Senegal - Sites and Services Project

Document of The World Bank F C FOR OFFICIAL USE ONLY Report No. 4768 PROJECT COMPLETION REPORT SENEGAL SITES AND SERVICES PROJECT (Credit 336-SE) October 31, 1983 Western Africa Region Urban Projects This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  SENEGAL SITES AND SERVICES PROJECT (Credit 336-SE) Principal Abbreviations and Acronyms OHLM - Office des Habitations a Loyer Modere CCCE - Caisse Centrale pour la Cooperation Economique BNDS - Banque Nationale de Developpement du Senegal SICAP - Societe Immobiliere du Cap Vert DPH - Direction de la Promotion Humaine DUH - Direction de l'Urbanisme et de l'Habitat SAC - Section d'Assistance aux Communautes DPA - Direction des Parcelles Assainies IDRC - International Development Research Center GOS - Government of Senegal BHS - Banque de l'Habitat du Senegal  FOR OFFICIAL USE ONLY SENEGAL SITES AND SERVICES PROJECT (CREDIT 336-SE) PROJECT COMPLETION REPORT TABLE OF CONTENTS Page No. Preface . ........................................................ i Basic Data Sheet ................................................ ii Highlights .......... v I. INTRODUCTION A. Preface ........................................... 1 B. Project Completion Report Evaluation Criteria ..... 1 C. Source of Information for this Report .... ......... 2 II. POLICY CONTEXT A. Dakar............. . . * . ... . 3 B. Macroeconomic Environment during the 1970's ....... 4 III. PROJECT IDENTIFICATION PREPARATION AND APPRAISAL A. Origins and Formulation of the Project. .... ........ 5 B. Appraisal . .....................*... .*******s**e* 7 C. Assessment .............. . .. . .. . 8 IV. IMPLEMENTATION A. Effectivenessi and Start-Up: 1972-1975 ............ 10 B. Detailed Design Construction: 1975-1981 .......... 12 C. Progress on Non-Physical Program ..... . .0 .......... 16 D. Occupancy and Consolidation ......... ............. 18 V. INSTITUTIONAL DEVELOPMENT A. Central Govenument Agencies ......... .............. 27 B. The Office des; Habitations a Loyer Modire (OHLM) - 27 C. The Direction des Parcelles Assainies (DPA) ....... 28 D. Assessment ............................................ 29 This document has a restricted disEtribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  VI. COSTS, DISBURSEMENTS, AND FINANCIAL PERFORMANCE A. Costs ........................................................ 31 B. Disbursements ......................................... 32 C. Financial Performance ....... .............. ..... . ..... 33 D. Recovery Performance and Replicability ............... 36 E. Financial Position of DPA and Results ............... . 37 VII. ECONOMIC REEVALUATION ..................... ............. 46 VIII. BANK PERFORMANCE ........49 ANNEX Government PCR .53 ATTACHMENT: Borrower Comments ..................................... 77 CHARTS World Bank 29315: Schedule of Disbursements World Bank 29328: Implementation Schedule at Appraisal World Bank 29329: Actual Implementation Schedule MAPS IBRD 16448 (PCR) IBRD 16449 (PCR) PHOTOGRAPHS  Project Completion Report SENEGAL SITES AND SERVICES PROJECT (Credit 336-SE) PREFACE Credit 336-SE for the Senegal Sites and Services Project for $8 million equivalent was signed on September 29, 1972 and was made effective on August 31, 1973. The Closing Date was extended several times and ultimately set at December 31, 1981. The Credit was fully disbursed except for a cancellation of US$450.000. This report consists of a Project Completion Report (PCR) prepared by the Western Africa Region Urban Projects Department. The PCR was prepared on the basis of the Government's Project Report on the Sites and Services Project, on information contained in the Project Appraisal Report, several reports produced by the Bureau d'Evaluation, a review of Bank files including supervision and progress reports, and interviews and discussion with Senegalese officials and Bank staff. In accordance with the revised procedures for project performance audit reporting, this Project Completion Report was read by the Operations Evaluation Department but the project was not audited by OED staff. The draft Project Completion Report was sent to the Borrower. The reply is reproduced as Attachment.  - ii SENEGAL SITES AND SERVICES PROJECT (Credit 336-SE) Project Completion Report Basic Data Sheet KEY PROJECT DATA Item Appraisal Actual or Expectation Current Estimate Total Project Cost 12.9 14.2 a/ (US$ million) Underrun or Overrun ) 10% / Loan/Credit Amount (US$ million) 8.0 8.0 Disbursed - 7.55 Cancelled .45 _ Repaid to IBRD (US$ million) - - Economic Rate of Return (%) 18 19 a/ As of 06/30/82. Converted a.t average rate of US$1 = CFAF 232.1. b/ The project was scaled down from 400 ha to 300 ha in Dakar and from 60 ha to 40 ha in Thies. The original number cf community facilities was reduced in Dakar and all community facilities were eliminated in. Thies. The upgrading component was also eliminated. There is over a 100% increase in per ha costs of the reduced project c/ As of 06/30/82. d/ Repayment begins 10/82. OTHER PROJECT DATA Item OriE;inal Plan Revisions Actual First Mention in Files - June 1969 Government's Application September 1970 Negotiations May 22-26/1972 May 22-June 2/1972 Board Approval - June 22, 1972 Credit Agreement Date Sept. 29, 1972 Sept. 29, 1972 Effectiveness Date March 01, 1972 Aug. 31, 1973 Closing Date June 30, 1979 Dec. 31, 1980 Dec. 31, 1981 June 30, 1981 Borrower : The Government of the Republic of Senegal Executing Agency : OHLM/DPA OHLM/DPA Fiscal Year of Borrower : July 1 - June 30  - iii - SENEGAL SITES AND SERVICES PROJECT (Credit 336-SE) Mission Data Month No. No. Staff Date of Mission Year Weeks Persons Weeks Report Identification 09/70 1 2 2 10/21/70 Preparation I 03/71 3 2 6 03/25/81 Preparation II 08/71 2 3 6 08/20/71 Preparation III 09/71 2 2 4 11/22/71 Appraisal 12/71 3 6 18 12/28/71 Pre-Negotiations 05/72 1 2 2 05/12/72 TOTAL 12 38 Supervision 1 07/72 1 3 3 07/12/12 Supervision 2 10/72 1 3 3 11/09/72 Supervision 3 12/72 2 2 4 01/05/73 Supervision 4 05/73 1 1 1 06/21/73 Supervision 5 08/73 2 1 2 10/08/73 Supervision 6 09/73 1 1 1 10/01/73 Supervision 7 03/74 2 2 4 04/24/74 Supervision 8 10/74 1 2 2 11/11/74 Supervision 9 01/75 2 1 2 03/18/75 Supervision 10 05/75 2 7 14 06/06/75 Supervision 11 07/75 1 3 3 08/01/75 Supervision 12 10/75 1 3 3 11/19/75 Supervision 13 02/76 1 2 2 03/29/76 Supervision 14 05/76 1 1 1 06/02/76 Supervision 15 10/76 1 3 3 11/09/76 Supervision 16 02/77 1 3 3 03/16/77 Supervision 17 07/77 2 1 2 09/06/77 Supervision 18 09/77 1 2 2 01/13/78 Supervision 19 05/78 1 4 4 06/26/78 Supervision 20 09/78 1 4 4 10/25/78 Supervision 21 02/79 1 2 2 03/22/79 Supervision 22 05/79 1 4 4 07/12/79 Supervision 23 01/80 1 4 4 02/05/80 Supervision 24 05/80 1 1 1 08/15/80 Supervision 25 07/80 2 4 8 07/14/80 Supervision 26 11/80 2 2 4 12/15/80 Supervision 27 02/81 2 2 4 03/13/81 Supervision 28 05/81 2 2 4 06/05/81 Supervision 29 11/82 2 3 6 02/02/82 Supervision 30 03/82 2 4 8 06/30/82 TOTAL 42 108  - iv - SENEGAL SITES AND SERVICES PROJECT (Credit 336-SE) Country Exchange Rate Name of Currency = CFA Franc (CFAF) Year: Exchange Rate Appraisal Year Average US$1 = 255.8 (1972) 1972-1973 242.50 1973-1974 241.75 1974-1975 208.00 1975-1976 220.50 1976-1977 249.00 1977-1978 236.75 1978-1979 218.14 1979-1980 212.22 1980-1981 225.50 1981-1982 297.50 Intervening Year Average US$1 = 232.1  HIGHLIGHTS 1. This Project Conmpletion Report reviews the performance of the Bank's first urban project according to the following criteria: (a) Policy: How fully did the project meet its policy objectives of (i) demonstrating the viability of the sites and services and upgrading approach and (ii) mobilizing private savings for house construction? Did the project contribute to a change in shelter policy as a result? (b) Provision of Basic Services: Were affordable, acceptable services delivered to the population as intended? (c) Institutions: Did the project succeed in creating a capacity to plan and execute sites and services projects over the longer term? (d) The Role of IDA: What has been the effect of IDA investment on the mobilization of public and private funds? and (e) Economic Justification: From an economic point of view, was the project justified in retrospect? 2. The experience of the Senegal Sites and Services Project is one of innovation and experiment. When the project was designed in 1970-71, Governments of developing countries were just beginning to recognize the impact of rapid urbanization. Projections suggested that the accelerating pace of urban growth was likely to continue, and that by the year 2000, would dramatically transform even the most rural of countries. 3. Throughout the 1.960's, Governments did little to prepare for the pressures of this urbanization which had begun to strain services, institutions, and financial resources. While urban policies were sweeping in their intentions, urban practice was, in cc'ntrast, quite limited. The projects which were financed were mostly expensive and heavily subsidized public housing schemes, reaching only a minority of the population. In contrast, the phenomenon of rural migration, which was bringing increasing waves of predominantly poor people to the cities, was largely ignored. 4. Senegal in the 1.960's was already one of the most urbanized countries in Africa, with 23% of its residents living in urban areas. The drought of 1968-74 had a marked effect on the country's predominantly agricultural economy, driving down rurELl incomes and accelerating rural- urban migration. The overall annual urban growth rate went from 3.4% in 1960 to 4% by 1975. Dakar increased its rate to about 6% per annum, well over double that of the country as a whole.  - vi - 5. At the same time, Government sought major investments for the modernization of Dakar. During 1965-1969, as a part of this policy, Government provided about 120 ha of free sites and minimal services on the outskirts of Dakar for low income households who were forced to relocate from the city center to make way for new commercial and high income housing development. This program, which was to be significantly expanded during the 1969-73 Plan period, attracted the attention of the Bank. 6. It was at this time that the Bank began receiving requests for assistance in the development of efficient and equitable solutions to the provision of urban setvices and the management of urban growth. Given the scarcity of public funds, agricultural development priorities, and the mushrooming urban sector, the Bank was especially interested in solutions which provided low cost, affordable and acceptable services to the urban poor. With the Bank's urban lending strategy just being defined, conditions in Senegal seemed to suggest that the environment was appropriate for a demonstration project. 7. The Bank agreed to support the Senegal effort and identified a project which was appraised in 1971. It included the following components: (1) 400 ha of sites and services including community facilities in the capital of Dakar, providing 14,000 plots for 140,000 people and 60 ha of serviced sites in the secondary center of Thias, providing 1,200 plots for 12,000 people; (2) technical assistance to assist in planning for urban growth over the longer term, in preparing detailed design, and in project execution; and (3) a study to determine criteria for upgrading of squatter neighborhoods and funds for a pilot upgrading scheme. The objectives of this project were to: (a) demonstrate an affordable and acceptable alternative to conventional housing solutions, (b) mobilize private savings for the production of shelter and (c) develop an institutional capacity to plan and develop similar projects for other urban areas. Project costs at appraisal were about US$12.9 million of which US$8 million was financed by an IDA Credit. The Project was to be executed over a 5 year period by the Direction des Parcelles Assainies (DPA) to be newly created under the project within the Office des Habitations a Loyer Modere (OHLM), the middle-income public housing agency. Approximately 100 ha were to be developed and settled each year after start-up. 8. Inadequate project preparation contributed to a difficult implementation. This was due primarily to the novelty of the approach and Bank inexperience in the sector. There were serious misunderstandings and misperceptions about project objectives and disagreements over design standards (paras. 3.18 & 4.11).  - vii - Fundamentally, the Government's prime focus was on the provision of subsidized middle income housing, while the Bank aimed to address the majority of the urban population which was overwhelmingly poor and had little or no access to adequate shelter and basic services. The Bank erred in not recognizing more clearly the extent of difference in views (para. 3.17). 9. Inadequate project preparation was dramatically translated by the steady backsliding on design standards. Because design standards were not pinned down at appraisal, agreements were fragile at best. Throughout the decade of the 1970's while real incomes were declining, Government hardly budged on the standards issue, and indeed, imposed unrealistic expectations concerning housing standards on project beneficiaries. Even when it was demonstrated that funds were scarce and standards excessive, Government held to its position and frequently pressed for changes upward. 10. These factors, combined with a shortage of counterpart funds caused by the economic stagnation of the 1970's, an overly large demonstration project, and an inordinately cumbersome bureaucracy repeatedly frustrated project execution. While projects in Senegal generally average a 2-3 year delay in implementation, the Sites and Services IProject is four years behind schedule and house construction is not yet completed. Many mid-course revisions to project design aimed to correct errors and oversights and improve implementation. The project was reduced from 400 ha to 300 ha in Dakar and from 60 ha to 40 ha in Thies. The number of community facilities was reduced in Dakar and construction of community facilities was postponed in Thies. The pilot upgrading scheme was eliminated and Lt was agreed to concentrate on completion of civil works on the Dakar site by December 31, 1981, the third and final Closing Date. However, even with these modifications, the major disagreements over design standards and ensuing delays contributed to over a 100% increase in per ha costs of the reduced project (para. 6.02). 11. The rate of housea construction and occupancy fell far short of appraisal estimates for four main reasons. Firstly, the estimated occupancy and consolidation pace was extremely optimistic. Secondly, there was a delay in completing physical worlks which delayed availability of water for house construction. The third reason was the lack of finance for house construction which was not included in the appraised project. A fourth reason was the unrealistic expectations oi the beneficiaries concerning the type of houses they would build. In 1977, the DPA developed a construction loan program, using downpayments and monthly payments on plot purchases to encourage house construction. As of December 31, 1981, this program had provided over 2,500 loans. Over 4,500 houses were under construction on 10,500 plots in Dakar; 2,000 houses were occupied and over 20,000 people were in residence. About 500 housing starts are undertaken each quarter and it is expected that the Dakar site will be fully occupied by 1986. To further accelerate construction and occupancy, Government lent US$2.5 million to OHLM under the Structural Adjustment Loan (SAL) of 1981. When this loan is fully disbursed, it will provide an additional 1,400 construction loans, bringing the total number of loans for house construction to 3,900. The remaining 8,320 beneficiaries (68%) are financing construction out of their own savings, demonstrating that the project has been successful in mobilizing private resources for house construction.  - viii - 12. The many difficulties and delays led to a proposal to cancel the project in 1976. The decision to continue was taken with considerable reluctance, and indeed, today is a subject of intense discussion. In terms of its regional implications, the decision appears sound. It was the Bank's first project in the urban sector and the first in the Western Africa Region. Several countries in the Region contemplating similar investments have studied and borrowed from the Dakar experience. Overall, the experience of Bank-assisted urban projects shows that the original aims in Senegal were overly ambitious and required major reforms in policy and practice which could not be realistically achieved in the space of one operation. Nevertheless, measurable gains have been made in meeting overall project objectives (para. 8.06). 13. Concerning policy objectives, the results are mixed. Inroads have been made in demonstrating that the provision of serviced sites is an acceptable and affordable solution. The Dakar population has responded favorably, filing over 40,000 requests for plots and continuing to pressure for new projects. Mayors and Governors from cities and regions outside the capital have also requested sites and services programs, and similar projects are being prepared for bilateral funding. However, there is yet to be full acceptance of this policy or its implications by the central ministry and other involved agencies. Although standards for this project are lower than conventional housing schemes, Government continues to advocate unrealistically high standards in isolation of affordability considerations for shelter. Concerning the policy of upgrading, no gains were made in demonstrating it as an alternative to slum removal. This issue is today as politically sensitive as it was in the early 1970's. 14. In the main, the project has contributed to a continuing policy dialogue and a shift towards policy reform. The longer than anticipated implementation period has permitted a continued Bank presence in the sector during a period of broad ranging discussions on macroeconomic issues. These circumstances have facilitated frequent discussions with Government officials on this project as well as other issues facing the sector. One important result is a Bank role in the preparation of a contrat-plan for OhLm, an agreement between Government and OHLM which aims to define its policy objectives, and its operations in relation to financial and institutional possibilities and macroeconomic development priorities (paras. 5.11 - 5.13). 15. As for the provision of affordable services, the project has delivered over 12,000 plots to a population whose access to services was previously limited. While standards are lower in this project than in OHLM's conventional schemes, they could be lower, and must be lowered if these projects are to be replicable. At appraisal, it was estimated that families between the 47th and 85th percentiles would qualify for plots. Selection criteria were modified to enable more (1) lower income families, (2) women and (3) informal sector households to participate. Actual beneficiaries fall between the 20th and 65th percentiles on the 1980 Dakar income distribution curve with the income ceiling at CFAF 55,000. The delays, the decline in real income over the period, and the push towards higher standards have cut into the private savings of those at the lower end of the income range, making it more difficult for these households to build a house. Cost recovery on plot charges is generally good compared to other projects,  - ix - with a 22% default rate, bu.t is poor (88%) on the construction loan program which was started in 1978 and which required additional procedures for eligibility and disbursement. It has been demonstrated that the level of recovery is not related to willingness or ability to pay but to cumbersome procedures and an inadequate institutional framework, a lack of political will to collect, ard effective sanctions (para. 6.19). 16. Concerning institutional development, the establishment of a growing capacity to implement and deliver sites and services projects has been most encouraging, despite the many difficulties and recognized weaknesses in some sections of the DPA. The DPA has developed new procedures to advertise the project, interview and select beneficiaries and recover plot charges. It has organized and supervised a construction loan program and mobilized beneficiaries into effective action groups for settlement. Most importantly, it has delivered over 12,000 serviced plots between 1976-1981 while its parent agency produced just over 2,000 houses for the middle and higher income groups duting the same period. It is currently preparing a sites and services project with German assistance for 3,000 househtolds in the secondary town of Kaolack. The team has had far less turnover i.n staff than either OHLM or its parent ministry, a continuity which should enhance the development of similar projects in the future, provided a suitable! policy framework is in place (paras. 5.07 - 5.10). 17. In contrast to institutional achievements at the project unit level, there have been few gains within the central ministry. Apart from disagreements over design standards, the Directorate of Urbanism in the Ministry of Urban Development contributed little towards project execution. The failure of the Ministry to take a more active role contributed to the isolation of the DPA and reinforced the widely held view that the DPA was "extra-governmental" and had little, if any support. There were also no gains made in creating a capacity to plan for the longer term management of urban growth in the central ministry, although there is probably a greater appreciation of the issues involved than there was at the beginning of the decade (paras. 5.02 & 5.03). 18. IDA participation in the Senegal project has been instrumental in mobilizing other resources to complement and augment its own participation. In 1980-81, the Caisse Centrale assisted Government in funding its counterpart obligations. In addition to the above-mentioned assistance provided by the German Government, other bilateral agencies have expressed interest in financing similar projects in other regions. By 1986, when the project is completed, each IDA dollar invested in. the project will have generated about US$1.80 of public funds. In addition and of greater interest is that each $ will generate about US$8.20 in private investment. The latter figure underscores the viability of sites and services as an alternative to conventional public shelter programs (involving 100% government finance) and answers some of the criticisms concerning its acceptability by the Senegalese population. 19. From an economic point of view, was the project justified in retrospect? The ex-post evaluation supports the view that this project was a sound investment, showing a recalculated rate of return of 19%. From a practical point of view, the learning by doing experience which typifies this project has had significant  -x - payoffs despite the many difficulties. The Government of Senegal has stated that this project is a success, a viewpoint which leaves the door open to a better dialogue in the future. For the World Bank, this project has made a valuable contribution to a better understanding of the design process and has provided important feedback on implementation. Having renewed the commitment to execute this project in 1976, it is essential now to consolidate the gains and strengthen the policy framework. Continuing discussions on the sector, on preparation of the contrat-plan and on preparation of subsequent operations should be the vehicles for achieving these objectives (paras. 7.01 & 7.06).  I. INTRODUCTION A. Preface 1.01 The experience of the Senegal Sites and Services Project is one of innovation and experiment. The approach embodied in this first Bank financed project in the urban sector marked a radical departure from other urban projects financed in the developing world during the 1960's. These projects focussed on the construction of high cost subsidized houses for a minority of the population. In marked contrast, the Senegal project aimed to provide basic, affordable services to the predominantly lower income families moving to the cities in accelerating numbers. 1.02 When the project was designed in 1970-71, few Governments had looked beyond the early warning signs of urban sprawl to the longer run implications of mushrooming urbanization. Fewer still acknowledged the financial burden posed by the generous subsidies accorded to the urban sector in economies overwhelmingly dependent on increased agricultural production. Development policies of the Government of Senegal seemed to suggest that conditions were ripe for change and that Senegal was an appropriate country for a first Bank urban intervention. B. Project Completion Report Evaluation Criteria 1.03 Although this pr:oject is officially closed, it is not yet complete. Some civil works, house construction, and occupancy are still in progress. This Report is therefore only an interim evaluation. Although the project was intended to be a demonstration effort, it was in reality an ambitious experiment to create a new town of 140,000 residents. Providing services, shelter and settling the new residenti; have proven to be much more difficult and to take considerably longer than originally expected. Neither the Government nor the Bank foresaw the range of problems which were encountered during implementation, essentially because neither had any experience in implementing sites and services projects. 1.04 The project was officially closed on December 31, 1981 and disbursements continued until March 31, 1982. As of the latter date, over 90% of the civil works had been completed and all of the plots had been distributed on the Dakar site. Over 20,000 people were in residence on 2,000 plots and over 4,500 plots were under construction. In the secondary center of Thies, about 84% of the civil works are completed. The outstanding civil works and community facilities which were not completed under the Credit will be completed and financed by Government. The required funding has been advanced to the project. 1.05 The length of intplementation has been a subject of considerable discussion inside the Bank.. While projects in Senegal generally average 2-3 years delay, this project is 4 years behind the original implementation schedule. The costs and benefits of our 1976 decision to continue our association in the face of seemingly insoluble difficulties have been argued for many years. This PCR will show that the decision to continue was sound.  - 2 - 1.06 The experience of this project will be evaluated on the following criteria: (1) Policy: How fully did the project meet its policy objectives of: (a) demonstrating the viability of the sites and services and upgrading approaches and (b) mobilizing private savings for house construction. Did the project contribute to a change in Government shelter policy as a result? (2) Provision of Basic Services: Were affordable, acceptable services delivered to the population as intended? (3) Institutions: Did the project succeed in creating a capacity to plan for and execute sites and services projects over the longer term? (4) The Role of IDA: What has been the effect of IDA investment on the mobilization of public and private funds? and (5) Economic Justification: From an economic point of view, was the project justified in retrospect? C. Source of Information for this Report 1.07 This Project Completion Report is based on information contained in the Project Appraisal Report (No. PS-lla of June 7, 1972), several reports produced by the Bureau d'Evaluation, a review of Bank files including supervision and progress reports, and interviews and discussion with Senegalese officials and Bank staff. Although not required under the legal agreements, the Government agreed to prepare a Project Completion Report as part of its own effort to evaluate this project. A draft translation of Government's PCR is attached as Annex 1. Government's report was prepared by a Committee headed by Ousmane Blondin Diop (Secretary General/OHL11), with the assistance of the Ministry of the Plan. It covers the initial phase of the project cycle up to appraisal.  - 3 - II. POLICY CONTEXT A. Dakar 2.01 Since its founding in the early 19th century, Dakar has been a prime magnet for growth and development in West Africa. Until 1960, it was the colonial capital of French West Africa, and, at Independence, was the largest urban agglomeration in francophone Africa. Economic prospects in other newly independent countries somewhat dampened growth in Dakar during the 1960's, but, by that time, Senegal was already one of Africa's most urbanized countries, with 23% of its residents living in urban areas. 2.02 Heavily reliant on groundnuts and livestock, Senegal's economy began to founder towards the end of the decade. The severe drought which struck in 1968 and lasted until 1974 caused agricultural production to slump. As rural incomes plummeted, rural-urban migration accelerated, especially towards Cap Vert, rapidly reversing the previous slowdown in urban growth. The urban share of the total population went from 24% to 28% by 1975. The overall annual urban growth went from 3.38% in 1960 to 4.03% by 1975. These trends represented annual increases of over 100,000 persons, most of whom moved to the Cap Vert Region. Dakar increased its rate to about 6% per annum, well over double that of the country as a whole. By 1975, the Cap Vert Region, which covered 1.3% of the national territory accounted for over 55% of GDP and 19% of the total population. This renewed urban growth increased the demand for shelter, services, and jobs. 2.03 Government began to seek a revitalized, regional role for the capital during this period. Major public investments were programmed and a campaign launched to promote investment from abroad, adding to the attraction of Dakar for those facing declining incomes in the rural sector. Deteriorating conditions in the rural areas and the drive to reassert Dakar's prominence led to a dramatic influx of predominantly poor households in search of shelter, basic services and job opportunities. By 1971, well over 60% of Dakar's population lived in squatter neighborhoods, which for the most part, were located in central Dakar. However, these areas were targeted as prime commercial and residential zones by the Ecochard Master Plan which had been prepared for Dakar just prior to Independence. 2.04 Government opted for a policy of relocation of the urban poor to the city's outskirts to clear a path for the more expensive development projects contemplated for the downtown area. However, this policy ran into an unforseen obstacle: traditional external funding sources of subsidized housing schemes for high income groups began to shift away from lending for housing in Africa. The Government suddenly found itself faced with a considerable housing demand and no sure sources of finance for its middle and upper class clientele, in addition to the rapidly expanding demand of the lower income groups. House construction by the Office des Habitations a Loyer Modere (OHLM), the middle income public housing agency fell sharply, and continued to stagnate throughout the 1970's. The Societe Immobiliere du Cap Vert (SICAP) the upper income housing authority, also began to experience similar difficulties during this period.  - 4 - B. Macroeconomic Environment during the 1970's 2.05 In Senegal, project execution, which is difficult across sectors under the best of circumstances, was made more so by a decade of economic stagnation. Between 1971-1981, real GDP increased by .8% per year, with an annual fall in per capita income of almost 2%. A policy of public sector expansion led briefly to higher growth rates in 1974-75. But, in 1977, terms of trade turned against Senegal, and serious public finance problems began to emerge. During 1978-1981, three crop failures triggered a sharp decline in GDP. While groundnut exports were sharply curtailed, the price of imported oil skyrocketed and high Eurodollar rates aggravated debt service problems. Despite exceptional aid from multilateral and bilateral donors, Senegal could not meet its obligations. 2.06 The Government instituted an economic stabilization program, which included liquidating some important parapublic enterprises and increasing taxes, tariffs for public services, and urban food prices to relieve the cash shortage. This program weighed heavily on the poor whose incomes did not keep pace with inflation. With increased expenditures for food and transport, available income for housing investment declined and private savings for house construction were strained. 2.07 This uncertain environment proved especially hostile to the introduction of sweeping reforms in urban shelter policy. The sites and services project was large for a demonstration project and required a major turnaround in Government's approach to urban development. The absence of a consensus on project objectives and of institutions and procedures for this new style project created inevitable difficulties. Given the length and depth of the country's economic decline, these difficulties are not surprising. What is more surprising is the measurable progress made in reaching some of the original objectives of the project.  - 5 - III. PROJECT IDEINTIFICATION PREPARATION AND APPRAISAL A. Origins and Formulation of the Project 3.01 The origin of th:is project can be traced to an economic mission which visited Senegal in Jtne 1969 and recommended that the Bank look into Government policies for planning and controlling urban growth. The following year, an urban reconnaissance mission from the Special Projects Department expressed interest in the (Government's evolving sites and services approach and recommended Bank support of its effort. Government's program was essentially a relocation project which provided free sites and minimal services on the outskirts of Dakar for households displaced from central zones. 3.02 The Government began implementation of its program during the Second Four Year Plan (1965-1969), developing about 120 ha of minimally serviced plots in Pikine, a large squatter neighborhood on Dakar's periphery. The Third Four-Year Plan (]969-1973) was more ambitious, providing for 350 ha in Dakar and 200 ha in selected regional centers. The Bank was encouraged by discussions with Government: on this program which seemed to suggest a commitment to a lower standard solution to shelter and services and appeared to reflect emerging Bank policy. It also appeared that the issue of relocation could be satisfactorily resolved. Tn Dakar, the program was to accommodate 100,000 of the expected population increase over the Plan period, the remaining 40% to be housed through existing public sector programs built by OHLM and SICAP and the private sector. The mission decided to support this program. 3.03 In January, 1971, a memorandum was put to the Loan Committee on Bank participation in urban development. Two major issues were raised: (1) should the Bank finance housing? and (2) was Senegal the right place to start? There appeared to be a consensus that the Bank be involved. The benefits centered around enmployment generation, the mobilization of private savings for shelter, and cost savings which would result from the efficient planning and management of development in fast growing urban areas. Concern was expressed about the extent of the housing demand in developing countries and the risk that the Bank would be overwhelmed with requests for which it had neither the funds nor the staff. It was generally agreed that the Bank should not finance housing superstructures, but should limit itself to the provision of basic infrastructure and services. 3.04 There was no clear position as to the choice of Dakar for an initial operation. Discussions held with USAID, the UN, and other agencies and private individuals revealed few project proposals of a similar nature which were further advanced than that of Senegal. It was recommended that the Bank inform the Government of Senegal that, in principle, it was willing to consider possible financing of its program provided the project could be economically justified. 3.05 Following the decision of the Loan Committee, a preparation mission visited Senegal in March, 1971, and identified a project providing for 400 ha of sites and services in Dakar, and 60 ha each in Kaolack and Thies. The objectives were to accommodate the population growth from 1969-1973 in Dakar and to support Government's regional development strategy which aimed to siphon off some of the population flows headed towards the Cap Vert.  - 6 - 3.06 The Bank spearheaded the preparation process, outlining the project scope and the necessary urban studies and specifying in detail such items as minimum service levels and plot layouts. Several consultants within the Ministry of Public Works and Urban Development were to be responsible for bringing the project to the appraisal stage. Parallel to the technical preparation, the Bank was concerned about several issues which were directly related to its emerging urbanization policy. These included: (i) justification of an urban intervention in a predominantly agricultural country, (ii) definition of the project population and (iii) land tenure mechanisms. These issues were the subject of frequent and often heated discussions between Government and the Bank. 3.07 An urban project in a predominantly agricultural environment. The course of Bank/Government dialqgue was heavily influenced by the fact that the project was to be the Bank's first in the urban sector, and would be undertaken in a country facing pressing problems in the rural areas, while, at the same time, urban expenditures were growing rapidly. These policy contradictions were of special concern to the Bank. 3.08 Considerable emphasis was therefore placed on the policy context and pressures were put on Government to commit itself to Ca) eliminate subsidies to SICAP frDm the start of the Fourth Plan, (b) reduce the number of "economic" 1/ units to be constructed by OHLM, (c) progressively eliminate subsidies on these units over 10 years and (d) restrict subsidies on OHLM's "super economic" 2/ units to the cost of land and infrastructure. Government was also urged to abandon its policy of slum eradication in favor of a slum upgrading program. Government was reluctant to commit itself initially but after much discussion and intense Bank pressure it finally agreed (see para. 3.15). 3.09 Definition of the Prolect Dopulation. This issue centered around the overwhelming needs of the majority of the urban population in contrast to Government's programs which were exclusively directed to the middle and upper income groups. Between 840 and 1,100 ha had been reserved for higher income housing while only 300 ha were reserved for sites and services for the lower income groups. The Bank pointed out that the area allocated to the higher income groups was adequate to accommodate this growth for 15-20 years, while the sites and services zones would be absorbed within a three year period. The Government countered that the concentration of 140,000 lower income households on one 400 ha site was socio-economically undesirable as well as a source of potential political difficulties. The Bank did not disagree but insisted nevertheless that the relative needs of the lower income groups were more pressing and should receive priority attention in this project. Government eventually agreed to 1/ Conventional heavily subsidized OHLM units. 2/ Lower cost unit to be introduced as a result of project.  this solution, although as Government's PCR clearly points out, there were many in Government who disagreed with the project formulation. 3.10 Land tenure arrangements. In keeping with a 1964 Government law, beneficiaries of plots in public housing schemes in Senegal were given occupancy rights rather than freehold. However, the scant information on sites and services led the Bank to conclude that ownership was a necessary incentive to private investment in shelter. The Bank consequently insisted that freehold be granted on the site. The Government eventually agreed to this proposal although it ran counter to both the legal context and the country's political philosophy. B. Appraisal 3.11 During the months prior to appraisal, several missions commented on the slow progress in preparation of preliminary engineering, It had been agreed that technical staff in OHLM would complete the required work, but little was done. A continuing question about the final location of the project site further slowed preparation. It was, nevertheless, decided to appraise the project in November-December 1971 with the team prepared to complete the bulk of the work. 3.12 The project was appraised in November-December 1971 and included the following components: (1) 400 ha of sites and services in Dakar, providing 14,000 plots for a total population of 140,000 people. Services included roads, water supply, sewerage by septic tanks and pit latrines, public toilets and community facilities such as schools health clinics and community centers. Households were expected to pay for plots at 7% per annum over 15 years. Individuals were expected to finance house construction out of their own savings. (2) 60 ha of sites and services similar to Dakar in the secondary town of Thies, about 100 km from Dakar providing 1,200 plots for 12,000 people. (3) Technical assistance to assist OHLM for preparation of detailed engineering, and in project execution and (4) a study to determine criteria to be used for the upgrading of the remaining slum neighborhoods in Dakar as well as funds for a pilot upgrading scheme. 3.13 Total project costs were about US$12.9 million of which US$8 million was to be financed by an IDA Credit to Government. The project was to be executed over a 5 year period by the Direction des Parcelles Assainies (DPA) to be created within OHLM. Approximately 100 ha were to be developed and settled each year after start-up. The first phase was to begin in mid-1973, with infrastructure works completed by late 1973. Occupation of the 1st Tranche  - 8 - (100 ha) was to begin in 1973 and be completed by 1974. Infrastructure for the entire project was expected to be completed by late 1976 and all sites allocated by end 1977. 3.14 Arrangements for implementation also specified that OLM was to be reorganized into two separate Directorates--a Housing Directorate to execute OHLM's conventional programs and a Sites and Services Directorate to carry out the IDA project. The Direction de la Promotion Humaine (DPH), then in the Ministry of Youth and Sports, was to manage and staff the Community Assistance Section to be created within the Sites and Services Directorate. A Monitoring Committee headed by the OHLM Director was to be established to assess progress. Conditions of effectiveness required: changes in the building codes, assignment of Senegalese counterpart staff to work with the expatriate technical assistance, approval of legal statutes governing the establishment of the Association of Plotholders and appointment of consultants for the criteria study. 3.15 Concerned about the project context and the macroeconomic climate in Senegal, the Loan Committee attached a number of conditions to the approval of the Credit. These were: (1) a reduction in conventional subsidized public housing in the Fourth Plan; (2) postponment of further demolition of slum housing until the conclusions of the criteria study were reviewed;and (3) establishment of the DPA. C. Assessment 3.16 Two elements characterize this period. The first one is the inadequate state of project preparation. The most significant example is the status of preliminary engineering. In the first instance, there appears to have been little or no association of Senegalese staff in its preparation either because of the lack of qualified staff or the time constraints of the appraisal process. Secondly, the issue of standards and its relation to affordability was not clearly defined with Government. Thirdly, the question of power supply was left untreated, and off-site requirements were only summarily addressed since it was assumed that Senegal would finance 100% of the required networks. This in itself was a serious error given the critical cash situation of the country at that time. Finally, the failure to have a portion of detailed engineering ready by Board presentation further weakened the already loose understanding on standards and was a built-in source of delay. This oversight, however, can be attributed to Bank inexperience, which was corrected in subsequent operations. 3.17 The conflicting and contradictory policy context is the second factor and reflects the serious misunderstandings and misperceptions concerning project objectives. Government policy and practice were oriented towards modern commercial development and the support to middle and upper income residents. Government's advocacy of a lower income solution was an expedient  - 9 - means of freeing up valuable downtown properties. Given the extensive debates which took place during the preparation phase, the failure to take note of the early warning signs is puzzling. These diverging policy views persisted throughout project execution and contributed significantly to the subsequent delays and cost overruns. 3.18 The Bank had very farsighted objectives. Given its emerging philosophy, its insistence on a sound policy framework was not difficult to understand. Yet the Bank sought major and pervasive policy changes which could not be realistically accomplished in the space of one operation. The impact of many of these changes was not fully understood by many in Government who were associated only peripherally in the preparation process but who had major roles to play in implementation. Other changes such as the introduction of slum upgrading were simply unacceptable.  - 10 - IV. IMPLEMENTATION 4.01 The implementation period can be divided into two distinct phases. The first period covers 1972-1975--effectiveness and start-up. The second phase spans 1976-1981 and covers the execution of civil works as well as consolidation and occupancy. Table 4.1 compares the appraisal schedule with actual implementation and is discussed in detail in the following paragraphs. The Credit was approved in June 1972. The implementation schedule at appraisal was based on an immediate start-up of detailed design, i.e., once approved in July 1972. However, because of insufficient IDA funds, Credit signature was delayed until September 1972. Effectiveness was scheduled for March 1, 1973, but was delayed until August 31, 1973 because of problems in fulfilling conditions of effectiveness. The Appraisal Report called for the consultants' arrival "soon after Credit approval". They arrived in April, 1973, nearly a year behind schedule. Therefore, from the beginning there was a delay of one year. 4.02 The project was expected to be completed by June 30, 1978 and to close six months later. Three extensions of the Closing Date to December 31, 1981 were required to complete civil works on the Dakar site, although by December 31, 1981, the civil works were not fully completed. The primary reason behind the extensions of the Closing Date was to bring the level of services on the project site up to that required for habitation under minimal sanitary conditions. This was thought to be a necessary incentive to spur occupancy and consolidation. A. Effectiveness and Start-up 1972-1975 4.03 Effectiveness was originally scheduled for March, 1973 but was delayed until August, 1973 because of numerous complicated and ambitious conditions of effectiveness. First, the Loan Committee conditions concerning the sweeping policy changes required a Government side letter that was delaved. Second, Government was required to pass three decrees: (i) establishing a new building code, permitting the use of low-cost materials; (ii) granting legal status to the sales contract to be used between OHLM and project beneficiaries; and (iii) establishing the Association of Plotholders. Unanticipated administrative and procedural complications blocked the process, and passage ultimately required a "procedure d'urgence" in May-June, 1973 to pass all three. 4.04 Third, a reorganization of OHLM also proved more complex than anticipated. Both the Government and the World Bank believed that the reorganization could be carried out through a Government decree, while the legal advisor to the President ruled that such a reorganization could only be accomplished through passage of a law by the Assemblee Nationale. Because of intervening elections, the law was not approved until June 12, 1973.  - 11 - Table 4.1 APPRAISAL ACTUAL Duration Duration Start Completion Months Start Completion Months DAKAR 1st Tranche Studies Oct. 72 March 73 6 April 73 Dec. 73 9 Constr. Civil Works Oct. 73 March 73 6 July 75 March 79 54 Community Facilities Apr. 74 March 75 11 Jan. 76 Site Occupation Apr. 74 March 75 11 March 78 incomplete 2nd Tranche Studies Oct. 72 March 73 6 July 73 March 74 9 Constr. Civil Works Oct. 74 March 75 6 Jan. 76 Dec. 81 72 Community Facilities Apr. 75 March 76 11 March 79 - Site Occupation Apr. 75 March 76 11 Oct. 79 incomplete 3rd Tranche Studies Oct. 74 March 75 6 March 74 Dec. 74 9 Constr. Civil Works Oct. 75 March 76 6 July 76 Dec. 81 66 Community Facilities Apr. 76 March 77 11 - - Site Occupation Apr. 76 March 77 11 July 79 incomplete THIES 1st Tranche Studies Jan. 73 March 73 3 July 78 March 80 21 Constr. Civil Works Jan. 75 March 75 3 April 80 Dec. 81 21 Community Facilities ApriL 75 Sept. 75 6 - - - Site Occupation ApriL 76 Sept. 75 6 not started TECHNICAL ASSISTANCE Project Manager April 73 March 77 48