Document of The World Bank FOR OFFICIAL USE ONLY Report No: 21538-MOR IMPLEMENTATION COMPLETION REPORT (CPL-36180; SCL-3618A; SCPM-3618S; CPL-36190; CPL-36200; CPL-36210) ON FOUR LOANS 1N THE AMOUNT OF US$130 MILLION TO THE KINGDOM OF MOROCCO, CREDIT IMMOBILIER ET H6TELIER, WAFABANK and SOCIETE GENERALE MAROCAINE DE BANQUE FOR A LAND DEVELOPMENT PROJECT FOR LOW-INCOME FAMILIES DECEMBER 28, 2000 Infrasturucture Develoment Group Middle East and North Africa Region This document has a restricted distribution and may be used by recipients only in the perfonmance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. CURRENCY EQUIVALENTS (Exchange Rate Effective ) Currency Unit = Moroccan Dirham (DH) DH 1 = US$ 0.12 at appraisal (January 1, 1993) 0.10 at closing date (June 30, 2000) US$ 1 = 8.50 DH at appraisal (January 1, 1993) 9.98 DH at closing date (June 30, 2000) FISCAL YEAR January 1 to December 31, up to FY96 January 1 to June 30, 1997 July 1 to June 30, FY 97/98 to FY99/00 ABBREVIATIONS AND ACRONYMS ANHE Agence'Nationale de Lutte contre l'Habitat Insalubre BMCE Banque Marocaine de Commerce Exterieur CIH Credit Inmmobilier et Hotelier ERAC Etablissements Regionaux d'Amenagement et de Construction FNAET Fonds National d'Achat et d'Equipement des Terrains GoM Government of Morocco SNEC Societe Nationale d'Equipement et de Construction ICB International Competitive Bidding LCB Local Competitive Bidding MoF Ministry of Finance MoI Ministry of Interior SGMB Societ6 Generale Marocaine de Banque TVA Taxe sur la Valuer Ajout6e VIT Valeur Immobiliere Totale Vice President: Jean-Louis Sarbib Country Director: Christian Delvoie Sector Manager & Acting Director: Sonia Hammam Task Team Leader/Primary Author: Jean-Hubert Moulignat/Ekaterina Massey FOR OFFICIAL USE ONLY CONTENTS Page No. 1. Project Data 1 2. Principal Performance Ratings 1 3. Assessment of Development Objective and Design, and of Quality at Entry 2 4. Achievement of Objective and Outputs 6 5. Major Factors Affecting Implementation and Outcome 17 6. Sustainability 18 7. Bank and Borrower Performance 20 8. Lessons Learned 23 9. Partner Comments 24 10. Additional Information 24 Annex 1. Key Perfornance Indicators/Log Frame Matrix 25 Annex 2. Project Costs and Financing 26 Annex 3. Economic Costs and Benefits 28 Annex 4. Bank Inputs 29 Annex 5. Ratings for Achievement of Objectives/Outputs of Components 31 Annex 6. Ratings of Bank and Borrower Performance 32 Annex 7. List of Supporting Documents 33 This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not be otherwise disclosed without World Bank authorization. Project ID: P005514 Pro]ect Name: LAND DEVELOPMENT Team Leader: Jean-Hubert Moulignat TL Unit: MNSID ICR Type. Core ICR Report Date: December 28, 2000 1. Project Data Name: LAND DEVELOPMENT L/C/TF Number: CPL-36180; SCL-3618A; SCPM-3618S; CPL-36190; CPL-36200; CPL-36210 Country/Department: MOROCCO Region: Middle East and North Africa Region Sector/subsector: UH - Urban Housing KEY DATES Original Revised/Actual PCD: 03/02/92 Effective: 09/21/93 12/13/93 Appraisal: 05/21/93 MTR: 05/13/98 05/13/98 Approval: 06/10/93 Closing: 06/30/2000 06/30/2000 Borrower/lImplementing Agency: Gov. of. Morocco, CIH/SNEC; Gov. of. Morocco, CIHICIH; Gov. of. Morocco, ClH/Ministry of Housing Other Partners: STAFF Current At Appraisal Vice President: Jean-Louis Sarbib Caio Koch-Weser Country Manager: Christian Delvoie Harinder Kohli Sector Manager: Sonia Hammamn Amir Al-Khafaji Team Leader at ICR: Jean-Hubert Moulignat Henri Beenhakker ICR Primary Author: Ekaterina Massey; Jean-Hubert Moulignat 2. Principal Performance Ratings (HS=Highly Satisfactory, S=Satisfactory, U=Unsatisfactory, HL=ighly Likely, L=Likely, UN=Unlikely, HUN=Highly Unlikely, HU=Highly Unsatisfactory, H=High, SU=Substantial, M=Modest, N=Negligible) Outcome: S Sustainability: L Institutional Development Impact: SU Bank Performance: U Borrower Performance: S QAG (if available) ICR Quality at Entry: U U Project at Risk at Any Time: Yes 3. Assessment of Development Objective and Design, and of Quality at Entry 3.1 Original Objective: The project had four main objectives: (a) provide serviced residential lots at affordable prices specifically targeted to squatter and low-income families; (b) provide medium-term financing to private sector property developers for construction of housing for low- income families; (c) foster the participation of private sector banks in the housing finance sector; and (d) improve the housing finance sub-sector, housing subsidy policies, and the land delivery process. At the time of the project appraisal those objectives were clear and responded to pressing borrower circumstances and development priorities, and supported CAS goals of enhancing the efficiency of public sector management, and alleviating poverty and improving social indicators through improved coverage and quality of essential services. However, the third objective, while in line with financial sector objectives, proved to be overly ambitious given housing real sector and policy constraints. The project was conceived in response to an emergency request (originally US$100 million) of the Government of Morocco to assist in the implementation of the 1988-92 Development Plan's action program. Its goal was to significantly reduce the number of squatter and informal settlements. It was to be inscribed in the comprehensive nationwide housing program of DH 7,180 million for the 1992-95 perio. However, at the time of project identification, the strategy was not yet ready, and the Moroccan authorities were yet to achieve a consensus on the policy measures and their sequencing. The same held true for the Bank team. Despite this fact, however, the management encouraged fast processing of the project and limited the number of issues that were to be addressed during its preparation. Strategic issues were to be addressed during the course of the project. The overall design of the investment component, and the choice of the implementing agency, in particular, reflected the importance, given by the GoM, to stepping up provision of the owner-occupied social housing for poor and low-income families. As noted in the Memorandum of the President, "the previous experience of the urban development projects had shown, among others, the need to overcome impediments to project implementation, such as shortages of adequately trained personnel, poor decision making channels, over-regulated project executing mechanisms, and prohibitive hiring and procurement restrictions. The choice of SNEC as an executing agency, besides other reasons described below, was intended to avoid these problems by placing the land development component under a semi-autonomous agency with highly qualified personnel which has been accorded recognition as a priority institution with easy access to the top levels of govermnent and a broad license to set its own operational policies and procedures". The differential pricing mechanism (cross-subsidization of lots) was supposed to optimize cost recovery, which had also been a problem in previous projects. The project was demanding for the implementing agency SNEC which, at the time of appraisal, had limited experience and low management capacity. Besides, the US$58.5 million, on-lent to SNEC, was too large an amount to be consumned in 7 years, given SNEC's implementation experience, and existing sector constraints. As noted, the attempt to channel medium-term finance through the banking sector to private sector property developers for construction of housing for low-income families was premature, despite the indisputable need for such a mechanism. The financial system in Morocco was not ready to provide sustainable incentives for private property developers and private banks to participate in financing - 2 - low-income housing. Though the project was not complex per se, it was surely over-ambitious in its expectations about the possible sector impact and reforms being put in place to facilitate the investment in land development. The main risk of slippage from low-income to middle-income beneficiaries was correctly identified, and, indeed, materialized. 3.2 Revised Objective: Original objective (c) was dropped as the two private banks (SGMB and Wafabank) never used the Bank loans (CPL 36200 to SGMB, for an amount of US$12 million, and CPL 36210 to Wafabank, for an amount of US$12 million). After formal restructuring, the revised objectives were as follows: (a) provide serviced residential lots at affordable prices specifically targeted to squatter and low-income families; (b) provide medium-term financing to private sector property developers for construction of housing for low-income families; and (c) improve the housing finance sub-sector, housing subsidy policies, and the land delivery process. 3.3 Original Components: To achieve the challenging original objectives, the project included the following components for the total Bank loan amount of US$130 million and the total estimated project cost of US$302.25 million. (1) Financing of land development sub-projects implemented by SNEC, to deliver about 17,100 serviced residential lots with a potential of private development of about 50,800 dwelling units, of which about 93 percent would be occupied by low-income families (including rehoused squatters) as owners and tenants (US$157.46 million - 53 percent of the project cost). The World Bank's financed this component out of its US$66 million loan to GoM, US$59 million of which (Part A of the Loan) was onlent to SNEC with US$58.5 million for refnancing sub-projects costs. (2) A line of credit to banks (CIH, and two private sector banks, SGMB and Wafabank) to provide: (a) medium-term financing to eligible private sector developers for the development of residential lots and housing for low-income families (including squatters), and (b) long-tern loans to owners of developed lots building their own dwelling. (US$141.10 million- 46 percent of the project cost). The World Bank financed this component out of its US$40 million loan to CIH (US$39.5 million credit line), US$12 million to SGMB, and US$12 million to Wafabank. (3) Financing of studies for the Ministry of Housing to address housing sector issues (Land Development and Housing Finance) and to carry out a socio-economic analysis of squatter rehousing (US$2.46 million - under 1 percent of project cost). The World Bank financed this component out of the US$66 million loan to GoM (Part B of the Loan) in the amount of US$2 million. (4) Institutional development of SNEC and CIH that includes training and technical assistance (US$1.64 million - under I percent of project cost). - 3- The World Bank financed this component out of its respective loans to CIH and GoM, onlent to SNEC: US$0.5 million for CIH and US$0.5 million for SNEC. (5) Unallocated: US$5 million. 3.4 Revised Components: (1) The first component, i.e., financing of land development was retained, however, its implementation was much slower than anticipated because of land acquisition delays and low demand for lots in small towns. In 1998, SNEC requested to use about US$2 million from the unallocated category (Part B of the Loan to GoM and US$10 million from the amount onlent to SNEC (Part A of the loan) for the construction of low income housing units. This request was not accepted by the Bank, and the amount of the loan to GoM was subsequently reduced, in May 1999, from US$66 million to US$44 million, the amount onlent to SNEC (Part A of the loan), being reduced from US$59 million to US$42 million. (2) The second component was limited to the loan to CIH and was fully disbursed by March 1996.The loans to Wafabank and SGMB, amounting to US$24 million, were canceled on January Ist, 1995 and November 29, 1995 respectively. (3) The third component was retained. The three studies were completed in 1998. Their conclusions and recommendations were presented and discussed during a two days seminar organized by the Ministry of Housing on May 27 and 28, 1998, with a large participation of the private sector. (4) The fourth component was also retained. Technical assistance to CIH, which included the improvement of its information system, was completed in 1996, and the amount allocated to this sub-component (US$0.5 million) was totally disbursed. Technical assistance to SNEC, which included the design and implementation of a new MIS (Management and Information System), was also disbursed and completed in 1998. In 1999, SNEC requested to use undisbursed Bank funds for additional technical assistance and training. However, SNEC did not manage to launch the consultation for this technical assistance on time to have it comnitted on the Bank loan before the closure date of June 30, 2000. Rating: Component; Cost S LAND DEVELOPMENT; $41,500,000.00 S LINE OF CREDIT THROUGH CIH TO PRIVATE DEVELOPERS; $39,500,000.00 S STUDIES FOR MINISTRY OF HOUSING; $2,000,000.00 S TECHNICAL ASSISTANCE TO SNEC; $500,000.00 NR TECHNICAL ASSISTANCE TO CIH; $500,000.00 3.5 Quality at Entry: The Quality at Entry at the ICR stage is rated unsatisfactory. At appraisal, three major project risks were identified. 1. Slippage of beneficiaries from low-income to higher income groups. Not properly mitigated. It was noted that the pressures of increased demand resulting from a shortage of housing could lead to - 4 - higher-income groups obtaining lots targeted for low-income households and squatters. The small size of lots and strict procedures for selecting beneficiaries were expected to mitigate this risk. The risk was correctly identified but the mitigation measures included in the design of the project were not adequate. As the projects were often located where demand was low, or where alternative informal sector developments were available, SNEC had difficulty selling the smaller lots. During project preparation, the agencies considered for the implementation of the land development component were ANHI, SNEC and ERACs. Of these only ANHI was created specifically to address the resettlement of shanty towns and upgrading of informal settlements. ANH[, however, received USAID financing for part of the GoM program. As a result, SNEC was retained as the implementing agency for the investment component, even though ANHI had much more extensive experience of cross-subsidization and land development projects targeting the poor. This selection outcome clearly exacerbated the risk of targeting slippage, as, at the time, SNEC had a very limited experience in identifying and implementing new land development projects targeted to the poor. Its funds were equally limited to acquire land most suitable for responding to the effective demand of the targeted group. 2. Financial viability of SNEC. The public developers, described above, were all undercapitalized and had a limited ability to borrow domestically and internationally. The financial viability of the public developers were examined during project preparation. The ERACs were not found financially viable. The situation of SNEC was carefully assessed and financial covenants were designed to monitor SNEC's financial viability: annual return on equity, ratio of net income to profit, ratio of total wage expenses to profit, ratio of total operating expenses to profit. However, while the financial viability of SNEC did not prove to be a major risk, the amount onlent to SNEC (US$58.5 million) was too high to be absorbed by that institution, given the requirements for financial viability and its share of national land development activities. 3. The possible delays in the implementation of the sector reforms. The delivery dates for studies have slipped, and most of their recommendations are still undergoing review. The tight schedule for the studies proved over-ambitious given the complications with identifying the consultants, and cumbersome payment procedures. The consultants fees had to be channeled through the Treasury, a procedure that created delays and caused several qualified consultants to drop out. Even if the studies had been delivered on schedule, however, it was clear that the conclusions and recommendations of these studies could not have been discussed, agreed upon, and, moreover, implemented in time to have an impact on the land development and housing finance components. The lack of coherent strategy for the housing sector reform, both at appraisal and even at the mid-term review, clearly indicated that the imnplementation of the sector reform aspect of the project was not well thought through. Additional issues also surfaced which were notfully recognized at appraisal: The inclusion of private commercial banks in the housing finance component: The financial sector reforms supported by the Bank's Financial Sector Development Program (FSDP - 1991) aimed at reducing directed credit. Thus exclusive lending to CIH, with its special privileges of VAT exemption on interest on loans for dwelling with a total real estate value of less than DH 500,000, and a special strearmlined foreclosure proceduress, was no longer warranted. The inclusion of the private banks was supposed to be in - 5- line with the FSDP. To put the CIH and the private banks on equal footing, it was agreed that, unlike in the previous Housing Finance projects where CIH onlent to parastatals and municipalities, both CIH and the private banks would onlend the proceeds of the loan to private sector developers to encourage their participation in land development and social housing projects. Though the CIHI privileges were eventually extended to the private banks, this did not prove to be enough of an incentive for the private banks still could not see any advantages in expanding their business in that sector. At the same time, the risks and transaction costs of reaching low-income clients were prohibitively high for both developers and banks. Thus, the inclusion of the private commercial banks in the project even as a pilot, was clearly premature, given sector constraints. Lack of appreciation of lessons learnedfrom previous projects and realism in project design. Given the fact that the Bank had had a long history of prior involvement in the urban and housing finance sector in Morocco, the lessons learned from the previous projects were available to the preparation team and could have been more fully integrated into the design of the project. At the time the team was working on the housing sector strategy (completed in 1995, but not approved by the GoM) and preparation of the follow-up projects. Prior to this project, the Bank had assisted the Kingdom of Morocco with the two housing finance projects and the two urban upgrading projects. In sum, though the design of the project had clearly attempted to incorporate some of the lessons of the past and address the sectoral gaps, this good intention did not materialize into a quality designed project. 4. Achievement of Objective and Outputs 4.1 Outcome/achievement of objective: The rnain development objectives have been met satisfactorily. The rating is justified by the fact that the project has been successful in meeting its revised development objectives, contributed significantly to the institutional reforms and advanced the policy dialogue in the housing sector in Morocco. The sector impact of the project is considerable, despite the fact that it has been restructured, the amount of the loan significantly reduced, and some of the outputs delivered with delay. A. The revised target of 11,433 serviced residential lots was met on the 20 subprojects around the country with around 70 percent of them serving lower-income families. There were few technical problems with this component. SNEC has proved itself fully competent to manage the works, and only in a few cases there were delays caused by the defaulting contractors. However, broader housing and land sector constraints caused delays and reduced the affordability of lots for low-income families. The project went into implementation prior to initiation of reformns that would have removed some of the bottlenecks, especially in access to land and carrying out land transactions. These reforms are at the core of the current policy debate in Morocco, but it is understood, that they will take some time to implement. One of the development tools the Bank had at its disposal to control the delivery on the investment component, but has not used fully, were the eligibility criteria for subprojects. The eligibility criteria, established in the loan agreement with SNEC, were mostly respected, though implementation showed them to be often insufficient to insure smooth subproject execution and reaching the target population. Some of them were non-binding by design, while others were ill-chosen or formulated in very general terns, and, therefore, were not properly enforced during supervision: - 6 - * SNEC was supposed to show that it had acquired or was in the process of acquiring the land on which the specific development project was to be carried out. This condition proved to be non-binding. Many projects experienced delays due to SNEC being in the "process of acquiring the land" for many years. Such subprojects had to be eventually dropped due to complications with the land transaction. The MoH study on land accessibility, commnissioned under the project, highlighted the issues that SNEC and the Bank supervision staff have encountered in the field. ,It also provided concrete recommendations, including the mass appraisal of land. This is still under consideration but would, naturally, take some time to introduce. It is unfortunate, that this fundamental housing sector constraint was not addressed in a more proactive manner in the design of the project, despite significant experience which demonstrated that access to land had always been one of the main hurdles in implementation in Bank urban projects. The OED report "Twenty years of lending for urban development 1972-92" that came out in June 1994 singled that out as "lessons learned" in the design of urban projects. * The specific development project should be technically feasible and economically, financially and commercially viable. This condition was stated in very general terms and, therefore, not enforced in its major requirement for identified sub-projects to be "commercially viable". As the effective demand and competing projects in the area were not analysed systematically, and the financing model based on cross-subsidization did not work for each site, SNEC often found itself developing land with no final consumer, especially in small towns, where demand is weak. It must be noted, however, that towards the end of the project SNEC became more selective and abandoned some of the sites that could not be commercialized. * At least 50 percent of the land to be developed under the specific land development project were to be reservedfor lots and sold. This criterion aimed at increasing the land utilization factor; i.e., the amount of land that would be used directly for housing lots (the other 50 percent are used for public space, subdivisions, etc.) The 50 percent requirement was on average observed (48 percent), but it could have been set higher to further reduce developed land costs for low income population and increase commercial viability of certain projects. * At least 60 percent of the marketable land were to be sold to families with a total income under DH 3,000 (in constant 1993 DH). The 60 percent targeting was generally observed, though the Bank supervision team agreed with the concern raised by SNEC, that given the nature of income generation process and family support nets in Morocco, the income level of DH 3,000 was more of an approximate, rather than a definite eligibility criteria. Therefore, (a) it was not be used as an accurate measure of income status, and (b) it was hard to enforce. One of the risks identified at appraisal was slippage from low-income to middle-income beneficiaries and it has surely materialized, though at this stage it is too early to give precise estimates of such slippage. It must be noted, additionally, that while complying with 60 percent targeting was important for the Bank project overall, its enforcement is less meaningful in the individual case of each sub-project. Middle-income housing is a carrying structure of the urban areas, and strengthens their social cohesion. Having such quality developments helps improve the housing stock and sets the standards of service provision in growing towns. Instead of creating isolated areas with high concentration of low-income families, promoting social mix of families with different levels of income and employment profiles will develop economically, financially and socially viable communities. A preliminary socio-economic analysis performed by a local consultant in April 2000 underlined the - 7 - structural constraints by locality (its size, demographics, social and professional composition, and the dominant land ownership patterns) that had been already identified by the supervision team that took over the project in 1998. The portfolio of sub-projects is very heterogeneous in tenns of their size, type of lots offered, required development time, possible rhythm of commercialization and the speed of housing construction. The success of commercialization is correlated with the dynamics of the larger urban area where the project is located. It seems that the fastest sales happened in the towns that had a combination of the following factors: * booming economy, or important administrative function (a lot of civil servants); * transition from primarily agricultural to urban activities, and predominance of irrigated agriculture; * location at the periphery of the large metropoles; * high incidence of "Moroccans abroad" ( residents marocains d l'etranger (RME); * fast population growth; * the low level of informal land activities (strict enforcement on the part of the local authorities); * low competition from other land developers and municipalities themselves. Ignoring these constraints in appraising the sub-projects has limited the impact of the land development component on the low-income households, especially on mass resettlement of shanty town dwellers. Only 4 subprojects in the end had some share of resettlement lots. B. Most of the US$40 million line of credit to CIH was used to refinance medium-term loans to private sector property developers for construction of housing: * 0.7 percent of the loan proceeds were used to refinance medium term loans (2 years maturity) to private land developers for two projects of serviced lots to be sold to low income families. The Project financed a total of 104 lots of 120 m2 or less. _ 10.9 percent of the loan proceeds were used to refinance long term loans for construction of 909 dwellings by owners of residential lots, with a real estate value of less than DH 300,000 per dwelling. * 88.4 percent of the loan proceeds were used to refinance medium term loans (maturity up to 3 years) to private sector developers for construction of about 5,500 dwellings with a real estate value of less than DH 300,000. CIH supported the private sector activities in housing construction. The Project allowed CIH to be active in financing housing projects by private developers. The line of credit was totally disbursed in less than two years, even though the economic situation was not propitious to high activity of the housing sector. From this point of view, the project can therefore be considered a success. However, private sector activity in the development of residential lots remains marginal, as it is crowded out by public developers like SNEC, ANI and ERACs. The impact on low income families is unclear. The eligibility criteria for dwelling units was a total real estate value of less than DH 300,000. CIH complied with this criterion. However, it must be recognized that this criterion is highly questionable as a measure of income level of homeowner. The eligibility criteria for lots (120 m2 or less) is more reliable, but this sub-component was marginal in the Project. C. The objective of increasing competition in the housingfinance sector by involving private banks was not achieved. -8 - Getting private banks interested in broad based housing finance and especially targeting of the low-income segment of the market proved difficult in the absence of a comprehensive reform of the sector and of the subsidy system. Only about 15 percent of household investment in housing is financed by bank loans, and the system of subsidized interest rates is not really targeted to low-income families, but benefits instead middle income families. So, it is not surprising, that the private banks, that are now becoming more active in housing finance, are serving the high and middle income end of the market. While the Bank loan provided long term resources, this was not a sufficient incentive and private banks were wary about assuming the foreign exchange risk. D. Reforms in the housing finance sub-sector, housing subsidy policies and land delivery process have been initiated The housing sector reform agenda was advanced jointly by the investment component with SNEC and by the studies conmmissioned by MoH. On the investment side some of the opportunities for sector improvement were missed: the project worked exclusively with SNEC, a public developer amongst a few others in the Moroccan system of public and private sector developers. It seems that while developing the portfolio and investing in the institutional capacity of SNEC, more attention could have been paid to SNEC's "competitive " and "special" role in the overall housing supply system. It is worth recalling here that two national agencies, SNEC and ANHI, and seven regional agencies, ERACs, while having different mandates, are responsible for providing mainly lots, and to a certain extent housing, to all classes of households, using the system of cross-subsidization ( perequation) of the poor by charging more for the higher income lots. The Government also has a housing developer in Casablanca, Attacharouk, and the publicly-owned bank CDG (Caisse de Depots et de Gestion ) has real estate branches dealing with promotion and construction of middle-income housing. Municipalities and Urban Agencies, regional inspections, and ultimately the Ministry of Interior, are responsible for urban land policies and urban plans. There are also private developers of various scales. During the course of the project's implementation no data were collected on how local markets were divided between different suppliers of land and housing and how, if at all, the operations were integrated into the local urban development plans (SDAUs), where the latter existed. In the new medium term plan for the Housing Sector, GoM is raising the issue of clarifying the mandates of the public developers - OSTs ( organismes sous tutelle), and streamlining land development for housing into the overall management of urban growth. On the technical assistance side: the three reports had a very positive impact on the advancement of housing reforms in Morocco, but only towards the end of the project, much later than originally planned (3 years later), and, thus, were out of sync with the investment components. Most of the recommendations of the studies were incorporated in the 2000-2004 GoM Plan for the Housing Sector: * The report on developing land for housing was delivered by the EDESA-URBAPLAN-IIrHEM and recommended: changing the role of the State from direct provision of lots and housing to providing critical regulatory conditions (land registration, norms of urban development, developing financing mechanisms for primary infrastructure, ensuring transparency of procedures) to capture the private capital that is largely diverted into the informal sector, and channel it into the social housing market; * The report on the financial and fiscal aspects of housing finance was delivered by Pluram Intemational Inc, Algoe, Promoconsultant and recommended: creation of a secondary mortgage market, revision of the system of state subsidies in favor of upfront grants, reform of the legal and regulatory framework for housing construction and rental market. -9- * The socio-economic impact of resettlement operations for shantytown dwellers study was delivered by Groupement d'etudes Atelier Atrium 3/Creuset/Cabinet Guerraoui for the first case study of resettlement in Yasmine n, in Khouribga. The second study is still in progress. Delays in the choice of the site and payment processing caused subsequent changes in the consultant team. As it is limited to a sample and a transversal approach, and not a longitudinal study as originally planned, its conclusions could not be widely generalized. The reforms that have ensued from these recommendations are described in detail in the institutional development section below. 4.2 Outputs by components: A. The targeted 11, 433 serviced residential lots were delivered on 20 sub-projects around the country and over 60 percent of the lots have been sold. 60.5 percent of lots are social lots, and 9.7 percent are lots for squatter resettlement. Though it is still too early for a definitive analysis of the socio-econornic impact of the sub-projects on its intended beneficiaries, a number of preliminary impact indicators can be assembled from the data readily available at SNEC: Average marketable area 48% of the total area of the site being developed (50% goal). Number of lots developed by type of Total 11,433 (100% goal) of which: lot Resettlement of shantytown dwellers: 1,112 ( 9.7%), 6,918 social lots (60.5%), 3,202 promotional lots (28%) and 201 lots for community facilities and activities (1.8%) Degree of physical completion of 6,640 lots are ready, 3,732 are basically completed (at 95%). All lots intended works for shantytown dwellers have been delivered. Degree of commercialization 7,156 lots are sold (63%). All lots intended for shantytown dwellers have been sold and construction of houses is in progress. Property titles 5,177 land titles have been created (45% of the lots, a good indicator, taking into consideration the average time lag for creating individual titles). B. CIH has financed, through private developers, 5,503 eligible housing units and 104 serviced residential lots, and, through owners of serviced lots, 909 housing units. Number of housing units with a total real estate 6412: 909 by lot owners and 5,503 through private value of less than DH 300,000 developers. Number of serviced residential lots of less than 120 2 projects with 104 lots m2 m Total number of housing units and serviced lots 261 private developers, 467 projects, in total 7,974 completed units and 104 lots According to the Loan Agreement, the Bank loan to CIH refinanced the loans to private developers for developing serviced residential lots not exceeding 120 m2 and construction of dwellings with the total estimated cost not exceeding DH 300,000. Though these requirements were supposed to insure that the refinanced projects reach the lower-income families, they are vague as indicators of the type of - 10 - end-consumer that would buy the lots or dwellings as they can mean very different income groups, depending on the location and type of development. As most CIH are medium tern loans, they were repaid after 2 or 3 years and the amount recovered by CIH was used by CIH to finance new projects before the end of the Bank loan grace period. Currently there are 42 loans that are at risk for the total loan value of DH 149 million. This represents 12 percent of the total value of the loans CIH made as part of the project (DH 1,250 million). The total cost of the projects financed is DH 3,300 million. At the time of the preparation of this report CIH was still looking at the causes that made 12 percent of the loans to become non-performing. C. Three studiesfor MoH were delivered and widely disseminated The studies were commissioned with a delay due to changes in the terms of reference, payment difficulties, and consequent problem in retaining the selected consultants, but all, with the exception of one case study, were successfully completed in 1998 and discussed at a seminar. D. Technical assistance to CIH and SNEC The technical assistance components to CIH and SNEC (US$0.5 million each) were designed to improve their data processing and management systems, in order to enable easy generation of monitoring documents for each operation, and to make their existing accounting and budgeting systems compatible. Training of staff was also included. CIH has installed computers and software to upgrade its data processing capacity. However, no technical assistance was provided to improve its management and risk monitoring. Technical assistance to SNEC included design and implementation of a new integrated Management and Information System with four components: Land Management, Marketing, Procurement and Contracts Management, Programming and Budgeting. This task was completed in 1998. However, in 1999, the new management of SNEC felt that the system was not used to its full capacity and towards the end of the project requested additional technical assistance and training to fully operationalize the MIS: to integrate the marketing module of the MIS, to develop its operational manual, and to provide operational training to SNEC staff. After some delay on the part of the Bank, a consultant prepared terms of reference for this additional assistance in November 1999. Unfortunately, SNEC was not able to launch these new components in time, due to delays in the approval of its budget for FY2000. Consequently, the request for financing could not be granted as the project closure date was set for June 30, 2000. Despite this setback, SNEC has decided to go ahead and finance the installation of the marketing module of the MIS out of its own budget. 4.3 Net Present Value/Economic rate of return: The following costs were included to compute the economic rate of return (ERR) of land development sub-projects: (a) land and site development, including site preparation and on-site infrastructure; (b) the share of off-site infrastructure allocable to the sites; (c) physical contingencies; (d) administration and supervision. The economic analysis was done on the basis of domestic market prices and a project life of 10 years. The assumption was made that land values would increase by 10 percent during the first year after project completion and 5 percent thereafter. The ERRs of a sample of 7 sub-projects were computed; they ranged from 29 to 48 percent. In the conservative assumption of a one-time increase of 10 percent in land - 11 - values, ERRs were expected to reach 15-40 percent. The ERRs computed for a sample of 10 houses financed by mortgage loans refinanced under the Second Housing Finance Project were applied to the CIH/private banks component for loans to the private sector land and housing developers. Such ERRs were assumed to be similar to estimated 15-20 percent. The assumption was used that half of the line of credit would be used for the construction of houses and half for the development of serviced lots, and overall ERR of the proposed project was expected to reach 27 percent under the conservative assumption of increased land values and ERR of 36 percent under the more optimistic assumption of increased land values. Both of this values were well above the prevailing cost of capital of 10 percent. It is too early yet to recalculate the ERR for land development sub-projects as only around 60 percent of the lots have been commercialized, and even less have been constructed, so land value increases are hard to observe with certainty and consistency, and other economic benefits could not be observed until the communities start functioning. A rough estimate, made using the methodology described in the SAR, gives comparable ERRs for the three sites where land value data is available (Ain Aouda, Kelaa Sraghna and Yasmine II). Resale land values there increased by 15-50 percent. The costs of the retained subprojects were close to the original estimates, and sale prices are also in line with the original estimates. 4.4 Financial rate of return: N/A 4.5 Institutional development impact: The institutional development impact is substantial. The two main achievements are: (a) advancing the policy debate and the reforms in the Housing Sector, including redefining the role of the State in providing serviced land and social housing, the relative roles of public and private developers, the housing finance system and the housing aid for low income families, and (b) the capacity building at SNEC. The impact on capacity building at CIH is more questionable. (a) Advancing the policy debate and the reforms in the Housing Sector Changing the role of the State and increasing the participation ofprivate developers As a result of the experience of the project, it becarne clear that SNEC as an unspecialized public developer, can not continue launching and commercializing many small and somewhat ad hoc projects, based on intemal cross-subsidization formula. SNEC has been charged with the piloting of ZUNs, defined as neighborhoods and new towns the primary infrastructure for which is to be developed "wholesale" by SNEC and later be sold to private developers that would further develop secondary and tertiary infrastructure on certain segments of the zone for individual sale. The future of public developers is called into question and their financial and operational records are under scrutiny as most of them are experiencing financial difficulties and find themselves in direct competition with the private developers and municipalities. The difficulty with provision of resettlement lots, and relatively high prices of SNEC's sites, lead the GoM to consider another mechanism for broadening the private sector involvement in land development: the launching of ZAPs, that would compete with the informal sector by offering the low-income families partially equipped lots for further improvement at more affordable prices. - 12- Though the concept of ZUNs and ZAPs is widely accepted, the regulatory and financial mechanisms of their practical implementation still need to be worked out. SNEC is experimenting with possible mechanisms for partnering with the private developers in the framework of ZUNs it has already launched. The issue of supply of urban landfor development has been recognized as a reform priority The project reconfirmed the existing awareness that the urban regulation of land, being very rigid and slow, causes major distortions and delays for the land and housing markets. The reform of the Land Law and launching of mass titling is underway. The Agency responsible for land title registration (Conservation Fonciere) will be given autonomous status by the law currently under consideration, and its efficiency will be improved. Creation of the National Agency of Land Promotion (Agence Nationale de la Promotion Fonciere) is underway. Assistance from the Bank was requested for the preparation of the reform of the administration of State Properties (Administration des Domaines). The new Rental Law was passed, redefining the relationship between the landlord and tenants in order to increase investment in the rental sector. That would in tum augment the supply of housing for low-and middle-income families, and increase the efficiency of use and incentives for the up-keep of existing housing stock. The laws and norms governing the fornal process of urbanization are being revised The GoM has started to reform the Urban Planning Law (Code d'Urbanisme), aiming at simplifying the documents and urban norms to make them more consistent with the social and economic dynamics of Moroccan urbanization. In the meantime, the articles of the Law and standard procedures for delivery of construction permits are being applied with more flexibility in order to speed up the unduly cumbersome process of approvals and permits. More attention is being paid to the social and economic analysis in the Urban Development Plans (SDAUs) and the need for more frequent revisions. The mandate of the existing Urban Planning Agencies and Inspections, established in the main regional capitals, was changed in 2000: they are charged now with urban planning issues of their regions. Such decentralization of planning will allow better and faster enforcement of planning laws and norms. The housing finance market is opening up to commercial banks and becomes more liquid The lack of interest of the private banks in participating in the project was not surprising, as they could not lend on the same terms as CIH. To level the playing field and encourage competition in mortgage provision, agreements (Conventions) were signed between the MoF and private banks to allow them to benefit from interest rate subsidies available to CIH special low-income program. At the same time, to increase liquidity in the housing finance market, a Law on securitization was passed. Today the private banks are more and more active in the housing sector, although their participation is still limited to the higher end of the market. Improving targeting and efficiency of the State Housing Subsidy The studies and consultant reports delivered during the project helped convince the GoM that the current system of the State housing subsidy (interest rate subsidy for social housing) is costly and poorly targeted. As a result of this finding, the GoM is considering replacing the indirect subsidy by a system of upfront grants that would be better targeted to the low-income households. The Bank is providing on-going assistance in the preparation for such transition. (b) Impact on the executing agency for the land development component: role and situation of SNEC - 13 - At the start of the project SNEC was established in 1987 to take over the program of FNAET from MoH accounts. During its first two years of operation (1987-89), SNEC concentrated almost entirely on FNAET activities: assessment of works completed to date, determination and verification of all outstanding liabilities, completion of construction activities for several of the sites. For these services SNEC received a commission of 7 percent of all transactions that it processed on behalf of FNAET. This commission was more or less consistent with the comparable fees charged by the private sector and was the most important source of income. At the start of the project SNEC operated with a staff of 43 people and was divided into three functional departments: technical, commnercial, and administrative. A staff increase to 70 people was made a condition of effectiveness. It had only 5 projects under construction for its own account, and thus its track record as a land developer was still relatively untested. As SNEC was undercapitalized at the start of the project, its paid-in capital was also required to increase annually to reach DH 20 million by the end of 1998. At the end of the proiect Scope of operations and expertise has grown significantly. A new strategy has been approved. SNEC's portfolio has increased to 85 projects (1998), with over 31,000 lots and 2,674 dwelling units for the total coast of about 4,390 MDH. From 1997 to 2000, 42 projects have been completed, with a total of 12,000 dwelling units. The extensive experience acquired by SNEC allowed the new management, nominated in 1998, to define a strategy and take into account the achievements and shortcomings of the previous modes of operation: (i) greater selectivity in the choice of project sites, and emphasis on marketing: the scarcity of public land in areas were the demand for serviced lots is high led to launching too many projects in small cities where land was available, but the demand was weak or already satisfied by competing projects managed by other public developers or by municipalities. The new SNEC management therefore decided to carefully screen all on-going projects, and to cancel those without commercial potential or stalled by property issues. This selectivity allowed SNEC to concentrate on a more aggressive marketing policy for retained projects, based on the decentralization of commercial services in regional agencies, and on the reinforcement of staffing and communication strategy. (ii) identification and launching of new modes of operation: to address the persistent scarcity of public land in large urban areas, SNEC designed, in close cooperation with the Ministry of Urban Planning and Housing, a new strategy based on the development of large scale projects outside urban perimeters, called ZUN (Zone d'urbanisation nouvelle). On these projects, SNEC will focus on the design and implementation of primary infrastructure, while secondary infrastructure, serviced lots and housing will be implemented by private developers. The first ZUN is already being implemented in Agadir, and others are being prepared near Rabat, Sale, Nador, and Fes. Management has improved To address the enlarged scope of operations, SNEC's staffing has increased steadily, to 121 people in 2000. Priority was given to the staffing of the new regional agencies and to commercial services. - 14 - Monitoring, internal audit and budgeting services were also reinforced. Growth in SNEC staffing SNEC 's new management is also focusing on the full integration of the MIS, especially on the marketing module and on reporting procedures between regional age-ncies and headquarters. Financial situation has dramatically improved and constitutes a sound basis for the implementation of the new strategy Concentration of SNEC activity on viable projects has improved the financial situation of the agency, and SNEC complies with all financial targets set in the loan agreement: * Return on equity is 15.7% (target: no less than 4%) * Income ratio is 58.7% (target: no less than 10%) * Wage expenses to profit is 26.2% (target: less than 30%) * Total operating expenses to profit is 41.3% (target: less than 50%). In compliance with the Loan Agreement, SNEC has increased its equity in successive amounts of about DH 4 m-illion throughout the life of the project. Moreover, in November 1999, SNEC's equity was increased from DH 22.5 million to DH 106.5 million, which is well above the target set by the loan agreement. This last equity increase is a highly important development in the way of institutional strengthening of SNEC. It conveys the trust of the GoM in the management of the agency and its capacity to imnplement the new ZUNs. By reducing its debt to equity ratio, it prepares SNEC for its future large scale development operations, which will require additional long term fmnancing. SNEC has already secured long termn loans from Arab development agencies for the first ZUN in Agadir. (c) Impact on the executing agency for loans to private developers: role and situation of CIHI The institutional imnpact on CIH as part of this project was negligible, as the CIH component was a simple line of credit with very limiited policy requirements. The technical assistance component was limited to upgrading its computer hardware and software. However, the Bank has a long history of working with CIH, and the role of CIH was central to the debates about the reform of the housing finance and subsidy systems in Morocco. As this reform is one of the major institutional achievements of the project, the institutional evolution of ClIH during the life of the project merits a brief account. - 15- At the start of the project At appraisal, CIH was found efficient and well managed, comparable to a private sector enterprise, and independent from the GoM, financially and in its management. CIH had doubled its branch network in the five years prior to the start of the project, and in 1992, it had 46 branches and a staff of 1,120. At appraisal the CIH policy statement and lending criteria were clearly defined and its portfolio supervision was found rigorous. Fifteen percent of the portfolio was earmarked towards the tourism sector, and housing loans made 75 percent. Loan arrears were increasing, but were found within acceptable parameters - 5.6 percent of its portfolio by the end of 1991 with more than two thirds of it arising from the tourism sector (bankrupt hotel chains). The targets set for financial ratios were the same as those set for the Second Housing Finance Project with CIH (Loan 3122-MOR approved in 1989 and completed in 1995). These targets included the interest rate spread, operating expenses to total assets, current assets to sight deposits, end equity to total assets. At the end of the project During project implementation, the Bank was getting increasingly concemed about the financial health of CIH. Up to FY97, CIH complied with financial targets, except for operating expenses as a percentage of total assets, but this was not the most serious issue, as it was due to the transformation of CIH into a depository bank. The main issue was CIH non compliance with the Bank Al-Maghrib (Central Bank) regulations on loan classification and provisions. Bank Al-Maghrib had agreed to postpone enforcement of these regulations by former specialized public financial institutions (OFS) such as CIH. However, the Bank insisted on resolving the problem, having reviewed the results of the audit report of CIH Financial Statements for FY94 In 1996, as the Bank was considering a new line of credit to CIH, the Bank missions to Morocco stated clearly that a special audit of CIH portfolio was a prerequisite to future lending to CIH, and the Bank prepared terms of reference for this audit. However, the on-going loan had already been totally disbursed, the idea of a new line of credit was abandoned in 1997, and the Bank's leverage on CIH was thus limited. The special audit was never commissioned by CIH. The situation became critical in 1998 when CIH began enforcing Bank Al-Maghrib regulations. The amount of non-perforning loans had soared since several years, as a result of the bankruptcies in the hotel industry, the crisis in the housing sector, but also as a result of bad management and weak risk management. The annual return on operations became negative in 1998, because of increased provisions. The other financial ratios still complied with the targets set by the Bank. The GoM nominated a new management team, and the audit of the FY99 financial statements was carried out according to intemational audit standards. It is qualified and raises questions of accountability for non-recoverable loans. Under-provisioning for doubtful debts could reach DH 5,700 million, against capital and reserves of DH 851 million, indicating that CIH is technically bankrupt. However, the main shareholders and the GoM have decided on a rehabilitation plan (Plan de redressement) which was approved by CIH's Board (Conseil d'Administration) in June 2000. The Plan includes: * injection of DH 2 billion in equity by shareholders in two tranches (2000 and 2001); * injection of DH 3 billion in loans by the banking sector, with State guarantee; * a State loan of DH 1 billion; * the write-off by the State of DH 200 million of exchange losses; * the write-off by the State of DH 400 million of ERAC debts; - 16 - * a proactive policy of nonperforming debt recovery; * extensive State support for the enforcement of the Plan. The Plan is expected to restore financial equilibrium to CIH and clean up its portfolio. Its objective is to transform CIH into a viable financial institution able to compete in a level playing field with commercial banks, that are increasingly active in the housing finance sector. The creation of a secondary mortgage market, with the new law on securitization, will contribute to a further development of the mortgage market and the housing finance sector. 5. Major Factors Affecting Implementation and Outcome 5.1 Factors outside the control of government or implementing agency: Macroeconomic situation During the project implementation period, the Moroccan economy experienced nearly zero real GDP per capita growth, with a 10 percent negative growth in 1995. That explains, in part the sluggishness in national and local investment demand. By the end of the decade, poverty is 50 percent higher than in 199 1. In the last two years Morocco witnessed a steep rise in unemployment, which reached 22.8 percent in urban areas at the end of 1999. In view of second consecutive drought year in 2000, it can be assumed that due to increased rural migration combined with low growth, unemployment in urban areas might exceed 2.5 percent. The housing sector has been performing below its potential, and currently accounts for only 5 percent of GDP. Economic stagnation, lack of public resources, increased number of the urban poor, and low opportunities for income generating activities jeopardized the project's design, that aimed at increasing ownership of land for self-construction and resettlement, and housing construction subsidies programs targeted at the poor. 5.2 Factors generally subject to government control: Land development component: access to land was identified as a major constraint at the start of the project, and it remained the main obstacle for the project's implementation and cause of slow disbursements of the land development component. The new mandate for SNEC and the first operations of ZUN, the project for the Land Agency (Agence Fonci&re), show that the government has taken this constraint int.o consideration and is responding to it. The housing finance (CIH) component: the bankrupt hotel chains and financial difficulties of ERACs, the large clients of CIH, jeopardized the overwhelming part of the CIH portfolio and along with other financial difficulties caused major problems for this institution. As CIH remains a major player on the housing finance market, its institutional viability is closely linked with the sustainability, costs, and success of implementing any current and future housing finance programs. The implementation of the recommendations of the MoH studies: the studies could have been delivered much faster and their recommendations translated into substantial reform even during the life of the project. The payment delays on the part of the Treasury caused many difficulties for the management of the consultant teams. 5.3 Factors generally subject to implementing agency control: - 17 - SNEC has performed well, and has increased its management effectiveness, pursued sound financial policy, rationalized its operations and staffing, and improved the quality of its growing portfolio. Such behavior naturally led to the reduction of the size of the project, as some projects identified initially were not viable. However, SNEC could have been more proactive in applying stricter eligibility criteria, undertaking market research to avoid engaging funds in non-viable projects, and finding ways to reduce the cost of serviced land to make its projects more competitive with the ample supply of lower cost informal developments. Bank funds were rather expensive for SNEC, as they were on-lent by GoM with a fixed interest rate of 12.5 percent, close to prevailing interest rate on the local market at the time of negotiations. However, while, consequently, domestic market rates dropped to 7-8 percent, no action was taken by the counterparts to reduce the on-lending rate. The on-lending rate was finally reduced to 11 percent only in 1999. At that time SNEC, whose financial situation had steadily improved, was already searching for cheaper resources on the local financial market. This was an important factor behind the low disbursement rate of the Bank loan. However, it must also be considered as a positive outcome of the sound SNEC management and as an indicator of the positive institutional impact of the project. 5.4 Costs andfinancing: The project has experienced a drop in costs due to reduced scope of the land development component and the withdrawal of two private banks. The final cost of all the retained sub-projects, including land costs, studies, works and financial fees, amount to US$65.5 million (DH 653.71 million at the exchange rate of DH 9.8 per US dollar). That represents 41.6 percent of the original estimates. Overall disbursements were very fast in the first three years due to the fast pace of the loan to CIH and slowed down significantly towards the end due to the slow progress of the land development component and the perception of the high price of the Bank loan by SNEC. The Bank reacted by reducing the loan by USS22 million in May 1999. Despite the reduction, the US$14.43 million was left undisbursed, due to the large gap between the appraisal estimate and actual disbursements. Loan Disbursements: Cumulative Estimated and Actual (US$ million equivalent) Land Development+CIH FY94 FY95 FY96 FY97 FY98 FY99 FYOO Appraisal Estimate 1.53 16.43 52.15 76.51 96.56 105.1 106.00 Actual 8.63 39.26 48.64 53.76 59.14 66.46 69.57 Actual as % of Estimate 564% 239% 93% 70% 61% 63% 66% 6. Sustainability 6.1 Rationale for sustainability rating: Sustainability is rated likely. The lessons learned during project implementation have already been taken into consideration by the implementing agency, the scale of the project has been reduced, and some of the land development practices have been modified accordingly. All that will insure sustainability of ongoing and future operations. Given the reduced availability of public land, the formula of cross-subsidization is losing its - 18- universal applicability, and new mechanisms are being explored. It must be noted, that the SNEC projects imposed no direct costs on the state Budget. In the Memorandumn of the President, sustainability of the project was mostly described in terms of the economic viability of the system of intemal cross-subsidies through differential pricing of lots and private sector participation, both in land development and housing finance. Lots for squatters and low-income families were to be sold at or below cost, cross-subsidized from the sale at higher prices of other lots within the same sub-project. To insure sustainability, total revenue was to be equal total costs for each sub-project. SNEC was to be not only developing, but also marketing its projects to the end consumer for private construction. The implementation showed that the cross-subsidization system works well in cases when there is a broad-based demand for all types of lots, from economy to "villa" type lots, with a large group of buyers for residential lots with commercial ground floors. That is mostly the case of fast growing middle to large cities. Where SNEC had managed to secure land in such areas, its projects have been a success. Unfortunately, cheap public land is hard to come by in such areas, and there is a strong competition from other developers. Many projects that had been included in the list at appraisal, being either inherited from the FNAET, or because it was easy for SNEC to get access to land, were eventually dropped, and the loan amount for land development component was scaled down to finance only the commercially viable sites. Both the Bank and the executing agency displayed vigilance and preferred to limit the size of the project than to continue financing what would be "white elephant" sites. By the end of the project, SNEC not only preserved but, in fact, increased its financial viability. Its role in the system of the OST (Organisme sous tutelle), has been reassessed to reflect its comparative advantage. SNEC has been charged with developing large tracts of land in the areas of high growth (e.g. Agadir peri-urban districts), so called ZUNs (Zones d'Urbanisation Nouvelle). This concept was discussed with the Bank throughout project implementation, as SNEC was becoming more and more financially and operationally sustainable. Developing "wholesale" primary infrastructure on large tracts of land and letting private developer provide secondary and tertiary infrastructure and then "retail" lots, will avoid the problems that surfaced in the current project. Equally, the policies of pursuing gradual land development, and using private capital to deliver further improvements, are conceptualized in the so called ZAPs (Zones d'amenagement progressij). It will reduce the cost of formally developed land and the burden on local budgets. The project was implemented during a period of shiffing sectoral orientation. Project experience proved that some of the sacred housing policy pillars inherited from the 80's did not work in the 90's and are unsustainable in the future: * the model of a national public developer for social housing at the same time striving for financial and commercial viability, with limited funds for accessing an artificially restricted pool of land, acting in limited, if any, coordination with municipalities and their infrastructure programs; * universal applicability of cross-subsidization system for individual sub-projects; * excessive reliance on policies of resettlement of shanty town dwellers without drastic reforns of the urban regulatory and subsidy system,. * the temporary emphasis on provision of state- and privately constructed social housing (the program of 200,000 units) proved costly and did not match the demand of the population, particularly in small and middle-size towns. 6.2 Transition arrangement to regular operations: - 19- The way the institutions have adjusted during the course of the project testifies that new modalities were foumd, and that in some way, the scaled down project has ended up serving as a pilot. It has produced limited outcomes that are, nonetheless, sustainable. The experience of the land development component of the project and parallel developments of the 90s caused MoH to reconsider its strategy. The implementation experience of the project, the studies completed as part of the project, and a seminar organized by the Bank to discuss them, have led the GoM to design and begin implementing important reforms that would provide a totally different operating environment for any future operation. The difficult situation of CIH is being addressed by the GoM in its rehabilitation plan for this institution, and the Bank will be monitoring the progress closely. 7. Bank and Borrower Performance Bank 7. 1 Lending: Bank performance at lending is rated unsatisfactory First. risks were underestimated during project preparation: i. Delays in the implementation of sector reforms: The Housing Sector Development Letter signed by the GoM was written in very general terms and did not include deadlines for specific reforms. 2. Many serviced lots produced do not benefit the low-income family group intended: The project did not include enforceable procedures for selecting beneficiaries. Moreover, previous projects analyzed at appraisal to infer the impact of land development projects on low- income groups were located in main agglomerations, whereas most projects financed by this loan were located in smaller cities. 3. Private developers launched a very limited number of social housing projects: No practical mechanisms were introduced to broaden the system of housing finance, so it is not surprising that private developers continued to specialize in housing projects for high and middle income families. Most issues mentioned above had been properly identified. However, it seems, that at the time of project preparation, there was a sense of urgency, from the executing agencies (SNEC and CIH) to secure long term financing from the Bank, and also from Bank management to approve the loan, even if the institutional framework was not ready and if there were no strict criteria ensuring that the Project was effectively targeting low-income families. Second, the staff skill mix at preparation and appraisal has been inadequate, as the team lacked specialized expertise in housing and land development. 7.2 Supervision: Bank performance at supervision is rated unsatisfactory. The supervision team was proactive in identifying together with SNEC new land development sub-projects that could be financed by the Project. However, both SNEC and the supervision team failed to focus on demand analysis when appraising sub-projects. When the supervision team realized that the implementation of land development projects was lagging behind schedule, the first reaction was to increase the percentage of works refinanced by the Bank loan from 65% to 85%. During mid-tern review (May 1998), the lingering financial problems of CIH were identified, but at that - 20 - time the Bank's leverage was already limited since this component had been totally disbursed. The supervision did not enforce the timely delivery of the project audit reports, that were all received after the component had been totally disbursed. The workshop organized to disseminate the results of the studies on housing sector reforms was successful, but no support was provided to the MoH to deal with the continuing delays by the Treasury in paying the consultants. The opportunity was missed to restructure the land development component. It was clear at the time that the delays in sub-projects implementation and the weak demand for lots in small cities would reduce the financing needs of SNEC. Instead, the supervision team proposed to use part of the loan to finance SNEC housing projects, which was not in line with Bank policy, did not represent SNEC main line of business, and was inconsistent with the project design. Lingering problems in implementation of some sub-projects were not addressed until the last quarter of 1998. The new team that assumed project supervision, acted proactively and dropped some of the non-viable sub-projects from the project list and canceled part of the loan, alleviating the financial burden for SNEC and the country. The Loan to the Government has been reduced from US$66 million to US$44 million on May 13, 1999. The fraction of the loan for Category 4 (unallocated), managed by the State Secretariat of Housing, was canceled (US$5 million) and the amount on-lent to SNEC was reduced from US$59 million to US$42 million. In the same period, weaknesses in the previous project management were identified and addressed: the threshold for the Banks prior review of contracts for civil works had been raised from US$300,000 to US$1,500,000 by a simple letter from the Bank, without a request from the Borrower and without a formal amendment. Some time later, another letter from the Bank had applied the new threshold to the disbursements on the basis of Statement of Expenditures. There had been also problems with the refinancing rate and the change of the annual interest rate charged to SNEC. The rate was reduced from 12.5 percent to 1 1 percent by an amendment in the subsidiary loan with SNEC without introducing the necessary changes in the Loan Agreement with the GoM. The timely delivery of account audits for SNEC and CIH was insisted upon, only after 1998, in order to stay in compliance with the legal covenants. 7.3 Overall Bank performance: Overall Bank performance is rated unsatisfactory. The preparation and appraisal stages of the project cycle were clearly not satisfactory. However, the supervision performance rating is more ambiguous. Despite problems outlined above, the Bank staff cooperated closely with the Government and the implementing agency for the land development component and behaved proactively towards the end of the project. The relationship with CIH was more arms-length due to the nature of the component - a simple line of credit. There were, however, significant problems in the overall quality of project supervision and institutional memory both for the Bank and the Borrower. The opportunity was lost to restructure the project at mid-term review, and though most problems were eventually rectified by FY99, valuable time was lost to make the best use of the available resources. The team has discussed the importance of the ICR preparation with the main executing agency. SNEC expressed a high level of engagement and initiative in taking stock and leaming from the project - 21 - implementation. Borrower 7.4 Preparation: Borrower performance at preparation is rated satisfactory. The project was designed in response to the Borrower's request for emergency Bank's assistance for its national social housing program. The Borrower was skeptical about the participation of the private sector in the operation, as at the time the role of the private sector in social housing was limited. The private banks component was included on the Bank's initiative and on a pilot basis in an attempt to encourage private sector participation, but the concerns of the borrower were borne out and private banks participation quickly tumed out to be premature, even as a pilot. Lack of a clear government strategy, however, can be considered one of the major weaknesses at the preparation stage. 7.5 Government implementation performance: Government implementation perforrnance is rated satisfactory. The launching and supervision of studies by the MoH was fully satisfactory, as was the organization of the dissemination workshop. The Government's commitment to reform has also been strong. 7.6 Implementing Agency: Implementing agencies' perfonnance is rated satisfactory. SNEC: the initial objectives of SNEC in launching the land development component proved over-ambitious, and SNEC failed to give enough importance to market analyses. In 1997/1998, under pressure from the Ministry of Housing, priority was given to social housing over land development, which contributed to the delays in project restructuring, as this policy was not in line with Bank orientations and not feasible, given the mandate of SNEC. Further changes in sector policy and a new management at SNEC led to more realistic views: cancellation of non viable projects, priority given to the marketing of viable projects, and improvement of internal management. Unfortunately, this change occurred too late to allow a full restructuring of the project, which should have occurred during mid-term review. CIH: the implementing agency disbursed quickly the housing finance component and complied with the eligibility criteria. However, CIH management did not fully cooperate with supervision teams to identify the lingering problems with its loan portfolio. MoH: the studies were delivered with large delays, that were caused largely by bureaucratic difficulties outside of control of the implementation team. 7.7 Overall Borrower perfornance: Overall Borrower performance is rated satisfactory. - 22 - There were some continuity and consistency problems in the implementation policy of the borrowers. However, they finally managed to built on positive and negative experiences of the project to define a new sector strategy, to define a new role for SNEC and CIH, and to improve their intemal management. 8. Lessons Learned 1. To be effective, land development projects should involve municipalities and other regional stakeholders and be integrated with the broader urban and environmental development programs for the agglomerations. In Morocco, as in most countries, social housing services are financed by the central government. Local governments are generally responsible for providing adequate municipal services, such as sewage, solid waste management, and finance local roads, lighting and some public and commercial facilities. More often than not, they have low progranmming and budgeting capacity to provide these services. Many limitations are also imposed by the urban development plans and norms that are too rigid and do not take into consideration local realities. It was decided at the preparation stage that the project will not include institutional development of municipalities, as this was being addressed under the Municipal Finance (FEC) project in preparation at the time. Implementation process proved that the two projects oftentimes provided mutually exclusive incentives for municipalities - a SNEC project will not sell, while the municipality would seek FEC credit to finance infrastructure upgrading of the informal settlement it allowed to grow spontaneously in the meantime. Therefore, the project has demonstrated that all these complimentary aspects of local management have to be taken into consideration for a land development project to respond to the targeted local effective demand and be sustainable. The mechanism of PLHDU (plans locaux d'habitat et de developpement urbain) is being considered by the GoM as a means to orchestrate the process of solving local housing, urban, and environmental problems at a regional level. 2. Combining housing sector and housing finance sector components requires careful timing. The project's three components, while all working towards the same goal, seem to have been designed with totally different timeframes in mind. Piloting private banks financing of private developers as part of the low-income housing project was not warranted as the financial system could not provide private banks with the incentives to compete with the CIH, a specialized public institution, and private developers were not interested in providing low-income housing as there were no economic incentives or mechanisms for them to do this. Besides, the banking system in Morocco reaches less then 20 percent of housing purchases and construction and the poor are the last ones to get access to credit. The studies component was supposed to advance the reforms in both, the housing and housing finance markets, but even without slippage in the delivery dates for the studies, realistic estimates should have been made as to the earliest possible impact of such studies on the sector. No expectations should have been placed on their contribution to the performance of the other two components. - 23 - 3. The institiutional and regulatory framework governing land use and housing sector was not ready for such a large scale project of land development for low-income families The issue of land accessibility, the cumbersome procedures that govern land use and increase land prices were raised at preparation, and even the Urban Law was critically reviewed, but nothing was streamlined during the life of the project. Bank's participation (versus USAID just supporting ANHI) in the project was justified by the argument of the comparative advantage of the Bank in bringing higher value added by financing more difficult programs with a well conceived strategy framework. It seems, that there should have been more conditionalities on the gradual reform of the regulatory framework and operational strategy at least by the mid-term of the project. Equally, the vast investment resources allocated could not have been realistically absorbed in the 7 years of implementation period, given the severe sector constraints. 9. Partner Comments (a) Borroiwer/implementing agency: No report has been received from the Borrower yet. (b) Cofinanciers: (c) Other partners (NGOsivrivate sector): 10. Additional Information - 24 - Annex 1. Key Performance Indicators/Log Frame Matrix Outcome / Impact Indicators: I lciftmwix Proed in last PSR Ac..,atU . Estlmae Increased supply of lots for housing Expeded delivery of 11,433 serviced Though not all identified sub-projects proved construction at affordable prices for residential lots on 20 sub-projects around the to be viable due to land market constraints, low-income households by a public developer country, around 70-% of them serving 11,433 serviced residential lots were low-income families. delivered on 20 sub-projects around the country with around 70% of them serving low-income families. Increased availability of medium term The line of credit totally disbursed in two CIH financed 261 private developers, 467 financing to private sector developers for years and financed 6412 eligible housing projects, and 6412 units with total real estate construction of housing for low-income units, however the private sector acivity in value less than 300, 000 DH. However, the families the development of residential lots for financial and housing sector was not ready to low-income families remains marginal. support private sector activity for low-income housing, so the impact is marginal. At the same time, increased private sector partidipation is observed on the high and middle income end of the market. Reforms in housing finance sub-sector, The investment component has strengthened SNEC is charged with implementation of housing subsidy policies, and land delivery SNEC as an institubon and helped to define ZUNs, the Land Registration agency is being process its proper role in the system of OSTs. reformed, the National Agency of Land The studies have been instrumental for Promoton is being created, the rental law supporting the sector dialogue and reforms. has been passed, the Urban Planning Law is under reform, and procedures are being simplified and decentralized, the housing finance market is opening up to equal competition for commercial banks, a secondary mortgage facility is under consideration. Output Indicators: in~~I~$QdMat~1X ?rqjc% ktdilaetPSR' cttLWtItlt (a) number of serviced lots delivered to 11400 lots built, including at least 60% 11,433 lots built, on 20 subprojects, squatters and low4ncome households (at resettlement and social lots including 60.5% social lots. 9.7% are least 60%) of total of lots delivered); resettlement lots. (b) number of dwellings constructed by About 6200 housing units with a total real 5, 503 eligible housing units and 104 private developers for low-income families; estate value of less than 300,000, about serviced residential lots were financed and 5,300 through private developers and 909 by through private developers, and 909 lot owners, 104 serviced residential lots of housing units through owners of serviced less than 120 sq.m. lots. (c) completion of the 3 projed-supported Studies delivered and disseminated Studies delivered and disseminated. A studies. seminar was held with govemment officials and developers. The studies' recommendations were discussed, and some are already implemented as laws. (d) Technical assistance to CIH and SNEC CIH TA completed. CIH upgraded its data Technical assistance delivered according to processing capacity. A consultant provided the TORs agreed at appraisal. CIH upgraded TORs for additional assistance to implement Rs data processing capacity. At SNEC a fully the marketng module of MIS and operatonal integrated MIS has been implemented, manual and training for SNEC staff. further reinforcement the marketing module and operational training was requested but could not be performed before the loan ciosed. End of project - 25 - Annex 2. Project Costs and Financing ProJect Cost by Component (in US$ million e uivalent) 0976r*c '~e 2y ( 4ompont Ut$ mte--illion US$ mnillion .____._ A. Land Development 1. Land Acquisition 38.06 14.69 38.6 2. Civil Works 75.73 35.26 46.6 3. Design Consultancies 11.29 2.11 18.7 4. Other Costs 32.38 9.03 27.9 B. Loans to Private Developers 1. Through CIH 87.78 115.80 131.92 2. Through private sector banks 53.32 0.00 0 C. Technical Assistance 1. To SNEC 0.62 0.08 13 2. To CIH 0.62 0.50 81 3. Studies for MoH 2.46 1.82 74 Total Baseline Cost 302.26 179.29 Physical Contingencies 9.35 Price Contingencies 9.92 Total Project Costs 321.53 Total Financing Required 321.53 179.29 Project Costs by Procurement Arrangements (Appraisal Estimate) (US$ million equivalent) 1. Works 7.50 81.90 0.00 0.00 89.40 . ~~~(4.88) (50.06) (0.00) (0.00) (54.94) 2. Goods OM0 O.Oo 0.00 0.00 0.00 (0.00) (0.00) (0.00) (0.00) (0.00) 3. Services 0.00 0.00 13.10 0.00 13.10 land project design (0.00) (0.00) (8.34) (0.00) (8.34) consultancies technical assistance 0.00 0.00 1.23 0.00 1.23 (0.00) (0.00) (1.00) (0.00) (1.00) 4. Studies for MoH 0.00 0.00 2.46 0.00 2.46 (0.00) (0.00) (2.00) (0.00) (2.00) 5. Sub loans 0.00 0.00 141.10 0.00 141.10 (0.00) (0.00) (63.50) 0.00 (63.50) Total 7.50 81.90 157.89 0.00 247.29 (4.88) (50.06) (74.84) (0.00) (129.78) The table does not include not bank-finance costs: land acquisition costs, financing costs and other costs as defined in the SAR, for the total sum of US $73.98 million. - 26 - Project Costs by Procurem Arrangements ActuallLatest Estimate) US$ million equivalnt 1. Works 0.00 35.26 0.00 0.00 35.26 (0.00) (24.40) (0.00) (0.00) (24.40) 2. Goods 0.00 0.00 0.00 0.00 0.00 (0.00) (0.00) (0.00) (0.00) (0.00) 3. Services 0.00 2.11 0.00 0.00 2.11 land project design (0.00) (1.76) (0.00) (0.00) (1.76) consultancies technical assistance 0.00 0.58 0.00 0.00 0.58 (0.00) (0.00) (0.00) (0.00) (0.00) 4. Studies for MoH 1.82 0.00 0.00 0.00 1.82 (1. 1 ) (0.00) (0.00) (0.00) (1. 1 0) 5. Sub loans 0.00 0.00 115.80 0.00 115.80 (0.00) (0.00) (39.50) (0.00) (39.50) Total 1.82 37.95 115.80 0.00 155.57 (1.10) (26.16) (39.50) (0.00) (66.76) The table does not include not bank-finance costs: land acquisition costs, financing costs and other costs Figures in parenthesis are the amounts to be financed by the Bank Loan. All costs include contingencies. 2 Includes civil works and goods to be procured through national shopping, consulting services, services of contracted staff of the project management office, training, technical assistance services, and incremental operating costs related to (i) managing the project, and (ii) re-lending project funds to local govemment units. Project Financing by C oponent (in US$ million equivalent) A. Land Development 66.00 91.46 24.40 36.65 37.0 40.1 B. Loans to private 63.50 77.60 39.50 76.30 62.2 98.3 developers C. Technical Assistance to SNEC 0.50 0.12 0.08 0.02 16.0 16.7 to CIH 0.50 0.12 0.50 0.12 100.0 156.5 100.0 MoH studies 2.00 0.46 1.10 0.72 55.0 156.5 Total 132.50 0.46 169.30 65.58 0.72 113.09 49.5 156.5 66.8 -27 - Annex 3: Economic Costs and Benefits N/A - 28 - Annex 4. Bank Inputs a) Missions: Stage of Project Cycle No. of Persons and Specialty Performance Rating (e.g. 2 Economists, I FMS, etc.) Implementation Development Month/Year Count Specialty Progress Objective Identification/Preparation 02/92 5 Economist/Fin.AnalystlUrban Planner/Procurement Spec./ San.Engineer 03/92 4 EconomistlMun.Engineer/ Fin. AnalystlFin.Spec. 07/92 6 Economist/Urban Planner/ Fin.Analyst/Fin. Specialist / Housing Spec./Lawyer 01/93 6 Economist/Fin.Analyst/ Financial Spec./Housing Spec./ Resettlement Spec. 02/93 1 Fin. Analyst Appraisal/Negotiation 11/92 Economist/Procurement Specialist/Urban Planner/ Social Scientist Supervision 08/93 1 Economist HS HS 02/94 6 Economist/Fin.Analyst/Urban HS HS Planner/Procurement Spec.! Mun.Engineer/Sociologist 07/94 3 Economist/Fin. Analyst / Urban S HS Planner 05/95 4 Economist/Fin. Analysts(2) S HS [Urban Planner 03/96 3 Economist/Social Scientist/ S HS Operations Analyst 08/96 2 EconomistlUrban Planner S HS 03/97 5 Economists(2)/Fin.Analyst/ S HS Urban Planner/Sociologist 06/98 (MTR) 4 Economist/Fin. Analyst/Urban S S Planner/Sociologist 02/99 5 Economist/Fin.Analyst/Sector U S Manager/Urban Planner/ Civil Engineer 11/99 1 Economist S S 02/99 3 Economist/Urban Planner/ S S Civil Engineer ICR 11/00 2 Economist/Urban Planner S - 29 - (I) Staff: Stage of Project Cycle Actua/LaVtest Estimate Identification/Preparation 70.5 228.9 Appraisal/Negotiation 51.4 161.6 Supervision 171.4 623.6 ICR 10.0 20.0 Total 303.3 1,034.1 - 30 - Annex 5. Ratings for Achievement of Objectives/Outputs of Components (H=High, SU=Substantial, M=Modest, N=Negligible, NA=Not Applicable) Rating O Macro policies O H OSUOM O N O NA Z Sector Policies O H OSU*M O N O NA Z Physical OH *SUOM ON ONA O Financial O H OSU*M O N O NA O Institutional Development 0 H * SU O M 0 N 0 NA O Environmental O H OSUOM ON O NA Social F Poverty Reduction O H OSUOM * N O NA OI Gender O H OSUOM * N O NA LI Other (Please specify) O H OSUOM O N O NA O Private sector development 0 H 0 SU * M 0 N 0 NA Z Public sector management 0 H O SUO M 0 N 0 NA El Other (Please specify) OH OSUOM O N O NA - 31 - Annex 6. Ratings of Bank and Borrower Performance (HS=Highly Satisfactory, S=Satisfactory, U=Unsatisfactory, HU=Highly Unsatisfactory) 6.1 Bank performance Rating F Lending OHS OS *U OHU Z Supervision OHS OS *U OHU Z Overall OHS OS * U O HU 6.2 Borrowerperformance Rating • Preparation OHS OS O U O HU Z Government implementation performance O HS O S 0 U 0 HU • Implementation agency performance OHS OS O U O HU 0 Overall OHS OS O U O HU - 32 - Annex 7. List of Supporting Documents The ICR is based on the documents in the project file and the relevant sector work. Early evaluation report was commissioned from a local consultant and was used in some of the analyses. -33 -