percent increase in construction materials in the period 1987-1990 alone. As a result, the CMa was unable to aske any withdrawals from the loan in 1983, and two years later, i.e. at the end of the third year, only D 32 million, or 6.25 percent of the total amount of the loan, had been disbursed. Consequently, CIN had to pay a substantial commitment charge on the undisbursed balance, amounting to DR 10,061,276, equivalent to US $2,016,057, or 3.36 percent of the total amount of the loan. The negotiations between the Bank and CIR led to a series of changes in the eligibility criteria in order to align them with market realities and take account of price increases; accordingly the upper limit for refinancable housing was raised as follows* Year Amount in DR 1983 80,000 1984 95,000 1985 115,000 1986 130,000 1987 140,000 1988 150,000 It must not be forgotten that this credit line was phased over a relatively long period. Although it was signed on April 11, 1983, the final disbursement did not occur until Jan. 29, 1990, following three successive extensions of the closing date. There was no alternative to revising the eligibility criteria, since if this had not been done the loan would not have been disbursed. This is confirmed by the refinancing criteria adopted for the second Urban Development Project, which includes two kinds of VIT, with ceilings of DR 150,000 and DR 300,000 respectively. 60 Annes. Pae 12 of 13 B - FINANCIAL ROA O H CIR The Performance Audit Report reters to the deteriorating financial performance of the CIN between 1983 and 1990. This observation is essentially based on the behavior of the net profit margin, which, however, cannot be used as a benchmark. The analysis of a series of management, profitability and structural indicators, as set out in the Appraisal Report prepared by the Bank in 1988 for the Second Rousing Finance Project (Loan 3122-MOR), and in the Project Completion Report on Loan 2245-MOR, indicates that in fact the financial performance of the CIS improved over the period 1983-87, an improvement that was reinforced in the years following the conversion of the CIR into a deposit bank. In any event, the downturn ir. the net profit margin indicated in the Bank's report, from 3.61 percent in 1983 to 2.57 percent in 1987, i of no significance, since this indicator is based on the net result. However, the not result includes a significant amount of provisions set aside each year as a prudent measure to offset arrears, although the risks involved are in fact minimal, in light of the guarantees given. Consequently, it that part of the provisions set aside for bad debts not involving any risk is restored, net profitability in fact continues to improve from one year to the next. This is confirmed by the perfoaance of the gross profit margin, which has behaved very favorably, as noted in the above- mentioned reports. Contrary to the assertion in the report regarding the low level of provisions against possible loses, the following table indicates a certain stability, since the ratio of provisions to total lending plus arrears has not worsened over the years. Year 1983 1985 1987 1990 1992 Provisions 116 191 225 327 489 Loans + Arrears 3,655 4,554 ,548 9,787 15,201 Provisions/Loans 3.17 4.19 3.43 3.34 3.21 As regards the debtsequity ratio, we would point out that this indicator, which did indeed rise from 10.35 to 13.97 between 1983 and 1987, has never reached its maximum permitted level of 15 percent. It should also be noted that two successive capital increases in 1988 and 1989, which provided DR 500 million of new money, were made for the purpose of increasing the CiR's lending capacity. As a result the debtaeguity ratio steadily improved to reach the very satisfactory figure of 11.32 in 1991 and then 10.13 in 1992. 61 Pae13 of 13 Pinally, we would point out that, as regards deposits, the share of this category of resources continues to increase in relation to other sources of funds. indeed, as a result of the importance accorded by CIS management to deposits, their share in total resources has increased over the last five years from 2.5 percent in 1988 to 7.5 percent in 1989, 9 percent in 1990, 12 percent in 1991 and 16 percent at the end of PT1992. A more comprehensive view of the CIa's situation, as measured by a number of important indicators, reveals a continuing improvement of which the following significant results are evidences - The constantly expanding level of activity, generating an ever- increasing cash flow over the years (DR 111 million in 1983, D 178 million in 1988 and DR 504 million in 1992); - The good performance of gross profits, particularly from 1988 onwards, in which year the CIR initiated its deposit operations this margin reached 3.65 percent at the end of 19928 - The maintenance of the debtsequity ratio at very reasonable levels - The very low level of administrative expenditure (below the established norm); - Satisfactory liquidity and debt-service coverage ratios; - The provision of real estate guarantees for all loans made by the Cie in addition to the State guarantee for all hotel loans. As regards the mismatch which indeed exists between the due date for repayment of Loan 2245-MOR and the maturity of the subloans to which its proceeds were applied, the CII will rely on the Bank to help it find a solution that matches the reimbursement of the Bank's loan to the repayments to be made by beneficiaries of the subloans in question. It should be noted that this mismatch involves a difference of 11 years, since the total maturity of developer-plu-mortgage loans on the one hand is 28 years and the rembursement period of the IBRD loan is 17 years.