Senegal - Urban Management and Rehabilitation Technical Assistance Project

Document of The World Bank FOR OFFICLL USiE ONLY Rqpt No. 11830 PERFORMANCE AUDIT REPORT SENEGAL URBAN MANAGEMENT AND REHABILITATION PROJECT (CREDITS 1458-SE AND SF-13) APRIL 26, 1993 Operations Evaluation Department 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.  EXCHANGE RATES (Currency Unit: CFA Franc) Annual Averages 1984 1985 1986 1987 1988 1989 1990 1991* 437 449 346 301 298 319 272 284 GLOSSARY OF ACRONYMS BC Traffic Bureau ("Bureau de Circulation'] BHS Senegal Housing Bank ["Banque de l'Habitat du S6negal"] CIDA Canadian International Development Agency CUD Greater Dakar Authority ["Communaut6 Urbaine de Dakar"] DCA Development Credit Agreement DCL Local Government Department (of MOI) ("Direction des Collectivites Locales'] DPA Serviced Lot Directorate (under Sites and Services Project) ("Direction des Parcelles Assainiesn] DPU Directorate of Urban Management and Rehabilitation Project ["Direction du Projet de Gestion et de D6veloppement Urbaine"] FECL Municipal Infrastructure Fund ['Fonds d'Equipement des Collectivites Locales"] GOS Government of Senegal GTZ German Cooperation Agency ["Gesellachaft fur Technische Zusammenarbeit"] IDA International Development Association MEFP Ministry of Economy, Finance, and Planning ("Ministere de l'Economie et des Finances et du Plan"] MEQ Ministry of Infrastructure ("Ministere de l'Equipement"] mOI Ministry of the Interior l"Ministere de l'Int6rieure"] MUE Ministry of Urban Development and Housing ("Minist4re de l'Urbanisme et de l'Habitat"] OHLM Low-cost rental housing agency ("Office des Habitations & Loyer Mod6r6"] PAR Performance Audit Report of OED SCAT-URBAM Central Bureau for Urban Land Development ["Socift4 Centrale d'Amenagement des Terrains Urbains") SNHLM National Low-cost Rental Housing Company (formerly OHLM) ("Socift6 Nationale d'Habitations & Loyers Moder6s"] TA Technical Assistance USAID United States Agency for International Development Fiscal Year of the Borrower July 01 - June 30  FOR OFFICIAL USE ONLY THE WORLD BANK Washington, D.C. 20433 U.S.A. Office of Director-General Operations Evaluation April 26, 1993 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Performance Audit Report on Senegal Urban Management and Rehabilitation Project (Credit Nos. 1458-SE and SF-13) Attached is a copy of the report entitled "Performance Audit Report on Senegal - Urban Management and Rehabilitation Project (Credits 1458-SE and SF- 13)" prepared by the Operations Evaluation Department. Senegal was the Bank's first borrower for an urban development operation i.e. a pioneering Sites and Services Project that was successfully completed in 1981. After approval of this first project, there was a hiatus of more than ten years in urban lending for Senegal, broken only by the project audited here. The audited project produced much important information relative to development of the urban sector in Senegal; strengthening and rehabilitating the operations of existing institutions. Its timing was opportune as well, as it provided critical inputs to the Seventh Plan of the GOS. Some of the studies, however, had a strong theoretical bias and more dissemination is necessary to make full use of them. The project highlights the potential role of a technical assistance operation in sustaining sector dialogue and supporting the preparation of future investments. Overall, the Audit concludes that the project's achievements were satisfactory, its institutional impact is rated as partial, and the sustainability of project benefits as uncertain. Project's focus on Dakar alone was, at the time, justified. The project led directly to a follow-on operation, the Municipal and Housing Development Project (Cr. 1884-SE), approved on March 15, 1988, and now being implemented. This document has a restricted distribution and my be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without WorLd Bank authorization.   PERFORMANCE AUDIT REPORT SENEGAL URBAN MANAGEMENT AND REHABILITATION PROJECT (CREDITS 1458-SE AND SF-13) TABLE OF CONTENTS PREFACE ................................................................ i BASIC DATA SHEET ....................................................... ii EVALUATION SUMMARY ........ ............................................. iv I. BACKGROUND.. 1 A. Introduction. 1 B. Macroeconomic Context. 1 C. The Urban Sector. 2 D. Bank Involvement. 3 II. THE PROJECT ............................................ 5 A. Stand-alone Technical Assistance Projects ........... 5 B. Issues in Project Preparation .................... 7 C. Project Appraisal .......................................... . 8 D. Project Objectives and Description ............................ 9 III. PROJECT IMPLEMENTATION AND RESULTS ...... ......................... 15 A. Scheduling .................................................... 15 B. Documents Produced ............................................ 15 C. Project Management ............................................ 16 D. Cost Accounting ............................................... 16 E. Results of Specific Components ................................ 18 1. Urban Investment Planning and Programming ............... ... 18 2. Housing and Urban Policy ...... ............................. 19 3. Municipal Policy .. . . .......... 20 P. Sustainability ................................................... 21 IV. OTHER POINTS OF INTEREST .............................................. 22 A. Delegation of Project Preparation and Implementation Responsibilities ................. .......................... 22 B. TA Operations to Sustain a Flagging Lending Program .......... . 22 C. The Need to Disseminate TA Results ............................ 23 V. BORROWER AND IDA PERFORMANCE ..24 A. Borrower and Executing Agency Experience .24 B. Role of IDA ..24 VI. CONCLUSIONS AND LESSONS LEARNED .................................... 25 A. Conclusions andRecoendation. .................... 25 B. Lessons and Recommendations ..................................... 25   i PERFORMANCE AUDIT REPORT SENEGAL URBAN MANAGEMENT AND REHABILITATION PROJECT (CREDITS 1458-SE AND SF-13) PREFACE 1. This document is a Performance Audit Report (PAR) on the Senegal: Urban Management and Rehabilitation Project, a technical assistance (TA) operation, which was approved by the Board of Directors on March 17, 1984, and partly financed by two credits, 1448-SE, in the amount of SDR 2.9 million and SF- 13-SE in the amount of SDR 2.8 million. Both were fully disbursed. 2. The PAR consists of an Evaluation Summary and a Performance Audit Report prepared by the Operations Evaluation Department (OED). A Project Completion Report (PCR) prepared by the Infrastructure Operations Division of the Sahelian Department of the Africa Regional Office was previously submitted to the Board as Report No. 10178, dated December 5, 1991. 3. The PCR provides an adequate account of the project experience, its successes and its failures. In order to give a wider and more recent perspective, the Audit provides additional information about the project, its eventual outcome, and the principal lessons learned. The PAR is based on the PCR, the President's Report, project legal documents, project files, legal files, a transcript of the Executive Directors' discussions at Board Presentation, and other relevant material. It also draws on field observations and on interviews with Bank operational staff and with Senegalese Government officials in Dakar during the audit mission of June 1992. 4. The Audit endorses PCR findings in most respects, and expands upon the lessons learned from the operation. It gives particular emphasis to the potential role of technical assistance projects in sustaining sector dialogue and supporting the preparation of new investment operations. It also highlights the need for the Government of Senegal, with IDA's support, to improve dissemination of the studies financed under the project and to give greater attention to training for urban management. 5. Following standard OED procedures, copies of the draft PAR were sent to the Borrower for comments on February 8, 1993, however, no co mnents were received.   ii PERFORMANCE AUDIT REPORT SENEGAL URBAN MANAGEMENT AND REHABILITATION PROJECT (CREDITS 1458-SE AND SF-13) BASIC DATA SHEET KEY PROJECT DATA Actual or Actual as x Appraisal Current of Appraisal Item Expectation Estimate Estimate Total Project Cost (US$ m) 6.65 8.33 125.2 Credit Amount (USS m) 6.00 7.49 124.8 Economic Rate of Return N/A N/A N/A Cofinancing -Total (US$ m) 0.213 CIDA 0.213 CUMULATIVE ESTIMATED AND ACTUAL DISBURSEMENTS Appraisal Estimate (US$M) 6.00 Actual (US$M) 7.49 Actual as Z of Appraisal (Z) 124.8 Date of Final Disbursement: December 31, 1989 PROJECT DATES Original Actual Identification 1982 02/83 Preparation - 03/83 Appraisal 7/83 10/83 Negotiation 11/83 03/84 Board Approval 01/84 4/17/84 Credit Signature 03/84 06/20/84 Credit Effectiveness 07/84 11/19/84 Project Completon 12/31/89 06130/90 Credit Closing 12/31/87 12/31/89  iii STAFF INPUTS (staffweeks) FY73 FY74 FY76 FY80 Fr81 FY83 FY84 FY85 FY86 FY87 FY88 FY89 FY90 FY91 FY92 Totat Preappraisal - - - - 1.1 10.9 - - - - - - - - 11.9 Appraisal - - - - - - 26.7 - - - - - - - - 26.7 Negotiation - - - - - - 3.7 - - - - - - - - 3.7 Supervision - - - - - - 1.1 12.4 11.7 5.2 4.3 3.2 7.9 12.6 - 58.3 Other .1 .6 - .7 .1 .9 6.8 .3 - - - - - 9.5 Li As per OED InstitutionaL Data Base/Reporting System MISSION DATA Date No. of No. of Month/Year Days Persons Identification 01/83 5 1 Preparation 03/83 5 2 Appraisal 07/83 10 2 Negotiation 03/84 5 2 Supervision 1 06/85 10 2 Supervision 2 01/86 10 2 Supervision 3 05/86 10 2 Supervision 4 01/87 7 2 Supervision 5 05/87 5 1 Supervision 6 06/87 5 1 Supervision 7 11/87 5 1 Supervision 8 04/88 7 1 Supervision 9 06/88 10 2 Supervision 10 11/88 5 2 Supervision 11 10/89 7 1 Supervision 12 02/90 10 2 Supervision 13 06/90 5 1 Supervision 14 10/90 5 1 Total 126 OTHER PROJECT DATA Borrower: Republic of Senegal Executing Agencies: Ministry of Planning and Cooperation  iv PERFORMANCE AUDIT REPORT SENEGAL URBAN MANAGEMENT AND REHABILITATION PROJECT (CREDITS 1458-SE AND SP-13) EVALUATION SUMMARY Introduction 1. As an active borrower in many urbanized area has already spread sectors, Senegal has been a preferred into neighboring municipal World Bank client. It was the Bank's jurisdictions. A Greater Dakar first borrower for an urban Authority (CUD) was created in 1983 development operation. After the to coordinate local affairs for the pioneering Sites and Services three affected municipalities. Project, a hiatus of more than a Institutional arrangements in the decade in urban lending, was broken urban sector in Senegal are complex, by the operation reviewed here, the directly involving at least three "TA Project" (para. 1.01). ministries and a number of specialized agencies as well as the 2. The tertiary sector, which is municipalities themselves. The TA mainly urban, is responsible for two- Project took this complexity as a thirds of Senegal's GDP. Despite the parameter rather than as a policy economic growth recorded for the variable over which it might have 1984-90 period when the project was exercised some influence (paras. being implemented, it is unlikely 1.05-1.08). that living conditions improved significantly for the urban 4. The first urban development population. Much of the increase in operation, was the first Bank Sites the US dollar value of GDP per capita and Services Project worldwide. An was due to the rapid appreciation of experiment, this project planned to the local currency, the CPA franc. deliver serviced lots to 140,000 This trend has eroded Senegal's people, creating what would become international competitiveness. the country's second largest urban Export promotion, in which the settlement near Dakar. Many problems capital city, Dakar, has an important were experienced during role to play, continues to be a key implementation, but an important feature of macroeconomic policy achievement was the general (paras. 1.02-1.04). acceptance of the basic philosophy of trying to reach low-income 3. In 1990, 2.8 million people beneficiaries with affordable housing lived in urban areas in Senegal, solutions. The main conclusion of about 38 percent of the country's the corresponding PCR, was that the population. Dakar, with 1.5 million operation was too ambitious in scope. inhabitants, was home to more than After completion of this operation, half the urban population. Dakar's the TA Project revived what had population is growing at more than become a flagging urban sector 4.0 percent per annum, and the dialogue, eventually leading to the  V follow-on Municipal and Housing the sites and services approach. Development Project. (Cr 1884-SE), Initiated largely as a stopgap currently under execution (paras. measure to put life back into an 1.09-1.14). ailing lending program, the technical assistance project was nevertheless Project Oblectives and Description expected to achieve great things on its own (paras. 2.05-2.06). 5. The present project was a stand-alone technical assistance 7. Project preparation was operation aimed at providing entrusted to foreign consultants. institutional support to Senegal's This arrangement was justified by urban sector. OED's study of free- Bank staff on the grounds that standing technical assistance for technical capacity in Senegal was institutional development in sub- insufficient to prepare an operation Saharan Africa (Report No. 8573, of the magnitude foreseen. GOS April 1990) assessed nineteen such officials gave a different opinion, operations, finding only five of them namely that the consultants' to have been successful. Among the participation was necessary to ensure reasons for failure were the lack of IDA approval of the proposed project. an institution building strategy and The consultants' preeminent role the limited absorptive capacity of ensured rapid project preparation. borrower institutions targeted for On the other hand, it alienated many improvement. Important GOS officials from an operation they recommendations included: (i) full did not regard as their own. Also, participation of target institutions lack of familiarity with Senegal's in project preparation; (ii) studies urban sector limited the relevance of should be undertaken only when some proposals, notably for training. clearly needed and in full Attempts were made to arrange co- collaboration with target financing for the project, but institutions; (iii) local consultants without success (paras. 2.07-2.08). should be used where possible. Experience elsewhere has demonstrated 8. Project appraisal was that technical assistance is more straightforward, involving a review effective when provided in of the consultants' preparation conjunction with physical report. The appraisal mission was investments. Projects that combine opportune, occurring at a time when both can reward institutional GOS needed urban sector inputs for progress with funding for popular the Seventh Plan. The appraisal investments, something that a stand- mission was strongly impressed by the alone technical assistance operation consultants' technical arguments, cannot offer. The experience of the neglecting a more direct policy present project was assessed in light dialogue with the Borrower itself. of these broader conclusions and The project's complexity exposed the lessons (paras. 2.01-2.04). inadequacy of project documentation, especially the absence of an 6. With completion of the appraisal report. The main technical pioneering Sites and Services Project document, the President's Report, was in 1981, the Bank found itself well prepared, but was too brief to without an urban lending program in be of much help to project Senegal. The preparation of a TA management, e.g. the technical project was an appropriate tactic in information on establishing a Traffic the absence of a clear alternative to Bureau (a US$1.9 million component)  vi was summarized in just ten lines alternatives for Land Development; (paras. 2.09-2.12). and pilot designs for above; and 9. The project's short-term (C) Municipal Policy, viz municipal objectives were: (i) to improve the organization and management; urban allocation of scarce public resources maintenance; traffic management in urban areas through better (creation of a Traffic Bureau); and investment planning and programing; training program. (ii) to formulate a housing and urban policy based on an appropriate 11. Despite the many studies allocation of resources; and (iii) to proposed, important aspects of urban maintain, with a view to ultimately management, such as basic sanitation improving, the existing level of and solid waste, were not covered. urban services in Dakar by enhancing The choice of components seems to its capacity to plan and program have been determined by a combination rehabilitation and maintenance works of GOS priorities, the consultants' and to mobilize resources. As skills, and Bank staff interests. formulated, project goals thus held The training component was a out the promise of spectacular surprisingly small one in the results without, however, providing circumstances (paras. 2.16-2.18). the means to achieve them. More modest operational aims could have ImDlementation ExDerience led to quantitative performance targets and been helpful to project 12. The project was completed two evaluation and supervision (paras. and a half years behind schedule. 2.13-2.15). Financial and technical weakness of the target institutions and poor 10. At appraisal, the project logistics were cited by the PCR as included some 20 separate tasks, reasons for the delays. Even so, grouped into three parts: Urban discussions between GOS and IDA Investment Planning and Programming concerning preparation of a follow-on (US$ 1,688,000 or 25.1 percent of project started during the project's total costs); Housing and Urban second supervision mission (paras. Policy (US$ 1,625,000 or 24.2 percent 3.01-3.02). of the total); and Municipal Policy (US$ 3,342,000 or 49.8 percent of 13. The project produced some total costs). Subcomponents were as eighty reports, a substantial output. follows: A well-organized documentation center at MEFP's Project Unit appears to (A) Urban Investment Plannina and have a complete set of these reports. Programminz, viz. steering committee Among target institutions in the (project management); and Seventh urban sector, however, project Plan (investment appraisal studies were not well disseminated. guidelines, updates); The project's US$8.3 million investment was intended to produce (B) Housina and Urban Policy, viz. more than just a library. To be Housing Finance (affordability effectively useful as management analysis, resource mobilization); instruments for urban sector management of state-owned properties; institutions, the Audit recommends rehabilitation of OHLM (the low-cost bibliographical work and housing agency); creation of a Land dissemination of this output (paras. Development Agency; design 3.03-3.05).  vii 14. Management of a project made up Results of some twenty different activities involving seventeen different 17. According to the PCR, the Urban agencies was a daunting task. At Investment and Programming component appraisal, the consultants were to was one of the most successful. advise an inter-ministerial Steering However, the audit found a strong Committee on its management. This theoretical bias in the reports. arrangement did not work out, Little attention was given to however, since the Committee met only surveying existing work and twice and project coordination knowledge. Authors justifiably disintegrated. Each subcomponent was complained about poor urban sector implemented as an isolated activity data availability in Senegal. Among (para. 3.06). the lessons learned from this experience is that future operations 15. Given the institutional should support institution building complexity of the urban sector in to develop data bases and permanent Senegal, the Steering Committee data collection mechanisms upstream approach to project management was of expensive consultants studies the only practical solution. Clearer (paras. 3.10-3.12). responsibilities and lines of authority would nevertheless have 18. Important studies were produced allowed the borrower to manage the under the Housing and Urban Policy project more efficiently. In the component. One produced an extensive difficult circumstances, the MEFP did and valuable listing of "reference a good job in seeing project prices" for house construction. The implementation through to the end. OHLM rehabilitation study, however, GOS' ability to induce changes in the had little impact on the agency's urban sector was reduced, however, operations, or on those of its after it relinquished some of its successor, SNHLM. Management of BHS, authority as policy-maker during in turn, was not familiar with the project preparation (paras. 5.01- findings of project studies relevant 5.04). to its operations. On the other hand, a small but practical study on 16. Project costs at completion GOS management of state-owned were US$8.3 million, compared with an properties brought significant appraisal estimate of US$6.7 million. potential savings to the Government. The increase was due to the The establishment of SCAT-URBAK, GOS availability of surplus funds as the new urban land development agency, US dollar devalued in terms of the was a successful institution building project's unit of account. Although achievement of the TA Project (paras. the Borrower maintained separate 3.13-3.17). project accounts, these do not allow precise identification of the costs 19. The Municipal Policy component of each component. Management risked was by far the largest, accounting losing control over costs, and did for half of project costs. The not learn how much should be paid for successful establishment of the particular components. This was a Traffic Bureau was a direct outcome disappointing result for a project of this component. Well-elaborated aimed at improving the management transportation studies also helped (including cost accounting) of other the Traffic Bureau's operations, agencies (paras. 3.07-3.09). which were slow to get off the ground. The study to make  viii recommendations concerning the operations, however, its formal Municipal Infrastructure Fund, FECL objectives were unrealistic. They did not appear to have much impact promised achievements of improved upon that entity. On the other hand, resource mobilization and allocation studies on maintenance were more in the urban sector that the project widely disseminated and have helped provided no means (beyond studies) to operations to a limited extent. attain. The Audit agrees with the Actual financial performance of the PCR that the expected results of the municipalities, which have recently operation were out of proportion to suffered sharp declines in revenues, the resources deployed. was a disappointing result for the project, which had hoped to 23. The project demonstrated the strengthen them (paras. 3.18-3.22). advantages and drawbacks of a Borrower delegating responsibilities 20. The Training component was very for project preparation and small, accounting for less than one implementation to foreign percent of project costs, and the consultants. The advantages for GOS quality of the materials produced was and the Bank included rapid poor. The absence of a training processing and greater chance of program based on an analysis of approval. The main drawback was the specific training needs in Senegal's alienation of the executing agencies urban sector was a serious flaw in from an operation which they did not project design (para. 3.23). consider to be their own. Institution building is likely to be Sustainabilitv unsuccessful in these circumstances. To succeed, Bank sector policy 21. Sustainability of the gains dialogue should be directly with from a TA operation depend on a policy-makers, not their agents. number of factors, including the dissemination of results to the 24. The project also illustrates target institutions, their how a TA operation can be used to willingness to implement revive a flagging urban lending recommendations, and the permanence program. The project not only of the institutions themselves. In renewed IDA's urban sector dialogue the present case, the necessary in Senegal, but led directly to a conditions for sustainability exist, follow-on operation, the Municipal but it is too early to draw and Housing Development Project (Cr. conclusions (para. 3.24). 1884-SE). If the aim of preparing a follow-on operation had been Conclusions explicitly acknowledged in project objectives, a more focussed, simpler, 22. The project produced much and less expensive operation might important information in relation to have resulted. Another missed Senegal's urban sector. Its focus on opportunity was the lack of effective Dakar, where most of the country's dissemination of TA Project outputs. GDP is produced and more than half of Bibliographical work should be done the urban population lives, was urgently before the information appropriate for an operation assembled under the project becomes ostensibly aimed at contributing to out of date. The Audit rates the improvements in the country's project's overall performance as economic performance. Like many satisfactory, its institutional stand-alone technical assistance impact as partial, and the  ix sustainability of project benefits (iii) In a large and complex (see para. 21 above) as uncertain. operation which involves This agrees with the ratings given in many agencies, it is the PCR. important t h a t i n s t i t u t i o n a 1 PrinciDal Lessons Learned responsibilities be made very clear to avoid 25. The main lessons to be drawn possible duplication of from this experience are (paras 6.01 functions; and 6.02): (iv) Nearly all the project (i) Technical assistance studies suffered from operations should be inadequate d a t a designed with the availability. Greater absorptive capacities of effort should be made to local institutions in establish permanent mind. Project objectives mechanisms to collect, need to be realistic and store, and disseminate operational. If well data on the urban sector thought out, they can prior to commissioning help provide the focus expensive studies; and that the present operation lacked. At the (v) A survey of training same time, they can help needs in the urban sector to identify explicit and is necessary to provide precise performance the basis for effective indicators to guide capacity building. It project management and should be undertaken in supervision; order to identify the professional skills (ii) A technical assistance available in the country, operation can be used to so that full use can be revive a flagging sector made of them in future dialogue. Had this been urban projects. made explicit in the present operation, project design could have been better focussed to achieve this goal in a cost-effective manner;  PROJECT PERFORMANCE AUDIT REPORT SENEGAL URBAN MANAGEMENT AND REHABILITATION PROJECT (CREDITS 1458-SE AND SP-13) I. BACKGROUND A. Introduction 1.01 Senegal was the first country to borrow from the Bank to help finance a physical urban development project. The resulting operation, the Sites and Services Project (Cr. 336-SE) was approved in June 1972. No further lending operations were undertaken in the urban sector until approval of the operation reviewed here, the Urban Management. and Rehabilitation Project in April 1984. Referred to in this report as the "TA Project," this operation ended a hiatus of more than a decade in urban lending to the Bank's earliest urban sector client. The TA Project was followed by the Municipal and Housing Development Project (Cr. 1884-SE), approved in 1988. Because of its role in reviving a laggard urban sector dialogue between the GOS (the Government of Senegal) and IDA, an assessment of the TA Project is of particular interest. B. Macroeconomic Context 1.02 The tertiary sector -- a largely urban phenomenon covering commerce and other services -- dominates Senegal's economy, accounting for two-thirds of the country's GDP when the TA Project was appraised in 1984. The performance of other sectors, notably agriculture, but also industry, suffers adversely from periodic droughts brought on by harsh climate conditions in the semi-arid regions of the country. During project implementation (1984-90), Senegal's GDP reportedly grew at an average rate of 3.4 percent per annum, despite a fall in 1989. As a result, GNP per capita, which had been US$380 in 1984, rose to US$710 by 1990. 1.03 This increase, which gave Senegal the sixth highest GNP per capita among sub-Saharan countries, may not have led to an improvement in the living conditions of the country's urban population however for two reasons. First, a good part of the increase in US dollar terms can be explained by the significant appreciation of the local currency, the CFA franc, against the dollar over the 1984-90 period.' Secondly, Senegal's GNP figures may have been over-reported. A recent Bank economic memorandum noted that performance according to the national accounts was not substantiated by the much lower growth reported for key I As member of UMOA ("Union Mon6taire Ouest-Africaine" - West African Monetary Union) Senegal, like some eleven other countries in French- speaking Africa, uses the CPA franc, which is tied to the French franc at a parity of 50:1. In 1984, the rate was CFA 437 - US$1.00; by 1990, it became CFA 256 - US$1.00, due to depreciation of the dollar against the French franc over the period.  2 sectors of the economy.2 This possible inconsistency is just one example of the severe data problems in Senegal, which the TA Project itself had to contend with (para. 3.11). On balance, however, the project was implemented during a difficult period for Senegal's economy. Its start-up coincided with the beginning of Senegal's macroeconomic stabilization and adjustment programs, which were supported by the IMF and the Bank respectively. 1.04 Visible trade deficits, of on the order of some 30 percent of exports, were a constant feature of Senegal's international payments during the 1984-90 period. Balance of payments equilibrium was achieved to a large extent through inflows of concessional bilateral and multilateral funding, of which Senegal has been a major recipient. The country has been an important borrower from the Bank itself (para. 1.09). In spite of these financial inflows, export promotion remained a key feature of Senegal's macroeconomic management. This had traditionally been through credit to support increased production of groundnuts and other cash crops, as well as light industrial and tourism projects, having important implications for the capital city, Dakar. Each of these sectors experienced some difficulties during the 1984-90 period, especially through the erosion of Senegal's international competitiveness due to the appreciation of the exchange rate vis A via the country's competitors. Notwithstanding these obstacles, an efficiently functioning urban system, especially in Dakar, remains a necessary condition for a sustained improvement in Senegal's macroeconomic performance. C. The Urban Sector 1.05 In 1990, some 2.8 million people, or 38 percent of the Senegalese population lived in urban areas. Dakar overwhelmingly dominates the urban structure of the country. With an estimated 1.5 million inhabitants in 1990, it was home to more than half of Senegal's urban population. During the colonial period, Dakar was the administrative center for French West Africa as a whole, and was the focus of significant investment in urban infrastructure. Today, the city is still one of the region's busiest commercial and transport centers. One of the challenges for urban sector policy is to build up the comparative advantages of Dakar, but in the context of a more modest role for the city at the national, rather than regional, scale. This task is not made easier by the fact that urban development in Senegal has rarely been a priority concern either for GOS or IDA.' 1.06 With rapid population growth in excess of 4 percent per annum, Dakar's urbanized area spread into neighboring city jurisdictions, as has occurred in many other major world cities. Thus, the Dakar metropolitan area is 2 World Bank, Senegal: Macroeconomic Update, Sahel Department, Country Operations Division, December 30, 1991, box 1. Thus, a recent Bank study of public expenditure in Senegal did not deal explicitly with urban investments or maintenance. See, World Bank, Senezal: Public Expenditure Review, Sahel Department, Country Operations Division, December 31, 1991.  3 comprised of three urban communes (of Senegal's thirty-eight), namely: Dakar itself, Pikine, and Rufisque-Bargny. Each one has its own indirectly elected mayor. To help coordination, a Greater Dakar Authority (CUD - "Communaut6 Urbaine de Dakar") was created in 1983 to oversee most local services, notably solid waste collection and disposal and street maintenance. These three communes make up the region of "Cap Vert" (the name of the peninsula vhere Dakar stands), which is frequently mentioned in project documentation. 1.07 Other cities in Senegal are much smaller than Dakar: the next largest Thies and Kaolack, for instance, have populations in the 100,000-150,000 range. After these, there are only three more cities in the country with more than 50,000 inhabitants. Difficult conditions in semi-arid agricultural zones especially have stimulated urban population growth through rural migration. Such migration has put pressure on the cities, Dakar most of all, to meet the growing needs of urban services, infrastructure, and housing. 1.08 Institutional arrangements for managing urban development in Senegal are complex. Although local government (through urban communes and CUD in the case of Dakar) have wide-ranging formal responsibilities, central government (GOS) remains the key player in the sector. GOS' own intervention in the urban sector is itself very complex, involving at least three ministries: (planning - MOEF; housing - MUH; and infrastructure - MEQ), and many specialized agencies, four of which should be specifically mentioned here (housing bank - BHS; urban land development - SCAT-URBAM; real estate - SICAP; rental housing - SNHLM). When responsibilities overlap, a multiplicity of institutions and departments risks leading to inter-agency conflict and duplication of efforts. The TA Project took this complicated administrative setting as a parameter, rather than a policy variable over which it might exercise a simplifying influence. D. Bank Involvement 1.09 The Bank has been an active lender in Senegal in many sectors. Since 1981, lending has been exclusively through IDA. According to an OED study of 1989: "...on a country comparative basis, Senegal has been a preferred client of the Bank: with just 2.5 percent of the total population of all countries in "low- income Africa," it has received 4.7 percent of total World Bank commitments to those countries during 1970-86." In this review, OED was critical of the performance of Bank lending to Senegal, citing in particular: (i) unfounded optimism about future outcomes in structural adjustment operations; (ii) the lack of clear vision about which areas and sectors of the economy had most potential and were in need of support; and (iii) the Bank's failure to fully recognize the immense constraints on development in Senegal. Although not made with the present project specifically in mind, as will be demonstrated below, these observations are pertinent to some of its weaknesses: (i) unduly ambitious and optimistic objectives (paras. 2.15-2.16); (ii) uncertain underlying direction (para. 2.14); and (iii) insufficient attention to institutional, political, and World Bank, The World Bank and Senezal 1960-87, Operations Evaluation Department, August 31, 1989, Pg. ii.  4 technical obstacles to change in the urban sector in Senegal (paras. 2.05, 2.13 and 5.02). 1.10 The Bank has been less active with respect to urban development than in many other sectors. To date, there have been just three urban lending operations. The first of these, the Sites and Service Project (Cr. 336-SE), was particularly important, however, as it was the first Bank-financed urban operation worldwide. For an experimental project, it was a very large operation, aiming to deliver serviced lots to 140,000 people in Camberene (Dakar) and to another 12,000 people in Thins. The Camberene site by itself would have become the second largest urban settlement in Senegal at that time. This fact clearly illustrates the extent of Bank ambitions for the urban sector in the early 1970s, when the project was being prepared. Through the Sites and Services Project in Senegal, moreover, the Bank sought to develop a blueprint for an approach to be replicated worldwide.5 1.11 The formal objectives of the Sites and Services operation were to: (i) demonstrate the affordability of sites and services; (ii) mobilize private savings for investment in shelter; and (iii) develop an institutional capacity to execute similar programs in other cities. The project was not audited by OED, but a detailed and critical PCR was issued in October 1983, which included the Government's own evaluation of its successes and failures. Among the PCR's many criticisms, those particularly relevant to the TA Project under review here were: (i) poor cost recovery through insufficient political will on the part of the authorities and the absence of sanctions for non-payment; (ii) inadequate involvement of Senegalese staff in project preparation; and, related to the previous point, (iii) Bank attention to project preparation was so close that the project was seen as a "Bank project" for which GOS felt neither responsibility nor attachment (Sites and Services PCR, paras. 3.16, 6.19, and 8.03). In its final remarks in the PCR (pg. 73), GOS noted: "...there is a general feeling on the part of the Senegalese agencies concerned that the Bank experts sought to influence, in a way favorable to their own views, Senegalese customs and ways of thinking. Referred to as the 'expert mentality,' the behavior of the Bank personnel in supervising projects financed by it is perceived as an attempt to impose analyses, concepts and philosophy options that do not take sufficient account of the country's own choices." Some Senegalese officials voiced similar concerns to the Audit mission with respect to TA Project, which cast some of them into the role of passive observers. (paras. 4.03 and 5.07). 1.12 The PCR nonetheless reported a positive side to the Sites and Services Project experience. Particularly important was the successful implementation of the smaller component in the city of Thias which, through considerable local effort, became known as the "Senegalese" component. Also significant was the general acceptance of the project's basic philosophy of trying to reach low-income beneficiaries who previously had no access to public housing. One of the main lessons of the experience was that the operation had The experience of preparing this project provided important inputs to the now classic document Sites and Services Proiects: A World Bank Paver of April 1974, which has guided Bank operations of this type ever since.  5 been on far too large a scale to be an experiment. In the words of the PCR (para. 8.06): "A smaller project, backed by solid government support, would have had a smoother implementation history. The ThiAs experience would appear to support this conclusion." It was nevertheless a very important learning experience. Thirty Bank supervision missions were fielded over a period of ten years, a coefficient of 13.5 staff weeks per million US dollars of credit amount, approximately three times that of the TA Project. Clearly, a tremendous effort went into what was, despite the difficulties, a worthwhile operation. Perhaps sheer exhaustion on both sides, after such an effort helps explain the lack of continuity of Bank-financed urban operations after the formal closure of the Sites and Services Project in 1981. It took the project reviewed here to revive a flagging urban sector dialogue. The TA Project also successfully led to a follow-on operation, partly financed by IDA. 1.13 The resulting operation was the Municipal and Housing Development Project (Cr. 1884-SE), approved by the Board in 1988. The project, expected to be completed in June 1994, aims to strengthen local government administration. especially in Dakar, to provide sites and services, and to stimulate private sector housing delivery. Its municipal management component consists of: (i) a line of credit to finance local income generating subprojects through FECL ("Fonds d'Equipement des Collectivit6s Locales" - Municipal Infrastructure Fund); (ii) strengthening the maintenance capability of CUD ("Communaut6 Urbaine de Dakar" - Greater Dakar Authority); (iii) continuing technical assistance to the Traffic Bureau established under the present project and urban road investments; and (iv) implementing a fiscal cadastre for the Dakar metropolitan area. The project's housing component, in turn, has two parts: (i) annual delivery of 1,800 sites and services plots; and (ii) a line of credit to the Senegal Housing Bank (BHS) to finance a program of housing construction loans. The majority of these components were identified in connection with the TA Project. 1.14 Apart from urban projects per se, Dakar has benefitted from Bank- supported operations in other sectors, notably air transport, investment promotion, tourism, and fishing. Given the city's overwhelming importance in the national economy, operations in other sectors inevitably concentrate there. It would be illuminating to evaluate the impact of all of these operations on the efficiency of the city as the Senegal's principal focus of economic development. II. THE PROJECT A. Stand-alone Technical Assistance Proiects 2.01 Like technical assistance operations in other sectors, the project reviewed here aimed to provide institutional support to fill gaps in particular areas and guide sector policy makers. A recent OED review examined the results of a number of stand-alone technical assistance operations of this kind in  6 Africa.' The principal focus of the report vas on operations linked to structural adjustment programs, but many of the findings, particularly in connection with project design and objectives and the performance of foreign consultants, are relevant to an evaluation of the present operation as well. 2.02 The OED review examined nineteen technical assistance projects in seven countries, including Senegal, but not the TA Project itself. Only five of the nineteen were found to have been successful. Among the reasons for the high failure rate, according to the report, were: (i) lack of institutional development strategy on part of borrowers and recipient institutions; (ii) project design failed to take borrower's limited absorptive capacity into account; (iii) unclear project objectives and vague terms of reference; (iv) uneven performance by foreign consultants and lack of local ones; and (v) inadequate Bank supervision for a type of operation that places heavy demands upon supervision. In the case of the project reviewed here, the first two reasons -- namely (i) lack of institutional development strategy and (ii) limited absorption capacity of borrower institutions -- constrained the success of the operation. Despite such difficulties, however, the OED report argued that continuing technical assistance operations in Africa was worthwhile, as long as some recommended improvements were made. 2.03 Among the recommendations of the OED study, those particularly relevant to the present project were: (i) during preparation, the Bank should give special attention to ensuring that target institutions are willing and able to benefit from the expertise proposed; (ii) studies should be commissioned only when they are clearly needed and should be designed and implemented in full collaboration with the target agency; (iii) since institutional development through technical assistance projects seeks behavioral changes in societies, cultures and institutions, the Bank needs to evaluate behavioral issues more thoroughly; and (iv) the Bank should continue to support a borrower's use of local consulting firms in technical assistance projects, and encourage partnerships between local and expatriate firms. With the benefit of hindsight, the TA Project could have benefitted from the effective participation of target institutions and local expertise in Senegal, whose absence was detrimental to the success of the operation (para. 2.10). The experience of the present project confirmed these broader conclusions and recommendations. 2.04 Technical assistance for institutional development can also be provided as a component either of a physical development project or of a line of credit operation. Past experience in other countries has shown that borrowers may be willing to adopt unpopular policy or institutional measures (such as broadening a tax base) in exchange for more popular (and usually more substantial) financing for physical works. Offering such a trade-off to an institution which already has minimal institutional capacity appears to have been the more successful approach according to an early Bank-supported study: "in general, technical assistance has been more effective and more readily accepted when associated with an investment and when addressed to institutions that World Bank, Free-Standing Technical Assistance for Institutional Development in Sub-Saharan Africa, Operations Evaluation Department, April 19, 1990.  7 already have a relatively high level of performance."' A stand-alone technical assistance project which has no funds for investment cannot reward the agencies that make most institution building progress with large-scale financing of physical components. In the case of the operation reviewed here, however, the promise of a larger follow-on project did serve as an incentive for efficient implementation. The TA Project's catalytic role is discussed in detail later in this report (paras. 4.05-4.07). B. Issues in Proiect Preparation 2.05 As formal completion of the pioneering Sites and Services Project approached in 1981, the Bank found itself in a dilemma. It wanted to continue urban lending to its first urban sector borrower, but slow progress with the Sites and Services Project (paras. 1.10-1.12) did not warrant an expansion or continuation of that operation. Bank files of 1980/81 show that discussions for extending sites and services operations to other cities in Senegal, as GOS initially proposed, did not get very far. It was clear that, to prosper, urban sector lending had to take a new direction, but which one was not clear. In the absence of an alternative strategy, preparation of an engineering credit was an appropriate tactic for IDA to adopt. It was conceived as stopgap measure to put life back into an ailing urban lending program. At the outset, however, there were quite unrealistic expectations about what a technical assistance credit would be able to achieve by itself: "The objectives of the engineering credit would be: (i) identify policy reforms which maximize public savings and improve the efficiency of public entities and local authorities; (ii) ensure the rational selection of investment packages in support of productive sectors with private sector participation."' Achievement of the second goal, for example, would have required the complete institutional and financial reform of Senegal's complex urban sector (para. 2.10), for which the proposed operation offered no financial support. Nonetheless, project preparation was stimulated by the idea that studies alone could bring about profound changes in the way urban investment was made and urban maintenance undertaken in Senegal. 2.06 Preparation of the operation was entrusted by GOS's ad hoc project Preparation Committee to foreign consultants. As GOS' agent, the consultants became IDA's principal technical interlocutor. It was not clear from project documentation why GOS subcontracted the preparation effort in this way. According to some Bank staff interviewed for this Audit, GOS' own institutions had insufficient technical capacity to prepare an operation of the magnitude and complexity foreseen at appraisal. Audit mission interlocutors in Senegal felt that local capabilities had been underestimated. They saw the subcontracting arrangement with the consultants as necessary, however, to increase the chances that IDA would approve the proposal. The calibre of the consultants and their cultural and professional affinities with IDA missions were seen by GOS officials as guarantees of IDA approval. Baum, Warren C. and Tolbert, Stokes M., Investing in Develoument: Lessons of World Bank Experience, Oxford University Press, New York, 1985, pg. 376. Urban projects division memorandum of December 7, 1982.  8 2.07 There vere advantages and disadvantages in having an external agency play a preeminent role in the preparation of an IDA-financed project. On the plus side, it speeded preparation which took only five months. In addition, project preparation could take advantage of the equipment and facilities of the consultants' installations in Europe. The arrangement had its negative side too, however. Many GOS officials became alienated from a project proposal for which they felt no responsibility, a point discussed in more detail below (para. 2.10). In this respect, the Audit agrees with the PCR that the borrower's technical agencies should be fully involved at all stages of a project (PCR para. 47). Another problem was the lack of familiarity of some consultants with the urban sector in Senegal, as mentioned by the PCR (para. 19). This limited the relevance of some proposals, notably in relation to training, for example (para. 3.23). Finally, project preparation, with considerable technical and professional resources at its disposal, risked producing a complex proposal exceeding the absorptive capacity of the borrower's target institutions. If project design assumes an installed institutional capacity greater than really exists, the project risks failure. 2.08 In a worthwhile effort to coordinate its operations with other donors, IDA sought to mobilize co-financing for the project, but with little success. GTZ (German technical cooperation) initially showed a lot of interest, but preferred to pursue an urban upgrading program in a secondary city on its own. On the other hand, Canadian Technical Assistance, CIDA, eventually agreed to finance part of the project's Traffic Bureau component. C. Prolect Appraisal 2.09 The principal task of the appraisal mission was the relatively straightforward one of assessing the consultants' preparation report, which provided the basis for subsequent project documentation. The timing of the appraisal was opportune, occurring at a time when GOS needed urban sector inputs for the Seventh Economic and Social Development Plan (1985-89). In this respect, GOS saw the operation in part as a project preparation facility for priority sector investments under the Plan. IDA's appraisal, however, focussed on the technical quality of the study proposals presented. 2.10 The consultants continued to play the key role of project agent, acting on GOS' behalf during appraisal. The consultants clearly made a strong impression on the appraisal mission, as can be seen from this extract from the President's Report: "the...preparation team strongly recommended improved investment planning, policy formulation, and strengthened institutions as preconditions to the better management of the Cap Vert Region." (PR para. 66) A reader might well ask whether the borrower and its urban sector agencies were in agreement with the recommendation. However technically well founded, the consultants' recommendation would not go very far if it ran into conflict with the policies and priorities of those institutions. In this context a more direct policy dialogue between IDA and GOS was missing. Its absence demoralized GOS institutions, discouraging their active support for what some government officials saw as someone else's project, a 'consultants' project." 2.11 As appraised, the TA Project was indeed a large and complex undertaking. For that reason, close attention was correctly paid to arrangements  9 for project management, the responsibility for which was to be entrusted to an inter-ministerial Steering Committee. The consultants would continue as technical advisor to this Committee, as it had been to the ad hoc Preparation Committee. In retrospect, a simpler, more coherent design, with fewer components might have eased the management task and, at the same time, produced results more easily absorbed by the target institutions in Senegal. 2.12 The PCR justified the complex project organization on the basis of the complex institutional arrangements prevailing in Senegal's urban sector (PCR para. 12). As mentioned earlier (para. 1.08), the project did not take on the simplification of these arrangements as one of its principal tasks, even though the obstacles posed by the multiplicity of institutions with overlapping responsibilities was recognized at appraisal (PR para. 32). 2.13 The complexity of the project also exposed the inadequacy of project documentation. As an engineering credit, no appraisal report was produced. The President's Report, as the main technical document, was well prepared. With some twenty tasks foreseen under the project, however, the ten pages the PR was able to devote to operation as a whole were insufficient. Space was simply insufficient to convey the minimum technical information about the proposed operation to the Borrower and others in the Bank. How can such a summary help the Borrower prepare terms of reference for a complex subcomponent, such as establishing a Traffic Bureau, for example? The PR succinctly summarized this component costing US$1.9 million in ten lines onlyl The Audit agrees with the PCR that a full appraisal report would have been beneficial to the operation (PCR para. 55). Such a report would have helped clarify the project's focus at the design stage and thereafter guided its management both locally and through IDA supervision missions. D. Prolect Oblectives and Description 2.14 According to the Development Credit Agreement, the long-term objective of the TA Project was to: "support improvements in the Borrower's economic performance by stimulating public savings and by improving the operational efficiency of public entities involved in urban management, maintenance and shelter."(DCA Schedule 2) As a general statement of the operation's aims, this reasonably expressed the intention of its authors. The short-term objectives were: (i) to improve the allocation of scarce public resources in urban areas through better investment planning and programming; (ii) to formulate a housing and urban policy based on an appropriate allocation of resources; and (iii) to maintain, with a view to ultimately improving, the existing level of urban services in Dakar by enhancing its capacity to plan and program rehabilitation and maintenance works and to mobilize resources. 2.15 Although these short-term goals were laudable in themselves, they were overly ambitious, involving improvements in the urban sector that a TA operation alone could not achieve. The project did not offer, for instance, the means to maintain, much less improve, the level of urban services in Dakar. Financial resources for investment and operations would be needed in order to bring about such improvements. True, mention was made of resource mobilization, but studies of ways to improve Dakar's financial management, studies of new sources of funding, or even a new fiscal cadastre -- all subcomponents foreseen  10 under the project -- cannot by themselves guarantee adequate funding to maintain the level of urban services. The implementation of fiscal strengthening requires a concerted institution building effort and a specific management program that the project had not intended to provide. 2.16 As formulated, the goals of the TA Project thus held out the promise of spectacular results without, however, providing the means to achieve them. In this respect, they were one more example of overly ambitious expectations in Bank-financed technical assistance operations, as OED's earlier review of TA projects in sub-Saharan Africa found (paras. 2.01-2.03). Had the real objectives of the project been better thought through, they might have led to quantitative performance targets (e.g., halving the unit cost of low-income housing designs; achieving an agreed minimum level of urban investment; halving the number of road accidents, etc.) which could have given the operation the focus it lacked. Such targets would not only have made an assessment of project performance more straightforward, but also have provided an important guide for supervision. 2.17 At appraisal, the project included some 20 separate tasks, grouped into three parts: (i) Urban Investment Planning and Programming (US$1,688,000 or 25.1 percent of total costs); (ii) Housing and Urban Policy (US$1,625,000 or 24.2 percent of the total); and (iii) Municipal Policy (US$3,342,000 or 49.8 percent of total costs). A very broad range of subcomponents was proposed, as can be seen from the project description reproduced here. To help assess their performance, the Audit report has added the estimated cost of each subcomponent as well as its category (man. - management of the TA Project itself; study; or inst. - institution building):' Part A: Urban Investment Planning and Programming (i) Steering Committee (US$561,000): (a) preparatory and start-up activities including, inter alia, the selection of specialists and consultants (man.); and (b) technical monitoring during the execution of the Project (man.). (ii) Seventh Plan: Preparation of the urban sector investment program of the Seventh Plan, including inter alia: (a) the carrying out of a methodological study to define guidelines for appraisal and selection of urban rehabilitation projects for inclusion in the Seventh Plan and to monitor physical and DCA Schedule 2 - note 1: "Dakar" in the DCA means the CUD - 'Communaut6 Urbaine de Dakar", or the Greater Dakar Authority; note 2: US dollar figures in parentheses are cost estimates at appraisal, including physical and price contingencies taken from President's Report, Annex V.  11 financial implementation of urban rehabilitation projects (US$689,000 - study); (b) completion of the Dakar/Cap Vert detailed urban development plan in order to program rehabilitation works for serviced land to be included in the Seventh Plan (US$311,000 - study); and (c) completion of the transport plan for the Cap-Vert region in order to program investments for the Seventh Plan (US$127,000 - study). Part B: Housing and Urban Policy (i) Housing Finance (US$347,000 - study): (a) affordability analysis (including calculation of costs and prices of subsidized housing) and review of existing mechanisms of resource mobilization and of low-cost housing finance; and (b) on the basis of such analysis and review, formulation of recommendations aimed at improving resource mobilization and use and at increasing private sector participation in the financing of low-cost housing. (ii) Management of State-owned ProDerties (US$81,000 - inst.): analysis of state-owned properties and related charges and revenues and formulation of recommendations on means to reduce subsidies and to increase rent collection by parapublic housing companies and the central government. (iii) Rehabilitation of OHLM (US$114,000 - inst.): preparation of a rehabilitation program for OHLM, and thereafter preparation of a contrat plan consisting of diagnostic studies, detailed recommendations, and a timetable for implementing the recommended actions, all aimed at improving financial and managerial performance and at redirecting OHLR's activities to lower-standard, lower- cost housing operations affordable by the bottom two- thirds of the urban population. (iv) Land Development Agencv (US$336,000 - inst.): carrying out a feasibility study for the establishment of a parapublic enterprise for the servicing of urban land with infrastructure and for community facilities and for residential, commercial, and industrial use, and focussing, inter alia, on standards that would make serviced land affordable to the bottom two thirds of the  12 urban population, on financial arrangements, and on the obligations of other parapublic and private entities engaged in infrastructure service and delivery. (v) Design Alternatives for Land Develoyment: carrying out of a study on alternative designs for land development at different service levels, including the formulation of recommendations on standards and pricing policies designed to make serviced land affordable to the bottom two thirds of the urban population. (vi) Pilot Designs (US$747,000 - study): on the basis of the recommendations developed under Part B (v) of the project, the carrying out of: (a) a feasibility land development study for one or two sites in Dakar; and (b) detailed design and engineering of an infrastructure rehabilitation plan for part of central Pikine, taking into account the detailed urban development plan referred to in Part A.(ii) (b) above. Part C: Municipal Policy: (i) Municipal Organization and Management (US$911,000): (a) review and analysis of relationships between the Borrower' s central goveriment and Dakar, including the allocation of authority, responsibility and obligations with respect to delivery and maintenance of infrastructure services (study); (b) identification and implementation of actions to improve financial management of Dakar (study); (c) identification of new financial resources, including development and establishment of a fiscal cadastre, and improvement of existing mechanisms to recover land and related taxes in Dakar (study); and (d) analysis of the role and operating procedures of the Borrower's Fonds d'Equipement des Collectivites Locales and formulation of recommendations on its suitability as a financing instrument for municipal works projects (study).  13 (ii) Maintenance (US$507,000): (a) identification of rehabilitation and maintenance works and equipment required to rehabilitate and maintain existing levels of infrastructure services in Dakar, and the formulation of recommendations for alternative solutions for regular maintenance of such services (inst.); (b) establishment of a system to plan and program the maintenance and rehabilitation of Dakar's infrastructure services equipment, including the purchase of equipment and spare parts (inst.); and (c) establishment of a cost accounting and annual budgeting system for Dakar's Technical Department (inst.). (iii) Traffic Management (US$1,873, 000- inst.) establishment of a traffic bureau within the Directorate of Studies and Programming of the Ministry of Equipment and carrying out by such traffic bureau of: (a) preparation of a priority action program for traffic management in Dakar designed to improve circulation for pedestrians, buses and municipal service vehicles and to improve access to and conditions in central Pikine; and (b) implementation and monitoring of the first year of said traffic management program, including the surfacing of footpaths, paving of roads to simple design standards to allow bus access, traffic control measures, junction improvements and provision of traffic signals. (iv) Training Program (US$51,000 - training): (a) identification of technical and administrative staffing requirements for Dakar and establishment of a priority training program; and (b) training of selected staff of Dakar through short-term traineeships, workshops, and seminars. 2.18 Despite the broad range of studies and actions proposed, important aspects of the urban sector were not covered by the project. Conspicuous by their absence, for example, are references to basic sanitation and solid waste collection and disposal, which are critical issues affecting Dakar. Furthermore, the training component was surprisingly small, given the frequent references made to the shortages of skilled personnel in urban sector institutions in Senegal.  14 The project could not, of course, be expected to cover everything, but it is not clear what criteria were used to select the components that were included. In practice, it seemed that a combination of GOS priorities, the consultants' skills, and Bank staff interests determined the final composition of the project. 2.19 Four kinds of activities were foreseen under the TA Project. First, studies were to be undertaken into various aspects of the urban sector (urban planning, housing finance, pilot designs, and municipal organization). Secondly, institution building actions included improved management of state-owned properties, rehabilitation of OHLM, and creation of a land development agency and a traffic bureau. Thirdly, there was a very small trainin-t component. The final activity was management of the project itself. The data in Table I gives a rough breakdown of the estimated costs of these activities. Approximately equal shares at appraisal were allocated to studies and institutional building. The small amount for training has already been mentioned. Project management by itself was expected to account for 8.4 percent of project costs, a reasonable overhead for a complex operation. A simpler design would have considerably reduced this cost however. Unfortunately, due to inadequate cost accounting during implementation (details paras. 3.07-3.09), it was not possible to reconstruct Table 1 with the actual costs incurred by these activities. ..:. i ; S ,.:: . .0.:ui ,.............................. i. T,,,W ............ . .L . Senegal~~~~ Uran- Maaeet and Rehbltto ehia sitnce Polec Tlf'-t'"',''-Xt----,''h-i -10-- liX ---~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~....... - -L i. .. . . . . . . . . . . . . . . . - y - , ?. . . . . . .. ...............i-,, . -.i--iXia Activity No. 6 Tof Tank AnrialCs US1: :..'..,.',.'.',,,,.,' '"'''"''''' ''''''........'.. ;'j'1''''-,",,""$'''''': ' ''',,0-4.'.t,""""-"'"'''"""""""'-X'"""""'"' .. ...,.-.',-,.. .... . ...,,.-.. ........ . InstLbI .t.on buldn 2,.7,000... R . . .~~~~~~~~~~..... . . ... . ... . ... ... . .... ..... ...... ..... -.i..... ~~~~~~~~~~~~~~~~~~... ..... -- 3 . .... *~~~u Proec manaemet....00 ) t ' . . R R ; R G - ,rR29. .'R ' G RGRR ,, ~. . .. .. .. . .. ..1 .. . ...) ... .G.''''' -,;..'''....'''''.,'''''<,,'' ~~...... .... "' I~iaL Z1LQQ ource~~~~~~~~~~ President's Re ort~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Annex ............  15 III. PROJECT IMPLEMENTATION AND RESULTS A. Scheduling 3.01 The project was scheduled for completion by December 1987. Start-up delays and slower progress than foreseen led to two extensions of the Credit's closing date. The operation was completed in June 1990, two-and-a-half years behind schedule. The PCR (paras. 18-19) highlighted some of the reasons for the delays, which included: (i) financial and technical weakness of the institutions targeted for technical assistance; (ii) shortage of counterpart funds; and (iii) inadequate logistical support. Other problems mentioned by the PCR, notably the poor remuneration of Senegalese staff assigned to the project discouraged their more active participation.10 In addition, some foreign consultants were unfamiliar with the Senegalese context. Interlocutors of the Audit mission in Senegal confirmed the existence of these problems, adding that the mobilization of consultant teams took much longer than expected. In spite of the many difficulties faced during implementation, all IDA supervision missions except the first one (July 1985) reported project performance as being "problem free." 3.02 In contrast to delays in implementing the TA Project itself, IDA was prompt to initiate discussions about a possible follow-on operation. Terms of reference for the second supervision mission, which took place in Paris some twelve months after Credit effectiveness, included instructions for the mission to begin to identify a second project. Despite the tardiness in the execution of its own components, the TA Project moved efficiently towards fulfilling its implicit aim of reviving the urban sector dialogue in Senegal. B. Documents Produced 3.03 The physical output of reports produced through the project was substantial. Today, some eighty documents constitute a veritable library of documentation on the urban sector in Senegal, and about Dakar in particular. The Audit mission was impressed by the well-organized documentation center of the Project Management Unit in MEFP in Dakar, which appeared to have a complete set of the project's reports. With the assistance of the librarian, the mission was able to locate selected reports from among the operation's output. MEFP's set, however, was the only complete collection in Senegal. Only a few of the project reports were available in the Bank's Resident Mission, for example. 3.04 During discussions with GOS and Dakar sectoral agencies targeted for institution building, the Audit mission inquired about present management's knowledge of the project and the relevant studies produced. Disturbingly, only '